What question did this study set out to answer?

To develop and evaluate a predictive risk model for fatal and nonfatal opioid overdose using linked statewide medical, behavioral health, and criminal justice records.

June 24, 2020JAMA PsychiatryOpen Access

Predictive Modeling of Opioid Overdose Using Linked Statewide Medical and Criminal Justice Data

View Full Paper

Ask AI

Bookmark

Authors

BSBrendan SalonerBrown University HCHsien‐Yen ChangJohnson & Johnson (United States)NKNoa KrawczykNew York University

Discussion

Member takes

Implication

Cross-sectional study uncovers high predictive accuracy for fatal and nonfatal opioid overdose in high-risk adults, highlighting the value of combining health and criminal justice records.

Key Points

To develop and evaluate a predictive risk model for fatal and nonfatal opioid overdose using linked statewide medical, behavioral health, and criminal justice records.
Analyzed linked 2015 administrative data across hospital discharges, prescription drug monitoring programs, public behavioral health treatment, and criminal justice records for N=2,294,707 Maryland adults aged 18 to 80 years.
Estimated multivariable logistic regression models with bootstrapping to predict fatal and nonfatal opioid overdoses documented in 2016.
In 2016, fatal opioid overdoses occurred in 1,204 individuals (0.05%) and nonfatal overdoses in 8,430 individuals (0.37%).
Fatal overdose was most strongly associated with prison release in 2015 (OR, 4.23; 95% CI, 2.10-7.11), hospital opioid use disorder diagnosis (OR, 2.93; 95% CI, 2.17-3.80), receiving addiction medication (OR, 2.81; 95% CI, 2.20-3.86), and male sex (OR, 2.40; 95% CI, 2.08-2.76).
Model discrimination yielded an area under the curve of 0.82 using hospital data alone, 0.86 with added prescription monitoring data, and 0.89 for the full model including criminal justice and behavioral health data (0.85 for nonfatal overdose).

Cite This Study

Saloner et al. (2020) studied this question.

synapsesocial.com/papers/6aa5766d66012019823a4ea5 https://doi.org/10.1001/jamapsychiatry.2020.1689

View Full Paper

Ask AI

Bookmark

Implication

Cross-sectional study uncovers high predictive accuracy for fatal and nonfatal opioid overdose in high-risk adults, highlighting the value of combining health and criminal justice records.

Predictive Modeling of Opioid Overdose Using Linked Statewide Medical and Criminal Justice Data

Discussion

Member takes

Implication

Key Points

Cite This Study

Predictive Modeling of Opioid Overdose Using Linked Statewide Medical and Criminal Justice Data

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider