Linda M Delahanty, MS, RD
Massachusetts General Hospital Diabetes Center
The DPP demonstrated that intensive lifestyle intervention achieved a 7% weight loss at 1 year and a 5% weight loss at study end and reduced the incidence of diabetes by 55% over 3.2 years in participants with impaired glucose tolerance. We examined which specific cognitive-behavioral factors predicted weight loss at 6 months, 1 year and study end for 274 of 1079 lifestyle participants from 18 of 27 DPP centers. The lifestyle subgroup completed questionnaires including the Exercise Self-Efficacy Scale, Weight Efficacy Lifestyle Questionnaire, Fat-Related Diet Questionnaire, Perceived Stress Questionnaire, Emotional Eating Questionnaire, Restraint Scale (Dutch Eating Behavior Questionnaire) and a subset of the Binge Eating Scale. Activity level was assessed using the Modifiable Activity Questionnaire (1 year recall) and the Low Level Physical Activity Recall (7 day recall). This subgroup was similar to the entire DPP lifestyle cohort at baseline: mean age was 52+/-12 (SD) yrs, mean baseline BMI 33.9+/-6.9 kg/m2, 65% were women, 62% were Caucasian, 20% African American, 14% Hispanic, and 4% Asian.
Psychological and Behavioral Predictors of 6 Month and 1 Year Weight Outcomes
The mean weight loss of this subgroup of DPP lifestyle participants was 7.3% at 6 months and 7.8% at 1 year with 53% achieving the 7% weight loss goal at 6 months and 51% achieving the 7% weight loss goal at 1 year, similar to the entire DPP lifestyle cohort. Stepwise logistic regression analyses adjusted for baseline weight found that the most important determinants of meeting the 7% weight loss goal at 6 months were greater dietary restraint (p=.035) and less frequent emotional eating (p=.002) at baseline, and greater improvements in dietary restraint (p=.0002) and low fat diet efficacy (p=.013) from baseline to 6 months. Stepwise multivariate regression analyses revealed that the most important predictors of greater mean percent weight loss at 6 months were fewer previous weight loss programs (p=.0002), more frequent weight cycling (p=.003), less high fat diet behavior at baseline (p<.0001) and improvements in low fat diet self-efficacy (p=.01), high fat diet behaviors (p<.0001) and dietary restraint skills (p=.015) from baseline to 6 months.
Significant baseline predictors of achieving a 7% weight loss at 1 year were lower perceived stress (p=.001), and less frequent high fat intake behaviors (p=.0006). Improvements in high fat intake behaviors (p=.029) and low fat diet self-efficacy (p=.01) from baseline to 6 months and improvements in dietary restraint skills (p=.003) from baseline to 1 year also predicted achievement of a 7% weight loss at 1 year. Stepwise multivariate regression analyses revealed that history of weight cycling (p<.0001) and greater baseline dietary restraint (p=.005) were positive predictors and frequency of previous self-imposed (p=.03) or formal weight loss programs (p=.015) were negative predictors of greater mean percent weight loss at 1 year. Improvements in dietary restraint skills from baseline to 6 months (p<.0001) and in low fat diet efficacy from baseline to 1 year (p=.001) were also independent predictors of greater mean percent weight loss at 1 year.
Conclusions: These results suggest that programs that foster dietary restraint skills and focus on the ability to change high fat intake behaviors and the self confidence to do so may yield the best weight loss results, especially in those who are not emotional eaters, have low levels of perceived stress and have not tried many weight loss programs in the past.
Psychological and Behavioral Predictors of Long-Term Weight Loss
Analysis after 3.2 years has revealed a mean weight loss of 5.3% +/- 7.3% with 41% (111 of 273) maintaining at least 7% weight loss in this subgroup, again similar to the entire DPP lifestyle cohort. Using stepwise linear regression analyses, the most important determinants of long-term mean % weight loss were: older age (p=.001); lower activity (p=.014) , higher weight (p=.009) and fewer previous weight loss programs (p=.009), all at baseline; less high fat diet behavior at 1 year (p=.0002) and improved dietary restraint baseline to 1 year (p=.013). Total R2 for all predictors combined was 22.4%. The most important determinants of meeting the 7% weight loss goal were: older age (p=.0001); lower baseline activity (p=.001) and higher baseline weight loss self-efficacy (p=.04); better dietary restraint (p=.02) at 1 year; and improvements in low fat diet self-efficacy (p= .009) and activity (p=.02) from baseline to 1 year.
Conclusions: Those DPP participants who were older and presented with higher initial weight, lower activity levels, higher self-confidence in ability to lose weight and had been in fewer weight loss programs lost more weight long-term. The results suggest that programs that foster dietary restraint skills, increased activity, reductions in high fat diet behaviors and increased self-efficacy regarding ability to follow a low fat diet may yield the best long-term weight loss results.
Completeness of Follow up
The completeness of follow up was excellent in this study with 98% of questionnaires returned at 6 months, 95% at 1 year and 93% at study end. The next steps are to analyze data on the psychological and behavioral factors associated with weight regain/relapse.
- Delahanty LM, Hayden D and Nathan DM, Psychological and Behavioral Correlates of Baseline Body Weight in the Diabetes Prevention Program. Diabetes 2001; 50 (Suppl. 2), A390.
- Delahanty L, Meigs J, Hayden D, Nathan D and the DPP Research Group. Psychological and Behavioral Predictors of Weight Outcomes in the Diabetes Prevention Program (DPP). Diabetes 2002; Vol 51 (Suppl 2): A447.
- Delahanty LM, Meigs JB, Hayden D, Williamson DA, Nathan DM and the DPP Research Group. Psychological and Behavioral Correlates of Baseline BMI in the Diabetes Prevention Program (DPP). Diabetes Care 2002; 25(11): 1992-1998.
- Delahanty L, Meigs J, Hayden D, Nathan D and the DPP Research Group. Psychological and Behavioral Predictors of Long-Term Weight Loss in the Diabetes Prevention Program (DPP). Diabetes 2003; Vol 52 (Suppl 1): A411.