COVID-19 Pandemic Short Interval National Survey Gauging Psychological Distress

Background It has become ubiquitous to describe the severe acute respiratory syndrome
coronavirus 2 (SARS-CoV-2) pandemic and the global measures to decrease mortality rates of
Coronavirus Disease (COVID-19) as 'unprecedented'. The Public Health Agency of Canada (PHAC)
models released on April 8th predict that even with strict epidemiologic controls
934,000-1,879,000 people will be infected in Canada and 11,000- 22,000 will die.
Disturbingly, the European and United States experience demonstrates that health care systems
in industrialized countries are at risk of becoming overwhelmed. It has long been recognized
that physicians are at risk for burnout, which can negatively affect their ability to care
for patients, their colleagues and themselves. Infectious disease pandemics present many of
the same challenges for physicians as natural disasters and mass casualty events, with the
addition of a prolonged time frame, potential need for isolation away from family, and
concern of becoming infected or infecting loved ones. During the 2003 severe acute
respiratory syndrome (SARS) outbreak, 18-57% of health care workers (HCWs) experienced
measurable psychological distress. Even 1-2 years following the SARS pandemic HCWs in Ontario
experienced high levels of burnout (19-30%), psychological distress (30-45%), and PTSD
symptoms (8-13%). Despite the overall high rates, there are several mitigating factors that
may decrease the rate or severity of psychological distress and burnout.

Given that these studies generally measure burnout and psychological distress at one or two
cross-sectional time points, it is difficult to tell how burnout develops over time.
Additionally, the bulk of the studies assessed HCWs following natural disasters, which tend
to present as a single time point with one large wave of patients. The current crisis is
distinct in its sustained level of demand on HCWs with potential for multiple waves of
patients or a sustained high number of patients outstripping the health care system resources
for months at a time. This study seeks to investigate the rates of burnout assessed at
multiple time points during the COVID-19 pandemic. Further, this study seeks to determine the
relationship between patient numbers, mitigating factors, and exacerbating factors and the
rates of burnout at various time points.

Objectives

1. What are the rates of psychological distress and burnout among physicians and residents
throughout the COVID-19 pandemic?

2. What baseline factors are associated with reductions and increases in psychological
distress and burnout over time (trajectories)?

Methods This study will be a national longitudinal survey of physicians in Canada. The
sampling frame will be all physicians currently practicing in Canada, whether they hold a
full, provisional, or post-graduate in-training license. The survey will be distributed by:
1. Advertisements on Twitter, 2. Provincial physician associations, and 3. National specialty
groups for higher risk specialties (anesthesia, critical care, emergency medicine).
Recipients will be directed to the survey site hosted by Redcap. Cross-sectional results of
the initial survey to give immediate feedback to provincial and national organizations about
the threat of physician burnout and psychological distress. Longitudinal results will analyze
how burnout and distress change over time and will analyze the relationship to mitigating and
exacerbating factors.

Consenting participants will complete an initial survey gathering demographic information
(age, gender, trainee, years in practice, specialty, chronic health conditions, practice type
(academic, community, rural), preparatory information (prior to first COVID-19 suspected
case: COVID-19 pathways/procedures, simulation, designated areas, scheduling changes), and
baseline measures (Maslach Burnout Inventory - General Survey (MBI-GS) short form, Pandemic
Experience & Perceptions Scale (PEPS), Hospital Anxiety and Depression Scale (HADS),
stressors, resources, coping strategies). Participants will then receive follow up
questionnaires every month until there is a sustained period of no new cases in their
province (1 month without a new case) up to a maximum of one-year involvement. These follow
up questionnaires will measure burnout (MBI-GS), psychological distress (HADS), pandemic
specific stressors (PEPS), workplace stressors, and resources/coping strategies. A final
follow-up survey will be sought one year after the end of the short- interval surveys to
assess the long-term effects of responding to the COVID pandemic.

Statistical Analysis Objective 1. Variables will be described using mean and standard
deviation when normally distributed or median and interquartile range when skewed as measured
by the Shapiro-Wilk test.

Objective 2. The cross-sectional results of the initial survey will be analyzed using linear
regression to quantify the strength of relationship between factors and burnout. In the
longitudinal analysis, trajectory analysis using latent growth mixed modelling (LGMM) will be
used to identify trajectories of burnout and psychological distress amongst physicians over
time. LGMM is ideally suited to longitudinal data as it is robust to the effects of missing
data. Furthermore, LGMM may prove very useful in an unprecedented situation such as this in
which there is scant data on which to base hypotheses as LGMM allows trajectories to be
identified more empirically based on the data rather than through a priori conceptual models.

Anticipated Outcomes The investigators will discover whether the characteristics of burnout
during this pandemic are comparable to the current body of knowledge about physician burnout
at baseline or during more limited periods of crisis. This will inform whether current
interventions for physician burnout are appropriate during widespread pandemic events, both
as physicians continue to deal with COVID-19 and in the future. Improved understanding of
stressors and protective factors will be essential not only for planning therapeutic studies
at an individual level but will also allow institutions to mobilize high yield supports for
physicians at risk.