General practice workload implications of the national service framework for coronary heart disease: cross sectional survey

Participants, methods, and results

We invited 65 practices randomly selected from the 51 primary care groups in the Trent region to participate; 24 practices volunteered and 18 were recruited.2 Ethical approval was obtained. We identified two target groups of high risk patients aged 35-74: patients with a Read code for ischaemic heart disease or stroke or at least one prescription for a nitrate (group 1),3 and patients with a computer recorded Read code for diabetes or hypertension (excluding those in group 1) who would be at high risk of developing coronary heart disease (group 2).

During March and April 2000, we extracted details of ischaemic heart disease, comorbidity (diabetes, hypertension, and stroke), drug treatment, and other coronary risk factors (age, sex, family history of cardiovascular disease, most recent smoking status, body mass index, blood pressure, glycated haemoglobin, and fasting lipid concentrations) from the practice computers. Descriptive data were analysed in SPSS (version 8) and Stata (version 5.0).

Of the 98 137 registered patients, 10 325 (10.5%) patients aged 35-74 years had at least one of ischaemic heart disease, stroke, diabetes, and hypertension. Fasting serum cholesterol concentrations had been recorded for 2267/4455 (50.9%) patients in group 1 and 2478/5870 (42.2%) in group 2. Of these, 1527/2267 (67.4%) patients in group 1 and 1809/2478 (73.0%) in group 2 had fasting serum cholesterol concentrations >5 mmol/l, with 1448 (63.9%) and 400 (16.1%), respectively, taking lipid lowering agents. Of the 1076 patients with a recorded history of myocardial infarction, 692 (64.3%) had received a β blocker. Aspirin was being taken by 3213 (72.1%) patients in group 1 and 1326 (22.6%) in group 2. Reducing the blood pressure target from 160/90 to 140/85 mm Hg (the current recommended target4) increased the number of high risk patients needing improved control from 2702/10325 (26.2%) to 6375 (61.7%). Of the 1839 patients with diabetes, 1518 (82.5%) had a glycated haemoglobin value recorded on computer, and of these, 729 (48%) had a value ≥7.5%, which indicated moderate to poor glucose control.

The number of patients who needed risk factors recorded on computer or further disease control measures was estimated on the basis of the proportion of patients in the whole study population (table). Practices varied 14-fold in recording risk factors on computer and fourfold in the need for further disease control measures. We estimate that in the average practice of 10 000 patients about 904 items will need recording and about 2221 disease control measures will be needed (see table).

Comment

The national service framework for coronary heart disease has profound implications for primary care. Changes in workload and funding vary substantially between general practices.

Our estimates are conservative for a number of reasons. Practices in the study had systems compatible with MIQUEST (a computer software programme for data extraction) and their data might be of better quality than those from an average practice. We looked at data recorded on computers; examination of manual records and other information systems might reveal higher numbers of cases and better standards of care. We did not include patients without heart disease, diabetes, or hypertension who may have an absolute cardiovascular risk of ≥3% a year based on other risk factors. Our data included patients aged 35-74, but elderly patients have a higher cardiovascular risk and are more likely to gain from risk factor modification.5

Substantial variations in the recording of risk factors and the need for further disease control measures between practices were not explained by differences in the age-sex structure of the practice population. Some variation may be due to differences in how computer Read codes are used.