14. Guideline Development Methodology
The National Heart, Lung, and Blood Institute’s (NHLBI) Obesity Education Initiative, in cooperation with the National Institute of Diabetes and Digestive and Kidney Diseases, convened the Expert Panel on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults.
These guidelines address the treatment of overweight and obesity only in adults, but it is the judgment of the panel that guidelines for treating obesity in children are equally important and should be drafted as soon as possible.
The panel’s charge was to develop evidence-based clinical guidelines for primary care practitioners; however, the guidelines should also be useful for certain specialists. The decision to develop “evidence based” guidelines was based on the increased attention being paid to clinical practice guidelines from methodologists, professional associations, third-party payers and policy makers, and the NHLBI’s mission to analyze research results with the goal of providing information that may enable health care practitioners to enhance their ability to detect, treat, and prevent disease.
In keeping with this approach, the panel defined topics to be included in the guidelines; developed an evidence model depicting the strategy of inquiry for each area of scientific interest; established criteria for searching and abstracting the literature; constructed and reviewed evidence tables of individual studies and summary tables of studies falling within a specific category of evidence; and identified the level or strength of the evidence that served as the basis for the recommendations.
The guidelines are based primarily on a systematic review of the published scientific literature in English found in MEDLINE from January 1980 through September 1997. This was done in the interest of time and economy. This information was supplemented by material provided by the panel and an ancestral search of appropriate references in eligible articles. The literature was searched and systematically reviewed by
- Establishing a priori eligibility criteria for inclusion of studies;
- Reviewing titles and abstracts to select promising articles;
- Reviewing these full articles; and
- Compiling evidence tables summarizing those articles that met the inclusion criteria.
As a priority, the panel identified randomized controlled trials (RCTs) as the strongest level of evidence for the evaluation of treatment efficacy.
Only RCTs lasting 4 months or more were considered.
The only exceptions were a few 3-month trials in the diet and pharmacotherapy sections. With the assistance of the San Antonio Cochrane Center*, 394 publications of RCTs were reviewed for data abstraction. RCT evidence serves as the basis for the many recommendations contained in these guidelines related to treatment efficacy. Instances when the panel had to make recommendations where RCTs were insufficient or absent are clearly indicated in the text. These instances most often pertain to issues of obesity assessment, classification, and measurement where RCT evidence would not be appropriate to answer the question. These issues are best addressed by epidemiological/observational studies of large population groups. In those few cases where the literature could not credibly support a recommendation, panel members relied on clinical experience and knowledge.
The panel recognizes that by relying primarily on only published literature, a publication bias (a positive result is more likely to be published than is a negative result) may exist, so that treatment efficacy may be overstated.
However, no other reliable sources of information were available.
The targeted population for the guidelines is all overweight and obese adults (18 years of age and older) with a BMI 25, with particular emphasis on those with cardiovascular risk factors.
While the guidelines are appropriate for patients with a BMI 40, their care is often complicated and may require surgery. These guidelines are not intended for pregnant women. Excluded from the analysis were adults with pharmacologically induced obesity and those with specific genetic syndromes associated with overweight and obesity.
The selection of weight loss interventions to be considered was determined by the literature review. Namely, the panel considered any topic for which articles meeting inclusion criteria were found, including diet, physical activity, behavior therapy, pharmacological therapy, surgery, and combinations of these modalities. No clinical interventions were excluded at the outset.
However, the panel did not consider other interventions such as acupuncture or hypnosis, for which no randomized trial articles were available.
Clinical interventions to prevent further weight gain in individuals already overweight were also considered relevant.
The panel also evaluated population factors and clinical situations that might potentially influence the physiological, medical, behavioral, or socio-cultural context for obesity identification and treatment. Evidence on special populations and situations was provided from RCTs and non-RCTs when available, but in many cases such evidence was meager.
Population factors and clinical situations selected for special consideration for obesity classification and treatment were age, gender, race/ethnicity, socioeconomic status, pregnancy, eating disorders, sleep apnea, and extreme obesity (BMI 40), concurrent treatment of other major conditions (such as heart disease or diabetes), and treatment of obesity in conjunction with smoking cessation. When evidence on these special populations or clinical conditions was insufficient to meet standards for inclusion in the main text of the guidelines, relevant issues are identified for the user, and in some cases are cross-referenced to an appendix or discussed in sidebar text as a commentary.
The panel recognized the possibility of an advocacy “bias” due to the large number of panel members drawn from organizations with an advocacy role in the treatment of obesity. As a result, it was agreed to obtain formal external reviews of the document from professional societies, consumer groups and government agencies representing a wide spectrum of expertise and concern about obesity.
The format for those sections of the report based on the RCT evidence begins with an evidence statement followed by the rationale for that statement. At the end of a series of related evidence statements, a recommendation is given.
Each evidence statement (other than those with no available evidence) and each recommendation is categorized by a level of evidence (A through D) as described below.
Statements for which there is no available evidence are so indicated.
Category A: Evidence is from endpoints of well designed RCTs (or trials that depart only minimally from randomization) that provide a consistent pattern of findings in the population for which the recommendation is made. Category A therefore requires substantial numbers of studies involving substantial numbers of participants.
Category B: Evidence is from endpoints of intervention studies that include only a limited number of RCTs, post-hoc or subgroup analysis of RCTs, or meta-analysis of RCTs. In general, Category B pertains when few randomized trials exist, they are small in size, and the trial results are somewhat inconsistent, or the trials were undertaken in a population that differs from the target population of the recommendation.
Category C: Evidence is from outcomes of uncontrolled or nonrandomized trials or from observation studies.
Category D: Expert judgment is based on the panel’s synthesis of evidence from experimental research described in the literature and/or derived from the consensus of panel members based on clinical experience or knowledge that does not meet the above-listed criteria. This category is used only in cases where the provisions of some guidance was deemed valuable but an adequately compelling clinical literature addressing the subject of the recommendation was deemed insufficient to justify placement in one of the other categories (A through C).
In applying these guidelines, the reader should note some caveats:
- The emphasis of these guidelines was to identify effective interventions, not to rank-order
- Them in terms of their efficacy or effectiveness.
- The panel chose not to emphasize comparisons among interventions because there were few studies that compared long-term outcomes. Also, since individual preferences and circumstances often dictate choice of therapy, the panel wished to present a menu of options rather than a ranked list of choices.
- When no evidence was available on the efficacy of a treatment, the panel usually rendered no opinion. An absence of studies should not be confused with an absence of effect. While clinicians may prefer to use proven therapies rather than untested ones, the lack of testing does not denote that the untested therapy does not work.
- The limitations of RCTs must be kept in mind. The RCT is the primary method for demonstrating efficacy. Often, participants enrolled in RCTs differ from the individuals in a primary care practice, and effectiveness in the community may differ from efficacy as measured in an RCT.
- The potential exists for misinterpretation of clinical trial results. Analysis of endpoints not specified at the outset, or post hoc or subgroup analysis, should be viewed as hypothesis-generating rather than hypothesis-testing.
* The San Antonio Cochrane Center is one of 12 centers around the world that comprise the Cochrane Collaboration. The Cochrane Collaboration is an international organization established in 1993 whose mission is to prepare, maintain, and disseminate systematic reviews and meta-analyses of health care interventions.
Next: Statement of Assumptions