Table 4

Multilevel regression analysis of the explanatory variables on the overall range of different medicines prescribed by GPs in 2001

Empty model

Full model


Estimate (standard error)

Estimate (standard error)

Hypothesis

practice level fixed effects

- prescriptions per patient

0.2583 (0.1005)*

Confirmed

GP level fixed effects:

- visits representatives industry

0.9289 (0.0456)*

Confirmed

- individual list size

0.0000765(0.0002074)*

Confirmed

variance components:

practice level intercept

0.1276(0.1521)

0.2684(0.1424)

GP level intercept

0.8419(0.1749)*

0.5914(0.1324)*

intraclass correlation

13.2%

31.2%


*p < 0. 05, one-tailed; N(practices) = 93; n(GPs) = 138.

de Bakker et al. BMC Health Services Research 2007 7:132   doi:10.1186/1472-6963-7-132

Open Data