Systematic Review of Health Literacy and Health Promotion in School-Aged Adolescents

Background: The literature has identified gaps in adolescent health literacy (AHL) measurements, as well as how the health literacy (HL) level is related to health promotion (HP) aspects. This study aimed to examine the tools used to measure HL and determine its relation with HP among adolescents. Methods: Three online databases (HINARI, PubMed, and DOAJ) were searched to conduct a systematic assessment of papers published between January 1, 2016, and January 1, 2021. In this review, 373 articles were identified from these databases. After removing duplicates and screening titles and abstracts of articles, 49 full texts were selected for full-text reading. After comprehensive reading, 23 papers were appraised for qualitative synthesis. Results: Of the 23 reviewed papers, 21 focused on assessing AHL measures, and 15 addressed the association between AHL and HP. Seven studies used the HL School-Aged Children instrument. The findings suggested that the methodological and conceptual underpinnings of HL measures are insufficient. Furthermore, HL acts as an independent and positive mediator for many facets of HP. Conclusion: This review offers a warning to practitioners and educationists interested in measuring HL as the number of measurement tools is quite huge with different tools applying different scales.

instruments (16). Guo et al conducted a review and found 29 HL instruments, among which functional HL (FHL) comprises half (21). Likewise, Ormshaw et al concluded that the majority (n = 14) assessed HL directly using an objective measure (22). All three studies lacked the quality assessment of studies included with these reviews nor did they cover the relationship between HL and the HP of adolescents. Therefore, it is uncertain which instrument is the most valid, reliable, and practical for use with adolescents.
To address these gaps in understanding, this systematic review has addressed two research questions (RQ), including what instruments are used to measure HL among school-based adolescents and how the HL and HP of adolescents are related to each other. The first part of this review analyses the major characteristics of HL measures, including place of operation, the types of scale/ score, method of assessment, competencies, reliability, the time of administration, methodology, and AHL situation. Similarly, the relationship of the HL with aspects of the HP is analyzed in the second part of the review. Our review of the literature was conducted to rigorously analyze all papers published between 2016 and 2021 to gain a better understanding of the literature. In our opinion, this study may have filled in some gaps in the measurement, research methodology, and how HL is related to HP. In addition, it will provide an opportunity for academics who are interested in collecting and assessing data or using them appropriately.

Materials and Methods
This systematic review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines (23) to ensure a high level of evidence. The PRISMA flow diagram (24) highlights the method of finding and incorporates the relevant articles (16,17).

Data Source and Searching Strategies
This study was performed using three online databases (HINARI, PubMed, and Directory of Open Access Journals) which published the articles between January 1, 2016, and January 1, 2021. Boolean operators (AND/ OR) were applied to combine search words around adolescence (including young people and teenagers), HL/ HP, and relationship/association. Adolescent OR school adolescent OR school teen health literacy AND health promotion were used as key search terms. The first searched term was "adolescent OR school teen health literacy", and the second term was "relationship" OR "correlation" OR "association" with health promotion. Three databases used the following algorithm: {((TitleCombined: (adolescent)) OR (TitleCombined: (school adolescent)) OR (school teen health literacy)) AND (relationship OR Correlation OR Association with health promotion)}

Eligibility Criteria
First, all authors contributed to the formulation of the eligibility criteria. The applied eligible criteria are provided in Table 1.

Screening, Data Extraction, and Analysis
Two separate authors (SPK and CBB) performed the search. Moreover, SPK and JA scanned the title and abstracts for relevance. Further, SPK, JA, and MKS performed the fulltext analysis. In addition, SPK and CW independently checked the result, while OO and EvT assessed the results to reach a consensus around the included studies. References were imported into the electronic database software tool (EndNote X9). Information identified in the relevant publications related to the measures of an HL level and the association between their HL level and HP of school adolescents were independently extracted by SPK and MKS.
The data were extracted from papers based on the characteristics of the included studies (e.g., the first author, publication year, and country, general characteristics of the instruments, psychometric properties, and relationships between HL and HP).
Each author separately evaluated papers for qualifying in accordance with the criteria to reduce the selection bias. To compare their findings, resolve any disagreements, and determine whether each article should be incorporated, we convened a group meeting. The study team considered contrary views and then concluded by consensus.
The data were collected and summarized for each paper, and then the summaries were tabulated by themes, including information such as the name of the author(s), year of publication, type, components, items, and mode of administration of measures, HL level, psychometric properties, and sample size.
The review of the literature yielded 373 articles ( Figure 1). A total of 278 articles were removed after identifying duplicates, and 44 articles were screened and reviewed for abstracts. Additionally, 49 articles were retrieved for full-text reviews. Next, 26 articles were excluded after reviewing the full text. Finally, 23 full-text articles were reviewed based on 373 database searches.

Results
The review of the literature yielded 373 articles ( Figure 1). Overall, 278 articles were removed after identifying duplicates, and a total of 44 articles were screened and reviewed for abstracts, and 49 articles were retrieved for full-text reviews. After reviewing the full text, 26 articles were excluded, and finally, 23 full-text articles were reviewed according to 373 database searches. Among the selected papers, 21 papers were assessed for research question 1 such as measures and the AHL level, and 15 papers described its relationship/association with HP (research question 2).

Health Literacy Measures and Level of Adolescent Health Literacy HL Measure
In total, 21 identified papers were suitable in 13 of these different measures and were identified to assess AHL. Based on data in Table S1, four instruments (25)(26)(27)(28) are new, and nine scales already existed or were adapted. Only three HL scales of Health Literacy Scale-European Union (HLS-EU), Functional Health Literacy (FHLs), and HLS for Japanese adults had been initially developed for adults. The other ten were specially developed for adolescents. Seven papers (29)(30)(31)(32)(33)(34)(35) used HL among School-Aged Children (HLSAC) scale. Three and two studies used Newest Vital Sign (NVS) (36)(37)(38) and Chinese Adolescent Interactive HL Questionnaire (CAIHLO) (39,40), respectively. However, other studies employed a different scale such as HLS-EU (25), an HLS for Thai children who were overweight (41), an HLS developed by Ran (26), the Rapid Estimate of Adolescent Literacy in Medicine Teen (REALM teen) measure (27), an HL screening instrument developed by Chew et al (42), pre-test and post-test tools (43), the HL Assessment Scale for Adolescents into Arabic Language (HAS-A-AR) (44), the FHLs (45), and HL Measure for Adolescent (HELMA) (28).

Population
Included school adolescents aged between 10 and 18 from any setting and any country. Age below 10 or over 18 Intervention Included evaluation and screening of generic HL situations and implementation of educational interventions to increase access, understanding, accounting, and use of health information in at least one aspect of HP was included.
Interventions that did not emphasize HL through education, HP, or public health.

Context
Included school setting studies along with any research designs studies related to generic HL. Domain-specific studies Outcome Aimed to measure the HL level, a psychosomatic test of measures, examine at least one aspect of HP, and analyze it with the relationship between HL an HP Only one outcome of interest was not even included.
Study design Included original paper published in English and in a peer-review study.
Any other language and non-peer-reviewed papers and other book chapters, case studies, protocol papers, and the like.

Methodological Characteristics
In the context of psychometric properties, six studies (27,31,37,42,43,45) did not provide reliability and validity on their instruments. Others reported the reliability of the instruments. Among these, the scales of nine studies had a strong ( > 0.85) value of reliability, and four studies had good reliability. Only one instrument (41) had a weak value ( < 0.70) of reliability. Seven studies did not have well-described psychometric properties and did not mention a reliability value. One study (25) mentioned the validity of the instrument.
Regarding sampling design and size in this review, only 13 papers mentioned the sampling design. Of those, there was greater use of a multi-stage sample (n = 5) (26,28,37,39,40) and cluster (n = 3) (25,32,38). Similarly, four studies used convenience (36,45) and stratified (41,44) sampling methods, and one study employed a snowball (30) sampling method. In the remaining seven reviewed papers (27,29,31,33,35,42,43), the sampling method was not reported at all. With regard to the sample size of the included studies, fourteen studies were conducted using large sample sizes. Only seven studies (27,28,(36)(37)(38)42,45) had a small sample size. The participants of the combined study were 11-19-year-old students studying at 7-12th levels of the school and were the most common groups (Table S1).

Situation of Adolescent Health Literacy
The eight reviewed studies conducted in Thailand (41), China (26), Belgium (33), Czechia (34), Turkey (35), Texas (36), Pakistan (37), and Iran (38) reported that the HL level of the school adolescents was inadequate/marginal. Six studies reported a moderate level of HL. Some papers had not cleared the HL situation of adolescents. There were large percentages of adolescents with inadequate HL in each nation. Several systematic review papers reported the same finding of the high prevalence of adequate HL among adolescents and youths (15,16,46). Eight studies demonstrated a statistically significant correlation between age, parental education, membership in the sports club, ethnic group, gender, grade, medication, older adolescents, countries, and AHL level. In another study, it was found that the perceived critical HL and perceived FHL were not directly related. Hence, the HL level of adolescents is determined by different factors (Table S1).

Health Literacy Level and Health Promotion of Schoolaged Adolescents
Fifteen studies reviewed the relationship of HL with 22 components of HP. Health behavior was included in eight studies (25,27,(29)(30)(31)38,40,41) and showed that some health behaviors were statistically significantly associated with HL. For example, HL is an independent factor/ mediator (1) for increasing HP behavior such as handwashing (30), healthy food (29,32), smoking and alcohol (25,32), and structurally stratified, health behavior, and physical activity (29). In relation to gender and sexual and reproductive health, HL served as a mediator (34) ( Table 2).
Three studies evaluated the significant relationship between health-related quality of life with HL (26,30,42). Likewise, three studies examined the significant relationship of family influence (31,32,34), physiological and mental well-being (26,39), self-care and asthma (27,42), and self-efficacy (38,42) with HL. One paper reported a significant association between HL and the management of a health condition, appropriate health decision-making, physiological, mental, and social well-being (26,39), street management, and health responsibility (35). Two studies showed that school achievement and availability of health provisions in school were significant predictors of HL (29,32).
Conversely, five articles found no association between HL and HP aspects (35,(37)(38)(39)(40). One study represented a weak relationship between HL and nutrition, exercise, social support, and life satisfaction (35). As per another study, HL was negatively associated with physical and psychological symptoms (39). Negative correlations were also observed between HL and health risk behavior (40), and health-seeking behavior (37), as well as daily physical exercise, self-efficacy, and four levels of the body mass index (38).

Discussion
This systematic review investigated HL status and how it is measured in the current studies of adolescent HL. The results will also be utilized to determine which scales are currently in use, as well as how they were developed, validated, tested, and employed in this regard (16). Second, the study sought to understand the relationship between adolescent HP and HL levels. According to the number of papers published worldwide during the given time period, the study of adolescent HL is progressing at an encouraging rate.
The review identified 13 HL measures that were used to assess the HL of adolescents in various settings. This inductive work in the field of HL demonstrates that there has been a noteworthy increase in the number of different HL measures for adolescents in the recent academic literature. It was found that most (n = 8) studies Systematic review of HL and health promotion HL, member sports club, physical activity, and family affluence. Perceived HL was higher among adolescents who participated in sports club activities (P < 0.001).

Multilevel mixed-effects logistic regression analyses
Intarakamhang and Intarakamhang (41) To develop a path model of HL for obesity preventive behaviors HL and obesity preventive behaviors, (eating, exercise, and emotional coping) HL affects the development of obesity prevention behaviors in three different paths.

Structural equation models and confirmatory factor analysis
Ran et al (26) To explore the relationship between QOL and different levels of HL HL, QOL, psychological wellbeing, and mental well-being Students who were equipped with higher HL had greater QOL (P < 0.01), and this discrimination remained significant in subscales such as physiological well-being (P < 0.01), mental well-being (P < 0.01), social well-being (P < 0.01), and pubertal wellbeing (P < 0.01).
Chi-squared tests, t-tests, or F tests were employed to examine the unadjusted associations and regression analysis Valerio et al (27) To investigate the relationships between HL and self-care for asthma. HL, asthma self-care, and self-efficacy REALM-Teen scores were strongly correlated with asthma knowledge and self-efficacy.
Bivariate associations of each predictor to the REALM-Teen through mixed-effect regression models.
Sukys et al (32) To examine the relationship between HL, school achievement, health education in schools, and family economic status.
HL, school achievement,health education (HE) in school, and family affluence A significant predictor of HL was school achievement (Pearson's r = 0.26). Family affluence (r = 0.12) also indicated a higher level of HL.  Descriptive, ANOVA, Spearman's correlation coefficient, and a two-sided P-value employed subjective HL measures, and the HLSAC was the most common one. Additionally, this study reported only five focused FHL instruments, and NVS was the most commonly used one. Many papers apply these two instruments because they are short and accessible. This result does not correspond with earlier findings (16,21,22). In the current review, as compared to earlier studies, the use of multidimensional tools appears to have intensified somewhat from a multidimensional perspective of adolescent HL tools. However, they only reperesented five of the 13 instruments. This review revealed the narrow focus of AHL measures as a solid research inconsistency similar to other studies, implying that the context should not be overlooked when establishing a tool based on a broad theoretical foundation of HL. Thus, the design of multidimeniosal and context-specific tools becomes desirable. This paper identified fewer HL scales than three other papers (16,21,22) exploring similar issues and more than one other previously conducted review paper (47). These discrepancies could be due to fewer database searches being included in the review and the timing of the review. In addition, only generic HL measures were included. Similarly, the other studies also included children in the sample. Most instruments used a broad and multifaceted notion of HL. It is noted that the number of AHL tools seems to have increased over time. This current review shows that the paradigm of HL measurement is now shifting from objective to subjective. This review identified a total of five new HL tools. Although all the identified scales were constructed in the context of schools, they did not conform to the school context and HP principle. The measures of HL exist for adolescents, but they are not generally accepted across settings. This may have posed a problem for this area of study. Therefore, more research is needed to involve schools' context, HP principles, and social determinants.
For adolescents, eight self-reported measures of HL were identified in this review. Likewise, few studies utilized tools mainly developed and validated for measuring the HL of adult people. This situation raises the question of whether an adult-specific questionnaire based on a subjective or objective approach can be effectively adopted to measure AHL. Moreover, most studies employed tools developed by others. Can scale made in one country or place be valid in another? Therefore, the HL measure is especially significant when analyzed in a practice-topractice situation. In the context of this review, there is a great deal of confusion about which of the 13 tools used in the various studies would be viable. Hence, there is a need to build context-based HL scales based on both subjective and objective perspectives. The findings of this study provided basic information on how to develop, test, use, and measure the reliability of HL tools. However, the full details of many of the applied tools have not been stated. The characteristics of the existing scales (e.g., their item number, eligibility requirements, scoring methods, Systematic review of HL and health promotion and administration) contain several contradictions and discrepancies. Thus, making the best tool selections is difficult. However, the results of this study will be taken into account to include more details for anyone who wishes to learn more about HL. There is a need to develop and validate a comprehensive or collective HL scale for adolescents contextually using HP perspectives. Based on the findings of this review, most studies (n = 19) had a cross-sectional design that only provides baseline data. Most studies reported the internal consistency of the applied instruments. Seven studies did not have welldescribed psychometric properties or failed to mention them. Only one study mentioned the validity of the instrument. HL scales focused on the quantitative validity and reliability of measures for functional, communicative, and critical HL. Due to the lack of methodological aspects and the quality of the research reports, most information on HL scale features remained unclear (21). There was a lack of qualitative measures of critical and interactive HL (4). Mixed-method and interventional study designs are also necessary.
In general, eight studies reported that the HL level of school adolescents was limited in certain countries. The same finding was also provided, indicating the high prevalence of limited HL among adolescents and youths (46,48). There is no evidence of AHL in many countries. Eight studies reported that there are statistically significant correlations with personal factors. The relationship between age, gender, and HL was inconsistent across studies (47), while another review's finding concurred with this (46).
To conclude this review, 15 studies comprised the relationship of HL with 22 components of HP. Overall, 13 articles examined the relationship of HL with different dimensions of health behaviors of school adolescents. Of these, eight cases showed that health behavior was statistically significantly associated with HL. HL is an independent and positive factor as a mediator for health-promoting behavior. The other reviewed papers demonstrated that there is a meaningful relationship between HL and adolescents' health behaviors (47), and poorer HL was associated with some adverse health outcomes (46). The findings of this review also represented that studies included only limited aspects of HP. No research was identified on the relationship between HL and HP in adolescents, highlighting the need for examining the relationship between HL and HP among adolescents.

Limitations and Strengths
The first limitation of this review was the omission of any examination of the conceptual perspective of HL. Second, only English-language HL papers were examined in this review. Furthermore, we did not conduct a quality assessment screening of the addressed studies. It was difficult to compare the findings because the age ranges, maturation, cognition, abilities, attitude, self-efficacy, educational level, decision-making, autonomy, and experiences varied between investigations. A further limitation of this study's search strategy was that it was only applied for five years. Only three reviewable datasets were employed in this investigation. However, to the best of our knowledge, this is the first study to identify the measurement techniques applied in AHL studies, which is considered a strong point. It also established the connection between HP and AHL.

Recommendations
In this study, HP principles were proposed as a basis for developing relevant and contextual measures to assess the level of HL among adolescents. The low level of HL among adolescents calls for the development and implementation of HL programs for them as suggested by Manganello (49). Despite its importance in improving people's health, HL lacks significant research, which highlights the need to incorporate and study the topic in the near future. Furthermore, compared to cross-sectional studies, experimental and qualitative studies are more limited, suggesting the need for such studies.

Conclusion
In the review, it was suggested that AHL measurement tools are becoming increasingly popular with different types of measurements, although the characteristics of the tools differ widely. As a result, researchers interested in this field face both opportunities and challenges. In addition, the evaluation of contemporary papers demonstrates evidence that children have a low HL. The findings of the current review study highlighted the need for interventional and action research for AHL promotion. It was concluded that the HL of school adolescents is statistically significant for their HP. It is an independent factor and a positive mediator in numerous areas of HP. The results of this review revealed that there is a scarcity of HL and HP research among school adolescents, emphasizing the need to conduct action and interventional research on the relationship between literacy and HP in the future, including multi-levels and multi-factors of HL.