Aims: To evaluate the natural history and management of patients diagnosed with chest pain of unspecified type or origin in primary care.

Design: Population-based case-control study.

Methods: The study included 13 740 patients with a first diagnosis of unspecified chest pain and 20 000 age- and sex-matched controls identified from the UK General Practice Research Database. Odds ratios (ORs) and 95% confidence intervals (CIs) were computed using unconditional logistic regression. Risk estimates were adjusted for age, sex and number of physician visits.

Results: The incidence of a new diagnosis of chest pain was 15.5 per 1000 person-years and increased with age, particularly in men. The risk of a chest pain diagnosis was greatest in patients with prior diagnoses of coronary heart disease (OR: 7.1; 95% CI: 6.1-8.2) and gastroesophageal reflux disease (OR: 2.0; 95% CI: 1.7-2.3). In the year after diagnosis, chest pain patients were more likely than controls to be newly diagnosed with coronary heart disease (OR: 14.9; 95% CI: 12.7-17.4) and heart failure (OR: 4.7; 95% CI: 3.6-6.1). A new diagnosis of chest pain was associated with an increased risk of death in the following year (RR: 2.3; 95% CI: 1.9-2.8).

Conclusions: Some causes of chest pain are underdiagnosed in primary care. This is of particular consequence for the minority of chest pain patients with cardiac disease.

Keywords: Chest pain; automated database; population-based sudy; primary care; incidence.