Table 2. Secondary outcome results.
App-based CBT-I, mean (SE)a | App-based HE, mean (SE) | Adjusted difference (95% CI) | p value | Cohen’s db | |
---|---|---|---|---|---|
PHQ-9 | |||||
Baseline | 12.1 (0.2); n = 354 | 12.1 (0.2); n = 354 | |||
Post-session 2 | 7.6 (0.2); n = 291 | 7.7 (0.2); n = 285 | −0.1 (−0.7, 0.5) | 0.72 | 0.06 |
Post-session 4 | 6.1 (0.2); n = 256 | 7.0 (0.2); n = 264 | −1.0 (−1.6, −0.4) | 0.002 | 0.50 |
Post-intervention | 5.1 (0.2); n = 314 | 6.2 (0.2); n = 319 | −1.0 (−1.6, −0.5) | <0.001 | 0.53 |
6 months | 5.4 (0.2); n = 273 | 6.3 (0.2); n = 268 | −0.8 (−1.4, −0.2) | 0.009 | 0.42 |
12 months | 5.5 (0.2); n = 249 | 6.0 (0.2); n = 227 | −0.5 (−1.1, 0.2) | 0.14 | 0.24 |
PHQ-8 c | |||||
Baseline | 9.6 (0.2); n = 354 | 9.6 (0.2); n = 354 | |||
Post-session 2 | 6.1 (0.2); n = 291 | 6.2 (0.2); n = 285 | −0.1 (−0.6, 0.5) | 0.83 | 0.03 |
Post-session 4 | 4.9 (0.2); n = 256 | 5.6 (0.2); n = 264 | −0.7 (−1.2, −0.1) | 0.02 | 0.39 |
Post-intervention | 4.1 (0.2); n = 314 | 5.0 (0.2); n = 319 | −0.9 (−1.4, −0.4) | 0.001 | 0.51 |
6 months | 4.5 (0.2); n = 273 | 5.1 (0.2); n = 268 | −0.7 (−1.2, −0.1) | 0.01 | 0.39 |
12 months | 4.5 (0.2); n = 249 | 4.9 (0.2); n = 227 | −0.4 (−1.0, 0.2) | 0.18 | 0.23 |
ISI | |||||
Baseline | 18.3 (0.2); n = 354 | 18.3 (0.2); n = 354 | |||
Post-session 2 | 11.5 (0.2); n = 291 | 12.1 (0.2); n = 285 | −0.6 (−1.3, 0.1) | 0.12 | 0.22 |
Post-session 4 | 9.0 (0.2); n = 256 | 10.7 (0.2); n = 264 | −1.7 (−2.4, −1.0) | <0.001 | 0.66 |
Post-intervention | 7.3 (0.2); n = 313 | 9.3 (0.2); n = 319 | −2.0 (−2.7, −1.3) | <0.001 | 0.78 |
6 months | 7.2 (0.2); n = 272 | 8.4 (0.2); n = 268 | −1.1 (−1.9, −0.4) | 0.002 | 0.45 |
12 months | 7.3 (0.2); n = 249 | 8.2 (0.2); n = 227 | −0.8 (−1.6, −0.1) | 0.03 | 0.32 |
Abbreviations: CBT-I, cognitive behavioral therapy for insomnia; HE, health education; ISI, Insomnia Severity Index; PHQ, Patient Health Questionnaire.
aAll values are adjusted for sex and insomnia severity at baseline.
bCohen’s d is defined as the adjusted intervention effect divided by the square root of estimated sample variance derived from the linear mixed-effects model.
cExcluding sleep item.