Divergent trends in attempted suicide and suicide deaths after the COVID-19 shock in Macao SAR, China: an interrupted time-series analysis
Year: 2026
Type: article
Study design: Observational Study
Abstract: Background The impact of the coronavirus disease 2019 (COVID-19) pandemic on suicidal behavior has diverged between fatal and non-fatal outcomes, yet parallel dual-outcome observation within a single population and long-window evidence are limited. Macao SAR, China—a micro-economy dependent on gaming—experienced the closure of casinos, cross-border travel restrictions, and a sharp decline in gross gaming revenue (GGR) in February 2020, an a priori composite breakpoint. Its police-recorded monthly suicide data have been published by age group and sex since 2015. We examined changes in attempted suicide and suicide death around this breakpoint and their coupling with income fluctuations. Methods Using monthly counts (January 2015–June 2025; 126 months) verified cell-by-cell against official data, we estimated breakpoint associations with an interrupted time-series analysis (segmented Poisson regression) and calculated net excess under dual counterfactuals (trend extrapolation; pre-pandemic seasonal level). Fourteen subgroup models underwent Benjamini–Hochberg correction; age heterogeneity was tested with an age-by-breakpoint interaction model; and coupling with GGR fluctuations was examined using distributed-lag models and cross-correlation functions. Results Attempted suicide rose immediately after the breakpoint [level rate ratio (RR) 1.47, q = 0.022], with a period-mean RR of 1.93 (95% CI 1.25–3.07) and net excess of approximately 556 cases (95% CI 240–784), persisting 5.4 years without decline. The increase concentrated among people under 25 (level RR 2.31 for ages 15–24, q = 0.003; joint age-interaction χ 2 = 78.0, 6 df, p < 0.00001). Suicide death showed no corresponding increase (level RR 0.80, q = 0.176; net excess CI crossing zero); those aged 65 + showed a marginal increase concentrated in the early pandemic period (RR = 1.81, q = 0.021), sensitive to the covariance specification and partly reflecting pre-existing drift. The rise did not track monthly GGR fluctuations, a level-shift pattern that cannot exclude effects of the structural income decline collinear with the breakpoint. Conclusion After the COVID-19 shock, Macao experienced a sustained non-fatal suicide crisis while suicide death showed no detectable net increase, with young people bearing most of the increase. Expanding youth mental health services and maintaining high-frequency, stratified surveillance are public health priorities.
Source: Frontiers in Public Health
Authors: Jingwen Cai(Formal analysis, Project administration, Validation, Writing – original draft, Writing – review & editing), Wai In Ho(Data curation, Writing – original draft, Writing – review & editing), Yifen Yin(Conceptualization, Supervision, Writing – review & editing), Yuanyuan Yu(Conceptualization, Methodology, Writing – review & editing), Chunning Wang(Methodology, Software, Visualization, Writing – original draft)
Institution: Macao Polytechnic University
Language: English
Subfield: Clinical Psychology
Field: Psychology
Domain: Social Sciences
Open Access status: gold
Estimated APC paid: $3,801
OpenAlex ID: W7235757200
