Majid Afshar, Emma J Graham Linck, Alexandra B Spicer, John Rotrosen, Elizabeth M Salisbury-Afshar, Pratik Sinha, Matthew W Semler, Matthew M Churpek
{"title":"Machine Learning-Driven Analysis of Individualized Treatment Effects Comparing Buprenorphine and Naltrexone in Opioid Use Disorder Relapse Prevention.","authors":"Majid Afshar, Emma J Graham Linck, Alexandra B Spicer, John Rotrosen, Elizabeth M Salisbury-Afshar, Pratik Sinha, Matthew W Semler, Matthew M Churpek","doi":"10.1097/ADM.0000000000001313","DOIUrl":null,"url":null,"abstract":"<p><strong>Objective: </strong>A trial comparing extended-release naltrexone and sublingual buprenorphine-naloxone demonstrated higher relapse rates in individuals randomized to extended-release naltrexone. The effectiveness of treatment might vary based on patient characteristics. We hypothesized that causal machine learning would identify individualized treatment effects for each medication.</p><p><strong>Methods: </strong>This is a secondary analysis of a multicenter randomized trial that compared the effectiveness of extended-release naltrexone versus buprenorphine-naloxone for preventing relapse of opioid misuse. Three machine learning models were derived using all trial participants with 50% randomly selected for training (n = 285) and the remaining 50% for validation. Individualized treatment effect was measured by the Qini value and c-for-benefit, with the absence of relapse denoting treatment success. Patients were grouped into quartiles by predicted individualized treatment effect to examine differences in characteristics and the observed treatment effects.</p><p><strong>Results: </strong>The best-performing model had a Qini value of 4.45 (95% confidence interval, 1.02-7.83) and a c-for-benefit of 0.63 (95% confidence interval, 0.53-0.68). The quartile most likely to benefit from buprenorphine-naloxone had a 35% absolute benefit from this treatment, and at study entry, they had a high median opioid withdrawal score ( P < 0.001), used cocaine on more days over the prior 30 days than other quartiles ( P < 0.001), and had highest proportions with alcohol and cocaine use disorder ( P ≤ 0.02). Quartile 4 individuals were predicted to be most likely to benefit from extended-release naltrexone, with the greatest proportion having heroin drug preference ( P = 0.02) and all experiencing homelessness ( P < 0.001).</p><p><strong>Conclusions: </strong>Causal machine learning identified differing individualized treatment effects between medications based on characteristics associated with preventing relapse.</p>","PeriodicalId":14744,"journal":{"name":"Journal of Addiction Medicine","volume":null,"pages":null},"PeriodicalIF":4.2000,"publicationDate":"2024-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11446670/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Addiction Medicine","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1097/ADM.0000000000001313","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2024/5/22 0:00:00","PubModel":"Epub","JCR":"Q1","JCRName":"SUBSTANCE ABUSE","Score":null,"Total":0}
引用次数: 0
Abstract
Objective: A trial comparing extended-release naltrexone and sublingual buprenorphine-naloxone demonstrated higher relapse rates in individuals randomized to extended-release naltrexone. The effectiveness of treatment might vary based on patient characteristics. We hypothesized that causal machine learning would identify individualized treatment effects for each medication.
Methods: This is a secondary analysis of a multicenter randomized trial that compared the effectiveness of extended-release naltrexone versus buprenorphine-naloxone for preventing relapse of opioid misuse. Three machine learning models were derived using all trial participants with 50% randomly selected for training (n = 285) and the remaining 50% for validation. Individualized treatment effect was measured by the Qini value and c-for-benefit, with the absence of relapse denoting treatment success. Patients were grouped into quartiles by predicted individualized treatment effect to examine differences in characteristics and the observed treatment effects.
Results: The best-performing model had a Qini value of 4.45 (95% confidence interval, 1.02-7.83) and a c-for-benefit of 0.63 (95% confidence interval, 0.53-0.68). The quartile most likely to benefit from buprenorphine-naloxone had a 35% absolute benefit from this treatment, and at study entry, they had a high median opioid withdrawal score ( P < 0.001), used cocaine on more days over the prior 30 days than other quartiles ( P < 0.001), and had highest proportions with alcohol and cocaine use disorder ( P ≤ 0.02). Quartile 4 individuals were predicted to be most likely to benefit from extended-release naltrexone, with the greatest proportion having heroin drug preference ( P = 0.02) and all experiencing homelessness ( P < 0.001).
Conclusions: Causal machine learning identified differing individualized treatment effects between medications based on characteristics associated with preventing relapse.
期刊介绍:
The mission of Journal of Addiction Medicine, the official peer-reviewed journal of the American Society of Addiction Medicine, is to promote excellence in the practice of addiction medicine and in clinical research as well as to support Addiction Medicine as a mainstream medical sub-specialty.
Under the guidance of an esteemed Editorial Board, peer-reviewed articles published in the Journal focus on developments in addiction medicine as well as on treatment innovations and ethical, economic, forensic, and social topics including:
•addiction and substance use in pregnancy
•adolescent addiction and at-risk use
•the drug-exposed neonate
•pharmacology
•all psychoactive substances relevant to addiction, including alcohol, nicotine, caffeine, marijuana, opioids, stimulants and other prescription and illicit substances
•diagnosis
•neuroimaging techniques
•treatment of special populations
•treatment, early intervention and prevention of alcohol and drug use disorders
•methodological issues in addiction research
•pain and addiction, prescription drug use disorder
•co-occurring addiction, medical and psychiatric disorders
•pathological gambling disorder, sexual and other behavioral addictions
•pathophysiology of addiction
•behavioral and pharmacological treatments
•issues in graduate medical education
•recovery
•health services delivery
•ethical, legal and liability issues in addiction medicine practice
•drug testing
•self- and mutual-help.