research thesis on health literacy

• Search Menu
• Themed Collections
• Editor's Choice
• Ilona Kickbusch Award
• Supplements
• Author Guidelines
• Submission Online
• Open Access Option
• Self-Archiving Policy
• About Health Promotion International
• Editorial Board
• Advertising and Corporate Services
• Journals on Oxford Academic
• Books on Oxford Academic

Article Contents

Introduction.

• < Previous

Improving health literacy in community populations: a review of progress

• Article contents
• Figures & tables
• Supplementary Data

Don Nutbeam, Bronwyn McGill, Pav Premkumar, Improving health literacy in community populations: a review of progress, Health Promotion International , Volume 33, Issue 5, October 2018, Pages 901–911, https://doi.org/10.1093/heapro/dax015

• Permissions Icon Permissions

Governments around the world have adopted national policies and programs to improve health literacy. This paper examines progress in the development of evidence to support these policies from interventions to improve health literacy among community populations. Our review found only a limited number of studies (n=7) that met the criteria for inclusion, with many more influenced by the concept of health literacy but not using it in the design and evaluation. Those included were diverse in setting, population and intended outcomes. All included educational strategies to develop functional health literacy, and a majority designed to improve interactive or critical health literacy skills. Several papers were excluded because they described a protocol for an intervention, but not results, indicating that our review may be early in a cycle of activity in community intervention research. The review methodology may not have captured all relevant studies, but it provides a clear message that the academic interest and attractive rhetoric surrounding health literacy needs to be tested more systematically through intervention experimentation in a wide range of populations using valid and reliable measurement tools. The distinctive influence of the concept of health literacy on the purpose and methodologies of health education and communication is not reflected in many reported interventions at present. Evidence to support the implementation of national policies and programs, and the intervention tools required by community practitioners are not emerging as quickly as needed. This should be addressed as a matter of priority by research funding agencies.

Defining and measuring health literacy

The past 25 years have seen extraordinary growth in interest in health literacy. A search on the term ‘health literacy’ in most publication databases shows negligible publications in the 1990s, rising steeply to many hundreds of papers published annually on the subject in the past few years. This surge in interest has been driven along by debate about the concept, definition and measurement of health literacy, and numerous studies that have investigated the relationship between health literacy and a wide range of health and social outcomes. A small but growing number of studies report on interventions to address the practical challenges of low health literacy in clinical settings, and describe approaches to improving health literacy in different clinical and community populations.

Health literacy is one of many domains of literacy—reflecting the fact that general literacy is both content and context specific. Literacy is generally understood as having two distinct components—those that are task-based, and those that are skills-based. It can be measured in absolute terms by distinguishing between those who can perform the tasks of reading and writing basic text and those who cannot, and in relative terms by assessing the skill differences between those who are able to perform relatively challenging literacy tasks and those who are not able to do so ( National Assessment of Adult Literacy, 2003 ). Individuals with higher levels of general literacy (high-level skills in reading, writing and understanding text) are more able to apply their skills in situations requiring specific content knowledge, or in new and unfamiliar contexts.

Health literacy can be described as the possession of literacy skills (reading and writing) and the ability to perform knowledge-based literacy tasks (acquiring, understanding and using health information) that are required to make health-related decisions in a variety of different environments (home, community, health clinic). Health literacy is also generally understood to include equivalent skills in numeracy. It has been defined and conceptualized in multiple ways ( Peerson and Saunders, 2009 ; Sorensen et al. , 2012 ). but almost all definitions of health literacy in common use have the same core elements describing the skills that enable individuals to obtain, understand and use information to make decisions and take actions that will have an impact on their health status. These are an observable set of skills that will vary from individual to individual.

These differences in skills have been categorized as functional , interactive and critical health literacy ( Nutbeam, 2000 ). Such a classification is derived from mainstream literacy studies and has the advantage of signalling the impact that differences in skill levels may have on health-related decisions and actions. Functional health literacy describes basic level skills that are sufficient for individuals to obtain relevant health information (for example on health risks, and on how to use the health system), and to be able to apply that knowledge to a range of prescribed activities. Interactive health literacy describes more advanced literacy skills that enable individuals to extract health information and derive meaning from different forms of communication, to apply new information to changing circumstances, and engage in interactions with others to extend the information available and make decisions. Critical health literacy describes the most advanced literacy skills which can be applied to critically analyse information from a wide range of sources, and information relating to a greater range of health determinants, and to use this information to exert greater control over life events and situations that impact on health. This differentiation between functional, interactive and critical health literacy have been used and refined by other authors ( Ishikawa et al. , 2008 ; Chinn, 2011 , Sykes et al. , 2013 ).

Such a categorisation also helps to distinguish between interventions that are task-based—designed to develop specific skills to manage prescribed activities (medication adherence, behaviour change), and interventions that are skills based—designed to develop generic, transferable skills that equip people to make a range of more autonomous decisions relating to their health, and to adapt to changing circumstances. The concepts of interactive and critical health literacy connect closely to modern concepts of health promotion. In this case, health literacy has been viewed as a personal and population asset offering a route to greater autonomy and control over health decision-making ( Nutbeam, 2008 ; Pleasant and Kuruvilla, 2008 ; Mårtensson and Hensing, 2012 ). It is through this focus on skills development and empowerment that the concept of health literacy has the potential to have a distinctive influence on the purpose and methodologies of health education and communication.

Developing a ‘universal’ measure of health literacy that can be applied to diverse populations is proving to be very challenging ( Jordan et al. , 2011 ; Haun et al. , 2014 ). Measurement tools need to be able to assess relative differences in health literacy skills, and the ability of individuals to apply those skills to achieve health outcomes in different circumstances. Several simple measures of functional health literacy have been tested, refined and validated over the past 20 years to provide short screening tools for clinicians to use in everyday practice with a broad range of populations ( Davis et al. , 1993 ; Parker et al. , 1995a ; Weiss et al. , 2005 ). These measures were designed and are most useful as screening tools in clinical practice, but are generally insufficient to measure relative differences in cognitive skills in population studies. Progress has also been made in measuring interactive and critical health literacy in clinical populations ( Ishikawa et al. , 2008 ). Currently work is underway in several countries to develop and adapt existing measurement tools for health literacy that can be applied to population studies, can discriminate between relative differences in health literacy, and importantly, can be used to assess change in individuals and populations following intervention. More sophisticated (and complex) tools are emerging but are not yet observably in use in intervention studies ( Chinn and Mccarthy, 2013 ; Jordan et al. , 2013 ; Osborne et al. , 2013 ; Sorensen et al. , 2013 ).

Improving health literacy in populations

In response to surveys that have indicated high rates of poor health literacy in populations, governments and national agencies in countries as diverse as the US, China, Australia and some European nations have developed national strategies and targets to improve health literacy in their populations ( Heijmans et al. , 2015 ; Us Department of Health and Human Services, 2010 ; Australian Commission on Safety and Quality in Health Care, 2014 ; Chinese Ministry of Health, 2008 ). As these policies and other government responses have emerged there has been increasing attention given to interventions to address the challenges posed by low health literacy in populations, and to improve health literacy in populations.

Health literacy can be improved through the provision of information, effective communication and structured education. It can be regarded as a measurable outcome to health education or patient education. Improvements in health literacy can be assessed through the measurement of changes to the knowledge and skills that enable well-informed and more autonomous health decision-making. Differences in communication methods, media and content will result in different learning outcomes and associated behavioural and health outcomes. In turn, individual responses to information and education will be moderated by the environment in which they occur.

To date, the great majority of reported interventions have been in clinical settings, and generally focus on task-directed, functional health literacy. As the number of reported intervention studies has increased there have been some helpful reviews ( Sheridan et al. , 2011 ; Manafo and Wong, 2012 ; Taggart et al. , 2012 ). Taken as a whole, these reviews provide broadly consistent evidence that comprehension of health information and advice among individuals with low health literacy can be improved through modifications to communication and other mixed-strategy interventions. These improvements are associated with better health outcomes including changes to identifiable risks for chronic disease, and among those with established disease, reduced reported disease severity, unplanned emergency department visits and hospitalizations.

These reviews also highlight some of the practical challenges experienced in the clinical environment that will often limit communication to the transmission of factual information on how to use medications and health care services, and the knowledge and skills necessary for successful self-management of a chronic disease such as diabetes and arthritis. These constraints often mean that the educational methods used do not enable interactive communication, nor support the learning that will enable a high level of autonomy in decision-making. Although the Manafo and Wong review (2012) specifically examined interventions that they judged to be directed at improving interactive health literacy, the reviews include few interventions outside of clinical settings, and very few that could be classified as skills-directed, designed to develop interactive and critical health literacy skills.

These reviews also include many interventions based on the well-established knowledge/attitudes/behaviour (KAB) conceptual framework, and studies that used proxy measures of health literacy, usually knowledge improvement. The Taggart et al. review (2012) includes many such interventions. There is nothing intrinsically wrong with this framework, and in the application of more sophisticated psycho-social theories of behaviour change that have been developed in response to the shortcomings of the traditional KAB model (some of which are employed in the interventions). However, such interventions do not always reflect the skills-directed methods and learning theories that are required to develop interactive and critical health literacy, and the use of proxy measures that are limited to knowledge improvement may present a risk that ‘health literacy’ is being used as a convenient, contemporary label to describe more traditional task-directed health education interventions.

Taken overall, these reviews reveal significant ‘work in progress’ in relation to health literacy, providing consistent evidence that individuals with lower health literacy can be identified in clinical settings, supported to make positive improvements in their understanding of specific clinical conditions and related risks, and helped to develop functional skills to modify the behaviours that produce improved health outcomes.

This paper builds on this earlier work by examining the progress to date in the development of interventions to improve health literacy with community (non-clinical) populations, searching in particular for interventions that are skills-directed and where improved interactive or critical health literacy is targeted as an outcome.

Literature search strategy

Electronic databases were searched for key words using search terms: ‘health literacy’ AND intervention AND (measure OR evaluation). Databases searched were Medline via Ovid, PsychInfo, SCOPUS, CINAHL and Web of Science. PubMed was searched by title using search terms: ‘health literacy’ AND intervention only, in an attempt to reduce the number of possible papers. Only studies accessible from the databases above in the English language were considered.

Study screening

Health literacy was identified as a significant outcome of interest

Intervention was designed to improve health literacy

Intervention was directed to an identified community population (non-clinical—no existing condition)

Evaluation included a well-defined measure of health literacy

Interventions directed at individuals and groups with established disease, and/or designed to improve specific clinical outcomes (compliance with medication, self-management of disease) were not included. Interventions targeted at health practitioners or practitioners in training were also excluded. We separately identified several papers that described interventions to improve mental health literacy. These were excluded and are discussed further below.

Study selection

The results of study selection are summarized in Figure 1 . It shows that from an initial pool of 1117 papers the majority were excluded because they did not report on interventions, or were interventions directed at a population with an established condition. Of the 57 papers subsequently considered for inclusion, 50 were excluded for the reasons indicated in the Figure 1 .

Flow chart describing included and excluded articles.

Reasons for exclusion

The review process confirmed that the use of the term ‘health literacy’ is common in searchable abstracts and key words. Even with the use of limiting search terms, the majority of the initial 1117 studies identified were descriptive studies of health literacy, many reporting on the relationship between that concept (variably defined and measured) and a range of social, demographic and health outcome variables.

Excluding papers describing an intervention involved some judgment and we may not have got this right every time. Many were straightforward, excluded on the basis that they described the results of a clinical intervention in a population with established disease. Others were more difficult to judge and worth acknowledging. For example, several of the excluded papers used health literacy to define or segment a population, but not as the target or outcome of the intervention. For example, Faruqi et al. (2015) aimed to enhance preventive care for patients identified as having low health literacy; Zoellner et al. (2015) investigated whether the health literacy status of employees influenced the outcomes of a worksite weight loss program; and Lanpher et al. (2016) reported on a weight loss intervention where the outcomes were weight and engagement by low or adequate health literacy levels. In addition, Banbury et al. (2014) grouped participants of similar health literacy levels to assess the acceptability of a telehealth literacy intervention; and Porter et al. (2016) determined the health literacy status of participants to compare self-monitoring skills between groups.

A number of papers described interventions to improve health literacy, but did not measure change in health literacy as an outcome of the study and were therefore also excluded. White et al. (2013) assessed the quality of educational materials aimed at obesity prevention among preschoolers; and Zoellner et al. (2016) reported the ‘effects of a behavioural and health literacy intervention to reduce consumption of sugar-sweetened beverages’. In this study health literacy was used as a variable to examine retention rates, engagement and behavioural outcomes.

During the course of the review we also identified several interventions to improve ‘mental health literacy’ ( Li et al. , 2013 ; Pinto-Foltz et al. , 2011 ; Reavley et al. , 2014 ; Skre et al. , 2013 ; Taylor-Rodgers and Batterham, 2014 ). These studies have been significantly influenced by the work of Jorm ( Jorm et al. , 1997 ; Jorm, 2015 ) who has argued that mental health literacy needs to be considered as a discrete sub-discipline of health literacy. Most focused on addressing stigma or improving knowledge and understanding of mental health and illness ( Pinto-Foltz et al. , 2011 ; Skre et al. , 2013 ; Taylor-Rodgers and Batterham, 2014 ). We excluded these studies from the review because of significant inconsistencies in the definition of mental health literacy, and wide variation in the measurement instruments used. We have taken a view that this field of study is sufficiently distinctive to warrant separate examination.

Finally, a number of papers were excluded because they described a protocol for an intervention, but did not provide results, ( Batterham et al. , 2014 ; McCaffery et al. , 2016 ; Parker et al. , 2012 ; Rogers et al. , 2014 ; Steckelberg et al. , 2009 ). These are important because they indicate that there may be a pipeline of future interventions to be reported, and that our review may be early in a cycle of interest in community intervention research.

Included studies

Summary of health literacy interventions

Only three of the reported studies ( Chervin et al. , 2012 ; Jay et al. , 2009 ; Soto Mas et al. , 2015 ) used previously established measures of health literacy (such as TOFHLA and S-TOFHLA) ( Parker et al. , 1995b ; Baker et al. , 1999 ); the others used customized measures of a set of variable skills that enable people to obtain, process and use information for health decisions and actions—judged to be broadly compatible with the more widely used formal definitions described above ( Sorensen et al. , 2012 ). One study ( Jay et al. , 2009 ) used the S-TOFHLA to determine health literacy status and then went on to measure change in health literacy using a study-specific customized measure. All included studies relied on self-reported measures of health literacy.

The examples we found covered a range of settings, including online programs, adult education, school and a supermarket-based multimedia program. All included education or communication strategies designed to develop functional health literacy skills directed towards specific improvements in knowledge and understanding, and most were also directed towards pre-defined behavioural responses. Five had elements that were compatible with the development of interactive and/or critical health literacy skills ( Austvoll-Dahlgren et al. , 2012 ; Chervin et al. , 2012 ; Carroll et al. , 2015 ; Fleary et al. , 2013 ; Xie, 2011 ). Educational methods varied considerably from formal classes ( Fleary et al. , 2013 ; Soto Mas et al. 2015 ), home visiting ( Carroll et al. , 2015 ) and study circles ( Chervin et al. , 2012 ), through multi-media ( Jay et al. , 2009 ) and eHealth/online interventions ( Austvoll-Dahlgren et al. , 2012 ; Xie, 2011 ).

The interventions focused on the needs of specific population groups throughout the life-course (parents, adult learners, older people), and addressed a range of topical health issues including food choices, physical activity and parenting. Most were also targeted at populations and in settings that have a higher proportion of individuals with lower health literacy.

The study designs were variable with the majority including some form of comparison or control group. Most described positive results from the intervention in terms of knowledge and skills gain compatible with the health literacy concept. Significant changes in health literacy were reported when comparing scores before and after the intervention ( Carroll et al. , 2015 ; Soto Mas et al. , 2015 ; Xie, 2011 ; Chervin et al. , 2012 ; Fleary et al. , 2013 ). One study ( Jay et al. , 2009 ) indicated differential results between adequate and limited health literacy participants.

Our review found only a limited number of studies that met the deliberately restrictive criteria, with many more that were evidently influenced by the concept of health literacy but had not demonstrably used the concept in the design and measurement of outcomes. Those included were diverse in setting, population and intended outcomes. The small numbers make it difficult to provide commentary that could be used as a general guide to future interventions, but some tentative observations can be made.

In terms of definition and measurement , all studies were designed to develop functional health literacy skills, and five of seven incorporated elements that were compatible with the development of interactive and/or critical health literacy skills. The measures of health literacy were conceptually narrow in some cases. For example, Carroll et al. (2015) and Austvoll-Dahlgren et al. (2012) used screening instruments not originally designed for population intervention studies. In most cases the studies used instruments that were customized to the specific intervention and outcome objectives of the study, for example those used by Fleary et al. (2013) and Jay et al. (2009) . These limitations on the measurement instruments are not surprising given the timing of the research reported. The more comprehensive measures of health literacy referred to above were not known or accessible for most of these studies, and we should anticipate that new studies will benefit from these and other more sophisticated measures ( Chinn and Mccarthy, 2013 ; Jordan et al. , 2013 ; Osborne et al. , 2013 ; Sorensen et al. , 2013 ).

In terms of education and communication methods , almost all of the examples used interactive communication methods to develop health literacy skills. Some were task directed, for example Jay et al. (2009) was focused on developing health literacy skills to improve competencies in comprehending food labels. Others such as Soto Mas et al. (2015) and Xie (2011) were more focused on the development of generic health literacy skills. In Carroll et al. (2015) , the communication was also personalized to the needs of individual participants.

Taken as a whole, it was disappointing to find so few studies that were actively using the concept of health literacy in their design and evaluation. The results suggest that the concept and rhetoric relating to health literacy has excited the interest of public health researchers, practitioners and policy makers, but that this interest has not yet been converted into substantive advances in public health interventions. Relatively few reported interventions directed at healthy (non-clinical) populations have incorporated the concepts of interactive and critical health literacy. The widespread use of the term in titles and key words indicates continuing scientific interest in the concept, but a risk that the term has become a fashionable way of relabeling established intervention methods and outcomes.

It is not difficult to understand why health literacy has become a subject of wide interest in the past decade. For researchers interested in health and disease causality, health literacy offers a convenient and logical summary definition of a pre-existing condition/risk that can be used to understand and explain variation in health and disease outcomes. There is a substantial and growing literature that confirms the importance of the concept in clinical practice and public health. For those interested in the evaluation of information, education and communication (IEC) interventions, health literacy has long been proposed as a useful outcome measure ( Nutbeam 1998 ).

For clinicians, work over many years, mainly in the USA, has established health literacy as an identifiable and manageable risk in clinical care, of particular importance in the management of long-term and complex conditions that depend upon successful patient engagement and management. For public health practitioners, health literacy is conceptually attractive in its fit with contemporary health promotion, understood as a personal ‘asset’ that can be developed through educational and other interventions to support greater personal and community control over a range of determinants of health.

For policy-makers, health literacy has the attraction of being a sufficiently malleable concept to be used to support a full spectrum of policy positions—at best to reflect a desire for greater patient and public engagement in health decision-making, and at worst, to shift the burden of protecting and improving health away from the state to the individual.

All of this attention is undoubtedly supporting advances in our knowledge and understanding of the concept, its relative importance as a health determinant, its measurement, and its potential for use to guide clinical practice, public health and public policy. However, this review provides a stark reminder that the academic interest and attractive rhetoric surrounding health literacy needs to be tested more often and more systematically through intervention experimentation in a wide range of populations using valid and reliable measurement tools.

We recognize that our review methodology may not have captured all the studies using health literacy to guide intervention development and assess outcomes in community populations. However, the paucity of intervention studies that clearly reflect the ‘skills-directed’ methods at the heart of the health literacy concept, and the continuing challenge of measurement should give cause for concern. The distinctive influence of the concept on the purpose and methodologies of health education and communication is not reflected in many reported interventions at present. That said, there is an encouraging pipeline of intervention studies that have reported protocols but not yet on results. Overall, evidence to support the implementation of national policies and programs, and the intervention tools required by community practitioners are not emerging as quickly as needed. This should be addressed as a matter of priority by research funding agencies.

Australian Commission on Safety and Quality in Health Care . 2014 . National Statement on Health Literacy [Online]. Available: http://www.safetyandquality.gov.au/wp-content/uploads/2014/08/Health-Literacy-National-Statement.pdf [Accessed 22 June 2016].

Austvoll-Dahlgren A. , Bjorndal A. , Odgaard-Jensen J. , Helseth S. ( 2012 ) Evaluation of a web portal for improving public access to evidence-based health information and health literacy skills: a pragmatic trial . Plos One 7 , e37715 .

Google Scholar

Baker D. W. , Williams M. V. , Parker R. M. , Gazmararian J. A. , Nurss J. ( 1999 ) Development of a brief test to measure functional health literacy . Patient Education and Counseling 38 , 33 – 42 .

Banbury A. , Parkinson L. , Nancarrow S. , Dart J. , Gray L. , Buckley J. ( 2014 ) Multi-site videoconferencing for home-based education of older people with chronic conditions: The Telehealth Literacy Project . Journal of Telemedicine and Telecare 20 , 353 – 359 .

Batterham R. W. , Buchbinder R. , Beauchamp A. , Dodson S. , Elsworth G. R. , Osborne R. H. ( 2014 ) The OPtimising HEalth LIterAcy (Ophelia) process: study protocol for using health literacy profiling and community engagement to create and implement health reform . BMC Public Health 14 , 694 .

Carroll L. N. , Smith S. A. , Thomson N. R. ( 2015 ) Parents as teachers health literacy demonstration project integrating an empowerment model of health literacy promotion into home-based parent education . Health Promotion Practice 16 , 282 – 290 .

Chervin C. , Clift J. , Woods L. , Krause E. , Lee K. ( 2012 ) Health Literacy in Adult Education A Natural Partnership for Health Equity . Health Promotion Practice 13 , 738 – 746 .

Chinese Minisry of Health 2008 . 66 tips of health: Chinese resident health literacy manual , Beijing , People's Medical Publishing House .

Google Preview

Chinn D. ( 2011 ) Critical health literacy: A review and critical analysis . Social Science & Medicine 73 , 60 – 67 .

Chinn D. , Mccarthy C. ( 2013 ) All Aspects of Health Literacy Scale (AAHLS): developing a tool to measure functional, communicative and critical health literacy in primary healthcare settings . Patient Education and Counseling 90 , 247 – 253 .

Davis T. C. , Long S. W. , Jackson R. H. , Mayeaux E. J. , George R. B. , Murphy P. W. et al. , ( 1993 ) Rapid estimate of adult literacy in medicine: a shortened screening instrument . Family Medicine 25 , 391 – 395 .

Faruqi N. , Lloyd J. , Ahmad R. , Yeong L. L. , Harris M. ( 2015 ) Feasibility of an intervention to enhance preventive care for people with low health literacy in primary health care . Australian Journal of Primary Health 21 , 321 – 326 .

Fleary S. , Heffer R. W. , Mckyer E. L. , Taylor A. ( 2013 ) A parent-focused pilot intervention to increase parent health literacy and healthy lifestyle choices for young children and families . ISRN Family Med 2013 , 619389.

Haun J. N. , Valerio M. A. , Mccormack L. A. , Sorensen K. , Paasche-Orlow M. K. ( 2014 ) Health literacy measurement: an inventory and descriptive summary of 51 instruments . Journal of Health Communication 19 Suppl 2 , 302 – 333 .

Heijmans M. , Uiters E. , Rose T. , Hofstede J. , Devillé W. , Van Der Heide I. et al.  ( 2015 ) Study on sound evidence for a better understanding of health literacy in the European Union . Brussels : European Commission .

Ishikawa H. , Takeuchi T. , Yano E. ( 2008 ) Measuring functional, communicative, and critical health literacy among diabetic patients . Diabetes Care 31 , 874 – 879 .

Jay M. , Adams J. , Herring S. J. , Gillespie C. , Ark T. , Feldman H. et al. , ( 2009 ) A randomized trial of a brief multimedia intervention to improve comprehension of food labels . Preventive Medicine 48 , 25 – 31 . 7p.

Jordan J. E. , Buchbinder R. , Briggs A. M. , Elsworth G. R. , Busija L. , Batterham R. et al.  ( 2013 ) The health literacy management scale (HeLMS): a measure of an individual's capacity to seek, understand and use health information within the healthcare setting . Patient Education and Counseling 91 , 228 – 235 .

Jordan J. E. , Osborne R. H. , Buchbinder R. ( 2011 ) Critical appraisal of health literacy indices revealed variable underlying constructs, narrow content and psychometric weaknesses . Journal of Clinical Epidemiology 64 , 366 – 379 .

Jorm A. F. ( 2015 ) Why We Need the Concept of “Mental Health Literacy” . Health Communication 30 , 1166 – 1168 .

Jorm A. F. , Korten A. , Jacomb P. , Christensen H. , Rodgers B. , Pollitt P. ( 1997 ) “Mental health literacy”: a survey of the public’s ability to recognise mental disorders and their beliefs about the effectiveness of treatment . Medical Journal of Australia 166 , 182 – 186 .

Lanpher M. G. , Askew S. , Bennett G. G. ( 2016 ) Health literacy and weight change in a digital health intervention for women: A randomized controlled trial in primary care practice . Journal of Health Communication 21 , 34 – 42 .

Li T. M. H. , Chau M. , Wong P. W. C. , Lai E. S. Y. , Yip P. S. F. ( 2013 ) Evaluation of a Web-Based Social Network Electronic Game in Enhancing Mental Health Literacy for Young People . Journal of Medical Internet Research 15 ,

Manafo E. , Wong S. ( 2012 ) Health literacy programs for older adults: A systematic literature review . Health Education Research 27 , 947 – 960 .

Mårtensson L. , Hensing G. ( 2012 ) Health literacy–a heterogeneous phenomenon: a literature review . Scandinavian Journal of Caring Sciences 26 , 151 – 160 .

Mccaffery K. J. , Morony S. , Muscat D. M. , Smith S. K. , Shepherd H. L. , Dhillon H. M. et al.  ( 2016 ) Evaluation of an Australian health literacy training program for socially disadvantaged adults attending basic education classes: study protocol for a cluster randomised controlled trial . BMC Public Health 16 , 1 – 10 .

National Assessment of Adult Literacy . 2003 . Definition of literacy. Available: http://nces.ed.gov/NAAl/fr_definition.asp [Accessed 8 January 2015].

Nutbeam, D. (1998) Evaluating health promotion - progress, problems and solutions. Health Promotion International 13 , 27–44.

Nutbeam D. ( 2000 ) Health literacy as a public health goal: a challenge for contemporary health education and communication strategies into the 21st century . Health Promotion International 15 , 259 – 267 .

Nutbeam D. ( 2008 ) The evolving concept of health literacy . Social Science and Medicine 67 , 2072 – 2078 .

Osborne R. H. , Batterham R. W. , Elsworth G. R. , Hawkins M. , Buchbinder R. ( 2013 ) The grounded psychometric development and initial validation of the Health Literacy Questionnaire (HLQ) . BMC Public Health 13 , 658.

Parker E. J. , Misan G. , Chong A. , Mills H. , Roberts-Thomson K. , Horowitz A. M. et al. , ( 2012 ) An oral health literacy intervention for Indigenous adults in a rural setting in Australia . BMC Public Health 12 ,

Parker R. M. , Baker D. W. , Williams M. V. , Nurss J. R. ( 1995a ) The test of functional health literacy in adults . Journal of General Internal Medicine 10 , 537 – 541 .

Parker R. M. , Baker D. W. , Williams M. V. , Nurss J. R. ( 1995b ) The test of functional health literacy in adults: a new instrument for measuring patients' literacy skills . Journal of General Internal Medicine 10 , 537 – 541 .

Peerson A. , Saunders M. ( 2009 ) Health literacy revisited: what do we mean and why does it matter? Health Promotion International 24 , 285 – 296 .

Pinto-Foltz M. D. , Logsdon M. , Myers J. A. ( 2011 ) Feasibility, acceptability, and initial efficacy of a knowledge-contact program to reduce mental illness stigma and improve mental health literacy in adolescents . Social Science & Medicine 72 , 2011 – 2019 .

Pleasant A. , Kuruvilla S. ( 2008 ) A tale of two health literacies: public health and clinical approaches to health literacy . Health Promotion International 23 , 152 – 159 .

Porter K. , Chen Y. , Estabrooks P. , Noel L. , Bailey A. , Zoellner J. ( 2016 ) Using teach-back to understand participant behavioral self-monitoring skills across health literacy level and behavioral condition . Journal of Nutrition Education and Behavior 48 , 20 – 26.e1 .

Reavley N. J. , Mccann T. V. , Cvetkovski S. , Jorm A. F. ( 2014 ) A multifaceted intervention to improve mental health literacy in students of a multicampus university: a cluster randomised trial . Social Psychiatry and Psychiatric Epidemiology 49 , 1655 – 1666 .

Rogers E. A. , Fine S. , Handley M. A. , Davis H. , Kass J. , Schillinger D. ( 2014 ) Development and Early Implementation of The Bigger Picture, a Youth-Targeted Public Health Literacy Campaign to Prevent Type 2 Diabetes . Journal of Health Communication 19 , 144 – 160 .

Sheridan S. L. , Halpern D. J. , Viera A. J. , Berkman N. D. , Donahue K. E. , Crotty K. ( 2011 ) Interventions for individuals with low health literacy: a systematic review . Journal of Health Communication 16 , 30 – 54 .

Skre I. , Friborg O. , Breivik C. , Johnsen L. I. , Arnesen Y. , Wang C. E. ( 2013 ) A school intervention for mental health literacy in adolescents: effects of a non-randomized cluster controlled trial . BMC Public Health 13 , 873.

Sorensen K. , Van Den Broucke S. , Fullam J. , Doyle G. , Pelikan J. , Slonska Z. et al.  ( 2012 ) Health literacy and public health: a systematic review and integration of definitions and models . BMC Public Health 12 , 80 .

Sorensen K. , Van Den Broucke S. , Pelikan J. M. , Fullam J. , Doyle G. , Slonska Z. et al.  ( 2013 ) Measuring health literacy in populations: illuminating the design and development process of the European Health Literacy Survey Questionnaire (HLS-EU-Q) . BMC Public Health 13 , 948 .

Soto Mas F. , Ji M. , Fuentes B. O. , Tinajero J. ( 2015 ) The Health Literacy and ESL study: a community-based intervention for Spanish-speaking adults . Journal of Health Communication 20 , 369 – 376 .

Steckelberg A. , Hulfenhaus C. , Kasper J. , Muhlhauser I. ( 2009 ) Ebm@school–a curriculum of critical health literacy for secondary school students: results of a pilot study . International Journal of Public Health 54 , 158 – 165 .

Sykes S. , Wills J. , Rowlands G , Popple K. ( 2013 ) Understanding critical health literacy: a concept analysis . BMC Public Health 13 , 1 .

Taggart J. , Williams A. , Dennis S. , Newall A. , Shortus T. , Zwar N. et al. , ( 2012 ) A systematic review of interventions in primary care to improve health literacy for chronic disease behavioral risk factors . BMC Family Practice 13 , 1 .

Taylor-Rodgers E. , Batterham P. J. ( 2014 ) Evaluation of an online psychoeducation intervention to promote mental health help seeking attitudes and intentions among young adults: randomised controlled trial . Journal of Affective Disorders 168 , 65 – 71 .

Us Department of Health and Human Services 2010 . National Action Plan to Improve Health Literacy . Washington DC : Office of Disease Prevention and Health Promotion .

Weiss B. D. , Mays M. Z. , Martz W. , Castro K. M. , Dewalt D. A. , Pignone M. P. et al.  ( 2005 ) Quick assessment of literacy in primary care: the newest vital sign . Annals of Family Medicine 3 , 514 – 522 .

White R. O. , Thompson J. R. , Rothman R. L. , Mcdougald Scott A. M. , Heerman W. J. , Sommer E. C. et al.  ( 2013 ) A health literate approach to the prevention of childhood overweight and obesity . Patient Education & Counseling 93 , 612 – 618 .

Xie B. ( 2011 ) Effects of an eHealth literacy intervention for older adults . Journal of Medical Internet Research 13 , 409 – 427 .

Zoellner J. , You W. , Almeida F. , Blackman K. C. , Harden S. , Glasgow R. E. et al.  ( 2015 ) The influence of health literacy on reach, retention, and success in a worksite weight loss program . American Journal of Health Promotion

Zoellner J. M. , Hedrick V. E. , You W. , Chen Y. , Davy B. M. , Porter K. J. et al.  ( 2016 ) Effects of a behavioral and health literacy intervention to reduce sugar-sweetened beverages: a randomized-controlled trial . The International Journal of Behavioral Nutrition and Physical Activity 13 , 38 .

Email alerts

Citing articles via.

• Recommend to Your Librarian
• Journals Career Network

Affiliations

• Online ISSN 1460-2245
• Print ISSN 0957-4824
• Copyright © 2024 Oxford University Press
• About Oxford Academic
• Publish journals with us
• University press partners
• What we publish
• New features  
• Open access
• Institutional account management
• Rights and permissions
• Get help with access
• Accessibility
• Advertising
• Media enquiries
• Oxford University Press
• Oxford Languages
• University of Oxford

Oxford University Press is a department of the University of Oxford. It furthers the University's objective of excellence in research, scholarship, and education by publishing worldwide

• Copyright © 2024 Oxford University Press
• Cookie settings
• Cookie policy
• Privacy policy
• Legal notice

This Feature Is Available To Subscribers Only

Sign In or Create an Account

This PDF is available to Subscribers Only

For full access to this pdf, sign in to an existing account, or purchase an annual subscription.

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here .

Loading metrics

Open Access

Peer-reviewed

Research Article

Trends of global health literacy research (1995–2020): Analysis of mapping knowledge domains based on citation data mining

Roles Conceptualization, Supervision, Validation, Writing – original draft

Affiliation Research Institute of Social Development, Southwestern University of Finance and Economics, Chengdu, China

Roles Methodology, Software, Visualization, Writing – original draft

Roles Data curation, Project administration, Writing – original draft

Affiliation Graduate School of Information Sciences, Tohoku University, Sendai, Japan

Roles Conceptualization, Supervision, Validation, Writing – review & editing

Roles Project administration, Validation, Visualization, Writing – review & editing

* E-mail: [email protected]

Affiliation School of Foreign Language, Huaiyin Normal University, Huai’an, China

• Shaojie Qi, 
• Fengrui Hua, 
• Shengyuan Xu, 
• Zheng Zhou, 

• Published: August 9, 2021
• https://doi.org/10.1371/journal.pone.0254988
• Reader Comments

During uncertainties associated with the COVID-19 pandemic, effectively improving people’s health literacy is more important than ever. Drawing knowledge maps of health literacy research through data mining and visualized measurement technology helps systematically present the research status and development trends in global academic circles.

This paper uses CiteSpace to carry out a metric analysis of 9,492 health literacy papers included in Web of Science through mapping knowledge domains. First, based on the production theory of scientific knowledge and the data mining of citations, the main bodies (country, institution and author) that produce health literacy knowledge as well as their mutual cooperation (collaboration network) are both clarified. Additionally, based on the quantitative framework of cocitation analysis, this paper introduces the interdisciplinary features, development trends and hot topics of the field. Finally, by using burst detection technology in the literature, it further reveals the research frontiers of health literacy.

The results of the BC measures of the global health literacy research collaboration network show that the United States, Australia and the United Kingdom are the major forces in the current international collaboration network on health literacy. There are still relatively very few transnational collaborations between Eastern and Western research institutions. Collaborations in public environmental occupational health, health care science services, nursing and health policy services have been active in the past five years. Research topics in health literacy research evolve over time, mental health has been the most active research field in recent years.

Conclusions

A systematic approach is needed to address the challenges of health literacy, and the network framework of cooperation on health literacy at regional, national and global levels should be strengthened to further promote the application of health literacy research. In the future, we anticipate that this research field will expand in two directions, namely, mental health literacy and eHealth literacy, both of which are closely linked to social development and issues. The results of this study provide references for future applied research in health literacy.

Citation: Qi S, Hua F, Xu S, Zhou Z, Liu F (2021) Trends of global health literacy research (1995–2020): Analysis of mapping knowledge domains based on citation data mining. PLoS ONE 16(8): e0254988. https://doi.org/10.1371/journal.pone.0254988

Editor: Arista Lahiri, College of Medicine and Sagore Dutta Hospital, INDIA

Received: January 28, 2021; Accepted: July 7, 2021; Published: August 9, 2021

Copyright: © 2021 Qi et al. This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are available in figshare: ( https://doi.org/10.6084/m9.figshare.15023226.v4 ).

Funding: The authors received no specific funding for this work.

Competing interests: The authors have declared that no competing interests exist.

1. Introduction

Health literacy is a concept that is constantly being developed. In recent years, governments, health professionals and researchers have paid increasing attention to it, and at the 9th Global Conference on Health Promotion in 2005, it was listed as the core topic [ 1 ]. Many countries have prepared or started to establish health literacy monitoring and evaluation systems, hoping to enhance the overall health of the population by improving people’s health literacy [ 2 ], and countries such as the United States, Canada, the United Kingdom, Australia and China have even made health literacy a national strategy. The WHO (2013) [ 3 ] defines this term as “the personal characteristics and social resources needed for individuals and communities to access, understand, appraise and use information and services to make decisions about health”. It is believed that health literacy can play an important role in encouraging people and communities to participate in health care and build their resilience, improve their health and well-being, address health inequities, etc. The WHO has pointed out that health literacy is generally low in both developed and developing countries [ 1 ]. Evidence supports that increasing age, low educational attainment, disadvantaged socioeconomic status and poor reading level are the main barriers to health literacy. In addition to other socioeconomic issues, the literature shows that people with low levels of health literacy around the world also have a misunderstanding of health information in English [ 4 ]. In contrast to developed areas that are actively exploring interventions to improve people’s health literacy, the relevant research in some developing countries is still in its infancy. Studies have shown that low-income populations primarily have low reading skills, leading to their low health literacy [ 4 , 5 ], especially in countries that are densely populated and ethnically and culturally diversified but heavily engaged in human development, economic stability and primary health care. At the same time, these countries are faced with great challenges in terms of providing health services to disadvantaged groups with low literacy rates and low socioeconomic status [ 6 ]. Thus, improving health literacy is an urgent need to facilitate and achieve the health-related United Nations Millennium Development Goals.

Generally, people with good health literacy can manage their health more effectively than those without or with poor health literacy [ 7 ]. The 2020 global pandemic of COVID-19 shows that low health literacy is a public health problem that has long been underestimated worldwide [ 8 – 10 ]. For example, nearly half of European adults mentioned that they had inadequate health literacy for taking care of their own health issues as well as those of others [ 11 ]. However, substantial evidence has proven that health literacy is an important tool for the prevention of noncommunicable diseases (NCDs), and sustainable long-term measures need to be implemented as early as possible [ 12 ]. In particular, in the current global epidemic, which is full of uncertainties, people need to quickly change their health cognition and behaviors to reduce the risk of infection and transmission of the disease. Therefore, how to improve the health literacy level of global citizens is of unprecedented importance.

Research on health literacy, as an emerging field, has grown rapidly over the past 30 years. The connotations have been enriched and specified, and the theories and methods involved have become much more extensive and complex. At the same time, research is expanding in the context of interdisciplinary approaches, with new theories and methods constantly being created, so determining field divisions might not be easy. At present, many scholars have reviewed this research topic in terms of basic concept and framework [ 7 ], measurement [ 13 ], evaluation methods [ 14 ] and impact mechanism [ 15 ]. To some degree, these studies have provided theoretical and methodological support for better understanding the involved interdisciplinary approaches to health literacy research, but the analytical perspectives have been relatively singular, and there is a lack of comprehensive and systematic studies on the development track of the discipline. Some questions still need to be answered: 1. Which main bodies (country, institution and author) promote research on health literacy, and what kinds of collaboration exists among them? 2. In the development process of health literacy research, which publications have acted as pioneers or played key roles? 3. What are the main topics in the research field, and how are the different research topics related to each other? Traditional review articles cannot comprehensively summarize and quantitatively analyze how knowledge in a certain field has developed across a large amount of literature, and their comments are usually qualitative in nature and prone to subjectivity.

As an important branch of science, bibliometrics has evolved into a relatively mature theoretical and methodological system after decades of development [ 16 ]. Based on scientific papers and citation data, it shows the features of attractiveness and objectiveness, explores the developmental characteristics and evolutionary rules of contemporary scientific research from different dimensions and perspectives, and provides an in-depth quantitative understanding of the relationships among scientists, research institutions and discoveries. Additionally, it plays a unique role in quantifying knowledge production and laws of scientific development [ 17 , 18 ]. In addition, the scientific citation database, which records all the progress that has been made, is constantly being enriched and improved, providing an objective data basis for the study of science [ 19 , 20 ]. Therefore, drawing knowledge maps of health literacy research through data mining and visualized measurement technology helps systematically present the research status and development trends in global academic circles and supplements the existing bibliometric research in this field.

On the basis of the previous unsolved questions, this study draws knowledge maps of global health literacy research by applying the existing bibliometric theory and data mining technology to outline the overall knowledge structure and development trends in the field and to better understand its research topics and evolution over time. The specific objectives of this study are as follows: 1. Construct the collaboration networks of health literacy research to understand the characteristics of research subjects, including the composition and collaboration modes of authors, the heterogeneity and collaboration modes of authors’ institutions, and the distribution and collaboration modes of authors’ countries. 2. Generate the cocitation networks of health literacy research to analyze the characteristics of key articles and citation structure. 3. Create the timeline view of health literacy research. Based on the different topics and changing trends of health literacy research, this paper reveals the research frontiers and future propositions of the field.

2. Methodology

2.1. data source and processing.

The scientific nature of mapping knowledge domains is rooted in their databases; that is, how to accurately and comprehensively retrieve all the documents on a subject is the key to data collection. Therefore, the data sources used for CiteSpace should be authoritative and massive at the same time. The Web of Science (WoS) Core Collection is selected in this paper since it has been trusted and selected as a global citation database over time [ 21 , 22 ]. First, WoS has been widely recognized as a great citation database for bibliometric research and the largest comprehensive scholarly information resource covering peer-reviewed journals with high impact factors [ 21 , 23 – 25 ]; the core WoS database consists of over 21,419 types of journals, books and conference proceedings with over 79 million records (WoS, October 2020). Chen (2016) [ 26 ] advocated that when people use scientometric analysis tools such as CiteSpace, the use of WoS as a data source can effectively prevent data loss and speed up data conversion. Our concrete retrieval strategies and data processing are as follows. First, publications with the subject term “Health Literacy” were searched in WoS, and the search was further optimized by the following conditions: language = English; document type = article + review. The number of search results was 9,888 (downloaded on September 19, 2020). Therefore, the period of the citing articles in our study is from January 1, 1995, to September 19, 2020. During the deduplication process, we excluded duplicate publications and articles with missing key fields, such as abstracts, keywords and references, resulting in 9,429 valid records for inclusion. Second, 245,234 citation datasets for document cocitation analysis were established from the 9,429 papers and divided into citing articles and cited references. Citing articles, also known as source articles, are the main parts of the data records in WoS, including information such as author, institution, country, title, keywords, category and cited references. Cited references are usually documents of high quality, including the author, title, journal, year and other information, and can be cited many times by others. The parameters of the collaboration analysis and document cocitation analysis in CiteSpace were set as shown in Table 1 .

• PPT PowerPoint slide
• PNG larger image
• TIFF original image

https://doi.org/10.1371/journal.pone.0254988.t001

2.2. Analysis

2.2.1 data mining and visualization..

Data mining is a process of exploring the potential relationship between large amounts of disordered data and determining the information that may be ignored, hidden or unknown [ 27 ]. In recent years, data mining techniques have been developed rapidly in new research fields such as social network analysis, image data mining, and structural and temporal data analysis [ 28 ]. The visualization techniques transform the data mining results from abstract results into graphics or images with the help of visualization models and form rich and meaningful infographics and mapped knowledge domains, thus building a “bridge” between data mining and knowledge discovery [ 29 , 30 ]. This paper uses the visualization software CiteSpace (CiteSpace 5.6.R4 Version) to carry out data mining and visual analysis of the scientific literature of health literacy to identify its knowledge bases, research hot spots and development trends on the basis of big data and provide references for future studies.

CiteSpace is a Java application for visualizing information in scientific literature, based mainly on cocitation analysis theory and pathfinder network scaling (PFNET). It facilitates a systematic review of a progressive knowledge domain with pivotal points and intellectual turning points as well as potential dynamic mechanisms and frontiers through a series of visualization maps [ 31 ]. Therefore, the software can effectively help readers better understand the research field in which they are engaged because it not only shows the overall situation of a certain area but also highlights specific important literature in the development process.

2.2.2 Multilevel collaboration network analysis.

Scientific collaboration refers to the research of scholars who create new scientific knowledge together, and their collaboration network represents the details of their research field. Specifically, the more frequently they collaborate, the greater depth their discipline develops [ 21 ]. In reality, scientific collaboration has multiple forms and manifestations, but this study focuses mainly on collaboration between different countries/regions, institutions or authors occurring in the same paper. Through CiteSpace, three levels of analysis of scientific collaboration networks are provided: macro co-country/region, meso coinstitution and micro coauthor. In addition, the dynamic structure of interdisciplinary collaboration in health literacy research over the past 25 years is drawn in this paper and can be used as a guide for new researchers as well as researchers seeking potential collaboration.

2.2.3. Multiperspective document cocitation analysis.

Cocitation analysis is one of the most commonly used methods in scientometrics, first proposed in 1973 by Henry Small, an American information scientist [ 33 ]. It refers to pairs of papers that are cited together in source articles. The higher the frequency of cocitation is, the closer the relationship and the more similar the academic background of the two documents. In this paper, document cocitation analysis, which means the process of mining the cocitations in a document’s structural data clusters, is mainly applied [ 34 ]. Through this process, researchers are able to better understand and quantitatively reveal the structure, relationship and evolution of science as well as the research frontiers of a discipline [ 31 ]. Additionally, this method provides a basic clustering mechanism for cocitation networks so that researchers can use it to identify different research branches and hot topics [ 21 , 35 , 36 ]. Traditional cocitation analysis generally emphasizes citing articles, which are used as the source and basis of identifying terms for clusters. This paper uses CiteSpace’s cocitation analysis method from multiple perspectives to explore the citation networks and cluster structure of health literacy research by considering both cited references and citing articles.

To realize the clustering networks, we use the log-likelihood ratio (LLR) provided by CiteSpace, which helps name the clusters by extracting similar items in titles, keywords or abstracts and calculating the similarity rates [ 37 ]. The LLR tends to reflect the uniqueness of a cluster and is more suitable for generating high-quality clusters with intraclass and interclass similarity [ 38 ]. Therefore, the LLR algorithm is mainly used in the clustering analysis section of our study. In addition, regarding the document cocitation analysis that reveals the relevant research frontiers, we adopt another key CiteSpace measure, i.e., burst detection. Burst detection can detect great changes in the amount of information being cited and regards the mutation of information as a means of measuring profound changes to determine the rise or fall of interest in certain cited references or a topic and to predict the frontier of a certain research field [ 21 , 31 , 39 , 40 ]. In particular, the continuous bursts in recent years may lead to new research trends, and the specific analytical framework of this study is shown in Fig 1 .

https://doi.org/10.1371/journal.pone.0254988.g001

3.1. Basic attributes of health literacy research

3.1.1. global publications..

The change in paper number is an important indicator to measure the development trend of a research topic over a specific period of time. It can directly present the variation in research heat, which is of great significance in analyzing and predicting the future. Generally, if the number of publications in a research field increases over the years, it often means that this field has received continuous attention from scholars. Fig 2 shows the annual distribution of 9,429 papers on health literacy collected from the WoS database since 1995. Before 2000, the number was scarce, and the annual output was less than 20. Since 2005, the papers have increased significantly, with more than 100 published in a single year, showing an exponential growth trend. By 2019, the annual output of global health literacy papers had exceeded 9,000 and continued to show a high level of growth. Due to the time delay of the database, papers published in 2020 are not fully included and had decreased in number compared with the previous year.

In the past 26 years, a total of 9,492 articles on health literacy were published in the WoS Core Collection.

https://doi.org/10.1371/journal.pone.0254988.g002

3.1.2. Main journals.

As a typical interdisciplinary field, health literacy research covers a wide range of disciplines. Therefore, the distribution of publications in this field is relatively scattered. Papers are published in more than 1,600 journals, most of which are professional and comprehensive journals in the fields of medicine and public health. Table 2 lists the top 10 journals with the largest number of articles in the field, which cover approximately 20% of all health literacy papers. Among them, Patient Education and Counseling is the most important, with 351 papers published. It is followed by Journal of General Internal Medicine , a medical journal with 262 published articles. Journal of Health Communication is third, with only two fewer articles than the second-place journal. It is worth noting that health literacy, as a new research topic, has attracted much attention from open access journals such as BMC Public Health , PLoS One and BMJ Open , which in turn have to some extent accelerated the spread of research results in this field.

https://doi.org/10.1371/journal.pone.0254988.t002

3.2. Collaboration network analysis of health literacy research

Science can be considered an activity of exploring new knowledge with intelligence and social dimensions. Group interaction is the basic mechanism for the generation of scientific knowledge, while scientific collaboration is the most direct form [ 41 ]. In this paper, the size of nodes represents the frequency of research content published by countries/regions, institutions or authors, and the thickness of the connecting lines between nodes reflects the strength of the co-occurrence relation between the main bodies. In addition, based on Freeman’s algorithms for node influence, betweenness centrality (BC) is used to discover and measure the influence of the nodes [ 42 , 43 ], and purple circles are used to mark these nodes. A node with high BC is usually a key hub linking two other nodes, also known as the turning point [ 31 ]. In addition, a color change in the diagram from a cold color (blue) to a warm color (yellow) represents the time sequence of collaboration.

3.2.1. National influence and co-country/region collaboration network.

From 1995 to 2020, 89 countries/regions participated in collaboration on health literacy research, and the top ten countries were selected according to the number of publications. There are differences in the numbers among countries. For example, as the largest producer, the United States published 4,939 papers, followed by Australia, with a total of 1,289.

Fig 3 shows the co-country/region collaboration network of health literacy research. It shows that most countries have carried out cross-border cooperation in this field. The nodes of the United States, Australia and the United Kingdom have purple outer circles, indicating that the three have high BC (BC > 0.1) and occupy an important position in global research cooperation on health literacy, along with great impact.

Each node represents a country/region, and the size of the node indicates the number of publications of the country/region. Each edge indicates a collaborative relationship between countries/regions.

https://doi.org/10.1371/journal.pone.0254988.g003

3.2.2. Core research institutions and the coinstitution collaboration network.

From 1995 to 2020, 790 institutions participated in research collaboration on health literacy. Among them, Northwestern University in the United States published 306 papers, followed by the University Melbourne, Australia, with 272 publications ( Table 3 ). There were 49 institutions with over 100 articles each. In general, the main forces in this field are concentrated in a few research institutions.

https://doi.org/10.1371/journal.pone.0254988.t003

Cooperation among authors from different institutions can form a collaboration network reflecting their relationship. Health literacy research also possesses the characteristics of collaboration among different institutions ( Fig 4 ). Specifically, the nodes of the University of North Carolina System and Northwestern University both have purple outer circles (BC > 0.1), indicating that the two are not only the main sources of publications on health literacy but also play an important role as bridges in the collaboration network.

Each node represents an institution, and the size of the node indicates the number of publications of the institution. Each edge indicates a collaborative relationship between two institutions.

https://doi.org/10.1371/journal.pone.0254988.g004

3.2.3. Core authors and the coauthor collaboration network.

From 1995 to 2020, 1,807 authors participated in research on health literacy. According to the number of publications, Table 3 lists the top 10 authors in the field, among whom Michael S. Wolf ranks first, with 167 articles published. Another fruitful author is Anthon F. Jorm, with 72 publications.

Authors and their social relations are the core elements as well as the key forces of a certain research field. Through the analysis of the coauthor collaboration network, we can determine which scholars cooperate closely in the field and further discuss the influence of team cooperation on their academic performance. From the coauthor collaboration network of health literacy research shown in Fig 5 , we can infer that Michael S. Wolf, an American scholar with the highest BC, has greatly changed the network in his field and formed a strong team. In addition, the top three authors in health literacy are all highly correlated.

Each node represents an author, and the size of the node indicates the number of publications of the author. Each edge indicates a collaborative relationship between two authors.

https://doi.org/10.1371/journal.pone.0254988.g005

3.2.4. Discipline evolution and interdisciplinary collaboration.

To better understand the expansion and evolution of health literacy research, all papers in the field of health literacy since 1995 are divided into five stages, namely, 1995–1999, 2000–2004, 2005–2009, 2010–2014 and 2015–2019. Fig 6 shows the interdisciplinary collaboration of health literacy papers during these stages and reflects their tendency to move from medicine to other disciplines. From 1995 to 1999, research on health literacy was in its infancy, with a relatively small number of publications. In addition, it was distributed mainly in the areas of medicine and health care, indicating that its disciplinary attributes during this stage were highly concentrated. Since 2000, research on health literacy has matured with the rapid increase in the number of papers and expanded research fields. Interdisciplinary collaboration has been extended and deepened, with medicine, public health and health policy as the core disciplines and psychology, education and social science as the supplements. Additionally, the close similarity between 2010–2014 and 2015–2019 in the knowledge map means that global health literacy research has formed its own discipline cluster. In the past five years, computer science in information science and social work in applied social sciences have both turned their attention to health literacy.

Each node indicates a category, and the larger the node, the more papers were published. Each edge indicates a collaborative relationship between two categories.

https://doi.org/10.1371/journal.pone.0254988.g006

3.3. Document cocitation analysis of health literacy

Citation is one of the core elements of academic work, through which scientific literature can either cite or be cited, forming an interconnected network of literature [ 44 ]. Citation analysis is a bibliometric method that takes citations as the research object and uses the methods of statistical analysis, network analysis and content analysis to examine the network pattern of scientific papers, authors, institutions, journals, etc. to reveal their quantitative characteristics and inherent laws and to study the dynamics of scientific documents. It is often used to identify and define research fields, discover and explore the knowledge community, analyze and predict research frontiers [ 34 ], etc. This study will identify the main research directions and core literature of health literacy through CiteSpace’s analytic function.

3.3.1. Cocitation networks and key references.

To further understand the characteristics of the structure of the cocitation network of health literacy, CiteSpace is used to retrieve information regarding cited references represented by network nodes. The visualized network reveals the overall structure of health literacy research in a broader context ( Fig 7 ). The network is composed of three differently colored regions: the right half of the network is basically blue, which indicates that the citation relationship of this part occurred mainly between 1995 and 2000; the middle part has blue and yellow links, most of which are from 2001 to 2010; and the network links in the left half of the network are mostly yellow, which means that these cocitations occurred most recently, between 2011 and 2020.

A total of 12 clusters were generated in the graph. Each node represents one cited reference, and each edge indicates the cocitation relationship. The color represents the date of publication: yellow indicates literature that is newly published, and green and blue indicate literature published in earlier years.

https://doi.org/10.1371/journal.pone.0254988.g007

Additionally, to better understand the network structure and its content, it is important to identify special nodes and links. Special nodes take important positions in the knowledge network and play a specific role in the evolution of the knowledge structure; they can be identified flexibly through cocitation frequency, BC and half-life in CiteSpace. Highly cocited literature plays a fundamental role in the discipline ( Table 4 ). A study with high centrality is an important turning point and milestone in the development of this research field ( Table 5 ). The cited references represent the research basis, and the longer the half-life is, the more classic a study can be [ 31 ].

https://doi.org/10.1371/journal.pone.0254988.t004

https://doi.org/10.1371/journal.pone.0254988.t005

First, Berkman’s Low Health Literacy and Health Outcomes : An Updated Systematic Review (n = 888), published in 2011, has been cocited most frequently. This paper also has high BC (BC = 0.14), showing that it not only has been cited many times but is also an important node connecting multiple studies. Based on a retrospective analysis of 111 English studies, this paper confirms that there is a correlation between low health literacy and poor health outcomes and poor use of health care services. Second, the study with the highest BC (BC = 0.19) is Adherence to combination antiretroviral therapies in HIV patients of low health literacy , published by Kalichman in 1999. The results of the intervention with HIV-seropositive patients showed that health literacy is an important independent predictor of treatment compliance. Generally, there was a burst of publication of these important node studies in 2005–2015.

3.3.2. Analysis of cocitation clusters.

According to the output results, the Q value of the map is 0.6239, greater than 0.3, meaning that the module structure of clustering is significant, and the S value is 0.85, greater than 0.7, meaning that the clustering result is reasonable [ 31 ]. In the cluster, the lowest S value is 0.577 (#0), and the highest is 0.996 (#7), meaning that the homogeneity is relatively high [ 31 ].

To provide a more intuitive understanding of the evolution of research hot spots over time, a timeline view of the clustering results ( Fig 8 ) sets each node in the position corresponding to the cluster to which the node belongs (axis of ordinates) and the publication time (abscissa axis) of the node. The nodes of the same cluster are arranged on the same level line in time order, displaying the historical results of the cluster. Compared with the relationship within the cluster, the timeline view pays more attention to the relationship among the clusters. In the timeline view, all 12 clusters are arranged from #0 to #11 in ascending order, with their color matching the years in which the clusters were active. Large nodes usually indicate a high level of referencing, a citation burst, or both.

This figure presents 12 clusters, which are arranged and numbered in ascending order from #0 to #11, with the colors corresponding to the average year in which the clusters were active. The larger the node, the more times it was cocited. Each cluster represents an area that has been developed or is developing Hence, the closer the arch body is to the right, the newer the topic.

https://doi.org/10.1371/journal.pone.0254988.g008

The automatic clustering function and the clustering tags automatically extracted from the titles of citing articles based on the log-likelihood ratio vb algorithm can help us understand the content of the health literacy cocitation network. Each cluster represents a field that has been developed or is developing. The closer the para-curve is to the right, the newer the topic is. The graph also shows that the durations of different clusters are not consistent. Some topics, such as socioeconomic factors (#2) and realm (#1), have left footprints in the history of promoting research on and the development of health literacy, while mental health literacy (#7) and patient education (#0) are clusters that have been active in recent years. In addition, there are emerging themes such as electronic health (eHealth) (#9).

4. Discussion

4.1. evolution of collaboration networks in health literacy research.

In recent years, an increasing number of scholars have become aware of the importance of scientific collaboration, which has become an important way to promote the development of science and technology, economy and society as a means of scientific knowledge production. This study provides a panoramic view of the collaboration network and academic influence of health literacy research through four types of collaboration network analysis provided by CiteSpace: countries/regions, institutions, authors and interdisciplines. First, from the perspective of geographical distribution and the number of papers, the distribution of health literacy research papers is highly unbalanced. From a global perspective, the United States and Australia have the largest number of health literacy papers. The two countries have attached great importance to this topic, resulting in an increasing number of papers on a yearly basis. In addition, Canada, the Netherlands and other developed countries have contributed a great number of papers. Some Asian countries, such as China and Japan, have also stepped up their efforts in health literacy research, and the number of papers from these two countries is in the top 10 globally.

Second, the results of the BC measures of the global health literacy research collaboration network show that the United States, Australia and the United Kingdom are the major forces in the current international collaboration network on health literacy, occupying important positions in the global collaboration on health literacy research and playing important roles in linking cross-border collaboration on health literacy research. BC measures of research institutions show that most of the collaboration networks are based in European and American countries. Northwestern University in the United States is the most productive and influential research institution in the field of health literacy, with a great number of publications and high BC. Additionally, the University of Melbourne and University of Sydney in Australia are also key forces in international health literacy collaboration, with BC measures and number of papers ranking in the top five, indicating that they play an important linking role in promoting collaboration among countries. It is worth noting that although there is close collaboration among universities in the same country/region, there are still relatively very few transnational collaborations between Eastern and Western research institutions. From the perspective of collaboration among authors, Michael S. Wolf is the most influential author in the field of health literacy research; he has established a strong collaboration team, which has had a great impact on the overall structure of the collaboration network of authors worldwide.

Although the importance of geographical distance in collaboration has been weakened by the development of information and communication technologies, the distribution of collaborative relationships in health literacy research remains closely linked to geographical locations. Within the international network of health literacy, North America, with the United States at its core, Asian countries, with Japan and China as their main forces, and European countries are just like the three legs of a tripod. Although collaboration between transnational institutions has become more common, collaboration within national institutions remains the dominant trend, and the geographically closer the institutions are located, the more prominent the collaboration is. Additionally, collaboration among authors also exhibits some geographical proximity, with cooperation generally arising between researchers at the same institution. Thus, the collaboration networks of country/region, institution, and author present some variability.

Finally, the analysis of interdisciplinary collaboration in health literacy research shows that interdisciplinary collaboration in the field of health literacy was very rare at an earlier stage but began to flourish recently, involving more fields. Collaborations in public environmental occupational health, health care science services, nursing, general internal medicine and health policy services have been active in the past five years. This shows that research on health literacy has shifted from mere macrolevel research to microlevel research, from an independent discipline to interdisciplinary research. Thus, interdisciplinary collaboration should be a trend of future health literacy research.

4.2. Hot spots in health literacy research

This paper identifies the hot topics in the field of health literacy by integrating the classification of representative studies and cocitation clusters. To simplify the analysis, clusters are divided into active clusters and silent clusters, and information about certain important clusters is selected for discussion. Active and silent clusters are relative concepts, with the former meaning that the research has entered a new stage, representing emerging topics of research, and the cited references constitute the knowledge base of the research, while the key nodes are the literature that needs to be highlighted. Active clusters are restricted to clusters that have witnessed citation bursts in the most recent decade, while other clusters are silent, with burst that occurred before 2010 ( Table 6 ).

https://doi.org/10.1371/journal.pone.0254988.t006

First, from the change in time order, it can be seen that the socioeconomic factors (#2) and AIDS (#11) clusters are the two themes that take form earliest and thus can be considered the knowledge base of health literacy. The socioeconomic factors cluster (#2) has 154 items, spanning from 1988 to 2006, 18 consecutive years. The S value is 0.821, showing high homogeneity. According to the timeline view, although the scale of the cluster is relatively large, it is no longer active. The starting point of the timeline is Emergency Department Patient Literacy and the Readability of Patient-Directed Materials , published by Powers R. D. in 1988. According to the annual average value, cluster #2 occurred mainly in the 1990s, which is in the early stage of health literacy research and development. The cluster is closely related to the network structure of several emerging clusters, establishing an important knowledge base for health literacy research. Cluster #11 contains 9 members, which emerged mainly in 1997. The relationship of the network links shows that this cluster is independent of all other clusters. The references in cluster #11 are all from a citing article– Barriers to HIV/AIDS Treatment and Treatment Adherence among African-American Adults with Disadvantaged Education , published by S. C. Kalichman in 1995. Through an intervention study of African American adult AIDS patients, this paper shows that education and health literacy are important factors in adhering to HIV treatment and gaining access to medical services.

Clusters #1, #4, #5, #6 and #8 are in the central area of the horizontal axis of the timeline view and have a close relationship with cluster #2, indicating a major stage of the development of health literacy research. Cluster #1 spans from 1997 to 2010, 13 consecutive years, with 165 items and an S value of 0.58, indicating low homogeneity and uncertainty in the clustering tags. Although clusters #4, #5, #6 and #8 are relatively independent of other clusters, they represent objects and characteristics of different subsets of health literacy research. Different intermediary and explanatory mechanisms are adopted to prove that health literacy plays an important role in the treatment of depressive moods [ 47 ], establishing a good relationship between doctors and patients [ 48 ], and rehabilitation of cancer patients [ 49 ].

In the third stage of health literacy research (2005–2020), patient education (#0) is the largest cluster, containing 167 documents. Cluster #0 spans from 2004 to 2018, showing that it is a topic that has enjoyed enduring popularity. The S value of the cluster is 0.541, which is relatively low among all the clusters, indicating that there is a tendency for this cluster to generate new topics. In addition, the articles with the highest cocitations are all from cluster #0. Another active and large topic that spans a long time is mental health literacy. Two clusters, #3 and #7, formed as a result of automatic clustering, which shows two different tendencies of citations on mental health literacy. References in cluster #3 are cited mainly by papers on assessment methods and practices of mental health literacy [ 50 , 51 ], while the citation relationship of cluster #7 shows that the mental health literacy of adolescents is an important research direction [ 52 , 53 ].

The eHealth cluster (#9) is an active cluster that began to emerge in 2011 and has been an emerging theme in the most recent decade. There are 24 items in the cluster, and its S value is 0.965, showing that the internal homogeneity is very high. With the rapid development of information technology, information and communication technology have had a great impact on all aspects of social life, including medical and health care. Therefore, how to improve the capabilities of the public to make full use of eHealth resources in an information-centered environment has gradually attracted the attention of researchers, and eHealth literacy has become a new field of health literacy research that has developed rapidly in recent years. In 2005, the World Health Organization defined eHealth as “the dissemination of health resources and health care information through electronic means, so that health professionals and users can disseminate and access health information” [ 54 ]. Based on the concept of eHealth, Norman and Skinner (2006) [ 55 ] first proposed the concept of eHealth literacy, which refers to the ability to search for, identify, understand and evaluate health information from electronic resources and process and apply the acquired information to solve health problems. In recent years, research on electronic health literacy has gradually been extended to a variety of different subgroups, such as teenagers [ 56 ], the elderly [ 57 ], medical professionals [ 58 ], and HIV-positive patients [ 59 ].

4.3. Trends and frontiers in health literacy research

Through a cocitation network and clustering analysis of the literature, this part defines the development trends of health literacy research over time. First, health literacy research has focused on conceptual frameworks and operational methods since 1995. At the second stage, between 2000 and 2004, the focus shifted to the comprehensive discussion of social factors of health literacy, making great contributions to research design, intermediary mechanisms and explanatory mechanisms. At the third stage, from 2005 to 2020, empirical research and theory on health literacy began to diversify gradually. On the basis of an enriched research design, scholars have applied more theoretical frameworks to explain health literacy.

This paper uses the burst detection technology provided by CiteSpace to reveal the frontier literature in the field of health literacy. The concept of the research frontier was first introduced by Price in 1965. A cocitation burst can detect emerging trends and sudden shifts of attention in the scientific development of disciplines. The algorithm behind burst detection is derived mainly from Kleinberg’s (2002) [ 60 ] emergent measurement algorithm, using numerical values to express the strength of the burst. The greater the value is, the greater the strength of the burst and the more obvious the development trend of the topic related to the burst. The large amount of citation data has led to a large number of emergent studies. We selected references that experienced great breakthroughs in the past five years, which represent the latest research frontiers of health literacy. Table 7 shows the top ten references in terms of strength of burst in the most recent five years, including empirical studies by Sorensen K. (2015) [ 61 ], Batterham R. W. (2016) [ 62 ], and Kutcher S. (2016) [ 51 ]. We predict that these citation bursts will continue to receive attention in the future because they are usually predictors of future research trends.

https://doi.org/10.1371/journal.pone.0254988.t007

5. Conclusions

This paper analyzes the network structure and trend of themes in health literacy research from 1995 to 2020 by using a multilayer collaboration network and multidimensional cocitation analysis. CiteSpace knowledge mapping analysis is panoramic. It can not only display the historical value of health literacy research but also show the main trends and future hot topics of health literacy research in recent years. The time distribution table of the literature shows that research on health literacy has been gradually expanding since the concept was first proposed. The main sources of health literacy literature are periodicals on medicine, public health and health-related fields, represented by Patient Education and Counseling , Journal of General Internal Medicine and Journal of Health Communication . The research on health literacy in the United States is way ahead of that in other countries, and its published literature accounts for more than 67% of the total literature. Moreover, interdisciplinary network analysis based on the map of scientific knowledge shows that research on health literacy is a process that shifts its focus from independent disciplinary research to interdisciplinary research. Finally, the key nodes in the literature and cluster classification reveal that research topics in health literacy research evolve over time: proposing and popularizing the concept of health literacy, introducing and improving evaluation tools, growing concerns for groups with low health literacy, generating a branch–mental health literacy, and putting forward emerging concepts such as eHealth literacy. In addition, mental health literacy, a relatively independent field in research and practice, has been the most active research field in recent years indicating that research in this field has matured over time. Sorensen K. (2015) [ 61 ], Batterham R. W. (2016) [ 62 ] and Kutcher S. (2016) [ 51 ] have provided frontier literature in the field of health literacy, which will be the main development trend of health literacy research and will also be the hot topics that will be the center of attention in the future.

Currently, knowledge about health literacy comes mainly through collaboration within a particular country or region, especially within an institution featured by a small group of 2–5 people. Moreover, the overall connectivity of co-authorship networks remains low, with subnetworks dominated by the dual-core and bridge pattern; international collaboration is mainly bilateral and trilateral, and the greatest level of international collaboration is found in developed areas such as Europe and the U.S. On a deeper level and in the long run, a systematic approach is needed to address health literacy issues, together with a strengthened framework of collaborative networks for health literacy at all regional, national and global levels. At the same time, more collaboration between research groups in different countries and institutions is encouraged to further drive the application of health literacy research forward. In the future, we anticipate that this research field will expand in two directions, namely, mental health literacy and eHealth literacy, both of which are closely linked to social development and issues.

First, mental health problems have become a key public concern, and one of the major causes of this phenomenon is the generally low mental health literacy of the population [ 63 , 64 ]. An increase in mental health literacy can provide an optimized pathway for early interventions to avoid worsening psychological problems [ 65 ]. Therefore, research in this field deserves more attention from the academic and social sectors. Second, with the rapid development of online consultations, eHealth literacy has become an important competency indicator that directly influences how netizens access health information, use it and make relevant health choices. We should actively promote the measurement of eHealth literacy and conduct surveys among different population groups [ 66 ]. We should also use different interventions to improve their access as well as their ability to screen health information. In summary, this study can significantly complement traditional literature reviews and provide useful information for future directions and perspectives in health literacy research.

There are some deficiencies in this study. First, the sources of data are limited. Only six online databases are used, which may not cover all studies on health literacy. In addition, non-English publications are not included in our analysis, so publications in other languages should be considered in future research. Second, the strategy of selecting “health literacy” as the only search term can be improved. Specifically, future research should search for more keywords in a more flexible manner to find different research databases. Third, because the citation analysis focuses on the representative literature that has been cited, it is impossible to objectively evaluate newly published high-quality papers. Fourth, the technical analysis of our study relies on bibliographic records without considering the differences between the results of bibliometric analysis and the practical research conditions as well as the rich stories behind the relations of literature.

Additionally, the intrinsic defects of CiteSpace software will lead to inevitable errors in the process of data processing and transformation, such as unintegrated synonyms or ununified literature types. The data used by CiteSpace for analysis are quantitative, so in many cases, it is difficult to interpret the clusters directly from a qualitative perspective, whether it is a delineated collaborative network, an automatically generated cluster or a group of clustering labels obtained from the cocitations based on the TF*IDF weighting algorithm. This is surely a common problem faced by scholars when they adopt CiteSpace for visual analysis [ 21 , 67 – 70 ]. Therefore, these descriptive labels usually require further manual classification, refinement and meaning.

• 1. WHO. Policy brief 4: health literacy. The 9th global conference on health promotion. Shanghai, China. 2017.
• 2. Roundtable on Health Literacy, Board on Population Health and Public Health Practice, Institute of Medicine. Appendix a, health literacy around the world: Part 1 health literacy efforts outside of the United States. health literacy: improving health, health systems, and health policy around the world: workshop summary. Washington (DC): National Academies Press (US). 2013.
• 3. WHO. Health literacy : the solid facts . Geneva: World Health Organization. 2013.
• View Article
• PubMed/NCBI
• Google Scholar
• 12. Kosir U, Sorensen K. COVID-19: the key to flattening the curve is health literacy. Perspectives in Public Health. 2020; Article Number: UNSP 1757913920936717.
• 19. Hicks D, Melkers J. Bibliometrics as a tool for research evaluation. Handbook on the Theory and Practice of Program Evaluation . 2013; 323–349. https://doi.org/10.4337/9780857932402.00019
• 20. Tijssen RJW. Measuring and Evaluating Science-Technology Connections and Interactions. In Moed H.F., Glanzel W, & Schmoch U (Eds.), Handbook of Quantitative Science and Technology Research : The Use of Publication and Patent Statistics in Studies of S & T Systems (pp. 695–715). Dordrecht: Springer Netherlands. 2005;.
• 24. Volet S. Emerging trends in recent research on motivation in learning contexts. Emerging trends in recent research on motivation in learning contexts. In: Volet S, and Järvelä S, (eds.) Motivation in learning contexts: Theoretical advances and methodological implications. Pergamon : An Imprint of Elsevier Science . Oxford, UK, 2001; pp. 319–334.
• 42. Alexander J, Bache K, Chase J, Freyman C, Roessner JD, Smyth P. An Exploratory Study of Interdisciplinarity and Breakthrough Ideas. In 2013 Proceedings of PICMET. Technology Management for Emerging Technologies (pp. 2130–2140).
• 43. Freeman LC. Encyclopedia of Social Network Analysis and Mining . New York: Springer. 2014;.
• 46. Rousseeuw PJ, Kaufman L. Finding Groups in Data. Series in Probability & Mathematical Statistics. 1990; Print ISBN:9780471878766.
• 54. Kickbusch I. Health literacy : the solid facts . Geneva: WHO Regional Office for Europe. 2013.
• 60. Kleinberg J. Bursty and hierarchical structure in streams. Paper presented at the Proceedings of the 8th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Edmonton, Alberta, Canada. 2002.

SYSTEMATIC REVIEW article

Health literacy among university students: a systematic review of cross-sectional studies.

• 1 Institute of Sports and Sports Science, Karlsruhe Institute of Technology, Karlsruhe, Germany
• 2 Central Scientific Institution for Key Competencies, Karlsruhe Institute of Technology, Karlsruhe, Germany

Objective: The aim of this systematic review was to provide an overview of cross-sectional studies that examined health literacy among university students and to identify possible determinants related to health literacy.

Method: The current review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). Three databases (PubMed, Scopus, and Web of Science) were systematically searched for cross-sectional studies that examined health literacy among university students. Results of included studies were narratively summarized.

Results: The systematic review includes twenty-one research studies. The majority of studies report health literacy scores among university students that are lower compared to reference samples. The health literacy of students is influenced by different variables (age, gender, number of semesters, course of studies/curriculum, parental education, and socioeconomic background).

Discussion: Health literacy activities should target all students. Universities should make use of their resources and offer health literacy courses for students in which content is used from disciplines available at the university (e.g., medicine, health, or psychology). To increase effectiveness, health literacy courses should be adapted according to the different needs and characteristics of the student subgroups.

Introduction

University students worldwide experience a high level of psychological stress that exceeds the level of non-students and physiological and psychological health problems ( 1 , 2 ). The reasons for this are academic responsibilities, financial worries, and adaptation to new life circumstances. These conditions can harm the health of the students ( 2 , 3 ). To counteract this, the Okanagan Charter for health-promoting universities and colleges ( 4 ) was created. Educational institutions that follow the idea of the charter, create campus cultures of wellbeing, equity, social justice, and improve the health of the people who live, learn, and work there. Furthermore, they also strengthen the ecological, social, and economic sustainability of their communities and the society as a whole, considering the responsibility students will later bear in their given environment.

It is important to stress that if people have to achieve their full health potential, they must also take control of its determinants ( 5 ). Health promotion is therefore defined by the Ottawa Charter ( 6 ) as a process that enables people to better control and promote their health on their own. This idea of empowerment can among other things be accomplished through the improvement of health literacy. The approach of promoting health literacy is indeed deeply rooted in health promotion per se : to empower people in a setting to make better decisions about their health and lives in general. A review showed that low health literacy is associated with poorer ability to understand and follow medical advice, poorer health outcomes, and differential use of some healthcare services ( 7 ). Educational institutions, such as universities, have the opportunity to optimize the health literacy of their students and empower them to make informed decisions for themselves and their environment ( 8 ).

According to Nutbeam ( 9 ), health literacy can be divided into three levels: functional, interactive, and critical health literacy. All three levels together comprise complex skills that enable an individual to extract, evaluate, and apply health-related information. Since the WHO introduced the concept of health literacy internationally in the glossary of health promotion ( 10 ), more and more definitions have been developed. Parker ( 11 ) defines health literacy as a relational concept that, while dependent on individual skills and abilities of a person, is also determined by the demands and complexity of health information and tasks. The most commonly used definitions of health literacy have been compiled by Sørensen et al. ( 12 ). In summary, all definitions address the importance of cognitive skills and competent skills that enable obtaining, understanding, and using health information.

There are a variety of reviews on health literacy in diverse populations and professional groups, such as men ( 13 ), older adults ( 14 ), immigrants ( 15 ), and librarians ( 16 ). The aim of this systematic review was to provide an overview of cross-sectional studies that examined health literacy among university students and to identify possible determinants. Additionally, we aimed to find out which theoretical frameworks and which different scales were used. Accordingly, the purpose of this review is 2-fold. First, we want to assess the state of research in this field and, second, we intend to identify starting points for decision-makers and health promoters at universities implementing health literacy interventions and adapting them to the needs of the target group.

With the specific target group of students, digital media should be highlighted as an especially relevant source of information, such as health information ( 17 ). However, skills required to collect information via the internet differ from those required to collect information from print media, e.g., books ( 18 ). Therefore, the definition of eHealth literacy will also be taken into account for this systematic review. It combines health literacy with media and computer-related skills ( 19 ).

For the purpose of this systematic review, we followed the guidelines described in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement ( 20 ). A review protocol has been prepared and can be requested from the authors. The study characteristics used to decide whether a study was eligible for inclusion in the review can be found below: cross-sectional studies (study design) examining the health literacy (outcome) of students in tertiary education of any age (population) and published since the publication of the Okanagan-Charter in 2015 were included in the review. No health status restrictions were imposed. The outcome variables of interest are health literacy and related influencing factors. The health literacy definition of Nutbeam ( 9 , 21 ) and common health literacy definitions ( 12 ) were used as a guiding principle in that respect. Regarding eHealth literacy, the definition of Norman and Skinner ( 19 ) served as a decisive criterion. In the studies, the outcome variables had to be given either as primary or secondary outcome variables. Studies were identified by searching three electronic databases (PubMed, Scopus, and Web of Science). The last search was run on February 19, 2020. Additionally, at the end of the search process, the already qualified studies were checked for additional relevant references. Combinations of the following keywords were used to search the databases: university; college; students; adolescents; health literacy; eHealth literacy. The search term was based on the review of Chesser et al. ( 22 ), which has a comparable research question but with regard to a different target population. Studies published in English and German were considered for this review. The complete search query can be found in the Appendix (see “Search term” ). The selection process (title, abstract, and full text) of the studies was conducted by two authors.

A data extraction sheet based on the patient/population, intervention, comparison and outcomes (PICOS) model was used to extract the desired data. Data items were [1] study-relevant information consisting of the name of the study, corresponding authors, the year of publication, and the country, [2] characteristics of participants (e.g., age, gender, study program, and course of studies), the underlying setting (university, college), [3] information on the outcome variables consisting of the theoretical background and the assessment instruments used, and [4] information on the results of the study regarding the health literacy of students and its determinants. The data extraction was always performed independently by at least two authors. Any discrepancies between the authors were resolved through discussion until consensus was reached.

The Appraisal Tool for Cross-Sectional Studies (AXIS) was used to assess the risk of bias of the included studies ( 23 ). Two authors independently assessed the quality of the studies. In case of disagreement, another author was consulted, and discussions were held until a consensus was reached. A scoring method was adapted to quantify the risk of bias in individual studies ( 24 , 25 ). According to this method, studies were categorized as very low risk of bias if they scored at least 19 of 20 questions correctly, as low risk of bias if they scored 17 or 18 out of 20 of the questions of the tool; as the moderate risk of bias if they scored 15 or 16 out of 20, and as high risk of bias if studies scored 14 or fewer points.

The narrative synthesis was based on data synthesis guidelines ( 26 ). First, a preliminary synthesis was developed, including initial descriptions of the results of the studies used, grouping the studies according to the PICOS scheme, preparing data and putting them into a common descriptive format, and identifying patterns along with the studies. Subsequently, relationships of the data within and between the studies were investigated. Overall health literacy, various factors that could contribute to health literacy and limitations and practical implications were identified. Also, plausible explanations were developed for the differences found within (characteristics) and between (results) the studies.

The search in the databases PubMed, Scopus, and Web of Science resulted in a total of 7,529 hits with the selected search terms. Out of those, 7,139 studies were excluded due to an inappropriate title, indicating an obviously different topic. Another 314 studies were excluded after the abstract review because they did not meet the necessary inclusion criteria. Thirteen further studies were removed after testing for duplicates. The full texts of the remaining 63 studies were then reviewed in detail. Forty-four of these did not meet the specified inclusion criteria. The remaining 19 studies were deemed suitable for inclusion in the review. In addition, further two studies could be identified by searching the references of these studies. Thus, a total of 21 studies were finally included in the review. Figure 1 presents a flow diagram summarizing the selection process.

Figure 1 . Flow diagram.

Seventeen studies were published in English and four in German. Studies had been conducted in Taiwan, Jordan, Denmark, the United States of America, Laos, Germany, Iran, Nepal, Portugal, Australia, Singapore, Lithuania, China, and Turkey. The selected studies were published in the period from 2015 to 2019. The included studies involved 13,772 students in higher education settings with the smallest sample size of 37 students and the biggest sample size of 2,892 students. The mean age of the students ranged from 20.1 to 24.1 years for the studies where data were available. Regarding student groups, twelve studies included students from various study programs, seven studies included students from various health-related study programs, and two studies included only one specific health-related program. Of the included studies, 17 were conducted in universities and two in colleges. Two studies provided no information about the setting. Theoretical frameworks for health literacy were the definition of the WHO ( 10 ), Nutbeam ( 21 ), Sørensen et al. ( 12 ), Baker ( 27 ), Kickbusch and Maag ( 28 ), Kickbusch, Maag, and Wait ( 29 ), Paasche-Orlow and Wolf ( 30 ), and Zarcadoolas, Pleasant, and Greer ( 31 ). Various scales were used to assess health literacy: The Turkey Health Literacy Scale ( 32 ), the Perception of Health Scale ( 33 ), the Health Literacy Questionnaire ( 34 ), the Danish version of the Health Literacy Questionnaire ( 35 ), concepts of Wieland and Hammes ( 36 ), Bässler ( 37 ), and Woll ( 38 ), the Iranian Health literacy questionnaire ( 39 ), the questionnaire of health-promoting lifestyle profile II ( 40 ), short version of the Test of Functional Health Literacy in Adults (S-TOFHLA) ( 41 ), the European Health Literacy Survey Questionnaire (HLS-EU)-Q16 ( 42 ), the HLS-Asia questionnaire ( 43 ), the HLS-EU-Portugal (PT) ( 44 ), The eHealth Literacy Scale (eHEALS) ( 45 ), the Taiwanese eHealth literacy scale ( 46 ), the dietary behaviors scale ( 47 ), and several self-made scales. With the exception of the performance-based S-TOFHLA ( 41 ), and a performance-based interview used by Kushalnagar et al. ( 48 ), these are all so-called self-reported health literacy instruments, i.e., instruments in which subjects are asked to self-assess their abilities ( 49 ). The survey instruments are largely based on rather broader definitions of health literacy and thus go beyond the functional aspect of it. The WHO definition is used as the theoretical basis in the Health Literacy Questionnaire (HLQ). The definition and model of Norman and Skinner ( 19 ) form the basis for eHEALS ( 45 ). Several different survey instruments are supported by the theoretical model of Sørensen et al. ( 12 ). The study by Kushalnagar et al. ( 48 ) also used its own survey instruments on the theoretical basis of Baker ( 27 ) and Nutbeam ( 21 ). Göring and Rudolph ( 50 ) assessed health literacy using a survey instrument based on the theory of Wieland and Hammes ( 36 ). The conceptual framework of the survey instrument used by Kaboudi et al. ( 51 ) was based on the theoretical considerations of Ratzan et al. ( 52 ).

In the study of Birimoglu and Cagalar ( 53 ), the health literacy of nursing students was insufficient compared to the data of other studies. Furthermore, working parents were associated with higher health literacy levels. Most students in the study by Budhathoki et al. ( 54 ) had only moderate health literacy and few individuals reported high health literacy according to their mean scores on the HLQ ( 34 ) scales. Thereby, higher age, being enrolled in a health-related course of study, higher educational level of parents, and male sex were associated with higher levels of health literacy. Elsborg et al. ( 55 ) showed that the health literacy scores of students were higher than the scores of the Danish population. Here, a higher study semester, female sex, being enrolled in a health-related course of studies, a higher educational level of the parents, and health-related experiences had a positive correlation with health literacy. The results of Göring and Rudolph ( 50 ) indicate that higher sports activity and male sex correlate positively with higher health literacy. Moreover, a finding of the study is that the mean health literacy values of common students are below the values of vocational school students. Kaboudi et al. ( 51 ) stated that in their study the mean and SD of the total health literacy of students were 4.04 ± 0.43 out of a score of five on the Iranian Health Literacy Questionnaire ( 39 ), indicating good health literacy. They found that healthy behavior is positively correlated with high health literacy. Due to their specific sample and measurement tools, Kushalnagar et al. ( 48 ) made no statement regarding the overall health literacy scores of deaf college students. The data showed a strong relationship between greater frequency of health-related discussions with friends and an accessible language during childhood and higher critical health literacy scores.

The results of Mullan et al. ( 56 ) suggest that different student groups have different health literacy profiles due to medical students demonstrating higher health literacy than students from other health-related courses of studies. Nevertheless, the authors conclude that students who are enrolled in a health-related course of studies, particularly nursing students, have gaps regarding their health literacy based on low to medium mean scores for the different HLQ ( 34 ) scales. Rababah et al. ( 57 ) also found limitations of health literacy among college students comparing the collected mean scores of the HLQ ( 34 ) to levels reported in the study of the measurement tool. Apart from the negative impact of smoking, health literacy was positively associated with higher age, higher study semester, female sex, and enrollment in a health-related course of studies. Compared with other population groups in Germany, there are more students with problematic health literacy according to Reick and Hering ( 58 ). Ninety-three percent of students in a study by Runk et al. ( 59 ) were found to have less than sufficient health literacy based on a reference index. According to the authors, accessible health services in the population and social understanding of health and disease and media distribution positively correlate with high health literacy levels. Santos et al. ( 60 ) made no statement regarding overall health literacy due to their specific research question, but found the internet as a poor source for information gathering among students. Compared to the adult population of North-Rhine-Westphalia and the German general population, students surveyed by Schricker et al. ( 61 ) have shown lower health literacy levels. While a higher subjective social status was positively correlated with the score, unfavorable financial situation and limited social support were negatively associated with health literacy by the authors. More than half of the students in the study by Schultes ( 62 ) have a high level of health literacy but are below the average in a European country comparison. Health-promoting behaviors of subjective health assessment and daily fruit and vegetable consumption were associated with better health literacy levels. The health literacy levels of the students in the study by Sukys et al. ( 63 ) were either lower, similar, or higher depending on international reference studies. A positive correlation with health literacy was found with the female sex and with enrollment in health-related courses. Suri et al. ( 64 ) did not make a statement regarding general health literacy in their study. Their work focused on the influence of the type of information gathering (traditional sources vs. internet) on health literacy and underlines that different domain-specific health literacy skills for different health sources are needed. According to Vamos et al. ( 65 ), there is a gap in health literacy among the sample groups based on the mean scores for the different HLQ ( 34 ) domains. In their data, older age, female sex, higher parental education, and higher socioeconomic status are associated with higher health literacy levels.

The general student population in the study by Zhang et al. ( 66 ) achieved a mean score of 131.89 ± 18.84 to the overall score of 197.00 in the HLQ ( 34 ). In addition, the data indicate that the health literacy levels of the medical students are insufficient. According to the authors, higher study semester, course of studies (engineering), higher educational level of the parents, and higher socioeconomic status are positively correlated with health literacy, while depression and anxiety disorders are negatively correlated. Zou et al. ( 67 ) described in their study that the health literacy level of the student group examined is suboptimal compared to other studies. Thereby, a higher study semester, a higher educational level of the parents, and a higher socio-economic status were associated with better health literacy levels. Yang et al. ( 68 ) made no statement regarding overall eHealth literacy but found that a medical course of study resulted in higher levels. Regarding critical eHealth literacy, a positive, health-promoting behavior was positively correlated. In the study by Luo et al. ( 69 ), eHealth literacy levels of students were medium to high due to the collected mean scores of 3.66 ± 0.70 for functional eHealth literacy, 3.67 ± 0.67 for interactive eHealth literacy, and 3.65 ± 0.69 for critical eHealth literacy each with a maximum score of five with eHEALS ( 45 ). Positive correlation for functional eHealth literacy was found with high frequency in the use of medical services, for interactive eHealth literacy with the selection of suitable types and locations and low intervals of health services utilization and for critical eHealth literacy with the selection of suitable types, locations, and purpose aspects of health services utilization. Medium-to-high levels of eHealth literacy for the student sample were described in the study by Yang et al. ( 70 ) indicated through the mean scores of functional eHealth literacy with 3.56 ± 0.77, interactive eHealth literacy with 3.57 ± 0.71, and critical eHealth literacy with 3.59 ± 0.72 out of a maximum score of five with the eHEALS measurement tool ( 45 ). Additionally, functional eHealth literacy was negatively related to unhealthy food intake, interactive eHealth literacy was positively related to a balanced diet, and critical eHealth literacy was positively related to regular eating habits. Also, interactive eHealth literacy and critical eHealth literacy were positively correlated with positive attitudes and decisions about food purchasing. Table 1 presents the results regarding the general levels of health literacy and possible determinants of these.

Table 1 . Results of individual studies.

To compile and interpret the results of the studies in a meaningful way, it is important to consider differences and similarities, especially in terms of the methods used. As these are exclusively cross-sectional studies, all studies are relatively homogeneous regarding study design. The greatest differences can be found in the selected samples (several health-related courses of study vs. one specifically health-related course of study vs. various courses of study and number of semesters) and the measuring instrument used. The results of the examined studies show a relatively homogeneous picture regarding their data on the health literacy of students. Eleven studies ( 50 , 53 , 54 , 56 – 59 , 61 , 65 – 67 ) report poor values or limited health literacy among students. A total of 8,089 students were involved in these studies. Regarding the study course, there is an even distribution between explicitly health-related and various study programs. Five studies include several health-related and five studies include all study programs. Only one study focuses on undergraduate nursing students solely.

For five studies, information on the number of semesters was available. Two studies explicitly included all semesters and three focused on students at the beginning of their study careers. These distributions about the course of study and the number of semesters must be taken into account when considering the results. The measuring instruments used in the studies are all assessed as valid and reliable, except for Göring and Rudolph ( 50 ), who used a self-made measuring instrument. The measurement instruments used were considered valid and reliable if they were sophisticated health literacy measurement instruments (e.g., HLQ) that had been previously tested, piloted, and repeatedly published.

The statements made on the health literacy of students are justified in each study due to comparisons with other populations. In fact, only two studies ( 51 , 55 ) report higher health literacy scores among students than among the national population. A total of 796 students were surveyed in the two studies with reliable and valid HLQ. It should be noted that these are exclusively health-related programs and therefore their results should be interpreted accordingly. The results of one of the studies were compared with the Danish rural population and the results of the second study with older studies and with a reference sample.

In the studies of Schultes ( 62 ) and Sukys et al. ( 63 ), no conclusion regarding the results was reached since the comparison with different reference samples brought different results. The long and the short form of the HLS-EU was used for measurement in both of these studies. In the study by Schultes ( 62 ), various bachelor's degree programs were included and in the study by Sukys et al. ( 63 ) different health study programs, except for medicine. In other three studies ( 48 , 60 , 64 ), no conclusion regarding general health literacy is given. Regarding eHealth literacy, authors of two studies ( 69 , 70 ) speak of medium or higher scores based on a score of their measurement instrument, and the third study ( 68 ) made no statement regarding general eHealth literacy levels. It should be noted that these three studies were conducted by the same research team.

Quantifying the risk of bias of the included studies using the AXIS tool (see Table 2 ), seven studies were classified as very low risk of bias ( 54 , 56 , 57 , 64 , 66 , 67 , 69 ), 11 studies were classified as low risk of bias ( 48 , 50 , 51 , 55 , 58 , 60 , 61 , 63 , 65 , 68 , 70 ), two studies were classified as the moderate risk of bias ( 53 , 59 ), and one study was classified as high risk of bias ( 62 ). In terms of quality, we are therefore dealing with a comparatively solid and homogeneous study situation, with only three out of 21 studies falling short. The main weaknesses of the included studies were the lack of sample size justification and not addressing non-responders.

Table 2 . Quality assessment of the included studies.

Possible Determinants of Health Literacy

Among the determinants presented, there was strong evidence for a relationship between health literacy and age, the semester of study, gender, course of studies, parental education, and socioeconomic background. Other possible determinants could be accessed to information, health-related experiences, financial situation, social support, housing situation, physical activity, smoking status, symptoms of depression and anxiety disorders, employment status of parents, and daily fruit and vegetable consumption. For students with impaired hearing or deafness, the frequency of health-related discussions with friends and access to a language in childhood play a critical role. Electronic health literacy may be related to a medical degree course. There are also several determinants for the respective sublevels of eHealth literacy. With regard to the length of this section, the methodology and conduct of individual studies are only discussed, if they involve a special sample or use a debatable measuring instrument.

Better health literacy with increasing age is shown in three studies ( 54 , 57 , 65 ) with 1,160 students overall, of which 419 come from health professional training programs ( 54 ). This correlation can be explained by increased experience with the healthcare system. With increasing age and experience, older students have an advanced ability to navigate within the healthcare system and engage with healthcare professionals. This results in increased awareness of health promotion resources in their environment and greater self-confidence when talking to healthcare professionals ( 54 , 65 ). One study with 127 students found no correlation between health literacy and age ( 58 ).

In terms of gender, there were four studies ( 55 , 57 , 63 , 65 ) with a total of 2,029 participants that measured higher health literacy among female students and two studies ( 50 , 54 ) with a total of 3,311 participants that measured higher health literacy among male students, whereby it should be mentioned that Göring and Rudolph ( 50 ) used a self-made measuring instrument. Except for two studies ( 54 , 55 ), these results refer to various study programs. These differences can be explained by variations in the educational system on the one hand, and sociocultural influences on the other ( 55 , 57 ). For example, in predominantly patriarchal societies, women have less influence on household decision-making. Also, male children are preferred to female children because of the idea that boys need more knowledge and therefore should be able to maintain their health ( 54 ). Another explanation could also be that women assess the individual ability to influence subjective health in a different way than men. For example, a different perception of complaints and specifically female complaints can influence one's own self-efficacy expectations regarding one's health in a different way to men ( 50 ). Two studies with 1,123 participants, however ( 58 , 61 ), could not find any differences between genders.

The Course of Studies

Six studies with a total of 3,873 students overall describe different levels of health literacy concerning the course of studies ( 54 – 57 , 63 , 66 ). Except for Rababah et al. ( 57 ), these results were found in studies that compared health-related courses of study. The results must, therefore, be interpreted carefully. These results can be explained by the specificity in certain health-related curricula. The contents of multiple health-related courses of study usually cover different areas of health promotion and disease prevention and individual political and organizational health areas. Students in health settings overall have better access to and understanding of health-related information, which facilitates decision-making and application of the decision. Besides, students in health-related courses of study often have a personal interest in the context of health promotion and the associated competencies due to their choice of study ( 54 , 55 , 63 ).

Study Semester

As the number of semesters of health students increases, so do the values of health literacy according to four studies ( 55 , 57 , 66 , 67 ) with a total of 2,783 participants. This supports the assumption that in addition to personal motivation, the curriculum has a major influence on acquiring skills and knowledge related to one's health. As the semester increases, so does the knowledge obtained. Late semesters already have more medical expertise and know-how to obtain quality information ( 55 , 66 , 67 ). One study with 127 students found no correlation regarding this determinant ( 58 ).

Parental Education

Five studies including a total of 2,903 students recorded higher health literacy if their parents have received higher education ( 54 , 55 , 65 – 67 ). Except for Vamos et al. ( 65 ), this concerns students from several health-related courses. Possible explanations could be the increased health awareness of the parents due to their education, which enables them to navigate their children through the health system and rubs off on the children ( 54 , 55 , 65 – 67 ). One study with 127 participants found no correlation between the education of parents and the health literacy of students ( 58 ).

Socioeconomic Background

Three studies including a total of 2,108 students found that higher socioeconomic groups have better access, understanding, and handling of health-related resources ( 65 – 67 ). Within this result, all three forms of existing samples are present (several health-related courses of study, one specifically health-related course of study, various courses of study, and the number of semesters). Due to their higher socioeconomic status, students are more likely to be exposed to or have access to relevant information from parents and other health promotion resources. Here too, parents play a decisive role, since the socioeconomic status of students reflects the socioeconomic status of their parents ( 65 , 67 ).

Access to Information

One study ( 60 ) with 485 students from all courses of the study found that while the internet is the most popular way for students to access information, it is also associated with the worst health literacy scores (compared to those, who appeal to family and friends or specialty journals as a source of health information). This is most likely due to the quality of information available on the internet. Information on the internet is often incorrect and hardly comprehendible.

Health-Related Experiences

According to one study ( 55 ) with a sample size of 376 participants, students in health-related programs who have already gained experience in healthcare (e.g., hospitalization) have better health literacy. The reason for this is the experience they have already had and the support they receive from healthcare providers and their assessment of their ability to find health-related information and communicate with healthcare professionals.

Physical Activity

Regarding physical activity, one study including 2,892 students ( 50 ) from various courses of study reports a positive relationship between health literacy and sporting activity due to increased self-efficacy expectations, measured with a self-made measuring instrument. One study with 533 students ( 62 ) also from various courses of study, on the other hand, does not report any correlation, this being the study with a high risk of bias.

Various other determinants of health literacy for several health-related and various courses of study were discussed in the involved studies: better financial situation ( 61 ) and positively perceived health behavior ( 62 ), non-smoking status ( 57 ), symptoms of depression and anxiety disorders ( 66 ), and daily consumption of fruits and vegetables ( 62 ). Social support should also be mentioned, as social exchange processes can lead to greater security in obtaining and handling health-related information ( 61 ). Lastly, the employment status of parents is of interest, as higher health literacy was found among students with working parents. This phenomenon could be explained by better access to technological resources ( 53 ).

No Influence on Health Literacy

In addition to the abovementioned missing correlations, no connection was found between health literacy and the migration background ( 61 ) or membership to a health profession ( 58 ). Contrary to another study ( 57 ), one study ( 62 ) found no correlation between higher levels of health literacy and smoking status and alcohol consumption. However, it should be noted that this is a study with a moderate risk of bias.

Special Student Groups

One study ( 48 ) measured health literacy in a group of 37 deaf students with the S-TOFHLA for functional literacy, two extra questions for interactive health literacy, and critical health literacy via the response to a self-made video. It was found that a higher frequency of health-related discussions significantly contributes to better critical health literacy. Language barriers can be avoided by healthy-literate peers who share a common language. The critical health literacy of deaf students was not influenced by the hearing ability of family members, so other social characteristics, such as the effort of the parents to communicate with the deaf individual, encourage participation in family discussions about health ( 48 ).

Possible Determinants of eHealth Literacy

Three studies ( 68 – 70 ) with a total of 1,858 students have specifically addressed determinants of eHealth. In each case, the different forms of health literacy, functional, interactive, and critical, were analyzed. According to Yang et al. ( 68 ), the only general determinant for higher eHealth literacy, in general, is belonging to a medical degree program.

Functional eHealth Literacy

In functional eHealth, a high frequency in the use of medical services was discovered. Poor understanding of medical care directions and poor problem-solving skills may lead to ineffective care and a lack of behavioral change when new information is available ( 69 ). However, a lower intake of unhealthy food could also be associated with higher functional eHealth literacy. Students are thus able to understand the risks associated with unhealthy food and can avoid its intake in everyday life ( 70 ).

Interactive eHealth Literacy

The selection of appropriate types and locations for health services and a low frequency of use of these have been measured at high interactive levels of eHealth literacy. Interactive eHealth literacy could help students to act independently, increase their motivation and self-confidence, thereby selecting appropriate types and locations for their health needs ( 69 ). It is also linked to a balanced diet and health aspects of consumers' dietary behaviors, as interactive eHealth literacy can lead to students actively participating in everyday activities and promoting healthy consumption patterns ( 70 ).

Critical eHealth Literacy

The highest level of eHealth literacy is linked to three possible determinants. First, the selection of appropriate types, locations, and purpose aspects of health services, as critical eHealth literacy allows individuals greater control over life events and situations, thus enabling them to evaluate health issues, as well as risks and benefits and advocate for themselves ( 69 ). Next comes regular eating habits and consumer health. By critically evaluating electronic health information, students can filter out the advantages and disadvantages of this information and apply them to their eating habits and activities ( 70 ). Finally, positive, health-promoting behaviors are associated with higher critical eHealth literacy. Through the highest level of eHealth literacy, students can engage in health-enhancing actions through critical examination and advocating for themselves, to engage in health-enhancing actions ( 68 ).

No Influence on eHealth Literacy

No link to eHealth literacy was found in gender and frequency of consumption of organic food. As this is an educated and age-limited group, possible gender differences may have been compensated ( 69 ). The frequency of organic food consumption is probably influenced more by perceptions of food safety than by knowledge about the food itself. Various food incidents worldwide may be the primary decision maker regardless of the level of eHealth literacy ( 68 ).

The general level of health literacy among university students seems to be insufficient and needs to be improved. Regarding the distribution of study courses, this observation seems to apply to both health-related and other study courses–although students from health-related study programs tend to have better health literacy. The health literacy of students is influenced by different variables. In this review, strong evidence for a relationship between health literacy and age, gender, number of semesters, course of studies, parental education, and socioeconomic background was found. These assumptions must be considered with regard to the respective samples selected. For example, regarding age and gender, more studies were represented with a general sample of students, while in course of study and parental education, more studies were represented with a sample of students studying health-related subjects. Concerning the number of semesters, only students from the health sector were represented, while concerning the socioeconomic background the distribution of students was equal among all sample types.

Students can benefit from increased health literacy for their own health. In addition to the personal added value, a social benefit can arise from health-competent multipliers in responsible positions. Besides, the results should always be considered in the context of the country's existing health system and social conception of health. Particularly concerning the results of gender differences, the cultural context must be considered. Health literacy can therefore possibly only be compared between populations if social, economic, and health systems are congruent ( 59 ). In general, however, it is recommended that universities pay more attention to the promotion of health literacy when planning the curriculum or additional offers for students. Electronic health literacy levels among students were high in the studies presented. However, this result should be interpreted with caution, as all three studies involved were conducted in the same country and possibly the same colleges and contradict the results regarding normal health literacy. A review ( 71 ) with six peer-reviewed articles and one doctoral dissertation with numbers of participants ranging from 34 to 5,030 on eHealth literacy also speaks of a high level of connection to the internet among students, but also of limited eHealth literacy. As the internet is the preferred way to obtain health information even if it does not lead to better health literacy or eHealth literacy, work is needed to promote the quality of the information and the ability of students to evaluate it ( 60 ). While the results of this review must be considered carefully, they can be used as a starting point for planning interventions and monitoring health literacy among students over the long term.

Concerning the studies, limitations in the performance of the measurements and the tests used were discovered. During the data collection process, practicability was prioritized, which meant that precision and quality had to suffer. This includes the use of incomplete questionnaires ( 70 ), or the inability to secure an appropriate, private space to take measurements ( 57 ). There were also limitations in the distribution of questionnaires. The use of social media can lead to self-selection bias and a lack of control over appropriate data ( 55 ). The self-reporting method may influence the accuracy of the results and the use of e-mail and online surveys may exclude students with low affinity to the internet ( 51 ). Some of the tests used had little or no evidence of their reliability or validity. A comparison between and within the studies is also difficult, because on the one hand HLQ-scores, for example, may not be comparable due to some scales being harder to score on ( 56 ), on the other hand, some studies used the long and other studies the more roughly measuring short form of their used test (e.g., HLS-EU-Q16 and HLS-EU-Q47). When using vignettes, participants may indicate what they think they have to indicate rather than giving their honest opinion ( 59 ). Another limitation was the exclusion of international students due to a language barrier.

The results of the study cannot readily be generalized, and its interpretation should only be applied to the respective groups of students. The reasons for this are the differences between the selected samples and the selected variables studied. For example, among the included studies there was often an uneven distribution in terms of gender or number of semesters. It should also be highlighted that some studies have examined students from various study programs and others only medical or health students. Due to a lack of time and money, very little information about the students was collected mostly. There may be other mediating or confounding variables that affect health literacy.

Also, this review is not without limitations. Overall, the quality of the included studies is sound. Nevertheless, there are three studies with moderate-to-poor study quality among them, and the majority of the high-quality studies lack sample size justification and addressing of non-responders as well. Differences regarding assessment methods, study population, and sample size hamper the comparison between the studies. Finally, it should be mentioned that only German and English language studies, and studies that have already been published or were available, were considered in this review.

Implications for Practice

Health literacy activities should target all students. Universities should make use of their resources and offer health literacy courses for students in which content is used from disciplines available at the university (e.g., medicine, health, or psychology). Multisectoral and multidisciplinary efforts are essential in promoting health for students, since not only synergies with regard to knowledge and resources are enabled, but also access to certain student subpopulations are made possible ( 72 ). To increase effectiveness, health literacy courses should be adapted according to the different needs and characteristics of the student subgroups and should be linked to evaluative research. The internet as well as gamification approaches, in particular, can help to make interventions interesting for the selected target group. Besides, social networks can provide an easy way to reach and connect students to promote their health and eHealth literacy, why peer-to-peer programs could play a role in this context. To consider special groups of students (e.g., deaf students), care should always be taken to include a suitable form of language or exchange with health literate, accessible peers in the interventions ( 48 ). Additionally, consideration should be given to the planning process when cross-curricular activities are offered for students with different backgrounds and courses of study. When planning interventions according to specific areas of health literacy, different needs of student groups can be taken into account. Furthermore, a central website of the university could be used to communicate accurate and actionable health-related information in a way that is appropriate for the target group, as has already been done during the corona pandemic through the development of corona landing pages for students with frequently asked questions.

Implications for Research

The results of this review suggest that students are a relevant target group for future health literacy studies. Furthermore, there is a need for appropriate measurement methods in the university setting that reflects the circumstances of the living situation for students. Additional variables (e.g., structural aspects, such as support services provided by the university) that may be possible determinants of student health literacy should be collected. Once interventions have been designed, they can be examined to determine which methods and media (despite the challenge of the fast-changing digital environment) are most effective and which determinants in the cultural and social context require particular attention. To ensure that interventions are accessible to all students on campus, more research is needed on accessibility and effectiveness for specific student groups. Appropriate tools must also be developed to regularly check the quality of information available online to counteract misinformation.

Data Availability Statement

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.

Author Contributions

All authors listed have made a substantial, direct, and intellectual contribution to the work and approved it for publication.

The study received funding by the Techniker Krankenkasse (German health insurance). This article has been funded through the Open Access Publishing Fund of the Karlsruhe Institute of Technology.

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher's Note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

1. Ribeiro ÍJS, Pereira R, Freire IV, de Oliveira BG, Casotti CA, Boery EN. Stress and quality of life among university students: a systematic literature review. Health Prof Educ. (2018) 4:70–7. doi: 10.1016/j.hpe.2017.03.002

CrossRef Full Text | Google Scholar

2. Stewart-Brown S, Evans J, Patterson J, Petersen S, Doll H, Balding J, et al. The health of students in institutes of higher education: an important and neglected public health problem? J Public Health. (2000) 22:492–9. doi: 10.1093/pubmed/22.4.492

PubMed Abstract | CrossRef Full Text | Google Scholar

3. Deasy C, Coughlan B, Pironom J, Jourdan D, Mannix-McNamara P. Psychological distress and coping amongst higher education students: a mixed method enquiry. PLoS ONE. (2014) 9:e115193. doi: 10.1371/journal.pone.0115193

4. Okanagan C,. Okanagan Charter: An International Charter for Health Promoting Universities Colleges. (2015). Available online at: https://bp-net.ca/wp-content/uploads/2019/03/Okanagan-Charter.pdf (accessed March 20, 2020).

5. World Health Organization. Promoting Health: Guide to National Implementation of the Shanghai Declaration . Geneva: World Health Organization (2018). Report No.: Contract No.: WHO/NMH/PND/18.2.

6. World Health Organization. Ottawa Charter for Health Promotion. (1986). Available online at: https://www.euro.who.int/__data/assets/pdf_file/0006/129534/Ottawa_Charter_G.pdf (accessed 20 March, 2020).

7. Berkman ND, Sheridan SL, Donahue KE, Halpern DJ, Crotty K. Low health literacy and health outcomes: an updated systematic review. Ann Intern Med . (2011) 155. doi: 10.7326/0003-4819-155-2-201107190-00005

PubMed Abstract | CrossRef Full Text

8. Brach C, Keller D, Hernandez L, Baur C, Parker R, Dreyer B, et al. Ten attributes of health literate health care organizations. NAM Perspectives. (2012) 2. doi: 10.31478/201206a

9. Nutbeam D. The evolving concept of health literacy. Soc Sci Med. (2008) 67:2072–8. doi: 10.1016/j.socscimed.2008.09.050

10. Nutbeam D. Health Promotion Glossary. Health Promot Int. (1998) 13:349–64. doi: 10.1093/heapro/13.4.349

11. Parker R. Measuring health literacy: what? So what? Now what? In: Hernandez L, editor. Measures of health literacy: workshop, summary, Roundtable on Health Literacy . Washington, DC, National Academies Press (2009). p. 91–8.

Google Scholar

12. Sørensen K, Van den Broucke S, Fullam J, Doyle G, Pelikan J, Slonska Z, et al. Health literacy and public health: a systematic review and integration of definitions and models. BMC Public Health . (2012) 12. doi: 10.1186/1471-2458-12-80

13. Oliffe JL, Rossnagel E, Kelly MT, Bottorff JL, Seaton C, Darroch F. Men's health literacy: a review and recommendations. Health Promot Int. (2020) 35:1037–51. doi: 10.1093/heapro/daz077

14. Zamora H, Clingerman EM. Health Literacy Among Older Adults. J Gerontol Nurs. (2011) 37:41–51. doi: 10.3928/00989134-20110503-02

15. Ward M, Kristiansen M, Sørensen K. Migrant health literacy in the European Union: a systematic literature review. Health Educ J. (2018) 78:81–95. doi: 10.1177/0017896918792700

16. Klem ML, Saleh AA, Devine PJ, Gutzman KE, Knehans AC, Mills TN, et al. Librarians and health literacy: A scoping review. Libr Inf Sci Res. (2019) 41:102–8. doi: 10.1016/j.lisr.2019.04.005

17. Heuberger RA, Ivanitskaya L. Preferred sources of nutrition information: contrasts between younger and older adults. J Intergener Relatsh. (2011) 9:176–90. doi: 10.1080/15350770.2011.567919

18. Karnoe A, Kayser L. How is eHealth literacy measured and what do the measurements tell us? a systematic review. KM&EL . (2015) 7:576–600. doi: 10.34105/j.kmel.2015.07.038

19. Norman CD, Skinner HA. eHealth literacy: essential skills for consumer health in a networked world. J Med Internet Res. (2006) 8:e9. doi: 10.2196/jmir.8.2.e9

20. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the prisma statement. PLoS Medicine. (2009) 6. doi: 10.1371/journal.pmed.1000097

21. Nutbeam D. Health literacy as a public health goal: a challenge for contemporary health education and communication strategies into the 21st century. Health Promot Int. (2000) 15:259–67. doi: 10.1093/heapro/15.3.259

22. Chesser AK, Keene Woods N, Smothers K, Rogers N. Health literacy and older adults: a systematic review. Gerontol Geriatr Med. (2016) 2:2333721416630492. doi: 10.1177/2333721416630492

23. Downes MJ, Brennan ML, Williams HC, Dean RS. Development of a critical appraisal tool to assess the quality of cross-sectional studies (AXIS). BMJ . (2016) 6:e011458. doi: 10.1136/bmjopen-2016-011458

24. Wunsch K, Fiedler J, Bachert P, Woll A. The Tridirectional relationship among physical activity, stress, and academic performance in university students: a systematic review and meta-analysis. Int J Environ Res Public Health. (2021) 18. doi: 10.3390/ijerph18020739

25. Taylor M, Masood M, Mnatzaganian G. Longevity of complete dentures: a systematic review and meta-analysis. J Prosthet Dent . (2020) 125:611–9. doi: 10.1016/j.prosdent.2020.02.019

26. Popay J, Roberts H, Sowden A, Petticrew M, Arai L, Rodgers M, et al., editors. Guidance on the conduct of narrative synthesis in systematic Reviews. A Product from the ESRC Methods Programme . Version 1. Lancaster: Lancaster University (2006).

27. Baker DW. The meaning and the measure of health literacy. J Gen Intern Med. (2006) 21:878–83. doi: 10.1111/j.1525-1497.2006.00540.x

28. Kickbusch I, Maag D. Health Literacy. In: Kris H, Stella Q, editors. International Encyclopedia of Public Health . 3. San Diego: Academic Press (2008). p. 204–11. doi: 10.1016/B978-012373960-5.00584-0

29. Kickbusch I, Maag D, Wait S. Navigating Health: The Role of Health Literacy : Alliance for Health and the Future, International Longevity Centre-UK (2005).

30. Paasche-Orlow MK, Wolf MS. The causal pathways linking health literacy to health outcomes. Am J Health Behav. (2007) 31:19–26. doi: 10.5993/AJHB.31.s1.4

31. Zarcadoolas C, Pleasant A, Greer DS. Advancing Health Literacy: A Framework for Understanding and Action : Wiley (2006).

32. Abacigil F, Harlak H, Okyay P, Kiraz DE, Gursoy Turan S, Saruhan G, et al. Validity and reliability of the Turkish version of the European Health Literacy Survey Questionnaire. Health Promot Int. (2019) 34:658–67. doi: 10.1093/heapro/day020

33. Diamond JJ, Becker JA, Arenson CA, Chambers CV, Rosenthal MP. Development of a scale to measure adults' perceptions of health: preliminary findings. J Community Psychol. (2007) 35:557–61. doi: 10.1002/jcop.20164

34. Osborne RH, Batterham RW, Elsworth GR, Hawkins M, Buchbinder R. The grounded psychometric development and initial validation of the Health Literacy Questionnaire (HLQ). BMC Public Health. (2013) 13. doi: 10.1186/1471-2458-13-658

35. Maindal HT, Kayser L, Norgaard O, Bo A, Elsworth GR, Osborne RH. Cultural adaptation and validation of the Health Literacy Questionnaire (HLQ): robust nine-dimension Danish language confirmatory factor model. SpringerPlus. (2016) 5. doi: 10.1186/s40064-016-2887-9

36. Wieland R, Hammes M. Gesundheitskompetenz als personale Ressource. In: Mozygemba K, editors. Nutzenorientierung-ein Fremdwort in der Gesundheitssicherung ? Bern: Huber (2008). p. 177–90.

37. Bässler R. Methodische Überlegungen zur empirischen Erhebung und Beschreibung der “Teilnahme der Bevölkerung am Sport”. SWS-Rundschau. (1988) 28:191–7.

38. Woll A. Gesundheitsförderung in der Gemeinde: eine empirische Untersuchung zum Zusammenhang von sportlicher Aktivität, Fitneß und Gesundheit bei Personen im mittleren und späteren Erwachsenenalter . Neu-Isenburg: LinguaMed-Verl-GmbH (1996).

39. Montazeri A, Tavousi M, Rakhshani F, Azin SA, Jahangiri K, Ebadi M, et al. Health literacy for iranian adults (HELIA): development and psychometric properties. Health Monitor. Iran J Health Sci. (2014) 13:589–99.

40. Walker SN, Sechrist KR, Pender NJ. The health-promoting lifestyle profile. Nursing Research. (1987) 36. doi: 10.1097/00006199-198703000-00002

CrossRef Full Text

41. Baker DW, Williams MV, Parker RM, Gazmararian JA, Nurss J. Development of a brief test to measure functional health literacy. Patient Educ Couns. (1999) 38:33–42.

PubMed Abstract | Google Scholar

42. Sørensen K, Pelikan JM, Röthlin F, Ganahl K, Slonska Z, Doyle G, et al. Health literacy in Europe: comparative results of the European health literacy survey (HLS-EU). Eur J Public Health. (2015) 25:1053–8. doi: 10.1093/eurpub/ckv043

43. Duong TV, Aringazina A, Baisunova G., Nurjanah, Pham TV, Pham KM, et al. Measuring health literacy in Asia: Validation of the HLS-EU-Q47 survey tool in six Asian countries. J Epidemiol. (2017) 27:80–6. doi: 10.1016/j.je.2016.09.005

44. Nunes LS, Sorensen K, Pelikan J, Cunha M, Rodrigues E, Paixão E. Cross-cultural adaptation and validation to Portuguese of the European Health Literacy Survey (HLS-EU-PT). Atencion Primaria. (2014) 46:13. Available online at: https://www.elsevier.es/en-revista-atencion-primaria-27-pdf-S0212656714700691

45. Norman CD, Skinner HA. eHEALS: the eHealth literacy scale. J Med Internet Res . (2006) 8:e27. doi: 10.2196/jmir.8.4.e27

46. Chiang C, Yang S, Hsu W. Development and validation of the e-health literacy scale and investigation of the relationships between e-health literacy and healthy behavior among undergraduate students in Taiwan Formosa. J Ment Health. (2015) 28:389–420. doi: 10.30074/FJMH.201509_28(3).0002

47. Luo Y-F, Yang S-C, Chiang C-H, Lu C-M. Development and validation of a food literacy self-report inventory and investigation of the relationships between food literacy and dietary behavior among college students. Taiwan Gong Gong Wei Sheng Za Zhi. (2018) 37:407. doi: 10.6288/TJPH.201808_37(4).107029

48. Kushalnagar P, Ryan C, Smith S, Kushalnagar R. Critical health literacy in American deaf college students. Health Promot Int. (2018) 33:827–33. doi: 10.1093/heapro/dax022

49. Kiechle ES, Bailey SC, Hedlund LA, Viera AJ, Sheridan SL. Different measures, different outcomes? a systematic review of performance-based versus self-reported measures of health literacy and numeracy. J Gen Intern Med. (2015) 30:1538–46. doi: 10.1007/s11606-015-3288-4

50. Göring A, Rudolph S. Die Gesundheitskompetenz von Studierenden in Abhängigkeit von sportlicher Aktivität. In: Göring A, Möllenbeck D, editors. Bewegungsorientierte Gesundheitsförderung an Hochschulen . 3. Göttingen: Universitätsverlag Göttingen (2015). p. 147–65. doi: 10.17875/gup2015-811

51. Kaboudi M, Kianipou N, Ziapour A, Dehngan F. A study of health literacy components and their relationship with health-promotin behaviors among students at kermanshah university of medical sciences in 2017. Int J Pediatr. (2017) 5:6721–9. doi: 10.22038/ijp.2017.26823.2313

52. Ratzan S, Parker R, Selden C, Zorn M. National Library of Medicine Current Bibliographies in Medicine: Health Literacy . Bethesda, MD: National Institutes of Health (2000).

53. Birimoglu Okuyan C, Caglar S. Investigation of health literacy levels and health perceptions of nursing students: a cross-sectional study. Int J Caring Sci. (2019) 12:270–9. Available online at: http://www.internationaljournalofcaringsciences.org/docs/30_birimoglu_original_12-1_1.pdf

54. Budhathoki SS, Pokharel PK, Jha N, Moselen E, Dixon R, Bhattachan M, et al. Health literacy of future healthcare professionals: a cross-sectional study among health sciences students in Nepal. Int Health. (2019) 11:15–23. doi: 10.1093/inthealth/ihy090

55. Elsborg L, Krossdal F, Kayser L. Health literacy among Danish university students enrolled in health-related study programmes. Scand J Public Health. (2017) 45:831–8. doi: 10.1177/1403494817733356

56. Mullan J, Burns P, Weston K, McLennan P, Rich W, Crowther S, et al. Health literacy amongst health professional university students: a study using the health literacy questionnaire. Education Sciences. (2017) 7:54. doi: 10.3390/educsci7020054

57. Rababah JA, Al-Hammouri MM, Drew BL, Aldalaykeh M. Health literacy: exploring disparities among college students. BMC Public Health. (2019) 19. doi: 10.1186/s12889-019-7781-2

58. Reick S, Hering T. Health literacy of students. Results of an online survey at the Bochum health university (Germany)/Gesundheitskompetenz Studierender–Ergebnisse einer Online-Befragung an der Hochschule für Gesundheit Bochum. Int J Health Prof. (2018) 5:44–52. doi: 10.2478/ijhp-2018-0007

59. Runk L, Durham J, Vongxay V, Sychareun V. Measuring health literacy in university students in Vientiane, Lao PDR. Health Promot Int. (2016) 32:360–8. doi: 10.1093/heapro/daw087

60. Santos P, Sa L, Couto L, Hespanhol AP. Sources of information in health education: a cross-sectional study in Portuguese university students. Med Stud J Aust . (2018) 11. doi: 10.21767/AMJ.2018.3435

61. Schricker J, Rathmann K, Dadaczynski K. Soziale Unterschiede in der gesundheitskompetenz von studierenden: ergebnisse einer online-studie an der technischen universität dortmund. Prävention und Gesundheitsförderung. (2019) 15:8–14. doi: 10.1007/s11553-019-00731-6

62. Schultes K. Gesundheitskompetenz, subjektive Gesundheit und Gesundheitsverhalten bei Studierenden. Public Health Forum. (2017) 25:84–6. doi: 10.1515/pubhef-2016-2115

63. Sukys S, Cesnaitiene VJ, Ossowsky ZM. Is Health Education at University Associated with Students' Health Literacy? Evidence from Cross-Sectional Study Applying HLS-EU-Q. BioMed Research International. (2017) 2017:1–9. doi: 10.1155/2017/8516843

64. Suri VR, Majid S, Chang Y-K, Foo S. Assessing the influence of health literacy on health information behaviors: a multi-domain skills-based approach. Patient Educ Couns. (2016) 99:1038–45. doi: 10.1016/j.pec.2015.12.017

65. Vamos S, Yeung P, Bruckermann T, Moselen EF, Dixon R, Osborne RH, et al. Exploring health literacy profiles of Texas university students. Health Behavior and Policy Review. (2016) 3:209–25. doi: 10.14485/HBPR.3.3.3

66. Zhang Y, Zhang F, Hu P, Huang W, Lu L, Bai R, et al. Exploring health literacy in medical university students of chongqing, china: a cross-sectional study. Plos ONE. (2016) 11. doi: 10.1371/journal.pone.0152547

67. Zou M, Zhang Y, Zhang F, Hu P, Bai R, Huang W, et al. The ability to obtain, appraise and understand health information among undergraduate nursing students in a medical university in Chongqing, China. Nursing Open. (2018) 5:384–92. doi: 10.1002/nop2.161

68. Yang S-C, Luo Y-F, Chiang C-H. The associations among individual factors, ehealth literacy, and health-promoting lifestyles among college students. J Med Internet Res. (2017) 19:e15. doi: 10.2196/jmir.5964

69. Luo YF, Yang SC, Chen A-S, Chiang C-H. Associations of eHealth Literacy with health services utilization among college students: cross-sectional study. J Med Internet Res . (2018) 20:e283. doi: 10.2196/jmir.8897

70. Yang SC, Luo YF, Chiang C-H. Electronic health literacy and dietary behaviors in Taiwanese college students: cross-sectional study. J Med Internet Res. (2019) 21:e13140. doi: 10.2196/13140

71. Stellefson M, Hanik B, Chaney B, Chaney D, Tennant B, Chavarria EA. eHealth literacy among college students: a systematic review with implications for eHealth education. J Med Internet Res. (2011) 13:e102. doi: 10.2196/jmir.1703

72. Bachert P, Wäsche H, Albrecht F, Hildebrand C, Kunz AM, Woll A. Promoting students' health at university: key stakeholders, cooperation, and network development. Front Public Health. (2021) 9:680714. doi: 10.3389/fpubh.2021.680714

Search Term

(college OR “college students” OR university OR universities OR student OR students OR “young adult” OR “young adults” OR adolescent OR adolescents) AND (“critical health literacy” OR “health literacy” OR “eHealth literacy” OR “functional health literacy” OR “health-related literacy” OR “health literacy education” OR “literacy programs”).

Keywords: health literacy, university students, health-promoting universities, systematic review, determinants of health behavior

Citation: Kühn L, Bachert P, Hildebrand C, Kunkel J, Reitermayer J, Wäsche H and Woll A (2022) Health Literacy Among University Students: A Systematic Review of Cross-Sectional Studies. Front. Public Health 9:680999. doi: 10.3389/fpubh.2021.680999

Received: 15 March 2021; Accepted: 29 November 2021; Published: 21 January 2022.

Reviewed by:

Copyright © 2022 Kühn, Bachert, Hildebrand, Kunkel, Reitermayer, Wäsche and Woll. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

*Correspondence: Philip Bachert, philip.bachert@kit.edu

Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.

Health literacy interventions and outcomes: an updated systematic review

• PMID: 23126607
• PMCID: PMC4781058

Objectives: To update a 2004 systematic review of health care service use and health outcomes related to differences in health literacy level and interventions designed to improve these outcomes for individuals with low health literacy. Disparities in health outcomes and effectiveness of interventions among different sociodemographic groups were also examined.

Data sources: We searched MEDLINE®, the Cumulative Index to Nursing and Allied Health Literature, the Cochrane Library, PsychINFO, and the Educational Resources Information Center. For health literacy, we searched using a variety of terms, limited to English and studies published from 2003 to May 25, 2010. For numeracy, we searched from 1966 to May 25, 2010.

Review methods: We used standard Evidence-based Practice Center methods of dual review of abstracts, full-text articles, abstractions, quality ratings, and strength of evidence grading. We resolved disagreements by consensus. We evaluated whether newer literature was available for answering key questions, so we broadened our definition of health literacy to include numeracy and oral (spoken) health literacy. We excluded intervention studies that did not measure health literacy directly and updated our approach to evaluate individual study risk of bias and to grade strength of evidence.

Results: We included good- and fair-quality studies: 81 studies addressing health outcomes (reported in 95 articles including 86 measuring health literacy and 16 measuring numeracy, of which 7 measure both) and 42 studies (reported in 45 articles) addressing interventions. Differences in health literacy level were consistently associated with increased hospitalizations, greater emergency care use, lower use of mammography, lower receipt of influenza vaccine, poorer ability to demonstrate taking medications appropriately, poorer ability to interpret labels and health messages, and, among seniors, poorer overall health status and higher mortality. Health literacy level potentially mediates disparities between blacks and whites. The strength of evidence of numeracy studies was insufficient to low, limiting conclusions about the influence of numeracy on health care service use or health outcomes. Two studies suggested numeracy may mediate the effect of disparities on health outcomes. We found no evidence concerning oral health literacy and outcomes. Among intervention studies (27 randomized controlled trials [RCTs], 2 cluster RCTs, and 13 quasi-experimental designs), the strength of evidence for specific design features was low or insufficient. However, several specific features seemed to improve comprehension in one or a few studies. The strength of evidence was moderate for the effect of mixed interventions on health care service use; the effect of intensive self-management inventions on behavior; and the effect of disease-management interventions on disease prevalence/severity. The effects of other mixed interventions on other health outcomes, including knowledge, self-efficacy, adherence, and quality of life, and costs were mixed; thus, the strength of evidence was insufficient.

Conclusions: The field of health literacy has advanced since the 2004 report. Future research priorities include justifying appropriate cutoffs for health literacy levels prior to conducting studies; developing tools that measure additional related skills, particularly oral (spoken) health literacy; and examining mediators and moderators of the effect of health literacy. Priorities in advancing the design features of interventions include testing novel approaches to increase motivation, techniques for delivering information orally or numerically, "work around" interventions such as patient advocates; determining the effective components of already-tested interventions; determining the cost-effectiveness of programs; and determining the effect of policy and practice interventions.

Publication types

• Research Support, U.S. Gov't, P.H.S.
• Systematic Review
• Cause of Death
• Emergency Medical Services / statistics & numerical data
• Health Literacy*
• Health Status
• Hospitalization / statistics & numerical data
• Influenza Vaccines / therapeutic use
• Mammography / statistics & numerical data
• Outcome Assessment, Health Care / statistics & numerical data*
• Patient Compliance / ethnology
• Patient Compliance / statistics & numerical data
• Randomized Controlled Trials as Topic
• Influenza Vaccines

Working nine to thrive

At a glance.

• Health can be meaningfully modified by factors outside traditional healthcare systems, including work factors.
• Employers have considerable opportunities to improve health through six modifiable drivers: social interaction, mindsets and beliefs, productive activity, stress, economic security, and sleep.
• Globally, improving employee health and well-being could create $3.7 trillion to $11.7 trillion in economic value.

Imagine a world in which employers make evidence-based investments in the health of their employees. In return, they reap a manifold benefit to those investments: their employees thrive, their business thrives, and the societies in which they operate thrive. There's a positive opportunity that arises when employers address the inherent interconnectedness between work and health.

The 23 drivers of health

The McKinsey Health Institute has identified 23 drivers of health across six categories. 1 Lars Hartenstein and Tom Latkovic, “ The secret to great health? Escaping the healthcare matrix ,” McKinsey Health Institute, December 20, 2022. All of them are considered modifiable]:[[footnote 2]

• physical inputs: diet, supplementation, and substance use
• movement: mobility, exercise, and sleep
• daily living: productive activity, social interaction, content consumption, and hygiene
• exposure: nature, atmosphere, sensory stimulation, materials, and stress
• state of being: mindsets and beliefs, body composition, physical security, and economic security
• healthcare: vaccination, detection and diagnosis, clinical intervention, adherence

Together, the drivers of health have a broad influence on holistic health (mental, physical, social, and spiritual health) and apply to settings beyond the workplace. Other research on holistic health  has explored a smaller ecosystem of factors that are directly measurable within an organization. 3 Jacqueline Brassey, Brad Herbig, Barbara Jeffery, and Drew Ungerman, “ Reframing employee health: Moving beyond burnout to holistic health ,” McKinsey Health Institute, November 2, 2023; Sanne Magnan, “Social determinants of health 101 for health care: Five plus five,” National Academy of Medicine, October 9, 2017.

The McKinsey Health Institute (MHI) has previously identified 23 drivers of health  (see sidebar “The 23 drivers of health”). 1 Lars Hartenstein and Tom Latkovic, “ The secret to great health? Escaping the healthcare matrix ,” McKinsey Health Institute, December 20, 2022. Employment can greatly influence some of these drivers, such as social interaction and sleep. In this article, we zoom in on six drivers of health that employers can influence and could be wise to support. By improving employees’ health, employers could add trillions of dollars to the global economy and have a positive impact on society. When employers and employees work together to improve modifiable drivers of health, everyone benefits.

Modifiable drivers of health in the workplace: What does the research say?

Six modifiable drivers of health in the workplace—social interaction, mindsets and beliefs, productive activity, stress, economic security, and sleep—were identified from the growing body of research that connects the dots among drivers of health and the workplace. Researchers are building a greater understanding of how employers can address modifiable drivers to create change in favor of optimal employee health.

Considering that the average person spends a third of their life at work (more than 90,000 hours in a lifetime), 2 “How many hours does the average person work per week?,” FreshBooks, April 17, 2023. employment can be a critical piece of the puzzle when working toward the goal of improving global health. MHI analyzed 26 workplace factors  to understand how they influence a range of health- and work-related outcomes across 30 countries. 3 Jacqueline Brassey, Brad Herbig, Barbara Jeffery, and Drew Ungerman, “ Reframing employee health: Moving beyond burnout to holistic health ,” McKinsey Health Institute, November 2, 2023. In this article, any McKinsey Health Institute research not otherwise cited comes from this source. That research showed there are important differences between the workplace factors that lead to poor health and those that lead to good health. Our analysis found that employee self-efficacy, adaptability, and feelings of belonging at work were top predictors of good health, whereas toxic workplace behaviors, role ambiguity, and role conflict at work were top predictors of poor health.

Previously, researchers at the University of Oxford’s Wellbeing Research Centre analyzed data from more than 15 million employees on their well-being and the underlying workplace factors driving it. 4 “How to use the Work Wellbeing Score on Indeed company pages,” Indeed, May 1, 2023. The researchers identified and tested 11 factors, including compensation, flexibility, purpose, inclusion, achievement, support, trust, belonging, management, and learning. The three top factors for the companies that scored best on well-being were feeling energized, belonging, and trust. Interestingly, they are different from the top drivers that employees think will make them happy and drive well-being at work: pay and flexibility. 5 “The key drivers of workplace wellbeing: Tapping into the hidden gems of happiness,” Indeed, July 6, 2021.

Together, all the research led us to identify six drivers of health that employers can most easily influence.

Employers can improve employee health through six modifiable drivers

Our analysis shows that employers can effect significant change through six modifiable drivers of health: social interaction, mindsets and beliefs, productive activity, stress, economic security, and sleep. 6 We recognize that employers can influence other modifiable drivers of health not specifically addressed here (for example, diet and mobility) but are focusing this article on the drivers most likely to create considerable opportunities for employers to improve health.

Social interaction

The positive effects of regular social interactions on health have been widely reported. For instance, a study reviewing mortality rates has documented an average 50 percent increase in likelihood for survival if participants have strong social relationships. 1 Julianne Holt-Lunstad, J. Bradley Layton, and Timothy B. Smith, “Social relationships and mortality risk: A meta-analytic review,  PLOS Medicine , July 2010, Volume 7, Number 7. Furthermore, social integration during childhood is related to lower blood pressure and body mass index in adulthood. 2 For more, see Jenny M. Cundiff and Karen A. Matthews, “Friends with health benefits: The long-term benefits of early peer social integration for blood pressure and obesity in midlife,” Psychological Science , May 2018, Volume 29, Number 5.

Social interactions at work experienced by employees strongly influence health and workplace outcomes. Feeling connected at work is associated with greater innovation, engagement, and quality of work—and may be especially impactful for those with smaller social networks outside of their jobs. 3 For more, see  Our epidemic of loneliness and isolation: The U.S. Surgeon General’s advisory on the healing effects of social connection and community , US Office of the Surgeon General, May 3, 2023. MHI’s 2023 research shows experiencing toxic workplace behavior is a strong predictor of negative health outcomes at work, including loneliness at work, the intention to leave an organization, and burnout symptoms.

Toxic workplace behavior is a critical workplace driver to combat. If left unaddressed, it can mitigate the benefits of any health and well-being initiatives pursued. Examples of interventions to counter toxic workplace behavior include establishing a zero-tolerance policy for it and creating anonymous feedback processes through which employees can report it—which also normalizes a culture of providing concrete, specific feedback to colleagues. 4 For more, see Amy Gallo, “How to manage a toxic employee,”  Harvard Business Review , October 3, 2016, and Deepa Purushothaman and Lisen Stromberg, “Leaders, Stop Rewarding Toxic Rock Stars,”  Harvard Business Review , April 20, 2022.

Meanwhile, experiencing psychological safety  on a team and support from coworkers and managers predicts positive health outcomes, including better holistic health . In 2023, MIT Sloan School of Management researchers outlined proven social-health initiatives that helped managers build psychological safety on their teams. 5 Chris Rider et al., “Proven tactics for improving teams’ psychological safety,”  MIT Sloan Management Review , March 27, 2023. They included training managers to use one-on-one meetings to increase employee individuation 6 “Individuation” refers to treating employees as unique individuals. by asking employees what was important to them and where they needed support. Another use of the meetings was to remove blockers for employees by helping them prioritize among tasks. Interestingly, individuation has been shown to increase psychological safety the most when psychological safety is relatively low, while removing blockers is more effective when psychological safety is relatively high.

Mindsets and beliefs

Research, including MHI analysis, has demonstrated a connection between positive mindsets and beliefs and better health experience. 1 For more, see Mathias Allemand, Patrick L. Hill, and Brent W. Roberts, “Examining the pathways between gratitude and self-rated physical health across adulthood,”  Personality and Individual Differences , January 2013, Volume 54, Number 1; Lisa A. Williams and Monica Y. Bartlett, “Warm thanks: Gratitude expression facilitates social affiliation in new relationships via perceived warmth,”  Emotion , February 2015, Volume 15, Number 1; and David S. Yeager et al., “A synergistic mindsets intervention protects adolescents from stress,”  Nature , July 2022, Volume 607, Number 7,919. This includes the positive effects of a growth mindset on mental health and the benefits of gratitude on physical health. Positive mindsets and beliefs in the workplace are also greatly influential in good holistic health.

In fact, good holistic health isn’t achieved by completely avoiding workplace stressors. Instead, it can be maintained through creating positive experiences at work, such as experiencing high self-efficacy, high adaptability, a feeling of meaning, and a feeling of belonging at work. For example, an individual may be able to tolerate the stress of a looming deadline on a big project if they believe that they have the support of their team.

Employers can foster meaning and belonging by engaging employees through compelling storytelling and fostering a connection to an organization’s mission. Purpose-driven companies that excel at this grow two times faster than their competitors do and achieve gains in employee satisfaction, employee retention, and consumer trust. 2 Scott Mautz, “Patagonia has only 4 percent employee turnover because they value this 1 thing so much,”  Inc. , March 30, 2019; Graham Staplehurst, “The evolution of purpose,” Kantar, August 27, 2020; “This is what work-life balance looks like at a company with 100% retention of moms,” Quartz, October 16, 2016. Some of these outcomes may be attributed to employees who are intrinsically motivated and able to maintain better well-being over time, creating a positive performance loop. 3 For more, see Emma L. Bradshaw et al., “A meta-analysis of the dark side of the American dream: Evidence for the universal wellness costs of prioritizing extrinsic over intrinsic goals,”  Journal of Personality and Social Psychology , April 2023, Volume 124, Number 4. Additionally, employee self-efficacy and adaptability are capabilities that can be cultivated among employees to make a more resilient and healthy workforce. 4 For more, see Jacqueline Brassey, Aaron De Smet, and Michiel Kruyt,  Deliberate Calm: How to Learn and Lead in a Volatile World , New York, NY: HarperCollins Publishers, 2022.

Productive activity

Productive activity includes employment- and nonemployment-related activities. Examples include volunteering, caregiving, spending time on hobbies, worshiping, spending time on activism, playing music, and traveling.

Employment has been linked to improved life expectancy. 1 For more, see “Relationship between employment and health,” Health Foundation, October 5, 2022. According to MHI research, one of the top contributors to productivity at work is an individual’s sense of self-efficacy—an employee’s belief that they can cope with difficult or changing situations. Self-efficacy can be improved through interventions, suggesting that employers can target self-efficacy to improve employee productivity. 2 For more, see Jacqueline Brassey et al., “Emotional flexibility and general self-efficacy: A pilot training intervention study with knowledge workers,”  PLOS One , 2020, Volume 15, Number 10.

Furthermore, employers have the opportunity to help the people in their communities connect to meaningful and productive activities that support their long-term health and well-being. Enjoyable leisure activities are also associated with improved psychosocial and physical measures that support good health and well-being, including greater life satisfaction and engagement and lower rates of depression, blood pressure, cortisol, and physical function. 3 For more, see Sarah D. Pressman et al., “Association of enjoyable leisure activities with psychological and physical well-being,”  Psychosomatic Medicine , September 2009, Volume 71, Number 7.

In discussing workplace stressors, it’s important to acknowledge that stress itself isn’t necessarily a bad thing, as it’s actually needed to learn, grow, and develop. 1 For more, see R. B. Zajonc, “Social facilitation,”  Science , July 1965, Volume 149, Number 3,681. Optimal levels of stress can contribute to better performance. After that point, the benefits diminish into worse well-being because of the excessive demands of high stress and lack of replenishment of energy resources. The employer’s role is to ensure that employees are stimulated, challenged, and motivated—but not overwhelmed—by the demands they experience in the workplace.

Chronically elevated levels of stress can increase the risk of cardiovascular disease, neurodegenerative disease, and metabolic disease. 2 For more, see Fan Tian et al., “Association of stress-related disorders with subsequent risk of all-cause and cause-specific mortality: A population-based and sibling-controlled cohort study,”  Lancet Regional Health–Europe , May 2022, Volume 18. Job strain and effort–reward imbalance can predict several common mental disorders. 3 For more, see Bridget Candy and Stephen Stansfeld, “Psychosocial work environment and mental health—a meta-analytic review,”  Scandinavian Journal of Work, Environment & Health , December 2006, Volume 32, Number 6. Additionally, MHI research shows that an increase in workplace demands is the driver most predictive of burnout and distress symptoms at work.

Some jobs are high in demand by structure. For example, some organizations have seasonal or other cyclical patterns in work demand. In these situations, interventions should focus on building in recovery time so that employees can regain their energy after high-demand periods.

Economic security

Economic opportunity and economic security can influence many facets of health and productivity. For example, high-income individuals are five times more likely than low-income individuals to report strong health. 1 Steven H. Woolf et al.,  How are income and wealth linked to health and longevity? , a joint report from Urban Institute and Virginia Commonwealth University, April 13, 2015. Employees who are struggling financially are more likely than others to experience signs of poor mental health that might affect their ability to function at work. 2 For more, see Lu Fan and Soomin Ryu, “The relationship between financial worries and psychological distress among U.S. adults,”  Journal of Family and Economic Issues , 2023, Volume 44, Number 1. A lack of job stability links with poor mental health, as well as poor physical well-being (for example, cardiovascular disease). 3 For more, see Susan J. Ashford, Guo-Hua Huang, and Cynthia Lee, “Job insecurity and the changing workplace: Recent developments and the future trends in job insecurity research,”  Annual Review of Organizational Psychology and Organizational Behavior , January 2018, Volume 5; Imma Cortès-Franch et al., “Employment stability and mental health in Spain: Towards understanding the influence of gender and partner/marital status,”  BMC Public Health , April 2018, Volume 18, Number 1; Marnie Dobson, Paul Landsbergis, and Peter L. Schnall, “Globalization, work, and cardiovascular disease,”  International Journal of Health Services , October 2016, Volume 46, Number 4; and Jose A. Llosa et al., “Job insecurity and mental health: A meta-analytical review of the consequences of precarious work in clinical disorders,”  Anales de Psicología , 2018, Volume 34, Number 2. Any short-term rise in employee performance fueled by job insecurity is often negated by the additional burden on employee physical and mental health. 4 Mindy Shoss et al., “Job insecurity harms both employees and employers,”  Harvard Business Review , September 6, 2022.

MHI research shows that the greatest contributor to employees’ feelings of financial insecurity is whether they are paid sufficiently to cover their basic needs. While what it takes to feel economically secure is unique to each person, employers can reduce feelings of financial insecurity by ensuring that compensation covers basic needs.

There’s a strong association between sleep hours and both employee health and workplace outcomes. The cost to employers when employees have insufficient or poor-quality sleep can be substantial.

Employees with untreated insomnia cost employers an average of $2,280 more annually than employees without untreated insomnia because of absenteeism, “presenteeism,” poor performance, and increased incidents of accident and injury. 1 Ronald C. Kessler et al., “Insomnia and the performance of US workers: Results from the America Insomnia Survey,”  Sleep , September 2011, Volume 34, Number 9. According to the MHI 2023 survey, 31 percent of employees across the world average fewer than seven hours of sleep per night. Although everyone has unique needs, this falls below the ballpark number of hours recommended to maintain good health. 2 We acknowledge that every individual is different, but there are some indications of recommended average hours of sleep that may benefit health. For more, see Eric J. Olson, “How many hours of sleep are enough for good health?,” Mayo Foundation for Medical Education and Research (MFMER), February 21, 2023. Researchers have shown severe sleep loss can even lead to death, as our bodies conduct necessary reparative processes when we sleep. 3 For more, see Alexandra Vaccaro et al., “Sleep loss can cause death through accumulation of reactive oxygen species in the gut,”  Cell , June 2020, Volume 181, Number 6.

The MHI survey found that one of the main contributors to an employee’s average number of sleep hours is the experienced volume of work required of them. Furthermore, one of the top contributors to an employee’s satisfaction with their sleep is their ability to adjust to unexpected changes. This may suggest that employee programs that look to improve adaptability may in turn improve employees’ satisfaction with their sleep.

Employers have additional interventions they can consider if their employees are struggling with getting consistent, high-quality sleep. They include creating work environments with ample natural light and access to healthy foods, limiting or disabling employees from being online after hours, creating incentives for employees who prioritize sleep, and encouraging and rewarding leaders who model the prioritization of sleep over work.

Many employers are already investing in employee health and well-being, but we would encourage them to reflect on where they currently provide support and if they might want to change resources or add more interventions. For example, many employee assistance programs (EAPs) provide coverage of interventions for factors such as stress and economic security but less coverage of those for factors such as social interactions at work. Additionally, while EAPs are widely available, they tend to be underused by employees and focus on a reactive instead of a proactive approach to health. 7 For more, see James Kenney, “Why most employee assistance programs don't work,” Forbes , July 6, 2022; and Stephen Sokoler, “Reimagining traditional employee assistance programs,” Forbes , March 17, 2023.

In rethinking a workplace strategy on employee health and well-being, current EAP offerings can be useful starting points for action but are unlikely to be the full solution. They are also unlikely, by themselves, to yield the ROI that employers increasingly expect. Strengthening the measurement of intervention outcomes may also help guide an organization’s overall investment strategy.

In rethinking a workplace strategy on employee health and well-being, current EAP offerings can be useful starting points for action but are unlikely to be the full solution.

Improving global employee health can create trillions of dollars of economic value

It makes good business sense to invest in employee health and well-being. We estimate that the total global opportunity for optimizing employee health and well-being is $3.7 trillion to $11.7 trillion, which is equivalent to raising global GDP by 4 to 12 percent. Together, high- and middle-income economies represent 95 percent of this total opportunity (exhibit).

While it may not be feasible in the near term to bring all employees everywhere to optimal well-being, capturing just 10 percent of the total opportunity could yield up to $1.17 trillion of annual value and raise the global GDP by more than 1 percent (see sidebar “Business case methodology”).

Business case methodology

To size the economic value that could be created if addressing health and well-being at the global level, we first established the metric for all economies by summing the positive economic effects of increased employee attraction, productivity, and retention with the savings created if absenteeism, attrition, and “presenteeism” were reduced.

Using a similar methodology, we sized the economic value of medium- and high-income economies. We calculated the economic value of low-income economies by subtracting that of the medium- and high-income economies from that of all economies. However, there’s low confidence in current estimates for low-income economies because of insufficient and unreliable data collected in these countries.

The economic value possible by addressing each driver was calculated as follows:

• attrition: the total turnover multiplied by the cost of turnover per employee
• absenteeism: the estimated number of working days lost because of work-related ill health and nonfatal workplace injuries multiplied by the average daily pay
• presenteeism: the estimated number of productive days lost when employees are present at work but can’t be fully productive multiplied by the average daily pay
• attraction: the total premium by employees for employers with an above-average happiness score multiplied by the total turnover
• productivity: the increase in productivity associated with increased well-being multiplied by the average value of productivity
• retention: the total turnover multiplied by the benefit of retention

In addition to contributing to increased productivity at work, our calculations indicate that investing in employee health and well-being provides a positive opportunity for attracting and retaining talent. As noted in McKinsey research, employees facing mental-health and well-being challenges are four times more likely than others to want to leave their organizations . 8 Patrick Guggenberger, Dana Maor, Michael Park, and Patrick Simon, “ The State of Organizations 2023: Ten shifts transforming organizations ,” McKinsey, April 26, 2023.

Better health correlates with higher productivity across countries and workplace settings and is also strongly correlated with workforce participation  at all ages. 9 For more, see Dan Chisholm et al., “Scaling-up treatment of depression and anxiety: A global return on investment analysis,” Lancet Psychiatry , May 2016, Volume 3, Number 5; Clément S. Bellet, Jan-Emmanuel De Neve, and George Ward, “Does employee happiness have an impact on productivity?," Management Science , May 11, 2023; Miriam Dickinson, Kathryn Rost, and Jeffrey L. Smith, “The effect of improving primary care depression management on employee absenteeism and productivity: A randomized trial,” Medical Care , December 2004, Volume 42, Number 12; and “ Prioritizing health: A prescription for prosperity ,” McKinsey Global Institute, July 8, 2020. Every 1 to 3 percent increase in global workforce participation is worth a further $1.4 billion to $4.2 billion, 10 Assumes additional labor force is employed at the same unemployment rate and generates the same average GDP per employee as the current labor force. benefiting employees, their health, the societies in which they live, and government finances. 11 For more, see Lixin Cai, “The relationship between health and labour force participation: Evidence from a panel data simultaneous equation model,” Labour Economics , January 2010, Volume 17, Number 1.

To capture these economic benefits fully, employers need to move from a sole focus of protecting against incidental risk and illness to helping employees achieve more optimal health. This is particularly important when considering that employees move along a continuum of health over time and may draw upon different workplace resources throughout their employment with a company. Ultimately, a focus on improving health could lead to a virtuous circle of positive change, as employees gain health literacy, and employers in turn respond to employee health concerns.

To capture the economic benefits of good health fully, employers need to move from a sole focus of protecting against incidental risk and illness to helping employees achieve more optimal health.

Acting now also reduces future brand and business risk. In Australia, a lawsuit resulted in a fine for an organization that tolerated a toxic workplace culture. 12 For more, see Naomi Neilson, “Court Services Victoria fined $380k for ‘toxic’ workplace,” Lawyers Weekly, October 19, 2023. Recently, the European Union adopted the European Sustainability Reporting Standards, requiring organizations by law to report on working conditions such as working time, social dialogue, and work–life balance. As employees develop higher standards for what is tolerable in the workplace, more pushback and litigation may be possible.

Furthermore, investors such as asset managers, private equity companies, and venture capitalists are increasingly weighing environmental, social, and governance (ESG) considerations in their investment decisions. They are guided by ESG ratings released by various agencies and standards issued by the International Sustainability Standards Board.

Improving employee health and well-being involves more than just employers

We have highlighted practical examples of how employers can play a role in changing norms and catalyzing innovation around employee health and well-being. However, employers alone can’t complete this task. Employees, policy makers, and local governments will need to help.

Employees can play a role in their own health by taking advantage of the workplace resources that do exist and helping cultivate a community and culture of healthy practices among colleagues. They can make their desires known to employers as a means of holding leaders accountable for responding to the health needs and aspirations of their workforces. These might include benefits such as paid parental leave and caregiving support, which aim to help employees balance work and family responsibilities while tending to their own overall health and well-being.

Policy and decision makers may want to consider a variety of ways to protect and promote employee health. Possibilities include mandating upper limits on total working hours, health coverage paid by employers, and employee access to therapy and other psychological resources. 13 For more, see Richard Layard, “Wellbeing as the goal of policy,” LSE Public Policy Review , December 2021, Volume 2, Number 2. Enhancing standards and transparency could enable employees to make informed choices about their employment while also allowing policy makers to audit progress on a wider scale.

Through investment in public health (such as funding and grants), policy makers can encourage and enable employers to take employee health seriously and professionalize how they track the impact of their initiatives on employee health and well-being. Finally, policy and decision makers can lead by example in acting to promote their own employees’ health. This may be done in partnership with both private and other public sector employers, such as those that play a critical role in educating individuals about health—school systems, healthcare systems, and community programs—down to the city level.

City governments can play an important role in unlocking positive health  outcomes. Given that most large employers are concentrated in cities, there’s a unique opportunity for companies and employees to come together to set broader aspirations on health and identify targeted interventions to pursue jointly.

Employment can and does have a profound impact on health, both positive and negative. Adapting how and where people work to support optimal employee health could result in billions of employees and their families around the world living longer, higher-quality lives—and simultaneously benefiting their employers and the societies in which they live.

Jacqueline Brassey is a coleader of employee health at the McKinsey Health Institute (MHI) and a senior fellow in McKinsey’s Luxembourg office, Barbara Jeffery is a coleader of employee health at MHI and a partner in the London office, Lars Hartenstein is a global leader of MHI and a senior fellow in the Paris office, and Patrick Simon is a senior partner in the Berlin office.

The authors wish to thank Erica Coe, Aaron De Smet, Martin Dewhurst, Arne Gast, Brad Herbig, Anna Hextall, Ashini Kothari, Tom Latkovic, May Lim, Robyn Macrae, Dana Maor, Roxy Merkand, Hannah Mirman, Lucy Pérez, and Brooke Weddle for their contributions to this article.

This article was edited by Hannah Buchdahl, an associate editor in the Washington, DC, office.

Explore a career with us

Related articles.

Reframing employee health: Moving beyond burnout to holistic health

The secret to great health? Escaping the healthcare matrix

The trends defining the $1.8 trillion global wellness market in 2024

• Research article
• Open access
• Published: 28 April 2020

Associations of health literacy with socioeconomic position, health risk behavior, and health status: a large national population-based survey among Danish adults

• Majbritt Tang Svendsen 1 , 2 ,
• Carsten Kronborg Bak 3 ,
• Kristine Sørensen 4 ,
• Jürgen Pelikan 5 ,
• Signe Juul Riddersholm 6 ,
• Regitze Kuhr Skals 7 ,
• Rikke Nørmark Mortensen 7 ,
• Helle Terkildsen Maindal 8 ,
• Henrik Bøggild 7 , 9 ,
• Gitte Nielsen 1 &
• Christian Torp-Pedersen 7 , 9  

BMC Public Health volume  20 , Article number:  565 ( 2020 ) Cite this article

21k Accesses

185 Citations

21 Altmetric

Metrics details

Health literacy concerns the ability of citizens to meet the complex demands of health in modern society. Data on the distribution of health literacy in general populations and how health literacy impacts health behavior and general health remains scarce. The present study aims to investigate the prevalence of health literacy levels and associations of health literacy with socioeconomic position, health risk behavior, and health status at a population level.

A nationwide cross-sectional survey linked to administrative registry data was applied to a randomly selected sample of 15,728 Danish individuals aged ≥25 years. By the short form HLS-EU-Q16 health literacy was measured for the domains of healthcare, disease prevention, and health promotion. Adjusted multinomial logistic regression analyses were used to estimate associations of health literacy with demographic and socioeconomic characteristics, health risk behavior (physical activity, smoking, alcohol consumption, body weight), and health status (sickness benefits, self-assessed health).

Overall, 9007 (57.3%) individuals responded to the survey. Nearly 4 in 10 respondents faced difficulties in accessing, understanding, appraising, and applying health information. Notably, 8.18% presented with inadequate health literacy and 30.94% with problematic health literacy. Adjusted for potential confounders, regression analyses showed that males, younger individuals, immigrants, individuals with basic education or income below the national average, and individuals receiving social benefits had substantially higher odds of inadequate health literacy. Among health behavior factors (smoking, high alcohol consumption, and inactivity), only physical behavior [sedentary: OR: 2.31 (95% CI: 1.81; 2.95)] was associated with inadequate health literacy in the adjusted models. The long-term health risk indicator body-weight showed that individuals with obesity [OR: 1.78 (95% CI: 1.39; 2.28)] had significantly higher odds of lower health literacy scores. Poor self-assessed health [OR: 4.03 (95% CI: 3.26; 5.00)] and payments of sickness absence compensation benefits [OR: 1.74 (95% CI: 1.35; 2.23)] were associated with lower health literacy scores.

Conclusions

Despite a relatively highly educated population, the prevalence of inadequate health literacy is high. Inadequate health literacy is strongly associated with a low socioeconomic position, poor health status, inactivity, and overweight, but to a lesser extent with health behavior factors such as smoking and high alcohol consumption.

Peer Review reports

During the past decades, an unsolicited socioeconomic divide has intensified in Europe, and the population is increasingly challenged with the growing complexity of the modern healthcare system and the rising expectations of the individuals to participate actively in promoting and maintaining their health [ 1 , 2 , 3 ]. The socioeconomic divide is not only about income. It remains a complex phenomenon involving health status which is also driven by education, employment status, and migration background [ 1 ]. Health literacy, a multidimensional concept covering the interacting capacities of the individuals and the systems to meet the complex demands of health in modern society, has been recognized as a key factor for reducing health inequality [ 3 ]. Despite the enormous implications of inadequate health literacy, knowledge of health literacy in the general population and how health literacy impacts health behavior and health status remain scarce [ 4 ].

In recent years, health literacy has gained importance on the European health agenda, and there has been a growing interest in the concept, accompanied by the increased emphasis on the role and responsibilities of the individual in health and healthcare [ 5 , 6 , 7 ]. Health literacy research has developed from investigations of general literacy or functional health literacy to expanded conceptual frameworks encompassing citizens´ knowledge, motivation, and competences to access, understand, appraise, and apply health information in everyday life to make decisions regarding healthcare, disease prevention, and health promotion [ 3 , 8 ]. Comprehensive measurement tools reflecting the multidimensional concept of health literacy have been developed, and a health literacy survey performed by the European Health Literacy Consortium (HLS-EU) in eight member states of the EU in 2011 suggested that nearly half of the European population included in this survey had limited or suboptimal health literacy competences [ 3 , 6 , 8 , 9 , 10 , 11 ].

In modern society, individuals with higher health literacy have advantages in health compared with those who have lower health literacy. Several studies have shown that limited health literacy is associated with adverse health-related outcomes, such as increased mortality [ 12 , 13 ], more hospitalizations [ 14 , 15 ], less use of preventive care [ 16 ], less efficient use of access to health services [ 17 ], and the lack of the ability to make informed health-related decisions [ 18 ] and follow medical instructions [ 19 ]. Limited health literacy has also been demonstrated in European countries to follow a social gradient [ 9 , 11 ], and the increasing demands on the individual seem inadvertently to raise social disparity in health as it favors those with adequate health literacy [ 20 ].

Beyond HLS-EU, the majority of previous research on health literacy is based on smaller samples, focusses on functional or specific dimensions of health literacy, or is centered on a specific population or patient groups [ 2 , 21 , 22 , 23 , 24 , 25 ]. There is a growing need to understand how health literacy, as a dynamic outcome of sociodemographic determinants as well as individual and environmental factors, manifests itself in the interaction between individuals and the modern health society at the population level [ 26 ]. Comprehensive knowledge of health literacy in the general population is needed for guiding health systems and health organizations in their response to the needs of the citizens.

To the best of our knowledge, the present study constitutes the largest national population-based study on health literacy using the short form (HLS-EU-Q16) of the comprehensive European Health Literacy Survey Questionnaire. Specifically, the study aims to examine the distribution of health literacy levels and investigate the association of health literacy with socioeconomic position, health risk behavior, and health status within a large representative sample of 15, 728 Danish adults.

Study design and sampling

A national, cross-sectional, representative survey with a random stratified sampling design was conducted involving Danish residents aged 25 years and older. A minimum age of 25 years was set in order to obtain a more stable life situation of participants with respect to socioeconomic indicators. Participants were drawn from the Danish Civil Registration System, and the sample was stratified according to age, gender, and geographical location (postal codes), leading to a probability of inclusion that was proportional to population size and density. Postal codes were included to ensure representative sampling from urban and rural residential areas. The sample was individually linked with information obtained from a variety of Danish administrative registries.

Study population

Between December 2016 and February 2017, 15,728 citizens were contacted; of these, 9007 were willing to participate, leading to a response rate of 57.3%. Participants were recruited using an electronic email system administered by public authorities including a link to a self-administered web-based questionnaire. The wording of the material was provided only in Danish and was not specifically targeted to low-literate individuals. A random part of the study population ( N  = 1082) were contacted by telephone to increase representation.

Overall, the survey instrument comprised 28 items divided into the following categories: (1) health literacy, (2) health behavior, (3) health risk indicators, and (4) perceived self-assessed health. Data on demographic characteristics, socioeconomic indicators, and health status were obtained from nationwide administrative registries.

• Health literacy

The HLS-EU-Q16 questionnaire developed by the HLS-EU Consortium for measuring health literacy in the general population was applied. Extensive information on the development and pre-testing is described elsewhere [ 4 , 8 ]. To ensure cross-cultural validity, translation and adaption of the Health Literacy Survey Questionnaire followed a standardized procedure completed by K. Sørensen, H. Maindal, and colleagues (Unpublished material. Please contact the third author, K. Sørensen, for further information). The short form of the HLS-EU-Q used in this present study consisted of 16 items measuring health literacy within the three domains of healthcare, disease prevention, and health promotion. Within these domains, questions focus on perceived difficulties or ease of accessing, understanding, appraising, and applying health information [ 4 , 27 ]. Each of the items had response categories on a 4-point Likert scale ranging from “very easy” to “very difficult.” Health literacy scores ranging from 0 to 16 were calculated by dichotomizing answer categories. “Very easy” and “easy” were given a score of 1, whereas “very difficult” and “difficult” were given a score of 0. The total health literacy score was classified according to three levels, namely inadequate, (0–8), problematic (9–12), and adequate (13–16). A “don’t know” answer option was provided in the telephone interviews and was used only when stated by the respondent spontaneously. The “don’t know” category was coded as a missing value. Health literacy scores were computed only for respondents who had answered a minimum of 14 of all health literacy items [ 4 ], corresponding to a total of 8455 (93.9%) respondents.

Demographic and socioeconomic characteristics

In Denmark, all citizens are identified with a unique civil registration number that enables individual linkage of information across Danish administrative registries. We retrieved information on age (year of birth), gender (male, female), origin (Danish, immigrant, descendant of immigrant), marital status (married/partnership, divorced, unmarried), and residence from the Danish Civil Personal Registration Registry [ 28 ]. Information on education, described using the International Standard Classification of Education (ISCED) nomenclature, was obtained from the Danish Education Registers [ 29 ]. Educational level was grouped into categories based on highest completed education level one year before the survey completion. Basic school, primary educations, lower secondary (ISCED 0–2), upper secondary, and vocational educations (ISCED 3–4) reflect the first and second education levels. Medium length educations including short and medium length tertiary, and bachelor’s educations (ISCED 5–6), and higher length education containing master’s level and PhD-level educations (ISCED 7–8) reflect the third and fourth levels of education. Annual income was obtained from the Danish Income Register [ 30 ] and was calculated as the mean of the respondents’ individual average income in the last three years before the survey completion. Based on the average income of 312,000 DKK [ 31 ] in the general Danish population (7.45 DKK equals 1 €), income was divided in two groups: below average (< 312,000 DKK) or above average (≥312,000 DKK). Finally, information on social benefits was obtained from a registry administered by the Danish Labour Market Authority (the DREAM database) [ 32 ]. Maternity leave compensation and sickness compensation were not considered social benefits.

Health behavior, health risk indicators, and health status

Different measures of health behavior including smoking habits, alcohol consumption, and physical activity were included from the survey. Smoking was classified as “daily smoker”, “infrequent smoker”, “former smoker”, or “never smoker”. Alcohol consumption was measured according to official national recommendations stated by the Danish Health Authorities. Weekly alcohol consumption above 14 drinks for men and seven drinks for women is considered high-risk behavior. Physical activity was classified according to daily physical activity level as “sedentary behavior”, “light activities”, “moderate training”, or “hard training” within the last year before the survey completion.

Self-reported height and weight were obtained to allow calculation of body mass index (BMI). BMI was regarded as a health risk indicator and classified as underweight (BMI ≤18.5 kg/m 2 ), normal (BMI 18.5–24.9 kg/m 2 ), overweight (BMI 25–30 kg/m 2 ), or obese (BMI ≥30 kg/m 2 ). Self-assessed general health stated as “How would you judge your current state of health?” included four categories ranging from very good to very poor. Information on sickness absence compensation benefits one year from the survey completion was obtained from the Dream database.

Statistical analysis

The categorical variables are presented using percentages and the continuous variables using medians with the 25th (Q1) and 75th (Q3) percentiles. Chi-square and Kruskal-Wallis tests were performed to test differences between respondents and non-respondents as well as differences between health literacy groups. Internal consistency and reliability of health literacy levels were assessed calculating Cronbach’s alpha coefficients based on Pearson correlations. The level of perceived difficulty was calculated in terms of health literacy competences and domains (sum score/number of items) using means with standard deviations and medians with the 25th (Q1) and 75th (Q3) percentiles. We estimated individual odds ratios (OR) between levels of health literacy (outcome variable) and demographic and socioeconomic measures, health risk behavior, and health status (exposure variables) using both univariable and multivariable multinomial logistic regression analyses compared to odds of adequate health literacy. Demographic and socioeconomic factors, including gender, age, migration background, civil status, education, and income were included as potential confounders in the regression analyses. We tested for interaction using the likelihood ratio test. The reference groups in the models were consistently chosen as the more prevalent ones. Sensitivity analyses were performed to test the consistency of the general health literacy score according to the method of distribution of survey material, including either web-based or telephone interview-based. A two-sided P -value < 0.05 was considered statistically significant. Statistical analyses were performed by the statistical software packages SAS version 9.4 (SAS Institute Inc., Cary, NC, USA), and R statistical software package, version 3.3.2 (R Development Core Team) [ 33 ].

According to Danish legislation (Law on ethical conduct in health science, Lovtidende:§14, section 2) application for ethical approval is not required for questionnaire-based and register-based studies [ 34 ]. The written provision of information about the survey communicated to participants, including information on data retrieval along with the voluntary completion by participants, constituted an implied consent. The data collection was approved by the Danish Data Protection Agency (j.no: 2008-58-0028) and was conducted in accordance with the Helsinki Declaration.

A total of 9007 residents (57.3%) participated in the survey. The median age of respondents was 53.2 years [Q1: 42.3, Q3: 63.7], slightly more women (54.5%) than men participated, and 7.2% were immigrants or descendants of immigrants. The majority of respondents were married (64.0%), approximately one-third (28.9%) had an annual income below the national average income, and more than one-fifth (21.9%) of participants received social welfare payments within the last year from the survey completion. Characteristics of respondents and non-respondents according to sociodemographic indicators are presented in Table 1 . The sociodemographic characteristics were distributed differently between respondents and non-respondents ( P < 0.001 ), and especially the youngest age group was underrepresented.

Distribution of health literacy within the population

The median health literacy score of respondents was 13.0 [ Q1: 11.0, Q3: 15.0 ] on the 16-item scale. The health literacy scale was further classified into three levels described previously. Overall, 8.2% ( N = 692 ) of the study population had inadequate health literacy, 30.9% ( N = 2616) had problematic health literacy, and 60.9% ( N = 5147 ) showed adequate health literacy. For individuals categorized within the inadequate health literacy category, the median health literacy score was 7.0 [ Q1: 6.0, Q3: 8.0] . Individuals within the problematic or adequate health literacy category presented with a median score of 11.0 [ Q1: 10.0, Q3: 12.0] and 14.0 [ Q1: 14.0, Q3: 16.0] , respectively. The Cronbach alpha coefficient indicated a high internal consistency of the assembled HLS-EU-Q16 questionnaire (α = 0.90) and good or acceptable internal consistency within each of the three health domains (healthcare: α = 0.82, disease prevention: α = 0.74, health promotion: α = 0.75). For the competences of accessing, understanding, appraising, and applying information, the median score per item was highest for accessing health information (Median: 3.3 [ Q1: 3.0, Q3: 3.5 ]) and lowest for appraising health information (Median: 2.7 [ Q1: 2.3, Q3: 3.0 ]). When comparing all competences over the health domains of healthcare, disease prevention, and health promotion, the median score per item was highest within the domain of healthcare (Median: 3.2 [ Q1: 2.9, Q3: 3.4 ]) and lowest in the domain of disease prevention (Median: 2.8 [ Q1: 2.6, Q3: 3.2 ]) (Table 2 ). The items that were rated least and that the respondents perceived most difficult were: ‘ judge if the information on health risks in the media is reliable’ ( N = 4668 , proportion experiencing difficulty: 62.3%) and ‘ judge when you may need to get a second opinion from another doctor ’ ( N = 4173 , proportion experiencing difficulty: 53.3%). In contrast, respondents experienced the least difficulty in relation to: ‘understand your doctor’s or pharmacist’s instruction on how to take a prescribed medicine’ ( N = 169 , proportion experiencing difficulty: 2.0%) and ‘understand health warnings about behavior such as smoking, low physical activity and drinking too much’ ( N = 171 , proportion experiencing difficulty: 2.0%).

Overall, health literacy varied between subgroups according to demographic and socioeconomic characteristics within the population (Table 3 ). Men, younger aged individuals (25–44 years old), non-ethnic Danes, unmarried individuals, people with a low education level, income below the national average, and individuals receiving public benefits reported statistically significantly lower levels of health literacy ( P < 0.001 ). The sensitivity analysis concerning the method of distribution of survey material showed that the general health literacy score was slightly lower among interview-based respondents (Median: 12.0 [ Q1: 11.0, Q3: 14.0 ]) compared to web-based respondents (Median: 13.0 [ Q1: 11.0, Q3: 15.0 ]). Socioeconomic characteristics and general health literacy score of interview-based respondents compared to web-based respondents are available in data supplement (Table S1 ).

Associations of health literacy with demographic and socioeconomic characteristics

Figure 1 shows odds ratios and confidence intervals for both univariable and multivariable logistic multinomial regression analysis estimating the odds of having inadequate health literacy compared to adequate health literacy, and further problematic health literacy compared to adequate health literacy. Males had significantly higher odds of inadequate [Adjusted OR: 2.30 (95% CI: 1.91; 2.79)] and problematic [Adjusted OR: 1.46 (95% CI: 1.31; 1.62)] health literacy compared to women. The odds of experiencing both inadequate and problematic health literacy diminished with higher age. The socioeconomic indicators, adjusted for covariates, showed that migration background, education, income, and transfer of public benefits were statistically significantly associated with health literacy. Individuals with high school or vocational, medium or high education had significantly lower odds of inadequate and problematic health literacy compared to an individual with only primary or basic education (ISCED level 0–4) as highest completed education level. Individuals with an annual income below the national average had higher odds of inadequate and limited health literacy. The adjusted multinomial logistic regression model showed that individuals receiving social benefits tend to have lower general health literacy scores compared to individuals who are self-supporting or receiving retirement benefits.

Associations of demographic and socioeconomic characteristics with health literacy. Forrest plot presenting multivariable multinomial logistic regression model describing odds ratios (OR), with corresponding 95% confidence intervals (CI), of inadequate and problematic health literacy compared to adequate health literacy. Unadjusted and model adjusted for all covariates. Statistically significant P -values ( P < 0.05 ) are flagged with star symbols (*)

Associations of health literacy with health behavior and health risk indicators

The association between health literacy and health behavior (Fig. 2 ) showed no associations of inadequate health literacy with smoking and alcohol consumption above national recommendations in the adjusted models. Alcohol consumption “never above recommendations” was associated with higher odds of inadequate health literacy . Significant associations were found between health literacy and physical activity. Individuals reporting sedentary behavior had higher odds of lower general health literacy scores compared to individuals reporting light activities as a physical behavior pattern. Contrarily, individuals reporting moderate exercise behavior had lower odds of both inadequate and problematic health behavior. Significant associations in both univariable and multivariable models with the long-term health risk indicator BMI were found. Obesity (BMI > 30) was associated with lower general health literacy scores [Inadequate health literacy: Adjusted OR: 1.78 (95% CI: 1.39; 2.28), problematic health literacy: Adjusted OR: 1.32 (95% CI: 1.14; 1.54)]. Significant associations with health literacy and overweight (BMI > 25) were also found, demonstrating that individuals with higher health literacy scores tend to have a normal BMI.

Associations of health risk behavior and health status with health literacy. Forrest plot presenting multivariable multinomial logistic regression model describing odds ratios (OR), with corresponding 95% confidence intervals (CI), of inadequate and problematic health literacy compared to adequate health literacy. Unadjusted and model adjusted for all covariates. Statistically significant P-values ( P < 0.05 ) are flagged with star symbols (*)

Health literacy and health status

A large proportion of respondents reported their health as good (64.7%) or very good (19.0%), compared to individuals reporting their health as poor (14.9%) or very poor (1.4%). A strong association between both inadequate [Adjusted OR: 4.03 (95%CI: 3.26; 5.00)] and problematic [Adjusted OR: 1.99 (95%CI: 1.71; 2.31)] health literacy with poor and very poor self-assessed health was found (Fig. 2 ), demonstrating that individuals reporting poor or very poor self-assessed health are more likely to have lower health literacy . Payment of sickness absence compensation benefits was used as a proxy for health status. In both univariable and multivariable regression models, significant associations between lower health literacy scores and payments of sickness absence compensation benefits were found (Fig. 2 ) .

Nearly four in 10 of the Danish population reports having difficulties managing and meeting the complex demands of health. Gender, age, ethnicity, education, income, and transfer of public payments were all associated with health literacy levels. Health literacy was strongly associated with physical activity, body weight, self-assessed health, and payments of sickness absence compensation, but not with smoking and to a lesser extent with alcohol consumption.

The present study is the first nationwide representative population study of health literacy in Denmark using the HLS-EU-Q16. Previous studies on health literacy were based on regional data and focused on specific dimensions of health literacy [ 23 ] or specific target groups [ 24 , 25 ]. Health literacy was measured on an international validated instrument, and to our knowledge, this is the largest sample applied in a single study using the HLS-EU-Q16. The study provided information on an individual’s self-perceived competences necessary for them to make empowered and informed decisions regarding health, reflected by the competences of accessing, understanding, appraising, and applying information in the domains of healthcare, disease prevention, and health promotion. The findings of this study indicate that the Danish population perceives the least difficulties within the domain of healthcare and that nearly every second respondent faced problematic or inadequate health literacy. This is consistent with findings from eight other European countries where 29% and 62% of the population (average: 47.6%) were categorized as having limited (inadequate or problematic) health literacy [ 9 , 11 ]. The health literacy survey of Dutch adults also showed that the mean score per item (over all domains) was lowest for appraising information, which is also consistent with findings in the Danish sample [ 35 ]. The general HL-score varied considerably between participating countries in the HLS-EU project.

Though the average health literacy score of the Danish population was within the range of other European countries, it seems that the general Danish population perceives lesser difficulties compared to the majority of other participating countries in the HLS-EU project, but this may reflect that the 16-item scale may have an overrepresentation of easier items than the 47-item scale. Given that the present study was conducted in a country with a universal healthcare system and multiple policies promoting health efforts among the general population, it is disturbing that a substantial part of the population experiences difficulties in making empowered and informed decisions regarding health. Besides the present study, another study using the HLS-EU-Q16 instrument on a national sample has been identified [ 36 ]. High internal consistency was found in both the Danish and the Israeli population, indicating that the 16-item scale is reliable and can be used instead of the larger 47-item scale. Inadequate health literacy competences were less pronounced in the Israeli population compared to both the Danish sample and the other eight European countries included in the HLS-EU project [ 9 , 11 ]. Our results may reflect the increasing complexity of being a health literate individual navigating in modern health society.

The analyses indicate that lower health literacy is associated with lower socioeconomic position, which is in accordance with existing literature [ 9 , 11 , 35 , 37 ]. The socioeconomic gradient in health literacy found in the present study is similar to the results of the HLS-EU survey in which health literacy is dependent on socioeconomic indicators such as social status, education, and financial resources [ 9 , 11 ]. The reason for this association is not well understood and still needs to be explored prospectively, as it may be critical in understanding the relationship to the socioeconomic divide and health inequalities. Interestingly, the present study indicated lower health literacy among the younger population groups, which is in contrast to results from the European countries of the HLS-EU study [ 9 , 11 ]. Mixed results regarding the association of age and health literacy have previously been discussed by van der Heide et al., who found that age is associated with lower health literacy within specific health literacy dimensions [ 35 ]. A previous survey on health literacy performed in Denmark using another health literacy questionnaire (HLQ™) also found that individuals aged 25–45 years perceive more difficulties with health literacy compared to older individuals [ 23 ]. They suggested that older individuals may have strengthened capabilities as a result of a more established relationship with their general practitioner and longer experience in navigating in the healthcare sector. The fact that the elderly population in Denmark is relatively highly educated could also contribute to the contradicting results of the relation of age to health literacy.

Another finding of this study showed that men perceive significantly more difficulties with health literacy compared to women with more than twice the odds of problematic health literacy. No consistent pattern between gender and health literacy has been reported in the literature. However, the finding of modest differences between men and women depending on specific health literacy dimensions has previously been reported [ 23 , 35 ]. Contrarily, the present study showed a strong association between health literacy and gender independent of specific health literacy dimensions.

Similar to the HLS-EU survey, the present study showed a positive association between health literacy and health-related behavior in the form of smoking, alcohol consumption, physical activity level, and body weight [ 9 , 11 ]. Yet, when controlling for socioeconomic factors; only physical activity and body weight were associated with health literacy. These results imply that although adequate health literacy competences help gain access to appropriate sources of health information critical for the adoption of health behavior, they are not the only factors influencing health behavior. In accordance with existing literature, a strong and positive association of health literacy with self-assessed health was found [ 9 , 36 ]. The present study supports those results from the other eight European countries of the HLS-EU study that found an association between health literacy and self-assessed health beyond sociodemographic and behavioral measures. Longitudinal data are needed to understand the exact relationship and pathways in which these two variables interact with each other. To our knowledge, no other studies have used payments of sickness absence compensation as a proxy for health status in relation to health literacy. A clear association was found, indicating that health literacy is strongly associated with health status. The significant relation of health literacy with health risk indicators and health status implies consistent attention to the risk of non-communicable diseases (NCDs) within the health system and society at large. The study reveals a triple burden for people with limited health literacy as there is a strong association between being poor, having poor health status, and poor levels of health literacy. These results are in line with previous European studies [ 11 ].

The strong socioeconomic divide remains a barrier for people to achieve and maintain good health, also in a welfare state such as Denmark. The divide calls for action in terms of targeted interventions that serve the specific need of people with insufficient and limited health literacy. While almost 40% of the population is challenged in terms of accessing, understanding, appraising, and applying information to manage their health in everyday life, universal health literacy precautions are recommended to facilitate a better match between people’s needs and the services and information offered through the health system [ 38 ].

Strengths and limitations

To our knowledge, the present study is the largest health literacy study of individual respondents using the HLS-EU-Q16 within a single country. A clear strength of this study is, therefore, the large sample size which allowed us to perform wide-ranging and robust investigations of health literacy across demographic, socioeconomic, and health-related indicators. Secondly, the use of nationwide administrative registries is an important strength. Thereby, we were able to adjust our analyses for a wide range of socioeconomic factors and other potential confounders. Another strength of using administrative registries is that we did not rely predominantly on self-reported information concerning socioeconomic indicators. Self-reported data could otherwise result in imprecision and biased results. Thirdly, a major advantage is the use of a short validated health literacy measurement tool, which is relatively easy to administer and allows for comparison with other population groups. However, a drawback of self-reporting questionnaires is that it requires a certain level of literacy and motivation to participate. We suspect that the most vulnerable population groups may not have participated in the study, but the study had the possibility to describe non-responders concerning their socioeconomic status.

Given that non-responders were differently distributed across sociodemographic determinants compared to responders, sensitivity analyses have been performed to evaluate the potential influence of selection bias. We examined the effect of having a web-based reporting tool. The general health literacy score was slightly lower among telephone interview-based respondents compared to web-based respondents, which at least partly can be explained by differences in distributions of the two samples of some factors related to health literacy like gender, age, education, and social benefits. Therefore, overestimation of the general population’s health literacy is possible when using a web-based collection of information. Further studies collecting information on health literacy in different ways or settings are recommended. A limitation is the cross-sectional design that precludes any causal conclusions. Longitudinal studies may provide a better basis to understand these aspects, especially regarding how health literacy may act as a mediator between social determinants and health. Further, the odds estimated in the present study could possibly exaggerate the true effect [ 39 ].

Implications

The evidence from the present study is important for shaping future health and healthcare in Denmark and other welfare societies. To bridge the gap of inequality, solutions need to be developed tackling the triple burden related to health that some population groups encounter. A systematic, organizational change using personalized approaches is required to overcome the barriers. Collaborative efforts are needed within all sectors regarding policy, research, practice, and education.

Despite a relatively educated population in Denmark, the prevalence of inadequate and problematic health literacy is high in our study. Notably, males, younger individuals, immigrants, individuals with basic education or income below the national average, and individuals receiving social benefits had a substantially higher risk of inadequate health literacy. An independent association between low socioeconomic position and low health literacy was demonstrated. Likewise low health literacy was associated with poor self-reported health, receiving sickness benefits, and with inactivity, but not with smoking and alcohol consumption. Finally, low health literacy was associated with overweight. A significant proportion of the general population faces serious problems in managing health demands. These findings emphasize that universal health literacy precautions are needed to facilitate a better match between people’s needs and the services and information offered through the Danish health system.

Availability of data and materials

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Abbreviations

Health Literacy Survey

Health Literacy Questionnaire

Body mass index

25th percentile

75th percentile

Confidence intervals

Standard deviation

Non-communicable diseases

Danish kroner

International Standard Classification of Education

Forster, M., Llena Nozal, A., & Thevenot C.. Understanding the socio-economic divide in Europe. OECD. 2017. http://oe.cd/cope-divide-europe-2017.

Kickbusch I, Jürgen MP, Franklin A, Agis DT. Health literacy : the solid facts. World Heal Organ World Heal Commun Assoc WHO Reg Off Eur 2013. http://www.euro.who.int/__data/assets/pdf_file/0008/190655/e96854.pdf.

Sørensen K, Van Den Broucke S, Fullam J, Doyle G, Pelikan J. Health literacy and public health: a systematic review and integration of definitions and models. BMC Public Health. 2012;12:80. https://doi.org/10.1186/1471-2458-12-80 .

Article   PubMed   PubMed Central   Google Scholar  

Pelikan JM, Ganahl K. Measuring health literacy in general populations: primary findings from the HLS-EU Consortium’s health literacy assessment effort. Stud Health Technol Inform. 2017;240:34–59. https://doi.org/10.3233/978-1-61499-790-0-34 .

Article   PubMed   Google Scholar  

Chinn D. Critical health literacy: a review and critical analysis. Soc Sci Med. 2011;73:60–7. https://doi.org/10.1016/j.socscimed.2011.04.004 .

Nutbeam D. Health literacy as a public health goal: a challenge for contemporary health education and communication strategies into the 21st century. Health Promot Int. 2006;15:259–67. https://doi.org/10.1093/heapro/15.3.259 .

Article   Google Scholar  

Nutbeam, D. Kickbusch I. Advancing health literacy: a global challenge for the 21st century. Helth Promot Int 2000;15:183–184. https://doi.org/10.1093/heapro/15.3.183.

Sørensen K, Van Den Broucke S, Pelikan JM, Fullam J, Doyle G, Slonska Z, et al. Measuring health literacy in populations: illuminating the design and development process of the European health literacy survey questionnaire (HLS-EU-Q). BMC Public Health. 2013. https://doi.org/10.1186/1471-2458-13-948 .

Pelikan, Jürgen M, Röthlin, F, Ganahl K. HLS-EU CONSORTIUM (2012): Comparative report of health literacy in eight EU member states. The European health literacy survey HLS-EU 2009. http:/www.Health-literacy.EU.

Osborne RH, Batterham RW, Elsworth GR, Hawkins M, Buchbinder R. The grounded psychometric development and initial validation of the health literacy questionnaire (HLQ). BMC Public Health. 2013;13. https://doi.org/10.1186/1471-2458-13-658 .

Sørensen K, Pelikan JM, Röthlin F, Ganahl K, Slonska Z, Doyle G, et al. Health literacy in Europe: comparative results of the European health literacy survey (HLS-EU). Eur J Pub Health. 2015;25:1053–8. https://doi.org/10.1093/eurpub/ckv043 .

Peterson PN, Shetterly SM, Clarke CL, Bekelman DB, Chan PS, Allen LA, et al. Health literacy and outcomes among patients with heart failure. JAMA. 2011;305:1695–701. https://doi.org/10.1001/jama.2011.512 .

Article   CAS   PubMed   PubMed Central   Google Scholar  

Sudore RL, Yaffe K, Satterfield S, Harris TB, Mehta KM, Simonsick EM, et al. Limited literacy and mortality in the elderly: the health, aging, and body composition study. J Gen Intern Med. 2006;8:806–12. https://doi.org/10.1111/j.1525-1497.2006.00539.x .

Baker DW, Gazmararian JA, Williams MV, Scott T, Parker RM, Green D, et al. Functional health literacy and the risk of hospital admission among Medicare managed care enrollees. Am J Public Health. 2002. https://doi.org/10.2105/AJPH.92.8.1278 .

Wu JR, Holmes GM, Dewalt DA, Macabasco-O’Connell A, Bibbins-Domingo K, Ruo B, et al. Low literacy is associated with increased risk of hospitalization and death among individuals with heart failure. J Gen Intern Med. 2013;8:1278–83. https://doi.org/10.1007/s11606-013-2394-4 .

Scott TL, Gazmararian JA, Williams MV, Baker DW. Health literacy and preventive health care use among Medicare enrollees in a managed. Care. 2002;40:395–404 http://www.jstor.org/stable/3767750 .

Cho YI, Lee SYD, Arozullah AM, Crittenden KS. Effects of health literacy on health status and health service utilization amongst the elderly. Soc Sci Med. 2008;66:1809–16. https://doi.org/10.1016/j.socscimed.2008.01.003 .

Schillinger D, Grumbach K, Piette J, Wang F, Dennis Osmond M, Daher C, et al. Association of Health Literacy with Diabetes Outcomes. JAMA. 2002;288:475–82. https://doi.org/10.1001/jama.288.4.475 .

Berkman ND, Sheridan SL, Donahue KE, Halpern DJ, Crotty K. Low Health Literacy and Health Outcomes: An Updated Systematic Review 2011:97–104. doi: https://doi.org/10.7326/0003-4819-155-2-201107190-00005 .

Protheroe J, Nutbeam DRG. Health literacy: a necessity for increasing participation in health care. Br J Gen Pract. 2009;59:721–3. https://doi.org/10.3399/bjgp09X472575 .

Jordan JE, Osborne RH, Buchbinder R. Critical appraisal of health literacy indices revealed variable underlying constructs, narrow content and psychometric weaknesses. J Clin Epidemiol 2011. doi: https://doi.org/10.1016/j.jclinepi.2010.04.005 .

Ishikawa H, Yano E. Patient health literacy and participation in the health-care process. Health Expect. 2008. https://doi.org/10.1111/j.1369-7625.2008.00497.x .

Bo A, Friis K, Osborne RH, Maindal HT. National indicators of health literacy: ability to understand health information and to engage actively with healthcare providers - a population-based survey among Danish adults. BMC Public Health. 2014. https://doi.org/10.1186/1471-2458-14-1095 .

Aaby A, Friis K, Christensen B, Rowlands G, Maindal HT. Health literacy is associated with health behaviour and self-reported health: a large population-based study in individuals with cardiovascular disease. Eur J Prev Cardiol. 2017. https://doi.org/10.1177/2047487317729538 .

Friis K, Vind BD, Simmons RK, Maindal HT. The relationship between health literacy and health behaviour in people with diabetes: a Danish population-based study. J Diabetes Res. 2016. https://doi.org/10.1155/2016/7823130 .

Kickbusch I. Responding to the health society. Health Promot Int. 2007:89–91. https://doi.org/10.1093/heapro/dam014 .

Rasch G. Probabilistic models for some intelligence and attainment tests (Expanded ed.). Chicago: Univ Chicago Press; 1980.

Pedersen CB. The Danish civil registration system. Scand J Public Health. 2011;39:22–5. https://doi.org/10.1177/1403494810387965 .

Jensen VM, Rasmussen AW. Danish Education Registers. Scand J Public Health. 2011;39(Suppl):91–4. https://doi.org/10.1177/1403494810394715 .

Baadsgaard M, Quitzau J. Danish registers on personal income and transfer payments. Scand J Public Health. 2011. https://doi.org/10.1177/1403494811405098 .

Statistics Denmark: https://www.statistikbanken.dk/INDKP101. Accessed 3 Sep 2019.

Hjollund NH, Larsen FBAJ. Register-based follow-up of social benefits and other transfer payments: accuracy and degree of completeness in a Danish interdepartmental administrative database compared with a population-based survey. S Scand J Public Heal. 2007;35:497–502.

R Core Team (2008), R: a language and environment for statistical computing. R Project for Statistical Computing, Vienna n.d. https://www.r-project.org/ .

Danish Ministry of Health. Lov om videnskabsetisk behandling af sundhedsvidenskabelige forskningsprojekter (Law on ethical conduct in health science) 2011;Lovtidende:§14, section 2.

van der Heide I, Rademakers J, Schipper M, Droomers M, Sørensen K, Uiters E. Health literacy of Dutch adults: a cross sectional survey. BMC Public Health. 2013. https://doi.org/10.1186/1471-2458-13-179 .

Levin-Zamir D, Baron-Epel OB, Cohen V, Elhayany A. The Association of Health Literacy with health behavior, socioeconomic indicators, and self-assessed health from a National Adult Survey in Israel. J Health Commun. 2016. https://doi.org/10.1080/10810730.2016.1207115 .

Duong TV, Aringazina A, Baisunova G, Nurjanah, Pham TV, Pham KM, et al. Measuring health literacy in Asia: Validation of the HLS-EU-Q47 survey tool in six Asian countries. J Epidemiol. 2017. https://doi.org/10.1016/j.je.2016.09.005 .

DeWalt DA, Leigh Callahan MF, Hawk VH, Kimberly Broucksou RA, Ashley Hink M, Rudd R, et al. Health Literacy Universal Precautions Toolkit 2010. https://www.ahrq.gov/sites/default/files/wysiwyg/professionals/quality-patient-safety/quality-resources/tools/literacy-toolkit/healthliteracytoolkit.pdf.

Davies HT, Crombie IK, Tavakoli M. When can odds ratios mislead? BMJ. 1998. https://doi.org/10.1136/bmj.316.7136.989 .

Download references

Acknowledgements

Our appreciation goes to individuals responding to the questionnaire.

This work was supported by Helsefonden (grant number: 15-B-0156), the Health Research Foundation of the North Denmark Region, and by the Maria Pedersen and Jensine Heiberg Foundation. The institutions did not influence the design or conduct of the study or its management, analysis, and interpretation of the data, and neither did it review or approve the manuscript before submission.

Author information

Authors and affiliations.

Department of Cardiology, North Denmark Regional Hospital, Bispensgade 37, 9800, Hjørring, Denmark

Majbritt Tang Svendsen & Gitte Nielsen

Centre for Clinical Research, North Denmark Regional Hospital / Clinical Institute of Medicine, Aalborg University , Hjørring, Denmark

Majbritt Tang Svendsen

Department of Research and Development, University College South, Kolding, Denmark

Carsten Kronborg Bak

Global Health Literacy Academy, Risskov, Denmark

Kristine Sørensen

Austrian Public Health Institute, Vienna, Austria

Jürgen Pelikan

Department of Anesthesia and Intensive Care, Aalborg University Hospital, Aalborg, Denmark

Signe Juul Riddersholm

Epidemiology and Biostatistics, Aalborg University Hospital, Aalborg, Denmark

Regitze Kuhr Skals, Rikke Nørmark Mortensen, Henrik Bøggild & Christian Torp-Pedersen

Department of Public Health, Section for Health Promotion and Health Services, Aarhus University, Aarhus, Denmark

Helle Terkildsen Maindal

Public Health and Epidemiology Group, Department of Health Science and Technology, Aalborg University, Aalborg, Denmark

Henrik Bøggild & Christian Torp-Pedersen

You can also search for this author in PubMed   Google Scholar

Contributions

MTS, CTP, KS, CKB, HB, and GN participated in the design. CTP, GN, HB, and CKB handled founding and supervision. MTS was responsible for the collection of data. MTS analysed and interpreted the data together with CTP, SJR, HB, RKS, and RNM. MTS and KS drafted the manuscript. CTP, JP, HTM, HB, SJR, KS, RNM, RKS, GN, and CKB made critical revision of the manuscript for key intellectual content. All authors read and approved the final version of the manuscript.

Corresponding author

Correspondence to Majbritt Tang Svendsen .

Ethics declarations

Ethics approval and consent to participate.

According to Danish legislation (Law on ethical conduct in health science, Lovtidende:§14, section 2) application for ethical approval is not required for questionnaire-based and register-based studies [ 33 ]. The written provision of information about the survey communicated to participants, including information on data retrieval along with the voluntary completion by participants, constituted an implied consent. The data collection was approved by the Danish Data Protection Agency (j.no: 2008-58-0028) and was conducted in accordance with the Helsinki Declaration.

Consent for publication

Not applicable

Competing interests

MTS reports grants from the Helsefonden, the Health Research Foundation of the North Denmark Region, and the Maria Pedersen and Jensine Heiberg Foundation, during the conduct of the study. CTP reports grants and speaker honorarium from Bayer. HB is a member of the editorial board in the BMC Public Health journal. The other authors report no disclosures.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Additional file 1: table s1..

Demographic and socioeconomic characteristics of Danish residents aged 25 years or older in 2016 and 2017 by interview or web-based distribution. Data are presented as medians with 25th (Q1) and 75th (Q3) percentiles (age) or number of residents and percentage (all others).

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Svendsen, M.T., Bak, C.K., Sørensen, K. et al. Associations of health literacy with socioeconomic position, health risk behavior, and health status: a large national population-based survey among Danish adults. BMC Public Health 20 , 565 (2020). https://doi.org/10.1186/s12889-020-08498-8

Download citation

Received : 03 May 2019

Accepted : 10 March 2020

Published : 28 April 2020

DOI : https://doi.org/10.1186/s12889-020-08498-8

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Socioeconomic position
• Health behavior
• Health status
• Health inequality
• Social position
• Health risk indicators
• Population survey
• Health literacy questionnaire

BMC Public Health

ISSN: 1471-2458

• Submission enquiries: [email protected]
• General enquiries: [email protected]

• Share full article

Advertisement

Supported by

What the Data Says About Pandemic School Closures, Four Years Later

The more time students spent in remote instruction, the further they fell behind. And, experts say, extended closures did little to stop the spread of Covid.

By Sarah Mervosh ,  Claire Cain Miller and Francesca Paris

Four years ago this month, schools nationwide began to shut down, igniting one of the most polarizing and partisan debates of the pandemic.

Some schools, often in Republican-led states and rural areas, reopened by fall 2020. Others, typically in large cities and states led by Democrats, would not fully reopen for another year.

A variety of data — about children’s academic outcomes and about the spread of Covid-19 — has accumulated in the time since. Today, there is broad acknowledgment among many public health and education experts that extended school closures did not significantly stop the spread of Covid, while the academic harms for children have been large and long-lasting.

While poverty and other factors also played a role, remote learning was a key driver of academic declines during the pandemic, research shows — a finding that held true across income levels.

Source: Fahle, Kane, Patterson, Reardon, Staiger and Stuart, “ School District and Community Factors Associated With Learning Loss During the COVID-19 Pandemic .” Score changes are measured from 2019 to 2022. In-person means a district offered traditional in-person learning, even if not all students were in-person.

“There’s fairly good consensus that, in general, as a society, we probably kept kids out of school longer than we should have,” said Dr. Sean O’Leary, a pediatric infectious disease specialist who helped write guidance for the American Academy of Pediatrics, which recommended in June 2020 that schools reopen with safety measures in place.

There were no easy decisions at the time. Officials had to weigh the risks of an emerging virus against the academic and mental health consequences of closing schools. And even schools that reopened quickly, by the fall of 2020, have seen lasting effects.

But as experts plan for the next public health emergency, whatever it may be, a growing body of research shows that pandemic school closures came at a steep cost to students.

The longer schools were closed, the more students fell behind.

At the state level, more time spent in remote or hybrid instruction in the 2020-21 school year was associated with larger drops in test scores, according to a New York Times analysis of school closure data and results from the National Assessment of Educational Progress , an authoritative exam administered to a national sample of fourth- and eighth-grade students.

At the school district level, that finding also holds, according to an analysis of test scores from third through eighth grade in thousands of U.S. districts, led by researchers at Stanford and Harvard. In districts where students spent most of the 2020-21 school year learning remotely, they fell more than half a grade behind in math on average, while in districts that spent most of the year in person they lost just over a third of a grade.

( A separate study of nearly 10,000 schools found similar results.)

Such losses can be hard to overcome, without significant interventions. The most recent test scores, from spring 2023, show that students, overall, are not caught up from their pandemic losses , with larger gaps remaining among students that lost the most ground to begin with. Students in districts that were remote or hybrid the longest — at least 90 percent of the 2020-21 school year — still had almost double the ground to make up compared with students in districts that allowed students back for most of the year.

Some time in person was better than no time.

As districts shifted toward in-person learning as the year went on, students that were offered a hybrid schedule (a few hours or days a week in person, with the rest online) did better, on average, than those in places where school was fully remote, but worse than those in places that had school fully in person.

Students in hybrid or remote learning, 2020-21

80% of students

Some schools return online, as Covid-19 cases surge. Vaccinations start for high-priority groups.

Teachers are eligible for the Covid vaccine in more than half of states.

Most districts end the year in-person or hybrid.

Source: Burbio audit of more than 1,200 school districts representing 47 percent of U.S. K-12 enrollment. Note: Learning mode was defined based on the most in-person option available to students.

Income and family background also made a big difference.

A second factor associated with academic declines during the pandemic was a community’s poverty level. Comparing districts with similar remote learning policies, poorer districts had steeper losses.

But in-person learning still mattered: Looking at districts with similar poverty levels, remote learning was associated with greater declines.

A community’s poverty rate and the length of school closures had a “roughly equal” effect on student outcomes, said Sean F. Reardon, a professor of poverty and inequality in education at Stanford, who led a district-level analysis with Thomas J. Kane, an economist at Harvard.

Score changes are measured from 2019 to 2022. Poorest and richest are the top and bottom 20% of districts by percent of students on free/reduced lunch. Mostly in-person and mostly remote are districts that offered traditional in-person learning for more than 90 percent or less than 10 percent of the 2020-21 year.

But the combination — poverty and remote learning — was particularly harmful. For each week spent remote, students in poor districts experienced steeper losses in math than peers in richer districts.

That is notable, because poor districts were also more likely to stay remote for longer .

Some of the country’s largest poor districts are in Democratic-leaning cities that took a more cautious approach to the virus. Poor areas, and Black and Hispanic communities , also suffered higher Covid death rates, making many families and teachers in those districts hesitant to return.

“We wanted to survive,” said Sarah Carpenter, the executive director of Memphis Lift, a parent advocacy group in Memphis, where schools were closed until spring 2021 .

“But I also think, man, looking back, I wish our kids could have gone back to school much quicker,” she added, citing the academic effects.

Other things were also associated with worse student outcomes, including increased anxiety and depression among adults in children’s lives, and the overall restriction of social activity in a community, according to the Stanford and Harvard research .

Even short closures had long-term consequences for children.

While being in school was on average better for academic outcomes, it wasn’t a guarantee. Some districts that opened early, like those in Cherokee County, Ga., a suburb of Atlanta, and Hanover County, Va., lost significant learning and remain behind.

At the same time, many schools are seeing more anxiety and behavioral outbursts among students. And chronic absenteeism from school has surged across demographic groups .

These are signs, experts say, that even short-term closures, and the pandemic more broadly, had lasting effects on the culture of education.

“There was almost, in the Covid era, a sense of, ‘We give up, we’re just trying to keep body and soul together,’ and I think that was corrosive to the higher expectations of schools,” said Margaret Spellings, an education secretary under President George W. Bush who is now chief executive of the Bipartisan Policy Center.

Closing schools did not appear to significantly slow Covid’s spread.

Perhaps the biggest question that hung over school reopenings: Was it safe?

That was largely unknown in the spring of 2020, when schools first shut down. But several experts said that had changed by the fall of 2020, when there were initial signs that children were less likely to become seriously ill, and growing evidence from Europe and parts of the United States that opening schools, with safety measures, did not lead to significantly more transmission.

“Infectious disease leaders have generally agreed that school closures were not an important strategy in stemming the spread of Covid,” said Dr. Jeanne Noble, who directed the Covid response at the U.C.S.F. Parnassus emergency department.

Politically, though, there remains some disagreement about when, exactly, it was safe to reopen school.

Republican governors who pushed to open schools sooner have claimed credit for their approach, while Democrats and teachers’ unions have emphasized their commitment to safety and their investment in helping students recover.

“I do believe it was the right decision,” said Jerry T. Jordan, president of the Philadelphia Federation of Teachers, which resisted returning to school in person over concerns about the availability of vaccines and poor ventilation in school buildings. Philadelphia schools waited to partially reopen until the spring of 2021 , a decision Mr. Jordan believes saved lives.

“It doesn’t matter what is going on in the building and how much people are learning if people are getting the virus and running the potential of dying,” he said.

Pandemic school closures offer lessons for the future.

Though the next health crisis may have different particulars, with different risk calculations, the consequences of closing schools are now well established, experts say.

In the future, infectious disease experts said, they hoped decisions would be guided more by epidemiological data as it emerged, taking into account the trade-offs.

“Could we have used data to better guide our decision making? Yes,” said Dr. Uzma N. Hasan, division chief of pediatric infectious diseases at RWJBarnabas Health in Livingston, N.J. “Fear should not guide our decision making.”

Source: Fahle, Kane, Patterson, Reardon, Staiger and Stuart, “ School District and Community Factors Associated With Learning Loss During the Covid-19 Pandemic. ”

The study used estimates of learning loss from the Stanford Education Data Archive . For closure lengths, the study averaged district-level estimates of time spent in remote and hybrid learning compiled by the Covid-19 School Data Hub (C.S.D.H.) and American Enterprise Institute (A.E.I.) . The A.E.I. data defines remote status by whether there was an in-person or hybrid option, even if some students chose to remain virtual. In the C.S.D.H. data set, districts are defined as remote if “all or most” students were virtual.

An earlier version of this article misstated a job description of Dr. Jeanne Noble. She directed the Covid response at the U.C.S.F. Parnassus emergency department. She did not direct the Covid response for the University of California, San Francisco health system.

How we handle corrections

Sarah Mervosh covers education for The Times, focusing on K-12 schools. More about Sarah Mervosh

Claire Cain Miller writes about gender, families and the future of work for The Upshot. She joined The Times in 2008 and was part of a team that won a Pulitzer Prize in 2018 for public service for reporting on workplace sexual harassment issues. More about Claire Cain Miller

Francesca Paris is a Times reporter working with data and graphics for The Upshot. More about Francesca Paris

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• J Womens Health (Larchmt)

Health Literacy and Women's Reproductive Health: A Systematic Review

Kimberly a. kilfoyle.

1 Division of Women's Primary Health Care, Department of Obstetrics & Gynecology, University of North Carolina School of Medicine, Chapel Hill, North Carolina.

Michelle Vitko

2 Division of Pharmaceutical Outcomes and Policy, Eshelman School of Pharmacy, University of North Carolina, Chapel Hill, North Carolina.

Rachel O'Conor

3 Health Literacy and Learning Program, Division of General Internal Medicine, Feinberg School of Medicine at Northwestern University, Chicago, Illinois.

Stacy Cooper Bailey

Background: Health literacy is thought to impact women's reproductive health, yet no comprehensive systematic reviews have been conducted on the topic. Our objective was to systematically identify, investigate, and summarize research on the relationship between health literacy and women's reproductive health knowledge, behaviors, and outcomes.

Methods: PRISMA guidelines were used to guide this review. English language, peer-reviewed research articles indexed in MEDLINE as of February 2015 were searched, along with study results posted on Clinicaltrials.gov . Articles were included if they (1) described original data-driven research conducted in developed countries, (2) were published in a peer-reviewed journal, (3) measured health literacy using a validated assessment, (4) reported on the relationship between health literacy and reproductive health outcomes, related knowledge, or behaviors, and (5) consisted of a study population that included reproductive age women.

Results: A total of 34 articles met eligibility criteria and were included in this review. Data were abstracted from articles by two study authors using a standardized form. Abstracted data were then reviewed and summarized in table format. Overall, health literacy was associated with reproductive health knowledge across a spectrum of topics. It was also related to certain health behaviors, such as prenatal vitamin use and breastfeeding. Its relationship with other reproductive behaviors and outcomes remains unclear.

Conclusions: Health literacy plays an important role in reproductive knowledge and may impact behaviors and outcomes. While further research is necessary, healthcare providers should utilize health literacy best practices now to promote high-quality care for patients.

Introduction

H ealth literacy is defined as the degree to which individuals have the “capacity to obtain, process, and understand basic health information and services needed to make appropriate health decisions.” 1 According to the Institute of Medicine, 90 million Americans have difficulty understanding and acting upon health information, which can negatively affect their health and well-being. 1 Low, or inadequate, health literacy has been linked to numerous poor health outcomes, such as increased hospitalizations and emergency department use, poor overall health status, and higher mortality. 2 Given the significant burden of low health literacy on the healthcare system, emphasis has been placed on identifying and addressing this modifiable risk factor to improve patient–provider interactions and health outcomes.

While many studies explore the relationship between health literacy and health outcomes, less attention has been focused specifically on the effects of health literacy on women's reproductive health. This is unfortunate as health literacy is likely to impact many facets of reproductive healthcare. Knowledge about contraception, safe sexual practices, healthy pregnancy and postpartum behaviors, and preventive care is important to keep women healthy and leading productive lives. Furthermore, as the demographic profile of women giving birth in the United States continues to evolve and include more women at greatest risk for limited health literacy, including non-English-speaking racial and ethnic minorities and low-income women, health literacy has been increasingly recognized as an important area of focus for women's reproductive health. 3 As such, the American Congress of Obstetricians and Gynecologists (ACOG) released a Committee Opinion that highlighted the need for physicians to consider patients' health literacy skills for health promotion and clinical care activities. 4

Despite a growing recognition of the importance of health literacy in reproductive health, no rigorous, systematic literature reviews have been conducted to date examining the relationship between health literacy and women's reproductive health behaviors and outcomes. Without a synthesis of the current body of research, it is difficult to understand the progress made to date and to identify gaps in research. The purpose of this study was therefore to systematically identify, investigate, and summarize research on the relationship between health literacy and reproductive health knowledge, behaviors, and outcomes in developed countries.

Materials and Methods

This systematic review was conducted according to PRISMA guidelines. 5 To select articles, a search in MEDLINE was performed in February 2015 using the following terms as text words or MeSH terms: literacy OR health literacy OR numeracy OR health literacy [MeSH] AND reproduct* OR obstetric* OR gynecolog* OR maternal OR pregnan* OR contracept* OR prenatal OR postnatal OR postpartum OR preconception OR women's health OR women's health [MeSH] OR women's health services [MeSH]. A filter of English language articles only was added to the search. An additional search of studies with results posted on Clinicaltrials.gov was conducted using the search term “health literacy.”

Eligibility criteria

Articles were included in the review if they met the following eligibility criteria: (1) described original data-driven research, (2) were published in a peer-reviewed journal, (3) measured health literacy using a previously validated health literacy or numeracy assessment, (4) provided evidence on the relationship between health literacy and reproductive health outcomes or related knowledge or behaviors, and (5) consisted of a study population that included reproductive age women, defined as less than or equal to 45 years old, or included a broader age range, but focused on an issue common to reproductive age women. For the purpose of this review, reproductive health was considered to encompass any topic broadly related to obstetric and gynecological care, including (but not limited to) family planning, perinatal and postnatal care, sexual health, and screenings for gynecological cancers. Articles describing research conducted in developing countries were excluded as they generally examined the effects of illiteracy on reproductive outcomes; this was considered conceptually distinct from health literacy and therefore beyond the focus of this review. Developing countries were identified based on categorization by the U.S. Agency for International Development (USAID). 6

Study selection

To select studies, one study author (M.V.) first conducted a title review of all articles, removing any that were ineligible for the review. Next, two study authors (M.V., S.C.B.) independently reviewed abstracts from remaining articles; only articles that were considered ineligible by both authors were eliminated. Last, all remaining articles underwent a full article review by two study authors. Any article receiving a discordant coding between the two authors was reviewed by a third author and a final determination of inclusion made by majority vote.

A secondary search was also performed to identify potentially eligible articles. Specifically, references for all articles that were selected for inclusion were hand-searched, along with those from commentaries and related systematic reviews identified during the primary search. Figure 1 describes the study selection process and the results from each stage of review.

Article search and review process.

Data abstraction, quality review, and analysis

Data were abstracted from selected articles by two study authors using a standardized form that captured information on study design, study population, measures utilized, and findings. Any coding discrepancies between the two authors were resolved through subsequent review. Abstracted data were then compiled, reviewed, and summarized in table format by one study author (K.K.).

Quality assessments were independently conducted for each article by two study authors. A modified version of guidelines published by the Agency for Healthcare Research and Quality (AHRQ) was used; these guidelines have been utilized in prior literature reviews conducted on health literacy and health outcomes. 7–9 Specifically, reviewers were asked to consider each study in terms of its design, methodology and measurement, consideration for confounders, and use of appropriate statistical methods for analyses related to health literacy. Study quality was considered specifically within the context of our question of interest, that is, how health literacy relates to reproductive health-related knowledge, behaviors, and outcomes. Study findings were rated according to the AHRQ assessment definition of good ( i.e. , conclusions are very likely to be correct given degree of bias), fair ( i.e. , conclusions are probably correct given degree of bias), or poor ( i.e. , conclusions are not certain because bias is too large). Any rating discrepancies between the two study authors were resolved by a third reviewer. Quality ratings were then compiled, reviewed, and summarized in table format by one study author (K.K.) and reviewed by two authors (S.C.B., K.K.) for consistency across studies.

A total of 1100 articles were returned from the primary search in MEDLINE ( Fig. 1 ). After the title review, 415 articles remained. The abstract review eliminated 202 articles; a total of 213 abstracts were selected by one or both reviewers for full article review. After reviewing the 213 articles, 31 were selected for inclusion in the systematic review. Three additional articles were identified via reference mining procedures for a total of 34 articles. All articles included in the final review are summarized in Table 1 . A total of 12 studies were identified by the search in Clinicaltrials.gov ; however, none of the studies met eligibility criteria.

Summary of Articles Included in the Review

ACOG, American Congress of Obstetricians and Gynecologists; CES-D, Center for Epidemiologic Studies–Depression Scale; EPDS, Edinburgh Postnatal Depression Scale; HPV, human papillomavirus; IUD, intrauterine device; NICU, neonatal intensive care unit; NVS, Newest Vital Sign; OTC, over-the-counter; REALM, Rapid Estimate of Adult Literacy in Medicine; SBSQ, Set of Brief Screening Questions; STI, sexually transmitted infection; S-TOFHLA, Short Test of Functional Health Literacy in Adults.

Twenty-nine (85%) of the 34 studies included were observational or cross sectional in design, with 2 studies 10 , 11 utilizing a mixed methods design that combined qualitative focus groups with cross-sectional observational data ( Table 1 ). There were three experimental studies that included patient randomization 12–14 ; these studies evaluated the effects of an educational tool on knowledge of a reproductive health topic using hypothetical scenarios. Last, two longitudinal cohort studies examined health outcomes in the postpartum period. 15 , 16

Overall, one of the included studies was given the highest quality rating of good and the rest were considered fair quality ( n  = 16) or poor quality ( n  = 17) due to suboptimal ( e.g. , cross-sectional) study designs, unadjusted analyses, and/or potential for bias ( Table 1 ). It is important to note that for many studies, examining the relationship between health literacy and reproductive knowledge, behaviors, or outcomes was not the primary purpose. Therefore, authors only reported results from bivariate or unadjusted analyses for these variables, which reduced their quality rating.

Twenty-eight (82%) of the 34 studies included in the review were conducted exclusively in the United States. Of the six remaining studies, three were large, internet-based multinational studies, 17–19 two were conducted in the United Kingdom, 20 , 21 and one in Japan. 22 Study samples ranged from less than 100 participants (5 total studies, the smallest consisting of 30 participants) to 8344 participants for one of the internet-based multinational studies. The majority of studies enrolled low-income women seen at university-based clinics or hospitals. Almost all included English-speaking participants only, although one study enrolled exclusively Spanish-speaking participants, 23 one enrolled both Spanish- and English-speaking participants, 24 and one Japanese study 22 and three large, internet-based multinational studies 17–19 enrolled participants in several different languages. All but one study 25 had a study population that was exclusively female; most were focused on adult women.

Health literacy measurement varied across studies. Twenty-one (62%) of the 34 studies used either the original, the short form, or a translated version of the Rapid Estimate of Adult Literacy in Medicine (REALM) 26 to assess patient literacy skills, while 6 studies used the Short Test of Functional Health Literacy in Adults (S-TOFHLA), 27 4 studies used the Newest Vital Sign (NVS), 28 and 3 studies used the Set of Brief Screening Questions (SBSQ). 29 A description of these measures and how they are commonly scored is provided in Table 2 . Finally, two studies used numeracy scales, in addition to one of the health literacy measures already mentioned. 10 , 30 Two studies 22 , 31 did not report the percentage of their sample with low health literacy.

Description of Commonly Used Health Literacy Measures Across Studies

Outcome measurement in the studies varied considerably. The majority of studies used questionnaires or self-report to obtain information about outcomes. Medical record review was used to determine episodes of screening, follow-up after abnormal testing, obstetric outcomes, and adequacy of prenatal care. Studies were grouped broadly according to topic area and are summarized below.

Family planning and fertility

Knowledge related to contraception and fertility.

Consistently, studies demonstrated that health literacy was associated with knowledge about contraception. Davis et al. examined whether knowledge regarding oral contraceptive pills was associated with health literacy among low-income women attending a family planning clinic in Louisiana. 32 After viewing a standard video shown to women interested in obtaining contraception, low health literacy was associated with less knowledge regarding the meaning, mechanism of action, and risks of oral contraception. Twenty-three percent of women with a sixth-grade reading level or below could identify the purpose of contraception, compared with 54% of women with seventh- to eighth-grade reading level, or 74% of women with a ninth-grade reading level or higher ( p  < 0.0001). 32

Similarly, two other studies found that women with low health literacy were less likely to know the time in their menstrual cycle when they were able to get pregnant. In a group of women in the United Kingdom, Rutherford et al. found that 61.5% of women with a seventh- to eighth-grade reading level knew the time in the menstrual cycle when a woman was able to get pregnant compared with 85.6% of women with a reading level equivalent to ninth grade or above ( p  < 0.001). 20 Similarly, Gazmararian et al. found that women with low health literacy had four times the odds of not knowing when a woman could get pregnant during her menstrual cycle (odds ratio [OR] 4.54, 95% confidence interval [CI] 2.18–9.48). 33 In regard to fertility and aging, Gossett et al. found that higher health literacy was associated with increased knowledge of how aging affects fertility and the need for assisted reproductive technology. 30

In terms of understanding how to use contraception, Yee et al. found that women with low health literacy were more likely to identify inadequate knowledge as a barrier to taking contraception correctly. 10 Two studies by Raymond et al. similarly found that low health literacy was associated with less knowledge regarding the indication for use of emergency contraception after viewing a prototype drug information label. 34 , 35 However, in their 2002 study, Raymond et al. found that overall knowledge and understanding of the use of emergency contraception was high, with over 80% of all participants understanding 8 of 11 objectives tested.

Contraceptive use and planned pregnancies

Regarding actual contraceptive use, no relationship was found between health literacy status and use of contraception. Gazmararian et al. found no association between health literacy and report of current use of contraception, 33 Rutherford et al. found that the rate of emergency contraception use was not associated with health literacy, 20 and Davis et al. found no difference in correct use of contraception by health literacy in women currently using oral contraceptive pills. 32 Gazmararian et al. also examined contraception choice. When asked about birth control method ever used in the past, a higher proportion of women with low health literacy were found among users of both long-term highly efficacious forms of birth control, such as an intrauterine device (17.9%) or implant (13.3%), and the less efficacious forms, such as douching (13.9%) or the rhythm method (13.7%), compared with the proportion reporting use of other methods such as the diaphragm (10.3%), medroxyprogesterone acetate (10.1%), or oral contraceptive pills (8.1%). 33

There were inconsistent results regarding rates of planned pregnancy by health literacy, although almost all studies on this topic were considered of poor quality; only one was rated as fair. 33 Endres et al. found that women with low health literacy were more likely to have an unplanned pregnancy compared with women with adequate health literacy (75% vs. 40%, p  = 0.02). 24 Lupattelli et al. similarly reported a difference in incidence of unplanned pregnancy by health literacy status. In their multinational cohort, 12.0%, 9.8%, and 8.2% of women with low, medium, and high health literacy, respectively, reported unplanned pregnancies ( p  < 0.05). 17 Several others, however, found no difference in rate of reported unplanned pregnancy by health literacy. 10 , 20 , 33

Sexual behavior, sexually transmitted infections, and cervical cancer screening

Sexual behavior and sexually transmitted infections.

Four studies examined the relationship between health literacy and sexual behavior and sexually transmitted infections (STIs) and found conflicting results. Two studies were considered of fair quality and two of poor quality; there were no systematic differences in findings between studies by their quality ratings. In evaluating sexual behavior that puts women at risk of STIs, Rutherford et al. found a statistically significant, but clinically narrow, difference in onset of sexual activity by health literacy status. Women with low health literacy had a reported mean age at first intercourse of 15.5 years compared with a mean age of 15.8 years in women with high health literacy ( p  < 0.001). 20 Women with low health literacy compared with women with high health literacy were also more likely to report unprotected intercourse with their first sexual encounter (16.3% compared with 10.2%, p  < 0.05) and two or more sexual partners in the last 6 months (19% compared with 9.5%, p  < 0.002). 20

Conversely, Sharp et al. found that high health literacy was associated with sexual behaviors that are risk factors for abnormal pap smears (1.8 risk factors in the low health literacy group compared with 2.3 risk factors in the high health literacy group), including the use of oral contraceptive pills, a higher reported number of lifetime sexual partners, and reported initiation of sexual activity before the age of 18 years. 36 A greater percentage of women with high health literacy reported more than five lifetime sexual partners compared with women with low health literacy (51.4% vs. 25.9%, p  < 0.01). Forty-three percent of women with low health literacy reported four to five lifetime partners. 36 There was no association between health literacy skills and age of sexual debut or history of prior STIs. 36 Needham et al. found no association between health literacy and report of risk factors for STIs, including age at first intercourse, number of lifetime sexual partners, history of previous sexually transmitted illness, and use of barrier protection with last episode of intercourse. 37

Rutherford et al. and Needham et al. found no differences in knowledge regarding safe sexual practices by health literacy status. 20 , 37 However, Rutherford et al. found that women with low health literacy were less likely to know how STIs could be passed from one individual to another; they were also less likely to know how to look for and identify signs of an STI. 20 In a separate study, women and men with low health literacy were more likely to perceive themselves at risk for acquiring gonorrhea, but were less likely to have obtained testing for gonorrhea in the past year. 25 Women with low health literacy were also more likely to report being interested in learning more about sexual health. 20

Cervical cancer screening

Three studies examined the relationship between cervical cancer screening and health literacy; all were considered of fair quality. Sharp et al. reported that the proportion of women with a prior abnormal pap smear and history of colposcopy did not vary by health literacy status. There was also no difference in severity of pap abnormalities on screening found during this study by health literacy status. 36 However, women with low health literacy were less likely to follow-up after receiving abnormal pap results and were more likely to report significant distress related to the diagnosis of an abnormal pap smear. 36

Lindau et al. similarly found an association with health literacy and follow-up of abnormal pap smears. Women with low health literacy compared with those with adequate health literacy were more likely to say they would not follow-up with a provider after abnormal pap smear testing (30% compared with 19%, p  < 0.036). Instead they would worry, panic, do nothing, or not know what to do. 38 Women with high health literacy had twice the odds of understanding the purpose of a pap smear as a screening test compared with women with inadequate health literacy (OR 2.25, 95% CI 1.05–4.80). 38

A follow-up study designed to investigate this relationship further, however, found no association between health literacy and follow-up within 1 year for abnormal pap results. 39 Additionally, although there were no statistically significant differences in severity of pap results by health literacy, trend data suggested a relationship. Among women with abnormal pap smears and inadequate health literacy, 29% had results that showed low-grade squamous intraepithelial lesions (LGSILs) and 55% had atypical squamous cells of unknown significance (ASCUS). Among women with abnormal pap smears and adequate health literacy, 50% had LGSIL results and 25% had ASCUS on pap smear testing. 39 This was likely not statistically significant due to a small study sample size.

Pregnancy and postpartum outcomes

Prenatal care utilization and quality.

Data are inconsistent as to whether there are differences in prenatal care by patient health literacy. While the quality ratings for these studies varied, there were no systematic differences in terms of findings by quality. Endres et al. evaluated prenatal care utilization among women with high-risk pregnancies due to pregestational diabetes. 24 Results indicated that women with low health literacy compared with women with adequate health literacy were less likely to have received preconception counseling with an obstetric provider (13% vs. 43%, p  = 0.01) and were more likely to initiate prenatal care at a later gestational age (12.8 weeks vs. 8.3 weeks, p  = 0.04). 24 In unadjusted analysis, Poorman et al. found that an increased proportion of women with limited health literacy reported missed prenatal care appointments compared with women with marginal and adequate health literacy (16.8% vs. 6.9% vs. 2.1%, respectively, p  = 0.03). 40 After controlling for income and education, this association was no longer significant.

In contrast, Bennett et al. reviewed medical records from a cohort of 202 postpartum African American women and found no difference by health literacy status in gestational age at initiation of prenatal care. Overall rates of delayed prenatal care in this sample were high, however, with 61% of women in this study initiating prenatal care after the first trimester. 11 This study also did not find an association between health literacy status and adequacy of prenatal care as measured by the Adequacy of Prenatal Care Utilization Index (APNCU), a measure that combines gestational age at initiation of prenatal care and number of visits attended. 41 Fifty-eight percent of women with low health literacy had inadequate prenatal care by the APNCU compared with 49% of women with high health literacy ( p  = 0.341). 11

In several studies, perceptions of prenatal care and care quality differed by health literacy categorization. Shieh et al. reported that women with low health literacy were less likely to believe that they could have an impact on their pregnancy's outcome, instead perceiving the healthcare provider to be in control of maternal and fetal outcomes. 42 A study conducted in Japan explored the relationship of health literacy among culturally diverse women and its association with reported quality of prenatal care. High health literacy was associated with less understanding, less perceived respect by the healthcare provider, and increased loneliness during and after pregnancy. 22 Finally, Bennett et al. conducted four focus groups with women who had low and high health literacy skills. Regardless of health literacy status, the theme of communication and partnership between provider and patient was consistent among all focus groups. Participants agreed that it was important that providers provide information in a way that was clear and accessible. Other themes that emerged included desire for trust, continuity of care, and a close patient–physician relationship. 11

Cho et al. found lower levels of understanding of aneuploidy screening options among women with limited health literacy. Pregnant women with low health literacy skills had lower average scores on the Maternal Serum Screening Knowledge (MSSK) Questionnaire compared with women with high health literacy (5.5 vs. 6.96, p  < 0.01) and were more likely to have inadequate knowledge as determined by a composite MSSK score (97.3% vs. 11.1%, p  < 0.01). Despite this, there was no difference by health literacy group in the rate of acceptance of first or second trimester serum screening. 43

Health information seeking during pregnancy was examined in two studies by Shieh et al. One study found no difference by health literacy status in broad practices related to media that women used to obtain health information during pregnancy, including books, brochures, TV, and the internet. 31 However, in a separate study, Shieh et al. found that women with low health literacy were less likely to use the internet as a source of information. Compared with women with high health literacy, pregnant women with low health literacy had less access to the internet, and among women who did have access, those with low health literacy skills were less frequent users of the internet for prenatal care information. 42 Studies did not differ in terms of their quality assessment.

Prenatal medication and supplement beliefs and use

Low health literacy was associated with not taking vitamin supplements during pregnancy. Endres et al. found a lower proportion of women with low health literacy compared with women with adequate health literacy who took folic acid before pregnancy and into the first trimester (31% compared with 83% p  = 0.001). 24 Poorman et al. found that women with low health literacy had greater odds of not taking a prenatal vitamin during pregnancy (adjusted OR 3.6, 95% CI 1.6–8.5) in comparison with those with adequate health literacy. 40

Pregnant women with low health literacy were also more likely to attribute risk to medication use in pregnancy. 17 , 21 In Ireland, Duggan et al. demonstrated that women with low health literacy had stronger beliefs regarding the harms of medication and that medications are overused in pregnancy compared with women with high health literacy. 21 In a large multinational study, Lupattelli et al. showed increased perception of risk and negative beliefs about medications and supplements in pregnancy, including herbal remedies, among women with low health literacy. 17 Furthermore, these negative beliefs and perceived risks appeared to mediate the increased rate of nonadherence to medications in pregnancy reported by women with low health literacy compared with medium health literacy and high health literacy (25%, 22.5%, and 19.2%, respectively). 17 In the same cohort of women, Hameen-Anttila et al. reported that women with limited health literacy skills desired greater information regarding medication use in pregnancy in comparison with women with adequate health literacy skills. 18

Smoking and pregnancy

It is unclear if health literacy status impacts smoking practices in pregnancy. Arnold et al. found that high health literacy was associated with increased knowledge and concern regarding smoking and its effects on pregnancy and children in the home. 44 Despite this, there was a nonsignificant trend toward a greater proportion of women with high health literacy smoking during pregnancy. Fifteen percent of women reading at or below a third-grade reading level compared with 14% reading at a fourth- to sixth-grade level, 18% of women reading at a seventh- to eighth-grade level, and 25% of women with a greater than ninth-grade reading level reported smoking during pregnancy. 44

Conversely, Smedberg et al. found that low health literacy was associated with a higher proportion of smoking during pregnancy. Among women who did smoke during pregnancy, there was no difference in the amount of smoking based on health literacy status. 19 Poorman et al., in unadjusted analysis, found that women with inadequate health literacy reported higher rates of smoking during pregnancy compared with women with marginal and adequate health literacy (19.8%, 15.8%, and 8.8%, respectively, p  < 0.01). This difference was not significant after adjusting for income and education. 40 The studies by Poorman and Arnold were considered of fair quality, while the Smedberg article was rated as poor.

Obstetric and postpartum outcomes

The effect of health literacy on obstetric outcomes has not been thoroughly evaluated. In a cohort of high-risk obstetric patients with pregestational diabetes, Endres et al. found that women with low health literacy had more frequent hospitalizations and higher birth weight babies. 24 Babies of women with pregestational diabetes and low health literacy weighed ∼450 g more than babies of mothers with adequate health literacy ( p  = 0.001). 24 However, there were no differences in other maternal and neonatal outcomes such as gestational age at delivery, percentage of women who delivered vaginally, rates of shoulder dystocia, APGAR scores, or neonatal intensive care unit admissions. Bennett et al. found no association between health literacy and obstetric outcomes, including low birth weight, preterm delivery, and cesarean section. 11

Women with low health literacy were more likely to suffer from depression during pregnancy 23 and at 6 weeks postpartum. 15 In a cohort of pregnant Latina women, those with inadequate health literacy compared with women with adequate health literacy had twice the odds of suffering from depression during pregnancy (OR 2.39, 95% CI 1.07–5.35). 23 In a separate postpartum cohort, a larger proportion of women with low health literacy met criteria for postpartum depression compared with women with high health literacy (26.1% vs. 8.8%). 15 After adjusting for history of depression and number of children at home, postpartum women with low health literacy had 3.45 greater odds (95% CI 1.1–11.8) of suffering from postpartum depression compared with women with high health literacy. 15

Postpartum practices were shown to vary by health literacy status across multiple studies. Health literacy was negatively associated with breastfeeding rates. Kaufman et al. found that women with low health literacy were less likely than those with adequate health literacy to report exclusively breastfeeding their infants at 2 months postpartum (23% compared with 54%, p  = 0.018). 45 Poorman et al. found that 30% of women with less than adequate health literacy never breastfed compared with 13% of women with adequate health literacy. 45 Finally, Ehrenthal et al. found that diabetic women with low health literacy were less likely to follow-up for glucose testing postpartum. 16

Interventions to improve understanding of reproductive health information

A number of studies tested interventions designed to reduce literacy-related disparities in understanding reproductive health information. Using a computer-based tool to improve prenatal screening knowledge, Yee et al. found improved patient understanding of screening methods after viewing the tool. The level of knowledge improvement did not differ by health literacy. 14

Similarly, You et al. developed an educational tool with text and icons to improve preeclampsia knowledge. 13 A previous study by this team found no difference in preeclampsia knowledge by health literacy in adjusted analysis, although knowledge about preeclampsia among all women in the study sample was low, with only 13% of participants able to identify signs and symptoms of this condition. 46 The educational tool was shown to improve overall levels of preeclampsia knowledge among pregnant women, and no difference in understanding was found by health literacy. 13 Similarly, in a study designed to improve interpretation of prescription drug warning labels by You et al., health literacy did not predict correct interpretation of an enhanced prescription drug warning label warning against medication use while pregnant. 47

Kakkilaya et al. examined the effect of a visual aid on patient knowledge of the morbidity and mortality associated with birth of a very preterm infant (23 weeks gestational age). Improvement in knowledge scores did not vary by health literacy after using the aid. 12

In a sample of women between the ages of 16 and 21, STI comprehension was assessed after reading an informational brochure regarding Chlamydia. 37 Women with low health literacy had significantly lower knowledge scores compared with women with high health literacy after reviewing the brochure (average knowledge score of 7.54 vs. 9.08 of 10, p  < 0.001). 37 Boxell et al. found a larger improvement in knowledge of symptoms of gynecologic cancer in women with high health literacy compared with those with moderate or low health literacy after viewing an ovarian and cervical cancer symptom brochure. 48

Findings from this systematic review suggest that health literacy is related to reproductive health knowledge across a number of topics, including contraception, fertility, prenatal screening, and STIs. Additionally, health literacy appears to be related to certain obstetric health behaviors, such as prenatal vitamin use and exclusive breastfeeding, but its relationship with other reproductive health behaviors, such as behavioral risk factors for STIs and smoking during pregnancy, is less clear. Evidence suggests a link between health literacy and postpartum depression, but too few studies have been conducted on other obstetric and gynecologic outcomes to draw firm conclusions. Finally, the few intervention studies conducted to date have shown promising results, indicating that the use of tailored educational materials can increase understanding of reproductive health topics for patients with both limited and adequate health literacy, perhaps even reducing literacy-related disparities in reproductive health knowledge.

Several studies included in this review demonstrated that women with low health literacy are more likely to exhibit avoidance behavior when it comes to care. This includes less screening for STIs, later initiation of prenatal care, less follow-up of abnormal test results after cervical cancer screening, and less overall likelihood to seek care. 24 , 25 , 36 , 38 This may be related to increased distress regarding results, lack of knowledge regarding risks and testing options, lack of health information seeking, poorer healthcare access, or lower level of self-efficacy. Studies in other populations have demonstrated a similar phenomenon. 8 , 49 It is important to determine the root cause of this avoidance, whether it is health literacy or some other factor, to better engage women in preventive care and follow-up.

While this review offers insight on the relationship between health literacy and reproductive knowledge, behavior, and outcomes, it is important to note that studies included in the review had numerous limitations. Specifically, many studies had small sample sizes that were often not representative of the general population and were recruited via convenience sampling. Small sample sizes also made it harder to evaluate certain clinical outcomes such as preterm birth that are not as common. Most studies also used cross-sectional and observational study designs and even the more rigorous experimental studies generally utilized hypothetical scenarios. Measurement was highly variable and often suboptimal, with studies relying upon self-report and measures that have not been validated to assess outcomes. Last, numerous studies solely reported unadjusted analyses and did not control for known factors that impact women's health outcomes. More rigorous studies are clearly warranted to further establish the relationship between health literacy and women's reproductive health outcomes. These studies should seek to use more objective measures of clinically relevant outcomes when possible and more advanced study designs or analyses to account for potential confounding. In addition, given the changing demographics of the United States, more studies should be conducted among Hispanic and Spanish-speaking women.

This systematic review itself also has limitations. As with any systematic review, our review is limited by publication bias. We only synthesized published findings and did not search the gray literature or other sources. It is also possible that articles that should have been included were not returned with our search strategy; however, additional reference mining was performed to minimize this possibility. As this review was restricted to research conducted in the United States and other developing countries, it is possible that relevant articles from other countries were unnecessarily excluded. Additionally, there were several topics that are important to women's health areas ( e.g. , menopause, incontinence) that were excluded as they were believed to be less relevant to women of reproductive age. Certainly, how health literacy impacts these topics should also be explored.

Studies included in this review reported that anywhere from 9% to 78% of women had less than adequate health literacy skills (average of 46%). This wide variation likely reflects the diversity of populations that study samples were drawn from, the variety in health literacy measures used, and differences in how individuals were categorized as having adequate or inadequate health literacy. Such variability makes it difficult to make comparisons across studies and may have led to conflicting results. This challenge is present for almost all literature reviews focusing on health literacy and has been previously acknowledged to be a limitation for the field. 50 Such measurement challenges highlight the need to identify the best health literacy measures available for use in this population and/or to consider the creation of new tools that effectively measure these skills in reproductive age women. Results from this review unfortunately do not offer clear guidance on which published instrument may be optimal for use among reproductive age women. Further refining, standardizing, and improving health literacy measurement will be crucial to advancing this line of research.

Despite measurement variation, findings from the studies included in this review indicate that limited health literacy is likely to be common in reproductive health practices and worthy of further attention and research. While the studies conducted to date offer insight on the relationship between health literacy and reproductive knowledge and related behaviors, more methodologically rigorous studies are needed to further investigate the association between health literacy and clinically meaningful outcomes and to further elucidate the causal mechanisms through which health literacy may affect these outcomes. Finally, more health literacy-informed interventions are needed to move beyond describing the problem of limited health literacy and into the realm of improving reproductive care and outcomes for women.

Our review suggests that health literacy plays an important role in reproductive health for women across a number of key contexts, including contraception, fertility, prenatal screening, and sexual health. Additional research is needed to fully understand these relationships, to further explore the association between health literacy and reproductive outcomes, and to evaluate interventions to reduce literacy-related disparities in reproductive health knowledge and outcomes. In the meantime, obstetricians and gynecologists should be aware of the potentially negative effects of low health literacy on reproductive health and should become more familiar with guidelines promoting clear communication and health literacy best practices for patient education and counseling. 51–53 Numerous organizations, including AHRQ, the Joint Commission, and The Centers for Disease Control and Prevention, have advised healthcare providers to utilize clear communication and plain language techniques with all patients regardless of their presumed health literacy skills, thus adopting a universal precautions approach. 51 This is essential as limited health literacy is likely to be prevalent in many patient populations and is not always readily recognized by healthcare providers. 38 , 39

Acknowledgments

Dr. Kilfoyle is supported by an institutional NIH training grant (T32 HD040672-15). Dr. Bailey is supported by the UNC Center for Diabetes Translation Research to Reduce Health Disparities (P30DK093002) from the National Institute of Diabetes and Digestive and Kidney Diseases.

Author Disclosure Statement

Dr. Bailey has served as a consultant to, and received grant support from, Merck, Sharp & Dohme Corp. for work unrelated to this manuscript. Dr. Kilfoyle's spouse was employed by GlaxoSmithKline at the time that this research was performed and the manuscript was written.