Correspondence

Age, Body-Mass Index, and Mortality

N Engl J Med 1998; 338:1158-1159April 16, 1998

Article

To the Editor:

Stevens et al. (Jan. 1 issue)1 present 12-year mortality in relation to body characteristics in a large cohort of Americans. The authors mention but do not discuss the implications of known errors of self-reporting of height and weight. Such errors are likely to be magnified when the questioner is not present. Of greater importance is the known error of the understating of body weight by heavier people.2 The groups with above-average body-mass indexes (calculated as the weight in kilograms divided by the square of the height in meters) in this study almost surely included some people with higher values, resulting in an overstated risk of mortality in these groups. In addition, the extent of such understating differs with age,2 further confusing the interpretation of these results.

Despite its wide use, body-mass index cannot be considered to represent some ideal and universally applicable description of body composition. Body-mass index varies inversely with height. Therefore, shorter people with the same relative body composition as taller people would appear to have greater mass simply on the basis of the calculation of body-mass index. Since the subjects' heights are not given, one cannot evaluate these effects for the various groups in this study. Clinicians will be further confused by use of the term “excess adiposity” as a synonym for “increased body-mass index.” It has long been known that excess body mass, derived solely from height and weight, may represent only fat, only lean tissue, or any combination of the two.3,4

Gerson T. Lesser, M.D.
Jewish Home and Hospital

Richard N. Pierson, Jr., M.D.
St. Luke's–Roosevelt Hospital, New York, NY 10025

4 References

1. 1

   Stevens J, Cai J, Pamuk ER, Williamson DF, Thun MJ, Wood JL. The effect of age on the association between body-mass index and mortality. N Engl J Med 1998;338:1-7
   Full Text | Web of Science | Medline

2. 2

   Rowland ML. Self-reported weight and height. Am J Clin Nutr 1990;52:1125-1133
   Web of Science | Medline

3. 3

   Van Itallie TB, Pierson RN Jr. Bridging the gap between laboratory and clinic. In: Pierson RN Jr, ed. Body composition in the third millennium. New York: Springer-Verlag (in press).
   

4. 4

   Lesser GT, Deutsch S, Markofsky J. Use of independent measurement of body fat to evaluate overweight and underweight. Metabolism 1971;20:792-804
   CrossRef | Web of Science | Medline

To the Editor:

The report of the effect of age on the association between body-mass index and mortality was interesting but somewhat misleading, because the relative risk of being overweight as related to death from all causes and death from cardiovascular causes was reported instead of the absolute risk. This could lead to the conclusion that people between the ages of 30 and 54 years have a higher risk of dying if they are overweight than do those 55 or older. Although this may be true relative to their peers, the people with the highest risk of death due to being overweight are most likely the older people. Therefore, from a public health perspective, these older age groups should be the ones targeted for long-term weight reduction.

Gene R. Pesola, M.D., M.P.H.
Columbia University School of Public Health, New York, NY 10032

To the Editor:

Stevens et al. neglected an important point in arriving at their conclusion that the association between body-mass index and mortality weakens with increasing age, particularly after the age of 74 years. The more probable explanation is rather that obese people are more likely to have died by the age of 74 because of their shortened life span. Therefore, the increased risk of the high body-mass index is less evident at older ages because so many of this cohort have already died.

Rita F. Redberg, M.D.
University of California, San Francisco, San Francisco, CA 94143-0214

Author/Editor Response

The authors reply:

To the Editor: It is well known that the use of body-mass index does not provide an ideal description of body composition. Nevertheless, it does give a reasonable estimate of the percentage of body fat (correlation, approximately 0.7), and in groups, high values indicate excess adiposity (albeit accompanied by increased lean body mass). The body-mass index was slightly negatively correlated with height in our data. We were not concerned by this finding, given that the percentage of body fat is also slightly negatively correlated with height.1,2 The addition of height to our models examining associations between body-mass index and mortality had very little effect on our results.

We do not agree with Dr. Pesola that the contrast in the absolute risk among the age groups that we studied indicated that older rather than younger people should be targeted for long-term weight reduction. It is true that, after 12 years of follow-up, the percentage of obese people who died was much higher in the older than in the younger groups. However, with sufficient follow-up the percentage who died would be identical among all age groups. Overweight young adults who survived 12 years of follow-up would most likely still be overweight and at elevated risk at the end of 12 years, because body weight tends to change little during adulthood. Long-term reduction of excess body weight in young adults implies that a leaner body weight would be maintained into old age. This reduction in weight would, one hopes, be accompanied by a lifelong reduction in risk. In addition, age-associated differences in the effect of death on the number of years of life lost and disability-adjusted life-years 3 lead us to believe that it is incorrect to conclude that programs for long-term weight reduction should be targeted to older rather than younger people.

It is true that obese people are more likely to have died by the age of 74; however, in our analysis, there were 1307 women and 293 men over the age of 74 with body-mass indexes of 27 or greater. We had adequate power to detect relatively small increases in the obesity-associated risk in the older groups (relative risk, 1.12 for women and 1.23 for men). Obese people who survived to the age of 75 may have been resistant to the effects of obesity.

June Stevens, Ph.D.
Jianwen Cai, Ph.D.
Joy L. Wood, M.S.
University of North Carolina, Chapel Hill, NC 27599-7400

3 References

1. 1

   Smalley KJ, Knerr AN, Kendrick ZV, Colliver JA, Owen OE. Reassessment of body mass indices. Am J Clin Nutr 1990;52:405-408
   Web of Science | Medline

2. 2

   Deurenberg P, Weststrate JA, Seidell JC. Body mass index as a measure of body fatness: age- and sex-specific prediction formulas. Br J Nutr 1991;65:105-114
   CrossRef | Web of Science | Medline

3. 3

   Murray CJL, Lopez AD, eds. The global burden of disease. Vol. 1 of Global burden of disease and injury series. Cambridge, Mass.: Harvard University Press, 1996.
   

Citing Articles (2)

Citing Articles

1. 1

   (2005) Clinical Q & A. Obesity Management 1:5, 207-208
   CrossRef

2. 2

   C. C. Wee, R. S. Phillips, A. T. R. Legedza, R. B. Davis, J. R. Soukup, G. A. Colditz, M. B. Hamel. (2005) Health Care Expenditures Associated With Overweight and Obesity Among US Adults: Importance of Age and Race. American Journal of Public Health 95:1, 159-165
   CrossRef