NIDA Sponsored Research – Precision Opiate Use Disorder Treatment Tools

Deaths as a result of overdose reflect a 2.6 fold increase in the past decade suggesting the crisis continues to worsen. Traditional treatment methods have shown limited success, with high relapse rates and low adherence. According to health economist and DBH partner, Dr. Jason Gibbons there are several factors as to why that may be occurring: First, existing research and opiate use treatment practice have primarily focused on static approaches with limited consideration for the significant temporal variation in treatment response and illness progression. Second, conventional statistical and care approaches commonly used to study and treat OUDs often fail to account for treatment effect heterogeneity based on individual characteristics. This has led to broad clinical recommendations not optimized for OUD patients facing unique clinical, social, and environmental circumstances. Finally, while there is a growing recognition of the importance of using patient-reported outcome measures to inform clinical practice, there is a current shortage in collecting these measures and integrating them into clinical decision-making processes. The NIDA sponsored research in collaboration with Discovery Behavioral Health led by Dr. Gibbons aims to address these gaps by using cutting-edge statistical and machine learning techniques in combination with a novel OUD severity measure to tailor treatments to individual patient needs over time, with the long-term goal of turning the resulting models into clinical decision support tools.

NIMH Sponsored Research – Precision Mental Health Tools to Inform Treatment Selection in Depression

Major depressive disorder is common and responsible for the most healthy life years lost in the US and worldwide, in part because up to 55% of patients fail to respond to >2 adequate antidepressant trials and are burdened by treatment-resistant depression (TRD). Although many treatment regimes for TRD are available, patients endure lengthy trial-and-error attempts on different regimes and increased risk for adverse outcomes, because clinicians lack decision support tools to select initial or subsequent treatments. Accurate and accessible prediction models are lacking in part because predictors of treatment response are static (time-fixed), despite varying over time. Clinical decision support systems (CDSS) need dynamic treatment regimes (DTRs): sequences of rules that dynamically update treatment response predictions as information becomes available over a patient’s course of treatment.

Precision Mental Health Tools to Inform Treatment Selection in Depression calls for a “pipeline to support initial tests of validation and feasibility of objective, easy-to-use, and widely accessible tools for predicting response to depression treatments.” The Investigators and national behavioral health system Discovery Behavioral Health [DBH] will achieve these goals by using National Real-World Electronic Health, Patient-Reported Outcome and Digital Biomarker

Data with Advanced Causal Inference Methods and we will study the comparative effectiveness of dynamic decision support rules for TRD, optimize dynamic decision support rules and then validate the DTRs via Target Trial Emulation and a Sequential Multiple Assignment Randomized Trial (SMART).

NIMH Sponsored Research – Advanced Laboratories for Accelerating the Reach and Impact of Treatments for Youth and Adults with Mental Illness (ALACRITY) In Process

The first goal of the study partnership with the Cambridge Health Alliance and Harvard’s Department of Psychiatry, Discovery Behavioral Health will develop dynamic treatment regimes to reduce suicide thoughts and behaviors in youth nationally. Next, is to design an empirically driven system of decision support rules that operationalize the dynamic treatment regimes. Lastly, a pilot trial will be conducted to optimize adaptive decision supports.

Our Research Team