Premium
Recommended anthropometric cut‐offs for population screening of diabetes and prediabetes need to be evaluated in resource‐limited settings
Author(s) -
Yu Elaine A,
Finkelstein Julia L.,
Bonam Wesley,
Glesby Marshall J.,
Bran Patsy M.,
Mehta Saurabh
Publication year - 2016
Publication title -
the faseb journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.709
H-Index - 277
eISSN - 1530-6860
pISSN - 0892-6638
DOI - 10.1096/fasebj.30.1_supplement.33.1
Subject(s) - waist , medicine , bioelectrical impedance analysis , anthropometry , body mass index , diabetes mellitus , circumference , population , prediabetes , type 2 diabetes , demography , endocrinology , environmental health , geometry , mathematics , sociology
Background Globally, waist circumference (WC) and body mass index (BMI) are widely used for population screening to predict disease risk, including for diabetes. We examined the association of anthropometric measurements with diabetes and pre‐diabetes in a cohort where the majority of participants would be classified as low risk based on either BMI or WC. Methods Study participants were recruited from an outpatient clinic in rural South India (n=297). Concentrations of glycosylated hemoglobin (HbA1c) were measured by high‐performance liquid chromatography. Diabetes and pre‐diabetes were defined as HbA1c ≥6.5% and ≥5.7% to <6.5%, respectively. Waist circumference was measured by non‐stretch measuring tape to the nearest 0.1 cm by study staff. Whole and segmental body composition were measured by bioelectrical impedance analysis (8 electrode; Tanita BC 418 MA). Binomial regressions were used to assess the associations between diabetes (and pre‐diabetes), and anthropometric indicators (WC, BMI). Restricted cubic splines were used to evaluate non‐linearity. Results Mean waist circumference was 73.2 cm (SD: 11.3) among males, and 69.7 cm (SD 12.1) among females. Among male study participants, more than 90% had a waist circumference < 90 cm; among females, 80% had a waist circumference <80 cm. Mean BMI was 19.2 kg/m2 (SD 4.1) among men, and 20.4 kg/m2 (SD 5.4) among women. Among males, 50.0% had BMI <18.5 kg/m2, and over 90% had BMI <25.0 kg/m2. Among females, nearly 50% of study participants had BMI <18.5 kg/m2, and more than 80% had BMI <25.0 kg/m2. Study participants with waist circumference in the two lowest tertiles had a lower risk of diabetes (Tertile 1 vs 3: RR 0.40 [95% CI 0.20–0.81]; Tertile 2 vs 3: RR 0.33 [95% CI: 0.15–0.73]), compared to those in the highest tertile (p=0.004). Waist circumference was not associated with pre‐diabetes; BMI was not associated with either diabetes or pre‐diabetes. Conclusion Based on widely used cutoffs of WC and BMI (World Health Organization, International Diabetes Federation for South Asia), nearly 90% of study participants would be considered to be at low or minimal risk for diabetes. However 13% of study participants had diabetes, and 21% had pre‐diabetes. Waist circumference was still associated with increased risk of diabetes among study participants. These findings emphasize the need for the development of population‐specific cutoffs for waist circumference in diabetes screening. Support or Funding Information Cornell University