Skip to main content An official website of the United States government Here's how you know Here's how you know Official websites use .gov A .gov website belongs to an official government organization in the United States. Secure .gov websites use HTTPS A lock ( Lock Locked padlock icon ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites. Search Log in Dashboard Publications Account settings Log out Search… Search NCBI Primary site navigation Search Logged in as: Dashboard Publications Account settings Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide PERMALINK Copy As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice BMC Psychol . 2026 Mar 5;14:513. doi: 10.1186/s40359-026-04258-2 Search in PMC Search in PubMed View in NLM Catalog Add to search Self-esteem patterns and the associations with health-related quality of life, self-rated health and lifestyle behaviour among junior high school students: a latent class analysis Ruowei Wang Ruowei Wang 1 School of Nursing, Shandong Xiehe University, 6277 Jiqing Road, Licheng District, Jinan, Shandong 250109 China Find articles by Ruowei Wang 1 , Shuna Zhai Shuna Zhai 1 School of Nursing, Shandong Xiehe University, 6277 Jiqing Road, Licheng District, Jinan, Shandong 250109 China Find articles by Shuna Zhai 1 , Chunhong Shen Chunhong Shen 1 School of Nursing, Shandong Xiehe University, 6277 Jiqing Road, Licheng District, Jinan, Shandong 250109 China Find articles by Chunhong Shen 1 , Jiaoyang Liu Jiaoyang Liu 1 School of Nursing, Shandong Xiehe University, 6277 Jiqing Road, Licheng District, Jinan, Shandong 250109 China Find articles by Jiaoyang Liu 1 , Pei Zhang Pei Zhang 1 School of Nursing, Shandong Xiehe University, 6277 Jiqing Road, Licheng District, Jinan, Shandong 250109 China Find articles by Pei Zhang 1 , Tong Zhang Tong Zhang 2 Affiliated Hospital of Jining Medical University, 89 Guhuai Road, Rencheng District, Jining, Shandong 272000 China Find articles by Tong Zhang 2 , Qingai Yang Qingai Yang 1 School of Nursing, Shandong Xiehe University, 6277 Jiqing Road, Licheng District, Jinan, Shandong 250109 China Find articles by Qingai Yang 1 , Xiangzheng Ding Xiangzheng Ding 1 School of Nursing, Shandong Xiehe University, 6277 Jiqing Road, Licheng District, Jinan, Shandong 250109 China Find articles by Xiangzheng Ding 1, ✉ , Aihua Wang Aihua Wang 3 School of Nursing, Shandong Xiandai University, 20288 Jingshi East Road, Licheng District, Jinan, Shandong 250104 China Find articles by Aihua Wang 3, ✉ , Xiuyun Wu Xiuyun Wu 1 School of Nursing, Shandong Xiehe University, 6277 Jiqing Road, Licheng District, Jinan, Shandong 250109 China Find articles by Xiuyun Wu 1, ✉ Author information Article notes Copyright and License information 1 School of Nursing, Shandong Xiehe University, 6277 Jiqing Road, Licheng District, Jinan, Shandong 250109 China 2 Affiliated Hospital of Jining Medical University, 89 Guhuai Road, Rencheng District, Jining, Shandong 272000 China 3 School of Nursing, Shandong Xiandai University, 20288 Jingshi East Road, Licheng District, Jinan, Shandong 250104 China ✉ Corresponding author. Received 2025 Oct 24; Accepted 2026 Feb 24; Collection date 2026. © The Author(s) 2026 Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/ . PMC Copyright notice PMCID: PMC13072603 PMID: 41787513 Abstract Background Previous studies have investigated associations of self-esteem with mental health among adolescents. Very few studies have examined the relationship of self-esteem patterns with health-related quality of life (HRQoL), mental health and lifestyle behaviour. This study aims to explore patterns of self-esteem, and analyze their associations with HRQoL, self-rated health, mental health and lifestyle behaviour among junior high school students. Methods We used the survey data for health behaviour and health among junior high school students in Jining city of Shandong province of China. Latent class analysis was used to identify students with heterogeneous response patterns of self-esteem. Multinomial and traditional logistic regressions and linear regression were applied to examine the relationships of self-esteem patterns with lifestyle behaviours, self-rated health and HRQoL respectively. Results Three subgroups of students with distinct patterns of self-esteem were identified. The first group (class 1) was named as “high self-esteem” group (55%). The second group (class 2) was labelled as “moderate self-esteem” group (33%). The third group was named as “low self-esteem” group (12%). Moderate and low self-esteem patterns (relative to high self-esteem) were associated with lower HRQoL and higher odds of poorer self-rated health and mental health. Skipping a breakfast, shorter sleep time and lower level of physical activity was respectively related to low self-esteem. Conclusions The findings in this study suggest that low self-esteem patterns are associated with poorer HRQoL, low mental health and self-rated overall health. Lower self-esteem patterns are related to unhealthy lifestyle behaviours characterized as breakfast skipping, insufficient sleep time and low level of physical activity among junior high school students. Further research is needed to extend prospective studies and intervention research to better elucidate the relationships between lifestyle behaviours and self-esteem, and between self-esteem and HRQoL, mental and physical health among adolescents. Health intervention programs that promote self-esteem levels and healthy lifestyles together are recommended to enhance HRQoL and health among junior high school students. Supplementary Information The online version contains supplementary material available at 10.1186/s40359-026-04258-2. Keywords: Self-esteem, Health-related quality of life, Self-rated health, Mental health, Lifestyle, Adolescents, Latent class analysis Background Adolescence is a key period to develop a healthy lifestyle and maintain better physical and mental health. Adolescents in this period experience significant changes and challenges in physical, psychological, and social health developments, such as puberty, increased psychological and emotional problems, low self-esteem, concerns of school academic performance or peer relations. Previous studies have shown that adolescents are vulnerable to low self-esteem and various mental health issues such as loneliness, stress, anxiety and depression [ 1 , 2 ]. Recent research has demonstrated that poor mental health among children and adolescents continues to be a global public health concern in the world after the COVID-19 pandemic [ 3 , 4 ]. Globally, it is estimated that approximately 14.3% of 10–19-year-old adolescents experience a mental disorder [ 5 ]. Research from large population-based surveys in China has reported that mental health disorders among Chinese children and adolescents contribute to a significant burden of diseases not only in China but also at the global level [ 6 – 8 ]. Self-esteem is an individual’s evaluation of self-value or personal abilities in specific domains (e.g. self-worth, self-image, self-respect), or an overall evaluation of self-esteem as a whole (referred as global self-esteem) [ 9 – 11 ]. Self-esteem is considered a sensitive indicator for mental health [ 12 ]. Studies have demonstrated that low self-esteem among children and adolescents is associated with poor mental health (e.g., depression, anxiety) and some severe mental illnesses, such as suicidal ideation and substance use [ 13 – 16 ]. Self-esteem is also related to a number of physical health outcomes, such as obesity [ 17 ], and other chronic diseases [ 18 ]. Research has also shown that adolescents with low self-esteem are more likely to exhibit poor academic performance in schools [ 19 ]. Health-related quality of life (HRQoL) is a multidimensional construct that encompasses individuals’ perception of the physical, psychological, functional and social aspects of health [ 20 ]. Assessment of HRQoL is essential to identify subpopulations among children and adolescents with poor health status and its associated risk factors. Such investigations could inform health intervention practice and health policy to prioritize health promotion programs in children and adolescents [ 21 ]. Besides, self-rated health (SRH) is a person’s self-perception of their overall health status. Prior research has documented that SRH is a stable and powerful indicator to predict morbidity and mortality in adults [ 22 ], and it has also been broadly used to assess children and adolescents’ health [ 23 ]. Some studies have reported associations between self-esteem and HRQoL and self-rated health among children and adolescents [ 24 – 26 ], and most of these studies investigated children and adolescents with chronic disease conditions (e.g., obesity, cancer) [ 25 , 27 ]. Very few studies have investigated the importance of self-esteem for HRQoL and self-rated health among general populations of children and adolescents [ 24 ]. More knowledge on how self-esteem correlates with HRQoL in adolescents is needed. A number of studies have investigated associations of lifestyle behaviours with obesity, mental health, self-rated health and HRQoL among children and adolescents, such as physical activity (PA), dietary behaviour and sleep [ 21 , 23 , 28 – 31 ]. Relatively fewer studies have examined the relationship between lifestyles and self-esteem in children and adolescents [ 17 , 32 , 33 ]. The studies demonstrated that higher level of PA was associated with higher self-esteem, skipping a meal or breakfast, and a lack of sleep respectively was related to low self-esteem in children and adolescents [ 17 , 33 , 34 ]. However, previous studies on the correlations of lifestyle behaviour with self-esteem in children and adolescents mostly examined a single lifestyle behaviour, such as PA, sleep duration [ 35 , 36 ]. Very few studies have investigated the importance of these behaviours (PA, skipping breakfast and sleep) simultaneously for self-esteem [ 37 ]. Particularly, the associations between breakfast habits, sleep duration and self-esteem in adolescents are still understudied. Furthermore, previous studies that evaluated self-esteem and associated health-related factors mainly examined global self-esteem with a continuous global composite score to be used [ 17 , 38 ]. Few studies have analyzed self-esteem in specific domains or components of self-esteem (e.g. self-worth, self-criticism) [ 32 , 39 ]. Scant studies have applied latent class analysis (LCA) to examine self-esteem clustering patterns. Self-esteem scales generally contain multiple items that constitute a global attribute score and several underlying domains [ 25 , 40 ]. Individuals’ responses to multi-items’ self-esteem measures among population are likely to be heterogeneous in a manner of clustering or latent classes [ 41 ]. LCA is a person-centered statistical method to classify a group of people into several mutually exclusive sub-groups (latent classes) with distinct response patterns on multiple health-related indicators [ 42 ]. LCA has been increasingly applied to investigate heterogeneous patterns or co-occurrence of health-related factors, such as health-related behaviours, mental health symptoms, eating habits among various populations [ 43 , 44 ]. To our best knowledge, no studies have reported how heterogeneous patterns of self-esteem are related to HRQoL and lifestyle behaviours in adolescents. Our literature review identified very few studies that used LCA to examine self-esteem items in combination with other mental health indicators like anxiety and depressive symptoms in children and adolescents [ 45 , 46 ], and one study used cluster analysis to analyze self-esteem patterns in a non-school based sample of adolescents [ 47 ]. To better understand the relationships of self-esteem with HRQoL, self-perceived health and lifestyle behaviours among adolescents, it is essential to explore the association with respect to patterns of self-esteem, specifically, to examine if individuals with patterns of higher self-esteem present better HRQoL and a healthier health-related behaviours than the those with a lower self-esteem pattern. Given a high prevalence of poor mental health and unhealthy lifestyle behaviours among adolescents [ 7 , 48 ], the information on the associations of self-esteem patterns on health and HRQoL and lifestyle behaviours will help guide health interventions to promote healthy lifestyles and self-esteem and improve mental health and HRQoL in adolescents. The present study aims to investigate heterogeneous patterns of self-esteem, and examine their associations with HRQoL, self-rated health, mental health and lifestyle behaviour (physical activity, breakfast eating and sleep) among junior high school students. Methods Participants Data used in this study were from the survey of junior high school students in four junior high schools in Jining city, Shandong province of China in June and July, 2020. The survey aimed to evaluate health-related behaviours and health status among students to promote their health and quality of life. The survey applied a two-stage randomized cluster sampling design. Four junior high schools were randomly selected from two districts (two schools in each of the major districts) of the city, and classes were then randomly selected within the schools. All students in the chosen classes were invited. In total, 800 questionnaires were distributed to students in the selected schools. Of these, 785 (98.1%) students with parents’ consent for participation completed the survey questionnaires. Of 785 questionnaires returned, two duplicate records were deemed invalid and excluded. Of 783 respondents, 16 presented missing data on all ten self-esteem items, and two, one, and three students had missing information on 4, 6, and 7 items, respectively. Respondents with missing values on more than five items ( n = 20) in the self-esteem scale were excluded, leaving a sample of 763 student observations (97.4% of 783 respondents) available in the present analysis. Fig. 1 presents the data collection and the sample. The survey and participants were previously reported [ 49 ]. Fig. 1. Open in a new tab Flow diagram of the data collection and the respondents Measures Self‑esteem scale The Rosenberg Self‑Esteem Scale ( RSES) was used to assess self-esteem [ 9 ]. The RSES comprises ten items: five positively worded items and five negatively worded items [ 9 ]. It is one of the most broadly used self-esteem instruments in the world [ 50 ] and has been validated in many countries [ 11 , 40 ]. The items of the scale have four response levels: strongly agree, agree, disagree and strongly disagree. The total score can be calculated by summing each item’s score of the 10 items, representing a global or overall self-esteem. The Chinese version of the RSES has been shown a high reliability and validity in adolescents and young adults in China [ 51 – 53 ]. The Cronbach’s coefficient α for the RSES is 0.81 in this study. Health-related quality of life The survey included questions on the EuroQol five-dimensional questionnaire, youth version, EQ-5D-Y [ 54 ]. The EQ-5D-Y was designed for use in children and youth aged between 8 and 18 years ( https://euroqol.org ). The EQ-5D-Y five-dimensional descriptive system has five questions: walking; looking after myself; doing usual activities; having pain or discomfort; and feeling worried, sad or unhappy [ 54 ]. Each question has three levels of responses: no problems, some problems or a lot of problems. A combination of the responses to the five questions represents an overall health state, and a total of 243 possible health states can be defined. Recently, a single index value set based on the EQ-5D-Y five-dimensional descriptive system has been developed in some countries to estimate the EQ-5D-Y utility values [ 55 , 56 ]. To calculate the EQ-5D-Y index, we used the recently developed Chinese value set for the EQ-5D-Y, which is based on time trade-off valuations among Chinese adults on behalf of hypothetical children (the hybrid model with A3 term, by Yang Z, et al., 2022) [ 57 ]. The index values range from −0.089 (lowest health) to 1.000 (highest health), with higher values indicating better HRQoL [ 57 ]. The EQ-5D-Y also includes a Visual Analogue Scale (VAS) with scores ranging between 100 (best imaginable health) and 0 (worst imaginable health) to indicate a self-rated overall health status. In this study, the EQ-5D-Y index and the EQ-5D-Y mental health dimension (feeling worried, sad or unhappy) were used. The Chinese version of the EQ-5D-Y has been validated and shown good feasibility, reliability and validity among children and adolescents in China [ 58 ]. In this study, the reliability coefficient (Cronbach’s alpha) is 0.70. Self-rated health Self-rated or self-perceived health was assessed by a single item question asking students to rate their general health with five-level response options: very good, good, fair, poor, very poor. The variable was dichotomized as two categories: very good/good; fair/poor/very poor. The self-rated health item has been widely used and presented good reliability and validity among adolescents [ 23 ]. Lifestyle behaviours Students were asked to select frequencies of moderate and vigorous physical activity (MVPA) outside of schools with the response levels: never or once a month, Once a week, 2–3 times a week and ≥ 4 times a week. The last two levels were combined into one group labeled as ≥ 2 times a week. The physical activity question was taken from the Chinese version of the Physical Activity Questionnaire for Adolescents (PAQ-A), which has been previously validated [ 59 ]. Breakfast habit was measured as frequencies of eating breakfast with the options: eating every day, eating often, and never or almost never eating breakfast. Sleep time was assessed as hours a day spent on sleep, and was collapsed to two levels: ≥ 7 h/day and < 7 h/day. The items of breakfast and sleep have been widely used and demonstrated good validity and reliability [ 33 , 60 ]. Control variables Demographic characteristics of students, including gender, grade level, highest level of parental education, were considered confounding variables. Grade was categorised as junior (first and second year in junior high school) and senior (third and fourth year in junior high school) levels. Highest parental education level was classified as junior high school or less; high school; and university/college or higher. Statistical analyses Latent class analysis was used to identify self-esteem pattern groups (latent classes) of students. All ten items in the aforementioned Rosenburg self-esteem scale were used in the LCA. As the number of students who responded to the “strongly disagree” level on some of the positive-wording items was very small (e.g., n = 15 for the item, I am able to do things as well as most other people), the four-category variables of the RSES scale were dichotomized by combing strongly agree and agree to one group, and strongly disagree and disagree to another group. Positively worded items were reverse coded such that a higher score indicates lower self-esteem. To select the latent class model with a best fit, a set of latent class models were estimated from one-class model to k -class model ( k is the number of latent class) until the model with best fit was determined. The model fit was assessed using fit statistics of the Bayesian Information Criterion (BIC), the sample size-adjusted BIC (aBIC), and the Lo–Mendell–Rubin adjusted likelihood ratio test (LMRALRT) [ 61 ]. A lower BIC and aBIC for a k -class model indicates a better fit of this model compared to the k-1 class model [ 61 ]. A p -value > 0.05 by the LMRALRT for a k -class model indicates that a better fit should be given to the k-1 class model relative to the k -class model. The latent class model was estimated with maximum likelihood estimation with robust standard errors (MLR). The final model selection was based on a better overall fit, parsimony and significant meaningful interpretation of the latent classes. Latent class analyses with four-response levels of the RSES items were also examined. As the predicted probability was zero on the categories with a sparse frequency for some items, the latent class models using the dichotomous variables were used. In this study, we applied a manual stepwise approach in the LCA to examine the associations of self-esteem patterns with health outcomes and lifestyle behaviour. In the first step, the latent class analysis was applied to the 10 self-esteem items to select the most appropriate latent class model. Subsequently, the predicted class indicator as well as the predicted probability of being in each class for an individual were saved and then linked with the original data (with all other variables needed). The third step was to analyze the associations of self-esteem patterns with HRQoL, mental health, self-rated health and lifestyle behaviours using a set of regression analyses. The analytical steps are visually presented in the supporting information (S1 Figure, Additional file 1). Chi-square test was used to test between class differences in the frequency distribution of lifestyle behaviours, mental health and self-rated health. One-way analysis of variance (ANOVA) test was used for difference in the EQ-5D-Y index (utility) score across classes. We applied multinomial logistic regression to examine the association of lifestyle behaviours (independent variables) with the self-esteem patterns (dependent variable). To examine the relationship between self-esteem patterns (independent variable) and HRQoL, mental health and self-rated health, we used linear regression for HRQoL (the EQ-5D-Y index) and logistic regression for mental health and self-rated health respectively. Univariate and multivariable regressions were performed respectively to examine the associations. The multivariable multinomial logistic regression included physical activity, breakfast and sleep time together as these variables did not show a significant collinearity (the variance inflation factor, VIF < 1.2 for each behaviour). All the multivariable regression models adjusted the effect of gender, grade level and parental education. Effect size was computed by Cohen’s d in the comparison of the HRQoL between latent classes [ 62 ]. A Cohen’s d value of 0.2, 0.5, and 0.8 indicates small, medium, and large effect, respectively [ 62 ]. Effect sizes for categorical variables based on Chi-square statistics were calculated with Cramer’s V [ 63 ]. A Cramer’s V value 0.1, 0.3 and 0.5 represents a small, medium, and large effect for a 2 × 3 table. A Cramer’s V value 0.07, 0.21 and 0.35 indicates a small, medium, and large effect for a 3 × 3 table. Latent class analyses were conducted using Mplus version 8 [ 64 ]. The regression analyses and other analyses (e.g., ANOVA) were performed using Stata/SE 15 (StataCorp LLC, College Station, TX, USA). Results Self-esteem patterns Of 763 student respondents, 54.1% were boys, 60.6% were in junior grade, and 50.7%, 29.6% of students had their parental education level in high school and university/college or higher, respectively. The fit statistics of the latent class models for estimation of the self-esteem latent classes that represent patterns of self-esteem is presented in S3 Table 1 (Additional file 3, supporting information). The BIC and a BIC decreased from one- to four-class models, while the magnitude of the decrease between four–class model and three-class model was smaller than the decrease between three-class model and two-class model. The LMRALRT p -value for the four-class model was not statistically significant (p = 0.1307). In contrast, the LMRALRT for the three-class model was statistically significant with a p -value of 0.0076. The fit statistics favored a better fit of the three-class model. The four-class model had a class with a very small class proportion (6%). The three-class model also provided better interpretability in terms of class separation; hence the 3-class model was selected as the best parsimonious model to extract latent classes for use in the subsequent analysis. The predicted class with the highest posterior probability in the model was assigned for each student observation as the final class membership representing a self-esteem pattern. Table 1 presents the within class frequency distribution of self-esteem items. Fig. 2 shows the proportion of students who reported low level of self-esteem on each item of the self-esteem scale across three classes. The first five items in Fig. 2 are positively worded and the remaining five items are negatively worded. Students in class 1 (self-esteem pattern 1, 54.52%) is characterized by a high probability of reporting high (versus low) self-esteem on nine out of ten items, thus is labeled as “high self-esteem” group (see Fig. 2 ). Students in class 2 (self-esteem pattern 2) represents 32.77% of respondents, with a mixed pattern of higher self-esteem on the positive-wording items, and lowest self-esteem on all negative-wording items except for the item, “All in all, I am inclined to feel that I am a failure” (47.2% reported lower self-esteem in class 2 versus 50.0% in class 3). This group is labeled as “moderate self-esteem” compared with other two groups. Students in class 3 (self-esteem pattern 3) accounts for 12.71% of respondents. Most of students in class 3 reported having low self-esteem on the positive-wording items and moderated (or intermediate) level of self-esteem on the negative-wording items. This class is labeled as “low self-esteem” group. The predicted probability of reporting low self-esteem on the items of the self-esteem scale based on the final three-class model is illustrated in the supporting information (S2 Figure, Additional file 2). The predicted probability of the responses of self-esteem items by the classes is presented in S3 Table 2 (Additional file 3). The frequency distribution of the RSES items with original our-response levels is presented in S3 Table 3 (Additional file 3). Table 1. Frequency distribution of the self-esteem variables by the self-esteem class membership Variables Class 1, n (%) Class 2, n (%) Class 3, n (%) Total Class size (%) (total) 416 (54.52) 250 (32.77) 97 (12.71) 763 (100.00) On the whole, I am satisfied with myself Agree 396 (95.19) 202 (80.80) 34 (35.05) 632 (82.83) Not agree 20 (4.81) 48 (19.20) 63 (64.95) 131 (17.17) Sometimes, I think I am no good at all Not agree 162 (38.94) 8 (3.20) 23 (23.71) 193 (25.29) Agree 254 (61.06) 242 (96.80) 74 (76.29) 570 (74.71) I feel that I have a number of good qualities Agree 394 (94.71) 224 (89.60) 26 (26.80) 644 (84.40) Not agree 22 (5.29) 26 (10.40) 71 (73.20) 119 (15.60) I feel that I do not have much to be proud of Not agree 294 (70.67) 12 (4.80) 20 (20.62) 326 (42.73) Agree 122 (29.33) 238 (95.20) 77 (79.38) 437 (57.27) I am able to do things as well as most other people Agree 394 (94.71) 239 (95.60) 18 (18.56) 651 (85.32) Not agree 22 (5.29) 11 (4.40) 79 (81.44) 112 (14.68) I certainly feel useless at times Not agree 408 (98.08) 98 (39.20) 54 (55.67) 560 (73.39) Agree 8 (1.92) 152 (60.80) 43 (44.33) 203 (26.61) I feel that I am a person of worth, at least on an equal plane with others Agree 386 (93.01) 205 (82.00) 36 (37.50) 627 (82.39) Not agree 29 (6.99) 45 (18.00) 60 (62.50) 134 (17.61) I wish I could have more respect for myself Not agree 346 (83.37) 45 (18.00) 45 (46.88) 436 (57.29) Agree 69 (16.63) 205 (82.00) 51 (53.13) 325 (42.71) All in all, I am inclined to feel that I am a failure Not agree 406 (97.83) 132 (52.80) 48 (50.00) 586 (77.00) Agree 9 (2.17) 118 (47.20) 48 (50.00) 175 (23.00) I take a positive attitude toward myself Agree 380 (91.57) 216 (86.40) 40 (41.67) 636 (83.57) Not agree 35 (8.43) 34 (13.60) 56 (58.33) 125 (16.43) Open in a new tab Class 1 (High self-esteem): highest self-esteem on nine of the variables. Class 2 (Moderate self-esteem): higher self-esteem on positive-wording items and lower self-esteem on negative-wording items. Class 3 (Low self-esteem): lower self-esteem on positive-wording items and moderated self-esteem on negative-wording items Fig. 2. Open in a new tab Within class proportion of students with response to low level of self-esteem on the items in the self-esteem scale. Response level of self-esteem: “Not agree” for the five positive-worded items; “Agree” for the five negative-worded items Associations of self-esteem patterns with HRQoL, self-rated health, mental health and lifestyle behaviours Table 2 displays the frequency distribution of lifestyle behaviours, self-rated health, and HRQoL by the self-esteem patterns (classes). Overall, there was a significant difference in the distribution of lifestyle behaviours, self-rated health, mental health problems and in the EQ-5D-Y utility score across three classes. The proportion of students who reported skipping breakfast, having shorter sleep time (< 7 h/day), lower level of physical activity, and lower self-rated health was larger in class 2 and class 3 than class one. The EQ-5D-Y utility score was significantly lower among students in class 2 and class 3 than the peers in class 1. The mean EQ-5D-Y utility score was 0.936 (SD 0.103), ranging between −0.089 and 1. Cohen’d was 0.41 (95% CI: 0.25, 0.57), and 1.05 (95% CI: 0.82, 1.28) for comparing the EQ-5D-Y utility score between class 2 or 3 and class 1 respectively. Students characterized as lower self-esteem (class 2 and 3) reported more problems of feeling worried, sad or unhappy. The prevalence of feeling worried, sad or unhappy among all students was 36.09%. The Cramér's V was 0.33 for feeling worried, sad or unhappy and 0.242 for self-rated health. Table 2. Frequency distribution of lifestyle behaviours, self-rated health, and the mean score of the EQ-5D-Y utility by the self-esteem patterns (class membership) Variables Class 1, % Class 2, % Class 3, % P -value Cramér's V Lifestyle behaviour MVPA ≥ 2 times a/week 50.96 47.60 37.11 0.009 0.094 Once a week 22.12 23.60 17.53 Never or once a month 26.92 28.80 45.36 Eating breakfast Eating everyday 63.70 50.40 32.99 < 0.001 0.179 Eating often 25.96 32.40 37.11 Never or almost never eating 10.34 17.20 29.90 Sleep time ≥ 7 h/day 58.55 51.20 30.93 < 0.001 0.163 < 7 h/day 41.45 48.80 69.07 Self-rated health Very good/good 86.51 74.09 57.29 < 0.001 0.242 Fair/poor/very poor 13.49 25.91 42.71 Feeling worried, sad or unhappy No problems 76.39 56.40 29.90 < 0.001 0.330 Some and a lot of problems 23.61 43.60 70.10 EQ-5D-Y index score: mean (SD) 0.960 (0.069) 0.927 (0.097) 0.857 (0.176) < 0.001 Cohen’ d (95% CI) comparing between classes' difference in the EQ-5D-Y index score Class 2 with Class 1: 0.41 (0.25, 0.57) Class 3 with Class 1: 1.05 (0.82, 1.28) Class 3 with Class 2: 0.56 (0.32, 0.80) Open in a new tab χ 2 test used for categorical variables in the table. ANOVA test used for the EQ-5D-Y index score. The percentages in the table are within class percentages across the levels of the explanatory variables SD standard deviation, CI confidence interval The linear regression result for the association of self-esteem patterns with health-related quality of life measured by the EQ-5D-Y index value is presented in Table 3 . In the multivariable regression, students in the moderate and low self-esteem patterns had 0.034 (95% CI: −0.049, −0.018) points, 0.101 (95% CI: −0.123, −0.079) points lower index score than those students in the high self-esteem pattern, indicating a dose–response association with respect to the levels of self-esteem. The univariate regression result is similar to the multivariable regression result. The HRQoL index was lower among girls than boys ( p = 0.018). Table 3. Linear regression results for the association between self-esteem patterns and health-related quality of life (the EQ-5D-Y index score) Variables Univariate model Multivariable model Coefficient (95% CI) p -value Coefficient (95% CI) p -value Self-esteem patterns Class 1 (reference group) Class 2 −0.033 (−0.049, −0.018) < 0.01 −0.034 (−0.049, −0.018) < 0.01 Class 3 −0.103 (−0.124, −0.081) < 0.01 −0.101 (−0.123, −0.079) < 0.01 Gender Boys (reference group) Girls −0.018 (−0.032, −0.003) 0.018 −0.011 (−0.025, 0.004) 0.143 Grade level Junior (reference group) Senior −0.004 (−0.019, 0.011) 0.569 −0.008 (−0.023, 0.007) 0.280 Parental education level Junior high school or lower (reference group) High school 0.006 (−0.013, 0.026) 0.530 0.006 (−0.013, 0.024) 0.547 University/College or higher 0.009 (−0.012, 0.031) 0.400 0.008 (−0.013, 0.028) 0.467 Open in a new tab The multivariable model adjusted for the effects of gender, grade level and parental education Table 4 shows the logistic regression result for self-esteem patterns in relation to self-rated health of students. Students in the moderate and low self-esteem patterns were significantly more likely to report poor health than the peers in the high self-esteem level. There was also a gradient effect between self-esteem levels and the odds of lower self-rated health: OR (95% CI): 2.27 (1.52, 3.39) for moderate self-esteem; 4.67 (2.84, 7.69) for low self-esteem relative to high self-esteem pattern. The logistic regression results for mental health problems showed that students with moderate and low self-esteem were significantly more likely to experience depression and anxiety problems than their peers with high self-esteem (Fig. 3 ). Table 4. Logistic regression results for the association between self-esteem patterns and self-rated health Variables Univariate model Multivariable model OR (95% CI) p -value OR (95% CI) p -value Self-esteem patterns Class 1 1.0 1.0 Class 2 2.24 (1.50, 3.35) < 0.01 2.27 (1.52, 3.39) < 0.01 Class 3 4.78 (2.92, 7.82) < 0.01 4.67 (2.84, 7.69) < 0.01 Gender Boys 1.0 1.0 Girls 1.27 (0.90, 1.81) 0.173 1.15 (0.80, 1.65) 0.454 Grade level Junior 1.0 1.0 Senior 1.01 (0.71, 1.44) 0.959 1.09 (0.75, 1.59) 0.653 Parental education level Junior high school or lower 1.0 1.0 High school 0.90 (0.57, 1.42) 0.665 0.89 (0.55, 1.42) 0.621 University/College or higher 0.84 (0.51, 1.39) 0.506 0.86 (0.50, 1.46) 0.568 Open in a new tab The multivariable model adjusted for the effects of gender, grade level and parental education Fig. 3. Open in a new tab Logistic regression results for the association of self-esteem patterns with mental health problems. Mental health measured by the EQ-5D-Y dimension of feeling worried, sad or unhappy. The multivariable model adjusted the effect of demographic variables: gender, grade and parental highest education level Table 5 presents the multinomial logistic regression results for the associations between lifestyle behaviours and self-esteem patterns. After adjusting for the confounding effect of gender, grade and parental education levels, students who skipped a breakfast often or always had a higher likelihood of having lower self-esteem (being in class 2 and 3) than students who ate breakfast everyday. Compared with students who had longer sleep time (≥ 7 h/day), students who experienced shorter sleep (< 7 h/day) were 2.5 timely more likely to report low self-esteem than their peers with better self-esteem (OR = 2.50, 95% CI: 1.52, 4.12, p < 0.01). Low level of physical activity was related to lower self-esteem (OR = 2.31, 95% CI: 1.41, 3.80, class 3 vs. class 1). Girls were more likely to experience low self-esteem than boys. The results in the unadjusted (univariate) model are largely consistent with the multivariable regression results. Table 5. Multinomial logistic regression results for the associations of lifestyle behaviours, demographic factors with self-esteem patterns Variables Univariate model Multivariable model Class 2 vs. Class 1 Class 3 vs. Class 1 Class 2 vs. Class 1 Class 3 vs. Class 1 OR (95% CI) OR (95% CI) OR (95% CI) OR (95% CI) Eating breakfast Eating everyday 1.0 1.0 1.0 1.0 Eating often 1.58 (1.10, 2.26) * 2.76 (1.63, 4.67) ** 1.55 (1.08, 2.24) * 2.27 (1.32, 3.91) ** Never or almost never eating 2.10 (1.31, 3.37) ** 5.59 (3.07, 10.15) ** 2.05 (1.26, 3.32) ** 3.93 (2.11, 7.33) ** Sleep time ≥ 7 h/day 1.0 1.0 1.0 1.0 < 7 h/day 1.35 (0.98, 1.85) 3.16 (1.97, 5.06) ** 1.28 (0.92, 1.78) 2.50 (1.52, 4.12) ** MVPA ≥ 2 times a/week 1.0 1.0 1.0 1.0 Once a week 1.14 (0.77, 1.70) 1.09 (0.58, 2.04) 1.18 (0.78, 1.77) 0.92 (0.48, 1.78) Never or once a month 1.15 (0.79, 1.66) 2.31 (1.41, 3.80) ** 1.10 (0.75, 1.62) 1.66 (0.98, 2.83) Gender Boys 1.0 1.0 1.0 1.0 Girls 0.93 (0.68, 1.27) 2.04 (1.30, 3.22) ** 0.84 (0.60, 1.18) 1.65 (1.01, 2.68) * Grade level Junior 1.0 1.0 1.0 1.0 Senior 0.86 (0.62, 1.19) 0.72 (0.46, 1.15) 0.85 (0.60, 1.19) 0.64 (0.39, 1.07) Parental education level Junior high school or lower 1.0 1.0 1.0 1.0 High school 1.29 (0.84, 1.98) 0.85 (0.49, 1.48) 1.34 (0.87, 2.07) 0.87 (0.49, 1.56) University/College or higher 1.20 (0.75, 1.91) 0.63 (0.34, 1.20) 1.37 (0.84, 2.21) 0.80 (0.41, 1.57) Open in a new tab MVPA: moderate and vigorous physical activity. vs: versus. The multivariable model adjusted for the effects of gender, grade level and parental education, and the lifestyle behaviour variables in the table. The dependent variable is self-esteem patterns (class membership) * p < 0.05, ** p < 0.01 Discussion The present study demonstrates that there are heterogeneous response patterns of self-esteem among junior high school students in a city of Shandong province. The lower levels of self-esteem correlate with lower HRQoL and self-perceived health, with largest negative effects on the health outcome observed for students with the lowest self-esteem. Skipping a breakfast, inadequate sleep and low level of physical activity are associated with low self-esteem. To the best of our knowledge, the present study is the first to investigate the correlations between HRQoL and self-rated health together and self-esteem in junior high school students. One of the main findings in this study is that lower self-esteem is strongly associated with lower HRQoL among students. The finding is consistent with previous studies reporting that self-esteem positively correlates to better HRQoL in children and adolescents [ 24 , 25 ]. This suggests that higher self-esteem may be an important protective factor for HRQoL in adolescents. This study contributes to the research on the effects of self-esteem on HRQoL by an investigation of co-occurred self-esteem patterns, and the use of HRQoL utility scores. The exploration of heterogeneous patterns of self-esteem among adolescents will help identify those individuals vulnerable to lower self-esteem, and with a high risk of being in poor HRQoL. This is essential for public health intervention strategies to implement effective and targeted health promotion programs among adolescents. Furthermore, the benefit of using the EQ-5D-Y utility score is that it can be used in economic evaluations, for example, it can be used to calculate quality-adjusted life years for cost-utility analyses [ 65 ]. This study adds a novel finding to health research for a relationship between adolescence self-esteem and HRQoL in school-based context given most of the previous studies examined clinical samples of adolescents [ 27 ]. This study also reveals a difference in the response patterns of self-esteem between the positive and negative worded items among subgroups of students. In both class one and two, the proportion of students who reported poor self-esteem on the negative worded items were mostly higher than those students who rated poor self-esteem on the positive-worded items (see Fig. 2 ). For example, 61.06%, 96.80% and 76.29% of students in class 1, 2, 3 respectively reported that they had low self-esteem on the item of “Sometimes, I think I am no good at all” (Table 1 ), which is higher than most of the other items. A possible explanation for the difference is that there may be wording effect from use of positive and negative wording. Some previous studies using factor analytical approach also reported that there may be method effects associated with negative-wording of the scale questions [ 66 , 67 ], although they measure a single construct of global self-esteem. This study observed a dose–response effect in the association between level of self-esteem and HRQoL. Students with lowest and moderated levels of self-esteem had lower HRQoL scores than the peers with high self-esteem, where the difference score in the HRQoL index comparing the lowest self-esteem and high self-esteem patterns was largest (decreased score = −0.103) among the three comparison groups. This finding suggests that the strength of the relationships between self-esteem and HRQoL follows the expected direction and the magnitude of the difference is in a dose–response manner. It is also important to consider whether the difference is clinically significant, or signifies minimally important difference (MID) as this is essential for informing and altering health intervention strategies [ 68 ]. To date, there is a lack of MID criterion for the EQ-5D-Y utility measure. Previous research has shown that the MID estimates for the adult version of EQ-5D index score ranged from 0.075 to 0.8 (distribution-based approaches), from 0.003 to 0.72 (anchor-based approaches) [ 69 ]. In this study, the EQ-5D-Y index scores comparing lowest (difference = 0.103) and moderated (difference = 0.033) levels of self-esteem with high self-esteem level respectively exceeds the MID threshold based on the EQ-5D index. Also, Cohen’d shows a large effect size (Cohen’d = 1.05) in the EQ-5D-Y index scores between high and low self-esteem patterns. The finding suggests that the difference in HRQoL across different self-esteem levels is not only statistically significant but also clinically important. There is a lack of research on the effect of self-esteem on the EQ-5D-Y index measure. More research on self-esteem and the EQ-5D-Y utility outcome in adolescents would warrant to strengthen the finding in this study. Along with the HRQoL outcome, the present study analyzed the relationship between self-esteem patterns and self-perceived health, and found a significant relation between low self-esteem patterns and poor self-rated health of students. This observation is in line with the findings in previous research [ 70 ], suggesting the correlation of self-esteem levels with inequalities of self-rated health in adolescents. As both the EQ-5D-Y index and self-perceived health represent an individual’s overall health state covering both physical health and mental health aspects, the finding suggests that low self-esteem may not only be an indicator for mental health, but also an important indicator for physical health in adolescents. Previous systematic reviews and individual studies have documented the impact of self-esteem on physical health outcomes in adolescents and adults [ 71 ]. A possible pathway for the effect of self-esteem on physical health may be through mental health aspect. The evidence suggests that higher self-esteem predicts fewer mental health problems [ 15 ]. We also found in this study that students in lower self-esteem levels reported more feeling worried, sad or unhappy problems (see Fig. 3 ). The findings support the previous research showing that poor self-esteem is linked with both poor mental health and low physical health in China and other countries [ 16 , 71 , 72 ]. This study further reveals the importance of key lifestyle behaviours, including breakfast habit, sleep time and physical activity for self-esteem. The finding highlights the association of poor breakfast habit like skipping a breakfast with low self-esteem, which supports the previous research documenting that skipping a breakfast or a meal is linked with lower self-esteem and more mental health problems [ 33 , 48 , 73 ]. The current study extends the previous research by investigating the relationship between sleep time and self-esteem. Our observation that insufficient sleep was related to low self-esteem patterns is in line with the finding from a few previous studies [ 36 , 37 ]. Similarly, our finding for the association between physical activity and self-esteem is consistent with previous studies demonstrating that a higher level of PA correlates with better self-esteem [ 17 , 32 , 35 ]. The observation that moderated to rigorous PA outside of schools is related to higher self-esteem is in line with a previous large-scale study in China reporting that moderate-to-vigorous PA correlated with better mental health outcomes [ 74 ]. Concerning the association of demographic characteristics with self-esteem, we observed that girls had a higher likelihood of experiencing low self-esteem than boys. The result coincides with other studies demonstrating that female adolescents had lower levels of self-esteem than boys [ 47 ]. This study analyzed cross-sectional associations between self-esteem and health outcomes and lifestyle behaviours. It is worth noting that the relationships between self-esteem, mental health and HRQoL, between lifestyle behaviour and self-esteem are likely bidirectional or in complex mediation pathways. For example, some prior studies reported that HRQoL was associated with self-esteem outcome or self-esteem impact lifestyle behaviour [ 26 , 34 ]. Future studies extending a longitudinal or prospective research design will warrant to explore bidirectional relationships or changes over time between the variables examined in this study. Future research that employs health intervention studies will help to ascertain causal relationships between self-esteem and health outcomes. In addition, studies that include other potential risk factors such as teachers’ interpersonal behaviour [ 75 ] will help to address how the teachers’ behaviour and support influence students’ mental health and HRQoL. Future research that includes diet and nutrients’ intake will help to further examine the effect of diet quality on self-esteem and HRQoL in adolescents. The data in this study were collected five years ago during the COVID-19 pandemic; thus, the results may not accurately represent the status that adolescents currently experience as self-esteem, lifestyle behaviours and health can change over time especially after the COVID-19 pandemic. However, research data after the COVID-19 pandemic showed that the prevalence of mental health disorders in Chinese adolescents remained high and a significant public health issue compared with the period of COVID-19 pandemic [ 4 – 6 , 8 ]. The 2022 national mental health survey in adolescents in China documented that the prevalence of depressive disorders in 2022 was 14.8% in adolescents, only slightly lower than the prevalence of 19% in 2020 ( https://dyh.sdu.edu.cn/info/1023/1250.htm ). In this study, the prevalence of low self-esteem (12%) of students is similar to the prevalence of depressive disorders in adolescents in 2022 when some schools were re-open for students. The recent study findings are largely consistent with our study results. More studies that compare data during and post-the pandemic will help to better delineate temporal changes and correlations of the aforementioned health-related behaviours and health outcomes. Strengths and Limitations Strengths of the study include the use of previously well-validated instruments for the outcome and exposure variables, the use of latent class analysis to address heterogeneity in self-esteem levels, and the examination of both HRQoL and self-rated health, and lifestyle behaviours simultaneously. Particularly, this study contributes to relevant research by revealing the presence of clustering patterns of self-esteem and the associations with HRQoL, perceived health and lifestyle behaviours in junior high school students. In addition, multivariable (logistic, linear and multinomial logistic) regressions were used to characterize the self-esteem patterns according to the health outcomes, lifestyle behaviours and demographic background variables among students, hence providing more robust results. Limitations of the present study should be acknowledged. This is a cross-sectional study, precluding ability to determine causal inference. Future research should employ a prospective study design and health intervention programs to address directionality and causal relationships between self-esteem and health outcomes, between lifestyle behaviours and self-esteem. As the survey was conducted in one city in the province, the generalizability of the study findings to junior high school students in other areas in China is limited. In addition, the study relied on students’ self-report measures although the measures used have been well validated, thus the results may be prone to recall bias. However, it is usually not feasible to use objective measures in population-based surveys due to limitations in human resources and financial supporting. Lastly, the study did not control for other possible confounders such as parental health-related behaviour or health, hence the results may have been impacted by residual confounding. Several implications from the findings in this study deserve to be summarized. School health promotion programs and health intervention strategies are recommended to place a priority in the interventions on those subgroups of junior high school students with unhealthy lifestyles and lower self-esteem patterns to enhance mental health and HRQoL. Interventions are needed to target promoting multiple healthy lifestyle behaviours, and self-esteem together rather than modifying single factor or consider the factors separately. Previous health intervention practice and research show that a comprehensive school health promotion approach that incorporates promoting physical activity, sleep quality and healthy eating simultaneously is more effective in improving health outcomes of children and adolescents than single behaviour intervention [ 76 ]. In addition, school educators, parents of students are needed to pay more attention to students’ behaviours and health, to work together to provide health education, healthy school and family environments, sufficient social and psychological supports to improve their mental health and HRQoL. Future research is needed to conduct more longitudinal and experimental studies among adolescents in various socio-economic and cultural contexts to better inform causal relations between self-esteem, lifestyle behaviours, and mental health and HRQoL. Conclusions The present study observed that self-esteem patterns are associated with health-related quality of life, self-rated health and mental health, breakfast habit, sleep and physical activity among junior high school students. Further research is needed to better elucidate the relationship between self-esteem and HRQoL and health outcomes by expanding longitudinal studies and health intervention programs among adolescents. Interventions that prioritize actions towards improving healthy eating, sleep quality and physical activity, and self-esteem will help address directionality and causality in the association of self-esteem with the lifestyle behaviours and health outcomes observed in the present study. Health interventions that target modifying low self-esteem and unhealthy lifestyle behaviours together may be more effective in enhancing HRQoL and health among adolescents than targeting a single influence factor. Supplementary Information 40359_2026_4258_MOESM1_ESM.pdf (136.9KB, pdf) Additional file 1: S1 Figure Illustration of the regression models. 40359_2026_4258_MOESM2_ESM.doc (276.5KB, doc) Additional file 2: S2 Figure Estimated probability of reporting low level of self-esteem on the Rosenberg self-esteem scale items across three classes. 40359_2026_4258_MOESM3_ESM.doc (99.5KB, doc) Additional file 3: S3 Table 1 Fit statistics of the latent class models using 10 self-esteem variables. S3 Table 2 Predicted probability on the response categories of self-esteem items by classes based on the three-class model. S3 Table 3 Frequency distribution of the self-esteem variables with four response categories. Acknowledgements The authors would like to thank the students, their parents and schools for their participation in this study. Authors’ contributions XYW designed the study, analyzed and interpreted data. RWW and XYW wrote a draft, and carefully revised the manuscript. PZH, TZH, QAY, AIHW, XZHD, CHSH, SNZH, JYL reviewed and critically revised the manuscript. TZH organized data collection. All authors participated in the writing of the manuscript, read and approved the final manuscript. Funding The present analyses were supported by the 2023 Shandong Higher Education Youth Research Innovation Team Program (No.: 2023KJ372). The survey was partly supported by a research grant to XYW provided by Weifang Medical University (grant no: 2017BSQD61). All interpretations and opinions in the present study are those of the authors. Data availability The datasets analyzed for this study are not publicly available due to privacy policies but are available on a reasonable request from the corresponding author. Declarations Ethics approval and consent to participate The study was conducted in accordance with the statement of ‘Declaration of Helsinki’. The survey and data collection for the present study was approved by the Human Research Ethics Committee of the Weifang Medical University (reference number 2020YX070), Shandong Province. Written informed consent for participation in this study was provided by the participants’ legal guardians/next of kin, and students themselves. 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. Contributor Information Xiangzheng Ding, Email: [email protected]. Aihua Wang, Email: [email protected]. Xiuyun Wu, Email: [email protected]. References 1. Surkalim DL, Luo M, Eres R, Gebel K, van Buskirk J, Bauman A, et al. The prevalence of loneliness across 113 countries: systematic review and meta-analysis. BMJ. 2022;376:e067068. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 2. Feiss R, Dolinger SB, Merritt M, Reiche E, Martin K, Yanes JA, et al. A Systematic Review and Meta-Analysis of School-Based Stress, Anxiety, and Depression Prevention Programs for Adolescents. J Youth Adolesc. 2019;48(9):1668–85. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 3. Kiviruusu O, Ranta K, Lindgren M, Haravuori H, Silén Y, Therman S, et al. Mental health after the Covid-19 pandemic among Finnish youth: a repeated, cross-sectional, population-based study. Lancet Psychiatry. 2024;11(6):451–60. [ DOI ] [ PubMed ] [ Google Scholar ] 4. Cai J, Wang Y, Wang C, Deng ZY, Mu YF, Deng AP, et al. Mental health problems and associated factors of students at different learning stages during and after Covid-19 pandemic in Sichuan, China. BMC Psychiatry. 2025;25(1):144. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 5. Mental Health of Adolescents. World Health Organization. Available from: https://www.who.int/news-room/fact-sheets/detail/adolescent-mental-health . Accessed on 13 December, 2025. Updated September 1, 2025. 6. Li Y, Chen Q, Liu L, Yang X, Mu H, Wang Q, et al. Disease burden of mental disorders among children and adolescents considering both co-morbidities and suicide in northeastern China. BMC Public Health. 2024;24(1):1359. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 7. Li F, Cui Y, Li Y, Guo L, Ke X, Liu J, et al. Prevalence of mental disorders in school children and adolescents in China: diagnostic data from detailed clinical assessments of 17,524 individuals. J Child Psychol Psychiatry. 2022;63(1):34–46. [ DOI ] [ PubMed ] [ Google Scholar ] 8. Burden of 375 Diseases and Injuries, Risk-Attributable Burden of 88 Risk Factors, and Healthy Life Expectancy in 204 Countries and Territories, Including 660 Subnational Locations, 1990–2023: A Systematic Analysis for the Global Burden of Disease Study 2023. Lancet. 2025;406(10513):1873–922. [ DOI ] [ PMC free article ] [ PubMed ] 9. Rosenberg M. Society and the adolescent self-image. Princeton, NJ: Princeton University Press; 1965. [ Google Scholar ] 10. von Soest T, Wichstrøm L, Kvalem IL. The development of global and domain-specific self-esteem from age 13 to 31. J Pers Soc Psychol. 2016;110(4):592–608. [ DOI ] [ PubMed ] [ Google Scholar ] 11. Moksnes UK, Espnes GA, Eilertsen MEB, Bjørnsen HN, Ringdal R, Haugan G. Validation of Rosenberg Self-Esteem Scale among Norwegian adolescents - psychometric properties across samples. BMC Psychol. 2024;12(1):506. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 12. Mann M, Hosman C, Schaalma H, Vries N. Self-esteem in a broad-spectrum approach for mental health promotion. Health Educ Res. 2004;19:357–72. [ DOI ] [ PubMed ] [ Google Scholar ] 13. McGee R, Williams S, Nada-Raja S. Low self-esteem and hopelessness in childhood and suicidal ideation in early adulthood. J Abnorm Child Psychol. 2001;29(4):281–91. [ DOI ] [ PubMed ] [ Google Scholar ] 14. Lee CG, Seo DC, Torabi MR, Lohrmann DK, Song TM. Longitudinal trajectory of the relationship between self-esteem and substance use from adolescence to young adulthood. J Sch Health. 2018;88(1):9–14. [ DOI ] [ PubMed ] [ Google Scholar ] 15. Sowislo JF, Orth U. Does low self-esteem predict depression and anxiety? A meta-analysis of longitudinal studies. Psychol Bull. 2013;139(1):213–40. [ DOI ] [ PubMed ] [ Google Scholar ] 16. Carlén K, Suominen S, Augustine L. The association between adolescents’ self-esteem and perceived mental well-being in Sweden in four years of follow-up. BMC Psychol. 2023;11(1):413. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 17. Wang F, Veugelers PJ. Self-esteem and cognitive development in the era of the childhood obesity epidemic. Obes Rev. 2008;9(6):615–23. [ DOI ] [ PubMed ] [ Google Scholar ] 18. Ferro MA, Boyle MH. Self-concept among youth with a chronic illness: a meta-analytic review. Health Psychol. 2013;32(8):839–48. [ DOI ] [ PubMed ] [ Google Scholar ] 19. Cvencek D, Fryberg SA, Covarrubias R, Meltzoff AN. Self-concepts, self-esteem, and academic achievement of minority and majority North American elementary school children. Child Dev. 2018;89(4):1099–109. [ DOI ] [ PubMed ] [ Google Scholar ] 20. Ravens-Sieberer U, Erhart M, Wille N, Wetzel R, Nickel J, Bullinger M. Generic health-related quality-of-life assessment in children and adolescents: methodological considerations. Pharmacoeconomics. 2006;24(12):1199–220. [ DOI ] [ PubMed ] [ Google Scholar ] 21. Wu XY, Han LH, Zhang JH, Luo S, Hu JW, Sun K. The influence of physical activity, sedentary behavior on health-related quality of life among the general population of children and adolescents: a systematic review. PLoS ONE. 2017;12(11):e0187668. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 22. Idler EL, Benyamini Y. Self-rated health and mortality: a review of twenty-seven community studies. J Health Soc Behav. 1997;38(1):21–37. [ PubMed ] [ Google Scholar ] 23. Zhang T, Lu G, Wu XY. Associations between physical activity, sedentary behaviour and self-rated health among the general population of children and adolescents: a systematic review and meta-analysis. BMC Public Health. 2020;20(1):1343. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 24. Mikkelsen HT, Haraldstad K, Helseth S, Skarstein S, Småstuen MC, Rohde G. Health-related quality of life is strongly associated with self-efficacy, self-esteem, loneliness, and stress in 14-15-year-old adolescents: a cross-sectional study. Health Qual Life Outcomes. 2020;18(1):352. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 25. Griffiths LJ, Parsons TJ, Hill AJ. Self-esteem and quality of life in obese children and adolescents: a systematic review. Int J Pediatr Obes. 2010;5(4):282–304. [ DOI ] [ PubMed ] [ Google Scholar ] 26. Meland E, Breidablik HJ, Thuen F, Samdal GB. How body concerns, body mass, self-rated health and self-esteem are mutually impacted in early adolescence: a longitudinal cohort study. BMC Public Health. 2021;21(1):496. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 27. Benzing V, Siegwart V, Spitzhüttl J, Schmid J, Grotzer M, Roebers CM, et al. Motor ability, physical self-concept and health-related quality of life in pediatric cancer survivors. Cancer Med. 2021;10(5):1860–71. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 28. Poorolajal J, Sahraei F, Mohamdadi Y, Doosti-Irani A, Moradi L. Behavioral factors influencing childhood obesity: a systematic review and meta-analysis. Obes Res Clin Pract. 2020;14(2):109–18. [ DOI ] [ PubMed ] [ Google Scholar ] 29. Biddle SJ, Asare M. Physical activity and mental health in children and adolescents: a review of reviews. Br J Sports Med. 2011;45(11):886–95. [ DOI ] [ PubMed ] [ Google Scholar ] 30. Wu XY, Zhuang LH, Li W, Guo HW, Zhang JH, Zhao YK, et al. The Influence of Diet Quality and Dietary Behavior on Health-Related Quality of Life in the General Population of Children and Adolescents: A Systematic Review and Meta-Analysis. Qual Life Res. 2019;28(8):1989–2015. [ DOI ] [ PubMed ] [ Google Scholar ] 31. Qin Z, Wang N, Ware RS, Sha Y, Xu F. Lifestyle-related behaviors and health-related quality of life among children and adolescents in China. Health Qual Life Outcomes. 2021;19(1):8. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 32. Wu X, Kirk SF, Ohinmaa A, Veugelers P. Health behaviours, body weight and self-esteem among grade five students in Canada. Springerplus. 2016;5(1):1099. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 33. Eckert KF, Asbridge M, Campbell LA, Stewart S, Bennett M, Loewen OK, et al. Meal Regularity Is Associated with Self-Esteem among Grade 5 Children. Am J Clin Nutr. 2021;113(2):467–75. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 34. Park J, Kim Y-H, Park S-J, Suh S, Lee H-J. The relationship between self-esteem and overall health behaviors in Korean adolescents. Health Psychol Behav Med. 2016;4(1):175–85. [ Google Scholar ] 35. Liu M, Wu L, Ming Q. How does physical activity intervention improve self-esteem and self-concept in children and adolescents? Evidence from a meta-analysis. PLoS ONE. 2015;10(8):e0134804. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 36. Jakobsson M, Josefsson K, Jutengren G, Sandsjö L, Högberg K. Sleep duration and sleeping difficulties among adolescents: exploring associations with school stress, self-perception and technology use. Scand J Caring Sci. 2019;33(1):197–206. [ DOI ] [ PubMed ] [ Google Scholar ] 37. Jiménez Boraita R, Gargallo Ibort E, Dalmau Torres JM, Arriscado Alsina D. Lifestyle habits, health indicators and sociodemographic factors associated with health-related quality of life and self-esteem in adolescents. Clin Child Psychol Psychiatry. 2024;29(2):493–512. [ DOI ] [ PubMed ] [ Google Scholar ] 38. Arsandaux J, Galéra C, Salamon R. The association of self-esteem and psychosocial outcomes in young adults: a 10-year prospective study. Child Adolesc Ment Health. 2021;26(2):106–13. [ DOI ] [ PubMed ] [ Google Scholar ] 39. Wild LG, Flisher AJ, Bhana A, Lombard C. Associations among adolescent risk behaviours and self-esteem in six domains. J Child Psychol Psychiatry. 2004;45(8):1454–67. [ DOI ] [ PubMed ] [ Google Scholar ] 40. Alessandri G, Vecchione M, Eisenberg N, Łaguna M. On the factor structure of the Rosenberg (1965) general self-esteem scale. Psychol Assess. 2015;27(2):621–35. [ DOI ] [ PubMed ] [ Google Scholar ] 41. Gong X, Zheng J, Zhou J, Huebner ES, Tian L. Global and domain-specific self-esteem from middle childhood to early adolescence: co-developmental trajectories and directional relations. J Pers. 2024;92(5):1356–74. [ DOI ] [ PubMed ] [ Google Scholar ] 42. Weller BE, Bowen NK, Faubert SJ. Latent class analysis: a guide to best practice. J Black Psychol. 2020;46(4):287–311. [ Google Scholar ] 43. Laxer RE, Brownson RC, Dubin JA, Cooke M, Chaurasia A, Leatherdale ST. Clustering of risk-related modifiable behaviours and their association with overweight and obesity among a large sample of youth in the Compass Study. BMC Public Health. 2017;17(1):102. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 44. Li X, Sun M, Yao N, Liu J, Wang L, Hu W, et al. Association between Patterns of Eating Habits and Mental Health Problems in Chinese Adolescents: A Latent Class Analysis. Front Nutr. 2022;9:906883. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 45. McChesney G, Toseeb U. Happiness, self-esteem, and prosociality in children with and without autism spectrum disorder: evidence from a Uk population cohort study. Autism Res. 2018;11(7):1011–23. [ DOI ] [ PubMed ] [ Google Scholar ] 46. Miyagawa Y, Kanemasa Y, Taniguchi J. A compassionate and worthy self: latent profiles of self-compassion and self-esteem in relation to intrapersonal and interpersonal functioning. Curr Psychol. 2024;43(16):14259–72. [ Google Scholar ] 47. Van Damme L, Colins OF, Vanderplasschen W. Gender differences in psychiatric disorders and clusters of self-esteem among detained adolescents. Psychiatry Res. 2014;220(3):991–7. [ DOI ] [ PubMed ] [ Google Scholar ] 48. Gong WJ, Fong DY, Wang MP, Lam TH, Chung TW, Ho SY. Skipping breakfast and eating breakfast away from home were prospectively associated with emotional and behavioral problems in 115,217 Chinese adolescents. J Epidemiol. 2022;32(12):551–8. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 49. Liu ZH, Wang YL, Yu YS, Ren Y, Zhang T, Liu HQ, et al. The individual and combined associations of health behaviours with health-related quality of life amongst junior high school students in China. Front Public Health. 2023;11:1283721. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 50. Schmitt DP, Allik J. Simultaneous administration of the Rosenberg Self-Esteem Scale in 53 nations: exploring the universal and culture-specific features of global self-esteem. J Pers Soc Psychol. 2005;89(4):623–42. [ DOI ] [ PubMed ] [ Google Scholar ] 51. Jiang C, Zhu Y, Luo Y, Tan CS, Mastrotheodoros S, Costa P, et al. Validation of the Chinese Version of the Rosenberg Self-Esteem Scale: Evidence from a Three-Wave Longitudinal Study. BMC Psychol. 2023;11(1):345. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 52. Lo MT, Chen SK, O’Connell AA. Psychometric properties and convergent validity of the Chinese version of the Rosenberg Self-Esteem Scale. J Appl Meas. 2018;19(4):413–27. [ PubMed ] [ Google Scholar ] 53. Wei J, Zhang J, Mao X. Psychometric properties of Rosenberg Self-Esteem Scale-Revised for middle school students. Chin J Clin Psychol. 2018;26:629–37. [ Google Scholar ] 54. Wille N, Badia X, Bonsel G, Burström K, Cavrini G, Devlin N, et al. Development of the Eq-5d-Y: A Child-Friendly Version of the Eq-5d. Qual Life Res. 2010;19(6):875–86. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 55. Roudijk B, Sajjad A, Essers B, Lipman S, Stalmeier P, Finch AP. A value set for the Eq-5d-Y-3l in the Netherlands. Pharmacoeconomics. 2022;40(Suppl 2):193–203. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 56. Pan T, Roudijk B, Devlin N, Mulhern B, Norman R. An Australian value set for the Eq-5d-Y-3l. Health Qual Life Outcomes. 2025;23(1):72. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 57. Yang Z, Jiang J, Wang P, Jin X, Wu J, Fang Y, et al. Estimating an Eq-5d-Y-3l Value Set for China. Pharmacoeconomics. 2022;40(Suppl 2):147–55. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 58. Pei W, Yue S, Zhi-Hao Y, Ruo-Yu Z, Bin W, Nan L. Testing measurement properties of two Eq-5d youth versions and Kidscreen-10 in China. Eur J Health Econ. 2021;22(7):1083–93. [ DOI ] [ PubMed ] [ Google Scholar ] 59. Qin L, Ho WKY, Khoo S. The Chinese version of the Physical Activity Questionnaire for Adolescents: a psychometric validity, reliability, and invariance study. BMC Public Health. 2024;24(1):3117. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 60. Li L, Zhang Y, Fan M, Cao B. Sleep and mental health among Chinese adolescents: the chain-mediating role of physical health perception and school adjustment. BMC Psychol. 2024;12(1):228. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 61. Nylund KL, Asparouhov T, Muthén BO. Deciding on the number of classes in latent class analysis and growth mixture modeling: a Monte Carlo simulation study. Struct Equ Modeling. 2007;14(4):535–69. [ Google Scholar ] 62. Cohen J. Statistical power analysis for the behavioral sciences. 2nd Ed. Routledge; 1988. 10.4324/97802037715871988. [ Google Scholar ] 63. Cramér H. Mathematical methods of statistics. Princeton: Princeton University Press; 1946. [ Google Scholar ] 64. Muthén LK, Muthén BO. Mplus User’s Guide. Eighth Edition. Los Angeles, Ca: Muthén & Muthén. (1998–2017). https://www.statmodel.com/download/usersguide/MplusUserGuideVer_8.pdf 65. de Melo do Espirito Santo C, Santos VS, Chiarotto A, Miyamoto GC, Yamato TP. Measurement properties of the Eq-5d instruments in children and adolescents: a systematic review. Appl Health Econ Health Policy. 2025;23(5):797–822. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 66. Wu Y, Zuo B, Wen F, Yan L. Rosenberg self-esteem scale: method effects, factorial structure and scale invariance across migrant child and urban child populations in China. J Pers Assess. 2017;99(1):83–93. [ DOI ] [ PubMed ] [ Google Scholar ] 67. DiStefano C, Motl RW. Further investigating method effects associated with negatively worded items on self-report surveys. Struct Equ Modeling Multidiscip J. 2006;13(3):440–64. [ Google Scholar ] 68. Jaeschke R, Singer J, Guyatt GH. Measurement of health status. Ascertaining the minimal clinically important difference. Control Clin Trials. 1989;10(4):407–15. [ DOI ] [ PubMed ] [ Google Scholar ] 69. Al Sayah F, Jin X, Short H, McClure NS, Ohinmaa A, Johnson JA. A systematic literature review of important and meaningful differences in the Eq-5d index and visual analog scale scores. Value Health. 2025;28(3):470–6. [ DOI ] [ PubMed ] [ Google Scholar ] 70. Jafflin K, Pfeiffer C, Bergman MM. Effects of self-esteem and stress on self-assessed health: a Swiss study from adolescence to early adulthood. Qual Life Res. 2019;28(4):915–24. [ DOI ] [ PubMed ] [ Google Scholar ] 71. Orth U, Robins RW. Is high self-esteem beneficial? Revisiting a classic question. Am Psychol. 2022;77(1):5–17. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 72. Li H, Hao F. The influence of physical activity on the mental health of high school students: the chain mediating effects of social support and self-esteem. Sci Rep. 2025;15(1):27537. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 73. Gratão LHA, Pessoa MC, Rodrigues da Silva TP, Rocha LL, Louise Cassimiro Inácio M, Resende Prado Rangel de Oliveira T, de Freitas Cunha C, et al. Dietary patterns, breakfast consumption, meals with family and associations with common mental disorders in adolescents: a school-based cross-sectional study. BMC Public Health. 2022;22(1):980. [ DOI ] [ PMC free article ] [ PubMed ] 74. Yan W, Wang Y, Yuan Y, Farid M, Zhang P, Peng K. Timing matters: a longitudinal study examining the effects of physical activity intensity and timing on adolescents’ mental health outcomes. J Youth Adolesc. 2024;53(10):2320–31. [ DOI ] [ PubMed ] [ Google Scholar ] 75. Tilga H, Hein V, Koka A, Hagger M. How physical education teachers’ interpersonal behaviour is related to students’ health-related quality of life. Scand J Educ Res. 2020;64:661–76. [ Google Scholar ] 76. Veugelers PJ, Dabravolskaj J, Khan MKA, Tran TT, Flynn J, Maximova K. From best practice to next practice: implementing comprehensive school health in rural and remote Northern communities. Health Promot Chronic Dis Prev Can. 2022;42(8):344–52. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials 40359_2026_4258_MOESM1_ESM.pdf (136.9KB, pdf) Additional file 1: S1 Figure Illustration of the regression models. 40359_2026_4258_MOESM2_ESM.doc (276.5KB, doc) Additional file 2: S2 Figure Estimated probability of reporting low level of self-esteem on the Rosenberg self-esteem scale items across three classes. 40359_2026_4258_MOESM3_ESM.doc (99.5KB, doc) Additional file 3: S3 Table 1 Fit statistics of the latent class models using 10 self-esteem variables. S3 Table 2 Predicted probability on the response categories of self-esteem items by classes based on the three-class model. S3 Table 3 Frequency distribution of the self-esteem variables with four response categories. Data Availability Statement The datasets analyzed for this study are not publicly available due to privacy policies but are available on a reasonable request from the corresponding author. Articles from BMC Psychology are provided here courtesy of BMC ACTIONS View on publisher site PDF (1.6 MB) Cite Collections Permalink PERMALINK Copy RESOURCES Similar articles Cited by other articles Links to NCBI Databases Cite Copy Download .nbib .nbib Format: AMA APA MLA NLM Add to Collections Create a new collection Add to an existing collection Name your collection * Choose a collection Unable to load your collection due to an error Please try again Add Cancel Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top