Intraobserver and Interobserver Reliability of Waist Circumference and the Waist‐to‐Hip Ratio

Abstract

Abdominal obesity correlates with numerous risk factors for cardiovascular disease and type 2 diabetes (1). It has been suggested that waist circumference (WC) is a better predictor of metabolic risk factors than the waist-to-hip ratio (WHR) (2, 3). Due to the importance of abdominal obesity in clinical practice, we evaluated the intra- and interobserver reliability of the WC measurement and compared it with the WHR in men and women. Participants were recruited from hospital volunteers, staff members, and students, all of whom provided institution-approved informed consent. Those younger than 19 years of age or with increased abdominal volume unrelated to increased adiposity were excluded. Participants (n = 218,121 women) were given two assessments within 1 week of each other: WC, hip circumference (HC), weight, and height were measured by observer 1, and WC and HC were reassessed by observer 1 and a second observer (observer 2). Both observers were blinded to the previous measures. WC was measured over the skin at the point of maximal narrowing from the anterior view (or at the 12th lateral rib when this was not obvious). HC was measured at the maximal gluteal protuberance from the lateral view over undergarments. Both WC and HC were measured to the nearest 0.1 cm. WHR was calculated as WC divided by HC. Weight and height were measured using standard methods. Intraclass correlation coefficients (ICCs) were calculated to determine reliability. The intraobserver reliability was evaluated between the results of the two assessments by observer 1. The interobserver reliability was assessed in 75 participants and evaluated between the results of the first assessment of observer 1 and the second assessment of observer 2. WC ICCs were compared with the WHR ICCs using the extension of Feldt's approach as described by Kraemer (4). The mean age and body mass index of the participants were 39 ± 11.9 years and 24.6 ± 4.4 kg/m2, respectively. WC and WHR from assessment 1 were 80.4 ± 10.7 cm (range, 57.2 to 119.4 cm) and 0.82 ± 0.08 (range, 0.58 to 1.04), respectively. The intraobserver ICC for WC was 0.987 (95% confidence interval [CI]: 0.983 to 0.990) and for WHR was 0.970 (95% CI: 0.960 to 0.977). The interobserver ICC for WC was 0.988 (95% CI: 0.982 to 0.993) and for WHR was 0.969 (95% CI: 0.951 to 0.980). Although WC and WHR demonstrated high intra- and interobserver reliability, WC was significantly more reliable than the WHR for both analyses (p = 0.0001, WC ICCs compared with WHR ICCs). This is the first report to compare both intra- and interobserver reliability between WC and WHR in such a large cohort. Our results confirm those reported in a smaller cohort (n = 51) in which the intraobserver WC ICC was greater than that of the WHR; however, the authors did not report the interobserver reliability (5). We demonstrated that WC also has a greater interobserver reliability than does the WHR. Because WC is more reliable than the WHR and a better predictor of cardiovascular risk factors, it is the preferable anthropometric tool for assessing patients and is recommended for routine use in physician offices.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 72 citations exist upstream.