Introduction

In recent years, a concerning global trend has emerged regarding the prevalence of reported health problems among adolescents. Adolescence, typically occurring between the ages of 10 and 24, marks the critical transition from childhood to adulthood (Sawyer et al. 2018). Despite the relatively low likelihood of serious illness among individuals in this age group, there has been a gradual increase in self-rated health problems (Ravens-Sieberer et al. 2009; Liu et al. 2023d) across various dimensions, such as mental and physical health (Mojtabai and Olfson 2020; Knaappila et al. 2021; Deighton et al. 2019; Houri et al. 2012). For instance, a survey of 2378 South Korean primary and middle school students revealed a gradual decline in average self-rated health during early adolescence (Kim and Choi 2018). Similar trends have been observed in both Europe and North America, indicating an increasing prevalence of subjective health complaints ranging from difficulty falling asleep to dizziness and headache (Cosma et al. 2020). World Health Organization (WHO) data from 298 school surveys in 146 countries and regions around the world reveal that more than four-fifths of adolescents do not meet the recommended one hour of physical activity per day, which is detrimental to their current and future health (Guthold et al. 2020). Notably, China is also facing similar challenges (Liu et al. 2023c; UNICEF 2021). A cross-sectional survey of more than 1000 senior high school students in Nanjing, China, revealed that nearly 50% of the respondents rated their general health as fair, while nearly 6% rated it poor (Qi et al. 2020). Several birth cohort studies in China have revealed a worrying decline in self-reported adolescent health since the early 1990s, which has been attributed to escalating social challenges, including inadequacies in the Chinese health insurance system, a rising Gini coefficient, and increased peer competition in academic and employment settings (Xin et al. 2012; Su and Liu 2022; Liu et al. 2023b). Additionally, nationwide policies implemented by the Chinese government during the COVID-19 pandemic, such as social distancing, home confinement and school closures, have had a significant impact on adolescent health, leading to increased reports of symptoms of depression and fatigue (Zhang et al. 2020). At present, despite a narrative suggesting a positive improvement in reported mental and physical health among adolescents (Potrebny et al. 2017; Duinhof et al. 2015), a larger body of evidence points to a general trend toward deterioration in self-rated health status with age in many countries globally (Jerdén et al. 2011; Wade and Vingilis 1999; Okada et al. 2022). Given the recurring challenges pertaining to adolescent health, an in-depth exploration of factors related to self-rated health has become increasingly urgent and imperative.

Previous evidence suggests that the Big Five personality traits, as an internal and persistent system of characteristics, can reflect an individual’s subjective perception of health and influence their health behaviors. The potential relationship between self-rated health and the Big Five personality traits deserves further attention (Tauber 2018; Stephan et al. 2020; Mund and Neyer 2016; Luo et al. 2022a; Heilmann et al. 2021). In the context of the Big Five personality traits, neuroticism is often considered a negative personality trait that is harmful to health, and high neuroticism is associated with more negative emotional experiences and frequent psychological disorders (Dahl et al. 2022). The relationships between conscientiousness, agreeableness, extraversion, openness, and health are more complex. Most related studies have focused on the positive and beneficial effects of these four personality traits on health. For instance, high conscientiousness indicates a high degree of self-control, adherence to healthy behaviors, and avoidance of unhealthy behaviors (Sutin et al. 2016; Bogg and Roberts 2004). Agreeableness refers to good interpersonal relationships and better adaptability, which may be beneficial for physical and mental health (Zhao et al. 2022). Extraversion represents a tendency to be sociable and outgoing, with high extraverts having strong social support networks and being more likely to perceive hope and happiness (Tan et al. 2018; Liu et al. 2023a). Openness emphasizes a tendency to be curious and open-minded, with highly open people able to use creative coping strategies to address health problems (Cao et al. 2023; Watson and Hubbard 1996). Relevant theories suggest that individual personality traits are not only associated with health, but that there may also be bidirectional relationships. According to the core principles of lifespan developmental theory, human development unfolds through a continuous interplay between the individual, biology, and culture, with personality traits exhibiting average levels and changes throughout the lifespan (Wurm and Kornadt 2019). The association between personality and physical well-being persists over long periods of time, spanning the entire lifespan (Baltes and Baltes 1990; Mroczek et al. 2020). Furthermore, the principle of the neo-socioanalytic model of personality development further purposes that individuals’ specific life experiences, including health-related encounters, are shaped by their personality traits. These experiences, in turn, may influence personality traits, creating a feedback loop in which changes in personality and health interact (Luo et al. 2022a). In summary, these theoretical perspectives provide a reference for further investigation of the bidirectional relationship between self-rated health and the Big Five personality traits.

However, there remains a scarcity of empirical evidence that thoroughly explores the directional nature of the relationship between self-rated health and the Big Five personality traits. Previous research has mainly used cross-sectional data for correlational analyses or examined unidirectional relationships. In addition, certain studies have focused exclusively on isolated dimensions of the Big Five personality traits concerning health, resulting in a dearth of comprehensive investigations into the bidirectional interplay between self-rated health and the entirety of the Big Five personality traits. This study uses a large-scale longitudinal survey dataset and employs a two-wave cross-lagged model to elucidate the relationships between self-rated health and the Big Five personality traits among Chinese adolescents. The findings enhance our holistic understanding of the interplay between these variables and serve as a cornerstone for the development of targeted interventions to improve adolescent health and personality traits.

Literature review

Self-rated health

Self-rated health encompasses an individual’s comprehensive subjective perception of his or her well-being across multiple facets, including physical, psychological and social dimensions, and aptly reflects related functional aspects (Wang et al. 2022). This concept is highly important for predicting an individual’s overall health status. On the one hand, self-rated health has a remarkable ability to predict future morbidity and mortality (Jylhä 2009; Hetlevik et al. 2020). Higher self-rated health is associated with better adult survival (Chipperfield 1993). On the other hand, there is concordance between self-rated health and objective measures of health. Poorer self-rated health is correlated with an increased incidence of abnormalities in laboratory parameters and conditions such as cardio-cerebral vascular diseases, digestive diseases, respiratory diseases, and mental illnesses(Wu et al. 2013). Self-rated health uniquely reflects objective health outcomes, suggesting that self-rated health may not only reflect diagnosed diseases but also capture preclinical symptoms that have not yet been diagnosed (Idler and Benyamini 1997). In this context, self-rated health is a more accurate predictor of mortality among young people than among older people (Franks et al. 2003). Given the urgency of addressing adolescent health concerns and the significant utility of self-rated health, it is imperative to investigate adolescents’ self-rated health and the key factors associated with it. This has important practical implications for disease prevention and health promotion in the early stages of life, particularly during adolescence.

Big five personality traits

The Big Five personality traits, which include openness, conscientiousness, agreeableness, extraversion, and neuroticism, are among the most widely accepted and discussed personality trait models (Goldberg 1990; Mccrae and John 1992). Known for its high accuracy, reliability and widespread use, the Big Five model has cemented its status as a universally recognized and highly valued measure across multiple disciplines worldwide (Marsh et al. 2010; Mccrae 2011; Kautz et al. 2014; Alderotti et al. 2023; Heineck and Anger 2010; Abbasi et al. 2023; Goldberg 1992). Evidence suggests that personality generally develops in childhood and matures in adulthood, with changes in the Big Five personality traits occurring more frequently during adolescence than at any other stage of life (Roberts et al. 2006). Regarding the development of average personality trait levels in adolescents, current research tends to support two opposing hypotheses that provide a theoretical basis for understanding potential fluctuations in the Big Five personality traits. The first is the maturity principle, which suggests a relatively steady increase in personality maturity from infancy through childhood, adolescence, and adulthood. The second is the disruption hypothesis, which proposes that certain aspects of personality maturity temporarily decline during the biological, social, and psychological transitions from childhood to adolescence (Soto and Tackett 2015). Prior literature focusing on adult populations has supported the maturity hypothesis. A meta-analysis of 92 studies revealed that social dominance (an aspect of extraversion), responsibility, and emotional stability increase in adulthood (Roberts et al. 2006). Cross-cultural research involving adult populations from 62 countries has also shown a continuing trend toward personality maturity, largely attributed to the transition from normative life to adult roles (Bleidorn et al. 2013). However, research on adolescent personality development tends to be more consistent with the disruption hypothesis. In particular, adolescents often exhibit decreases in conscientiousness, agreeableness (Ibáñez et al. 2016), openness (Denissen et al. 2013) and extraversion (Soto et al. 2011), coupled with increases in neuroticism during early adolescence (Van Den Akker et al. 2021). Some studies have yielded mixed conclusions. For example, a study of 576 Dutch adolescents found that during the transition from late childhood to early adulthood, personality development followed the maturity principle, as evidenced by increased self-reported and parent-rated agreeableness and conscientiousness. When examining the adolescent phase, the conclusions supported the disruption hypothesis, manifested by a decline in parent-observed agreeableness (Luan et al. 2017). However, most of the above studies are rooted in Western countries, and the findings among adolescents in the Chinese context need further verification.

Relationships between adolescent health and the Big Five personality traits

Recent research has suggested a potential correlation between adolescent health status and the Big Five personality traits. According to a large-scale survey conducted in Norway, higher levels of neuroticism among adolescents were closely associated with a greater incidence of chronic fatigue, anxiety, or depression, obesity, and daily smoking (Dahl et al. 2022). A study of students from 20 schools in India showed that higher levels of neuroticism and lower levels of conscientiousness among adolescents were associated with greater involvement in health risk behaviors such as alcohol consumption, poor diet and substance abuse (Singh 2022). Evidence from a study in the United Kingdom showed that greater openness in adolescence was associated with better cognitive function both at the same time and even 30 years later (Sutin et al. 2021). A prospective meta-analysis revealed that higher levels of agreeableness and extraversion and a lower risk of fatigue were associated with better health outcomes (Stephan et al. 2020). Other studies further discussed the correlation between self-rated health and the Big Five personality traits, but the specific relationship between these two variables remains unclear. Based on investigations with large samples from the United States, Europe, and Japan, it has been found that higher neuroticism is associated with lower self-rated health, while higher levels of extraversion, openness, agreeableness, and conscientiousness are associated with better self-rated health (Stephan et al. 2020). Similar results were found in a study in Hawaii, where increases in agreeableness, conscientiousness, and extraversion and decreases in neuroticism were associated with improved self-rated health in people of the same age group, even in the relatively stable middle-aged phase of personality and health (Letzring et al. 2014). However, there are also contradictory findings. For instance, a study of German adolescents found that those with higher neuroticism and conscientiousness self-rated their health as poorer, while openness was unrelated to self-rated health status (Heilmann et al. 2021). Adolescents’ extraversion and agreeableness were associated with life satisfaction but not with self-rated health. In addition, very few longitudinal studies have revealed a one-way relationship between self-rated health and the Big Five personality traits. On the one hand, the Big Five personality traits may predict people’s subjective assessment of their health. Personality can influence self-rated health by shaping individuals’ positive or negative subjective interpretations of their objective health status and by influencing subjective health through health-related behaviors and stress responses (Löckenhoff et al. 2012). For example, lower neuroticism is associated with more positive emotional experiences and greater happiness and less depression and anxiety (Diaconu-Gherasim and Mardari 2022). Conscientiousness is related to organization and discipline, which makes highly conscientious individuals more likely to engage in health-promoting behaviors such as physical exercise (Sutin et al. 2016) and less likely to engage in health-damaging behaviors such as alcohol and drug abuse or reckless driving (Bogg and Roberts 2004). Extraversion, which represents sociability, is associated with increased smoking, alcohol and marijuana use and an increased risk of cardiovascular disease. It may also lead to greater participation in health-promoting physical activities (Stephan et al. 2012; Yañez et al. 2020). Agreeableness and openness, which represent social harmony and curiosity, creativity, and openness of thought, respectively, have been associated with better self-rated health outcomes (Sirois 2015; Stephan et al. 2020; Tauber 2018). However, contrasting findings have also suggested that agreeableness or openness may not be associated with self-rated health (Löckenhoff et al. 2012; Heilmann et al. 2021). This further highlights the complex predictive role of the Big Five personality traits on self-rated health. These inconsistent conclusions from existing research highlight the need for further investigation into the direction of the relationships between various dimensions of personality and self-rated health. On the other hand, there is a paucity of research discussing the predictive role of self-rated health on the Big Five personality traits. Longitudinal investigations conducted in Boston, the Netherlands, and Sweden suggest that an increase in self-rated health can predict subsequent increases in conscientiousness and extraversion, providing evidence for the predictive role of prior self-rated health outcomes in subsequent changes in personality traits (Luo et al. 2022a). Furthermore, few studies have examined the bidirectional relationship between self-rated health and the Big Five personality traits. Analysis of data from the Interdisciplinary Longitudinal Study on Adult Development and Aging (ILSE) found that baseline extraversion and conscientiousness positively predicted subsequent self-rated health levels, whereas baseline health predicted higher levels of subsequent extraversion (Tauber 2018). Similarly, a 15-year longitudinal study in Germany showed that higher neuroticism predicted lower levels of subjective health in later years, and better subjective health predicted lower levels of neuroticism 15 years later (Mund and Neyer 2016). Unfortunately, these conclusions are largely based on Western contexts, with limited evidence from China, prompting further exploration in this study.

The present study

The purpose of the present study is to examine the directional relationship between adolescents’ self-rated health and the Big Five personality traits using a two-wave cross-lagged model. In addition, given the influence of gender on adolescents’ personality and health development, as indicated by the literature (Moksnes and Espnes 2020; Wilson et al. 2005; De Bolle et al. 2015; Soto et al. 2011; Iimura and Taku 2018; Kim 2022; Luo et al. 2022b), we also tested the robustness of the relationships between self-rated health and the Big Five personality traits after controlling for gender. Building upon the literature review, we have identified several aspects of previous studies that warrant further investigation. For instance, some studies have used cross-sectional data to analyze the correlation between these factors. The limited longitudinal studies have focused mainly on adult populations, with a scarcity of analyses involving samples of Chinese adolescents. Additionally, the literature often focuses on isolated facets of the Big Five personality traits, lacking a comprehensive analysis that encompasses all dimensions. Furthermore, although existing theories suggest a correlation between individual personality traits and health, possibly in a dynamic relationship (Mroczek et al. 2020; Baltes and Baltes 1990; Luo et al. 2022a), empirical research is relatively scarce. In the context of China, little is known about the average developmental levels of self-rated health and the Big Five personality traits among adolescents, as well as about their bidirectional relationship. The present study uses two-wave longitudinal survey data from the same cohort of adolescents to empirically analyze the directional relationship between self-rated health and the Big Five personality traits over time. By exploring these relationships, we can provide a more comprehensive, empirically supported perspective on promoting adolescent health and personality development and make targeted recommendations. On the basis of the literature, the purpose of the study and the relevant theoretical reference, we propose the following hypotheses: (1) self-rated health can significantly positively predict openness, conscientiousness, agreeableness, and extraversion while significantly negatively predicting neuroticism; (2) openness, conscientiousness, agreeableness, and extraversion can significantly positively predict self-rated health, while neuroticism can significantly negatively predict self-rated health.

Methods

Participants

The data used in this study are from the China Education Panel Survey (CEPS) database. The survey was designed and conducted by the National Survey Research Center (NSRC) of Renmin University of China and is nationally representative. For the CEPS, a multistage probability proportional to size and a whole-class sampling design were used. The survey started with two simultaneous cohorts: grade 7 and grade 9 of middle school. Based on the average level of education and the proportion of the floating population, 28 county-level units were randomly selected as survey points from across the country. The survey was school-based and involved the random selection of 112 schools and 438 classes from the selected county-level units, with all students in the selected classes included in the sample. In the follow-up survey, only students in grade 7 at baseline were followed. To date, data from two waves have been published: the baseline survey for 2013–2014 and the follow-up survey for 2014–2015. In this study, we refer to the baseline and follow-up survey time points as Time1 and Time2, respectively. In accordance with the research objectives, we selected students who participated in both waves of the survey. In the end, 7967 student samples were retained for the empirical analysis. The mean age at baseline was 13.5 years (standard deviation=0.686).

Measures

Self-rated health. Self-rated health is an important indicator of individual health status. In this study, a single item in the student questionnaire, “Which one of the following best describes your general health condition at present?”, was used to measure the self-rated health status of adolescents. This item is a 5-point assessment (ranging from “1=very poor” to “5=very good”), and the higher the score is, the better the adolescent’s self-rated health.

In terms of the Big Five personality traits, higher scores for openness, conscientiousness, agreeableness, and extraversion are usually desirable, while lower scores for neuroticism are better. Based on the Big Five personality model and the CEPS database, the mean values of the relevant items under each indicator were standardized to generate openness, conscientiousness, agreeableness, extraversion, and neuroticism to measure the development of adolescents’ Big Five personality traits.

Openness mainly describes the characteristic tendency of an individual with respect to comprehension, imagination, emotional richness, and creativity. The items “I was able to express myself clearly”, “I was able to give quick responses”, “I was a fast learner” and “I was curious about new stuff” were used to measure openness. Each item was evaluated with 4 points (ranging from “1=strongly disagree” to “4=strongly agree”). The reliability coefficients α of the openness scale in the first and second waves were 0.709 and 0.671, respectively.

Conscientiousness primarily describes an individual’s tendencies toward diligence, self-discipline, prudence, achievement orientation and self-efficacy. The items “I would try my best to go to school even if I was not feeling very well or I had other reasons to stay at home”, “I would try my best to finish even the homework I dislike” and “I would try my best to finish my homework, even if it would take me quite a long time” were used to measure conscientiousness. Each item was rated on a 4-point scale, with response options ranging from “1=strongly disagree” to “4=strongly agree”.

Agreeableness mainly describes individuals’ tendency to have traits such as trust, altruism, cooperation, and empathy. The three items in the questionnaire “I feel close to people in this school”, “Most of my classmates are nice to me” and “My class is in good atmosphere” were used to measure agreeableness. Each item was evaluated with 4 points (ranging from “1= strongly disagree” to “4= strongly agree”). The reliability coefficients α of the agreeability scale in the first and second years were 0.737 and 0.752, respectively.

Extraversion mainly describes an individual’s tendency to be enthusiastic, sociable, active, and optimistic. The item “I often take part in school/class activities” was used to measure extraversion, and it was a 4-point assessment (ranging from “1= strongly disagree” to “4= strongly agree”).

Neuroticism mainly describes emotional traits such as anxiety, depression, and fragility. The items “Do you have the feelings below (feeling blue) in the last seven days?”, “Do you have the feelings below (depressed) in the last seven days?”, “Do you have the feelings below (unhappy) in the last seven days?”, “Do you have the feelings (not enjoying life) below in the last seven days?”, and “Do you have the feelings below (sad) in the last seven days?” were used to measure neuroticism. The options of the questions ranged from “1=never” to “5=always”, all of which were 5-point assessments. The reliability coefficients α of the neuroticism scale in the first and second years were 0.854 and 0.908, respectively.

Data analysis

This study used Stata 17.0 and Mplus 7.4 software for data processing and analysis. First, this study conducted descriptive statistical analysis and bivariate correlation analysis for the variables self-rated health, openness, conscientiousness, agreeableness, extraversion, and neuroticism using Stata 17.0 software. Then, based on the cross-lagged model, we empirically tested the bidirectional relationship between self-rated health and the Big Five personality traits. We explicitly constructed five models using Mplus 7.4 software, and schematic diagrams of models 1 to 3 are shown in Fig. 1. Model 1 is an autoregressive model designed to test the stability of openness, conscientiousness, agreeableness, extraversion, neuroticism, and self-rated health between Time1 and Time2. Model 2 and Model 3 include not only autoregressive paths but also cross-lagged paths from the Big Five personality traits to self-rated health and from self-rated health to the Big Five personality traits. Model 2 is used to elucidate the predictive effects of openness, conscientiousness, agreeableness, extraversion, and neuroticism at Time 1 on self-rated health at Time 2. Model 3 aims to elucidate the predictive effects of self-rated health at Time1 on openness, conscientiousness, agreeableness, extraversion, and neuroticism at Time2. We then constructed Model 4, which integrates all the paths from Model 2 and Model 3, to clarify the bidirectional relationship between self-rated health and the Big Five personality traits. Finally, Model 5 introduces gender as a control variable on the basis of Model 4 to examine the robustness of the previous conclusions. We also report the fit indices for the models, which can be found in the appendix.

Fig. 1: Cross-lagged models between self-rated health and the Big Five personality traits (Model 1 to Model 3).
figure 1

In the specific analysis, we have also taken into account the pairwise correlations between the different variables in each wave.

Full size image

Results

Descriptive statistics of self-rated health and the Big Five personality traits among adolescents

Table 1 shows the mean, standard deviation, and correlation analysis results for self-rated health and the Big Five personality traits. From Time1 to Time2, the mean score for self-rated health decreased from 4.118 to 3.865, indicating an overall decline in Chinese adolescents’ health during this period. In terms of the Big Five personality traits, the mean scores for agreeableness, openness, and conscientiousness all exhibited a decreasing trend, while the mean scores for extraversion and neuroticism showed an increasing trend. Moreover, the results of independent samples t-tests confirm significant differences between the first and second waves for self-rated health (t = 17.567, p = 0.000), extraversion (t = −3.702, p = 0.000), agreeableness (t = 2.702, p = 0.007), openness (t = 21.542, p = 0.000), conscientiousness (t = 17.310, p = 0.000) and neuroticism (t = −13.236, p = 0.000). At the same time, the correlation coefficients reveal that in each wave, self-rated health is significantly positively correlated with extraversion, agreeableness, openness, and conscientiousness (p < 0.05), while self-rated health is significantly negatively correlated with neuroticism (p < 0.05). Self-rated health is consistently highly positively correlated across the years (p < 0.05), and the Big Five personality traits also show a strong positive correlation at both time points (p < 0.05).

Table 1 Descriptive statistics and correlations of self-rated health and the Big Five personality traits.
Full size table

Model results

This study examines the relationships between self-rated health and the Big Five personality traits by constructing cross-lagged models. Table 2 presents the standardized stability and cross-lagged coefficients of the models. The data reveal that all the autoregressive paths are significant (p < 0.05). Specifically, in each autoregressive path within Models 1–4, the standardized stability estimates are as follows: for extraversion, 0.317, 0.320, 0.310, and 0.313; for agreeableness, 0.329, 0.333, 0.319, and 0.323; for neuroticism, 0.377, 0.389, 0.365, and 0.377; for openness, 0.171, 0.176, 0.156, and 0.161; and for conscientiousness, 0.185, 0.187, 0.182, and 0.183; and for self-rated health, 0.362, 0.331, 0.395, and 0.362. In Model 2, we examined the predictive effects of the Big Five personality traits at Time 1 on self-rated health at Time 2. The results reveal that the lagged paths from openness (β = 0.036, p < 0.05) and neuroticism (β = −0.081, p < 0.05) at Time 1 to self-rated health at Time 2 are significant. The lagged paths from extraversion, agreeableness, and conscientiousness to self-rated health are not significant (p > 0.05). In Model 3, we tested the predictive effects of self-rated health at Time 1 on the Big Five personality traits at Time 2. We found that the lagged paths from extraversion (β = 0.088, p < 0.05), agreeableness (β = 0.101, p < 0.05), neuroticism (β = −0.074, p < 0.05), openness (β = 0.128, p < 0.05), and conscientiousness (β = 0.061, p < 0.05) at Time 1 to the Big Five personality traits at Time 2 are significant. Next, Model 4 integrates all the paths from Model 2 and Model 3 to examine the mutual interactions between the Big Five personality traits and self-rated health. These results are consistent with the conclusions drawn from Models 2 and 3. The lagged effects of openness and conscientiousness on self-rated health are significant at 0.034 and −0.079, respectively (p < 0.05). Self-rated health has significant lagged effects on extraversion, agreeableness, neuroticism, openness, and conscientiousness at 0.087, 0.100, −0.069, 0.127, and 0.061, respectively (p < 0.05). In addition, following the relevant literature(Moksnes and Espnes 2020; Wilson et al. 2005; De Bolle et al. 2015; Soto et al. 2011; Iimura and Taku 2018; Kim 2022), we included gender as a control variable in Model 4 to control for its influence on the Big Five personality traits and self-rated health at both Time 1 and Time 2 (Model 5). The results of Model 5 show that our conclusions remain robust (see Fig. 2). Finally, it is worth noting that the models we have constructed show good model fit. The model fit indices are presented in detail in the appendix (see Appendix A: Fit indices of the models).

Table 2 Overview of the standardized stability and cross-lagged coefficients.
Full size table
Fig. 2: The results of Model 5.
figure 2

Model 5 builds on the reciprocal model (Model 4) by including gender as a control variable. Solid lines indicate significance, while dashed lines indicate insignificance in the figure. There is also a correlation between each pair of variables in each wave.

Full size image

Discussion

This study uses a cross-lagged model to clarify the bidirectional relationship between self-rated health and the Big Five personality traits among Chinese adolescents. The findings provide novel evidence in the Chinese context and enrich our understanding of the relationships between self-rated health and the Big Five personality traits. Descriptive statistical analysis reveals a declining trend in self-rated health among Chinese adolescents, which is consistent with previous findings reported in various regions, such as Sweden, Japan, Europe, and North America (Jerdén et al. 2011; Wade and Vingilis 1999; Okada et al. 2022; Cavallo et al. 2015). According to classical psychological viewpoints, adolescence is a period marked by intensified “storms and stresses” in life, which increases the likelihood of experiencing difficulties such as conflicts with parents, emotional disturbances, and risky behaviors (Arnett 1999; Cao and Liu 2023b). In China, the prevalence of conditions such as obesity, poor eyesight, anxiety, and depression among adolescents has increased significantly in recent years (UNICEF 2021; Liu et al. 2023c). The emphasis on exam-oriented education within Confucian culture, coupled with heightened parental expectations, leads to prolonged experience of academic pressure among adolescents (Fu et al. 2022; Cao 2023; Cao and Liu 2023a). This phenomenon may conceivably influence their subjective evaluations of health. Regarding the dimension of the Big Five personality traits, Chinese adolescents demonstrate a decrease in agreeableness, conscientiousness, and openness and an increase in extraversion and neuroticism. Overall, there is a decrease in the maturity of the Big Five personality traits among Chinese adolescents. This finding suggests that the developmental trajectory of personality during adolescence in China is consistent with the disruption hypothesis. In essence, while personality traits tend to mature gradually from adolescence to adulthood, the transitional period of early to mid-adolescence may lead to a temporary decline in the maturity of personality traits due to changes in adolescents’ biological, social, and psychological characteristics (Soto and Tackett 2015). Our findings are consistent with the conclusions drawn by Denissen et al. (2013), Ibáñez et al. (2016)., and Van Den Akker et al. (2021). At the cross-sectional level, self-rated health is significantly positively correlated with openness, conscientiousness, extraversion, and agreeableness, while self-rated health is significantly negatively correlated with neuroticism. This finding is consistent with previous findings from the United States, Europe, and Japan (Stephan et al. 2020). However, there is a discrepancy with the conclusions of Heilmann et al., who found that openness, extraversion, and agreeableness were not associated with self-rated health (Heilmann et al. 2021). It is worth noting that many of the previous studies were based on cross-sectional data, which could reveal only correlations between variables and could not capture the longitudinal relationship between self-rated health and the Big Five personality traits.

To further elucidate the bidirectional predictive relationship between self-rated health and the Big Five personality traits, we used a two-wave cross-lagged analysis. To our knowledge, there is currently a lack of comprehensive research on the role of self-rated health in predicting subsequent personality development. Our cross-lagged path analysis results demonstrate that Chinese adolescents’ self-rated health can predict their subsequent Big Five personality development. Specifically, self-rated health significantly positively predicts extraversion, agreeableness, openness, and conscientiousness. Self-rated health significantly negatively predicts neuroticism. A previous study by Luo et al. (2022a) showed that individuals’ self-rated health levels could predict a decrease in conscientiousness and extraversion levels. Most of the prior literature has focused predominantly on specific facets of the Big Five personality traits, and comprehensive analyses covering openness, conscientiousness, agreeableness, extraversion, and neuroticism have been relatively scarce. The results of this study provide further evidence for the predictive effects of self-rated health on all dimensions of the Big Five personality traits. It is of significance to emphasize that poorer self-rated health conditions may signify unfavorable trends in personality development among adolescents, highlighting the need to increase awareness in educational and family settings.

We also conducted empirical investigations of reverse and bidirectional relationships. Using cross-lagged path analysis, we found that openness positively predicted self-rated health, whereas neuroticism negatively predicted self-rated health. This suggests a bidirectional relationship between self-rated health and neuroticism and openness. In addition, the bidirectional relationship between neuroticism and self-rated health is consistent with previous research suggesting a reciprocal negative predictive relationship between neuroticism and subjective health levels (Mund and Neyer 2016). However, in contrast to the findings of Tauber (2018), who found reciprocal predictive relationships between extraversion and self-rated health, our study found no inverse or reciprocal relationships between self-rated health and extraversion, conscientiousness and agreeableness. This divergence may reflect differences in the demographics of the samples, as their study focused on individuals in middle to late adulthood, whereas our investigation targeted adolescents in the middle school phase. Finally, we also conducted a gender control at Time 1 and found that the above conclusions remained robust.

In conclusion, the findings of this study have empirically unveiled the bidirectional relationship between self-rated health and the Big Five personality traits among Chinese adolescents. This contributes to the literature on health and personality and provides some empirical support for relevant theories. Specifically, our research results partially echo the core principles of life span development theory (Mroczek et al. 2020; Baltes and Baltes 1990), emphasizing the intricate interrelations between personality and bodily functions. Additionally, our findings align with the neo-socioanalytic model of personality development, which emphasizes how an individual’s personality traits shape their unique health experiences, which in turn may influence their personality traits (Luo et al. 2022a). In summary, this study highlights the dynamic interplay between improved health status and personality development in adolescents. The results highlight health indicators as crucial predictors of personality traits and their development, with higher levels of openness and lower levels of neuroticism emerging as protective factors for adolescent health.

Extending the findings of this study, an exploration of the potential mechanisms by which the Big Five personality traits influence adolescents’ self-rated health is particularly important, as it may provide a roadmap for practical application and inspire future research in this area. Healthy lifestyle behaviors and the availability of professional social support are postulated to be key pathways through which the Big Five personality traits facilitate improvements in self-rated health (Kelly et al. 2011). For example, individuals characterized by high levels of conscientiousness may be more likely to adopt healthy habits, such as eating a regular diet and engaging in vigorous physical activity, which may have a positive impact on their self-rated health (Sutin et al. 2016). Conversely, individuals with high levels of neuroticism may be more likely to engage in prolonged sedentary activities, such as excessive television viewing, which can increase the risk of experiencing stress and negative emotions and thus negatively affect the development of self-rated health (Lodewyk 2018). In addition, individuals with high levels of extraversion tend to engage in frequent social interactions, with previous literature suggesting that extraversion can lead to increased subjective well-being through perceptions of increased social support and hope (Tan et al. 2018). An inclination toward openness may encourage individuals to adopt novel health practices or seek professional health advice, potentially enriching their understanding and management of their personal health. These considerations suggest that the Big Five personality traits may plausibly improve self-rated health through the adoption of healthy lifestyles and the receipt of beneficial external support, highlighting the need for future comprehensive research to elucidate these potential mechanisms of influence.

On a practical level, our findings also provide valuable insights for educational administrators and health professionals. They should prioritize improving the health status of adolescents, with particular emphasis on caring for their inner well-being and cultivating positive personality traits. To achieve this goal, targeted interventions can be implemented that take advantage of the bidirectional positive interaction mechanism between health and personality. For example, previous research has shown that active engagement in physical activity contributes to increased openness and reduced negative emotions in adolescents (Cao et al. 2023). It is imperative for both schools and parents to promote healthy lifestyle habits among adolescents while ensuring adequate time for physical activity or other beneficial social engagements. Moreover, the establishment of psychological counseling services and the implementation of mental health curricula are essential for fostering openness and reducing neuroticism, thereby promoting healthy development.

Limitations

First, the data on self-rated health and the Big Five personality variables in this study are derived from questionnaire responses provided by students, which may be subject to certain biases, such as subjectivity and social desirability bias. Additionally, the conclusions about the decline in self-rated health in this study are derived solely from descriptive statistics. However, as relying on descriptive analysis alone does not allow us to infer meaningful conclusions or generalizations, caution should be exercised when extrapolating the findings.

Second, only one item was used to measure the self-rated health of adolescents, which can only be used directly as a manifest variable. Future research could consider developing relevant scales to comprehensively measure adolescent health conditions.

Third, due to limitations of the CEPS database, only two-wave data were used in this study. In future studies, researchers may investigate the bidirectional dynamic relationship between self-rated health and the Big Five personality traits across a wider range of measurement occasions. We are committed to exploring additional longitudinal models as more measurement occasions become available (Hamaker et al. 2015; Lucas 2023; Mund and Nestler 2019).

Fourth, given the lack of information on emotional variables in the database and the prevalence of emotional changes during adolescence that may affect perceived health, future investigations into the influence of emotional factors could be undertaken as relevant research data become available. Similarly, given the potential relevance of sociopolitical factors such as authoritarian governance, income distribution systems and competition for educational resources for mental health in the Chinese context (Xin et al. 2012; Su and Liu 2022; Zhang et al. 2020; Cao and Liu 2022), the current study is limited by data constraints in its empirical exploration of these aspects. Future research should comprehensively examine other potential factors that may influence the health of Chinese adolescents.

Finally, although this study provides new evidence for understanding the bidirectional relationship between self-rated health and the Big Five personality traits, the exploration of this relationship remains preliminary, and a theoretical culmination of this relationship has not yet been achieved. We anticipate the future development of more precise theoretical frameworks with the aim of providing a thorough theoretical interpretation of the dynamic interplay between these two factors.

Conclusions

First, the self-rated health of Chinese adolescents in the middle school stage shows a downward trend. Agreeableness, conscientiousness, and openness decrease, while extraversion and neuroticism increase. We observed significant positive correlations between self-rated health and openness, conscientiousness, extraversion and agreeableness, and significant negative correlations between self-rated health and neuroticism.

Second, the results of the cross-lagged analysis indicate that self-rated health significantly predicts the Big Five personality traits among Chinese adolescents. Specifically, self-rated health positively predicts subsequent extraversion, agreeableness, openness, and conscientiousness but negatively predicts subsequent neuroticism. Furthermore, neuroticism is found to significantly and negatively predict subsequent self-rated health, whereas openness significantly and positively predicts subsequent self-rated health.

Third, there is an urgent need for targeted interventions by educational authorities, schools, and parents to improve adolescents’ well-being and facilitate their personality development. Our findings underscore the importance of increasing overall health as an effective means of promoting the development of their Big Five personality traits. In addition, cultivating openness and reducing levels of neuroticism have been shown to have significant benefits in improving health outcomes. These findings may serve as a promising starting point for future intervention strategies.