Workload implications of identifying patients with ischaemic heart disease in primary care: population-based study.

BACKGROUND The aims of this study were to develop ischaemic heart disease (IHD) registers in three primary care groups (PCGs) in SW London; to determine what proportion of patients with IHD were already identified; and to estimate the workload in producing an IHD disease register. METHODS A population-based cross-sectional study was carried out in 46 out of 49 general practices in three PCGs in SW London, using computerized and paper medical records. Outcome measures were proportion of patients with IHD on existing disease registers, and workload and cost of producing complete registers. RESULTS Of 3803 patients with a pre-existing IHD Read code, 570 (15 per cent) were found to have no evidence of IHD, leaving 3233 patients with confirmed or probable IHD. A search of 7726 patients prescribed one of five cardiovascular drugs but not already coded as having IHD identified a further 1447 confirmed or probable cases. On average, coders spent 4.9 hours per 1000 list size verifying IHD cases or finding uncoded cases. Each additional IHD case required about 0.68 hours (41 minutes) of coder's time to identify and one case of IHD was identified or confirmed for about every five sets of notes examined. The cost of each additional case identified was about pounds sterling 10.20. At practice level, there was a wide variation in the proportion of IHD patients already on the register or wrongly coded as having IHD. CONCLUSIONS A centralized search programme can identify patients with IHD efficiently and at relatively low cost. As the identification of cases is an essential first step in implementing effective secondary prevention, other primary care trusts may also find this method useful in improving the management of patients with IHD.