Kazakhstan’s First National Youth Mental Health Survey Reveals Stark Gender and Ethnic Divides

Kazakhstan has, for the first time, a nationally representative picture of the mental health of its young people, and the results carry a message that is both reassuring and deeply troubling. A new population-based study published in BMC Psychology analyzed data from 5,079 Kazakhstani youth aged 15 to 24 years, drawn from the 2024 Kazakhstan Multiple Indicator Cluster Survey, a large household survey implemented by the country’s Bureau of National Statistics with technical and financial support from UNICEF. The headline numbers are comparatively low: weighted national prevalence stood at 1.70 percent for potential depression, 1.59 percent for potential anxiety, and 0.96 percent for potential suicidality. Yet beneath those modest averages lies a pattern of stark inequality. Young women carried nearly three times the risk of depression and anxiety compared with young men, and young people of non-Kazakh ethnicity faced almost double the risk of both conditions, disparities large enough to dominate the study’s statistical models.

The significance of the study lies less in its absolute figures than in the fact that they exist at all. Mental health research across Central Asia has been critically under-prioritized, according to a recent scoping review cited by the authors, and empirical data on youth aged 15 to 24 in the region have remained sparse. This gap is particularly concerning given the epidemiological context. Globally, roughly 300 million people aged 5 to 24 live with a mental health disorder, contributing nearly 30 million years lived with disability, and suicide remains a leading cause of death among those aged 15 to 29, with three quarters of these deaths occurring in low- and middle-income countries. In Eastern Europe and Central Asia, suicide is the foremost cause of mortality among young people. UNICEF has previously reported that the risk of suicidal behavior among Kazakhstani adolescents aged 15 to 19 is three times higher than the average across the Commonwealth of Independent States, making the absence of rigorous national data all the more striking.

The methodological machinery behind the new estimates is considerable. The 2024 MICS employed a two-stage, stratified cluster sampling design that generated representative estimates at national, regional, and urban-rural levels across all 17 regions and three major municipalities of Kazakhstan. In the first stage, census enumeration areas were systematically selected with probability proportional to size; in the second, households were systematically sampled within each area. The survey achieved a household response rate of 96.3 percent, with response rates of 97.4 percent among women and 94.4 percent among men aged 15 to 24. After excluding participants with more than half of the screening items missing, the analytical sample comprised 5,079 young people. Missing data were minimal, below 1 percent for all three outcomes, and were handled through multiple imputation by chained equations with five imputations. All analyses incorporated survey weights and clustering by enumeration district to preserve national representativeness.

Depression and anxiety were screened using two ultra-brief instruments: the two-item Patient Health Questionnaire, known as PHQ-2, and the two-item Generalized Anxiety Disorder Scale, GAD-2. The PHQ-2 asks about the frequency of depressed mood, defined as sadness and hopelessness, and anhedonia, the loss of pleasure in previously enjoyable activities, over the preceding two weeks. The GAD-2 probes feelings of nervousness, anxiety, or being on edge, together with the inability to stop or control worrying. Each item is rated on a four-point scale from zero to three, producing total scores from zero to six, with a cut-off of three or higher flagging potential depression or anxiety. Suicidality was assessed as a binary self-reported measure, classifying participants as positive if they had tried to harm themselves with intent to end their life in the past 12 months or had thought about doing so in the past two weeks. While formal validation of these instruments in Kazakhstan is limited, systematic reviews support the reliability of ultra-brief screeners across diverse cultural settings.

To identify independent risk factors, the researchers applied modified Poisson regression models, an approach chosen because it directly estimates risk ratios, which are more interpretable than odds ratios and align with the calculation of population attributable fractions, a statistic that quantifies how much of a condition’s burden in the population could theoretically be eliminated if a risk factor were removed. The results were unambiguous. In fully adjusted models, female sex was the strongest predictor of both potential depression, with an adjusted risk ratio of 2.72, and potential anxiety, with an adjusted risk ratio of 2.80. Non-Kazakh ethnicity remained significantly associated with both outcomes, at 1.88 for depression and 1.99 for anxiety. The population attributable fractions were striking: female sex alone accounted for 45.6 percent of the population risk of depression and 46.8 percent of the risk of anxiety, while non-Kazakh ethnicity contributed 19.5 percent of depression risk.

Anxiety showed a further wrinkle that sets it apart from depression: a U-shaped relationship with household wealth. Both lower wealth, with an adjusted risk ratio of 1.89, and higher wealth, at 2.20, were associated with elevated anxiety risk, and the corresponding attributable fractions were 33.0 percent for low wealth and 21.7 percent for high wealth. The authors suggest that the mechanisms at the two ends of the wealth spectrum may differ. At the lower end, financial hardship and material deprivation act as persistent stressors that erode mental health, a relationship well documented in longitudinal research. At the upper end, evidence indicates that parents with higher socioeconomic status tend to hold higher expectations for their children’s achievement, and that pressure to perform can breed worry about failure and fear of disappointing parents, feeding internalizing problems. In other words, scarcity and pressure may converge on the same anxious outcome through very different routes.

Stratified analyses reinforced how pervasive the gender gap is. Among 15 to 19 year olds, females reported significantly higher prevalence of potential depression, 1.8 percent versus 0.8 percent in males, and anxiety, 1.9 percent versus 0.6 percent. The disparity widened in the 20 to 24 age group, where depression reached 3.5 percent in females against 1.0 percent in males, and anxiety 3.0 percent versus 1.2 percent. The same pattern held in both rural and urban settings, suggesting that gender-specific vulnerabilities operate independently of residential context. The authors frame this within the internalizing-externalizing framework of psychopathology, which conceptualizes depression and anxiety as manifestations of a broader internalizing dimension to which females show greater liability, shaped by pubertal hormonal changes, differential exposure to stressors, adverse childhood experiences, gender inequity, and higher exposure to maltreatment, including childhood sexual abuse.

Geography mattered too. Regional prevalence varied significantly for depression and anxiety, though not for suicidality. North Kazakhstan recorded the highest prevalence of potential depression at 5.44 percent, while West Kazakhstan led on anxiety at 4.51 percent and also showed the highest, though statistically non-significant, suicidality figure of 2.35 percent. The authors point to regional heterogeneity in economic development, service access, and culture as plausible explanations: North Kazakhstan has experienced lower economic development and higher unemployment, potentially exposing residents to chronic stressors that manifest as depression, whereas West Kazakhstan is characterized by industrial activity, migration fluctuations, and relative social isolation, conditions that may heighten exposure to acute stress and uncertainty. Cultural factors, including stigma, help-seeking norms, and religiosity, may further shape which symptoms are experienced and reported in surveys.

Perhaps the most enigmatic finding concerns suicidality. Despite the region’s alarming background rates of youth suicide, no sociodemographic factor, not sex, age, residence, education, marital status, wealth, or ethnicity, was significantly associated with potential suicidality in either unadjusted or adjusted analyses. The authors caution that the low prevalence of 0.96 percent may have limited statistical power to detect associations, and they stress that suicidal behavior may not simply be an extension of depression or anxiety but may involve complex, multidimensional interactions among social, psychological, and life-event factors. They also acknowledge the study’s limits: screening tools identify potential symptoms rather than confirmed clinical diagnoses, self-report in a stigma-laden context may drive underreporting, the cross-sectional design precludes causal inference, and the analysis was restricted to sociodemographic variables available in the survey, leaving room for residual confounding. Even so, the study delivers the first national benchmarks for youth mental health in Kazakhstan, and its implications are clear. Prevalence rates lower than global averages should not obscure the concentration of risk among young women and ethnic minority youth. Meeting Sustainable Development Goal Target 3.4, which mandates the promotion of mental health and well-being across the lifespan, the authors argue, will require culturally sensitive, targeted interventions and strengthened early detection for precisely these vulnerable subgroups.

Subject of Research: National prevalence and sociodemographic correlates of depression, anxiety, and suicidality among young people in Kazakhstan

Article Title: National prevalence and sociodemographic correlates of potential depression, anxiety, and suicidality among the young population in Kazakhstan: a national population-based study

Article References: Li, X., Stickley, A., Cai, Y., Gai, Y., Bjørklund, G., Li, Y., Jiang, F., & Chen, S. (2026). National prevalence and sociodemographic correlates of potential depression, anxiety, and suicidality among the young population in Kazakhstan: a national population-based study. BMC Psychology, 14(1), Article 1421. https://doi.org/10.1186/s40359-026-05618-8

Image Credits: AI Generated

DOI: 10.1186/s40359-026-05618-8

Keywords: Kazakhstan, youth mental health, depression, anxiety, suicidality, PHQ-2, GAD-2, MICS survey, gender disparities, ethnicity, population attributable fraction, Central Asia