Luther Elliott

Error in Conflicts of Interest Statement

Objectives: To evaluate the performance of modified Poisson regression using cluster-weighted generalized estimating equations (MP-CWGEE) for estimating risk ratios in longitudinal data with informative cluster sizes (ICSs). Methods: Using a comprehensive Monte Carlo simulation study, we evaluated the finite-sample performance of MP-CWGEE relative to five competing approaches: (1) modified Poisson regression using (unweighted) generalized estimating equations (MP-GEE), (2) log-binomial regression using cluster-weighted generalized estimating equations (LB-CWGEE), (3) log-binomial regression using (unweighted) generalized estimating equations (LB-GEE), (4) log-binomial mixed effects model (LB-MM), and (5) Poisson mixed effects model (Pois-MM). We conducted 1,000 simulations across 324 simulation scenarios, spanning different ICS mechanisms, informativeness degrees, sample sizes, exposure effects, within-person correlations, and exposure-covariate associations. Performance metrics included convergence rate, empirical bias, standard error ratio, coverage rate, and empirical Type I error rate. MP-CWGEE’s practical application was demonstrated in a cohort study of 423 people who used illicit opioids in New York City. Results: MP-CWGEE recovered the individual-averaged marginal risk ratio with negligible bias, near-nominal coverage and Type I error, and perfect convergence across all scenarios, including under high ICS. The unweighted marginal estimators were biased under ICS, and the log-binomial approaches converged less reliably. The Pois-MM performed comparably to MP-CWGEE, but requires a random-effects specification and yields a conditional (cluster-specific) rather than a marginal interpretation. Conclusions: MP-CWGEE offers a robust modeling approach for estimating risk ratios in longitudinal data with ICSs. Implementation code for R, Stata, and SAS is provided to facilitate wider adoption.

Background: In 2021, NYC implemented overdose prevention center (OPC) services at two existing syringe exchange programs, allowing people to use pre-obtained drugs on-site. Although OPCs in Canada, Western Europe, and Australia have demonstrated their feasibility and benefits towards reducing overdose risk and drug-related harm, there is less data on how people who use drugs (PWUD) conceptualize the benefits and any potential drawbacks of using OPCs. Methods: In June–August 2022, we conducted 26 semistructured interviews with people in New York City who used unprescribed opioids. Interviews lasted 30–60 min and were conducted remotely using Zoom and later transcribed by a professional service. Data were then coded, using AtlasTi, into meaningful categories using a thematic approach based on the aims of the study and existing literature. Results: Most participants had heard of OnPoint and reported a willingness to use it. They described the ability of OnPoint staff to reverse an overdose quickly and the presence of naloxone, oxygen, and other supplies as the primary benefits. Yet, many also noted that OPCs provide PWUD with a place to escape from the weather and/or avoid law enforcement. Participants also reported concerns about how far PWUD would be willing to travel or wait to use an OPC and for the autonomy of PWUD in the context of formal, sanctioned OPCs. Conclusions: Results suggest that many PWUD in NYC are well-informed about OnPoint and are willing to use OPCs. Yet, to fulfil their potential, OPCs must be located near to where PWUD live, and should be made as low-threshold as possible. However, since it is unlikely that OPCs will be expanded enough to meet the need and because some PWUD will never choose to use in sanctioned OPCs, expanding the reach of alternative strategies, such as Mobile Overdose Response Services, is recommended.

Variation in binary outcomes over time by cluster size arises across various biomedical disciplines, including reproductive health, dental medicine, and psychiatric epidemiology. This study formally integrates modified Poisson regression with cluster-weighted generalized estimating equations (MP-CWGEE) for computing risk ratios in longitudinal studies with informative cluster sizes. Using a comprehensive Monte-Carlo simulation study, we empirically evaluated MP-CWGEE’s statistical properties against alternative modeling approaches: MP-GEE, log-binomial CWGEE (LB-CWGEE), and log-binomial GEE (LB-GEE). We conducted 1,000 simulations across varying sample sizes, risk ratios, and informativeness degrees. MP-CWGEE demonstrated superior performance in model convergence, empirical bias, average estimated standard error, coverage, and Type 1 error control. While LB-CWGEE showed comparable results, its convergence rates were slightly inferior. The benefits of cluster-weighted models (MP-CWGEE and LB-CWGEE) over unweighted models (MP-GEE and LB-GEE) were pronounced in scenarios with informative cluster sizes. We demonstrated MP-CWGEE’s practical application to a cohort study of people who used illicit opioids in New York City. We also provided implementation code for R, Stata, and SAS to facilitate wider adoption of the MP-CWGEE approach.

Background/Objectives: The opioid crisis has disproportionately impacted U.S. military veterans, who face heightened risks of opioid use disorder and overdose due to chronic pain and mental health conditions. The pharmaceutical industry’s role in misrepresenting opioid risks—leading to over USD 50 billion in legal settlements—has included targeted marketing to vulnerable populations. This study examines Janssen Pharmaceuticals’ “Imagine the Possibilities Pain Coalition” (IPPC), which aimed to increase opioid use among veterans with chronic non-cancer pain. Insights from this public health industry document analysis offer guidance for military medicine and healthcare policymaking. Methods: Using the Opioid Industry Document Archive (OIDA), housed at Johns Hopkins University and the University of California, San Francisco, researchers conducted retrospective content analysis. Documents referencing veterans were identified through keyword searches on Johns Hopkins’ SciServer portal and reviewed using CoCounsel, an AI-based legal document platform using a human-in-the-loop approach. Relevant documents were examined by the authors to extract material aligned with the research focus. Results: The IPPC employed strategies to influence opioid prescribing for veterans. These included educational materials that minimized addiction risks and exaggerated long-term benefits and empathy-driven narratives prioritizing immediate pain relief over potential harms. Ghostwriting ensured favorable perspectives on opioids in scientific literature, aligning with broader industry strategies to promote opioids for chronic pain. Conclusions: The targeted marketing of opioids to veterans has exacerbated the opioid crisis, as documented in government reports and litigation. Rigorous oversight of industry-funded coalitions and evidence-based practices are critical to insulating military healthcare from corporate influence and addressing the opioid crisis among veterans.

Background: Innovative analytic approaches to drug studies are needed to understand better the co-use of opioids with non-opioids among people using illicit drugs. One approach is the Bayesian kernel machine regression (BKMR), widely applied in environmental epidemiology to study exposure mixtures but has received far less attention in substance use research.Objective: To describe the utility of the BKMR approach to study the effects of drug substance mixtures on health outcomes.Methods: We simulated data for 200 individuals. Using the Vale and Maurelli method, we simulated multivariate non-normal drug exposure data: xylazine (mean = 300 ng/mL, SD = 100 ng/mL), fentanyl (mean = 200 ng/mL, SD = 71 ng/mL), benzodiazepine (mean = 300 ng/mL, SD = 55 ng/mL), and nitazene (mean = 200 ng/mL, SD = 141 ng/mL) concentrations. We performed 10,000 MCMC sampling iterations with three Markov chains. Model diagnostics included trace plots, r-hat values, and effective sample sizes. We also provided visual relationships of the univariate and bivariate exposure-response and the overall mixture effect.Results: Higher levels of fentanyl and nitazene concentrations were associated with higher levels of the simulated health outcome, controlling for age. Trace plots, r-hat values, and effective sample size statistics demonstrated BKMR stability across multiple Markov chains.Conclusions: Our understanding of drug mixtures tends to be limited to studies of single-drug models. BKMR offers an innovative way to discern which substances pose a greater health risk than other substances and can be applied to assess univariate, bivariate, and cumulative drug effects on health outcomes. Antecedentes: Se necesitan enfoques anal & iacute;ticos innovadores para estudios de medicamentos que permitan comprender mejor el consumo simult & aacute;neo de opioides y no opioides entre personas que usan drogas il & iacute;citas. Un enfoque es la regresi & oacute;n bayesiana de m & aacute;quina kernel (BKMR), ampliamente utilizada en epidemiolog & iacute;a ambiental para estudiar mezclas de exposiciones, pero que ha recibido mucha menos atenci & oacute;n en la investigaci & oacute;n sobre el uso de sustancias.Objetivo: Describir la utilidad del enfoque BKMR para estudiar los efectos de las mezclas de sustancias en los resultados de salud.M & eacute;todos: Simulamos datos para 200 individuos. Utilizando el m & eacute;todo de Vale y Maurelli, simulamos datos de exposici & oacute;n a drogas multivariadas no normales: concentraciones de xilazina (media = 300 ng/mL, SD = 100 ng/mL), fentanilo (media = 200 ng/mL, SD = 71 ng/mL), benzodiazepina (media = 300 ng/mL, SD = 55 ng/mL) y nitazeno (media = 200 ng/mL, SD = 141 ng/mL). Realizamos 10,000 iteraciones de muestreo MCMC con tres cadenas de Markov. Los diagn & oacute;sticos del modelo incluyeron gr & aacute;ficos de trazado, valores r-hat y tama & ntilde;os efectivos de muestra. Tambi & eacute;n proporcionamos relaciones visuales del efecto exposici & oacute;n-respuesta univariada y bivariada y del efecto global de la mezcla.Resultados: Niveles m & aacute;s altos de concentraciones de fentanilo y nitazeno se asociaron con niveles m & aacute;s altos del resultado de salud simulado, controlando la edad. Los gr & aacute;ficos de trazado, los valores r-hat y las estad & iacute;sticas del tama & ntilde;o efectivo de muestra demostraron la estabilidad de BKMR en m & uacute;ltiples cadenas de Markov.Conclusiones: Nuestra comprensi & oacute;n de las mezclas de medicamentos tiende a limitarse a estudios de modelos de un solo medicamento. BKMR ofrece una manera innovadora de discernir qu & eacute; sustancias representan un mayor riesgo para la salud que otras y se puede aplicar para evaluar los efectos univariados, bivariados y acumulativos de los medicamentos en los resultados de salud.

This commentary seeks to evaluate existing knowledge about the relationship between brain injury (BI) and overdose (OD), to unify distant bodies of literature, and to enhance prevention and treatment for opioid OD among individuals with BI. There is a hidden epidemic of undiagnosed BI in the United States. Due to lack of screening, the vast majority of BI sufferers do not know they have a BI. Not only are those with BI at elevated risk for opioid use, misuse, and opioid use disorder, but also they are at elevated risk for OD. Conversely, those with OUD and those who experienced an OD, are more likely to sustain BI. Key Findings/Conclusions: The existing literature suggests that primary strategies to reduce ABI (Acquired Brain Injury)/TBI (Traumatic Brain Injury) harms involve addressing: screening, stigma, racial disparities, and popular misconceptions about OD. The association between TBI and OD is an underexamined public health issue, exacerbated by the bidirectional nature of the relationship. Not only is TBI a risk factor for opioid OD; opioid OD was also found to be a major cause of ABI, which can have lifelong effects similar to Alzheimer's disease. Screening tools for BI were underutilized and inconsistently implemented across reviewed studies. Enhanced screening population wide is a promising intervention, complemented with expanded treatment and research. Black individuals face worse outcomes in BI and treatment outcomes. Anti-racist strategies must fight inequity while addressing social and structural drivers of overdose and BI within the opioid and opioid overdose crises.

This analysis identifies factors associated with overdose risk behaviors and non-fatal overdose among a sample of 577 adult-age people who use illicit opioids and live in NYC. Survey data--which included outcome measures assessing (1) past 30-day non-fatal overdose and past 30-day overdose-related risk behaviors and (2) predictors representing potential risk and protective factors—were analyzed using bivariate and Poisson regression techniques. Results indicate being 41–56 years in age, being widowed, using cannabis, injecting, and having greater pain severity and mental health challenges were associated with greater risks. Current employment, homelessness, and prescription of medications for opioid use disorder were associated with fewer risks. Being 57+ was negatively associated with past-month overdose; higher pain severity and opioid related withdrawal were positively related, as were employment, cannabis use, and injection. Findings suggest the importance of expanding access to MOUD and tailoring OD prevention interventions for mental health and pain management services.