Reporting of harms in systematic reviews focused on naltrexone: a cross-sectional study

Joseph Schnitker¹Lindsey Purcell¹Morgan Garrett¹Holly Flores²Audrey Wise³Micah Kee⁴Brayden Rucker⁵Adam Khan^1*Jason Beaman⁶Matt Vassar⁶

• ¹Office of Medical Student Research, Oklahoma State University Center for Health Sciences, Tulsa, Oklahoma, United States
• ²Department of Obstetrics, McGovern Medical School, University of Texas Health Science Center at Houston, Houston, Texas, United States
• ³Department of Obstetrics & Gynecology, Baylor Scott and White Health, Temple, New Hampshire, United States
• ⁴Department of Internal Medicine, Oklahoma State University Medical Center, Tulsa, Oklahoma, United States
• ⁵Department of Anesthesiology, College of Medicine, University of Oklahoma, Oklahoma City, Oklahoma, United States
• ⁶Department of Psychiatry and Behavioral Sciences, Oklahoma State University Center for Health Sciences, Tulsa, Oklahoma, United States

Background Naltrexone is a pharmacological intervention widely used for Alcohol Use Disorder (AUD), Opioid Use Disorder (OUD), and several off-label conditions. Systematic reviews (SRs) play a critical role in synthesizing data on the efficacy and safety of such interventions to inform clinical guidelines and decision-making. However, adequate reporting of harms in SRs remains inconsistent, limiting the ability to fully assess the safety profile of naltrexone. This study evaluates the completeness of harms reporting and methodological quality in SRs focusing on naltrexone. Methods A comprehensive search of MEDLINE, EMBASE, Epistemonikos, and the Cochrane Database for Systematic Reviews was conducted. The study employed masked, duplicate screening and data extraction. Included SRs were evaluated for completeness of harms reporting using the PRISMA harms checklist and other established frameworks. Methodological quality was appraised using the AMSTAR-2 tool, and primary study overlap among SRs was assessed through corrected covered area (CCA) analysis. Results 87 SRs were included in the analysis. Only 1.1% (1/87) utilized severity scales to classify harms, and 4.6% (4/87) defined harms in their methods. Nearly half (48.3%) of SRs failed to address harms as either a primary or secondary outcome. A total of 82.8% (72/87) of SRs were rated as "critically low" quality by AMSTAR-2. Statistical analysis revealed a significant relationship between "critically low" AMSTAR-2 ratings and incomplete harms reporting (P=.0486). Additionally, 4 SR pairs demonstrated "high" overlap (>50%) of primary studies, accompanied by inconsistencies in harms reporting. Conclusion Our findings underscore the critical need for improved and standardized harms reporting in SRs on naltrexone. Inconsistent and incomplete reporting limits the ability of clinicians to fully assess the safety profile of naltrexone within systematic reviews. Adopting established frameworks such as PRISMA harms extensions and severity scales is imperative to enhance transparency and reliability in SRs. This study advocates for methodological improvements in SRs to support comprehensive safety evaluations and evidence-based prescribing of naltrexone.