1. Update on the Methods of the U.S. Preventive Services Task Force: Methods for Understanding Certainty and Net Benefit When Making Recommendations
- Author
-
Tracy Wolff, Michael L. LeFevre, Russell Harris, Chien-Wen Tseng, Daniel E Jonas, Carol M. Mangione, Alex H. Krist, Alex R. Kemper, and David C. Grossman
- Subjects
Epidemiology ,Service delivery framework ,media_common.quotation_subject ,Advisory Committees ,Population ,Primary care ,Outcome and Process Assessment ,Medical and Health Sciences ,01 natural sciences ,Education ,Indirect evidence ,03 medical and health sciences ,0302 clinical medicine ,Preventive Health Services ,Humans ,030212 general & internal medicine ,0101 mathematics ,Set (psychology) ,education ,media_common ,education.field_of_study ,Evidence-Based Medicine ,Actuarial science ,Task force ,010102 general mathematics ,Public Health, Environmental and Occupational Health ,Certainty ,Outcome and Process Assessment (Health Care) ,United States ,Health Care ,Outcome and Process Assessment, Health Care ,Expert opinion ,Public Health ,Psychology - Abstract
Since the 1980s, the U.S. Preventive Services Task Force (USPSTF) has developed and used rigorous methods to make evidence-based recommendations about preventive services to promote health and well-being for all Americans. Recommendations are based on the evidence of magnitude of net benefit (benefits minus harms). Expert opinion is not substituted when evidence is lacking. Evidence gaps are common. Few preventive services are supported by high-quality studies that directly and comprehensively determine the overall magnitude of benefits and harms in the same study. When assessing the body of evidence, studies may not have been conducted in primary care settings, studies may not have sufficiently included populations of interest, and long-term outcomes may not have been directly assessed. When direct evidence is not available, the USPSTF uses the methodologies of applicability to determine whether evidence can be generalized to an asymptomatic primary care population; coherence to link bodies of evidence and create an indirect evidence pathway; extrapolation to make inferences across the indirect evidence pathway, extend evidence to populations not specifically studied, consider service delivery intervals, and infer long-term outcomes; and conceptual bounding to set theoretical lower or upper limits for plausible benefits or harms. The USPSTF extends the evidence only so far as to maintain at least moderate certainty that its findings are preserved. This manuscript details with examples of how the USPSTF uses these methods to make recommendations that truly reflect the evidence.
- Published
- 2018
- Full Text
- View/download PDF