Ensemble learning to predict opioid-related overdose using statewide prescription drug monitoring program and hospital discharge data in the state of Tennessee.

OBJECTIVE: To develop and validate algorithms for predicting 30-day fatal and nonfatal opioid-related overdose using statewide data sources including prescription drug monitoring program data, Hospital Discharge Data System data, and Tennessee (TN) vital records. Current overdose prevention efforts in TN rely on descriptive and retrospective analyses without prognostication. MATERIALS AND METHODS: Study data included 3 041 668 TN patients with 71 479 191 controlled substance prescriptions from 2012 to 2017. Statewide data and socioeconomic indicators were used to train, ensemble, and calibrate 10 nonparametric "weak learner" models. Validation was performed using area under the receiver operating curve (AUROC), area under the precision recall curve, risk concentration, and Spiegelhalter z-test statistic. RESULTS: Within 30 days, 2574 fatal overdoses occurred after 4912 prescriptions (0.0069%) and 8455 nonfatal overdoses occurred after 19 460 prescriptions (0.027%). Discrimination and calibration improved after ensembling (AUROC: 0.79-0.83; Spiegelhalter P value: 0-.12). Risk concentration captured 47-52% of cases in the top quantiles of predicted probabilities. DISCUSSION: Partitioning and ensembling enabled all study data to be used given computational limits and helped mediate case imbalance. Predicting risk at the prescription level can aggregate risk to the patient, provider, pharmacy, county, and regional levels. Implementing these models into Tennessee Department of Health systems might enable more granular risk quantification. Prospective validation with more recent data is needed. CONCLUSION: Predicting opioid-related overdose risk at statewide scales remains difficult and models like these, which required a partnership between an academic institution and state health agency to develop, may complement traditional epidemiological methods of risk identification and inform public health decisions.

Leiomyomas in Pregnancy and Spontaneous Abortion: A Systematic Review and Meta-analysis.

OBJECTIVE: To systematically review studies reporting the risk of spontaneous abortion among pregnant women of typical reproductive potential with and without uterine leiomyomas. DATA SOURCES: We searched PubMed, EMBASE, Web of Science, and ClinicalTrials.gov for publications from January 1970 to December 2016. METHODS OF STUDY SELECTION: We excluded studies that did not use imaging to uniformly document leiomyoma status of all participants, did not have a comparison group without leiomyomas, or primarily included women seeking care for recurrent miscarriage, infertility care, or assisted reproductive technologies. TABULATION, INTEGRATION, AND RESULTS: Two authors independently reviewed eligibility, extracted data, and assigned overall quality ratings based on predetermined criteria. Of 1,469 articles identified, nine were eligible. Five enrolled general obstetric populations and four included women undergoing amniocentesis. In five studies in general obstetric populations that included 21,829 pregnancies (1,394 women with leiomyomas and 20,435 without), only one adjusted for potential confounders. This meta-analysis revealed no increase in risk of spontaneous abortion among those with leiomyomas compared with those without (11.5% compared with 8.0%; risk ratio 1.16, 95% CI 0.80-1.52). When bias from confounding was estimated for nonadjusted studies, the aggregate calculated risk ratio was 0.83 (95% CI 0.68-0.98). CONCLUSION: Leiomyoma presence was not associated with increased risk of spontaneous abortion in an analysis of more than 20,000 pregnant women. Failure of prior studies to adjust for confounders may have led to the common clinical belief that leiomyomas are a risk factor for spontaneous abortion.