Have conflicts of interest in manual therapy trials for low back pain been transparent? A meta-epidemiological review.

Abstract

Importance Accurate disclosure of conflicts of interest (COI) is vital for trust and transparency in clinical research, but few studies have evaluated these disclosures in the manual therapy literature. Objective The primary aim was to review study-disclosed COIs in manual therapy trials for low back pain (LBP) and compare them with those discovered by external searches. Data sources PubMed, CINAHL, CENTRAL, and PEDro were searched for trials investigating manual therapy for nonspecific LBP published between 2000 and 2025. Study selection Trials were included if joint mobilization/manipulation was part of the intervention for LBP and excluded if they were not in English or participants were not adults. Data extraction and synthesis Strength of trial conclusion, funding sources, and study-disclosed COIs were recorded. An independent online external search was conducted to identify any undisclosed COI. COIs were categorized as academic or as consulting-related, ownership-related, or salary-related financial. Main outcome(s) and measure(s) Our primary outcome was frequency of COI by category disclosed in trials and discovered by external search. Strength of conclusion and type of funding source were also summarized based on presence of COI. Results Of 3488 trials, 169 (806 unique authors) were included. Of those authors, 600 (74.4%) had verifiable online information. Disclosures were compared at the study level (946 author listings). Academic interests were disclosed in 80 (8.5%) listings, but 151 (16.0%) were found by external search. Consulting-related and ownership-related financial interests were each disclosed in 20 (2.1%) listings, but our external search identified 115 (12.2%) and 89 (9.4%), respectively. Salary-related financial interests were declared in 709 (75.0%) listings and identified in 573 (60.6%) by external search. Conclusions Our review suggests a lack of transparency with COI reporting in manual therapy research, with authors disclosing nonsalary COIs only about 20-50% as often as those identified via external search. Relevance Transparency of COI in manual therapy trials for LBP is lacking. Efforts to improve reporting are warranted.