SummaryBackgroundRising rates of suicide, the most dreadful consequence of mental health effects elicited by the coronavirus pandemic (COVID-19) are cause for grave concern. However, the exact association between mental health problems and suicide remains largely unknown in relation to COVID-19.MethodsTo determine the impact of COVID-19 on suicide trajectory, we used an interrupted time-series design to analyze monthly suicides rates extracted from Japan’s national database. We next used mixed-effects regression models to investigate the relationship between the nationwide suicide increase in August 2020 and psychiatric states of 4,348 individuals from an online survey performed immediately before (December 2019) and during (August 2020) the pandemic. Psychiatric states included depression, anxiety, and COVID-19-related PTSD, a form of severe event-related stress.FindingsIn Japan, suicides had gradually decreased before COVID-19 (β = −0·7×10−3, t57 = −14·2, p = 8·6×10−46), but increased drastically after a state of emergency was declared in April 2020 (β = 0·9×10−2, t57 = 17·3, p = 2·3×10−67). We found that PTSD symptoms reliably predict COVID-19’s impact on suicide rates (β = 6·3×10−4, t3936 = 5·96, p = 2·7×10−9). In contrast, depression scores are a reliable indicator of stress vulnerability (i.e. future suicide increases, β = 0·001, t3936 = 6·6, p = 4·5×10−11). Simulations revealed that a one-point reduction in PTSD score could decrease suicides by up to 3·1 per ten million people per month in Japan.InterpretationPTSD symptoms may help to identify high-risk groups so as to increase efficacy of prevention policies.FundingKDDI collaborative research contract, the Innovative Science and Technology Initiative for Security (JPJ004596), ATLA and AMED (JP20dm0307008).Research in contextEvidence before this studyWe searched PubMed on December 2, 2020, for “COVID” and “suicid*” in the titles or abstracts of published articles and obtained 269 hits. No language restrictions were applied to the search. Nearly all previous articles on suicide and COVID-19 have reported simulation studies of suicide counts and rates in case studies, editorials, letters, and commentaries. To date, no study has analyzed the association between psychiatric states and suicide increases in the context of the COVID-19 pandemic.Added value of this studyTo the best of our knowledge, this is the first study reporting a concrete approach to predict suicide rate increases from psychiatric states during the COVID-19 pandemic. Our findings indicate that PTSD symptoms are a reliable surrogate endpoint of pandemic-related suicide increase.Implications of all available evidenceThis work provides a new perspective on preparing guidelines for suicide prevention. Efforts should focus on reducing PTSD severity for single individuals and populations to reduce the overall suicide risk.