One of the largest adolescent mental health studies ever conducted has revealed that the combination of family structure and clustered health-risk behaviors can dramatically multiply a teenager’s odds of depression and anxiety—by more than twenty-fold in the most vulnerable group. The research, drawing on responses from nearly 300,000 students across every district and county of Shandong Province, China, offers the most detailed picture yet of how broken family environments and risky behaviors interact to shape adolescent mental health, and it suggests that clinicians and schools may need to look at both factors together rather than in isolation.
The study, published in BMC Public Health by a team of researchers from Shandong University, the Shandong Center for Disease Control and Prevention, and collaborating institutions, surveyed 292,536 students enrolled in junior, senior, and vocational high schools across all 136 districts and counties of Shandong Province between 2023 and 2024. Participants completed self-administered questionnaires covering sociodemographic characteristics, ten categories of health-risk behaviors, and five types of non-intact family structures, including single-mother, single-father, stepfather, stepmother, and grandparent-headed households. Depressive symptoms were measured using the Center for Epidemiologic Studies Depression scale, a twenty-item instrument widely used in population research, while anxiety was assessed with the seven-item Generalized Anxiety Disorder scale, a validated clinical screening tool.
To make sense of the behavioral data, the researchers turned to latent class analysis, a statistical technique that identifies hidden subgroups within a population based on patterns of responses across multiple observed variables. Rather than treating each risky behavior—such as smoking, drinking, poor diet, or physical inactivity—as a separate problem, this approach recognizes that behaviors tend to cluster together in predictable ways. The analysis revealed four distinct classes of health-risk behavior among the adolescents, with the most prevalent being an Unhealthy Diet class that encompassed just over forty percent of all students, at 40.13 percent.
The most alarming class, however, was the one the researchers labeled Violent Victimization Experience and Addictive Behavior, abbreviated VVAB. Students in this group reported experiences of being victims of violence alongside addictive behaviors such as tobacco, alcohol, or other substance use. While the study did not report the exact prevalence of this class, its association with mental health outcomes proved to be by far the strongest of any behavioral pattern identified.
When the researchers examined the independent association between family structure and behavioral risk, striking differences emerged among the different types of non-intact families. Adolescents living in stepmother families showed the highest odds of belonging to the VVAB class, with an odds ratio of 3.21 and a ninety-five percent confidence interval ranging from 2.39 to 4.31, meaning these students were more than three times as likely as those from intact families to fall into the violent victimization and addiction cluster. Stepmother families were also linked to elevated odds of physical inactivity, with an odds ratio of 1.89. In contrast, adolescents in stepfather families showed the highest odds of unhealthy diet patterns, with an odds ratio of 2.40, while only single-mother families showed a modest elevation in odds of overweight status, with an odds ratio of 1.09.
The associations between family structure and mental health outcomes followed a similar pattern of differentiation. Adolescents from stepmother families had the highest odds of anxiety symptoms, at an odds ratio of 1.43, while those from stepfather families showed the highest odds of depressive symptoms, at 1.73, and of comorbid depression and anxiety, at 1.75. These findings challenge a common assumption in the literature that all non-intact family types carry uniform risk. Instead, the data suggest that the specific configuration of a broken family matters, with stepparent households—whether stepmother or stepfather—appearing to confer greater risk than single-parent or grandparent-headed arrangements in this population.
The behavioral classes themselves proved to be powerful predictors of mental health, independent of family structure. Adolescents in the VVAB class had an odds ratio of 9.53 for depressive symptoms, 6.04 for anxiety symptoms, and 10.18 for comorbid symptoms, each with tight confidence intervals reflecting the enormous sample size. These magnitudes dwarf the associations attributable to family structure alone, underscoring that the clustering of violence exposure and addictive behavior represents a profound risk state for adolescent psychological wellbeing. The Unhealthy Diet class, by contrast, was the most common but carried far weaker associations with depression and anxiety, suggesting that not all behavioral clusters are equally threatening to mental health.
The most consequential finding of the study, however, emerged when the researchers combined the two dimensions and examined the joint associations of family type and behavioral class. Among adolescents who belonged to the VVAB class and lived in stepfather families, the odds of depressive symptoms reached 21.10, with a confidence interval of 17.32 to 25.60. The odds of anxiety symptoms in this same group were 9.94, and the odds of comorbid depression and anxiety climbed to 22.70, with a confidence interval of 18.50 to 27.80. In practical terms, a teenager exposed to violence and struggling with addictive behaviors while living in a stepfather household faced more than a twenty-fold increase in the likelihood of experiencing both depression and anxiety simultaneously compared with peers from intact families without those behavioral risks.
This multiplicative pattern carries important implications for how scientists understand the architecture of adolescent mental health risk. Neither family disruption nor behavioral clustering alone tells the full story. The data indicate that these two factors operate jointly, such that the presence of both creates a risk profile far exceeding the sum of its parts. This kind of statistical interaction is consistent with theoretical models in which family instability erodes protective factors—such as parental monitoring, emotional support, and household routine—that might otherwise buffer adolescents from the psychological consequences of violence exposure and substance use. When those buffers are absent, the mental health consequences of risky behaviors appear to escalate steeply.
The scale of the study lends considerable weight to its findings. With nearly 300,000 participants drawn from every administrative subdivision of a province home to more than 100 million people, the survey captures extraordinary geographic and socioeconomic diversity, from dense urban centers to rural agricultural counties. The inclusion of vocational high school students, a population often underrepresented in adolescent health research, further broadens the picture. Tight confidence intervals across virtually all reported estimates reflect the statistical power afforded by such a massive sample, making it unlikely that the observed associations are artifacts of random variation.
The researchers acknowledge that the cross-sectional design of the study limits causal inference. Because family structure, behavioral patterns, and mental health symptoms were all measured at a single point in time, it is impossible to determine from these data alone whether non-intact family structures cause risky behaviors, whether pre-existing mental health difficulties contribute to both behavioral problems and family breakdown, or whether unmeasured factors—such as household income, parental conflict preceding separation, or genetic vulnerability—drive the observed associations. Longitudinal studies that follow adolescents over time will be needed to disentangle these pathways. Self-reported questionnaires also carry the possibility of recall bias and underreporting, particularly for sensitive topics such as violence victimization and substance use, although the anonymity of school-based administration typically reduces such distortion.
Nevertheless, the findings arrive at a moment of rising concern about adolescent mental health worldwide, and at a time when divorce rates in China and many other countries continue to climb, steadily increasing the proportion of young people growing up in non-intact households. Global data have documented rising prevalence of depression and anxiety among teenagers across high-, middle-, and low-income countries alike, prompting public health agencies to search for identifiable and modifiable risk factors that can guide prevention.
The authors suggest that adolescent health and mental health services should consider strengthening family relationships, addressing clustered risky behaviors, and supporting mental wellbeing in an integrated fashion, rather than treating these domains separately. For schools, the results point to the value of screening students from stepparent households for both behavioral risks and psychological symptoms. For clinicians, the study argues that a teenager presenting with depression or anxiety should be assessed not only for individual behaviors but for the broader pattern of violence exposure, substance use, diet, and activity—and for the family context in which those behaviors are embedded. For policymakers, the message is that support programs aimed at non-intact families, and stepparent families in particular, may yield disproportionate returns for the mental health of the next generation.
What the Shandong study makes unmistakably clear is that the most endangered adolescents are not simply those with a broken home or a risky habit, but those carrying both burdens at once—a convergence that transforms statistical odds into an urgent call for integrated care.
Subject of Research: Joint associations of non-intact family types and clustered health-risk behaviors with depressive, anxiety, and comorbid mental health symptoms among adolescents in Shandong Province, China
Subject of Research: Medicine
Article Title: Joint associations of non-intact family types and health-risk behaviors with mental health outcomes: evidence from a large-scale survey among adolescents in Shandong Province, China
Article References: Wang, K., Yu, M., Lv, C., Wei, Z., Qi, Y., Sun, L., Qin, A., Yu, L., Liu, Z., Zheng, S., Li, S., & Xu, M. (2026). Joint associations of non-intact family types and health-risk behaviors with mental health outcomes: evidence from a large-scale survey among adolescents in Shandong Province, China. BMC Public Health. https://doi.org/10.1186/s12889-026-29171-6
Image Credits: AI Generated
DOI: 10.1186/s12889-026-29171-6
Keywords: health-risk behavior, non-intact family types, depressive symptoms, anxiety symptoms, comorbid symptoms, adolescents, latent class analysis, stepfamilies, mental health, China
Cite Scienmag News
APA MLA Chicago
Glenn Wilkins. (September 6, 2026). Family structure and health-risk behaviors jointly shape adolescent mental health in China. Scienmag. https://scienmag.com/family-structure-and-health-risk-behaviors-jointly-shape-adolescent-mental-health-in-china/
Glenn Wilkins. “Family structure and health-risk behaviors jointly shape adolescent mental health in China.” Scienmag, 6 September 2026, https://scienmag.com/family-structure-and-health-risk-behaviors-jointly-shape-adolescent-mental-health-in-china/. Accessed 6 September 2026.
Glenn Wilkins. “Family structure and health-risk behaviors jointly shape adolescent mental health in China.” Scienmag. September 6, 2026. https://scienmag.com/family-structure-and-health-risk-behaviors-jointly-shape-adolescent-mental-health-in-china/
Copy citation Download RIS
Tags: Adolescent Mental Healthadolescent mental health in Chinaadolescent mental health risk factors in Chinaclustering of health-risk behaviorscomprehensive assessment of adolescent well-beingdepression and anxiety in teenagerseffects of family environment on mental healthfamily dynamics and mental healthfamily structurehealth-risk behaviorsimpact of broken family environmentslarge-scale epidemiological studylarge-scale youth mental health studyrisk factors for adolescent depressionrole of schools and clinicians in mental health preventionschool-based mental health interventionssocio-demographic influences on youth mental healthsociodemographic influences on adolescent mental health
