Thursday, January 05, 2012

Is Grandma to Blame for Obesity Today?

Posted by Darin Hayton on 01/05 at 01:58 PM

In her recent “New Year’s Roundup” post, Historiann pointed out an article on obesity that appeared in The Denver Post. As the title suggests, Shari Roan’s article, “1950s Women May Have Triggered Obesity Epidemic,” blames mothers for the recent rise in obesity. The article was originally published in last month’s Los Angeles Times, where it was titled “Obesity Epidemic May Have Roots in 1950s.” Although the title has been changed, the content is identical. Roan’s article purports to summarize Melinda Sothern’s “obesity trinity” theory.

In this post I am concerned with criticizing Roan’s article and her presentation of Sothern’s theory. The article does not provide enough details of the theory or how it might or might not explain obesity rates to assess the theory itself. And despite a long list of published articles on Sothern’s department webpage, which unhelpfully was not linked to in either version of the article, I was unable to determine which of the articles discussed her “obesity trinity.” Most of her publications seem to be some form of commonsensical advice about more exercise will help reduce obesity rates in children.1 I think Roan’s article is particularly problematic because it will be taken by many people as validation for their prejudices, often retrograde and oppressive prejudices. Further, Roan’s article represents bad science reporting (bad reporting in general, actually).

Shari Roan’s article on obesity in the LA Times

Roan reports that the rise in obesity in the 1980s resulted from “a generation of young women decades earlier [who] smoked, spurned breast-feeding and restricted their weight during numerous, closely spaced pregnancies” 2 Roan offers few details beyond recounting Sothern’s anecdotal family experience:

  • In the 1950s Sothern’s mother smoked to keep her weight gain down during pregnancy.

  • Sothern’s mother did not breast feed her children but relied instead on bottle-fed formula.

  • Sothern and her two siblings were born within four years.

The conclusion, according to the article, is inescapable: “All three children — Sothern thinks it’s no coincidence — battled with their weight as adults: Her brother is diabetic and her sister is obese.” But surely personal anecdote doesn’t carry much scientific or logical force.

Yes, as the article points out, smoking cigarettes was not uncommon, often even advised by doctors. And yes smoking “is thought to contribute to obesity risk in offspring.” And yes, as the article points out, by the 1970s (note, not the 1950s) breast feeding was certainly out of fashion. And yes, breast feeding can prevent ovulation and as a result reduce the frequency of pregnancies. But this list of facts and possibilities does not demonstrate a causal link between these factors and a rise in obesity. Moreover, this list represents only a tiny subset of the conditions that have changed over the last 50 years. Roan makes no effort to justify these.

And despite Roan’s bold claim that “the statistics tell the tale,” her statistics—four numbers—don’t tell any tale. She points out that on average, middle-aged men and women and 10-year-old boys and girls are heavier today than they were in the 1960s. Of the various problems with these statistics I am most interested in two:

  • Roan confuses weight with obesity. Other factors need to be considered—such as height—and ruled out before weight can be correlated to obesity.

  • Although Roan is trying to demonstrate a cause for rising obesity rates, her evidence is that people weigh more today than they did 50 years ago. If weight can be directly correlated to obesity, all Roan has shown is that today people are more obese.

These statistics may be related to rising obesity rates, but as stated in the article they are unhelpful.

Also unhelpful is Roan’s repeated dismissal of social and cultural factors. She admits that

other changes were afoot in the mid-20th century, of course: the growth of suburbs, a car culture and modern conveniences. The fast-food craze was launched with the first McDonald’s in 1961.

And further, she even suggests that Sothern recognizes multiple causes for the rise in obesity:

The obesity epidemic has multiple causes, Sothern acknowledges. Food has changed in the last five decades. Americans have become much more sedentary.

But in the end, these cultural factors play no role because, according to Roan, “Sothern thinks there must be more to the story for the changes to have happened the way they did, at warp speed: ‘There had to be physiological and metabolic changes in our bodies.’” But wait. These seem like just the sort of factors that would be relevant: changes in food availability, quantity, and processing; decrease in calorie expenditure; significant changes in lifestyles. Can we really reject these in favor of a reductive, material causality?

Any theory of obesity needs to take into account, at a minimum, calorie consumption and expenditure. Calorie consumption alone seems to have contributed significantly to our weight gain.

According to the USDA’s Agricultural Fact Book 2001-2002:

Americans at the beginning of the 21st century are consuming more food and several hundred more calories per person per day than did their counterparts in the late 1950s (when per capita calorie consumption was at the lowest level in the last century), or even in the 1970s. The aggregate food supply in 2000 provided 3,800 calories per person per day, 500 calories above the 1970 level and 800 calories above the record low in 1957 and 1958.3

Adjusting for possible waste, Americans are still consuming around 2700 calories per day, about 530 more than in 1970.

The USDA Agricultural Fact Book 2001-2002 is full of other interesting information about changing dietary patters that certainly must be accounted for in any effort to understand changing obesity rates. Some are surely relevant, such as:

  • the record levels at which we consume caloric sweeteners—152 pounds of caloric sweetener per year in 2000;
  • rising tendency to eat out—1994-96 Americans ate 32% of their calories from restaurants (it is unclear if this number includes fast food restaurants), up from 18% in 1977-78;
  • the increase in refined grain in our diet—200 pounds in 2000 compared to 155 pounds in the 1950s.4

Those extra 500-800 calories per day would add up quickly. We are consuming something around 20% more calories than we were in 1970 and approaching 30% more than we were in the 1950s. And what happens when we complicate this picture by considering relative consumption. The USDA statistics are for the average American. Is there any evidence that all Americans have increased their consumption? Or any evidence that different ethnic and socio-economic groups have different consumption patterns, both in quantity and quality of calories? Regardless, given that the average American is consuming so much more, should we be surprised that we weigh more today?

Then there is decreased daily calorie expenditure. As Roan indicates, we are living more sedentary lives now than a few decades ago. Much of Sothern’s work seems to focus on sedentary living as the key problem. Judging from the titles of her publications, Sothern argues consistently that children need to increase their activity levels precisely to increase calorie expenditure. Yet Roan minimizes the role of our increasingly inactive lifestyles.

Roan should have shown that she had considered these broader issues rather than ignore them. Her task in reporting on science is to assess and evaluate the theory. She needs to tell her readers what the theory does well and what it does not do so well. She can’t simply promote a particular agenda. If she accepts the theory—here the “obesity trinity”—then she needs defend it by marshaling evidence, by showing its strengths and advantages, by arguing for it, rather than merely being a cheerleader for that theory.

Regrettably, Roan does not present the case for Sothern’s “obesity trinity.” Instead, she presents statements devoid of any context and relies on implication. And even out of context her statements do not demonstrate her claims. Instead, they suggest that current behavioral and cultural factors are as important if not more important than smoking in the 1950s and bottle-feeding in the 1970s. Her experts seem to focus on changing prenatal and neonatal nutritional practices. And that brings us back to social, cultural, and socio-economic factors. Practices are socially and culturally constituted. How, then, do those practices correlate to different demographic, geographic, and economic factors? How different would this article have looked if Roan had reported on the different obesity rates in the U.S.? What would happen if you take seriously the demographic and geographic and economic distribution of obesity? That might have been interesting.

Although Roan ignores socio-economic factors, there is considerable disturbing rhetoric in the article that suggests there is a social agenda involved. At one point Roan quotes Sothern: “Significantly overweight women should not have babies. Women should be physically active and have a healthy diet for at least a year before pregnancy.” Later Roan recounts one of Sothern’s “prescriptions for change: Women who are significantly overweight should be discouraged from having babies until they shed some pounds.” As presented, such statements are disturbing for the way they recall earlier efforts to decide who is and is not fit to procreate.

In addition to suggesting who should and should not become a mother, Roan’s article seems to advance two social agendas: mothers should stay at home with their children and mothers should breastfeed their children. A luxury that many families cannot afford, Roan claims that “[w]omen should breast-feed for at least six months after childbirth or — better yet — take one year off from work and breast-feed.” While that advice may be possible for the mothers who can relate to Sothern’s “need to exercise” and her “plans to rent a bicycle in Coronado and ride, fast and far,” for many mothers that ideal is just not possible.

Roan’s article is problematic on a number of levels: it is fallacious in various ways, most obviously because it is grounded on a post hoc, ergo propter hoc fallacy; it promotes a particular agenda—Sothern’s “obesity trinity”—rather than assessing or analyzing a scientific theory; it offers a misleading picture of what constitutes a scientific theory; and it ignores a considerable amount of relevant information. The article masquerades as factual reporting when, in fact, it functions as propaganda for a particular social agenda. But in the end, blaming somebody for our “obesity epidemic” is easier than trying to understand the multifarious factors that contribute to obesity.

Unfortunately, Roan’s use of the term “epidemic” already casts the issue of obesity in social and moral terms that include assigning blame, casting aspersions on certain people, and condemning “risky” behaviors. Epidemics are social phenomena rather than biological. Plague and epidemic narratives reflect social judgements and prejudices, stigmatizing groups and classes of people and justifying their marginalization and repression.
1Although the “get more exercise advice” seems commonsensical, it is not necessarily easy to implement. Multiple studies have shown that there is a significant gap between the knowledge that certain activities carry increased health risks and the ability to act on that knowledge. Moreover, as numerous studies have shown, there are often powerful social and cultural factors that prevent people from acting on that seemingly commonsensical advice.
2All quotations are from the LA Times story, “Obesity Epidemic May Have Roots in 1950s.”
3From the chapter “Profiling Food Consumption in America” in the Agricultural Fact Book 2001-2002.

Tags: melinda sothern, obesity, science journalism, shari roan, usda


Comment posted by Historiann on 01/05 at 03:46 PM

Darin, thanks for the link and the excellent takedown of Roan’s article here.  You provide lots of crunchy data points for us to consider regarding the real scope and cause of increased average weight in the U.S. over the past thirty years.  Furthermore, your post points to the problematic journalism on display here.

As many of my readers have pointed out, grappling with the many complex roots of increasing weight gain among Americans is hard.  Blaming and shaming mothers is really easy by comparison!

Comment posted by Darin Hayton on 01/05 at 10:14 PM

Dear Historiann,

Thanks for reading and for pointing me to the article in the first place.

Page 1 of 1 pages