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

Introduction

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.

Results

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).

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).

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.

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)).

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).

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.

Discussion

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.