A multicenter study of short-term changes in mental health emergency services use during lockdown in Kitchener-Waterloo, Ontario during the COVID-19 pandemic

Abstract

Background: The COVID-19 pandemic and subsequent lockdown measures have led to increasing mental health concerns in the general population. We aimed to assess the short-term impact of the pandemic lockdown on mental health emergency services use in the Kitchener-Waterloo region of Ontario, Canada.

Methods: We conducted an observational study during the 6-month period between March 5 and September 5, 2020 using National Ambulatory Care Reporting System metadata from mental health visits to three regional Emergency Departments (ED); mental health and substance related police calls; and calls to a regional mental health crisis telephone line, comparing volumes during the pandemic lockdown with the same period in 2019. Quasi-Poisson regressions were used to determine significant differences between numbers of each visit or call type during the lockdown period versus the previous year. Significant changes in ED visits, mental health diagnoses, police responses, and calls to the crisis line from March 5 to September 5, 2020 were examined using changepoint analyses.

Results: Involuntary admissions, substance related visits, mood related visits, situational crisis visits, and self-harm related mental health visits to the EDs were significantly reduced during the lockdown period compared to the year before. Psychosis-related and alcohol-related visits were not significantly reduced. Among police calls, suicide attempts were significantly decreased during the period of lockdown, but intoxication, assault, and domestic disputes were not significantly different. Mental health crisis telephone calls were significantly decreased during the lockdown period. There was a significant increase in weekly mental health diagnoses starting in the week of July 12 - July 18. There was a significant increase in crisis calls starting in the week of May 31 - June 6, the same week that many guidelines, such as gathering restrictions, were eased. There was a significant increase in weekly police responses starting in the week of June 14 - June 20.

Conclusions: Contrary to our hypothesis, the decrease in most types of mental health ED visits, mental health and substance-related police calls, and mental health crisis calls largely mirrored the overall decline in emergency services usage during the lockdown period. This finding is unexpected in the context of increased attention to acutely deteriorating mental health during the COVID-19 pandemic.

Keywords: COVID-19; Lockdown; Pandemic; Psychiatric emergency.

Fig. 1

Daily total Emergency Department volumes in three hospitals in Kitchener-Waterloo and daily new COVID-19 cases in the community from March 5 to September 5, 2019–2020. Trends were smoothed using the LOESS method (see methods section), with a 95% confidence interval for predictions shaded in gray. The shaded areas indicate the periods of lockdown (red) and diminishing restrictions (green)

Fig. 2

Year-over-year change between 2019 and 2020 in weekly mental health diagnoses in three Emergency Departments in the Kitchener-Waterloo region: alcohol related, involuntary Form 1, mood related, psychosis related, self-harm related, situational crisis, and substance related. Total weekly mental health diagnoses were also compared. Shaded areas indicate periods of lockdown (red) and diminishing restrictions (green)

Fig. 3

Year-over-year change in weekly Waterloo Regional Police responses between 2019 and 2020 for assault, domestic dispute, intoxication, and suicide attempts. Total police responses and calls to the crisis line were also compared. The shaded areas indicate the periods of lockdown (red) and diminishing restrictions (green)

Fig. 4

Changepoints in daily ED visits, total weekly mental health diagnoses, total weekly crisis calls, and total weekly police responses from March 5 to September 5, 2020. Changepoints indicate significant changes in mean and variance, as calculated by the PELT method. Changepoints are denoted by dashed vertical lines and segment means are denoted by horizontal lines. The shaded areas indicate the periods of lockdown (red) and diminishing restrictions (green)