Association of primary care factors with hospital admissions for epilepsy in England, 2004–2010: National observational study

PURPOSE There has been little research on the accessibility and quality of primary care services for epilepsy and emergency hospital admissions for epilepsy. METHODS We examined time trends in admissions for epilepsy in England between 2004-2005 and 2010, and the association of admission rates with population and primary care factors. The units of analysis were the registered populations of 8622 general practices. We used negative binomial regression to model indicators from the Quality and Outcomes Framework, the UK's primary care pay for performance scheme, to measure the accessibility and quality of care for epilepsy, and supply of general practitioners, after adjustment for population factors. RESULTS The mean indirectly standardised admission rate decreased from 122.9 to 102.6 (-16.5%; P<0.001) over the study period, while the mean percentage of patients seizure free increased from 65.3% to 74.9% (P<0.001). In the multivariable analysis, a one unit increase in the percentage of seizure free adult patients on epilepsy drugs predicted a 0.20% decrease (IRR=0.9980; 95% CI: 0.9974-0.9986) in admission rate. The percentage of patients who were able to book a GP appointment over two days ahead predicted a 0.12% decrease (IRR=0.9988; 95% CI: 0.9982-0.9994). The deprivation score of practice populations (IRR=1.0179; P<0.001) and general practitioner supply (IRR=1.0022; P<0.001) were both positively associated with admission rates. CONCLUSION Patient access to primary care appointments and percentage of patients who have been recorded as seizure free for 12 months were associated with lower admission rates. However the effect sizes are small relative to that of population deprivation.