Outcome reporting in bariatric surgery: an in‐depth analysis to inform the development of a core outcome set, the BARIACT Study. Issue 1 (30th November 2014)

• Record Type: Journal Article
• Authors: Hopkins, J. C.; Howes, N.; Chalmers, K.; Savovic, J.; Whale, K.; Coulman, K. D.; Welbourn, R.; Whistance, R. N.; Andrews, R. C.; Byrne, J. P.; Mahon, D.; Blazeby, J. M.; By‐Band Trial Management Group
• Is Part Of: Obesity reviews. Volume 16:Issue 1(2015)
• Page Start: 88
• Page End: 106
• Publication Date: 2014-11-30
• Subjects: Obesity -- Periodicals; 616.398005
• DOI: 10.1111/obr.12240
• Languages: English
• ISSNs: 1467-7881

Abstract

<abstract abstract-type="main"> <title>Summary</title> <p>Outcome reporting in bariatric surgery needs a core outcome set (COS), an agreed minimum set of outcomes reported in all studies of a particular condition. The aim of this study was to summarize outcome reporting in bariatric surgery to inform the development of a COS. Outcomes reported in randomized controlled trials (RCTs) and large non‐randomized studies identified by a systematic review were listed verbatim and categorized into domains, scrutinizing the frequency of outcome reporting and uniformity of definitions. Ninety studies (39 RCTs) identified 1, 088 separate outcomes, grouped into nine domains with most (<italic>n</italic> = 920, 85%) reported only once. The largest outcome domain was 'surgical complications', and overall, 42% of outcomes corresponded to a theme of 'adverse events'. Only a quarter of outcomes were defined, and where provided definitions, which were often contradictory. Percentage of excess weight loss was the main study outcome in 49 studies, but nearly 40% of weight loss outcomes were heterogeneous, thus not comparable. Outcomes of diverse bariatric operations focus largely on adverse events. Reporting is inconsistent and ill‐defined, limiting interpretation and comparison of published studies. Thus, we propose and are developing a COS for the surgical treatment of severe and complex obesity.</p> </abstract>