Table 5.
Path coefficients from the parallel mediation model illustrated in Fig. 3
| Predictor variable | VC credibility | VC attractiveness | Message relevance | CRC screening intention | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| B (SE) | P* | 95% CI | B (SE) | P* | 95% CI | B (SE) | P* | 95% CI | B (SE) | P* | 95% CI | |
| Interaction | ||||||||||||
| Race-matching × participant race | 0.56 (0.22) | 0.01 | (0.12, 1.00) | 0.93 (0.18) | <0.001 | (0.58, 1.29) | 0.47 (0.18) | 0.01 | (0.13, 0.82) | 0.29 (0.18) | 0.10 | (−0.05, 0.64) |
| Mediator effects | ||||||||||||
| Message relevanceM1 | – | – | – | – | – | – | – | – | – | 0.91 (0.04) | <0.001 | (0.83, 0.99) |
| VC attractivenessM2 | – | – | – | – | – | – | – | – | – | 0.51 (0.05) | <0.001 | (0.42, 0.60) |
| VC source credibilityM3 | – | – | – | – | – | – | – | – | – | 0.92 (0.06) | <0.001 | (0.80, 1.05) |
| Adjusted interaction effects | ||||||||||||
| Race-matching × participant raceM1 | – | – | – | – | – | – | – | – | – | 0.22 (0.19) | 0.25 | (−0.15, 0.58) |
| Race-matching × participant raceM2 | – | – | – | – | – | – | – | – | – | 0.07 (0.18) | 0.70 | (−0.28, 0.43) |
| Race-matching × participant raceM3 | – | – | – | – | – | – | – | – | – | 0.17 (0.18) | 0.34 | (−0.18, 0.53) |
Abbreviation: VC, virtual clinician. CRC, colorectal cancer.
Note: Em-dash represents a relationship that was not regressed upon as part of the model.
*
Parallel mediation conducted using the ordinal package in R 4.0.1. All statistical tests were two-sided. B = unstandardized beta-coefficient; CI = confidence interval.