Table 3.
Point estimates and 95% confidence intervals of standardized group-specific intercepts and slopes in longitudinal regression model predicting weight change since baseline at the 6-month, 12-month, and 18-month follow-up.
| Variable | Group | Value | Lower Confidence Interval | Upper Confidence Interval |
|---|---|---|---|---|
| Intercept | Newsletter | .0649 | .0019 | .1279 |
| Internet | .1104 | .0406 | .1803 | |
| Face-to-Face | .0273 | -.0402 | .0948 | |
| Slope: Time | Newsletter | .0985 | .0733 | .1238 |
| Internet | .0698 | .0427 | .0969 | |
| Face-to-Face | .0595 | .0334 | .0856 | |
| Slope: Change in Weighing Frequency | Newsletter | -.0099 | -.0452 | .0255 |
| Internet | -.1085 | -.1438 | -.0732 | |
| Face-to-Face | -.0713 | -.1146 | -.0280 | |
| Slope: Change in Restraint | Newsletter | -.0558 | -.0891 | -.0224 |
| Internet | -.1249 | -.1577 | -.0921 | |
| Face-to-Face | -.0291 | -.0688 | .0106 |
NOTE: Time has been coded as 0, 1, 2 respectively at the 6-month, 12-month, and 18-month follow-up. All other continuous variables have been standardized by subtracting off their baseline mean and dividing by their baseline standard deviation, using values listed in Table 1. Study group has been coded with the control arm (N=Newsletter) as the reference group. Treatment contrasts are given by a) the difference between the two active treatment arms (F=Face-to-Face, I=Internet), and b) the difference of their average of the two active treatment arms from the control arm.