Table 3.
Multivariable logistic regression modelsa of predictors of digital health technology adoption by gender and COM-Bb domain.
| Predictorc | Telehealth, aORd (95% CI), P value | mHealthe app, aOR (95% CI), P value | Wearable device, aOR (95% CI), P value | Chatbot/virtual assistant, aOR (95% CI), P value | e-Pharmacy, aOR (95% CI), P value |
| Gender (1=female) | 1.51 (0.58-3.95), .403 | 2.59 (1.28-5.25), .008f | 0.61 (0.33-1.12), .113 | 0.32 (0.18-0.59), <.001f | 0.48 (0.22-1.01), .052g |
| Digital literacy (z-score) | 0.47 (0.24-0.92), .026f | 0.74 (0.42-1.29), .288 | 0.70 (0.40-1.21), .199 | 0.88 (0.54-1.44), .613 | 0.84 (0.50-1.38), .487 |
| Capability index (z-score) | 0.88 (0.44-1.73), .703 | 1.44 (0.81-2.58), .215 | 0.77 (0.47-1.26), .315 | 0.56 (0.35-0.89), .016f | 0.70 (0.43-1.15), .158 |
| Opportunity index (z-score) | 1.80 (1.15-2.83), .010f | 3.18 (2.11-4.77), <.001f | 1.93 (1.33-2.79), <.001f | 1.16 (0.84-1.59), .380 | 1.26 (0.89-1.76), .184 |
| Motivation index (z-score) | 1.50 (0.92-2.45), .101 | 0.93 (0.70-1.25), .627 | 1.02 (0.72-1.45), .914 | 1.26 (0.93-1.70), .135 | 1.50 (1.05-2.15), .026f |
aEach model was adjusted for age, education, occupation, region, and rural/remote residence.
bCOM-B: Capability, Opportunity, Motivation – Behavior.
cPredictors were standardized (z-scores), except gender, so each odds ratio represents the change in odds for a 1-SD increase in the predictor. Gender was entered as a binary categorical variable and was not standardized.
daOR: adjusted odds ratio.
emHealth: mobile health
fStatistically significant results at P<.05.
gP=.05-.10 (trend).