Applying Causal Inference Methods in Psychiatric Epidemiology: A Review

Area of Science:

• Psychiatric epidemiology
• Causal inference methodology

Background:

• Associations between risk factors and psychiatric disorders are common but may not imply causation.
• Relying solely on observational data for prevention is hazardous.

Purpose of the Study:

• To review and evaluate causal inference methods for psychiatric epidemiology.
• To inform the development of realistic etiologic models for psychopathology.

Main Methods:

• Discussion of randomized clinical trials (RCTs), natural experiments, and statistical models.
• Examination of methods like propensity scoring and marginal models.
• Review of quasi-experimental designs such as instrumental variables and regression discontinuity.

Main Results:

• RCTs are the gold standard but have limitations in generalizability and ethical feasibility for psychiatric research.
• Statistical and quasi-experimental methods offer alternatives but control for different confounders.
• Each method possesses unique strengths and limitations regarding assumptions and generalizability.

Conclusions:

• Causal inference is vital for advancing etiologic understanding and prevention in psychiatry.
• A simplistic view favoring only RCTs is inadequate; diverse methods have value.
• Triangulation of multiple causal inference methods can strengthen confidence in findings.