Intended for healthcare professionals

General Practice

Randomised controlled trial of health promotion in general practice for patients at high cardiovascular risk

BMJ 1994; 309 doi: (Published 15 October 1994) Cite this as: BMJ 1994;309:993
  1. M E Cupples,
  2. A McKnight
  1. Department of General Practice, Queen's University of Belfast, Dunluce Health Centre, Belfast BT9 7HR
  1. Correspondence to: Dr Cupples
  • Accepted 24 August 1994


Objective: To assess the value of health education for patients with angina in reducing risk factors for cardiovascular disease and lessening the effect of angina on everyday activities.

Design: Randomised controlled trial of personal health education given every four months.

Setting: 18 general practices in the greater Belfast area.

Subjects: 688 patients aged less than 75 years and known to have had angina for at least six months; 342 randomised to receive education and 346 to no education.

Main outcome measures: Restriction of everyday activities, dietary habit, smoking habit, frequency of physical exercise; blood pressure, body mass index, and serum total cholesterol concentration at entry to trial and after two years.

Results: 317 in the intervention group and 300 in the control group completed the trial. At the two year review more of the intervention group (140, 44%) reported taking daily physical exercise than the control group (70, 24%). The intervention group also reported eating a healthier diet than the control group and less restriction by angina in any everyday activity. No significant differences were found between the groups in smoking habit, systolic or diastolic blood pressure, cholesterol concentration, or body mass index.

Conclusion: Despite having no significant effect on objective cardiovascular risk factors, personal health education of patients with angina seems to increase exercise and improve dietary habits and is effective in lessening the restriction of everyday activities.


  • implications

  • Health education for patients with angina

  • In this study personal health education improved patients' dietary habits, increased their frequency of taking physical exercise, and decreased their restriction of activity by angina

  • Patients receiving personal health education reported increased use of prophylactic drugs

  • No improvements were seen in blood pressure, cholesterol concentration, or rate of stopping smoking

  • Health education can improve the quality of life of patients with angina and should be incorporated into routine care in general practice


Cardiovascular disease is the main cause of premature death in Northern Ireland.1

Although the risk of cardiovascular disease can be reduced by primary prevention,2,3 trials of interventions in general practice have reported poor success.*RF 4–7* Secondary prevention can also be effective.*RF 8–10* Changes in exercise, diet, and smoking have been shown to produce regression of stenotic lesions on coronary arteriography.11

A mass public health education programme is aiming to increase awareness of healthy lifestyles in Northern Ireland.12 We studied the value of secondary prevention for high risk patients in general practice by giving personal health education to patients with angina. Angina is associated with increased cardiovascular mortality and restriction of everyday activities.*RF 13–18*

Patients and methods

Patients were identified from 18 group general practices in the greater Belfast area. General practitioners were asked to identify patients aged under 75 years who had had angina for at least six months and did not have any other severe illness. Angina was defined as recurrent, transient, and reproducible discomfort in the chest, arms, jaw, or shoulders, the discomfort being reproduced by physical exertion or emotional excitement and relieved by rest or drugs.

We sent letters to 1431 patients, asking for their consent to be interviewed by a research worker interested in angina. The initial appointment took place in the patient's local health centre or surgery or in their own home. Trained health visitors asked questions about the effect of angina on everyday activities, the frequency of attacks of angina, drugs taken, smoking exercise, and diet. They also administered a questionnaire to determine intake of various foods based on that used in a Department of Health and Social Services survey in Northern Ireland.19

Exercise levels were rated as the number of episodes each week of physical activity sustained for at least 20 minutes. Patients' height (Microtois tape, Raven Equipment, Dunnow), weight (Sea scales, Seca, Birmingham), blood pressure (random zero sphygmomanometer, Hawksley and Son, Lancing), and breath carbon monoxide concentration (Smokerlyzer, Bedfont Technical Instruments, Sittingbourne) were measured. An electrocardiogram was recorded and a sample of blood taken for measurement of serum cholesterol and thiocyanate concentration and a urine sample for cotinine assay.

Each subject was then randomly allocated to one of two groups. The health visitor opened an opaque, sealed, and numbered envelope containing the allocation, which had been generated by a computer program using random permuted blocks.

For the control group the interview ended at this point. Patients in the intervention group were given practical relevant advice regarding cardiovascular risk factors. They were reviewed at four monthly intervals and given appropriate health education. After two years both groups were reviewed by a research worker who had not previously been involved with the subjects.

Statistical methods

Comparisons between the control and intervention groups were done by X2 analysis. Analysis of variance was used to compare measurements of blood pressure, cholesterol, and body mass index between the two groups. Paired t tests were used to compare the differences within the control and intervention groups. Logistic regression was used to compare the mortality in the two groups.

A previous study of patients with angina in general practice identified 22% with “severe” angina - that is, they had attacks once or more a day when walking on the level and during sexual activity, sport, housework, or shopping.18 We estimated that effective intervention should reduce this to about 10%. We calculated that we would require 325 patients in each group to show a significant difference at the 5% level of probability. To allow for numbers diminishing through default or death we tried to recruit 350 to each group.


A total of 688 patients entered the study. Of the 342 in the intervention group, 317 completed the study, 13 died, and 12 defaulted from review. Of the 346 in the control group, 300 completed, 29 died, and 17 defaulted. No significant differences were found between the two groups at baseline (table I).


Characteristics of intervention and control groups at entry to trial

View this table:

Table II shows that the reported frequency of physical exercise was similar in the two groups initially. At review, more of the intervention group (140, 44%) than the control group (70, 24%) took daily physical exercise (seven or more times per week). More of the intervention group reported increased activity and fewer reported decreased activity compared with the control group (table III).


Frequency of physical exercise in patients with angina at baseline and review after two years. Values are numbers (percentages)

View this table:

Changes in frequency of physical exercise in patients with angina between baseline and review at two years

View this table:

Initially the eating patterns of the two groups did not differ, but at review more of the intervention group reported improved dietary habits (table IV). More people in the control group reported an increase than a decrease in the frequency of eating poultry, green vegetables, and high fibre foods - that is, an improved dietary habit - but more people also reported eating increased amounts of fried food, biscuits, sweets, and saturated fat.


Changes in eating habits among patients with angina between baseline and review at two years. Values are numbers of patients

View this table:

Of the 688 patients who entered the study, 187 stated that they had never smoked and 501 said that they had ever smoked. Of these 501, 148 said that they still smoked cigarettes and 31 that they smoked cigars or a pipe. Measurements of breath carbon monoxide, serum thiocyanate, and urinary cotinine concentrations were fully completed for 460 of the reported non-smokers and showed that 28 (6%) of the reported non- smokers had values that were more than two standard deviations above the mean for non-smokers and similar to values for those who reported smoking cigarettes.

The reported rate of stopping smoking at review was 21.5% (14/65) in the intervention group and 17.7% (11/ 62) in the control group. This difference was not statistically significant (P=0.82). There was also no difference in the number of patients who had increased or decreased their smoking between the groups.

At baseline there was no significant difference between the two groups in reported restriction of activities (table V). At review 27 (8.5%) of the intervention group claimed no restriction compared with eight (2.7%) of the control group (P=0.003). The numbers with severe angina fell from 39 (5.7%) initially to 22 (3.6%) at review, but there was no significant difference between the groups. The mean number of episodes of angina per week in the intervention group decreased from 3.2 (95% confidence interval 2.7 to 3.7) at baseline to 2.6 (1.7 to 3.5) at review (P=0.04), but no significant change was seen in the control group (2.5 (2.1 to 2.9) at baseline and 2.14 (1.7 to 2.5) at review (P=0.13)).


Effect of angina on everyday activities. Values are numbers (percentages) of patients

View this table:

At baseline 252 (36.6%) patients reported taking prophylactic drugs for angina with no significant difference between the groups. At review significantly more of the intervention group (166 (53%)) than the control group (120 (40%)) did so (P=0.004).

The groups did not differ significantly either initially or at review with regard to systolic or diastolic blood pressure, serum cholesterol concentration, or body mass index (table VI). Diastolic blood pressure and body mass index fell in both groups over the study period (P<0.0001). The intervention group showed a significant reduction in serum cholesterol concentration (P=0.003) of 2.4% but the fall was not significantly different from that in the control group (P=0.06).


Mean values of cardiovascular risk factors in patients with angina at baseline and review

View this table:

There were 29 deaths the control group and 13 in the intervention group. The relative odds of death in the control group was 2.32 (95% confidence interval 1.18 to 4.53). Ten of the deaths in the intervention group and 28 in the control group were attributed to cardiovascular causes. The relative odds of death was 2.20 (1.06 to 4.57) after age, sex, history of myocardial infarction, blood pressure, cholesterol, body mass index, smoking status, family history, social class, diabetes, and recent worsening of angina were adjusted for.


We have shown benefits from secondary prevention of cardiovascular disease for a group of high risk patients. Although there was no difference in values of risk factors between the intervention and control groups at review, patients in the intervention group were significantly less restricted by their angina than the control group.

The slight reductions in systolic blood pressure and significant reductions in diastolic blood pressure seen in both groups may have been partly due to an accommodation effect from repeated measurements20,21 but also to the reduction in body mass index.22 The reductions in diastolic blood pressure and body mass index in both groups suggest that the present medical care of patients with angina in combination with a background public health education campaign in effective. Additional personal health education could not be justified on the basis of these observations.

Lifestyle habits

Though elderly patients with angina might be expected to become less active over two years, only 28% of the intervention group compared with 54% of the control groups did so, and significantly more of the intervention group increased their frequency of physical exercise. We did not validate patients' reporting of their activities and some reporting bias may have occurred. However, the difference between the groups is in keeping with the intervention group's reporting of less restriction of activity.

The patterns of restriction of activities differed and were not directly related to physical effort or exercise tolerance. Psychological factors are obviously also important.17 The percentage of patients in the intervention group who took drugs prophylactically increased significantly. This behaviour could have caused a reduction in symptoms.

There was no significant difference in reported rates of giving up smoking cigarettes between the two groups. Education aimed at smoking may be more effective in specific programmes than as part of a multifactorial programme.6

The intervention group reported eating a healthier diet than the control group at review. The education may have encouraged the intervention group to report a more healthy diet than they were actually eating. We tried to minimise such bias by using someone unknown to the patient to make the review observations. Corroboration of dietary habit was not feasible in this study, but questionnaires are often used to assess diet.23

Changing diet has been shown to reduce serum cholesterol concentration,24,25 but we found no such correlation. This may be because food frequency questionnaires reflect trends in dietary habits but cannot accurately quantify intake.


Logistic regression analysis did not explain the mortality difference between the two groups. Although there were no significant differences in objective risk factors, the improved wellbeing resulting from increased physical exercise and a healthy diet may have had an effect on coronary mortality. The observation could be a chance occurrence, but we should remember that psychological factors affect patients with angina.


  • This work was funded by the Medical Research Council. We wish to thank our statistician, Dr D Merrett; the health visitors who carried out the fieldwork; and our secretary, Miss C Agnew.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.