Analysis of the Quality of Prospective Randomized Controlled Trials for Treatment of Boxer’s Fractures

Abstract

BACKGROUND: This study aimed to review level I and II therapeutic studies on boxer's fractures to measure variation in quality among the highest level study designs. METHODS: We used quantitative measures of study quality to evaluate prospective randomized controlled trials (RCTs) of treatments of boxer's fractures. A search of PubMed, using terms "boxer's fracture" and "fifth metacarpal neck fracture" identified 164 articles from 1961 to 2019. From this list, we identified 6 RCTs. Two observers classified each trial according to 3 systems: the Oxford Levels of Evidence, the modified Coleman Methodology Score, and the revised Consolidated Standards of Reporting Trials (CONSORT) score. RESULTS: The 2 reviewers were consistent in their use of the Oxford Levels of Evidence (100% agreement). The differences between the average modified Coleman Methodology scores and the average CONSORT scores assigned by the 2 observers were not significant (46.2 vs 45.3 points, κ = 0) and (13.7 vs 14.3 points, κ = 0.33), respectively. Both observers rated all the studies as level I and as unsatisfactory according to the Coleman Methodology Score (100% and 100%), and less than half as unsatisfactory according to the CONSORT score (50% and 17%). Areas of deficiency included randomization, blinding, group comparability, clinical effect measurements, and allocation into treatment arms. CONCLUSION: Classifying orthopedic scientific reports according to the levels of evidence implies a degree of respect for level I and II studies that may not always be merited. Our data suggest that the quality of higher level studies, namely those involving boxer's fractures, varies and may often be unsatisfactory when critically evaluated.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 6 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.