Parameter estimates for trial arm, time, and their interaction, intention to treat analysis
| Outcome Measure | Complete case cohort (n=759) | Available case cohort (n=1201) | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Trial arm* | Time† | Time×trial arm | Trial arm* | Time† | Time×trial arm | ||||||||||||
| Estimate (SE) | P | Estimate (SE) | P | Estimate (SE) | P | Estimate (SE) | P | Estimate (SE) | P | Estimate (SE) | P | ||||||
| PCS scale (US 1998 NBS) | +1.59 (2.26) | 0.480 | +0.12 (0.40) | 0.771 | +0.38 (1.47) | 0.795 | +2.32 (1.73) | 0.181 | +0.08 (0.38) | 0.843 | −0.13 (1.36) | 0.923 | |||||
| MCS scale (US 1998 NBS) | −1.55 (2.93) | 0.597 | −0.83 (0.50) | 0.097 | −2.37 (1.85) | 0.200 | −1.74 (2.25) | 0.440 | −0.70 (0.49) | 0.147 | −1.02 (1.73) | 0.557 | |||||
| EQ-5D scale | −0.01 (0.07) | 0.904 | +0.01 (0.01) | 0.360 | +0.01 (0.05) | 0.807 | +0.05 (0.06) | 0.383 | +0.01 (0.01) | 0.570 | −0.01 (0.05) | 0.820 | |||||
| Brief STAI scale | +0.18 (1.07) | 0.866 | −0.01 (0.19) | 0.953 | +0.21 (0.72) | 0.765 | +0.10 (0.84) | 0.905 | +0.05 (0.18) | 0.791 | +0.09 (0.67) | 0.897 | |||||
| CESD-10 scale | −0.35 (1.46) | 0.812 | +0.04 (0.25) | 0.865 | +1.34 (0.95) | 0.157 | −1.20 (1.14) | 0.293 | -0.01 (0.24) | 0.974 | +0.99 (0.88) | 0.262 | |||||
PCS=physical component score; MCS=mental component score; NBS=norms based scoring; SE=standard error.
Data are based on multilevel models controlling for the intraclass correlation, all covariates, and the relevant baseline outcome measure. Parameter estimates can be interpreted as the observed difference in an outcome measure (for example, PCS) between levels of a predictor variable (for example, telehealth v usual care) when the intracluster correlation and all covariates are taken into account. For example, parameter estimate +1.59 for trial arm on the PCS scale indicates that patients receiving telehealth had a score 1.59 units higher than patients receiving usual care (reference category) when the intraclass correlation, all covariates, and the baseline PCS score are taken into account.
*Telehealth=0; usual care=1 (reference category).
†Baseline assessment=1, short term assessment (at four months)=2, long term assessment (at 12 months)=3 (reference category). For the time variable, the main effect tests the hypothesis that outcome measure differs between short and long term assessments while controlling for baseline scores and other covariates, including trial arm (that is, testing the effect of time on the outcome measure, while the effect of trial arm and all other covariates held constant).