Skip to main content
. 2016 Jun 13;9(2):559–581. doi: 10.1007/s12561-016-9150-3

Table 1.

Relative risks (RR) of respiratory mortality, with 95 % confidence intervals for an increase of 10 ppb of BS over the previous 27 years

Without deprivation With deprivation
Method 1: observed exposures only
   RR 95 % CI RR 95 % CI
   1.037 1.025–1.050 1.038 1.023–1.049
Method 2: predictions
   RR 95 % CI RR 95 % CI
   1.022 1.014–1.030 1.021 1.013–1.029
Method 3: observed data and predictions combined
   RR 95 % CI RR 95 % CI
   1.011 1.004–1.018 1.010 1.003–1.017

Exposure values are obtained using three methods: (1) using observed data; (2) using predictions from a spatio–temporal model; (3) using observed data combined with predictions to fill in missing values. Risks are estimated with and without adjustment for deprivation. Results for methods 2 and 3 are from multiple imputation using 100 datasets (see text for details)