Harvard Data Science Review Podcast Mental Health Challenges: How Can Data Science Help?
Jul 16, 2021
Margarita Alegria, Chief of the Disparities Research Unit at Massachusetts General Hospital, and Robert Gibbons, a biostatistics expert at the University of Chicago, delve into the intersection of data science and mental health. They discuss how adaptive testing can swiftly assess mental health risks like depression and anxiety. The conversation touches on the cultural nuances in reporting mental health issues, the unexpected suicide trends during the pandemic, and the promise of psychedelics as potential treatments, emphasizing the need for better data methodologies.
AI Snips
Chapters
Transcript
Episode notes
Population-Wide Drug Screening Reveals Risk Signals
- Screening all drugs simultaneously can reveal both increased and decreased suicide risk associations.
- Rob Gibbons used claims data from 150 million people to identify signals like alprazolam raising risk and folic acid lowering it.
Data Scientists Should Join Policy Decisions
- Data scientists should take an active role in policy discussions to ensure correct causal interpretation.
- Rob Gibbons urged statisticians to sit at the policy table and help translate complex evidence for decisions.
Adaptive Testing Tailors Assessment By Severity
- Computerized adaptive testing (CAT) tailors items to respondent severity, improving measurement efficiency.
- Rob Gibbons extended CAT to multidimensional models to better capture complex constructs like depression.