It’s the Fast Food, Stupid!

Abstract & Commentary

Synopsis: The optimum body mass index (BMI) for adults is probably 18.5 to 21.9 kg/m2.

Source: Field AE, et al. Arch Intern Med. 2001;161:1581-1586.

This was a prospective 10-year follow-up of 77,690 wo-men from the Nurses’ Health Study and 46,060 men from the Health Professionals Follow-up Study. Both are ongoing studies. The women were enrolled starting in 1976, and were 30 to 50 years of age at the time of enrollment. They are sent questionnaires every other year; the questionnaires include queries about diet, physical activity, smoking, contraception, and medical diagnoses. Field and colleagues also obtain medical records (with permission) for diseases of particular interest, follow-up with relatives in the event of death, and send supplementary questionnaires to subjects who report diabetes. The men were enrolled starting in 1986, and were 40 to 75 years of age when the study began. Data collection in the Health Professionals study is similar to that of the Nurse’s Health Study.

For the current analysis, Field et al used logistic regression to correlate body mass index (BMI) with high cholesterol levels, hypertension, gallstones, type 2 diabetes, heart disease, stroke, and colon cancer. They controlled for age, smoking, and ethnicity. Subjects who reported being diagnosed with cancer before 1986, those who smoked, were missing data, or who had BMIs of less than 18.5 were excluded from analysis.

For both men and women, the risk of developing diabetes, gallstones, hypertension, heart disease, and stroke increased with the severity of overweight. Being overweight was a particularly strong risk factor for diabetes (duh). When those with a BMI between 18.5 and 21.9 were used as a reference group, men and women with a BMI of 22-24.9 were found to be at increased risk of diabetes, hypertension, and high cholesterol level. Women were at increased risk for gallstones and heart disease, and men for colon cancer. Further, both men and women who had BMI’s between 25 and 29 were at significantly increased risk of developing all conditions reported except for stroke.

Comment by Barbara A. Phillips, MD, MSPH

This study clearly demonstrates that obesity not only kills, but also maims. What is new about this paper (besides the excellent methods and large sample size) is the association of morbidity with BMI levels previously believed to be "safe." Field et al point out that in studies correlating weight with health, the higher the BMI is for the reference group, the lower the risks of obesity will appear to be. This is because some people who are actually at risk will be placed in the reference group. The US Department of Agriculture has recently suggested an overweight cutoff of a BMI of 25, instead of 27. Field et al support that proposal with this data.

This is by no means the first paper to point out the risks of moderate obesity. Among the health hazards reported here, overweight has previously been shown to be a risk factor for diabetes,1,2 hypertension,3,4 dyslipidemia,5 coronary heart disease,6 stroke,7 and gallbladder disease.8 In addition to the risks studied and reported here, obesity has been linked to osteoarthritis,9 obstructive sleep apnea,10 breast cancer,11,12 endometrial cancer,12 and prostate cancer.13

We are the most obese nation on the planet. We are also the home of McDonald’s, Coca-Cola, and jumbo portions. All-cause mortality rises above a BMI of 25.14 I think we have to add fast food to the (long) list preventable risk factors for premature mortality.


1. Colditz GA, et al. Ann Intern Med. 1995;122:481-486.

2. Holbrook TL, Barrett-Conner E, Windard DL. Int J Obes. 1989;13:723-729.

3. Field AE, et al. Am J Epidemiol. 1999;150:573-579.

4. Folsom AR, et al. Int J Epidemiol. 1989;18:361-367.

5. Pi-Sunyer FX. Med Sci Sports Exerc. 1999;31(Suppl): S602-608.

6. Harris TB, et al. Am J Epdemiol. 1993;137:1318-1327.

7. Rexrode KM, et al. JAMA. 1997;277:1539-1545.

8. Stampfer MJ, et al. Am J Clin Nutr. 1992;55:652-658.

9. Cooper C, et al. Am J Epidemiol. 1998;147:516-522.

10. Peppard PE, et al. JAMA. 2000;284:3015-3021.

11. Suga K, et al. Jpn J Cancer Res. 2001;92:127-134.

12. Bergstrom A, et al. Int J Cancer. 2001;91:421-430.

13. Rodriguez C, et al. Cancer Epidemiol Biomarkers Prev. 2001;10:345-353.

14. Calle EE, et al. N Engl J Med. 1999;341:1097-1105.