Evidence‐Based Recommendations for Best Practices in Weight Loss Surgery

Weight loss surgeries (WLSs)11 Nonstandard abbreviations: WLS, weight loss surgery; RCT, randomized controlled trial. grew 625% between the early 1990s and 2003 and are expected to increase another 14%, to over 140, 000 procedures, in 2004 ((1)). This growth belies the serious risks involved in undergoing WLS and heightens the potential for medical errors that could threaten the safety and well being of WLS patients. This issue of Obesity Research includes a series of evidence-based reports on recommendations for best practice in WLS. Findings in these reports are based on a comprehensive review of all available clinical trials and observational studies. They directly link patient safety ((2)) to methods for setting evidence-based guidelines based on peer-reviewed scientific publications ((3)). We believe that this approach is a valid route to best practice recommendations not only in the field of WLS, but in other fields of medicine as well. The genesis of these reports is a February 2004 request by Christine Ferguson, the Massachusetts Commissioner of Public Health, that the Betsy Lehman Center for Patient Safety and Medical Error Reduction (Lehman Center) convene an Expert Panel to study weight loss surgical programs and procedures as they directly relate to patient safety. The panel was charged with reviewing WLS operations, identifying potential safety issues, and recommending specific actions to reduce medical errors and improve patient outcomes. It used a state-of-the-art model of evidence-based medicine ((4), (5), (6)) to analyze current literature and develop best practice recommendations. The approach involved extensive electronic searches of MEDLINE and other pertinent databases for studies [e.g., systematic reviews, randomized controlled trials (RCTs), prospective cohort studies, case series reports] published from 1980 to 2004. It also involved use of a classification system developed from well-established models for classifying and grading the quality of evidence ((6), (7)). This methodology is considered state-of-the-art in efforts to synthesize research findings as a basis for practice guidelines and recommendations. Evidence-based medicine is designed to promote clinical decision-making based on the best available scientific evidence ((8)); its strongest findings come from well-designed and executed double-blinded RCTs because they minimize potential biases, especially confounding caused by nonrandomized allocation of subjects to treatment groups ((9), (10)). In the field of WLS, however, there are few well-conducted RCTs. Thus, in many instances, we had to rely on the next best evidence, i.e., prospective observational studies or population-based registries. Hill ((11)) developed criteria for establishing scientific causation in the absence of RCTs—broad guidelines that have served since 1965 as a model for ensuing standards developed to assess the quality of research evidence. In addition to developing guidelines for scientific causation, Hill also authored an influential textbook on medical statistics ((12)) and played a pivotal role in the evaluation of streptomycin in the treatment of tuberculosis ((13)). The results of that trial, published in 1948 ((13)), marked the start of the modern era of clinical research. Since that time, methodologies for summarizing, assessing, and judging the strength of scientific evidence have evolved from Hill's inference of etiology in a broad biological context to evidence-based medicine's current focus on efficacy of treatment ((14), (15)). To convert these criteria into clinical decisions, many international and national organizations including the World Health Organization, the NIH, the Agency for Healthcare Research and Quality, and the U.S. Preventive Services Task Force have developed grading systems for evidence-based recommendations. These grading systems all rank RCTs as the highest-level evidence. However, in many areas of medical practice, RCTs may not be practical or ethical to undertake, and, thus, one has to rely on other types of study as the best evidence. This is especially so in the field of WLS. For most topics we reviewed, including Pediatrics, Patient Selection, Patient Education/Informed Consent, Nursing, Coding/Reimbursement, Facilities, and Data Collection, there were few or no RCTs. In these areas, observational evidence from prospective studies, in conjunction with expert opinions, was given the greatest weight in making best practice recommendations. In developing evidence-based recommendations, we were cognizant of the pros and cons of different types of study design. Pocock and Elbourne ((16)) report that problems of heterogeneity and publication bias are relevant to all comparisons of evidence from RCTs and observational studies; that any study, whether randomized or observational, may have flaws in design or analysis that can cast doubt on the generalization of the results. They also note that the addition of observational data to RCTs enhances the quality of both approaches. Thus, evidence levels should be based not only on study design, but also on the methodological quality of individual studies ((17)). The methodology for the Lehman Center report specified that all selected studies be critically assessed for internal validity and ranked according to levels of evidence based on a grading system similar to that used by NIH and U.S. Preventive Services Task Force. For example, we used Category D evidence—including clinical experience, opinions of respected authorities, reports from expert committees, and the consensus of Expert Panel members—in conjunction with evidence from RCTs or observational studies to develop recommendations. Because study designs and outcomes were too dissimilar for pooled analysis, we relied primarily on narrative summaries for the literature review. A formal meta-analysis was not performed because of tremendous heterogeneity in surgical procedures performed, outcome measures, and length of follow-up ((18)). This report from the Expert Panel on WLS was requested and funded by the Massachusetts Department of Public Health's Betsy Lehman Center for Patient Safety and Medical Error Reduction to improve the safety and well-being of patients who undergo WLS in the state. Toward that end, >80 of the Commonwealth's obesity experts and health care professionals came together to produce its contents. This is the first time that such a group has carried out a comprehensive, in-depth, and systematic review of the medical literature related to WLS. We expect these recommendations to inform health care policy and clinical practice and have far-reaching effects on both state and national levels. In the absence of a requirement for new surgical procedures to undergo institutional review or the scrutiny of Investigational New Drug and New Drug Applications ((19)), this framework for evidence-based recommendations ((2)) may be applicable to most new developments in surgical procedures.