Hospital discharge rates for suspected acute coronary syndromes between 1990 and 2000: population based analysis

BMJ 2004; 328 doi: (Published 10 June 2004) Cite this as: BMJ 2004;328:1413
  1. N F Murphy, research fellow1,
  2. K MacIntyre, specialist registrar2,
  3. S Capewell, professor3,
  4. S Stewart, professor4,
  5. J Pell, consultant5,
  6. J Chalmers, consultant6,
  7. A Redpath, statistician6,
  8. S Frame, principal analyst6,
  9. J Boyd, record linkage project manager6,
  10. J J V McMurray, professor (j.mcmurray{at}
  1. 1 Department of Cardiology, Western Infirmary, Glasgow G12 8QQ
  2. 2 Department of Public Health, University of Glasgow, Glasgow G12 8RZ
  3. 3 Department of Public Health, University of Liverpool, Liverpool L69 3GB
  4. 4 Division of Health Sciences, University of South Australia, Adelaide 5000, Australia
  5. 5 Greater Glasgow Health Board, Glasgow G3 8YU
  6. 6 Information and Statistics Division, Edinburgh EH5 3SQ
  1. Correspondence to: J J V McMurray
  • Accepted 22 January 2004


Although hospital discharge rates for acute myocardial infarction are falling,14 no contemporary studies compare temporal trends in these rates for angina and other types of chest pain. We examined recent trends in population discharge rates for myocardial infarction, angina, and chest pain (“suspected acute coronary syndromes”) between 1990 and 2000.

Participants, methods, and results

We got data from the Scottish morbidity record for Scottish residents aged at least 18 years with a “first” emergency hospitalisation for myocardial infarction (codes ICD-9 (international classification of diseases, ninth revision) 410, ICD-10 I21 or I22), angina (ICD-9 411 or 413; ICD-10 I20 or I24.9) or “other chest pain” (ICD-9 786.5; ICD-10 R07), between 1990 and 2000.5 We analysed discharges coded only in the principal position. A “first” hospitalisation was one with no discharge diagnosis of coronary heart disease or chest pain in the previous 10 years.

We calculated rates using annual official age and sex specific population estimates for 1990-2000 and tested the significance of trends in discharge rates with linear regression.

We found 225 512 first hospitalisations for suspected acute coronary syndrome with a discharge diagnosis of myocardial infarction in 96 026 (43%), angina in 37 403 (17%), and other chest pain in 92 083 (41%). Patients with angina were on average 11 years older and those with myocardial infarction 13 years older than patients with chest pain (both P < 0.001).

The population discharge rate for myocardial infarction fell by a third, from 260 to 173 per 100 000 between 1990 and 2000 (regression coefficient (average change per year in population hospitalisation rate) −9.9; 95% confidence interval −11.5 to −8.2). Conversely, the rate for angina increased by 79%, from 59 to 105 per 100 000 (5.2; 4.3 to 6.1). The discharge rate for other chest pain increased by 110%, from 114 to 296 per 100 000 (coefficient 13.6; 11.6 to 15.5). The rate for myocardial infarction or angina decreased by 12%, from 319 to 278 per 100 000 (−4.7; −6.6 to −2.7), though that for any suspected acute coronary syndrome increased by a quarter, from 460 to 574 per 100 000 (8.9; 6.0 to 11.8).

The greatest change, across all diagnostic subgroups, was in elderly people (figure). The decline in myocardial infarction was much greater in men than in women in young and older age groups. The increase in angina was slightly greater in men than in women in both age groups. For chest pain, the increases were comparable in men and women in both age groups.


Population discharge rates per 100 000 per year for acute myocardial infarction, angina, and chest pain in Scotland, 1990-2000 by age group


The fall in hospitalisation rates for myocardial infarction is consistent with reports from other countries but contrasts with the rise in rates for angina and chest pain.14 The Minnesota heart survey, the only other study to show a rise in discharge rates for angina, reported a 56% increase in men (30% in women) during 1985-1995 (with a 20% decline for myocardial infarction in men and women).4 We found no similar reports of an increase in chest pain discharges.

Possible reasons for this apparent shift are increased survivorship from previous acute coronary events (unlikely given that we looked only at first admissions); a change in the clinical course of coronary heart disease (overall decline and shift from presentation with infarction to angina—for example, because of wider use of aspirin); better diagnostic exclusion of infarction; or “threshold” changes for both admission and diagnosis over the period of study.

The increase in hospitalisations for angina and chest pain has enormous implications for resources, finances, and services—for example, expansion of coronary intervention for increasing angina and development of observation units for chest pain to avoid admission for non-coronary chest pain.

What is already known on this topic?

Incidence of acute myocardial infarction has declined in recent years

What this study adds

Between 1990 and 2000, population hospitalisation rates for angina increased by 79%, for chest pain by 110% (acute myocardial infarction declined by 33%), and for any suspected acute coronary syndrome (myocardial infarction, angina, or chest pain) by 25%

These increases have important financial and service implications

This article was posted on on 21 May 2004:


  • Contributors NM, KM, SC, and JMcM designed the study. NM coordinated the study, analysed the data, and led the writing team. NM and KM contributed to the study design, data analysis, and report drafting. SS, JP, JC, AR, and JB helped analyse and interpret the data and draft the report. SF contributed to the study design and extracted the data. All investigators contributed to the final report. JM, KM, and NM are guarantors

  • Funding NM is funded by the British Heart Foundation, SS is supported by the National Health and Medical Research Council of Australia and the National Heart Foundation of Australia

  • Competing interests None declared

  • Ethical approval Not needed


View Abstract