Innovations in suicide prevention research (INSPIRE): a protocol for a population-based case-control study

Area of Science:

• Public Health
• Data Science
• Epidemiology

Background:

• Suicide deaths have increased 29% in the USA since 1999, reaching 45,979 in 2020.
• Lack of data integration between health insurers, institutions, and corrections impedes comprehensive suicide prevention strategies.

Purpose of the Study:

• To link diverse administrative datasets with death records for enhanced suicide surveillance.
• To estimate associations between risk factors and suicide outcomes.
• To develop predictive algorithms for identifying at-risk individuals and establish long-term surveillance workflows.

Main Methods:

• Combining six North Carolina data sources (death records, VDRS, insurance claims, Medicaid, EHR, justice data) from 2006 onwards.
• Utilizing a nested case-control design to identify short- and long-term suicide risk factors.
• Developing machine learning algorithms to predict suicide risk.

Main Results:

• Establishing benchmarks for suicide incidence, attempts, and ideation across subpopulations.
• Identifying key risk factors associated with suicide attempts and mortality.
• Developing predictive models for suicide risk identification.

Conclusions:

• Creating an integrated suicide surveillance system by linking multiple large databases.
• Addressing prior study gaps through comprehensive data integration and analysis.
• Enabling ongoing surveillance, predictor identification, and prevention effort evaluation.