Susan L Johnson1, Bahman P Tabaei, William H Herman. 1. Department of Internal Medicine, Michigan Diabetes Research and Training Center, University of Michigan Health System, 1500 E. Medical Center Dr., 3920 Taubman Center, Ann Arbor, MI 48109-0354, USA.
Abstract
OBJECTIVE: To simulate the outcomes of alternative strategies for screening the U.S. population 45-74 years of age for type 2 diabetes. RESEARCH DESIGN AND METHODS: We simulated screening with random plasma glucose (RPG) and cut points of 100, 130, and 160 mg/dl and a multivariate equation including RPG and other variables. Over 15 years, we simulated screening at intervals of 1, 3, and 5 years. All positive screening tests were followed by a diagnostic fasting plasma glucose or an oral glucose tolerance test. Outcomes include the numbers of false-negative, true-positive, and false-positive screening tests and the direct and indirect costs. RESULTS: At year 15, screening every 3 years with an RPG cut point of 100 mg/dl left 0.2 million false negatives, an RPG of 130 mg/dl or the equation left 1.3 million false negatives, and an RPG of 160 mg/dl left 2.8 million false negatives. Over 15 years, the absolute difference between the most sensitive and most specific screening strategy was 4.5 million true positives and 476 million false-positives. Strategies using RPG cut points of 130 mg/dl or the multivariate equation every 3 years identified 17.3 million true positives; however, the equation identified fewer false-positives. The total cost of the most sensitive screening strategy was $42.7 billion and that of the most specific strategy was $6.9 billion. CONCLUSIONS: Screening for type 2 diabetes every 3 years with an RPG cut point of 130 mg/dl or the multivariate equation provides good yield and minimizes false-positive screening tests and costs.
OBJECTIVE: To simulate the outcomes of alternative strategies for screening the U.S. population 45-74 years of age for type 2 diabetes. RESEARCH DESIGN AND METHODS: We simulated screening with random plasma glucose (RPG) and cut points of 100, 130, and 160 mg/dl and a multivariate equation including RPG and other variables. Over 15 years, we simulated screening at intervals of 1, 3, and 5 years. All positive screening tests were followed by a diagnostic fasting plasma glucose or an oral glucose tolerance test. Outcomes include the numbers of false-negative, true-positive, and false-positive screening tests and the direct and indirect costs. RESULTS: At year 15, screening every 3 years with an RPG cut point of 100 mg/dl left 0.2 million false negatives, an RPG of 130 mg/dl or the equation left 1.3 million false negatives, and an RPG of 160 mg/dl left 2.8 million false negatives. Over 15 years, the absolute difference between the most sensitive and most specific screening strategy was 4.5 million true positives and 476 million false-positives. Strategies using RPG cut points of 130 mg/dl or the multivariate equation every 3 years identified 17.3 million true positives; however, the equation identified fewer false-positives. The total cost of the most sensitive screening strategy was $42.7 billion and that of the most specific strategy was $6.9 billion. CONCLUSIONS: Screening for type 2 diabetes every 3 years with an RPG cut point of 130 mg/dl or the multivariate equation provides good yield and minimizes false-positive screening tests and costs.
Authors: Deborah J Wexler; David M Nathan; Richard W Grant; Susan Regan; Allison L Van Leuvan; Enrico Cagliero Journal: J Clin Endocrinol Metab Date: 2008-08-12 Impact factor: 5.958