The U.S. "Obesity Epidemic" and the Lasting Value of BMI
What's Hot in Medicine 2011
by David W. Sharp
The "red hot" research papers for 2010 listed in Science Watch’s March/April 2011 issue were dominated by genetics, which is no surprise perhaps, but the leader of the pack was a far cry from that largely laboratory world. Papers on exome and metagenomic sequencing, at second and third places, trailed well behind a study (now at #9 in the accompanying Medicine Top Ten) on trends in adult obesity from the U.S. National Center for Health Statistics (NCHS).
Slightly less warm, at seventh in that March/April year-end ranking and published in the same issue of JAMA, lies a companion paper covering children and adolescents (C.L. Ogden, et al., JAMA, 303[3]: 242-49, 2010; as yet outside the Top 25, with 24 citations this period and 70 overall). Articles published in January, as these were, have an advantage in annualized rankings; nonetheless, the attention given to these two public-health studies is striking.
Definitions of overweight and obesity have shifted over the years, but generally agreed criteria today for adults are a body-mass index (BMI) of 25.0 or more for overweight and of 30.0 or more for obesity. BMI has proved to be a very useful tool but it is not itself a direct measurement of adiposity.
" Katherine Flegal Discusses the Prevalence of Obesity in the US," in an interview from the Special Topic of Obesity (Sep. 2010).
The high-ranking NCHS paper (#9) indicates that roughly two-thirds of the U.S. adult population are overweight and half of those are obese. In an interview with Science Watch last year Dr. Katherine Flegal, whose published work was third in overall citations within the Special Topic of Obesity admitted that the first of her major papers on trends in adult obesity (in JAMA in 1994) had "caused quite a commotion."
Health professionals had not been thinking of obesity in epidemic terms. Just how devastating the epidemic might be was highlighted in December 2009, in an extrapolation suggesting that if trends continue, 45% of the population of the United States would, by the year 2020, be classified as obese (S.T. Stewart, et al., New Engl. J. Med.,361[23]: 2252-60, 2009). Recent public-health gains resulting from reductions in cigarette smoking were being negated by increases in body weight.
"...if trends continue" is the key qualification to otherwise gloomy predictions, which is where Flegal and colleagues’ paper (#9) comes into play. The U.S. National Health and Nutrition Survey (NHANES) collects measurements of height and weight, from which BMI is calculated.
The most recent data are from 2007-08 and hint at a plateau—a plateau at an uncomfortably high altitude, true, but not an inexorable rise. A similar, faintly encouraging pattern may be emerging for children and adolescents. Dr. Flegal and Dr. Cynthia Ogden share (with collaborators) a significant co-publication history in obesity, the latter having first authorship on studies of younger age-groups (an example being the companion paper in JAMA referred to above).
The distribution of BMI might be expected to be bell-shaped but it is not. Very high BMIs are over-represented, and smoothed curves provided in Flegal and colleagues’ paper show this. The median BMI for men shifted somewhat to the right for men aged 40 to 59 between 1999-2000 and 2007-08, but for women there was very little change.
"The BMI has been used in several very large collective studies of overweight and mortality."
BMI is simple to measure and was devised back in the 19th century. A scientometric analysis using Clarivate Web of Science® data uncovered no fewer than 63,845 articles on the index published between 1900 and 2008 (A.B. Bohlen, et al., Z Gastroenterol.,48[11]: 1285-92, 2010), yet for many years physicians and actuaries relied on weight-for-height charts and tables.
The BMI has been used in several very large collective studies of overweight and mortality. These reveal a J- or distorted U-shaped relationship. For example in the very latest of these to be published (March 26, 2011) overall mortality was found to be lowest at a BMI in the range 22.5 to 25.0; it was higher below 22.5, and above 25.0 there was an increase of about 30% in mortality for every 5 points rise in BMI (Emerging Risk Factors Collaboration. Lancet, 377[9771]: 1085-95, 2011). Furthermore, measures of central obesity such as waist-to-hip ratio did not add much to BMI alone as an indicator of future mortality.
Most of the big BMI/mortality studies have been done on North American and European populations. The pattern revealed by them is also found in east Asia but not, for higher BMIs, in India and Bangladesh (W. Zheng, et al., New Engl. J. Med.,364[8]: 719-29, 2011).
So, is BMI turning into a stand-alone risk factor, especially for cardiovascular disease? Apparently not. If body size is added to risk models that already include blood pressure, diabetes, and cholesterol, predictive accuracy is not improved. Nor can BMI measurement replace testing for total and high-density-lipoprotein cholesterol, as some had hoped (R.R. Huxley, D.R. Jacobs, Jr., Lancet, 377[9771]: 1051-52, 2011). We may not know exactly how BMI and fatal illnesses are related but the hope must be that some future NCHS or similar study will show that overweight and obesity rates really are falling at last.
A former deputy editor of The Lancet, David W. Sharp, M.A. (Cambridge), is a freelance writer living in Minchinhampton, Gloucestershire, U.K.
What's Hot in Medicine | |||
---|---|---|---|
Rank | Paper |
Cites This Period Nov-Dec 10 |
Rank Last Period Sep-Oct 10 |
1 | Novel Swine-Origin Influenza A (H1N1) Virus Investigation Team (F.S. Dawood, et al.), "Emergence of a novel swine-origin influenza A (H1N1) virus in humans," New Engl. J. Med., 360(25): 2605-15, 18 June 2009. [Writing group: Ctrs. for Disease Control & Prevent., Atlanta, GA] *458WR | 62 | 1 |
2 | F.H. Schröder, et al., "Screening and prostate-cancer mortality in a randomized European study," New Engl. J. Med., 360(13): 1320-8, 26 March 2009. [15 institutions worldwide] *423VP | 60 | 2 |
3 | NICE-SUGAR Study Investigators (S. Finfer, et al.), "Intensive versus conventional glucose control in critically ill patients," New Engl. J. Med., 360(13): 1283-97, 26 March 2009. [Writing Committee: 4 Australian, New Zealand, and Canadian institutions] *423VP | 52 | 4 |
4 | G.L. Andriole, et al., "Mortality results from a randomized prostate-cancer screening trial," New Engl. J. Med., 360(13): 1310-9, 26 March 2009. [16 U.S. and Canadian institutions] *423VP | 42 | † |
5 | R.J. Garten, et al., "Antigenic and genetic characteristics of swine-origin 2009 A(H1N1) influenza viruses circulating in humans," Science, 325(5937): 197-201, 10 July 2009. [26 institutions worldwide] *468FK | 42 | † |
6 | P.C. Fong, et al., "Inhibition of poly(ADP-ribose) polymerase in tumors from BRCA mutation carriers," New Engl. J. Med., 361(2): 123-34, 9 July 2009. [7 U.K.. and European institutions] *467OS | 39 | 3 |
7 | T.S. Mok, et al., "Gefitinib or carboplatin-paclitaxel in pulmonary adenocarcinoma," New Engl. J. Med., 361(10): 947-57, 3 September 2009. [14 institutions worldwide] *490EI | 38 | 6 |
8 | P.J. Turnbaugh, et al., "A core gut microbiome in obese and lean twins," Nature, 457(7228): 480-4, 22 January 2009. [7 U.S. and French institutions] *395JA | 38 | † |
9 | K.M. Flegal, et al., "Prevalence and trends in obesity among US adults,1999-2008" JAMA, 303(3): 235-41, 20 January 2010. [Ctrs. Disease Control & Prevent., Hyattsville, MD] *544VM | 38 | † |
10 | S.J. Connolly, et al., "Dabigatran versus warfarin in patients with atrial fibrillation," New Engl. J. Med., 361(2): 1139-51, 17 September 2009. [12 institutions worldwide] *494QA | 37 | 7 |
SOURCE: Thomson Reuters Hot Papers Database. Read the Legend. |