Appendix 3.
Linear multiple regression models of demographic variables, clustered by PCT (n = 596)
| Linear regression: satisfaction Multiple regression — demographic Variables (R2 = 0.13) 2008/2009 | Coefficient | Standard error | Beta | P>|z| | 95% CI |
|---|---|---|---|---|---|
| Practice size | 0.00 | 0.00 | -0.08 | 0.31 | 0.00 to 0.00 |
| Index of Multiple Deprivation | 0.00 | 0.00 | 0.31 | 0.02a | 0.00 to 0.00 |
| QOF score | 0.00 | 0.00 | 0.08 | 0.08 | 0.00 to 0.00 |
| Percentage patients >45 years | 0.22 | 0.07 | 0.30 | 0.01a | 0.06 to 0.38 |
| Proportion UK graduate doctors | 0.01 | 0.01 | 0.04 | 0.49 | -0.01 to 0.03 |
| _cons | 0.57 | 0.07 | <0.001 | 0.41 to 0.74 | |
| Linear regression: satisfaction Multiple regression — demographic Variables (R2 = 0.02) 2007/2008 to 2008/2009 | Coefficient | Standard error | Beta | P>|z| | 95% CI |
| Practice size | 0.00 | 0.00 | -0.09 | 0.17 | 0.00 to 0.00 |
| Index of Multiple Deprivation | 0.00 | 0.00 | -0.11 | 0.15 | 0.00 to 0.00 |
| QOF score | 0.00 | 0.00 | 0.01 | 0.85 | 0.00 to 0.00 |
| Percentage patients >45 years | 0.03 | 0.05 | 0.05 | 0.56 | -0.07 to 0.13 |
| Proportion UK graduate doctors | 0.00 | 0.01 | 0.02 | 0.75 | -0.01 to 0.01 |
| _cons | -0.02 | 0.05 | 0.72 | -0.13 to 0.10 | |
| Linear regression: satisfaction Multiple regression — demographic Variables (R2 = 0.12) 2008/2009 | Coefficient | Standard error | Beta | P>|z| | 95% CI |
| Practice size | 0.00 | 0.00 | 0.19 | 0.02a | 0.00 to 0.00 |
| Index of Multiple Deprivation | 0.00 | 0.00 | -0.15 | 0.23 | 0.00 to 0.00 |
| QOF score | 0.00 | 0.00 | 0.01 | 0.85 | 0.00 to 0.00 |
| Percentage patients >45 years | -0.32 | 0.14 | -0.33 | 0.04a | -0.62 to-0.01 |
| Proportion UK graduate doctors | 0.00 | 0.01 | 0.01 | 0.79 | -0.02 to 0.03 |
| cons | 0.66 | 0.11 | <0.001 | 0.43 to 0.89 | |
a
These relationships could not be confirmed in simple linear regression models, clustered at PCT level and should be interpreted with caution. The relationships in these models are however compatible with national data, indicating that an older patient profile is associated with lower demand and higher satisfaction (beta −0.35 and 0.30 respectively, n = 77 862 practices). Deprivation is associated with higher satisfaction, but is not related to lower demand, in keeping with national data (beta 0.13 and 0.02 respectively for satisfaction and demand, n = 7 862 practices).