To estimate current prevalence levels of stress, and to identify related characteristics among urban residents in China.

A cross-sectional, multilevel study. Selected through multi-stage quota-sampling, survey participants were 4,735 urban residents aged 15 years and older who resided in one of six selected Chinese capital cities. Data were collected on stress levels and sociodemographic characteristics. Stress was assessed using the Perceived Stress Scale, Chinese version (CPSS). A multilevel variance component model was employed to analyze associations between sociodemographic variables and stress.

The mean stress score for urban residents was 22.34 (SD: 3.22), and 36.8% of those surveyed (95% CI: 33.5–40.2%) were severely stressed (>25). Multilevel regression analysis indicated that residents aged 55 years and older were less stressed than residents under age 25. The most educated and higher income earners had lower stress levels than the least educated and poorest. High levels of stress were apparent among all other occupational groups, relative to managers and clerks, except retirees and operational workers. Residents in the north of China exhibited higher stress levels than counterparts in the south.

This study suggests that higher stress levels are positively associated with social class in China. Our findings could inform health policy, guide prevention strategies, and justify the design and implementation of targeted interventions.