July 11, 2013

Is childhood obesity contagious?

Every parent knows that young children catch lots of things at school: chicken pox, the flu and, of course, the annual back-to-school cold.

Now there’s evidence that kids can catch something else from their classmates:  obesity.

Of course there isn’t a fat virus, or at least one we know about.  But a research team from the University of Arkansas has tracked the pattern of weight gains and obesity among 341,876 elementary school students and found that the typical student gained extra pounds in grades with a larger share of obese classmates but slimmed down when they were in classes with a larger proportion of skinny kids.

Hispanic and black children were particularly susceptible to exposure to fat peers, they found. Even an increase in the share of children who were considered overweight but not obese had a corresponding effect on other students’ weight, Jebaraj Asirvatham reported in a paper the team will deliver next month at the Agricultural & Applied Economics Association meeting in Washington, DC.

The effect is striking.  As the proportion of obese students in a grade doubles, “a typical normal-weight child nearly becomes overweight,” Asirvatham said in an email.

Asirvatham says their study shows that changes in a student’s weight are strongly correlated with the weight of their peers. But he cautions that more work is needed before researchers can claim a causal relationship between the two or pinpoint what the underlying mechanisms causing the change might be.

Their conclusions flow from an analysis of a unique database that contains the body mass index (BMI) scores of every Arkansas public elementary student from 2004 to 2010. These BMI scores, which measure body fat based on an individual’s height and weight, were collected every two years, allowing researchers to track individual children and their classmates over time.

To help ensure they measured the effects of the school environment, they controlled for the presence of fast food restaurants, convenience stores and other fatty food venues in the area surrounding the school and in the student’s home neighborhood. Overall, Arkansas grade school students rank in the 69th percentile among students nationally in terms of BMI scores.

Asirvatham and his colleagues discovered that as the share of overweight students in a class rose, so did the BMIs of other students. For example, they found that if the percentage of obese students doubled, the percentile ranking of the typical student’s BMI rose from 69 to 82—“only 3 percentile ranks below being classified as overweight,” he said.

The weight gain effect was particularly strong among Latino and black students.  “An average Hispanic student at the 70th percentile moved to the 85th percentile or became overweight,”  Asirvatham and his colleagues found.  “The increase in body weight for an African American student was a little lower, i.e., he [or] she moved to the 84th percentile.  The increase was lowest for a Caucasian student, who moved to the 79th percentile.  Thus Hispanic and African American students seem to be more influenced by peers than Caucasians.”

Category: Social Studies

Topics: Health, Teens and Youth

  1. Photo of Rich Morin

    is a senior editor focusing on social and demographic trends at Pew Research Center.

Leave a Comment


All comments must follow the Pew Research comment policy and will be moderated before posting.


  1. LFB, MD2 years ago

    The choice of title for this article is most unfortunate and misleading as Dr Gross also indicated in his comment of 7/11/13. Over my years as a primary care physician I was well aware of the misperception, incomplete or incorrect information people acquire from scanning headlines, article titles, or by the news medias’ use of “snippets” or “teasers” prior to the fuller report, which often does not get heard or viewed. For many individuals, those information “snippets” become their perception, knowledge or belief of a particular entity being reported.

    From an anecdotal perspective, my recollection of days when the office phone lines would be inundated, were those immediately following the reporting of certain health related matters, and especially so if the perception was that one could “catch” or “become infected” in some unknown way by those around them! Consider the public response in 2003 to SARS and in 2006 to the H5N1 “Bird Flu” media reporting. Similar scenarios occurred with media reporting of a “previously unknown”, possibly dangerous effect from a prescription medication, but some individuals simply stopped the now allegedly “dangerous drug” following such reports, leading to an underlying condition, usually hypertension or diabetes, to become uncontrolled with the dangers for the individual in that circumstance being far greater than the one being reported in the media. The preceding anecdotes should suffice for the true intent of the paragraph that follows.

    Journalists and media reporting on medical or other health related subjects an have exceptionally important functions for the general public, however this ought to equally carry an exceptional obligation for close attention to factual accuracy, avoiding “sensationalizing” or “spin” in titles or headlines, inclusion of scientifically based estimations of risk, harm, etc. having the appropriate emphasis [or perhaps that should be “underemphasis”] when reporting on the real critical concern for much of the public “how common, dangerous or likely is this?” At least that’s how I see it from my years as a family practice physician. [this ends the Andy Rooney-ish portion of my comment! ]

    Comments on the study reported in this article:
    The study’s purpose was to investigate aspects of “peer pressure”, “peer modeling” and peer influences within the social networks of school age children regarding weight in light of the increasing incidence of overweight and obesity among children and adolescents. The importance of peer influences, particularly”positive peer influences” have been shown in several studies to help reduce smoking, substance abuse, sexual activity and other risky behaviors among children in adolescents when implemented in the school setting. Of course “negative peer influence or modeling” would have the opposite effect, or to use parent lingo “hanging out with the wrong crowd.”

    Their findings indicate that peer group effects or peer modeling behaviors seem to have some role in increasing or decreasing students weights, thus trials of “positive peer influence” programs similar to those showing favorable effects in reduce smoking, substance abuse etc might be a good strategy to include within the schools.
    This would seem intuitive, however in the era of evidence based medical practice it is wise to ensure one’s intuition lines up with some substantiating scientific evidence prior to implementation of programs or policies within school systems and this study has provided a start on that component. Such information would be useful for parents and teachers to be aware of as option to try in addressing the “obesity epidemic” among children and adolescents, and a very cost effective one as well!

    1. Rich Morin2 years ago

      Very thoughtful comments from both LFB and Dr. Gross. As for the headline, it was inspired by and a reference to the headline “Is obesity contagious? Social networks vs. environmental factors in the obesity epidemic” (Cohen-Cole Fletcher), an article that appeared in the Journal of Health Economics in 2008. The piece raised questions about a widely publicized study on the the spread of obesity “The Spread of Obesity in a Large Social Network over 32 Years” (Christakis Fowler) that appeared the previous year in the New England Journal of Medicine.

  2. Dr Stan Gross2 years ago

    With all due respect, a contagion suggests a pathogen like a virus or bacteria. However, close proximity to a person struggling with excess weight might just mean eating the same foods and experiencing the same lifestyle. If there IS a contagion, it is being transmitted by television advertisement displaying fast foods and junk food as being so delicious. Case in point, China is starting to experience an obesity issue. While still under 22% of the total population has a weight issue, China regards that statistic as significant. How did they go from 8% to 10% obesity to over 20%? Simple! They began eating like Americans. They began experiencing greater financial stress and they were drive to work longer hours and exercise less. The social pathogen is not a virus or bacteria, but a lifestyle that has been glamorized by the media. However, I am hoping in the near future they discover a connection between looking younger and being subjected the youth virus. Please keep me informed of any progress in that area of healthcare.