Transitivity and Treatment-Effect Modifiers in Network Meta-Analysis of Co-Occurring ADHD and Stimulant Use Disorder: Reply to Yen and Shan

See letter by Yen and Shan and article by Oliva et al

To the Editor: We thank Drs Yen and Shan for their thoughtful comments¹ on our network meta-analysis² of pharmacotherapies for co-occurring stimulant use disorder (StUD) and attention-deficit/hyperactivity disorder (ADHD). Their point is well taken: Stimulant subtype is a clinically plausible treatment-effect modifier, and cocaine use disorder and amphetamine-type stimulant use disorder should not be assumed to be fully interchangeable when interpreting pharmacotherapy effects.² As emphasized by Jansen and Naci, valid indirect comparisons require sufficient comparability of important effect modifiers across treatment comparisons.³

In our analysis, mixed amphetamine salts (MAS) ranked most favorably for stimulant-use outcomes, particularly negative urine toxicology, whereas methylphenidate (MPH) ranked most favorably for ADHD symptom reduction.¹ These rankings require caution because MAS was evaluated only in cocaine-use trials, while MPH was evaluated in both cocaine-and amphetamine-type stimulant use disorders.¹ Consequently, apparent differences between formulations may partly reflect stimulant subtype, geography, treatment setting, or trial design.

We attempted to address transitivity by evaluating stimulant type, treatment setting, concomitant psychosocial treatment, and baseline ADHD severity as potential effect modifiers.¹ Trials involving amphetamine dependence were conducted in Sweden, including criminal-justice settings, whereas cocaine trials were conducted in US outpatient settings; all nevertheless enrolled treatment-seeking adults with confirmed ADHD and StUD and provided cognitive behavioral therapy.¹ These similarities support, but cannot establish, transitivity. This limitation is particularly difficult to resolve in sparse networks. An empirical study of 209 published networks found study dissimilarity to be common and formal statistical assessments of transitivity feasible in only 61% of datasets, underscoring the importance of transparent, cautious interpretation rather than implying that transitivity can be definitively verified.⁴

Stimulant subtype is also unlikely to be the only relevant effect modifier. Sex may influence medication response in cocaine use disorder. A review highlighted differential outcomes for some pharmacotherapies and noted that sex-specific analyses are often unavailable or underpowered.⁵ Co-occurring substance use may also matter: Modafinil showed efficacy in a trial excluding alcohol dependence, consistent with prior data suggesting different responses among patients with versus without comorbid alcohol dependence.⁶ These examples illustrate how multiple clinical characteristics may shape pharmacotherapy effects yet remain difficult to examine when only aggregate data from 6 small trials are available. Accordingly, our rankings should be considered exploratory and hypothesis-generating, not evidence that one formulation is superior across all StUD subtypes. Future trials should stratify or prespecify analyses by stimulant subtype and other plausible modifiers, and head-to-head comparisons of long-acting amphetamine formulations and MPH within the same StUD subtype would be especially informative. Until then, treatment should be individualized according to stimulant subtype, ADHD severity, co-occurring substance use, sex-related considerations when evidence supports them, cardiovascular risk, diversion risk, treatment setting, and psychosocial care, consistent with the American Society of Addiction Medicine/American Academy of Addiction Psychiatry guideline’s substance-specific framework.⁷

Article Information

Published Online: November 11, 2026. https://doi.org/10.4088/JCP.26lr16713a
© 2026 Physicians Postgraduate Press, Inc. J Clin Psychiatry 2026;87(4):26lr16713a
To Cite: Oliva HNP, Angarita GA. Transitivity and treatment-effect modifiers in network meta-analysis of co-occurring ADHD and stimulant use disorder: reply to Yen and Shan. J Clin Psychiatry 2026;87(4):26lr16713a
Author Affiliations: Department of Psychiatry, Yale University School of Medicine, New Haven, Connecticut (Oliva and Angarita); Connecticut Mental Health Center, New Haven, Connecticut (Oliva and Angarita); Clinical Neuroscience Research Unit, Connecticut Mental Health Center, New Haven, Connecticut (Oliva and Angarita).
Corresponding Author: Gustavo A. Angarita, MD, MHS, Department of Psychiatry, Yale University School of Medicine, Connecticut Mental Health Center, 34 Park Street, New Haven, CT 06519 ([email protected]).
Financial Disclosure: The authors report no conflicts of interest with respect to the content of this manuscript.
Funding/Support: This work was supported in part by grants R01 DA052454-03 and R33DA053592 (Dr Angarita), from the National Institute on Drug Abuse and by the State of Connecticut, Department of Mental Health and Addiction Services.
Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Disclaimer: This publication does not express the views of the Department of Mental Health and Addiction Services or the State of Connecticut. The views and opinions expressed are those of the authors.