Up to half of U.S. adolescents engage in disordered eating behaviors, such as frequent dieting or vomiting, to control their weight. Understanding how these behaviors link to the body mass index (BMI) of people across their lifetime could help researchers design approaches to stop unhealthy eating and weight-management habits. A new study by the School of Public Health recently examined the relationship and found that teens who engaged in disordered eating behaviors often have higher BMI numbers as adults.

The study was led by postdoctoral fellow Cynthia Yoon and published in the Journal of Adolescent Health.

Yoon used data from the school’s long-running Project Eating and Activity in Teens and Young Adults (Project EAT) for her research. Project EAT has followed groups of teens since the 1990s to learn more about their health and eating habits as they mature into adulthood. 

Yoon used 15 years of data from 1,230 people who began completing Project EAT surveys as teens (11-18 years) and continued into adulthood (27-33 years). Yoon reviewed survey answers to count the number of seven particular disordered eating behaviors participants reported using during adolescence.

1. Valuing importance of weight and shape: Reporting that weight and shape were the main or most important things to them over the past six months. 
2. Unhealthy weight control behaviors: Using behaviors such as fasting, taking diet pills, self-induced vomiting, using laxatives or diuretics (water pills).
3. Frequent dieting: Dieting five or more times in the prior year.
4. Episodic overeating: Eating an amount of food in a short period of time they believe would be embarrassing if others saw them doing it.
5. Distress about overeating: Feeling upset about eating more than they thought was best for them.
6. Loss of control when overeating: Feeling that they could not stop eating or control what or how much they were eating.
7. Frequency of overeating with loss of control: Feeling that they ate large amounts of food in an out of control way multiple times a month.

Yoon also used the Project EAT data to determine the BMI of participants as adults. She then compared how the numbers of disordered eating behaviors tracked with their BMI.

The study found:

• 34% of males and 51% of females in the study said they used at least one disordered eating behavior in adolescence; 
• 12% of males and 29% of female adolescents used two or more behaviors; 
• people who reported engaging in two or more behaviors in adolescence had higher BMIs in adulthood than those who did not use those behaviors; 
• males typically had the highest BMI levels. 

“The results mean that counting the number of disordered eating behaviors might be a simple but adequate approach for examining the severity of those behaviors and its consequences, especially in terms of BMI and body-weight related health issues,” says Yoon.

One limitation of the study is it couldn’t pinpoint specifically which disordered eating behaviors lead to the development of higher BMI. 

“However, the study suggests the total number of disordered eating behaviors contributes to developing higher BMI because the study results show us that the more behaviors participants are engaged in, the higher their BMI will be over time,” says Yoon. 

Yoon added that adolescents who engaged in disordered eating behaviors typically had higher BMI compared to those who didn’t do the same things. She said it is likely that the behaviors lead to greater BMI, rather than the reverse association of higher BMI leading to disordered eating.

“One thing we could not determine is whether adolescents had higher BMI before their eating behaviors were first assessed when they began the Project EAT study,” says Yoon. “However, if BMI diverged before the assessment of eating behaviors, it’s possible that a feedback loop may exist. For instance, a higher BMI during childhood may lead to disordered eating behaviors, which may further place participants at greater risk for higher BMI trajectories.

For further insight into how the relationship between disordered eating and BMI plays out, Yoon recommends that follow-up research measure the prevalence of health problems related to excessive weight and obesity diagnosed within the study group.