Weight Control Practices and Disordered Eating Behaviors Among Adolescent Females and Males With Type 1 Diabetes

Dianne Neumark-Sztainer, PHD, Joan Patterson, PHD, Alison Mellin, PHD, Diann M. Ackard, PHD, Jennifer Utter, MPH, Mary Story, PHD and Joseph Sockalosky, MD

Disclosures

Diabetes Care. 2002;25(8) 

In This Article

Results

More female respondents (92.4%) than male respondents (53.6%) reported engaging in weight control behaviors over the past year to lose weight or keep from gaining weight (P < 0.001). Disordered eating behaviors, as assessed with the DEPS, also tended to be higher among females than males, although differences were not statistically significant (females: 44.8 ± 10.7%; males: 41.7 ± 8.0%; P = 0.068).

The frequencies for each of the specific weight control behaviors used by females and males are shown in Table 1 . Healthy weight control behaviors were the most commonly reported practices, especially for females. Among the unhealthy behaviors used to control weight, skipping meals was used the most frequently. Of the very unhealthy weight control behaviors, 10.3% of the females reported skipping insulin, and 7.4% of the females reported taking less insulin as a way to control weight. Only one male reported engaging in any method of insulin mismanagement for weight control.

In classifying respondents into one of the four weight control behavior categories, 7.6% of females and 46.4% of males used no weight control behaviors, and 54.5% of females and 37.7% of males used only healthy weight control behaviors. Among females, 18.2% engaged in unhealthy weight control behaviors, and an additional 19.7% used very unhealthy behaviors to control their weight. For males, these last two categories of behaviors were less frequent with 13.0% and 2.9% reporting unhealthy behaviors and very unhealthy weight control behaviors, respectively.

When the categories of weight control behaviors (none, healthy only, unhealthy, and very unhealthy) were compared separately for females and males by age, parental level of education, and race, there were no differences between the categories (data not shown). Although the difference was not statistically significant, it is noteworthy that 66.7% of the nonwhite females (6 of 9) reported unhealthy or very unhealthy weight control behaviors in comparison to 31.6% of the white females (18 of 57). Similarly, associations between sociodemographic characteristics and disordered eating behaviors, as assessed with the DEPS, were not statistically significant (data not shown). Family structure was not associated with weight control behaviors or DEPS scores.

For males, BMI differed by weight control behavior category; males in the very unhealthy weight control category had a significantly higher mean BMI than those in the other three categories (see Table 2 ). There were significant differences in weight perception by category of weight control behaviors for both females and males; respondents with more unhealthy weight control behaviors reported perceiving their weight as being heavier. Weight dissatisfaction differed significantly across the weight control behavior categories for females; those engaging in unhealthy behaviors reported the highest rates of weight dissatisfaction.

Slightly different patterns were found in examining associations between disordered eating behaviors, using scores from the DEPS, and BMI, weight perception, and weight dissatisfaction. Among females, DEPS scores were associated with weight dissatisfaction (r = 0.40; P = 0.001) but were not associated with BMI (r = 0.10; P = 0.442) or weight perception (r = 0.22; P = 0.090). Similarly, among males, DEPS scores were associated with weight dissatisfaction (r = 0.31; P = 0.013), but were not associated with BMI (r = 0.10; P = 0.424) or weight perception (r = 0.10; P = 0.417).

Family cohesion scores differed significantly between the weight control behavior categories among the females. Females using only healthy weight control behaviors reported the highest levels of family cohesion, while females using very unhealthy behaviors reported the lowest levels of family cohesion (see Table 3 ). Similarly, among the females, family cohesion was significantly associated with the DEPS scores (r = -0.52; P < 0.001). Other measures of the family social environment (control, independence, and responsibility for diabetes management) were not significantly different between the weight control behavior categories. The correlations between DEPS scores and the family environment were also nonsignificant.

Among the males, family cohesion did not differ significantly between the weight control behavior categories, although (like the females) the highest levels of family cohesion were reported by males using only healthy weight control behaviors, followed by males not using any weight control behaviors (see Table 3 ). For males, family cohesion was significantly associated with DEPS scores (r = -0.41; P < 0.001); males from more cohesive families reported fewer disordered eating behaviors. Other measures of family social environment (control, independence, and responsibility for diabetes management) were neither significantly different between weight control behavior categories nor correlated with DEPS scores.

1c and Weight Control/Disordered Eating

Among females, there was a statistically significant difference between weight control behavior categories and laboratory measures of HbA1c; females reporting only healthy behaviors had significantly lower HbA1c levels than females reporting unhealthy weight control behaviors (see Table 4 ). The correlation between disordered eating behaviors (as assessed with the DEPS) and HbA1c levels also was statistically significant (r = 0.33; P < 0.01). Females reporting more disordered eating behaviors had poorer metabolic control, as evidenced by their higher HbA1c levels.

Among males, the difference between weight control behavior categories and HbA1c level was not statistically significant, although a trend similar to that found among females was noted. The correlation between DEPS scores and HbA1c levels for males was statistically significant (r = 0.26; P < 0.05).

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.

processing....