Objective To investigate gender differences of health literacy in individuals with a migration background.
Design Systematic review and meta-analysis. OVID/MEDLINE, PsycINFO and CINAHL were searched in March 2018 and July 2020.
Setting Studies had to provide health literacy data for adult women and men with a migration background, collected with a standardised instrument, or report results that demonstrated the collection of such data. Health literacy data were extracted from eligible studies or requested from the respective authors. Using a random-effects model, a meta-analysis was conducted to assess standardised mean differences (SMDs) of health literacy in men and women. Two researchers independently assessed risk of bias for each included study using the Appraisal Tool for Cross-Sectional Studies.
Results Twenty-four studies were included in this systematic review. Thereof, 22 studies (8012 female and 5380 male participants) were included in the meta-analyses. In six studies, gender-specific health literacy scores were reported. The authors of additional 15 studies provided their data upon request and for one further study data were available online. Women achieved higher health literacy scores than men: SMD=0.08, 95% CI 0.002 to 0.159, p=0.04, I2=65%. Another 27 studies reported data on female participants only and could not be included due to a lack of comparable studies with male participants only. Authors of 56 other eligible studies were asked for data, but without success.
Conclusion Men with a migration background—while being much less frequently examined—may have lower health literacy than women. As heterogeneity between studies was high and the difference became statistically insignificant when excluding studies with a high risk of bias, this result must be interpreted with caution. There is a paucity of research on the social and relational aspects of gender in relation to health literacy among people with a migration background, especially for men.
PROSPERO registration number CRD42018085555.
- public health
- health economics
- education & training (see medical education & training)
Data availability statement
Data are available upon reasonable request. Data may be obtained from a third party and are not publicly available.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
Strengths and limitations of this study
This is the first systematic review to investigate gender differences of health literacy in persons with a migration background.
Our study incorporates previously unpublished gender-specific health literacy data of 15 studies.
Data of 56 potentially eligible studies could not be retrieved.
Heterogeneity between studies was high and statistical significance of gender differences vanished when excluding studies with high risk of bias; results must be interpreted with caution.
Health literacy can be described as having the ‘knowledge, motivation and competencies of accessing, understanding, appraising and applying health-related information within the healthcare, disease prevention and health promotion setting, respectively’.1 This broad definition is often referred to as ‘comprehensive health literacy’. A somewhat narrower concept, designated as ‘functional health literacy’, focuses on a person’s ability to read and understand written health information and perform simple arithmetic tasks in a health context.2 Empirical research has shown that limited health literacy is associated with more frequent hospitalisation and emergency treatments, reduced use of preventive measures, poor adherence to medical treatment, and an increased risk of morbidity and mortality.3–6 Studies examining overall health literacy in the USA7 and eight European countries (Austria, Bulgaria, Germany, Greece, Ireland, the Netherlands, Poland and Spain)8 have found inadequate or problematic levels of self-reported health literacy in 30%–50% of the general population, which are thought to result in substantial additional costs in healthcare systems.9 Importantly, health literacy is subject to the influence of societal, environmental, personal and situational factors.1 The exchange of health information, for example, in the treatment setting, depends on the respective social context.10 Thus, rather than being an individual skillset, health literacy should be regarded as a social-relational concept.11
Limited health literacy is not evenly distributed among populations; current studies have found some populations to be more vulnerable than others, especially migrants.12 13 For the purpose of this review, we defined persons with a migration background as either first-generation or second-generation migrants. For first-generation migrants, we follow the definition of the International Organization for Migration (IOM): ‘IOM defines a migrant as any person who is moving or has moved across an international border or within a State away from his/her habitual place of residence, regardless of (1) the person’s legal status; (2) whether the movement is voluntary or involuntary; (3) what the causes for the movement are; or (4) what the length of the stay is.’ Meanwhile, second-generation migrants are defined as persons with at least one parent being a migrant; it is important to include this group as the health-related effects of migration can also affect the second generation.14 15 This broad definition is in line with the term ‘person with a migratory background’, as defined by the European Commission.16
In a representative, cross-sectional study in Germany, 71% of persons with a migration background reported major difficulties in processing health information and translating it into healthy choices.17 Lower health literacy scores compared with the native-born population have also been measured in immigrants in Canada,18 Sweden19 and in some immigrant subpopulations in the USA, for example, elderly20 and Hispanic/Latinx immigrants.21 With the number of international migrants worldwide rising steadily, reaching an estimated 272 million in 2019,22 and a constant high level of global migration expected for the near future,23 understanding the factors that influence health literacy in migrants is considered a highly relevant task. As plenty of research yielded profound differences between men and women regarding health information processing,24 25 health behaviour26 27 and health outcomes,28 gender could be one of these factors.
However, the usage and understanding of the terms gender and sex within medical research appears to be inconsistent.29 For conceptual clarity, we refer to gender as relating to social aspects concerning gender identities, norms and relations, while sex is reserved for biological differences between men and women.30 Given the social-relational character of health literacy,1 possible differences between men and women are much more likely to reflect socially influenced gender dissimilarities than biologically determined sex differences. Therefore, we will use the term gender consistently throughout this review, even though we expected most studies to dichotomise gender in a biologically inspired way as male/female.31
Correlations have been found between health literacy scores and gender,4 32–35 with ambiguous results regarding the strength and direction of these results; some reported higher health literacy scores for women,33 while others reported higher scores for men.35 Moreover, for migrant populations, studies found health behaviour36 and health information-seeking strategies37 differed between the genders. Men and women migrate for different reasons and their experiences during and after migration differ as well, including their interactions with the health systems of the receiving countries.38 Consequently, researchers have repeatedly called for taking gender aspects into account when it comes to examining the health literacy of persons with a migration background.39 40
Accordingly, this systematic review and meta-analysis aims to investigate gender differences of health literacy in persons with a migration background, which were assessed using standardised instruments to measure health literacy.
This study is part of an overarching research project on ‘Gender-specific health literacy in individuals with migration background’ (GLIM). It includes two qualitative focus group studies, one on gender aspects of health literacy in migrants,41 and one on systemic aspects of intercultural treatment settings.42 Furthermore, a Cochrane review on interventions targeting the health literacy of migrants43 and a qualitative Cochrane review on gender differences in the health literacy of migrants44 are currently conducted within GLIM.
This systematic review was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.45 The protocol has been registered in advance on the International Prospective Register of Systematic Reviews,46 including descriptions of the review question, search strategy and inclusion and exclusion criteria of studies and participants.41
As a first step, we defined and set the search terms in English, addressing the main concepts, that is, ‘health literacy’ and ‘migration background’. As a second step, we developed two-parted search strings—one section referring to health literacy and the other encompassing migration background—and pretested them in PubMed. All parts were combined using the Boolean operator ‘AND’. Following the preliminary search, we identified key publications and analysed them for wording used in the title and abstract as well as index terms to expand and adapt the search terms to cover additional and divergent wording. After the search terms and search strings were finalised for PubMed, the search strategy was adapted to each of the additional databases and the actual search was conducted in the OVID (MEDLINE), PsycINFO and CINAHL databases. The first search was conducted in March 2018, followed by an update search in July 2020. Further details on the search terms and search strategy are provided in online supplemental file 1.
Types of studies
In this review, we included primary research studies that used quantitative methods such as observational, prospective and retrospective cohort studies, randomised controlled trials and controlled trials. The inclusion criteria were as follows. (1) Studies had to include health literacy data collected with a standardised instrument that had been validated according to objectivity, validity and reliability. In cases where a validated assessment tool had been translated into another language but had not been validated in the target language, the minimum requirement was a forward-translation and back-translation process, as recommended by the WHO.47 (2) Studies had to provide gender-specific health literacy scores, health literacy levels or they had to report results that demonstrated the collection of extractable health literacy data for women and men with a migration background. In the latter case, we requested the respective authors send us mean health literacy scores, SDs, and the number of male and female participants included in their study.
Types of participants
We included all adults with a migration background aged ≥18 years. Many studies from the USA focus on so-called Latinos/Latinas or Hispanics, blurring the categories of migration background and ethnic minority. Drawing on statistics related to the use of language within these populations,48 49 we decided to include such studies only if participants (self-)identified as Latinos/Latinas/Latinx, Hispanics or Latin-American (eg, Mexican Americans) and stated that they speak Spanish as their first language at home or in medical consultations. We excluded studies that focused on ethnic minorities (eg, Roma, Asian Americans) if these studies did not state that the participants were first-generation or second-generation migrants. The inclusion and exclusion criteria are listed in table 1.
Types of outcome measures
Our primary outcome was gender differences in the health literacy of persons with a migration background assessed at baseline with standardised instruments to measure health literacy.
No time or language filters were applied.
Study selection and screening
The studies retrieved were exported to Covidence, a web-based systematic review tool.50 Two researchers (DC, AB) independently screened the studies’ titles and abstracts for eligibility. In a second step, they individually reviewed the full texts of the studies identified in the screening process using the predefined inclusion and exclusion criteria. Disagreements were resolved through involvement of a third author (AA).
We adapted a data extraction sheet provided from Cochrane51 and tested it using the first three included studies. Study characteristics and results were extracted for each study, including authors, country of research, description of the population, number of male/female participants, type of health literacy measurement instrument and baseline mean and SD of health literacy scores for men and women. In addition, we extracted further study details according to PROGRESS (an acronym for place of residence, race, ethnicity/culture/language, occupation, gender/sex, religion, education, socioeconomic status and social capital), a framework for incorporating equity aspects into systematic reviews.52 As we focused on quantitative gender differences, we did not extract data for the additional items described in PROGRESS Plus,53 except for age. All extracted data were double-checked against the full text of the studies by a second researcher, as recommended in Chapter 5 of the Cochrane handbook.54 In the case that a study did not include gender-specific health literacy scores, the authors of the study were emailed and asked to provide these data. If baseline health literacy scores and SDs were reported for more than one validated measurement instrument, we extracted the data produced by the tool measuring the broader conceptualisation of health literacy. Thus, we favoured comprehensive over functional health literacy and differentiated assessment tools, such as the Test of Functional Health Literacy in Adults (TOFHLA), over screening instruments such as the Brief Health Literacy Screen (BHLS).
Risk of bias in individual studies
Two authors (DC, AB) independently assessed the risk of bias for the studies included in the meta-analysis by using the Appraisal Tool for Cross-Sectional Studies (AXIS).55 Differences were reconciled discursively. To categorise studies into low, medium and high risk of bias, we built a composite score, as proposed by Boxberger and Reimers.56 AXIS comprises 20 criteria. We considered studies to have high risk of bias if less than 50% of the criteria were fulfilled. Medium risk of bias was ascribed to studies meeting between 50% and 66% of the criteria and low risk of bias was reserved for studies fulfilling more than 66% of the criteria.57
Synthesis of results
For meta-analyses, we imported gendered health literacy scores and SDs and numbers of male and female participants into Review Manager (V.5.4), a software provided by the Cochrane Collaboration for conducting systematic reviews and meta-analyses.58 As there are different tools for measuring health literacy in various ways, we used standardised means and a random-effects model59 to estimate the gender differences in health literacy scores, as recommended by the Cochrane Collaboration.60 Heterogeneity between studies was assessed using Q and I2 statistics.61 For better interpretation, we transformed statistically significant (p<0.05) standardised mean differences (SMDs) into a commonly used scale (S-TOFHLA) using the pooled SD of scores in included studies that had applied this instrument.62 63
Tests for subgroup differences were carried out for region of origin, type of health literacy assessment tool and functional versus comprehensive health literacy. We undertook two kinds of sensitivity analyses: (1) excluding studies considered to have high risk of bias and (2) using a fixed-effects model instead of a random-effects model.
For visualisation, data were exported into RStudio V.1.3.64
We identified 5742 studies, of which 2013 were excluded as duplicates. Thus, 3729 articles were checked for titles and abstract, of which 3437 were excluded, leaving 292 studies for full-text screening. At the full-text review stage, we excluded a further 268 studies, including 56 otherwise eligible studies that did not report gender-segregated health literacy data and whose authors did not provide these scores upon request. This includes the only study that made it into the full-text screening and was written in a language other than English (in this case, Chinese). Among the excluded studies were 40 that used unsuitable definitions of the term migrant (ie, focusing on ethnic or ‘racial’ minorities without providing information about whether the participants or their parents had migrated themselves). We also had to exclude 27 studies that included female participants only; as no studies with exclusively male participants met our inclusion criteria, there were no comparable counterparts for these studies. Finally, a total of 24 studies were included in the systematic review. Six studies65–70 reported gender-separated mean health literacy scores and SDs as well as the number of participants for each gender. For 15 of the included studies,71–85 the data were provided by the respective authors via email; and for 1 study,86 we obtained it from a publicly available data set.87 Two studies19 88 reported health literacy levels (eg, low vs high health literacy) instead of scores and could not be meta-analysed. For the PRISMA45 flow diagram, see figure 1.
Characteristics of included studies
Of the 24 studies included in this review, 16 were conducted in the USA, 6 in Europe and 2 in Asia. Participants included in the studies were of Hispanic/Latin (n=14), diverse (n=5), Korean (n=2), Chinese, Somali and Russian (each n=1) origin. Most studies (n=20) measured functional health literacy, while comprehensive health literacy was measured in five studies (one study reported results for both functional and comprehensive health literacy). Health literacy was measured using different instruments. Varieties of the BHLS,89 containing 1–16 questions for self-assessment of functional health literacy, were used in seven studies. Meanwhile, five studies made use of the European Health Literacy Survey Questionnaire (HLS-EU-Q),90 containing either 16 or 47 items to self-assess participants’ comprehensive health literacy. Four studies made use of the Rapid Estimate of Adult Literacy in Medicine (REALM),91 a performance-based oral reading and recognition test, or versions of the Short Assessment of Health Literacy in Spanish (SAHLSA, SAHL-S)92 93 which was developed on the basis of REALM. The performance-based TOFHLA94 was applied in three studies, including a short version called S-TOFHLA.95 The TOFHLA measures functional health literacy by assessing the participants’ reading comprehension of medical information. Further measurement tools for determining functional health literacy, each used in one study, were the Health Literacy Assessment Using Talking Touchscreen72 and the Swedish Functional Health Literacy Scale,96 both self-assessment instruments, and the performance-based Korean Health Literacy Scale.97 Finally, the All Aspects of Health Literacy Scale,98 an instrument for self-assessing comprehensive health literacy, was used in one study. All scales were reported with higher scores representing higher health literacy. None of the included studies focused on gender aspects. The characteristics of the included studies are presented in table 2.
Of the 24 studies included in this review, 219 88 reported health literacy categories (eg, low vs high health literacy) instead of numerical scores. These studies are not included in the meta-analysis; their results are reported narratively. The risk of bias was low, medium, and high in 16, 4, and 1 study, respectively, as represented in online supplemental file 2. We meta-analysed 22 studies with 13 392 participants reporting health literacy scores for women (n=8012) and men (n=5380).
Gender differences of health literacy in persons with a migration background
A small but significant gender difference (SMD=0.08; 95% CI 0.002 to 0.159; p=0.04) in health literacy scores could be identified, with women achieving higher scores than men. In S-TOFHLA units, which range from 0 to 36, the mean difference between scores of women and men was 0.90 (95% CI 0.03 to 1.78). A considerable level of heterogeneity (I2=65%) between studies was found. Detailed results and a forest plot are depicted in figure 2.
Two studies not included in the meta-analysis reported health literacy levels instead of scores. Geltman et al88 found significantly (p<0.001) higher health literacy levels in male (n=184; low health literacy: n=118; high health literacy: n=66) compared with female Somali refugees in the USA (n=255; low health literacy: n=208; high health literacy: n=47). Wångdahl et al19 included refugees of diverse origins living in Sweden (n=455; 242 male; 204 female; 9 unknown) and measured functional as well as comprehensive health literacy. They categorised the results into inadequate, problematic and sufficient health literacy: sufficient functional health literacy was found in 17.6% and 22.3% of male and female participants (p=0.06), respectively, while sufficient comprehensive health literacy was reported for 39.3% and 48.1% of male and female participants (p=0.07), respectively. However, neither of these studies posited possible reasons for the gender differences they found.
Analysis of subgroups
Deviating from the protocol, we refrained from conducting tests for subgroup differences based on migratory status (eg, labour migrant, refugee, asylum seeker) as only two studies74 84 included in the meta-analysis reported participants’ migratory status. As we included baseline measures of health literacy only, we did not perform tests for subgroup differences based on the study design. Detailed data and forest plots for the tests on subgroup differences can be found in online supplemental file 3.
Region of origin
Most studies included migrants from diverse countries of origin without reporting separate scores for the respective groups. Gendered scores were not broken down by country of origin. Thus, we categorised the included studies into three groups regarding the following ethnicities:
(1) The category Hispanic/Latinx with 14 studies65 68 69 71 73 74 76 77 79 81 82 86 99 including 10 858 participants (female: n=6593; male: n=4265) who were of Latin American origin (including the Caribbean, Guyana, French Guyana and Suriname) or were denoted as Hispanics or Latinos/Latinas/Latinx in the paper; (2) the category Asian entailing three studies66 80 84 with 982 participants (female: n=658; male: n=324) from Asian countries; and (3) the category Mixed/Others containing five studies70 75 78 83 85 with participants from diverse countries and regions of origin within the same study and one study with participants from the former Soviet Union.83 The Mixed/Other category contained 1552 participants (female: n=761; male: n=791).
Women with a Hispanic/Latinx background scored significantly higher in health literacy than men of the same background (SMD=0.12; 95% CI 0.02 to 0.23; p=0.02) with considerable heterogeneity between studies (I2=71%). No significant gender differences were found for participants of Asian (SMD=−0.01; 95% CI −0.29 to 0.27; p=0.93; I2=75%) and Mixed/Other origins (SMD=0.04; 95% CI: −0.07 to 0.15; p=0.48; I2=14%).
The differences between these subgroups were not significant: Χ²=1.48, df=2 (p=0.48), I²=0%.
Type of health literacy measurement instrument
We grouped instruments to measure health literacy into five main subgroups: BHLS, HLS-EU-Q, SAHLSA/REALM, TOFHLA and a residual category (Others). Within the studies selected for the meta-analysis, six66 67 77 78 83 86 belonged to the BHLS subgroup. For this subgroup, we did not find any significant gender difference in health literacy (SMD=0.015; 95% CI: −0.05 to 0.08; p<0.085, I2=0%). The same was true for the subgroup HLS-EU-Q75 85 100 (SMD=0.04; 95% CI: −0.20 to 0.28; p<0.001, I2=49%), TOFHLA65 68 73 76 79 (SMD=0.08; 95% CI: −0.25 to 0.41; p<0.001, I2=84%) and the residual category Others80 84 99 (SMD=0.08; 95% CI: −0.22 to 0.38; p<0.001, I2=0%). Only the SAHLSA/REALM subgroup71 74 81 82 showed a significant gender difference in health literacy scores (SMD=0.14; 95% CI: 0.09 to 0.20; p<0.001, I2=0%), with women achieving higher scores than men. Again, the differences between these subgroups were not significant: Χ²=9.19, df=4 (p=0.06), I²=56.5%.
Functional versus comprehensive health literacy
Of the studies included in the meta-analysis, 1865–69 71 73 74 76–79 81–84 86 99 reported functional health literacy and 4 studies70 75 80 85 measured comprehensive health literacy. There was no significant gender difference found in the functional (SMD=0.08; 95% CI: −0.01 to 0.17; p=0.10; I2=69%) nor the comprehensive health literacy group (SMD=0.10; 95% CI: −0.08 to 0.28; p=0.03; I2=51%). Here too, the differences between these subgroups were not significant: Χ²=0.03, df=1 (p=0.86), I²=0%.
One study73 was considered to have high risk of bias. When omitting this study from the main analysis, the gender differences in the health literacy of persons with a migration background ceased to be statistically significant (SMD=0.07; 95 % CI: −0.01 to 0.15; p=0.07; I2=66%) and differences between subgroups remained statistically insignificant. As recommended in the Cochrane handbook,101 we also calculated all meta-analyses using a fixed-effects model, which yielded very similar results. Gender differences in the health literacy of persons with a migration background were somewhat more pronounced in the fixed-effects model (SMD=0.10; 95% CI 0.10 to 0.13; p<0.001), and the subgroup differences regarding type of health literacy measurement instrument became significant (p=0.008) with women now also scoring significantly higher in the TOFHLA category (SMD=0.20, 95% CI 0.09 to 0.31, p<0.001). All other subgroup differences remained statistically insignificant.
This systematic review aimed to investigate gender differences of health literacy in individuals with a migration background assessed using standardised instruments. Overall, we found health literacy in female persons with a migration background to be higher than in their male counterparts. Tests for subgroup differences regarding region of origin, type of health literacy (functional or comprehensive) and measurement instrument did not reveal clues to potential explanations for this finding. As further indicators possibly interacting with gender (eg, education) were not reported separately for men and women in the included studies, the reasons for slightly higher health literacy scores in migrant women remain unclear.
Of the 22 studies included in the meta-analysis, 6 reported gender-separated health literacy scores but without providing explanations for possible gender differences or relating them to further criteria such as age or migratory status. The remaining scores were retrieved from the authors of the respective studies (n=15) by request and, in one case,86 taken from a publicly available data set also used by the respective study. With 56 of our 71 data requests remaining unanswered or turned down, there seems to be a high number of unreported cases. We found a further 27 studies that investigated the health literacy of only female migrants but could not include them, as we could not find eligible studies on migrant men. Thus, it appears there is a severe lack of research on health literacy in male migrants. This finding may not be restricted to migrants: a current scoping review found only 12 studies on men’s health literacy worldwide.102 In contrast to that, another systematic review focusing on health literacy in women living in Iran revealed 34 studies.103 Furthermore, even within the studies included in the meta-analysis, the number of female participants (n=8012) far exceeded that of men (n=5380). Thus, in the context of male migration, gender might be a blind spot in health literacy research. This may also be seen from the fact that none of the studies included in this review provided a definition of gender. Most studies (n=30) did not mention how gender was assessed and only one study81 reported having assessed genders beyond the male/female dichotomy. This is in line with the observation that a lack of theoretical foundation and adequate operationalisation of gender still is a common phenomenon in research.104 For example, gender roles are likely to exert an influence on health literacy105 ; for the context of migration this has been described in qualitative studies.11 106 107 Nevertheless, gender roles were not mentioned in the included studies; not considering gender roles appears to be a severe omission in health literacy research.
Implications for research
This systematic review revealed a need for more studies on the influence of gender-specific aspects on the health literacy of persons with a migration background. Future research should provide thorough theoretical foundations for examining gender in this context and operationalise the construct gender accordingly, thus evaluating personal, situational, cultural and societal aspects of gender. This is necessary to explore the influence of gender and its interactions with other factors such as education, age and culture, in relation to health literacy, which suggests the inclusion of a qualitative research methodology.108 For example, a higher health literacy in migrant women indicates that there might be advantageous gender-specific traits or strategies for processing health-related information. These strengths should be further explored to enhance migrant women’s health literacy skills. Furthermore, there is an urgent need for more research on the health literacy of (migrant) men in general, who may have lower health literacy than women. Further research should aim determining the causes of this possible disadvantage and how to enable men to improve their health literacy. Lastly, it is remarkable that sex (as the male/female dichotomy) often seems to be assessed but not as frequently reported. Therefore, publishing data of single studies in publicly available repositories such as the one provided by the Centre for Open Science109 may help researchers investigate relationships beyond the purpose of the respective study, for example, when conducting systematic reviews and meta-analyses.
Implications for practice
Participative intervention development involving the respective migrant communities has proven to be an effective approach to foster health literacy in migrant populations.43 A gender-sensitive methodology might help to further improve the effectivity of such interventions.110 111 Specifically, promoting the health literacy of migrant men and further strengthening that of migrant women seem promising.
Strengths and limitations
To our knowledge, this is the first systematic review to investigate gender differences of health literacy in persons with a migration background. The results of our review revealed important gaps in health literacy research about gender aspects—the first and foremost being the neglect of migrant men’s health literacy. As research on health literacy in non-migrant men is likewise scarce,102 this insight might reach beyond the populations examined. A significant strength of this review lies in the acquisition of unpublished data on gendered health literacy scores, which were retrieved thanks to the cooperation of the authors of the respective studies.
This study also has some important limitations. First, we were unable to retrieve data from 56 studies that could have contributed to the meta-analyses. However, we were able to successfully incorporate previously unpublished data on health literacy scores of men and women from 15 studies into our review. Second, the gender differences of health literacy in migrants we found in our meta-analyses were extremely small (SMD=0.08, p=0.04). On the other hand, health literacy is a complex construct and subject to the influence of many variables,1 which indicates that a small effect might nevertheless be of some importance.112 Furthermore, a small effect gains importance if it is constant and long-lasting113 —as it is with gender, being present throughout a person’s entire life span. Nevertheless, as statistical significance disappeared when excluding the one study considered to have a high risk of bias, these results must be interpreted with caution. Third, the heterogeneity of the studies was high (I2=65%), insinuating a weak comparability of the included studies. This cannot be denied as persons with a migration background are a highly heterogeneous population and health literacy is subject to different definitions and measurement tools. Therefore, this review can only shine a spotlight on gender differences of health literacy in migrants. Nevertheless, our study contributes to the exploration of the continuously evolving concept of health literacy.
The results of this systematic review indicate that migrant women might have higher health literacy than their male counterparts. Furthermore, two research gaps can be identified: (1) a thorough theoretical foundation, operationalisation, analysis and reporting of gender are rarely found in health literacy research on migrants and (2) there is little research on the health literacy of male migrants. Adequately defining, measuring, analysing and reporting gender seem mandatory when designing research and interventions in the realm of health literacy.
Data availability statement
Data are available upon reasonable request. Data may be obtained from a third party and are not publicly available.
Patient consent for publication
We would like to thank the researchers who shared their data upon our request. We also thank our student assistant, Constanze Huebner, for her assistance throughout the project and our colleague, Görkem Anapa, for helping us by providing his expertise on migration.
Contributors NS, CW, EK and ÜSS conceived the overarching project GLIM. DC, AB and AA conceived the study and wrote the protocol. IM designed and conducted the database searches. DC, AB and AA screened the articles for eligibility. DC and AB extracted the data from the individual studies for analysis. DC analysed the data, conducted the meta-analysis, produced forest plots and wrote the first draft of the manuscript. AB double-checked the meta-analysis. All authors revised the manuscript and approved the final version. DC is the guarantor for this work and accepts full responsibility for the conduct of the study, had access to the data and controlled the decision to publish.
Funding This study is part of the GLIM project which was funded by the German Ministry of Education and Research (grant number 01GL1723).
Disclaimer The funding institution had no influence on the study design, data collection and analysis, the decision to publish or the preparation of the manuscript.
Competing interests All authors have completed the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf and declare the receipt of a grant from the German Ministry of Education and Research for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous 3 years and no other relationships or activities that could appear to have influenced the submitted work.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.