How the Obesity Paradox Works

Checking the waist of an obese — Obesity is a growing epidemic, but could it be beneficial in some situations? Peter Dazeley / Getty Images

About 65 percent of Americans are either obese or overweight, and the Centers for Disease Control has classified obesity as an epidemic in the United States. According to NIDDK/NIH, obesity costs Americans more than $117 billion annually in health care [source: NIDDK/NIH]. If you are obese, you have a 50 to 100 percent increased risk of premature death than someone of normal weight. Obesity is a risk factor in other conditions, like high blood pressure, heart disease and type-2 diabetes. However, recent studies have shown that obese people with chronic diseases have a better chance of survival than normal-weight individuals do. This finding has been called the obesity paradox. But before you reach for those extra doughnuts or postpone going on that diet, let's examine obesity.

Obese people have excess body fat. Overweight people have excess body weight (weight includes bone, fat, and muscle). Generally, women have more body fat than men do. Women with more than 30 percent body fat and men with more than 25 percent body fat would be considered obese.

Scientists can measure body fat with X-ray absorption techniques and underwater weighing, which are based on the fact that fat tissue has a different density than bone or muscle. But these methods aren't practical for routine doctor's visits. So, primary health care providers use other methods (like height, weight and skin-fold thickness).

The most popular and convenient method for estimating obesity is the body mass index (BMI). BMI is a ratio of weight to height. This is the formula:

BMI = weight (lb) / [height (in)]² x 703 (English measurements)

BMI = weight (kg) / [height (m)]² (metric measurements)

For example, a 5-foot-5-inch, 150-pound woman would have a BMI of 25. According to these BMI categories, she is overweight but not obese.

Less than 18.5 = underweight
18.5 to 24.9 = normal weight
25 to 29.9 = overweight
More than 30 = obese

There are several online charts based on BMI calculations that you can use to categorize your weight.

Obesity affects men and women of all racial and ethnic backgrounds, but women have a higher percentage of obesity than men. In the United States, African-Americans have the highest percentage of obesity, followed by Mexican-Americans and non-Hispanic whites. Obesity affects about 11 to 28 percent of children, who show the same racial and ethnic obesity patterns. Obesity increases the risk for hypertension (high blood pressure), cardiovascular disease, stroke, cancer, gallbladder disease and diabetes. Obese patients can have higher levels of cholesterol and lipids circulating in their bloodstreams. This can lead to the buildup of atherosclerotic plaques in blood vessels, which increases the risks of high blood pressure, heart attack and stroke. So, obesity is a well-known risk factor for developing cardiovascular disease.

Next, we'll learn how scientists discovered the obesity paradox.

Obesity Paradox Research

In 2001, A. Mosterd and colleagues from the Netherlands studied the prognosis of patients diagnosed with heart failure. They did statistical analyses on more than 5,000 patients, some of whom had heart failure. They found that patients with low BMIs and low blood pressure had more in-hospital deaths than patients with higher BMIs. The researchers claimed that their discoveries supported similar findings from a 1993 study in Massachusetts, and since 2001, at least eight studies have supported the findings. So, even though obesity is a well-known risk factor for heart failure and would be expected to cause problems for obese heart-failure patients, it seems that the opposite could be true.

The obesity paradox extends to other conditions besides heart failure. Patients with chronic kidney disease most often undergo hemodialysis, where a machine filters impurities out of the blood, and dialysis. About 20 percent of dialysis patients die each year from cardiovascular complications. Studies by researchers at UCLA Medical Center have shown that dialysis patients with higher BMIs have a better chance of survival than those with lower BMIs [source: Kalantar-Zadeh].

To summarize, the obesity paradox goes something like this. Obesity is a major risk factor for cardiovascular disease (like hypertension, congestive heart failure, coronary artery disease) and chronic renal disease. However, in patients with these chronic diseases, it appears that obesity is associated with better survival. If this finding is actually true, it could have important implications for how physicians treat patients with chronic diseases. Doctors could conceivably stop putting patients on diets and recommending that they lose weight.

So, why does the obesity paradox happen? We'll find out on the next page.

Obesity Paradox Explanations

The medical community has had mixed reactions to the idea of the obesity paradox. Many physicians and scientists are skeptical because the findings go against what is expected from the normal population. Researchers at the University of Texas School of Public Health and Baylor Medical College reviewed many published reports of the obesity paradox and came up with six reasons to explain the findings -- and possibly to be skeptical of them [source: Habbu].

The numbers of people studied in the reports of the obesity paradox were generally small. Therefore, do these results apply or hold in much larger populations?
The statistical techniques show associations between factors, but not cause-and-effect conclusions. So, the results may or may not be real.
In many studies, congestive heart failure was diagnosed from clinical symptoms (difficulty breathing, swelling in extremities) instead of from laboratory tests (like echocardiography, cardiopulmonary testing, cardiac catheterization).

These clinical criteria to diagnose congestive heart failure have not been validated in obese populations and may not be applicable.
In some of the studies where lab tests were done, the obese patients had slightly better heart functions (pumping ability, oxygen delivery) than normal or underweight patients.
Therefore, the obese patients may have been either slightly "healthier" with respect to CHF or in earlier stages of CHF than their normal/underweight counterparts. So obese survival rates were better.

Congestive heart failure (and chronic kidney disease) is a wasting disease. Patients are so sick that they lose weight (fat, muscle mass) over the course of the disease. This could lead to two conclusions:

Again, obese patients could be "healthier" or in earlier stages of these chronic diseases. Chronic disease patients with low BMIs do not have low BMIs intentionally, but because of the nature of the wasting disease. None of the studies discriminated between intentional weight loss (from diet and exercise) and unintentional weight loss (from disease).
Obese patients may have a better metabolic reserve than their normal/underweight counterparts.

Few of the studies looked at extreme obesity (BMI more than 35). In some studies that did, the extremely obese did not have a greater chance of survival than the underweight. Therefore, the survival curves may be U-shaped. Normal and overweight patients would have better survival probabilities than those at the extremes -- underweight and extremely obese.
Some recent studies have questioned whether BMI is the best way to categorize obesity. Maybe waist circumference or waist-to-hip circumference ratios may be better indicators of obesity. These criteria are based on observations that body fat stored in the waist is worse with respect to the risks of obesity than fat stored elsewhere. [source: Romero-Coral].

On the next page we'll learn about more explanations for and criticisms of the obesity paradox theory.

Obesity Paradox Criticisms

There are many reasons to be skeptical of the obesity paradox, but the UCLA researchers have posed some biological explanations for it [source: Kalantar-Zadeh].

The adverse effects of obesity take more time to develop than those of chronic disease (heart failure, kidney disease). Therefore, the consequences of wasting kill patients much faster than obesity does.

In both CHF and chronic kidney disease, malnutrition and inflammation are common. These conditions alone could reduce the survival of these patients. Therefore, weight gain could be an indicator of better nutrition and, therefore, improved chances of survival in obese patients.

Dr. Kalantar-Zadeh argues that dietary restrictions placed on CHF and chronic kidney disease patients my actually be harmful to these patients. He advocates a "reverse epidemiology" approach to these conditions.

So, the obesity paradox has been observed in several clinical settings (CHF, chronic kidney disease), but physicians and scientists still aren't sure that it's a real phenomenon. Not many researchers have proposed biological hypotheses to explain the obesity paradox. There have been no direct animal studies or clinical trials that could address cause-and-

effect relationships. So, based on evidence available at this time, we can't conclude that the obesity paradox is real, certainly not enough to alter treatment for patients with CHF and chronic kidney disease. But many scientists think that research should be aimed at resolving the obesity paradox. It may have implications for changing treatment options for these patients.

Dr. Kalantar-Zadeh and colleagues also argue that the risk factors for the general population may not be applicable to the obese population. The obesity paradox is one example of this idea, which has been called reverse epidemiology. Imposing dietary restrictions on obese chronic disease patients, they say, may actually be harmful. The group cites a number of conditions where phenomena like the obesity paradox have been observed [source: Kalantar-Zadeh].

Lots More Information

Sources

Bray, GA, Medical consequences of obesity, JCEM: 89: 2583-2589, 2004. http://jcem.endojournals.org/cgi/content/full/89/6/2583
Cordtlandt Forum: Extra weight: A benefit for some patients? http://www.cortlandtforum.com/content/index.php?id=26&no_cache=1&tx_cortlandtmagissue_pi1%5BshowUid%5D=1489InCites
Cossrow, N and B Falkner, Race/ethnic issues in obesity and obesity-related comorbidities, JCEM: 89: 2590-2594, 2004. http://jcem.endojournals.org/cgi/content/full/89/6/2590
Davey Smith, G et al, Meta-analysis: Beyond the Grand Mean?, British Medical Journal: 315:1610-1614, 1997. http://www.bmj.com/cgi/content/full/315/7122/1610
Egger, M AND G Davey Smith, Meta-analysis: Potentials and Promise, British Medical Journal: 315:1371-1374, 1997. http://www.bmj.com/cgi/content/full/315/7119/1371
Egger, M et al, Meta-analysis: Principles and Procedures, British Medical Journal: 315:1533-1537, 1997. http://www.bmj.com/cgi/content/full/315/7121/1533
Habbu, A et al, The Obesity Paradox: Fact or Fiction?, Am J Cardiol 98: 944-948, 2004. http://www.ncbi.nlm.nih.gov/sites/entrez?cmd=Retrieve&db=PubMed&list_uids=16996880&dopt=Citation
Hall JA et al, The Paradox of Obesity in Patients with Heart Failure, Journal of the American Academy of Nurse Practitioners 17, 542-546, 2005. http://www.blackwell-synergy.com/doi/abs/10.1111/j.1745-7599.2005.00084.x?journalCode=jaan
InCites: An Interview with Dr. Kamyar Kalantar-Zadeh. http://in-cites.com/scientists/KamyarKalantar-Zadeh.html
Kalantar-Zadeh, K et a. Reverse epidemiology of cardiovascular risk factors in maintenance dialysis patients, Kidney International 63: 793-808, 2003. http://www.nephrology.rei.edu/RevEpi_01.pdf
Kalantar-Zadeh, K et al, Racial and survival paradoxes in chronic kidney disease, Nature Clinical Practice Nephrology 3, 493-506, 2007. http://npg.nature.com/ncpneph/journal/v3/n9/full/ncpneph0570.html;jsessionid=9D829C9B81FD63FE14FE8385A996A823
Kopple, JD, The phenomenon of altered risk factor patterns or reverse epidemiology in persons with advanced chronic kidney failure, American Journal of Clinical Nutrition 81: 1257-1266, 2005. http://www.ajcn.org/cgi/content/full/81/6/1257?maxtoshow=&HITS=10&hits=10&RESULTFORMAT=&searchid=1&FIRSTINDEX=0&minscore=5000&resourcetype=HWCIT
Liu, Y et al, Association Between Cholesterol Level and Mortality in Dialysis Patients, JAMA; 291: 451-459, 2004. http://jama.ama-assn.org/cgi/content/full/291/4/451
Mosterd, A et al, The Prognosis of Heart Failure in the General Population: The Rotterdam Study, European Heart Journal 22, 1318-1327, 2001. http://eurheartj.oxfordjournals.org/cgi/reprint/22/15/1318
NHLBI/NIH, Guidelines on Overweight and Obesity: Electronic Textbook. http://www.nhlbi.nih.gov/guidelines/obesity/e_txtbk/txgd/4142.htm
NIDDK/NIH Weight-control Information network, Understanding Adult Obesity. http://win.niddk.nih.gov/publications/understanding.htm
Romero-Coral, A et al, Diagnostic performance of body mass index to detect obesity in patients with coronary artery disease, European Heart Journal 28: 2087-2093, 2007. http://eurheartj.oxfordjournals.org/cgi/content/abstract/28/17/2087
Ryan, D, Risks and benefits of weight loss: challenges to obesity research, European Heart Journal Supplements 7 (Supplement L), L27-L31, 2005. http://eurheartjsupp.oxfordjournals.org/cgi/content/full/7/suppl_L/L27
Slyper, AH, The pediatric obesity epidemic: causes and controversies, JCEM: 89: 2540-2547, 2004. http://jcem.endojournals.org/cgi/content/full/89/6/2540
Special Obesity Issue, Journal of Endocrinology and Metabolism (JCEM): 89, June, 2004. http://jcem.endojournals.org/content/vol89/issue6/
Stein, CJ and GA Colidtz, GA, The epidemic of obesity, JCEM: 89: 2522-2525, 2004. http://jcem.endojournals.org/cgi/content/full/89/6/2522