The provision of care for patients with epilepsy in the UK is fragmented and inadequate. Lack of support in the community, poor communication across the primary–secondary care interface and inaccessibility of specialist services are contributory factors. The inclusion of epilepsy as a quality marker in the new general practitioner (GP) contract provides an opportunity to improve co-ordination of services.Prior to the acceptance of the New GP Contract in April 2004, the first part of this audit aimed to (a) assess the documentation of the process of care of patients with epilepsy in the city of Chester; and (b) provide GPs with a clinical resource to facilitate review of patients and targeting of referrals to secondary care.The case notes of 610 patients were scrutinised to identify the proportion (a) receiving regular review, (b) in whom diagnosis was uncertain, (c) in documented remission, (d) who were women of child-bearing age, (e) who were non-compliant with treatment. A computerised template, including specific suggestions, was produced for every patient but responsibility for review remained with each GP. Re-audit in 2 years was planned.Two hundred and fifty (41%) had been seen by a GP in the previous year. One hundred and sixteen (19%) had regular review arrangements. One hundred and thirty-eight (22.6%) had shared care of whom 79 had seen their GP in previous year. Three hundred and one (49%) had not seen any doctor concerning their epilepsy in the previous year. Forty-three percent of adults who should be receiving shared care were not doing so.In 67 (11%) cases, there was diagnostic uncertainty. Only 178 (29.2%) patients were in documented remission. One hundred and nine (17.8%) patients were women of child-bearing age of whom 37 were receiving sodium valproate. In 84 (14%), there was evidence of non-compliance.Haphazard follow-up arrangements preclude review of diagnostic accuracy, estimation of remission rate, assessment of adherence to prescribed treatment and awareness of level of patients’ knowledge especially women of child-bearing age on teratogenic drugs. One cannot identify those patients who can be managed solely in the community and those who merit reassessment in hospital clinics. Assuming these findings are representative, primary and specialist secondary care face an enormous task in reorganising services to ensure that all patients receive appropriate care. This project describes a method of improving local co-ordination of patient care.