Table 2.
Ratios of vegetable-rich meal orders and daily restaurant sales (during the intervention period vs the control period)
| Non-adjusted | Model 1 | Model 2 | |
|---|---|---|---|
| Vegetable-rich meal orders | |||
| Intervention (ref. control) | 1.33 (1.18, 1.49) | 1.35 (1.21, 1.52) | 1.50 (1.29, 1.75) |
| Weekends (ref. weekdays) | 1.18 (0.98, 1.42) | ||
| Temperature (per 1 °C increase) | 0.91 (0.86, 0.97) | ||
| Humidity (per 10% point increase) | 1.06 (0.99, 1.13) | ||
| Weather: Rain (ref. not rain)a | 0.89 (0.74, 1.08) | ||
| Adjusted for fixed effects of restaurants | No | Yes | Yes |
| Restaurant sales | |||
| Intervention (ref. control) | 1.14 (0.63, 2.05) | 1.15 (0.82, 1.61) | 1.77 (1.11, 2.83) |
| Weekends (ref. weekdays) | 1.22 (0.75, 1.98) | ||
| Temperature (per 1 °C increase) | 0.81 (0.69, 0.96) | ||
| Humidity (per 10% point increase) | 0.92 (0.78, 1.07) | ||
| Weather: Rain (ref. no rain)a | 0.70 (0.44, 1.12) | ||
| Adjusted for fixed effects of restaurants | No | Yes | Yes |
aBased on weather during business hours. We estimated the ratios for vegetable-rich meal orders and restaurant sales separately. Model 1 adjusted for fixed effects of restaurants by adding the dummy variables identifying restaurants. In Model 2, we further added daily temporal and climatic data as covariates