Changes in suicide rates across COVID-19 phases by sex and age group: a population-based time-series analysis

Background

Early population-based studies suggested that suicide rates remained largely stable during the initial phases of the covid-19 pandemic, despite concerns about its impact on mental health [1]. However, aggregate findings may obscure heterogeneous patterns across demographic groups and time periods, and evidence on age- and sex-specific changes is mixed, particularly across different pandemic phases. Social isolation, increased unemployment, and disruption to services may have differential effects, but subgroup-specific temporal patterns remain unclear.

Aims

To examine changes in suicide rates across multiple covid-19 phases and to assess whether these changes differed by sex and age group.

Methods

Monthly suicide counts for the years 2019-2022 were obtained from the Israeli Central Bureau of Statistics and supplemented by Ministry of Health updates. Annual population estimates were used to calculate monthly rates per 100,000. The year 2019 served as a pre-pandemic baseline. Pandemic phases were defined to capture periods before, during, and after three national lockdowns. General linear models were used to analyze monthly suicide rates, including phase, sex, age group (25-44, 45-64, and 65+), and calendar month to adjust for seasonality, with time as a continuous covariate. Interaction terms were included to assess whether temporal changes differed across demographic groups, with a particular focus on age- and sex-specific patterns over time.

Results

The overall model was significant (adjusted R² = .65; p < .001). As expected, suicide rates differed significantly by sex and age group (both p < .001), with higher rates observed in males and in older age groups. No overall effect of covid-19 phase was observed (p = .552), and no immediate change was detected during the first lockdown (p = .489). A significant three-way interaction between phase, sex, and age group was identified (p < .001), indicating group differences in temporal patterns. Estimated marginal means showed a marked increase in suicide rates among males aged 65+ during the period between the first and second lockdowns, rising from 1.09 (95% CI 0.89-1.29) pre-pandemic to 2.10 (95% CI 1.81-2.39) post-lockdown. No comparable increase was observed during the lockdown itself (0.87 [95% CI 0.57–1.16]) or across later phases. Other groups demonstrated minimal or inconsistent changes across time, with no clear patterns of increase or decrease observed.

Conclusions

No overall increase in suicide rates was observed across covid-19 phases. Instead, changes were time-limited and specific to subgroups, with a pronounced increase observed among older males during the period between the first and second lockdowns that was not observed in later phases. These findings suggest that early pandemic conditions may have had differential effects across demographic groups and highlight the importance of targeted monitoring and intervention during and immediately after large-scale public health emergencies. These findings should be interpreted cautiously in light of small numbers and potential underreporting of suicide deaths.

References

1. Pirkis J, John A, Shin S et al. Suicide trends in the early months of the COVID-19 pandemic: an interrupted time-series analysis of preliminary data from 21 countries. The Lancet Psychiatry, 2021; 8, 579-588.