Table 2.
CBT for Adolescent ADHD
| Observed Data | Multiple Imputationsa | ||||||
|---|---|---|---|---|---|---|---|
|
| |||||||
| Outcomes | Time-Months | Initial Study Assignment
|
Estimated Effect of CBT on Outcomes (95% Confidence Interval) p-value | Estimated Effect of CBT on Outcomes (95% Coverage Intervalb) | |||
| Wait List | CBT | ||||||
|
| |||||||
| WL/CBT | mean +/- SD (n) | WL/CBT | mean +/- SD (n) | ||||
| CGI | 0 | WL | 4.64 +/- 0.85 (22) | WL | 4.75 +/- 0.68 (24) | -1.18 (-1.31, -1.05)c | |
| 4 | WL | 4.55 +/- 0.80 (22) | CBT | 3.71 +/- 0.85 (21) | -1.17 (-1.39, -0.94) | ||
| p<.0001 (***) | |||||||
| 8 | CBT | 3.20 +/- 0.86 (15) | CBT | 3.71 +/- 0.96 (21) | |||
|
| |||||||
| Parent Rating-Total | 0 | WL | 28.05 +/- 9.57 (22) | WL | 25.30 +/- 9.17 (24) | -11.47 (-12.50, -10.44-)c | |
| 4 | WL | 24.18 +/- 8.23 (22) | CBT | 16.76 +/- 10.40 (21) | -10.93 (-12.93,-8.93) | ||
| p<.0001 (***) | |||||||
| 8 | CBT | 12.82 +/- 6.51 (15) | CBT | 13.71 +/- 9.65 (21) | |||
|
| |||||||
| Adolescent Rating-Total | 0 | WL | 14.41 +/- 8.35 (22) | WL | 16.46 +/- 10.90 (24) | -5.81 (-6.50, -5.11)c | |
| 4 | WL | 13.86 +/- 8.12 (22) | CBT | 11.95 +/- 8.89 (21) | -5.24 (-7.21, -3.28) | ||
| p<.0001 (***) | |||||||
| 8 | CBT | 7.42 +/- 3.71 (15) | CBT | 9.71 +/- 7.21 (21) | |||
Notes:
a
This set of results is from a mixed model using multiple imputations to estimate the main effects of treatment and group assignment on the outcome.
b
The estimated 95% coverage interval is the 2.5th and 97.5th percentiles of the sampling distribution of the estimated effects.
c
In an analysis of 100 data sets with imputed values for cases with missing data, there was a statistically significant (p<0.0001) effect of CBT on this outcome in all 100 iterations.
Abbreviations: WL-Waitlist, CBT-Cognitive Behavioral Therapy, SD-Standard Deviation, CGI-Clinical Global Impression