Table 1

 Baseline characteristics of intervention and control practices. Figures are means (SD) unless otherwise stated

Intervention group (n=64) Control group No in control group Standardised difference* P value†
Outcomes in high risk cohort‡ before intervention (period 3§)
Emergency admission/patient/year 0.56 (0.50) 0.58 (0.36) 7811 −0.05 0.70
Emergency bed days/patient/year 6.96 (5.25) 6.63 (7.37) 7799 0.05 0.63
HES mortality 0.09 (0.09) 0.10 (0.13) 7710 −0.07 0.436
Outcomes in population aged ≥65 (period 3§)
Emergency admission/patient/year 0.23 (0.07) 0.21 (0.07) 7695 0.38 0.01
Emergency bed days/patient/year 3.042 (0.86) 2.56 (1.00) 7672 0.482 0.00
HES mortality 0.05 (0.02) 0.05 (0.02) 7653 0.212 0.11
Other characteristics of practices
Average growth in admissions¶ 0.13 (0.26) 0.03 (0.20) 7589 0.48 0.00
Total on practice list 7596 (3325) 6148 (3745) 7589 0.39 0.00
Index of multiple deprivation-health 0.64 (0.89) 0.26 (0.93) 7589 0.40 0.00
Proportion aged ≥65 0.16 (0.05) 0.15 (0.05) 7589 0.06 0.60
Proportion of women aged ≥65 0.57 (0.04) 0.56 (0.04) 7589 0.19 0.08
Number in high risk population 32.24 (17.17) 24.00 (18.69) 7589 0.44 0.00
Proportion of high risk population 0.03 (0.01) 0.026 (0.01) 7589 0.32 0.03
List/whole time equivalent GP 2138 (628) 2125 (599) 7589 0.02 0.87
Low income scheme index** 12.56 (7.12) 11.48 (7.86) 7589 0.14 0.23

*Standardised difference: intervention group mean minus control group mean divided by pooled SD.

t test for differences in means.

‡N=62 as HES data for 2000-1 for two intervention practices could not be found.

§October 2002 to March 2003. Pre-intervention high risk cohort: 65+, 2+emergency admissions in 13 months to 31 January 2001, not recorded as dead in HES by 1 July 2001. Post-intervention high risk cohort population: number of patients 65+, 2+ emergency admissions in 13 months to 31 January 2003, not recorded as dead in HES by 1 July 2003.

¶Growth rate in emergency admissions in those aged ≥65 is geometric mean of growth rates in four years to 2002-3.

**Proportion of prescriptions dispensed without charge on grounds of low income.