Table 2.
Univariate and multivariate results of the Generalized Estimating Equation (GEE) model. The incidence ratio and the respective 95% confidence interval are shown.
| Univariate | Multivariate † | |||||
|---|---|---|---|---|---|---|
| Variable | Incidence Ratio | 95% C.I. | p-value | Incidence Ratio | 95% C.I. | p-value |
| Male Gender a | 1.14 | (1.07, 1.22) | <0.001 | 1.12 | (1.09, 1.15) | <0.001 |
| PODP intervention b | 1.34 | (1.25, 1.44) | <0.001 | 1.49 | (1.34, 1.64) | <0.001 |
| Post-PODP b | 0.86 | (0.80, 0.92) | <0.001 | 0.88 | (0.81, 0.97) | 0.007 |
| Lunch-time c | 0.40 | (0.21, 0.77) | 0.006 | 1.16 | (0.28, 4.75) | 0.83 |
| Evening rush hour c | 1.50 | (0.54, 4.19) | 0.44 | 1.10 | (0.39, 3.04) | 0.86 |
| Number of Trains | 1.06 | (1.05, 1.07) | <0.001 | 1.06 | (1.05, 1.07) | <0.001 |
a female gender is reference; b no PODP is reference; c morning rush hour is reference; † multivariate model contains gender, PA message, post-PA message, and number of trains.