Table 1.
Model 1 (base) | Model 2 | Model 3 | Model 4 | Model 5 | Model 6 | Model 7 | |
---|---|---|---|---|---|---|---|
COVID-19 ICU use* | 2.84 [0.6, 5] | 2.71 [0.4, 5] | 1.79 [0.7, 2.9] | 2.48 [1.3, 3.7] | 2.46 [1.3, 3.6] | 2.99 [0.7, 5.3] | 3.31 [0.9, 5.7] |
COVID-19 non-ICU use* | 17.84 [11.1, 24.6] | 17.23 [10.1, 24.4] | 18.89 [12.3, 25.5] | 6.15 [2.2, 10.1] | 6.41 [2.7, 10.2] | 17.81 [11, 24.6] | 17.37 [10.7, 24.1] |
3-day lags of COVID-19 ICU use | Yes | Yes | No | No | No | Yes | Yes |
3-day lags of COVID-19 non-ICU use | Yes | Yes | Yes | No | No | Yes | Yes |
Cumulative # of cases | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
Cases in days 0 to −6 | Yes | No | Yes | Yes | No | Yes | Yes |
Week of year indicators | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
Day of week indicators | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
State indicators | Yes | Yes | Yes | Yes | Yes | Yes | Yes |
Estimator | GEE | GEE | GEE | GEE | GEE | GEE | GEE |
Specification |
Log link Neg Bin |
Log link Neg Bin |
Log link Neg Bin |
Log link Neg Bin |
Log link Neg Bin |
Log link Neg Bin |
Log link Neg Bin |
N | 1056 | 1062 | 1056 | 1118 | 1131 | 1056 | 1056 |
# Of states | 23 | 23 | 23 | 23 | 23 | 23 | 23 |
Error correlation | AR1 | AR1 | AR1 | AR1 | AR1 | AR2 | AR3 |
Standard errors | Robust | Robust | Robust | Robust | Robust | Robust | Robust |
Notes: In our analysis sample, mean (SD, 25th percentile–75th percentile) of key variables were the following: COVID-19 ICU use was 19.6 (19.22, 7.8–22.1); COVID-19 non-ICU use was 5.5 (6.3, 1.5–8.8); total ICU bed capacity, 1678 (1677, 643–1891); total non-ICU bed capacity, 12,978 (12,418, 5581–15,462); COVID-19 deaths over the next 7 days, 234 (358, 40–276)
Minimum number of days of data contributed by a state was 6. Maximum number was 65
GEE, generalized estimating equation; AR1, autoregressive one-error structure (also known as first-order autoregressive process); AR2 and AR3 are similarly defined
*The ICU use was calculated as the number of current ICU patients with COVID-19 and then expressed as a percentage of the total ICU beds of that state. The non-ICU use was calculated similarly