Outcome and use of health services four years after admission for acute myocardial infarction: case record follow up study

BMJ 1999; 319 doi: (Published 24 July 1999) Cite this as: BMJ 1999;319:230
  1. Martin Melville, research assistant (martin.melville{at},
  2. Nigel Brown, specialist registrar in cardiologya,
  3. David Gray, reader in medicinea,
  4. Tracey Young, research assistantb,
  5. John Hampton, professor of cardiologya
  1. a Department of Cardiovascular Medicine, University Hospital, Nottingham NG7 2UH
  2. b British Heart Foundation Cardiovascular Statistics Unit, Department of Mathematics, University of Nottingham, Nottingham NG7 2RH
  1. Correspondence to: M Melville
  • Accepted 10 February 1999

Acute myocardial infarction affects around 250 000 people each year in the United Kingdom. To our knowledge, medium term outcome and use of resources, other than revascularisation rates,1 have not been reported in a non-selected population.

Subjects, methods, and results

All patients resident in Nottingham Health District who had been admitted in 1992 for acute myocardial infarction to either of Nottingham's two hospitals were identified from the Nottingham heart attack register.2 We reviewed all hospital and general practitioner case notes for investigations, interventions, readmissions, clinic visits, and symptoms up to August 1996. Data on deaths were obtained from the Office for National Statistics.

Overall, 900 patients were admitted for myocardial infarction (mean age 66.6 years; 561 men). Data extraction was completed in 899 (99.9%). The table shows the outcomes in the 695 patients who were discharged alive.

In all, 537 patients received a clinic appointment on discharge (eight did not attend and seven others had died). The remaining 158 did not receive follow up—medical records did not indicate why.

Only 126 (24%) patients who were followed up had had a previous myocardial infarction compared with 66 (42%) of those who were not (χ2=20.47, P<0.001). The two groups did not differ in size of infarct (as measured by rise in creatine kinase concentrations (χ2=1.51, P=0.219), location of infarct (χ2=0.72, P=0.399), or Killip score at hospital presentation (χ2=2.27, P=0.132)). Patients without follow up were, however, less likely to have received thrombolysis (χ2=25.01, P<0.001) and to have been under the care of a cardiologist; 142 of the 519 (27%) patients managed by a physician and 16 of the 176 (9%) managed by a cardiologist were not followed up (χ2=24.97; P<0.001). These patients were no more likely to require readmission in the four years after infarction, but after adjustment for age, sex, and previous infarction 79 (50%) had died compared with 130 (24%) (z=3.44, P=0.001). There were no differences in the proportion of deaths from coronary heart disease in the two groups (52 of the 78 deaths (67%) in those not followed up v 88 of the 129 deaths (68%) in those followed up; χ2=0.54, P=0.817). By the end of the study 135 patients had never had an outpatient cardiology review and 62 had had no further hospital contact.

Of the 488 patients alive at August 1998, 282 were recorded as having or not having angina. Ninety eight had documented ongoing anginal symptoms, of whom 45 required two or more antianginal drugs; none of the 21 patients under the care solely of their general practitioner but 20 of the 24 patients under specialist review were being investigated.


Survivors of myocardial infarction comprise a mixed group with varying degrees of underlying coronary disease, cardiac impairment, and socioeconomic status, all of which influence health care needs.3 The prospects for a patient surviving an infarction are not particularly favourable, and patients require hospital based care over years.

Use of health services and outcome in unselected cohort of patients admitted for acute myocardial infarction during median of 4 years of follow up

View this table:

Guidelines recommend formal follow up after discharge,4 but we found that clinical review was not universal. Opportunities were missed to optimise secondary prophylaxis and expedite cardiac rehabilitation, not least among those who did not receive routine follow up whose mortality was inexplicably high.

Demand did not fall with time: many survivors continued to attend clinic years after their infarction, reflecting the long term nature of coronary disease. Two thirds were readmitted with symptoms suggestive of further infarction (most on more than one occasion) or heart failure. Half underwent some form of cardiac investigation. Our angiography rate of 840 per 1 million population (63% of whom have had myocardial infarction) is close to British Cardiac Society recommendations5 but low by standards in the United States.

Opportunities to reduce the impact of disease are being missed. The least that should be offered is to review all patients, optimise treatment to minimise symptoms and cardiac risk, and advise general practitioners when to refer for a specialist opinion.


We thank the general practitioners of Nottingham for their cooperation.

Contributors: MM was responsible for study design, collecting follow up data, data analysis, and writing the paper. NB had the original idea for the study, was responsible for study design and collecting follow up data, and contributed to writing the paper. DG had the original idea for the study, was responsible for collecting the initial data on the register and for data analysis, contributed to writing the paper, and is guarantor for the study. JH originated the heart attack register and contributed to writing the paper. TY analysed the data and contributed to writing the paper. Jean Barton and Caroline Gray collected initial data for the heart attack register.


  • Funding The Nottingham heart attack register has been supported by the Department of Health since its inception in 1973.

  • Competing interests None declared.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
View Abstract