When clever wording can lead us to jump to the wrong conclusions

A recent news story made it sound like a new study supported what everyone believes to be true: that skipping breakfast and eating fast food makes teens fat. It was written from a press release, as most news stories are, and both led with the headline:

No breakfast and frequent fast food leads to extra pounds in aging teens

This is one of the most cleverly-worded journal articles I’ve seen in awhile and its findings weren’t anything like the impression we were left with. Yet once again, no reporter appears to have actually read the study to give us the straight scoop.

This study, published in the Journal of Adolescent Health, was conducted by professors at the Weight Control and Diabetes Research Center in Providence, Rhode Island. They used data in 9,919 questionnaires from Add Health, a North Carolina school-based project, gathered during two collection periods. The first sample (Wave II) came from the summer of 1996 and the second sample (Wave III) from the 2001-2002 school year. There were large ranges in ages within the two waves, with the first from 11-21 year olds and the second from young adults 18-27 years of age. No information was provided on how many were boys and girls. The researchers determined fast food consumption by asking the young people in each wave only once how often they’d eaten breakfast and how many days in the past week they’d eaten at a fast food place — nothing was gathered about what was eaten, how much, how many calories or how much fat they consumed.

They also gathered questionnaires asking them how often they participated in structured team sports, skating and bicycling, and exercising — nothing was gathered about how long or how intensely they worked out or any other types of physical activities, and the researchers made no attempt to calculate their energy expenditures but instead, said that they created their own “composite physical activity variable.” Then they didn’t report any of their findings about activity and weight! They appear to only have used the information to control for physical activity in their statistical analyses.

According to the press release, lead researcher, Heather Niemeier, Ph.D., a psychologist at Miriam Hospital, said: “We found that both fast food consumption and breakfast skipping significantly increased between Waves 2 and 3. More importantly, both behaviors were associated with increased weight gain during this time.”

But of course, ages 11 to 27 (with average ages of 15 to 21 in this study) is a time when young people naturally gain weight as they grow from childhood into adulthood, especially boys.

Adolescence to adulthood is also the age group known to eat more fast food and skip breakfast, as this is a transition period when they’re off on their own for the first time, assuming more responsibility for food preparation, and usually busy with school and first jobs. In this study, the frequency of fast food dining increased over the five year age difference among these young people by less than half a serving a week, and breakfast decreased from 4 to 3 days a week.

But an association between these dietary changes and growth does not mean that those caused the weight gain or caused any of the young people to become fatter. Nor will eating less fast food stop these young people from growing.

In fact, the actual results in this study found that increases in fast food intake during this age group did not predict their BMI (body mass index) as they grew. The most the researchers could statistically create was a correlation, they said, between fast food and a 0.02% variance in BMIs. Of meaningless clinical relevance. Similarly, breakfast changes impacted BMIs by 0.01 points. The researchers concluded:

[T]he magnitude of the relationships between fast food consumption and breakfast skipping and body weight in this study was relatively small (predicting increases in zBMI of 0.01-0.02).

Despite these actual findings, the authors asserted that “these behaviors could be easily assessed and targeted... and may provide a useful tool in the prevention of weight gain during adolescence.”

In trying to make their case for this speculation, they cited a Journal of the American College of Nutrition study published in 1986 finding that breakfast consumption had dropped over the previous 30 years. They also described theories that this decline has coincided with an increase in prevalence of obesity and could be a contributing factor. Except, the study they cited twice in making their case for this theory didn’t actually support that. The longitudinal study published in the International Journal of Obesity in 2003 found that skipping breakfast was not related to changes in BMI status among overweight teens, that teens who skipped breakfast consumed fewer calories than those who ate breakfast, and that the kids who ate more calories were also more active. Concerning fast food, the researchers mentioned other papers which discussed portion sizes of fast food, its caloric density and palatability which could be associated with higher energy intake. Except none of those papers ever found that it made any difference in the weight status of the children.

The most notable finding in this study corroborated the body of evidence concerning determinations of body weight. Yet we didn't hear about this finding.

The most significant factor correlating with the children’s BMI beyond where they fell on their growth curve throughout childhood, was their mother’s body size — reaffirming once again, that our size is largely determined by genetics not diet.

© 2007 Sandy Szwarc