Summary of estimated posterior distributions for the fixed effects, containing the model estimates with associated uncertainty, on the rate ratio scale
| Mean | SD | 2.5% | 50% | 97.5% | Mode | |
| α—iIntercept | 0.208 | 0.008 | 0.192 | 0.208 | 0.225 | 0.208 |
| β1—log(GP practice patient list size) | 0.915 | 0.002 | 0.911 | 0.915 | 0.919 | 0.915 |
| β2—drug group (5-α reductase inhibitors) | 0.536 | 0.009 | 0.518 | 0.535 | 0.554 | 0.536 |
| β3—GP run practices | 1.269 | 0.042 | 1.190 | 1.268 | 1.352 | 1.269 |
| β4—dispensing GP practice | 0.918 | 0.024 | 0.871 | 0.917 | 0.966 | 0.918 |
| β5—Men aged ≥45 years | 1.049 | <0.001 | 1.049 | 1.049 | 1.050 | 1.049 |
| β6—socially deprived area | 1.017 | 0.007 | 1.003 | 1.017 | 1.031 | 1.017 |
| β7 —remote/rural GP practice | 1.002 | <0.001 | 1.002 | 1.002 | 1.003 | 1.002 |
The posterior distribution associated to the patient list size effect estimate β1 was not transformed and should be read on the outcome scale, as explained in-text.
GP, general practitioner.