Pengelly and Morris (Scottish Medical Journal 2009) have reviewed the relative merits of measurements of body mass index and weight distribution.
It has been accepted for many years that being overweight or obese, as indicated by a body mass index (BM I) of 25 or over for the former and 30 or over for the latter, is associated with impairment of long term health and prognosis. The World Health Organisation (WHO) has indicated that, in Caucasians, waist measurements of 94cm or more in men, and 80cm or more in women have similar adverse effects on health, with increased risks at 102 cm or more in men and 88 cm in women. The role of waist-hip ratio (W/H) and whether it represents a better index than waist (W) measurement alone is being debated; many papers favour waist measurement alone. But two papers in 2005 discussing 27,098 subjects, 12,461 of whom had myocardial infarction and 14,637 controls, come down firmly in favour of W/H and were followed by a Lancet Editorial entitled Farewell to Body Mass Index? Life assurance companies at medical examination usually request height and weight measurements (and therefore BMI). Most ask for waist measurements and a few hip measurements in addition (and therefore W/H). The authors have reviewed the data in 816 consecutive subjects for life assurance examination in whom BMls, Ws and W/Hs were all recorded. In these the evidence supports the use of W as the best indicator of risk in men (634 cases), but not in the relatively small number of women (182 cases) in whom H appeared better. We believe that BMI, W and W/H should be recorded in every subject at life assurance examination so that the insurance companies in the long term will be able to reach valid conclusions about their individual and collective value.
Pengelly and Morris Body mass index and weight distribution . Scottish Medical Journal 2009 vol 54 p 17-21