An analysis of reporting quality of prospective studies examining community antibiotic use and resistance

Abstract

BackgroundAntibiotic resistance is a global problem, but the relationship between antibiotic use and resistance development and decay is not well understood. This knowledge is best provided by prospective studies, but to be useful they must be both conducted and reported well. Little is known about the reporting quality of these studies. This study aimed to assess the quality of reporting in prospective studies that investigated antibiotic resistance following antibiotic exposure in community-based individuals.MethodsThe quality of reporting of prospective studies (17 randomised trials, eight cohort studies) identified in a systematic review of the relationship between antibiotic use and resistance were assessed independently by two researchers using checklists (one for trials, one for cohort studies) developed from existing reporting guidelines for these designs and this field.ResultsThe mean percentage (SD, minimum-maximum) of mandatory items that were adequately described by the included studies was 59% for trials (14%, 36–84%) and 52% for cohort studies (17%, 13–70%). Most studies adequately described the study background and rationale, the type, combination, and duration of the antibiotic intervention, and the sampling procedures followed to isolate resistant bacteria. Most studies did not report the incident numbers of resistant and susceptible isolates analysed at each time point. Blinding and sample size calculation was inadequately reported in almost half of the trials and all cohort studies.ConclusionsThe quality of reporting in prospective studies investigating the association between antibiotic exposure in the community and isolation of resistance isolates is variable. Some details were missing in more than half of the studies, which precludes a complete risk of bias assessment and accurate interpretation and synthesis of results.