A transferable programme of nutritional counselling for rehabilitation ...

European Journal of Clinical Nutrition (2004) 58, 778–786

& 2004 Nature Publishing Group All rights reserved 0954-3007/04 $25.00 www.nature.com/ejcn

ORIGINAL COMMUNICATION A transferable programme of nutritional counselling for rehabilitation following myocardial infarction: a randomised controlled study WS Leslie1*, CR Hankey1, D Matthews2, JEP Currall3 and MEJ Lean1 1 Human Nutrition, University of Glasgow Division of Developmental Medicine, Glasgow Royal Infirmary, Glasgow, UK; 2Department of Medicine and Endocrinology, Monklands Hospital, Lanarkshire Acute Hospitals NHS Trust, Glasgow, UK; and 3Department of Computing Service, University of Glasgow, Glasgow, UK

Objective: To evaluate the response to simple innovative dietary counselling in post myocardial infarction patients. Design: Randomised controlled trial. Setting: Cardiac rehabilitation programmes of two acute hospitals in Lanarkshire, Scotland. Patients: A total of 69 men and 29 women aged 35–75 y who survived acute myocardial infarction and participated in the cardiac rehabilitation programmes of the study hospitals between 1st September 1997 and 1st August 1998. Results: At 12 weeks follow-up, diet composition had improved significantly in intervention subjects, but no such change was evident in the control group. The target of five portions of fruit and vegetables per day was achieved by 65% of intervention subjects but only 31% of control subjects (P¼0.004). Between-group differences in food intakes were no longer evident at 1 y. Diet composition did however remain in line with current dietary targets in intervention subjects. Conclusion: Compared to conventional cardiac rehabilitation, this intervention, focused on targets known to improve mortality, improved diet in post myocardial infarction patients. However, a more sustained programme is required to maintain improvements. Delivery of the intervention was expensive and further research is required to determine the feasibility of a group approach using the same package. Sponsorship: The study was supported by Chief Scientist Office of the Scottish Executive Department of Health.

European Journal of Clinical Nutrition (2004) 58, 778–786. doi:10.1038/sj.ejcn.1601876 Keywords: myocardial infarction; nutrition; cardiac rehabilitation

Introduction Previous research has identified four specific dietary measures known to be effective in the secondary prevention of coronary heart disease. These are: increased oily fish (Burr

*Correspondence: WS Leslie, Human Nutrition, University of Glasgow, Division of Developmental Medicine, Glasgow Royal Infirmary, 10 Alexandra Parade, Glasgow G31 2ER, UK. E-mail: [email protected] Guarantor: WS Leslie. Contributors: WL recruited and randomised patients, collected, analysed and interpreted the data, and prepared the report. CH provided nutritional expertise, carried out the nutritional analysis, interpreted the data and assisted with the preparation of the report. JC provided statistical expertise. DM facilitated access to patients. ML generated the hypothesis and devised the study design. All the authors contributed to the final version of the paper. Conflict of interest: None. Received 10 April 2003; revised 3 July 2003; accepted 31 July 2003

et al, 1989), increased fruit and vegetable consumption (Singh et al, 1992), modification of dietary fat intake (Watts et al, 1992; De Lorgeril et al, 1994) and weight loss (Singh et al, 1996). There is clinical trial evidence for each of these measures individually being effective in secondary prevention. This research is of variable quality, but for all these measures there is supportive observational evidence (Kromhout et al, 1985; Williamson, 1997; Law & Morris, 1998; Hooper et al, 2001) and also experimental evidence to provide mechanistic and biological plausibility for modifying atheroma and thrombosis risk (Schmidt et al, 1990; Folsom et al, 1993; Zino et al, 1997; Miller et al, 1998; 1989). No previous study has tested all these modalities collectively as a therapeutic package within cardiac rehabilitation. The present study aimed to establish the responses to innovative dietary counselling focused on achieving dietary change of known value, in post myocardial infarction (MI) patients.

Dietary advice, post myocardial infarction WS Leslie et al

779

Methods Study population Recruitment was carried out from those patients attending the established cardiac rehabilitation programmes of two acute hospitals in Lanarkshire, Scotland, a region with a particularly high incidence of coronary heart disease (Scottish Office, 1997). Patients were recruited consecutively within 3 months of MI, over a 12-month period to this randomised, controlled trial. Lanarkshire Health Board Ethics of Research Committee awarded ethical approval for the study. A one-page outline of the study and invitation for volunteers was included in the cardiac rehabilitation folder received by all patients while in hospital. At 4–8 weeks after discharge, at a prerehabilitation programme clinic, patients were approached in person by the study nurse and invited to join the study. Exclusion criteria were dementia, alcoholism, malignant disease and any other disease requiring specific dietary advice. The hypothesis to be tested was that as patients who survive an MI are especially susceptible to messages about their health, would intensive dietary advice provide additional benefit over standard cardiac rehabilitation in terms of dietary change? The primary outcome measures were changes in the consumption of oily fish, fruit and vegetables, and saturated fat and changes in body weight. Secondary outcome measures were changes in plasma lipid concentrations and blood pressure.

Assignment Eligible subjects were individually randomised to either intervention or control group by means of a computergenerated random number list. The study statistician (JC), who had minimal initial contact with the study investigators and no contact with study participants, generated the randomisation list. Participants were recruited and then assigned to a treatment group by the study nurse in accordance with the random number list. Given the nature of the intervention, it was not possible to blind either participants or the study nurse to the treatment allocation.

Intervention Patients assigned to the control group received standard cardiac rehabilitation, which included a single group session (approx. 30–60 min) with a hospital dietitian. The session provided an overview of a ‘cardioprotective’ diet, was accompanied by a handout to reinforce the advice and also provided practical tips on implementing the dietary advice. In addition to the nutritional advice received during the cardiac rehabilitation programme, patients randomised to the intervention group were invited to attend four 1-h sessions of dietary counselling over an 8-week period. To try and encourage and facilitate dietary change, patients were asked to bring a ‘supporter’ (spouse, friend, other family member). Dietary advice focused on quantitative targets: a

minimum of five portions of fruit and vegetables daily, 2–3 portions weekly of oily fish, and a reduction in saturated fat intake to below 10% dietary energy. Patients with a body mass index (BMI) greater than 25 kg/m2 were given specific advice for weight management using the current clinical guidelines (Scottish Intercollegiate Guidelines Network (SIGN), 1996). Over the 8-week programme, each subject compiled a folder containing information, recipes, ideas and tools to encourage, reinforce and consolidate the achievement of advised dietary changes. A one-to-one approach was considered the most appropriate for the vulnerable post infarction patient and allowed a more, individualised approach. For the results of this study to be sustainable and transferable, for national application, existing personnel must be utilised rather than importing additional specialists. With this in mind, a nurse who had received appropriate nutritional training undertook the nutritional counselling for all intervention subjects.

Measurements and follow-up Baseline assessment, made prior to attendance at the cardiac rehabilitation programme, included measurements of height, weight, waist circumference, skin-fold thickness, fasting plasma lipids and blood pressure. Seven-day weighed food intake inventories were also completed. All measurements were repeated 3 and 12 months after nutritional counselling in intervention subjects and supporters, and 3 and 12 months after cardiac rehabilitation in control subjects. Dietary data were assessed using the dietary analysis programme COMPEAT (Lifestyle London) to provide information on the nutritional components of the food consumed. The data were also examined to ascertain the achievement or nonachievement of Scottish Dietary Targets (Scottish Office, 1996).

Statistical analysis Published data on the dietary changes of interest did not provide the changes and standard deviations necessary for a formal power analysis. On the basis of our own recent data, significant and clinically important changes have been achieved with study numbers of 40–50 (Anderson and Lean, 1996; Cox et al, 1997; Hankey et al, 2002a). A total of 40 subjects equally allocated to intervention and control groups would have at least 80% power of detecting differences equal in size to the population standard deviation for each variable at Po0.05. In all, 100 subjects were sought to be certain of sufficient numbers for analysis. The design of the study was a repeated measures design with three time points (baseline, 12 weeks and 1 y), with two groups of subjects (intervention and control). The most appropriate analysis of results is by a repeated measures European Journal of Clinical Nutrition


780 analysis of variance (ANOVA), partitioning the variation into components relating to time, treatment and the interaction between time and treatment. From a clinical viewpoint, it is the last that is the most interesting. For the sake of clarity, in the text and tables we present results based on Student’s ttests to show changes from baseline on a particular measure (paired sample t-test), and differences in the changes between intervention and control groups (independent samples t-test). Although the P-values presented in the body of the text are not corrected using a Bonferroni correction, they may be corroborated against figures in the appendix of significance derived by repeated measures ANOVA. w2 test was used to determine the significance of the differences between groups in the number of subjects achieving Scottish Dietary Targets.

Results During the period 1st September 1997 and 1st August 1998, 263 patients survived MI and were discharged from hospital. Of these survivors, 94 were ineligible to participate in the rehabilitation programme leaving 169 available for recruitment. A total of 32 patients refused to participate and 39 although referred for rehabilitation were not yet attending the programme. Thus we recruited 98 patients: 50 were randomly assigned to the intervention group and 48 to the control group. Of the patients, 13 (six controls and seven intervention subjects) withdrew before their baseline appointment. It was not possible to follow up nine patients at 12 weeks post intervention/cardiac rehabilitation and a further six patients were unavailable for follow-up at 1 y (Figure 1). Following randomisation, the physical characteristics of intervention and control groups were similar (Table 1). The majority of patients in both the intervention and control group were males (74%). Drug therapy in both groups was also similar as was the incidence of type II diabetes (two intervention, two control).

Dietary measurements Mean reported dietary intakes and their macronutrient composition were broadly similar in intervention and control subjects at baseline (Table 2). Differences between the intervention and control group were apparent in the changes in both reported food intakes and macronutrient composition at 12 weeks (Table 2). A total of 65% of intervention subjects achieved the target of five portions of fruit and vegetables per day compared to only 31% of control subjects (P¼0.003). Macronutrient content improved significantly in the intervention group in line with current dietary targets. No such change was observed in the control group. Oily fish consumption increased in all subjects, with no significant difference according to treatment group. In the intervention group, no differences were seen between those with and those without a supporter in terms of dietary intake and composition. European Journal of Clinical Nutrition

Differences in food intakes between groups observed at 12 weeks were no longer evident at 1 y (Table 2). However, diet composition in intervention subjects remained in line with current dietary targets, while it deteriorated in control subjects. Significant differences were observed at 1 y between the intervention and control group respectively in the number of subjects achieving the Scottish Dietary Targets for total fat (85 vs 58%, P¼0.01), saturated fat (66 vs 36%, P¼0.008) and carbohydrate (56 vs 22%, P¼0.004) (Figure 2).

Physical and anthropometric measurements Little change was seen in the physical and anthropometric measurements of the intervention group over the study period; however, increases in BMI and body weight were observed in the control group at 1 y (Table 3). Analysis of the subset of 53 subjects, who were overweight at baseline, is presented elsewhere (Hankey et al, 2002b).

Biochemical measurements At 12 weeks, triglyceride concentrations were reduced in intervention subjects (P¼0.04), with a significant difference between groups evident in their changes from baseline (Table 3). Follow-up at 1 y showed significant between-group differences in their changes, from baseline, in total and LDL cholesterol concentrations (Table 3). These between-group differences are corroborated by repeated measures ANOVA (Appendix B).

Post hoc analysis—‘supporters’ A total of 28 intervention subjects were accompanied by a ‘supporter’. Significant improvements were seen in food consumption and macronutrient composition of reported food intakes. In all, 57% of supporters achieved the target of five portions of fruit and vegetables per day at 12 weeks compared to 29% at baseline (P¼0.03). Diet composition remained in line with current dietary targets at 1 y.

Discussion Much effort on the part of health professionals is spent trying to affect dietary change as part of primary and secondary prevention strategies. The present study makes a clinically important comparison of the effectiveness of dietary advice given in routine care with that of a more intensive intervention using a randomised-controlled design and provides evidence to guide practice. Cardiac rehabilitation has a recognised value in post MI care and provides a natural opportunity for secondary prevention. Both the Scottish Diet Report (Scottish Office, 1993) and Scottish Diet Action Plan (Scottish Office, 1996) recommend incorporating dietary change into cardiac rehabilitation. Recent UK guidelines and audit standards for cardiac rehabilitation do not specify any nutrition or


781 263 survived Myocardial Infarction

94 patients ineligible for entry

169 eligible for entry

32 refused 39 not yet attending cardiac rehabilitation programme 98 patients randomised 48 assigned to control group

50 assigned to intervention group

2 died 7 withdrew

1 died 12 withdrew 39 reviewed at 12 weeks

37 reviewed at 12 weeks

1 died 2 withdrew

3 withdrew

34 reviewed at 1 year

Figure 1

36 reviewed at 1 year

Trial profile.

Table 1 Comparison of physical, anthropometric and biochemical measurements between intervention and control subjects at baseline Control subjects (n¼43) Measurement Age (y) BMI (kg/m2) % body fat Weight (kg) Waist (cm) Systolic BP (mmHg) Diastolic BP (mmHg) Total cholesterol (mmol/l) HDL (mmol/l) LDL (mmol/l) Triglyceride (mmol/l)

57.0 28.0 35.8 79.9 97.7 115 69 5.1 1.3 2.8 1.7

(9.7) (4.7) (8.5) (15.7) (13.3) (20) (15) (1.1) (0.3) (0.8) (0.8)

Intervention subjects (n¼42)

58.0 27.8 33.7 77.3 94.3 118 73 5.1 1.3 2.8 1.8

(9.1) (4.0) (5.3) (11.4) (10.2) (18) (9) (1.2) (0.4) (0.8) (1.2)

Significance of difference between groups P

0.62 0.81 0.20 0.37 0.19 0.51 0.15 0.98 0.77 0.84 0.61

95% CI 3.06 2.13 5.32 8.61 8.54 5.60 1.50 0.54 0.19 0.45 0.36

to to to to to to to to to to to

5.08 1.67 1.18 3.23 1.71 11.15 9.42 0.52 0.14 0.37 0.61

Data are expressed as mean (s.d.).

European Journal of Clinical Nutrition


3.05 1.21 7.64 0.15 0.16

Upper

95% CI

All subjects at baseline Intervention subjects at 1 year Control subjects at 1 year 0.27 0.23 0.09 6.53 2.31 2.83 1.28 7.04 1.88 0.007

0.10 0.18 0.04 0.06 0.08

% of subjects achieving target

80

0.50 0.56 1.66 7.43 2.65

Lower Upper Lower

95% CI

12 weeks

95% CI

P

95% CI

1y

90

70 60 50 40 30 20

0.006 0.44 0.002 0.001 0.04 (3.6) (1.5) (7.2) (6.3) (2.2)

0

Fruit & Vegetables

þ 0.6 þ 0.4 þ 0.9 0.9 0.4

> 5/day

Total Fat 50% energy

þ 0.8 þ 0.4 þ 3.7 4.2 1.4

(2.8) (1.4) (4.9)* (5.1)* (2.6)*

Figure 2 Percentage of subjects achieving Scottish Dietary Targets at baseline and one year.

dietetic elements or targets (Thompson et al, 1996). Adherence to current guidelines was also found to be poor, audit of the elements of cardiac rehabilitation limited and no formal audit has been carried out on any dietary components (Lewin et al, 1998). The primary aim of this study was to determine the response, in terms of dietary practices, of post MI patients to a programme of dietary counselling. The programme was based on existing evidence for benefit from specific dietary changes (fat, fruit and vegetables, oily fish and body weight), all of which feature in the Scottish Dietary Targets (Burr et al, 1989; Singh et al, 1992; Watts et al, 1992; De Lorgeril et al, 1994; Singh et al, 1996).

*Po0.05. a Average portion size¼80 g. b Weekly frequency of eating oily fish. c CHO¼carbohydrate.

(2.1)* 0.7 (3.1) 4.7 (2.8) (2.4) 0.05 (1.4) 1.2 (1.0) (6.6) 2.8 (8.4)* 49.1 (6.8) (6.8) þ 2.2 (7.5) 30.5 (6.60) (3.0) þ 0.6 (2.8) 9.1 (2.9) Portions of fruit and vegetablesa 4.6 (2.8) 0.7 Oily fishb 1.1 (1.2) þ 0.1 % energy CHOc 47.0 (7.1) 0.5 % energy total fat 32.3 (6.2) þ 0.4 % energy saturated fat 10.1 (3.2) 0.1

1 year (n¼34) 12 weeks (n¼37) Baseline (n¼42) 1year (n¼36) 12 weeks (n¼39) Baseline (n¼43)

Change from baseline Change from baseline


Mean consumption

Intervention subjects

P

10

Control subjects

Table 2 Changes from baseline in dietary intake measures

Significance of change in intervention group vs change in control group

782

Scottish dietary targets In our study population, the frequency of consuming key foods in the Scottish Dietary Targets (Scottish Office, 1996) was achieved by most intervention and control subjects at baseline, prior to commencement of cardiac rehabilitation. This may reflect an early adoption of appropriate health messages following acute MI, since the reported consumption of foods was very different from that in the general population (Ministry of Agriculture, Fisheries and Food (MAFF), 1999). The sources of these messages will have been numerous and are likely to have been reinforced, to a receptive audience, during and after hospitalisation for MI.


1.4 to 0.2 3.6 to 0.3 3.9 to 0.5 4.1 to 0.8 8.7 to 7.3 3.4 to 4.9 1.3 to 0.01 0.01 to 0.1 1.0 to 0.02 1.2 to 0.01

95% CI

to 0.1 to0.6 to 0.3 to 0.4 to 4.6 to 2.1 to 0.1 to 0.1 to 0.2 to 0.1

P

0.13 0.30 0.12 0.13 0.34 0.22 0.17 0.93 0.37 0.008*

0.7 2.0 2.2 2.8 12.9 8.9 0.7 0.1 0.5 0.8

95% CI

Oily fish A slight increase was seen in mean oily fish consumption to just under the target of two occasions per week recommended for secondary prevention (Burr et al, 1989). Mean weekly consumption was 180 g at 52 weeks. These figures are encouraging in relation to the Scottish Diet Target consumption target of 88 g per person per week (Scottish Office, 1996) and the reported average UK weekly consumption of 15 g oily fish (MAFF, 1999). As in the Dart study (Burr et al, 1989), palatability was an important barrier to achieving the oily fish targets. A total of 30% of intervention subjects reported a dislike of oily fish and therefore did not eat it.

(1.6) (3.2) (4.7) (3.1) (15.7)* (8.6)* (1.2) (0.2) (0.9) (1.0)

P

Data are expressed as mean (s.d.). *Po0.05.

BMI (kg/m2) % body fat Weight (kg) Waist (cm) Systolic BP (mmHg) Diastolic BP (mmHg) Total cholesterol (mmol/l) HDL (mmol/l) LDL (mmol/l) Triglyceride (mmol/l)

28.0 35.8 79.9 97.5 115. 68 5.0 1.3 2.8 1.6

(4.8) þ 0.2 (1.0) þ 0.7 (1.5)* (8.5) 0.2 (2.4) þ 0.7 (3.1) (16.3) þ 0.5 (2.8) þ 2.0 (4.5)* (13.8) 0.0 (3.3) þ 1.5 (4.6) (21.0) þ 3.0 (17.0) þ 7.0 (18.0)* (15.0) þ 2.3 (15.0) þ 4.3 (8.0)* (1.1) þ 0.01 (0.8) þ 0.2 (1.0) (0.3) þ 0.01 (0.1)* 0.01 (0.20) (0.8) þ 0.01 (0.7) þ 0.2 (0.8) (0.8) þ 0.1 (0.6) þ 0.3 (1.3)

27.2 33.7 76.8 93.8 119 73 5.0 1.3 2.8 1.8

(3.6) 0.1 (0.9) (5.3) 0.9 (3.2) (11.7) 0.4 (2.5) (10.7) 1.2 (3.7) (19.0) 1.1 (20) (10.0) 1.0 (7.0) (1.1) 0.1 (0.9) (0.4) 0.08 (0.2) (0.9 0.1 (0.8) (1.2) 0.3 (0.9)*

þ 0.1 0.9 þ 0.2 þ 0.8 þ 6.5 þ 5.1 0.3 0.0 0.2 0.3

1 y (n¼34) 12 weeks (n¼37) Baseline (n¼42) 1year (n¼36) 12 weeks (n¼39) Baseline (n¼43) Measurement

Change from baseline

Fruit and vegetables Fruit accounted for the majority (68%) of fruit and vegetable portions consumed by both intervention and control subjects at baseline, and this pattern did not change. Intervention subjects reported eating more than one vegetable with meals; however, this did not conform to the standard portion size of 80 g (Williams, 1995). This finding was frustrating, given that the target portion size of 80 g was carefully demonstrated to all participants using photographs and household measures. Increasing vegetable portion size is a message which requires further emphasis since vegetables or salad, if eaten, are customarily only consumed with the main meal of the day.

0.13 0.09 0.12 0.18 0.86 0.73 0.03* 0.69 0.03* 0.05

1y 12 weeks Change from baseline

Significance of change in intervention group vs change in control group Intervention subjects Control subjects

Table 3 Comparison of physical, anthropometric and biochemical measurements between intervention and control subjects over time

783 Additionally, although all subjects had been alerted to the existence of the programme, those who volunteered were likely to have been already interested in healthy eating, and already ‘health aware’. For these reasons, the impact of this intervention is likely to have been minimised.

Dietary composition: macronutrients Recent UK survey data indicate that the nutritional value of household food fails to reach macronutrient targets. National averages for Scotland are reported as: total fat 38% of total energy, saturated fat 15% of total energy and carbohydrate 46% of total energy intake (MAFF, 1999). The package of nutritional counselling, including verbal and written advice along with practical visual tools, successfully facilitated the achievement of the Scottish Dietary Targets of Z50% energy from carbohydrate and o35% energy from total fat in the majority of all intervention subjects, and these were maintained at 1 y.

Effect of dietary change on biochemical measurements The health benefits of dietary change (reduction of dietary fat and salt) and moderate weight loss in the overweight on blood pressure and plasma lipids have been reviewed elsewhere (Scottish Office, 1993). Some significant changes were seen in the biochemical measurements in the present European Journal of Clinical Nutrition


784 study; however, the assessment of other possible outcome measures (eg plasma vitamin C and E concentrations) was outside the scope of the project. Of the subjects, 60% were established on cholesterol lowering medication prior to their participation in this study and this continued throughout the study. Despite this, at 1 y follow-up, there were significant differences between groups in the changes from baseline in total and LDL cholesterol concentrations. This suggests that nutritional counselling, while not lowering total or LDL cholesterol concentrations significantly in the intervention group, conferred an additional benefit by preventing the rise observed in the control subjects. No other relevant changes, such as in drug treatment, occurred. The impact of LDL cholesterol lowering of 35% by statins is about 3–4% in terms of absolute mortality risk reduction (4S Study Group, 1994; LIPID Study Group, 1998). The effect on lipids is much greater than the differences between populations, but the effect on mortality is much less than differences between populations, so factors other than cholesterol are obviously important. Dietary change cannot match the cholesterol lowering effect of a statin but affects many other modalities (macronutrients, vitamins, minerals, fibre, bioactive compounds) whose combined effect may be greater than cholesterol lowering alone. Studies comparable in power to the statin trials would be needed to demonstrate or exclude such effects from dietary intervention.

Physical and anthropometric measurements While weight management as part of a comprehensive nutritional counselling programme failed to induce weight loss, it was successful in preventing the weight gain, which was demonstrated in the control group at 2 kg over 12 months. Weight management studies have demonstrated that the maximum period over which weight will continue to be lost is 12 weeks (SIGN, 1996). Anecdotally, intervention subjects often reported significant weight loss during and shortly after hospitalisation. Some subjects will also have tried to lose weight, and it is reasonable to speculate that in the 6- to 8-week period between discharge and participating in this study, most of any weight reduction had already been achieved. Weight maintenance in the intervention group should therefore be viewed as a positive outcome.

Supporters Although the study was not designed or powered to test the response of ‘supporters’ to the dietary intervention, significant improvements were seen in their reported dietary intake at 12 and 52 weeks, mirroring those of the intervention group. While the presence of a supporter made no difference to subjects in the intervention group in terms of improvements in both dietary intake and composition, these data suggest the possibility of ‘added value’ to a rehabilitation programme since families may represent an accessible and receptive secondary target group, sharing the same socioEuropean Journal of Clinical Nutrition

demographic risks for coronary heart disease as their partners.

Overall conclusion The impact of this intervention was significant at 12 weeks in terms of between-group differences in dietary intake measures but small in terms of biochemistry. While demonstrating that improvements in diet can be achieved, the deterioration in dietary habits between 12 weeks and 1 y observed in the intervention group suggests the need for a more sustained programme to maintain improvements. Available evidence suggests that delivery of advice must be carefully focused to produce behavioural change (Neil et al, 1995). The present study did not ration advice to individuals assessed as ‘ready to change’. In previous studies that have reported the beneficial effects of dietary modification, even higher intensity interventions have been used, with patients being seen regularly beyond 12 weeks, and in some cases as frequently as once per week (Burr et al, 1989; Singh et al, 1992; De Lorgeril et al, 1994). The one-to-one approach used in the present study would however present an additional burden on the resources currently available to cardiac rehabilitation programmes, with relatively modest gains. Health economics analysis of the present study has shown that the net cost of the intervention programme was d73.35 per patient (Andrew Walker, personal communication). This is high in comparison to an average cost of d10.29 per patient seen for one group session within a cardiac rehabilitation programme. It is however low in relation to the cost of the recommended pharmacological intervention following MI (aspirin, betablocker, ace inhibitor and statin) (SIGN, 2000), which is approximately d530 per patient per annum (Audrey Thompson, personal communication). The potential health gains from dietary interventions in cardiac rehabilitation and in primary prevention over a lifetime are less well recognised but still potentially substantial (Puska et al, 1998). Further investigations are therefore required to determine whether using the same package with a group treatment approach, incorporating individual target setting, regular follow-up and reinforcement would be feasible and yield sustained improvements in dietary habits.

Acknowledgements We thank Hairmyres and Stonehouse NHS Trust for providing access to patients, Sister Margaret Davidson, Cardiac rehabilitation nurse and Christine Proudfoot, Senior Physiotherapist, Hairmyres Hospital, for continued help and cooperation, Dr Andrew Walker, Robertson Centre for BioStatistics, University of Glasgow, for health economics analysis, and Audrey Thompson, Medicines Management Advisor, Greater Glasgow Primary Care Trust, for information on drug therapy costs.


785 References Anderson AS & Lean MEJ (1996): Changing dietary intake: a plate model approach. Proc Nutr. Soc. 55, 76A (abstract). Burr ML, Gilbert JF, Holliday RM, Elwood PC, Fehily AM, Rogers S, Sweetnam PM & Deadman NM (1989): Effects of changes in fat, fish, and fibre intakes on death and myocardial infarction: diet and reinfarction trial. Lancet 2 (8666), 757–761. Cox DN, Anderson AS, Reynolds J, McKellar S & Mela DJ (1997): Lean ME. Measuring fruit and vegetable intake: is five-a-day enough? Eur. J. Clin. Nutr. 51, 177–180. De Lorgeril M, Renaud S, Mamelle N, Salen P, Martin JL, Monjaud I, Guidollet J, Touboul P & Delaye J (1994): Mediterranean alphalinolenic acid-rich diet in secondary prevention of coronary heart disease. Lancet 343, 1454–1459. Folsom AR, Qamhieh HT, Wing RR, Jeffery RW, Stinson VL, Kuller LH & Wu K (1993): Impact of weight loss on plasminogen activator inhibitor (PAI-1), factor VII, and other haemostatic factors in moderately overweight adults. Arterioscler. & Thromb. 3, 162–169. Hankey CR, Lean MEJ, Lowe GDO, Rumley A & Woodward M (2002a): The effect of moderate weight loss on angina symptoms and indices of coagulation and fibrinolysis in overweight patients with angina pectoris. Eur. J. Clin. Nutr. 56, 1039–1045. Hankey CR, Leslie WS, Currall JEP, Matthews D & Lean MEJ (2002b): Weight change after myocardial infarction: statistical perspectives for future study. J. Hum. Nutr. Dietet. 15, 439–444. Hooper L, Summerbell CD, Higgins JPT, Thompson RL, Capps NE, Davey Smith G, Riemersma RA & Ebrahim S (2001): Dietary fat intake and prevention of cardiovascular disease: systematic review. BMJ 322, 757–763. Kromhout D, Bosschieter EB & Coulander CL (1985): The inverse relationship between fish consumption and 20 year mortality from coronary heart disease. N. Engl. J. Med. 312, 1203–1209. Law MR & Morris JK (1998): By how much does fruit and vegetable consumption reduce the risk of ischaemic heart disease? Eur. J. Clin. Nutr. 53, 549–556. Lewin RJP, Ingleton R, Newens AJ & Thompson DR (1998): Adherence to cardiac rehabilitation guidelines: a survey of rehabilitation programmes in the United Kingdom. BMJ 316, 1354–1355. Miller ER, Appel LJ & Risby TH (1998): Effect of dietary patterns on measures of lipid peroxidation. Circulation 98, 2390–2395. Miller GJ, Cruickshank JK, Ellis LJ, Thompson RL, Wilkes HC, Stirling Y, Mitropoulos KA, Allison JV, Fox TE & Walker AO (1989): Fat consumption and factor VII coagulant activity in middle-aged men. Atherosclerosis 78, 19–24. Ministry of Agriculture, Fisheries and Food (1999): National Food Survey 1999. London: The Stationery Office. Neil HAW, Roe L, Godlee RJP, Moore JW, Clark GMG, Brown J, Thorogood M, Stratton IM, Lancaster T, Mant D & Fowler GH (1995): Randomised trial of lipid lowering dietary advice in general practice: the effects on serum lipids, lipoproteins and antioxidants. BMJ 310, 569–573. Puska P, Vartiainen E, Tuomilehto J, Salomaa V & Nissinen A (1998): Changes in premature deaths in Finland: successful long-term prevention of cardiovascular diseases. Bull. World Health Org. 76, 419–425. Scandinavian Simvastatin Survival Study Group (1994): Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet 344, 1383–1389. Schmidt EB, Varming K, Ernst E, Madsen P & Dyerberg J (1990): Dose–response studies on the effect of n-3 polyunsaturated fatty acids on lipids and haemostasis. Thromb. Haemostasis 63, 1–5. Scottish Intercollegiate Guidelines Network (1996): Obesity in Scotland: Integrating Prevention with Weight Management. Edinburgh: Scottish Intercollegiate Guidelines Network. Scottish Intercollegiate Guidelines Network (2000): Secondary Prevention of Coronary Heart Disease Following Myocardial Infarction. Edinburgh: Scottish Intercollegiate Guidelines Network.

Scottish Office (1996): Eating for Health: A Diet Action Plan for Scotland. Edinburgh: HMSO. Scottish Office (1993): The Scottish Diet. A report of a working party to the Chief Medical Officer in Scotland. Edinburgh: HMSO. Scottish Office Department of Health (1997): The Scottish Health Survey 1995. Edinburgh: HMSO. Singh RB, Rastogi SS, Verma R, Verma R, Laxmi B, Singh R, Ghossh S & Niaz MA (1992): Randomised controlled trial of a cardioprotective diet in patients with acute myocardial infarction: results of a one-year follow up. BMJ 304, 1015–1019. Singh RB, Rastogi V, Niaz MA & Beegom R (1996): Effect of diet and moderate exercise on central obesity and associated disturbances, myocardial infarction and mortality in patients with and without coronary artery disease. J. Am. Coll. Nutr. 15, 592–601. The Long-Term Intervention with Pravastatin in Ischaemic Heart Disease (LIPID) Study Group (1998): Prevention of cardiovascular events and death with pravastatin in patients with coronary heart disease and a broad range of initial cholesterol levels. N. Engl. J. Med. 339, 1349–1357. Thompson DR, Bowman GS, Kitson AL, de Bono DP & Hopkins A (1996): Cardiac rehabilitation in the United Kingdom: guidelines and audit standards. Heart 75, 89–93. Watts GF, Lewis B, Brunt JN, Lewis ES, Coltart DJ, Smith LDR, Mann JI & Swan A (1992): Effects on coronary artery disease of lipidlowering diet, or diet plus cholestyramine, in the St Thomas’ Atherosclerosis Regression Study (STARS). Lancet 339, 563–569. Williams C (1995): Healthy eating: clarifying advice about fruit and vegetables. BMJ 310, 1453–1455. Williamson DF (1997): Intentional weight loss: patterns in the general population and its association with morbidity and mortality. Int. J. Obes. Relat. Metab. Disord. 21(Suppl 1), S14–S19. Zino S, Skeaff M, Williams S & Mann J (1997): Randomised controlled trial of effect of fruit and vegetable consumption on plasma concentrations of lipids and antioxidants. BMJ 314, 1787–1791.

Appendix A Nutritional counselling following a heart attack You are now recovering from your heart attack and will attend the cardiac rehabilitation programme after you have been discharged from hospital. We are carrying out a study that will investigate if more detailed nutritional (food) advice to people and their families, given alongside the cardiac rehabilitation classes, could be practical and effective. The cardiac rehabilitation teams in both Hairmyres and Stonehouse hospitals have kindly agreed to participate in our study. However, it is your help that is vital to the success of the study. We would value your help in this Scottish Office funded project and will write personally to you after your discharge inviting you to take participate. We look forward to meeting you in the near future and wish you a speedy recovery. Wilma Leslie and Catherine Hankey Nutritional Counselling Team European Journal of Clinical Nutrition


786

Appendix B

Appendix C

Supplement to Table 2

Calculation of basal metabolic rate (BMR) A. Calculation of the basal metabolic rate (in MJ/day)

Significance of changes in biochemical measurements in intervention group vs control group from repeated measures ANOVA

Variable Total cholesterol (mmol/l) HDL cholesterol (mmol/l) LDL cholesterol (mmol/l) Triglyceride (mmol/l)

Time P

Treatment P

Treatment time P

0.73 0.84 0.77 0.77

0.68 0.73 0.89 0.39

0.03* 0.69 0.04* 0.05

Women

18–30 y¼0.0621 actual weight in kg þ 2.0357 31–60 y¼0.0342 actual weight in kg þ 3.5377 460 y¼0.0377 actual weight in kg þ 2.7545

Men

18–30 y¼0.063 actual weight in kg þ 2.8957 31–60 y¼0.0484 actual weight in kg þ 3.6534 460 y¼0.0491 actual weight in kg þ 2.4587

*Po0.05.

Supplement to Table 3 Significance of changes in dietary intake measures in intervention group vs control group from repeated measures ANOVA

Variable Portions of fruit and vegetables Weekly frequency of oily fish % energy carbohydrate % energy total fat % energy saturated fat *Po0.05.


Time P Treatment P Treatment time P 0.86 0.29 0.31 0.43 0.76

0.03* 0.20 0.002* 0.001* 0.007*

0.10 0.18 0.04* 0.06 0.08

To convert to kcal/day, multiply the result by 240 B. Correction for daily activities — total daily energy expenditure would equal BMR 1.3 for obese subjects engaged in mild to moderate daily activity. (For subjects engaged in strenuous activity, the total daily energy expenditure would equal BMR 1.5.) Prescribed diet — the prescribed diet will equal the estimated total daily energy expenditure minus 600 kcal/ day. The minimum number of calories prescribed will be 1200 kcal/day.