To Control Weight, Adjust 'Diet Trajectory'

    • October 26, 2011

Faced with mounting evidence that being overweight is bad for their health, many Americans still find it difficult to change eating habits that pile on the pounds. For them, a new study by a Robert Wood Johnson Foundation (RWJF) Health & Society Scholar brings this good news: Even modest changes from less healthy to more healthy eating habits can help control weight gain.

Mark Pachucki, PhD, reached that conclusion after analyzing the food choices of more than 2,000 people over a 10-year period. His study found that it was "extremely uncommon for adults to drastically change from an unhealthy diet to a very healthy diet." But the study also found that even "incremental changes toward a healthier diet" can stave off weight gain, says Pachucki, an RWJF Health & Society Scholar (2010-2012) at the University of California Berkeley School of Public Health and University of California San Francisco Center for Health & Community.

With support from RWJF, the National Institutes of Health and the National Science Foundation, Pachucki set out to examine eating patterns in a way that previous research had not. "In the last 20 years, there have been hundreds of studies that charted what a given population was eating, with the general goal of linking particular food items or elements of a diet with some kind of health outcome," he says. "The thrust has been to say, 'Let's figure out what a group of people were eating at a given point in time. We can classify their food choices into these four diet boxes, and then we can look to see if they have higher risk for, say, cardiovascular disease if they're eating the foods in Box A than in Box D.' "

What that research approach fails to consider, Pachucki says, is what happens when eating patterns change. He designed his study to look not just at people's diets at one point, but a "diet trajectory" that showed how people changed what they ate over a 10-year span.

Pachucki's study used data from the Framingham Heart Study, a longitudinal study designed to investigate risk factors for cardiovascular disease. His study cohort began with 3,418 Americans who had a median age of 54. On the scale that uses weight and height to calculate body mass index (BMI)—where a BMI of 18.5 to 24.9 is healthy, 25 to 29.9 is overweight, and 30 or more is obese—the study subjects' median BMI was 26.7. At three points between 1991 and 2001, 2,437 subjects completed "food frequency questionnaires" describing what foods they ate most and least often.

By crunching the data on hundreds of food choices, Pachucki arrived at seven food groupings that characterized the study subjects' eating patterns. The least healthy were diets heavy in meat and soda, in sweets, and in alcohol and snacks. In the healthiest diets, study subjects consumed chiefly fruits, vegetables and low-fat proteins, or at least enough of those healthy foods to offset the less-healthy ones they sometimes ate.

Pachucki said he began the study with the expectation that in the 50-plus age group, "people's taste preferences are pretty much set and they would tend to eat more of the same things over time." So he was surprised to find that, over the decade, about two-thirds of study subjects showed enough change in what they ate to classify them into a different eating pattern.

By analyzing subtle shifts over the decade between diets in the seven food groupings, Pachucki found that the roughly 23 percent of study subjects with a healthful diet trajectory—people who made even modest changes to more healthy eating patterns—ended up gaining less weight. Meanwhile, the 34 percent of study subjects whose diets were consistent through the decade, and the 26 percent whose trajectories were mixed (moving up and down between healthier and unhealthier diets over time) all gained weight, on average.

As Pachucki had expected, after controlling for possible confounders, those who changed from eating the more-healthy to the less-healthy diets—about 17 percent of study subjects—gained weight. The study found that individuals who followed an unhealthful eating trajectory over time were significantly more likely than their study subject peers to have a higher BMI at the end of the study; they were 1.8 times more likely to be overweight, and were 2.4 times more likely to be obese. In addition, individuals whose trajectory was mixed were 1.5 times more likely to be overweight.

"Although on average no trajectory was linked to weight loss, those in the healthful trajectory showed the least weight gain," Pachucki concluded. "As physicians struggle to help their patients improve their eating habits, these findings confirm the importance of considering a patient's diet history and reminding patients that even small, more healthful changes in diet can benefit their waistline."

The study, "Food Pattern Analysis Over Time," was published online July 26, 2011 in the International Journal of Obesity.