Governments implemented lockdowns and other physical distancing measures to stop the spread of SARS-CoV-2 (COVID-19). Resulting unemployment, income loss, poverty, and social isolation, coupled with daily reports of dire news about the COVID-19 pandemic, could serve as catalysts for increased self-harm deaths (SHD). This ecological study examined whether observed SHD counts were higher than predicted SHD counts during the pandemic period in the Canadian provinces of Alberta, British Columbia, Ontario, and Quebec. The study also explored whether SHD counts during the pandemic were affected by lockdown severity (measured using the lockdown stringency index [LSI]) and COVID-19 case numbers. We utilized publicly available SHD data from January 2018 through November 2020, and employed AutoRegressive Integrated Moving Average (ARIMA) modelling, to predict SHD during the COVID-19 period (March 21 to November 28, 2020). We used Poisson and negative binomial regression to assess ecological associations between the LSI and COVID-19 case numbers, controlling for seasonality, and SHD counts during the COVID-19 period. On average, observed SHD counts were lower than predicted counts during this period (p < 0.05 [except Alberta]). Additionally, LSI and COVID-19 case numbers were not statistically significantly associated with SHD counts.

Governments implemented lockdowns and other physical distancing measures to stop the spread of SARS-CoV-2 (COVID-19). Resulting unemployment, income loss, poverty, and social isolation, coupled with daily reports of dire news about the COVID-19 pandemic, could serve as catalysts for increased self-harm deaths (SHD). This ecological study examined whether observed SHD counts were higher than predicted SHD counts during the pandemic period in the Canadian provinces of Alberta, British Columbia, Ontario, and Québec. The study also explored whether SHD counts during the pandemic were affected by lockdown severity (measured using the lockdown stringency index [LSI]) and COVID-19 case numbers. We utilized publicly available SHD data from January 2018 through November 2020, and employed AutoRegressive Integrated Moving Average (ARIMA) modelling, to predict SHD during the COVID-19 period (March 21 to November 28, 2020). We used Poisson and negative binomial regression to assess ecological associations between the LSI and COVID-19 case numbers, controlling for seasonality, and SHD counts during the COVID-19 period. On average, observed SHD counts were lower than predicted counts during this period ( p < 0.05 [except Alberta]). Additionally, LSI and COVID-19 case numbers were not statistically significantly associated with SHD counts.