Characteristics of clinical trials assessing antimicrobial treatment of bovine respiratory disease, 1970-2005

Abstract

Objective: To evaluate reporting of key study design features and study outcomes in trials of antimicrobial treatment of bovine respiratory disease (BRD) in North American feedlots. Design: Systematic review. Sample Population: 29 manuscripts (41 studies) reporting antimicrobial treatment of BRD in North American feedlot cattle. Procedures: A search of the electronic citation databases AGRICOLA, Commonwealth Agricultural Bureau, and PubMed was conducted to identify relevant manuscripts published between 1970 and 2005. Key study design features were extracted by 2 reviewers. Results: 12 of 29 (41%) manuscripts did not disclose a funding source, and 21 (72%) had an author clearly identified as an employee of a pharmaceutical company. At the study level, 36 of 41 (88%) studies reported a random method of treatment allocation, 9 (22%) described the method of allocation sequence generation, 20 (49%) reported that study investigators were blinded to treatment, and 3 (7%) included a study size justification. No studies described the null hypothesis to be tested. Thirty-seven (90%) studies reported at least 3 outcomes; the largest number of outcomes reported was 14. It was not possible to conduct the statistical analysis as originally planned because it was not possible to discern the primary outcome for the majority of studies. Conclusions and Clinical Relevance: Many studies did not report key study design features that would assist critical evaluation by readers. It was not clear whether the studies failed to use the design features or failed to report them. Several nondesign features, such as reporting of the null hypothesis, a primary outcome, and sample size rationale, represent relatively new standards for reporting; however, reporting these features would substantially clarify the study objective.