Adjusted HRs of COVID-19 risk according to healthful Plant-Based Dietary Index scores stratified by sociodemographic and clinical characteristics
| Factor | No of events/person-months* | HR per 1 SD increase in diet quality score | P value |
| Age | |||
| <60 | 25 329/2 285 329 | 0.94 (0.93–0.95) | |
| ≥60 | 6486/1 600 945 | 0.94 (0.92–0.97) | 0.63 |
| Sex | |||
| Male | 9338/1 232 656 | 0.95 (0.93–0.97) | |
| Female | 22 428/2 647 254 | 0.96 (0.95–0.98) | 0.21 |
| Race | |||
| White | 30 335/3 736 972 | 0.96 (0.95–0.97) | |
| Non-white | 1480/149 303 | 0.96 (0.91–1.01) | 0.95 |
| Socioeconomic deprivation† | |||
| High | 5244/456 271 | 0.94 (0.91–0.96) | |
| Intermediate | 13 172/1 567 516 | 0.96 (0.94–0.98) | |
| Low | 13 399/1 862 489 | 0.97 (0.95–0.99) | 0.04 |
| Population density, km2 no | |||
| <2000 | 10 581/1 490 084 | 0.96 (0.94–0.98) | |
| ≥2000 | 21 234/2 396 190 | 0.96 (0.94–0.97) | 0.74 |
| Healthcare worker | |||
| Yes | 2908/140 087 | 0.95 (0.92–0.99) | |
| No | 28 907/3 638 588 | 0.96 (0.95–0.97) | 0.78 |
| Body mass index, kg/m2 | |||
| <25 | 13 989/1 905 517 | 0.96 (0.94–0.97) | |
| 25–30 | 9854/1 252 222 | 0.96 (0.94–0.98) | |
| ≥30 | 7972/728 536 | 0.96 (0.94–0.98) | 0.73 |
| Physical activity | |||
| <1 day/week | 6751/683 562 | 0.94 (0.91–0.96) | |
| 1–4 days/week | 19 476/2 425 198 | 0.96 (0.94–0.97) | |
| ≥5 dasy/week | 5588/777 515 | 0.99 (0.96–1.01) | 0.01 |
Association between predicted COVID-19 and diet quality according to sociodemographic and clinical characteristics.
*Number of observations varies among imputations.
†Socioeconomic deprivation categories were based on deciles of the deprivation index (methods). Cox models were adjusted for the same covariates as previous model 3. P values obtained using the Q test for heterogeneity.