Table 2 Mean predicted incidence rate per period for the entire study area (for 1,000 person-week), and CAR random effects parameters estimated by the full model (M4) with their 95% credibility interval.

From: Predictive quality of census-based socio-economic indicators on Covid-19 infection risk at a fine spatial scale in France

Variable

Period

Low incidence

Growth

Peak and decrease

Stabilization

Mean predicted incidence ratea

0.01 [0.01, 0.01]

0.82 [0.79, 0.86]

2.81 [2.75, 2.88]

1.41 [1.37, 1.45]

Standard deviation \(\tau \in {\mathbb{R}}_{+}\)

1.16 [1.07, 1.25]

0.58 [0.56, 0.60]

0.51 [0.49, 0.52]

0.53 [0.52, 0.55]

\(\text{exp}\left(2\tau \right)\)b

10.2 [8.6, 12.3]

3.2 [3.1, 3.3]

2.8 [2.7, 2.9]

2.9 [2.8, 3.0]

Spatial correlation coefficient \(\rho \in (\text{0,1})\)

0.46 [0.28, 0.63]

0.78 [0.69, 0.85]

0.88 [0.83, 0.93]

0.59 [0.52, 0.66]

  1. aIncidence rate of SARS-CoV-2 infection in an IRIS of 1,000 inhabitants and at mean covariates level (/1000 person-week).
  2. bInterpretation: the standard deviation is additive around the mean log-incidence rate and multiplicative around the mean incidence rate: \(\text{incidence}\times \text{exp}(\pm 2\uptau )\) approximately gives the interval containing 95% of the incidence rates predicted in every IRIS.
Back to article page