May 22, 2012 (CIDRAP News) – A new study suggests that obesity does not increase a person's risk of getting sick with influenza—a finding that may come as a surprise to those who are familiar with findings about obesity as a risk factor during the 2009 H1N1 flu pandemic.
Studies during the pandemic showed that those who caught the virus had a higher risk of severe illness if they were very obese. Those findings prompted some to wonder if obesity is a risk factor for contracting flu in the first place. The new study suggests the answer is no.
The researchers, from two clinics in Wisconsin and Minnesota and from the Centers for Disease Control and Prevention, studied adults who sought care for respiratory illnesses during two flu seasons and the 2009 pandemic. They found that patients who tested positive for flu were no more likely to be obese than patients who tested negative.
"Our study demonstrates for the first time that among community-dwelling adults seeking outpatient medical care for an acute respiratory illness, obesity is not a risk factor laboratory-confirmed influenza," says the report, published May 14 in Influenza and Other Respiratory Viruses. The first and senior authors are Laura A. Coleman, PhD, and Edward A. Belongia, MD, of the Marshfield Clinic Research Foundation in Marshfield, Wis.
Obesity tied to immune defects
The authors comment that obesity has been linked to various human immune defects, including reduced responses to tetanus, hepatitis B, and 2009 H1N1 vaccines. In addition, some studies have shown that obese mice are more likely to die of flu.
In the 2009 pandemic, various studies linked severe obesity to an increased risk of complications and death in flu patients. For example, a study focusing on the first 16 weeks of the pandemic in California showed that extremely obese patients were three times as likely to die of the infection as normal-weight patients were.
The authors of the recent study prospectively studied community-living adults (age 20 and up) who visited the Marshfield Clinic for treatment of respiratory illness during the 2007-08 and 2008-09 flu seasons and the 2009 pandemic (May to November of 2009). Those who had a fever, chills, or cough were enrolled and tested for flu.
Obesity was defined by body mass index (BMI). Patients were classified as underweight (BMI less than 18.5), normal (BMI 18.5 to 24.9), overweight (BMI 25.0 to 29.9), obese (BMI of 30.0 to 39.9, or extremely obese (BMI 40 or over).
A total of 2,623 patients were included in the analysis. Their mean BMI was 30.8, and average age was 45. By BMI category, 23% of patients were normal weight, 28% were overweight, 37% were obese, and 11% were extremely obese. Twenty-four percent of the patients had a high-risk medical condition. None of these variables differed significantly by study period.
Of the 2,623 patients, 705 (26.9%) tested positive for flu. The number of flu cases was much higher in the 2007-08 season (454) than in the ensuing season (132) or the pandemic period (119).
No difference in BMI
In their principal finding, the researchers determined that the mean BMI was nearly the same for flu case-patients (30.58 plus or minus a standard deviation of 7.31) and test-negative controls (30.93 plus or minus 7.55).
Using logistic regression models, the authors found that obesity was not a predictor of flu in any of the seasons or in all three combined. The models included adjustments for gender, flu vaccination status, age, and presence of any high-risk condition.
For example, compared with normal-weight patients, extremely obese patients had a 10% higher likelihood of contracting flu (adjusted odds ratio [OR], 1.10), but the wide 95% confidence interval (0.80 to 1.52) meant that the increase was not significant. In contrast to extremely obese patients, those classified as obese had a 5% lower risk of flu than normal-weight patients, but again the difference was not significant.
The authors also ran a secondary analysis covering only the flu case-patients to see if obesity was associated with a serious outcome, meaning pneumonia or hospital admission within 30 days after illness onset. They found no link, but there were very few serious outcomes, which limited their ability to detect such an association.
In contrasting their findings with the evidence that severe obesity increased the risk of complications in 2009 pandemic patients, the authors comment, "We suggest that the risk factors for severe complications of influenza-associated illnesses may be fundamentally different from those for less severe, non-hospitalized influenza."
They conclude that their findings need confirmation in a larger patient population in the interest of understanding the implications of obesity for flu prevention and treatment.
Lone Simonsen, PhD, an infectious-disease epidemiologist and research professor at George Washington University, praised the general design of the study but expressed some reservations.
She said she wonders about the effect of controlling for high-risk medical conditions in the statistical model, commenting, "Surely obese patients are highly likely to have co-morbidities and so would be classified as 'high risk' mostly." This could potentially bias the findings in unpredictable ways, she added.
"What the authors perhaps could have included that would have been very helpful would be a 'positive control,' " Simonsen said. "For example, we can all agree that diabetes is a risk factor of influenza; thus, a sub-analysis demonstrating that diabetes is a risk factor in this prospective cohort study would add confidence that the study design could identify a true risk factor."
She also said she was surprised that the study did not identify young age as a flu risk factor during the pandemic, compared with the previous two preceding flu seasons. (The report says the mean age for confirmed flu patients was 41 in the pandemic season, 41 in 2008-09, and 44 in 2007-08.) Other studies made clear that seniors had a low risk for the 2009 pandemic virus, she noted.
Coleman LA, Waring SC, Irving SA, et al. Evaluation of obesity as an independent risk factor for medically attended laboratory-confirmed influenza. Influenza Other Respi Viruses 2012; early online publication May 14 [Abstract]
Jan 5, 2011, CIDRAP News story "Study confirms extreme obesity as novel H1N1 risk factor"
Aug 16, 2011, CIDRAP News story "Study reveals obesity link to seasonal flu hospitalizations"