Two women, in their 70s, are walking on Blackstone Boulevard. Both are 5 feet 2 inches tall. One weighs 146 pounds; the other, 118 pounds. Corresponding body mass index (BMI) is 26.5 (overweight) for the first and 21.6 (normal) for the second. They are both healthy without any chronic medical conditions. Each takes vitamins and calcium supplementation. Which woman has the lower mortality risk? Many would choose the woman with the normal BMI. The standard weight recommendations support this

more »

... ion; these guidelines suggest that a BMI of 25 is the upper limit of ideal weight for all adults. 1,2 Recent evidence, however, suggests that in spite of the increased risk for diabetes and hypertension in those who are overweight, BMI >25 may be protective with regard to mortality. (Table 1) A study published in February 2010 supported the findings of other recent work indicating that older adults who are overweight by BMI calculations are at reduced mortality risk, compared with those who have a lower BMI. 3 Flicker et al analyzed a cohort of adults (4,677 men and 4,563 women) aged 70 to 75 years old who were followed for 10 years or until death if sooner. 3 With regards to gender differences, the major differences were that more men were married (more than 80% men and 54% women) and more men reported a history of bronchitis or emphysema. The lowest all-cause mortality risk was at a BMI of 26.6 kg/m2 in men and 26.26 kg/m2 in women and the risk of death increased as BMI decreased. 2 The degree of medical illness or co-morbidity did not alter the relationship between BMI and mortality risk, even though mortality risk increased with degree of illness. Degree of physical activity also reduced risk for mortality, but again the lowest risk of mortality was seen in those who were classified as overweight, regardless of degree of physical activity. 3 The authors identified limitations; e.g., they catalogued height and weight only once during the study period, and they did not record fluctuations in participants' weight. 3 However, the results coincided with other recent studies and reviews. One, published in 2007, looked at 32 observational studies. 4 The systematic review and meta-analysis suggested that a BMI in the overweight range is not associated with increased mortality in elderly men and women. The relative risk (RR) of all-cause mortality was 1.00 (95% confidence intervals, 0.97-1.13). 4 Only a modest increase in mortality was noted for those in the obese category RR 1.10 (95% confidence intervals, 1.06-1.13). These results were similar to those reported in an earlier systematic review published in 2001. 5 Published articles that examined the relationship between BMI and all-cause mortality and cardiovascular or coronary heart disease mortality in adults over 65 years of age were reviewed. The review included 13 studies and demonstrated several uniform findings. First, most studies failed to show a significant association between higher BMI and increased mortality. 5 Also, the association tended to be U-shaped: lower BMIs and those on the higher end were more closely associated with higher mortality. However, the higher BMIs that were associated with higher mortality were in the obese category; again implying that being "overweight," in contrast to "obese," is not associated with higher mortality. 5 The authors did identify limitations that could affect the results, including lack of control for smoking, lack of identification of underlying disease in the underweight participants, and lack of control of some weight-related conditions, such as hypertension, diabetes and dyslipidemia. 5 They chose studies that controlled for these factors for the review. The review, similar to the others discussed, demonstrated that low BMI is more consistently associated with higher mortality than high BMI in older adults. A major limitation in the interpretation of these results is that the BMI measurement does not take into account body composition. As people age, the amount of body fat mass increases and fat free mass, mostly muscle, decreases. However, these studies suggest that perhaps having excess body fat in older age is not as dangerous as in young age. The excess body fat may have less of an effect on mortality in older age. 4 It appears that being underweight is more closely associated with mortality. The other alternative consideration is that the BMI does not take the fat distribution into account, and waist circumference, a better measure of fatness in older persons, would correlate with increased mortality in older adults. Further study using waist circumference and other measures are needed. It has also been hypothesized that there is a selection bias in these patient populations. Perhaps those who were going to die of Table 1. BMI Weight Category < 18.5 Underweight 18.5 -24.9 Normal weight 25 -29.9 Over weight > 30 Obese BMI (kg/m2) = weight in kilograms/height in meters 2