Rates of discontinuation and non-publication of trials for the pharmacologic treatment of alcohol use disorder

Abstract

Objective: With approximately 15 million individuals in the United States meeting criteria for Alcohol Use Disorder (AUD), advancing effective medication-assisted treatment options is crucial. This advancement stems from the publication of clinical trial (CT) results. The primary objective of this study was to assess the rates of discontinuation and non-publication of results in CTs focused on the pharmacologic treatment of AUD and to assess associated factors. Design: A cross-sectional study was completed after acquiring trials focused on AUD within the ClinicalTrials.gov database. Inclusion criteria for CTs were the use of a pharmaceutical intervention with an outcome measure of alcohol intake or craving, conducted between October 2008 through September 2018. The primary outcome measures were the frequency of trial discontinuation and non-publication. Discontinuation was assessed as the listed status on ClinicalTrials.gov. Publications were identified through verification of listings on ClinicalTrials.gov, or via searches of PubMed, Embase, and Google Scholar. If publications were not found, correspondence to trial contacts were sent. Data analysis occurred on December 5th, 2020. Results: Of 235 trials returned from the search, 87 met inclusion criteria. Our study found that 12.6% (11) of CTs involving pharmaceutical treatments for AUD were prematurely terminated, and 39.1% (34) had no published results. Recruitment and lower cutoff of targeted age groups were significantly associated with discontinuation and non-publication, respectively. Conclusions: Scientific evidence advances faster when all results are known—furthering the progress of positive studies, while avoiding duplicative efforts to test the same hypotheses with the same methods, thereby reducing scientific waste. Given the number of unpublished AUD trials, potentially useful information regarding treatment for individuals with AUD may be inaccessible to clinicians while also adding to the abundance of research waste.

Key points

• Question: What are the rates of discontinuation or non-publication of clinical trials for pharmacological treatments for AUD and associated factors?

• Findings: Among 87 trials, 11 (12.6%) were prematurely terminated and 34 (39.1%) did not reach publication, with trials that included participants 21 years and older more likely to reach publication than those with younger participants.

• Meaning: Low publication rates of CTs for pharmacological treatments of AUD may (1) stunt the advancement of AUD research, (2) decrease the value in funding AUD research from government entities, and (3) needlessly expose participants to potentially harmful interventions.

Keywords:

• Alcohol
• clinical trials
• discontinuation
• research waste

Disclosure statement

No potential conflict of interest was reported by the author(s).

Author contributions

Micah Hartwell was responsible for the study’s conceptualization, data extraction, formal analysis, and overall manuscript development. John Whelan collected and extracted data. Nicholas Sajjadi and Samuel Shepard developed the original draft of the manuscript. Alicia Ford, Jason Beaman, and Jamie Roberts provided critical revision based on their respective expertise. Matt Vassar provided supervision.

Additional information

Funding

No financial or other sources of support were provided during the development of this Protocol. Micah Hartwell receives research support through the National Institutes of Justice. Matt Vassar reports grant funding from the National Institutes of Health, the U.S. Office of Research Integrity, and Oklahoma Center for the Advancement of Science and Technology, all outside the present work.