The sample consisted of 35 healthy, perimenopausal women 40 to 59 years old. The mean age for the participating sample was 49.6 years old (SD = 6.8). Most participants had at least some college education (89%) with an annual family income of $50,000 or more (58%). The sample was tri-ethnic: 54% white, 29% Hispanic/Latino, and 17% African American. The average BMI was 31.8 kg/m2 (SD = 6.8), ranging from 22.1 kg/m2 to 54.4 kg/m2. Participants reported eating out on average 5.6 times per week (SD = 3.4). Current eating patterns at the start of the study varied, with 31% dieting, 23% eating to maintain weight, and 46% eating whatever they wanted. Of the 43 participants who consented to participate, 35 completed the study for an attrition rate of 19%. There were no significant differences at baseline in age, BMI, or reported frequency of eating out between those who completed and those who dropped out of the study.
There were no significant differences at baseline between the intervention and control groups for age, educational level, ethnicity, income, reported frequency of eating out, time spent dieting, EES scores, SEEB scores, or BarriersRE scores. Additionally, there were no significant differences at baseline between the intervention and control groups for daily caloric and fat intake or amount of calories and fat consumed during restaurant eating episodes. However, the intervention group had a significantly [t(33) = 2.12; P = .04] higher average BMI of 33.9 kg/m2 (SD =7.2) at baseline compared to the BMI (M= 29.3 kg/m2, SD = 5.5) of the control group. The waist circumference was also significantly [t(33) = 2.44; P = .02] higher for the intervention group at baseline than for the control group. The means and standard deviations for the dependent variables for Time 1 and Time 2 by group assignment are presented in .
Means and Standard Deviations for Time 1 and Time 2 Outcome Variables by Group Assignment (n=35)
Weight and Waist Circumference Outcomes
Results of the regression analysis indicated that the intervention group had significantly less weight gain at Time 2 than the control group after controlling for Time 1 weight (β=−.04, t=−2.23, P=.03, R2=.989) (). In addition, the intervention group also had a decrease in waist circumference at Time 2, relative to the control group which was not statistically significant (β=−.10, t=−1.72, P=.096, R2=.917).
Summary of Regression Analysis for Variables Predicting Intervention Outcomes (n = 35)
Caloric and Fat Intake Outcomes
Compared to the control group, the average number of calories consumed during a 3-day period was significantly lower at Time 2 for the intervention group (β=−.44, t=−3.43, P=.002, R2=.486). The average fat intake was also significantly lower for the intervention group relative to the control group (β=−.45, t=−3.66, P=.001, R2=.521). Despite a trend for the intervention group to have a decrease in caloric (β=−.28, t=−1.87, P=.07, R2=.280) and fat intake (β=−.27, t=−1.59, P=.12, R2=.085) per restaurant eating episode at Time 2, it did not reach statistical significance in comparison to the control group.
Impact on Emotional Eating, Self-Efficacy, and Barriers
There was no significant difference in total EES scores at Time 2 (β=−.073, t=−.55, P=.59, R2=.477) between the intervention and control group. At the end of the intervention, diet-related self-efficacy was significantly higher among intervention participants (β=.36, t=2.54, P=.016, R2=.397), and the intervention group experienced fewer barriers to managing intake when eating out (β=−.38, t=−3.61, P=.001, R2=.687) than the control group.
Intention to Treat Analysis
The intention to treat analysis modeled the results as if all 43 participants in the study, including the 8 that withdrew prior to completing Time 2 data collection, had completed the study. Results of the multiple imputations analysis were consistent with the findings for the 35 participants who completed the study. For instance, when all 43 cases were considered, the intervention group still reported significantly less weight gain, and lower average daily caloric and fat intake than the control group. Models based on the imputed data also indicated that the intervention group perceived greater diet-related self-efficacy and fewer barriers at Time 2 relative to the control group.