Abstract
Background
Health literacy and health behaviors are closely related and play key roles in health promotion. Previous studies have reported that low health literacy is associated with unhealthy behaviors. It is important to identify which latent dimensions of health literacy are particularly associated with health behaviors. This study aimed to examine the association between components of health literacy and health behaviors among South Korean adults.
Methods
A cross-sectional study design was used with data from the Korea Health Panel Survey. The principal component analysis was applied to the HLS-EU-Q16 to extract dimensions of health literacy, and multivariate logistic regression was conducted to assess their associations with health behaviors.
Results
Health literacy tertiles were significantly associated with socioeconomic status, insurance coverage, chronic disease, disability, self-rated health status, and health behaviors (p < .05). Principal component analysis identified two distinct components. Component 1 represented active engagement and the ability to access, appraise, and apply health information. Component 2 reflected a compliance-oriented dimension involving understanding and applying health-related information. Individuals with higher levels of active engagement were more likely to participate in regular physical activity (OR = 1.08, p < .01) and to attempt weight control (OR = 1.05, p < .05). By contrast, compliance-oriented health literacy was negatively associated with regular physical activity (OR = 0.93, p < .01) and positively associated with current alcohol consumption (OR = 1.10, p < .01). Similar patterns were observed in analyses stratified by educational attainment, particularly among individuals without a college education.
Conclusions
This study used principal component analysis to identify two underlying components of health literacy and demonstrated that each component had a distinct relationship with health behaviors. The findings highlight the need for policies that foster proactive health literacy and encourage active information-seeking.
Keywords: Health literacy, Healthy weight, Smoking behavior, Alcohol consumption, Principal component analysis
Background
Social determinants of health refer to a range of social and economic factors, such as economic stability, access to quality education and healthcare, neighborhood conditions, and community support, that shape health outcomes and contribute to health inequities [1]. Researchers have examined how these determinants contribute to inadequate access to healthcare and poorer health among individuals with low socioeconomic status [2]. Among these determinants, particular attention has been paid to the potential of health education to promote health and prevent disease, with health literacy increasingly recognized as a means of enhancing the effectiveness of health education and improving health outcomes [2, 3].
Health literacy is one of the key outcomes of health promotion. It refers to the cognitive and social skills that determine an individual’s ability and motivation to access, understand, and use information to promote and maintain health [4]. Sørensen et al. (2012) defined health literacy as the capacity to access, understand, evaluate, and apply health-related information for health management, disease prevention, and health promotion, highlighting its growing importance in public health and healthcare [5].
Various tools have been developed to assess health literacy worldwide. In South Korea, researchers have used instruments such as the European Health Literacy Survey Questionnaire [6, 7], the Newest Vital Sign [8], the measure developed by Chew, Bradley, and Boyko (2004) [9], and the Korean Health Literacy Assessment Tool [10]. At the national level, the 2021 Korea Community Health Survey has been used to evaluate linguistic and textual health literacy, and the Korea Health Panel Survey (KHPS) has been used to assess health literacy based on the 16-item European Health Literacy Survey Questionnaire (HLS-EU-Q16).
Health literacy refers to an individual’s overall ability to process health-related information for health promotion, whereas health behaviors are key determinants that directly influence health outcomes [10, 11]. The 1965 Alameda cohort study demonstrated that healthy habits, such as regular meals, adequate sleep, maintaining a healthy weight, abstaining from smoking, moderating alcohol consumption, and engaging in regular physical activity, significantly extend life expectancy [12]. In South Korea, the socioeconomic costs of tobacco use, alcohol consumption, and obesity amounted to approximately 31 billion USD in 2021 [13]. Globally, estimates indicate that, annually, an elevated body mass index (BMI) contributes to 3.7 million deaths due to noncommunicable diseases, alcohol consumption contributes to 2.6 million deaths, and smoking contributes to over 7 million deaths, including 1.6 million non-smokers [14, 15, 16].
Building on these findings, this study examined the association between health literacy and health behaviors to assess their potential public health contributions. Specifically, we focused on three key lifestyle behaviors related to health literacy: maintaining a healthy weight, alcohol consumption, and smoking.
In exploring the relationship between health literacy and health behaviors, we identified opportunities to make methodological contributions. Most previous studies that analyzed the HLS-EU-Q16 data converted responses into binary scores, assigning 0 points to “very difficult” and “fairly difficult” and 1 point to “fairly easy” and “very easy”. Total scores were grouped into three categories: adequate (13–16), problematic (9–12), and inadequate (0–8). Although this is a relatively straightforward approach to summarizing data, it may oversimplify individuals’ complex responses to the 16 items, potentially resulting in information loss. Furthermore, a previous study that applied confirmatory factor analysis (CFA) to health literacy supported the view that health literacy is not a unidimensional construct and explored the lower-dimensional structure of the HLS-EU-Q [7]. Therefore, rather than simply categorizing HLS-EU-Q16 scores into predefined groups, we focused on identifying the latent structure underlying the items to assess health literacy.
Sørensen et al. (2012) proposed a conceptual model of health literacy encompassing four competencies—accessing or obtaining, understanding, processing or appraising, and applying or using information—across three domains: healthcare, disease prevention, and health promotion [5]. Building on this framework, prior research mapped each item of the HLS-EU-Q16 onto the corresponding domains and competencies [17]. This study aims to examine whether identifying the latent structure of health literacy from empirical data can provide meaningful new insights, rather than relying solely on structures derived deductively from theoretical models.
Accordingly, we applied principal component analysis (PCA) to reduce the dimensionality of the data while preserving its underlying patterns and structure, thereby summarizing its key features [18]. Some studies have applied PCA to the HLS-EU to synthesize the diverse health literacy items and assess the reliability, validity, or structure [19, 20]. However, the primary focus of prior research using PCA has been on evaluating the psychometric properties of health literacy items. Therefore, we used PCA to statistically distinguish different dimensions of health literacy, explore their underlying meanings, and examine their relationships with health behaviors. Finally, drawing on previous research, including Nutbeam (2000), which highlighted the close relationship between education and health literacy, we examine heterogeneity in the association between health literacy and health behaviors across educational levels [3].
Methods
This cross-sectional study used the data from the 2021 KHPS [21]. In this study, responses marked as “don’t know” were treated as missing values [22], and respondents with missing data for any of the 16 items were excluded from our PCA. Cronbach’s alpha was 0.94, indicating a high level of internal consistency in the response data.
This study applied PCA to identify correlations among the HLS-EU-Q16 items, enabling the extraction of principal components that effectively capture the underlying structure of the items. The components were then used as independent variables to analyze their associations with the dependent variables.
The dependent variables, representing various aspects of health behavior, were measured using several indicators of the three key lifestyle habits mentioned above. Participants’ regular physical activity was assessed using self-reported regular exercise during the past year. Weight control efforts refer to attempts to lose, maintain, or gain weight. Alcohol consumption was assessed by categorizing those who consumed alcohol regularly as regular drinkers and those who did not drink regularly or ever as non-drinkers. Smoking status was defined as a current smoker for regular smokers and a non-smoker for both former smokers and never-smokers. Smoking cessation attempts were analyzed among current smokers. Based on previous literature, this study included the following covariates: gender, age, marital status (married and living together), educational attainment (college or higher), health insurance type (National Health Insurance or Medial Aid), private health insurance enrollment, participation in economic activities, household income, place of residence (one of seven major cities in South Korea), subjective health status, and presence of disabilities and chronic conditions among 30 major diseases [23, 24, 25, 26].
Socioeconomic characteristics, health-related variables, and HLS-EU-Q16 score tertiles were analyzed using descriptive statistics and chi-square tests. Logistic regression models were used to explore the associations between the identified principal components of health literacy and health behaviors, accounting for sample weights. All data were analyzed using Stata/MP 17 software, and a p-value < 0.05 was considered statistically significant.
Results
In the 2021 KHPS, a total of 12,874 participants responded to the questionnaire. After excluding observations with missing data on explanatory variables, those who reported being bedridden due to illness or injury, and those with incomplete responses to the 16 health literacy items, the final analytic sample included 7178 respondents (Fig. 1).
Fig. 1.
Study population inclusion/exclusion criteria
Table 1 presents the distributions of sociodemographic, socioeconomic, and health-related characteristics according to response completeness for the 16 health literacy items, based on chi-square tests. Small but statistically significant differences were observed by gender, age, education level, private health insurance coverage, economic status, and residence in a metropolitan city at the 5% significance level. Participants aged 65 years or older, those without higher education, those not covered by private health insurance, those in the lowest income tertile, and those residing outside metropolitan cities were slightly more likely to have missing responses to the health literacy items and were therefore less likely to be included in the analytic sample. These differences suggest that relatively vulnerable groups may have been underrepresented in the analytic sample, potentially introducing selection bias and warranting caution in the interpretation of the findings.
Table 1.
Participant characteristics by response completeness
| Variables | All | Answer | p | |||
|---|---|---|---|---|---|---|
| N | < 16 | = 16 | ||||
| Sociodemographic and Socioeconomic Characteristics | ||||||
| Gender | Female | 4596 | 595(13.0) | 4001(87.0) | 0.006 | |
| Male | 3739 | 562(15.0) | 3177(85.0) | |||
| Age (years) | 19–34 | 797 | 120(15.1) | 677(84.9) | < 0.001 | |
| 35–49 | 1652 | 182(11.0) | 1470(89.0) | |||
| 50–64 | 2420 | 311(12.9) | 2109(87.1) | |||
| ≥ 65 | 3466 | 544(15.7) | 2922(84.3) | |||
| Married and Living together | Yes | 5915 | 796(13.5) | 5119(86.5) | 0.080 | |
| No | 2420 | 361(14.9) | 2059(85.1) | |||
| Educational level of college or higher | Yes | 2826 | 351(12.4) | 2475(87.6) | 0.006 | |
| No | 5509 | 806(14.6) | 4703(85.4) | |||
| Health insurance | Yes | 8027 | 1115(13.9) | 6912(86.1) | 0.899 | |
| No | 308 | 42(13.6) | 266(86.4) | |||
| Private health insurance | Yes | 6246 | 822(13.2) | 5424(86.8) | 0.001 | |
| No | 2089 | 335(16.0) | 1754(84.0) | |||
| Economic activity | Yes | 5579 | 785(14.1) | 4794(85.9) | 0.477 | |
| No | 2756 | 372(13.5) | 2384(86.5) | |||
| Economic status | T1 | 3141 | 490(15.6) | 2651(84.4) | 0.002 | |
| T2 | 2730 | 347(12.7) | 2383(87.3) | |||
| T3 | 2464 | 320(13.0) | 2144(87.0) | |||
| Residence in a Metropolitan City | Yes | 3331 | 357(10.7) | 2974(89.3) | < 0.001 | |
| No | 5004 | 800(16.0) | 4204(84.0) | |||
| Health-Related Characteristics | ||||||
| Chronic diseases | Yes | 4673 | 676(14.5) | 3997(85.5) | 0.081 | |
| No | 3662 | 481(13.1) | 3181(86.9) | |||
| Disability | Yes | 574 | 87(15.2) | 487(84.8) | 0.360 | |
| No | 7761 | 1070(13.8) | 6691(86.2) | |||
| Self-rated health status | Very good or good | 3172 | 386(12.2) | 2786(87.8) | 0.001 | |
| Moderate | 3797 | 582(15.3) | 3215(84.7) | |||
| Very poor or poor | 1366 | 189(13.8) | 1177(86.2) | |||
| Maintain a healthy weight | Regular physical activity | Yes | 4346 | 534(12.3) | 3812(87.7) | < 0.001 |
| No | 3989 | 623(15.6) | 3366(84.4) | |||
| Weight control attempts | Yes | 5089 | 682(13.4) | 4407(86.6) | 0.113 | |
| No | 3246 | 475(14.6) | 2771(85.4) | |||
| Alcohol consumption | Yes | 5164 | 720(13.9) | 4444(86.1) | 0.836 | |
| No | 3171 | 437(13.8) | 2734(86.2) | |||
| Smoking | Current smoking | Yes | 1313 | 202(15.4) | 1111(84.6) | 0.086 |
| No | 7022 | 955(13.6) | 6067(86.4) | |||
| Smoking cessation attempts | Yes | 205 | 40(19.5) | 165(80.5) | 0.075 | |
| No | 1108 | 162(14.6) | 946(85.4) | |||
| N | 8335 | 1157(13.9) | 7178(86.1) | |||
Values are presented as counts, with row percentages shown in parentheses
Table 2 presents differences in participants’ general characteristics across HLS-EU-Q16 score tertiles. Table 3 presents differences in health-related variables across health literacy tertiles. All variables differed significantly across health literacy tertiles, except for regular physical activity. Health literacy was lower among individuals aged 65 years and older, without health insurance or private health insurance, with disabilities or chronic diseases, and with poor subjective health status. Higher educational and economic levels were associated with greater health literacy.
Table 2.
Descriptive statistics for general characteristics and health literacy
| Variables | All | Health Literacy | p | |||
|---|---|---|---|---|---|---|
| N | T1 | T2 | T3 | |||
| Total | 7178 | 2429(33.8) | 2644(36.8) | 2105(29.3) | ||
| Gender | Female | 4001 | 1528(38.2) | 1433(35.8) | 1040(26.0) | < 0.001 |
| Male | 3177 | 901(28.4) | 1211(38.1) | 1065(33.5) | ||
| Age (years) | 19–34 | 677 | 44(6.5) | 271(40.0) | 362(53.5) | < 0.001 |
| 35–49 | 1470 | 147(10.0) | 582(39.6) | 741(50.4) | ||
| 50–64 | 2109 | 512(24.3) | 902(42.8) | 695(32.9) | ||
| ≥ 65 | 2922 | 1726(59.1) | 889(30.4) | 307(10.5) | ||
|
Married and Living together |
Yes | 5119 | 1634(31.9) | 1969(38.5) | 1516(29.6) | < 0.001 |
| No | 2059 | 795(38.6) | 675(32.8) | 589(28.6) | ||
| Educational level of college or higher | Yes | 2475 | 263(10.6) | 979(39.6) | 1233(49.8) | < 0.001 |
| No | 4703 | 2166(46.1) | 1665(35.4) | 872(18.5) | ||
| Health insurance | Yes | 6912 | 2283(33.0) | 2564(37.1) | 2065(29.9) | < 0.001 |
| No | 266 | 146(54.9) | 80(30.1) | 40(15.0) | ||
| Private health insurance | Yes | 5424 | 1391(25.6) | 2191(40.4) | 1842(34.0) | < 0.001 |
| No | 1754 | 1038(59.2) | 453(25.8) | 263(15.0) | ||
| Economic activity | Yes | 4794 | 1371(28.6) | 1838(38.3) | 1585(33.1) | < 0.001 |
| No | 2384 | 1058(44.4) | 806(33.8) | 520(21.8) | ||
| Economic status | T1 | 2651 | 1480(55.8) | 796(30.0) | 375(14.2) | < 0.001 |
| T2 | 2383 | 637(26.7) | 992(41.6) | 754(31.6) | ||
| T3 | 2144 | 312(14.6) | 856(39.9) | 976(45.5) | ||
| Residence in a Metropolitan City | Yes | 2974 | 855(28.7) | 1132(38.1) | 987(33.2) | < 0.001 |
| No | 4204 | 1574(37.4) | 1512(36.0) | 1118(26.6) | ||
Values in the T1–T3 columns are presented as counts, with row percentages shown in parentheses
Table 3.
Descriptive statistics for health-related variables and health literacy
| Variables | All | Health Literacy | p | |||||
|---|---|---|---|---|---|---|---|---|
| N | T1 | T2 | T3 | |||||
| Total | 7178 | 2429(33.8) | 2644(36.8) | 2105(29.3) | ||||
| Chronic Diseases | Yes | 3997 | 1922(48.1) | 1348(33.7) | 727(18.2) | < 0.001 | ||
| No | 3181 | 507(15.9) | 1296(40.7) | 1378(43.3) | ||||
| Disability | Yes | 487 | 288(59.1) | 136(27.9) | 63(12.9) | < 0.001 | ||
| No | 6691 | 2141(32.0) | 2508(37.5) | 2042(30.5) | ||||
| Self-Rated Health Status | Very good or good | 2786 | 613(22.0) | 1094(39.3) | 1079(38.7) | < 0.001 | ||
| Moderate | 3215 | 1072(33.3) | 1264(39.3) | 879(27.3) | ||||
| Very poor or poor | 1177 | 744(63.2) | 286(24.3) | 147(12.5) | ||||
| Maintain a healthy weight | Regular physical activity | Yes | 3812 | 1267(33.2) | 1450(38.0) | 1095(28.7) | 0.079 | |
| No | 3366 | 1162(34.5) | 1194(35.5) | 1010(30.0) | ||||
| Weight control attempts | Yes | 4407 | 1165(26.4) | 1736(39.4) | 1506(34.2) | < 0.001 | ||
| No | 2771 | 1264(45.6) | 908(32.8) | 599(21.6) | ||||
| Alcohol consumption | Yes | 4444 | 1140(25.7) | 1747(39.3) | 1557(35.0) | < 0.001 | ||
| No | 2734 | 1289(47.2) | 897(32.8) | 548(20.0) | ||||
| Smoking | Current Smoking | Yes | 1111 | 284(25.6) | 431(38.8) | 396(35.6) | < 0.001 | |
| No | 6067 | 2145(35.4) | 2213(36.5) | 1709(28.2) | ||||
| Smoking cessation attempts | Yes | 165 | 36(21.8) | 59(35.8) | 70(42.4) | < 0.134 | ||
| No | 946 | 248(26.2) | 372(39.3) | 326(34.5) | ||||
Values in the T1–T3 columns are presented as counts, with row percentages shown in parentheses
Two components were retained based on the Kaiser criterion (eigenvalues > 1) and visual inspection of the scree plot. Following promax rotation, items with component loadings of 0.30 or higher were considered salient. The results of the analysis are presented in Table 4. The following items with significant implications were adopted: Items 1, 2, 5, 6, 8, 11, 12, and 13 for Component 1 and Items 3, 4, 7, 10, and 14 for Component 2.
Table 4.
Component loadings of the HLS-EU-Q16 items after promax rotation
| Item | Component | Competency* | |
|---|---|---|---|
| 1 | 2 | ||
| 1. Find information on the treatment of the illness that concerns you? | 0.337 | -0.055 | Obtain/Access |
| 2. Find out where to get professional help when you are ill? | 0.311 | -0.030 | Obtain/Access |
| 3. Understand what your doctor says to you? | 0.010 | 0.396 | Understand |
| 4. Understand your doctor’s or pharmacist’s instructions on taking prescribed medicine? | -0.058 | 0.487 | Understand |
| 5. Judge when you may need a second opinion from another doctor? | 0.323 | -0.045 | Process/Appraise |
| 6. Use the doctor’s information to decide about your illness? | 0.319 | -0.007 | Apply/Use |
| 7. Follow instructions from your doctor or pharmacist? | -0.047 | 0.458 | Apply/Use |
| 8. Find information on how to manage mental health problems like stress or depression? | 0.332 | -0.031 | Obtain/Access |
| 9. Understand health warnings about behavior such as smoking, low physical activity, and drinking too much? | 0.160 | 0.215 | Understand |
| 10. Understand why you need health screenings? | -0.017 | 0.435 | Understand |
| 11. Judge if the information on the health risks in the media is reliable? | 0.307 | -0.026 | Process/Appraise |
| 12. Decide how you can protect yourself from illness based on information in the media? | 0.332 | -0.034 | Apply/Use |
| 13. Find out about activities that are good for your mental well-being? | 0.324 | -0.016 | Obtain/Access |
| 14. Understand advice on health from family members or friends? | 0.025 | 0.375 | Understand |
| 15. Understand information in the media on how to get healthier? | 0.253 | 0.098 | Understand |
| 16. Judge which everyday behavior is related to your health? | 0.265 | 0.063 | Process/Appraise |
| Eigenvalue | 8.95 | 1.50 | |
| Proportion of variance explained | 0.56 | 0.09 | |
| Cumulative variance explained | 0.56 | 0.65 | |
*The 16 items of the HLS-EU-Q16 have been identified as measuring specific competencies related to health literacy. Sørensen et al. (2013) identified four competency domains: access, understand, process/appraise, and apply/use
Items 9, 15, and 16 did not exceed 0.3, the minimum threshold used to indicate a meaningful loading in the PCA, on either of the two components. This suggests that the two components selected using the scree plot and Kaiser criterion were only weakly associated with these items. Although this pattern may indicate the presence of an additional latent dimension of health literacy, it also suggests that, according to our criteria for determining the number of components, this additional dimension did not contribute substantially to the overall structure of health literacy.
The components were interpreted using two approaches. First, the components were assigned meanings and compared with the conceptual framework proposed by Sørensen et al. (2013), partially adopting the methodology of Gustafsdottir et al. (2020) and Kim S et al. (2019) [7, 19, 27]. Based on previous studies that categorized the HLS-EU-Q16 items into domains according to this framework [17], the components were interpreted as follows. Component 1 mainly represented the ability to access, appraise, and apply health-related information. In contrast, Component 2 mostly reflected the ability to understand and apply health-related information.
Furthermore, we derived the conceptual meaning of each component by examining the common themes and similarities among the items that loaded on it. Consequently, each component was interpreted as representing the participants’ underlying attitudes toward health information. Component 1 primarily reflected an active and leading role in assessing and evaluating health information. In contrast, Component 2 appeared to reflect a compliance-oriented tendency, as it was more closely associated with understanding and following expert advice and with interpreting information from family or friends.
Based on a comprehensive review of the competency domains used to measure health literacy in previous studies, together with the pattern of loadings in the PCA described above, we interpreted the two components presented in Table 4 as follows: Component 1 was interpreted as reflecting participants’ agency in accessing, appraising, and applying health information, whereas Component 2 was interpreted as reflecting a compliance-oriented tendency in understanding and applying such information.
Table 5 presents the results of the logistic regression analyses examining the associations between the identified principal components and health behaviors. Component 1 was positively associated with regular physical activity (OR = 1.08, p < .01) and weight-control attempts (OR = 1.05, p < .05). In contrast, Component 2 was negatively associated with regular physical activity (OR = 0.93, p < .01) and positively associated with current drinking (OR = 1.10, p < .01).
Table 5.
Logistic regression of health literacy components and health behaviors
| Variables (N = 7178) | OR (95% CI) | ||||||
|---|---|---|---|---|---|---|---|
| Component 1 | Component 2 | ||||||
| Alcohol consumption | Yes | 0.97 | (0.92–1.01) | 1.10** | (1.05–1.16) | ||
| No | (reference) | ||||||
| Smoking | Current smoking | Yes | 1.01 | (0.95–1.07) | 0.96 | (0.89–1.03) | |
| No | (reference) | ||||||
| Smoking cessation attempts† | Yes | 1.12 | (0.97–1.30) | 0.89 | (0.74–1.06) | ||
| No | (reference) | ||||||
| Regular physical activity | Yes | 1.08** | (1.04–1.13) | 0.93** | (0.89–0.98) | ||
| No | (reference) | ||||||
| Weight control attempts | Yes | 1.05* | (1.01–1.10) | 1.04 | (0.99–1.09) | ||
| No | (reference) | ||||||
p < .05*, p < .01**
† N = 1111
Given the well-documented association between educational attainment and health literacy [28, 29], we conducted stratified analyses by educational level (Table 6). In the lower educational attainment group, alcohol consumption was significantly associated with Component 2 (OR = 1.11, p < .01). Regular physical activity was also significantly associated with both Component 1 (OR = 1.10, p < .01) and Component 2 (OR = 0.93, p < .05), and weight-control attempts were associated with Component 1 (OR = 1.09, p < .01). In the higher educational attainment group, smoking cessation attempts were associated with Component 1 (OR = 1.26, p < .05) and Component 2 (OR = 0.74, p < .05). In addition, weight-control attempts were associated with Component 2 (OR = 1.10, p < .05).
Table 6.
Logistic regression of health literacy components and health behaviors by educational level
| Variables | OR (95% CI) | |||||
|---|---|---|---|---|---|---|
| College or higher (N = 2475) | Less than college (N = 4703) | |||||
| Component 1 | Component 2 | Component 1 | Component 2 | |||
| Alcohol consumption | Yes | 0.91 | 1.11 | 1.00 | 1.11** | |
| (0.83–1.01) | (0.99–1.24) | (0.95–1.05) | (1.05–1.17) | |||
| No | (reference) | |||||
| Smoking | Current smoking | Yes | 0.99 | 0.93 | 1.01 | 0.99 |
| (0.89–1.10) | (0.82–1.05) | (0.93–1.10) | (0.90–1.09) | |||
| No | (reference) | |||||
| Smoking cessation attempts† | Yes | 1.29* | 0.74* | 0.98 | 1.07 | |
| (1.04–1.60) | (0.57–0.96) | (0.81–1.20) | (0.84–1.36) | |||
| No | (reference) | |||||
| Regular physical activity | Yes | 1.07 | 0.94 | 1.10** | 0.93* | |
| (0.99–1.14) | (0.87–1.02) | (1.04–1.15) | (0.88–0.99) | |||
| No | (reference) | |||||
| Weight control attempts | Yes | 0.98 | 1.10* | 1.09** | 1.01 | |
| (0.91–1.07) | (1.01–1.20) | (1.04–1.15) | (0.96–1.08) | |||
| No | (reference) | |||||
p < .05*, p < .01**
† N = 433 (College or higher), N = 678 (Less than college)
Discussion
This study analyzed the associations between health literacy components and health behaviors among South Korean adults. This investigation assigned meanings to the components and incorporated them into a regression model to estimate their relationships.
This study revealed significant associations between HLS-EU-Q16 score tertiles, socioeconomic characteristics, and health-related variables, including age, educational and economic levels, insurance coverage, presence of diseases, self-rated health status, and health behaviors. Our results align with those of previous research showing that limited health literacy is associated with older age, lower education, lower income, and perceived poor health status [23, 28, 29, 30].
This study found that a compliance-oriented tendency toward provided health-related information was positively associated with current alcohol consumption, and a similar pattern was observed among individuals with lower educational attainment. This finding appears to contrast with previous research suggesting that frequent smoking and drinking are more strongly associated with the ability to locate health-related information than with the ability to understand, appraise, or apply such information [31]. Further research is therefore needed to better understand how different dimensions of health literacy are related to alcohol consumption.
In contrast, among individuals with higher educational attainment, an active orientation toward seeking health information was positively associated with smoking cessation attempts, whereas a compliance-oriented tendency was negatively associated with such attempts. Prior studies have reported that higher educational attainment is associated with a greater likelihood of smoking cessation [32, 33]. In addition, Upadhyay et al. (2019) found that seeking health information was positively correlated with intention to quit smoking [34]. These findings suggest that smoking cessation may be more difficult to achieve than other health-related behaviors and that attempts to quit smoking are shaped by a complex set of individual and contextual factors.
The findings for physical activity were particularly noteworthy. Individuals with a greater tendency to seek out and evaluate health information were more likely to engage in regular physical activity, whereas those with a stronger tendency to comply with provided information were less likely to do so. The patterns were exclusively observed in the lower educational attainment group, suggesting that attitudes toward health information may be more strongly associated with regular physical activity among individuals with lower educational attainment than among those with higher educational attainment.
In particular, the negative association between regular physical activity and Component 2—especially the statistically significant association observed among individuals without a college education—merits attention. Because physical activity is a self-directed health behavior that requires individuals to allocate time, make plans, and sustain their efforts independently, this finding suggests that compliance-oriented health literacy, as represented by Component 2, may be insufficient to promote more active health behaviors such as exercise. Consistent with a previous study that distinguished health literacy from patient empowerment in relation to health behaviors, the type of health literacy captured by Component 2—reflecting accurate understanding of information and adherence—may be less conducive to behaviors such as exercise, which require a higher degree of self-determination [35].
A similar pattern of heterogeneity was observed for weight-control attempts. In the lower educational attainment group, the likelihood of attempting weight control increased with Component 1, whereas in the higher educational attainment group, it increased with Component 2. These findings suggest that individuals with lower educational attainment may be more likely to engage in beneficial health behaviors when they adopt a more active, self-directed approach to health information. In contrast, those with higher educational attainment may do so even when their orientation is more receptive and compliance-based.
Several recommendations can be drawn from the findings of this study. First, individuals with lower educational attainment were more likely to engage in positive health behaviors when they could access and evaluate health information. Previous studies have shown that individuals with lower educational attainment are less likely to seek health information actively and tend to rely on sources such as television and religious organizations rather than government agencies [36]. Therefore, targeted interventions tailored to vulnerable groups are warranted to enhance their ability to obtain and critically assess reliable health information.
Second, our findings are broadly consistent with previous research suggesting that adequate information may encourage individuals to adopt new health practices and that those who take a proactive approach to their health are more likely to seek health information [37]. However, we also found a positive association between weight-control attempts and Component 2 among individuals with higher educational attainment. Further research is needed to identify the factors underlying the association between Component 2 and positive health behaviors.
Finally, the findings of this study suggest that the latent structure of health literacy identified in this analysis is related to health behaviors in complex ways. Future research using diverse datasets from different countries may further clarify the multifaceted relationships embedded in the latent structure of health literacy, thereby shedding light on the underlying theoretical pathways and potential causal mechanisms.
This study has several limitations. First, our dataset, the KHPS, assessed health literacy exclusively among adults and therefore excluded individuals younger than 19 years from the sample. Second, simple descriptive statistics suggested the presence of non-random differences between respondents who answered all HLS-EU-Q16 items and those who did not. Such non-random missingness highlights the need to interpret the results with caution, given the potential for selection bias. Third, because alcohol consumption was not analyzed by drinking frequency or intensity, the variable was treated as dichotomous, which may have reduced the measure’s sensitivity and led to some loss of information. Fourth, because the data were collected in 2021 during the COVID-19 pandemic, certain health behaviors may have differed from those observed in a typical year [38], potentially limiting the generalizability of the study’s findings. Fifth, the interpretation of components derived from principal component analysis involves a degree of subjectivity, which may introduce interpretive bias. Finally, given the cross-sectional design of this study, the findings should not be interpreted as establishing causal relationships. Despite these limitations, this study is noteworthy in that it applies PCA to multiple survey items assessing health literacy, extracts key underlying factors, and examines their relationships with health behaviors.
Conclusion
Health literacy encompasses the ability to access, understand, appraise, and apply health-related information for health promotion. Efforts to improve health literacy should prioritize individuals with low socioeconomic status or poor health status. This study underscores the importance of fostering individuals’ capacity to seek out, appraise, and effectively apply health information. In particular, encouraging interest in health-related issues may help strengthen self-efficacy and motivation to seek health information. Furthermore, Pálsdóttir (2010) noted that seeking health information provides individuals with the opportunity to take responsibility for their own health [39].
We hope that future research will extend our analysis by exploring the relationship between health literacy components and health-related outcomes. In addition, we expect that future research will adopt other statistical methods to examine various aspects of health literacy. Such efforts may contribute to a deeper understanding of health behaviors and overall health, further advancing health promotion strategies.
Acknowledgements
The authors would like to thank Wordvice for providing English language editing assistance.
Abbreviations
- KHPS
Korea Health Panel Survey
- HLS-EU-Q16
The 16-item European Health Literacy Survey Questionnaire
- BMI
Body Mass Index
- CFA
Confirmatory Factor Analysis
- PCA
Principal Component Analysis
Authors’ contributions
Hyein Shin: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Data Curation, Writing – Original Draft; Woohyeon Kim: Conceptualization, Methodology, Software, Validation, Writing – Review & Editing, Supervision.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Data availability
The dataset supporting the findings of this study is available from the Korea Health Panel Survey. Access to the data requires completion of a data use agreement on the official website, submission of user information, and consent to the collection and use of personal information (https://www.khp.re.kr:444/).
Declarations
Ethics approval and consent to participate
This study was conducted in accordance with the ethical principles of the Declaration of Helsinki. In the Korean Health Panel survey, informed consent was obtained from participants at the time of data collection. The present study analyzed de-identified secondary data; therefore, additional informed consent for the secondary analysis was not required. This study was exempt from review by the Institutional Review Board of the University of Seoul (UOS 2024-09-002).
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.World Health Organization. Social determinants of health. World Health Organization. 2026. Available from: World Health Organization. https://www.who.int/health-topics/social-determinants-of-health.
- 2.Vrtikapa K, Urmy FH, Hoque F. Social Determinants of Health: The Impact of This Overlooked Vital Sign. J Brown Hosp Med. 2025;4(3):138072. [DOI] [PMC free article] [PubMed]
- 3.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(3):259–67. [Google Scholar]
- 4.Nutbeam D. Health promotion glossary. Health Promot Int. 1998;13(4):349–64. [DOI] [PubMed] [Google Scholar]
- 5.Sørensen K, den Broucke SV, 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;12:80. [DOI] [PMC free article] [PubMed]
- 6.Chun H, Lee J. Factors associated with health literacy among older adults: Results of the HLS-EU-Q16 measure. Korean J Health Educ Promotion. 2020;37(1):1–13. [Google Scholar]
- 7.Kim S, Choi J, Park D. The relationship between sub-dimensions of health literacy and health-related behaviors among Korean adults. Health Social Welf Rev. 2019;39(1):334–64. [Google Scholar]
- 8.Kim J. Measuring the level of health literacy and influence factors: Targeting the visitors of a university hospital’s outpatient clinic. J Korean Clin Nurs Res. 2011;17(1):27–34. [Google Scholar]
- 9.Kim Y, Park B, Lee H. A study on predicting health literacy of Korean elderly using Andersen’s Health Behavior Model. J Welf Aged. 2014;65:35–57. [Google Scholar]
- 10.Kim Y. Association between health literacy and health-promoting behavior (eating habits, physical activity, and stress) of university students. Korean J Community Nutr. 2022;27(2):94–104. [Google Scholar]
- 11.Olisarova V, Kaas J, Staskova V, Bartlova S, Papp K, Nagorska M, et al. Health literacy and behavioral health factors in adults. Public Health. 2021;190:75–81. [DOI] [PubMed] [Google Scholar]
- 12.Housman J, Dorman S. The Alameda County Study: A Systematic, Chronological Review. J Health Educ. 2005;36(5):302–8. [Google Scholar]
- 13.Ra G, Ghang H, Eom T, Lee S. A study on the socioeconomic cost and policy priority of health risk factors. Korean J Health Econ Policy. 2024;30(1):21–50. [Google Scholar]
- 14.World Health Organization. Obesity and overweight. World Health Organization. 2025. Available from: https://www.who.int/news-room/fact-sheets/detail/obesity-and-overweight
- 15.World Health Organization. Alcohol. World Health Organization. 2024. Available from: https://www.who.int/news-room/fact-sheets/detail/alcohol
- 16.World Health Organization. Tobacco. World Health Organization. 2025. Available from: https://www.who.int/news-room/fact-sheets/detail/tobacco
- 17.Pelikan JM, Röthlin F, Ganahl K. Measuring comprehensive health literacy in general populations: validation of instrument, indices and scales of the HLS-EU study. In Proceedings of the 6th Annual Health Literacy Research Conference Bethesda, Rockville, MD: Hyat Regency. 2014.
- 18.Lever J, Krzywinski M, Altman N. Principal component analysis. Nat Methods. 2017;14(7):641–3. [Google Scholar]
- 19.Gustafsdottir SS, Sigurdardottir AK, Arnadottir SA, Heimisson GT, Mårtensson L. Translation and cross-cultural adaptation of the European Health Literacy Survey Questionnaire, HLS-EU-Q16: the Icelandic version. BMC Public Health. 2020;20:61. [DOI] [PMC free article] [PubMed]
- 20.Bergman L, Nilsson U, Dahlberg K, Jaensson M, Wångdahl J. Validity and reliability of the swedish versions of the HLS-EU-Q16 and HLS-EU-Q6 questionnaires. BMC Public Health. 2023;23:724. [DOI] [PMC free article] [PubMed]
- 21.Korea Institute for Health and Social Affairs. National Health Insurance Service. Korea Health Panel Annual Data 2019. 2020.
- 22.Okan O, Bauer U, Levin-Zamir D, Pinheiro P, Sørensen K. International handbook of health literacy: Research, practice and policy across the lifespan. Policy; 2019.
- 23.Toçi E, Burazeri G, Kamberi H, Toçi D, Roshi E, Jerliu N, et al. Health literacy and body mass index: a population-based study in a South-Eastern European country. J Public Health. 2021;43(1):123–30. [DOI] [PubMed] [Google Scholar]
- 24.Kim S, Lee E. The influence of functional literacy on perceived health status in Korean older adults. J Korean Acad Nurs. 2008;38(2):195–203. [DOI] [PubMed] [Google Scholar]
- 25.Song I. Relationship between health literacy and health-related quality of life in Korean adults with chronic diseases. PLoS ONE. 2024;19(4):e0301894. [DOI] [PMC free article] [PubMed]
- 26.Li M, Sonoda N, Koh C, Yasumoto R, Morimoto A. Association between health literacy and current smoking among middle-aged Japanese ever smokers. Public Health Toxicol. 2022;2(2):1–7. [Google Scholar]
- 27.Sørensen K, Broucke Stephan Van den, Pelikan JM, Fullam J, Doyle Gerardine, 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;13:948. [DOI] [PMC free article] [PubMed]
- 28.Svendsen MT, Bak CK, Sørensen Kristine, Pelikan J, Riddersholm SJ, Skals RK 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. 2020;20:565. [DOI] [PMC free article] [PubMed]
- 29.Garcia-Codina O, Juvinyà-Canal Dolors, Amil-Bujan P, Bertran-Noguer C, González-Mestre MA, Masachs-Fatjo E et al. Determinants of health literacy in the general population: results of the Catalan health survey. BMC Public Health. 2019;19:1122. [DOI] [PMC free article] [PubMed]
- 30.Protheroe J, Whittle R, Bartlam B, Estacio EV, Clark L, Kurth J. Health literacy, associated lifestyle and demographic factors in adult population of an English city: a cross-sectional survey. Health Expect. 2017;20(1):112–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Brandt L, Schultes M-T, Yanagida T, Maier G, Kollmayer M, Spiel C. Differential associations of health literacy with Austrian adolescents’ tobacco and alcohol use. Public Health. 2019;174:74–82. [DOI] [PubMed] [Google Scholar]
- 32.Ruokolainen O, Härkänen T, Lahti J, Haukkala A, Heliövaara M, Rahkonen O. Association between educational level and smoking cessation in an 11-year follow-up study of a national health survey. Scand J Public Health. 2021;49(8):951–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Lee E, Seo H. The Factors Associated with Successful Smoking Cessation in Korea. J Korean Acad Fam Med. 2007;28(1):39–44. [Google Scholar]
- 34.Upadhyay S, Lord J, Gakh M. Health-Information Seeking and Intention to Quit Smoking: Do Health Beliefs Have a Mediating Role?. Tobacco use insights. 2019;12:1–10. [DOI] [PMC free article] [PubMed]
- 35.Eyüboğlu E, Schulz PJ. Do health literacy and patient empowerment affect self-care behaviour? A survey study among Turkish patients with diabetes. BMJ Open. 2016;6:e010186. [DOI] [PMC free article] [PubMed]
- 36.Richardson A, Allen JA, Xiao H, Vallone D. Effects of Race/Ethnicity and Socioeconomic Status on Health Information-Seeking, Confidence, and Trust. J Health Care Poor Underserved. 2012;23:1477–93. [DOI] [PubMed] [Google Scholar]
- 37.Rakowski W, Assaf AR, Lefebvre RC, Lasater TM, Niknian M, Carleton RA. Information-Seeking about Health in a Community Sample of Adults: Correlates and Associations with Other Health-Related Practices. Health Educ Q. 1990;17(4):379–93. [DOI] [PubMed] [Google Scholar]
- 38.Park E, Park J, Park N, Kwak Y, Chun H, Oh Y. Analysis of changes in health behavior, health status and medical use during COVID-19 pandemic. Korea Institute for health and social affairs; 2023.
- 39.Pálsdóttir Á. The connection between purposive information seeking and information encountering: a study of Icelanders’ health and lifestyle information seeking. J Doc. 2010;66(2):224–44. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The dataset supporting the findings of this study is available from the Korea Health Panel Survey. Access to the data requires completion of a data use agreement on the official website, submission of user information, and consent to the collection and use of personal information (https://www.khp.re.kr:444/).