Waist to Stature Ratio is More Strongly Associated with
Cardiovascular Risk Factors than Other Simple Anthropometric Indices
, TAI-HING LAM,
, AND EDWARD D JANUS,
, AND FOR THE
HONG KONG CARDIOVASCULAR RISK FACTOR PREVALENCE STUDY STEERING
PURPOSE: To determine which is the best anthropometric index among body mass index (BMI), waist
circumference (WC), waist to hip ratio (WHR) and waist to stature ratio (WSR) in relation to cardiovascu-
lar risk factors.
METHODS: A representative sample of 2895 Hong Kong Chinese aged 25 to 74 years received medical
examinations in 1995 and 1996. Anthropometric indices and cardiovascular risk factors in blood were
measured, and partial correlation and Receiver Operator Characteristic (ROC) curves were used in analysis.
RESULTS: Among 11 cardiovascular risk factors in partial correlation analysis, including ties WSR had
the highest r in 6 in men, and 5 in women; followed by WC with 4 in men and 6 in women. In ROC
analyses of 21 risk factors and health conditions, the area under curve (AUC) of WSR was the largest
for most (13 of 21) factors in men and 10 in women; followed by WHR with 14 in women but only 5
in men. The optimal WSR cutoff value was 0.48 for both men and women.
CONCLUSION: WSR is the best simple anthropometric index in predicting a wide range of cardiovascu-
lar risk factors and related health conditions. A simple message that one’s waist circumference should
not exceed half the stature is recommended for the public.
Ann Epidemiol 2003;13:683–691. ą 2003 Elsevier Inc. All rights reserved.
: Anthropometry, Cardiovascular Risk Factors, Cross-sectional Studies, Hong Kong, Obesity.
Obesity is an important risk factor for cardiovascular diseases
and diabetes mellitus and various anthropometric indices of
obesity have been suggested to predict their risks. These
indices can be classified by the type of obesity that they
were used to measure. Body mass index (BMI) (weight in
Kg/square of height in meters) measures overall obesity,
skin-fold at different body parts assesses regional obesity,
while waist circumference (WC), waist to hip ratio (WHR),
conicity index (WC/Ί
Weight(kg)/Height(m)) waist to stature
ratio (WSR), abdominal sagittal diameter (ASD) and vis-
ceral adipose tissue area measured by computer tomography
all indicate abdominal obesity (1–7).
Of these, BMI is perhaps the index most thoroughly
studied and its relation with cardiovascular risk factors
and outcomes have been well elucidated by cross-sectional
From the Department of Community Medicine, Clinical Biochemistry
Unit, University of Hong Kong Medical Centre, Pokfulam, Hong Kong
(S.-Y.H, T.-H.L.); and Wimmera Base Hospital, Baillie Street, Horsham,
VIC 3400, Australia (E.D.J.).
Address correspondence to: Prof. T.-H. Lam, Department of Community
Medicine, University of Hong Kong Medical Centre, 21 Sassoon Road,
Pokfulam, Hong Kong. Tel.: (852) 2819-9287; Fax: (852) 2855-9528.
Received March 18, 2002; accepted March 31, 2003.
ą 2003 Elsevier Inc. All rights reserved. 1047-2797/03/$–see front matter
360 Park Avenue South, New York, NY 10010 doi:10.1016/S1047-2797(03)00067-X
and prospective studies (1, 8). However, there are increasing
doubts about the appropriateness of this overall obesity
index in predicting cardiovascular events, and evidence is
mounting for indices of abdominal obesity such as WC,
WHR, WSR, and ASD, and new indices are being suggested
from time to time (7).
Obesity is increasing in all countries with higher levels of
socio-economic development but the debate about the most
valid index remains unresolved. Little is known about how
the public assess their body size and shape. For any obesity
index to be accepted by the public and hence to make an
impact on health risk behavior, a few prerequisite criteria
are necessary. Obviously it has to be simple and nothing
more complicated than BMI is likely to work. It should be
easily self-measurable and perceived to be a useful indicator
of obesity and predictor of related problems.
Only BMI, WC, WHR, and WSR can satisfy these re-
quirements as other indices are either too complicated or
difficult to measure. The association of each of these indices
with cardiovascular risk factors has been demonstrated by
previous studies mostly based on Caucasian populations.
The few existing local studies on Chinese were based on
relatively small and non-representative samples (9, 10).No
study has examined which of the four indices is most closely
associated with the common cardiovascular risk factors in
Chinese. The present study aims to compare the four indices
in a representative Hong Kong Chinese population to deter-
mine which is the best.