A Unique and Rigorous Data Platform for Depression Screening Research

Methods Notes 

All eligible studies must conduct multivariable evaluations of factors associated with mental health symptoms, which may include, for instance, pre-COVID-19 participant or community characteristics, pre-existing health characteristics, and aspects of participants’ experience during COVID-19 (e.g. infection, exposure to infected individuals, work status, lockdown status). Mental health outcomes must be based on validated diagnostic interviews, scales, or sets of items. Both cross-sectional and longitudinal designs are eligible, although cross-sectional designs will only be included if factors assessed were present prior to the outcome (e.g., sex, pre-existing medical condition) and not measured only concurrently. Thus, studies with multivariable analyses that include concurrent mental health or behavioural variables (e.g. information seeking, exercise) will be excluded, as these do not permit interpretation of directionality. Studies with < 100 participants will be excluded.

 

Summary of Results

Preliminary results have been removed from the website as they no longer reflect up-to-date evidence

 

Comment

As of September 12th, 2020, the team has stopped reviewing new references for research question 2, “Factors Associated with Levels or Changes in Symptoms”. Since the project’s beginning, question 2 has posed challenges, as it assesses findings from cross-sectional studies, not studies which evaluate changes over time. Even in well-conducted studies, it is impossible to ascertain the degree to which risk factors identified reflect COVID-19-specific factors or replicate associations present prior to the pandemic. Additionally, the very high volume of low-quality studies eligible for question 2 has posed a challenge to the team’s ability to keep evidence current. Since cross-sectional model-based studies are highly dependent on completeness and design of the models, many eligible studies have been at high risk of bias. This has negatively impacted our ability to highlight trends and report useful conclusions. Due to these concerns, including the rapidly expanding volume of potentially eligible studies for question 2 and our team’s limited resources, we will analyze studies published as of September 12, 2020, but will not maintain a living systematic review for question 2 after that date.