PROVIDENT: Development and Validation of a Machine Learning Model to Predict Neighborhood-level Overdose Risk in Rhode Island.

BACKGROUND: Drug overdose persists as a leading cause of death in the United States, but resources to address it remain limited. As a result, health authorities must consider where to allocate scarce resources within their jurisdictions. Machine learning offers a strategy to identify areas with increased future overdose risk to proactively allocate overdose prevention resources. This modeling study is embedded in a randomized trial to measure the effect of proactive resource allocation on statewide overdose rates in Rhode Island (RI). METHODS: We used statewide data from RI from 2016 to 2020 to develop an ensemble machine learning model predicting neighborhood-level fatal overdose risk. Our ensemble model integrated gradient boosting machine and super learner base models in a moving window framework to make predictions in 6-month intervals. Our performance target, developed a priori with the RI Department of Health, was to identify the 20% of RI neighborhoods containing at least 40% of statewide overdose deaths, including at least one neighborhood per municipality. The model was validated after trial launch. RESULTS: Our model selected priority neighborhoods capturing 40.2% of statewide overdose deaths during the test periods and 44.1% of statewide overdose deaths during validation periods. Our ensemble outperformed the base models during the test periods and performed comparably to the best-performing base model during the validation periods. CONCLUSIONS: We demonstrated the capacity for machine learning models to predict neighborhood-level fatal overdose risk to a degree of accuracy suitable for practitioners. Jurisdictions may consider predictive modeling as a tool to guide allocation of scarce resources.

Translating Predictive Analytics for Public Health Practice: A Case Study of Overdose Prevention in Rhode Island.

Prior applications of machine learning to population health have relied on conventional model assessment criteria, limiting the utility of models as decision support tools for public health practitioners. To facilitate practitioners' use of machine learning as a decision support tool for area-level intervention, we developed and applied 4 practice-based predictive model evaluation criteria (implementation capacity, preventive potential, health equity, and jurisdictional practicalities). We used a case study of overdose prevention in Rhode Island to illustrate how these criteria could inform public health practice and health equity promotion. We used Rhode Island overdose mortality records from January 2016-June 2020 (n = 1,408) and neighborhood-level US Census data. We employed 2 disparate machine learning models, Gaussian process and random forest, to illustrate the comparative utility of our criteria to guide interventions. Our models predicted 7.5%-36.4% of overdose deaths during the test period, illustrating the preventive potential of overdose interventions assuming 5%-20% statewide implementation capacities for neighborhood-level resource deployment. We describe the health equity implications of use of predictive modeling to guide interventions along the lines of urbanicity, racial/ethnic composition, and poverty. We then discuss considerations to complement predictive model evaluation criteria and inform the prevention and mitigation of spatially dynamic public health problems across the breadth of practice. This article is part of a Special Collection on Mental Health.

Neighborhood-level association between release from incarceration and fatal overdose, Rhode Island, 2016-2020.

The association between recent release from incarceration and dramatically increased risk of fatal overdose is well-established at the individual level. Fatal overdose and. arrest/release are spatially clustered, suggesting that this association may persist at the neighborhood level. We analyzed multicomponent data from Rhode Island, 2016-2020, and observed a modest association at the census tract level between rates of release per 1000 population and fatal overdose per 100,000 person-years, adjusting for spatial autocorrelation in both the exposure and outcome. Our results suggest that for each additional person released to a given census tract per 1000 population, there is a corresponding increase in the rate of fatal overdose by 2 per 100,000 person years. This association is more pronounced in suburban tracts, where each additional release awaiting trial is associated with an increase in the rate of fatal overdose of 4 per 100,000 person-years and 6 per 100,000 person-years for each additional release following sentence expiration. This association is not modified by the presence or absence of a licensed medication for opioid use disorder (MOUD) treatment provider in the same or surrounding tracts. Our results suggest that neighborhood-level release rates are moderately informative as to tract-level rates of fatal overdose and underscore the importance of expanding pre-release MOUD access in correctional settings. Future research should explore risk and resource environments particularly in suburban and rural areas and their impacts on overdose risk among individuals returning to the community.

Thirty-day Treatment Continuation After Audio-only Buprenorphine Telehealth Initiation.

OBJECTIVES: Before the coronavirus disease 2019 pandemic, federal law required in-person evaluation before buprenorphine initiation. Regulatory changes during the pandemic allow for buprenorphine initiation by audio-only or audiovisual telehealth. Little is known about treatment engagement after buprenorphine initiation conducted via audio-only telehealth. METHODS: A retrospective cohort study of 94 individuals who received initial treatment through an audio-only encounter between April 2020 and February 2021 was performed. Participant demographics, substance use history, withdrawal symptoms, 30-day treatment engagement, and adverse outcomes were determined by an electronic chart and REDcap database review. Subsequent buprenorphine prescriptions filled within 30 days of the initial encounter were tracked through the Rhode Island Prescription Drug Monitoring Program. RESULTS: Buprenorphine was prescribed for 94 individuals. Most (92 of 94 [97.9%]) filled their prescription within 30 days. Most had previously taken buprenorphine, including prescribed (42 of 92 [45.7%]) and nonprescribed (58 of 92 [63.0%]). Two thirds were in opioid withdrawal at the time of the call (61 of 92 [66.3%]) with a mean Subjective Opioid Withdrawal Scale of 26.8 (range, 4-57). Four individuals experienced precipitated withdrawal (4 of 94 [4.3%]), and 2 reported persistent withdrawal at their follow-up visit (2 of 94 [2.1%]). More than 70% filled a subsequent prescription for buprenorphine within 30 days of the end of their hotline prescription (65 of 92 [70.7%]), on average of 5.88 days (range, 0-28) after completion of their telehealth prescription. CONCLUSIONS: Expanding telehealth-delivered buprenorphine care has the potential to address treatment gaps and facilitate delivery of on-demand services during peak motivation. This evaluation of audio-only buprenorphine initiation found high rates of unobserved buprenorphine initiation and treatment continuation with low rates of complications.

Divergence Between Individual- and Neighborhood-Level Fatal Overdose Burden: A Population-Based Statewide Study.

OBJECTIVES: To compare the characteristics of individual overdose decedents in Rhode Island, 2016-2020 to the neighborhoods where fatal overdoses occurred over the same time period. METHODS: We conducted a retrospective analysis of fatal overdoses occurring between January 1, 2016 and June 30, 2020. Using individual- and neighborhood-level data, we conducted descriptive analyses to explore the characteristics of individuals and neighborhoods most affected by overdose. RESULTS: Most overdose decedents during the study period were non-Hispanic White. Across increasingly more White and non-Hispanic neighborhoods, rates of fatal overdose per 100,000 person-years decreased. An opposite pattern was observed across quintiles of average neighborhood poverty. CONCLUSIONS: Rates of fatal overdose were higher in less White, more Hispanic, and poorer neighborhoods, suggesting modest divergence between the characteristics of individuals and the neighborhoods most severely affected. These impacts may not be uniform across space and may accrue differentially to more disadvantaged and racially/ethnically diverse neighborhoods.

Preventing Overdose Using Information and Data from the Environment (PROVIDENT): protocol for a randomized, population-based, community intervention trial.

BACKGROUND AND AIMS: In light of the accelerating drug overdose epidemic in North America, new strategies are needed to identify communities most at risk to prioritize geographically the existing public health resources (e.g. street outreach, naloxone distribution efforts). We aimed to develop PROVIDENT (Preventing Overdose using Information and Data from the Environment), a machine learning-based forecasting tool to predict future overdose deaths at the census block group (i.e. neighbourhood) level. DESIGN: Randomized, population-based, community intervention trial. SETTING: Rhode Island, USA. PARTICIPANTS: All people who reside in Rhode Island during the study period may contribute data to either the model or the trial outcomes. INTERVENTION: Each of the state's 39 municipalities will be randomized to the intervention (PROVIDENT) or comparator condition. An interactive, web-based tool will be developed to visualize the PROVIDENT model predictions. Municipalities assigned to the treatment arm will receive neighbourhood risk predictions from the PROVIDENT model, and state agencies and community-based organizations will direct resources to neighbourhoods identified as high risk. Municipalities assigned to the control arm will continue to receive surveillance information and overdose prevention resources, but they will not receive neighbourhood risk predictions. MEASUREMENTS: The primary outcome is the municipal-level rate of fatal and non-fatal drug overdoses. Fatal overdoses will be defined as unintentional drug-related death; non-fatal overdoses will be defined as an emergency department visit for a suspected overdose reported through the state's syndromic surveillance system. Intervention efficacy will be assessed using Poisson or negative binomial regression to estimate incidence rate ratios comparing fatal and non-fatal overdose rates in treatment vs. control municipalities. COMMENTS: The findings will inform the utility of predictive modelling as a tool to improve public health decision-making and inform resource allocation to communities that should be prioritized for prevention, treatment, recovery and overdose rescue services.