Skip to main content
. 2021 Nov 29;9:754696. doi: 10.3389/fpubh.2021.754696

Table 2.

Model predicted policy effects on case rates, doubling time, and death rates for all US counties.

Case rates Doubling time Death rates
Log beta (95% CI) Log beta (95% CI) Log beta (95% CI)
Fixed intervention effect
Level 1 0.768 (0.56–0.97)††† 0.059 (−0.03–0.15) 2.371 (1.18–3.57)††
Level 2 1.373 (1.14–1.6)††† 0.082 (−0.03–0.19) 2.999 (1.8–4.2)†††
Level 3 1.516 (1.28–1.75)††† 0.072 (−0.04–0.19) 3.187 (2–4.38)†††
Level 4 1.707 (1.47–1.94)††† 0.143 (0.03–0.26) 3.539 (2.35–4.72)†††
Duration of intervention effect
Level 1 0.03 (0.01–0.05)†† 0.004 (−0.01–0.01) −0.018 (−0.05–0.01)
Level 2 −0.018 (−0.04–0) 0.012 (0–0.02) 0.028 (0–0.05)
Level 3 −0.012 (−0.02–0) 0.011 (0–0.02) 0.062 (0.04–0.08)†††
Level 4 −0.032 (−0.04–0.02)††† 0.020 (0.01–0.03)††† 0.056 (0.04–0.07)†††
†††

p < 0.0001,

††

p < 0.01,

p < 0.05.

The log beta values represent the log change in the outcome when an intervention policy is in place or the change for the number of days the policy is in effect. Results are from three negative binomial models using all counties from the US (Supplementary Table B0). Case and Doubling Time Models included the same covariates, except for the addition of population density in the Doubling Time Model. The model for Death Rates was reduced to ease model fit.