Drugs 36 (Suppl , 3): 105-109 (1988)00 12-6667/88/0300-0I05/$2 .50/0© ADIS Press LimitedAll rights reserved.

Strategies for Dietary and Anti-Smoking AdvicePractical Experiences from the Oslo Study

Ingvar HjermannDepartment of Medicine, University of Oslo Medical School, Ulleval Hospital,Oslo, Norway

Summary One of the 2 controlled preventive trials within the Oslo Study was a non-drug trialon the effect ofdiet and smoking intervention on coronary heart disease in J232 middle­aged. normotensive. healthy men . All had elevated serum cholesteroland 4 out of5 smokedevery day.

The participants in the intervention group met every 6 months during the 5-year studyfor clinical examination and for dietary and smoking counselling. The effect on serumcholesterol was a 13% lowering (i.e. 10% net reduction compared with control group), andabout a 50% reduction in cigarette consumption. The effect on coronary heart disease(fatal and non-fatal myocardial infarction or sudden death) was a 47% lowering of thes-vear incidence in the intervention group compared with the control group (p = 0.02.2-sided).

The main intervention strategies were:• information about the risk factor concept to participant and spouse (in groups)• individual diet and anti-smoking counselling after finish ing an extensive clinical andelectrocardiographic examination . including exercise-ECG• basis and background for the counselling strategy: the total situation ofthe participant.Some ofthe most important items: personality, motivation, diet history, bodyweight, bloodlipids and blood sugar• anti-smoking advice given individually to all smokers in the intervention group• advice that smoking cessation was expected to be of special importance for those withelevated blood lipids,

One of the controlIed preventive trials withinthe Oslo Study was a non-drug trial on the effectof diet and anti-smoking advice on the first eventof coronary heart disease (CHD) in 1232 healthymen aged between' 40 and 50 years who werenormotensive (Hjermann et al. 1981; Leren et al.1975). 80% of the participants smoked cigarettesdaily, and serum cholesterol concentrations rangedfrom 6.9 to 9.0 mmol/L (290 to 380 rng/dl) , Base­line characteristics for the 2 groups are shown intable I.

1. Study Design

The men , selected from about 16,000 who wereinitialIy screened in this age group, were informedabout their statisticalIy increased risk of CHD ontheir return for their first re-examination. They werefurther informed about the design of the study, andthat there would be a 50% chance of being ran­domised to the intervention or control group.

The men in the intervention group were thenindividually advised about the risk factor concept,

Dietary and Ant i-Smoking Advice: the Oslo Study

Table I. Comparison of study groups on entry to trial

SexMean age (years)Age range (years)History/symptoms of CHDMean daily cigarette consumptionSmokers (%)

Sodyweight (kg)Height (em)Serum cholesterol enzymatic method range (mmol/L)Serum triglycerides (mmol/L) at first re-examination

(fasting)SSP (mm Hg)Sedentary workers (%)

Diet score

Intervention (n = 604)

Male45.240-49None13.079.177.3 ± 10.3177.4 ± 6.06.9-9.0

2.21 ± 0.9< 1505014.8 ± 6.1

Control (n = 628)

Male45.240-49None12.579.678.2 ± 9.8176.9 ± 6.36.9-9.0

2.25 ± 1.1< 1504814.1 ± 6.1

106

and that the purpose of the study was to investigatethe effect ofcholesterol reduction and smoking ces­sation on CHD.

The dietitian subsequently established a diet re­cord for each man by means of a standardisedquestionnaire. Dietary advice, given in a 30­minute talk , was based on this record, bodyweight,serum cholesterol and triglyceride concentrations,and the man's general background. For those sub­jects whose only risk factor was high cholesterolconcentrations, diet change consisted mainly of areduction in saturated fat intake and a slight in­crease in polyunsaturated fat intake. For bothoverweight and normal weight subjects with ele­vated fasting triglyceride concentrations, a reduc­tion in total energy intake (mainly sugar, alcoholand fat intake) was also recommended.

For sandwiches we recommended fibre-richbread with no margarine or a thin layer of low fator high polyunsaturated margarine; a fish or veg­etable filling was preferable, but low fat cheese ormeat was acceptable. We also recommended thatsubjects should use skimmed milk and not eat morethan I egg weekly. For main meals we advised sub­jects to eat fish, whale meat and low fat meat withpotatoes and vegetables . We advised that polyun­saturated oil should be used for cooking, bakingand sauces, and that fruit should be eaten for des­sert. Reduced intakes of sugar, sweet drinks and

alcoholic beverages were specifically recommendedwhen reduction in calorie intake was advisable (i.e.for overweight subjects and those with hypertri­glyceridaemia).

Anti-smoking advice was given individually toall smokers in the intervention group. They wereinformed that cessation of smoking might be ofspecial importance for those with high blood lipidlevels; also, the danger of increasing bodyweightafter smoking cessation was stressed . In additionto individual counselling, participants and spouseswere invited in groups to information meetingsabout low cholesterol diet and the importance ofstopping smoking.

2. Results and Conclusions

Subjects in the intervention group were re­examined every 6 months and control subjectsevery 12 months. Follow-up included a shortclinical examination, with special emphasis on car­diovascular symptoms, bodyweight, blood pres­sure, serum cholesterol and triglyceride concentra­tions, and a 12-1ead resting ECG. At each follow­up the men in the intervention group were askedabout their eating and smoking habits. Adherenceto diet and non-smoking were assessed, and thena 'cholesterol curve' was made for each man andshown to him.

Dietary and Anti-Smoking Adv ice: the Oslo Study 107

At the 5-year follow-up a random sample of halfthe men in the intervention group was given thesame diet interv iew as at the start of the trial. Thebest responders (those whose reduction in choles­terol concentrations was in the highest quintile ofthe distribution) were found to have consumedmore fish and vegetables, more skimmed ratherthan whole milk, less saturated fat margarine andmore polyunsaturated fats in cooking and in mar­garine. The effects of intervention on serum lipidconcentrations and smoking habits are shown in

figure I. In general, intervention produced desir­able decreases in both cholesterol and triglycerideconcentrations and in daily tobacco consumption.

The impact of the intervention strategy on theincidence of cardiovascular events (myocardial in­farction or sudden death) is shown in figure 2. Atthe end ofthe study the total number of events was47%lower in the intervention group, compared withthe control group (p = 0.028, 2-sided test). All diag­noses were confirmed by an expert panel, accord­ing to defined criter ia, and in a blind procedure.

543

Intervention

1 2Years

o

30

,-r-.....

--

~ - 15

I I30 20 10 10

b Decrease Increase (%

d

Number of men

15

~0E§.c0

.~

C 8.0 Control'"0c ........ ...00 - - _ ..... '0.. .... _- -"0... ..... -- ... 0 - - - -- ... 000

07.50;

;;;

'"0s:o 7.0

Invervention

A"I Ii I i

12 Start 1 2 3 4 5a Visit Years

~ 2.80E§. 2.6c.Q~ 2.4C'"0c

2.200

'":'20; 2.00>-g..= 1.8

Fasting intervention

~I i I i

12 Start 1 2 3 4 5c Visit Years

Fig. 1. Effects of intervention on (a) and (b) serum cholestero l concentrations, (c) serum triglyceride concentrations and (d) tobaccoconsumption. Pipe smoking is included, as 50g tobacco/week == 7 cigarettes/day (Hjermann et al. 1981, with permission).

Dietary and Anti-Smoking Advice: the Oslo Study 108

Intervention: n = 604

Fig. 4. Tobacco consumption at follow-up 2 to 3 years after the

end of the trial (Hjermann et at, 1986. with permission).

IIII

~Intervention: n =604 :

t

873 4 5 6I

IEnd of trial

Start 1 2Years

15>-

'"~~ 10~e'"OlU 5

806020 40

Time (months)

Control: n = 628

oz 10

l/)

E~~ 40

'""5ol/)

~ 30

~

'"o~ 20

Fig. 3. Serum cholesterol concentrations at follow-up 2 to 3 years

after the end of the trial (Hjermann et al. 1986. with permission).

Fig. 2. Cumulative incidence of cardiovascular end-points (first

myocardial infarction or sudden death) in control (e--e) and

intervention (0-0) groups during the study period (Hjermann

et al. 1986. w ith permission) .that a large number of people in the higher socio­economic groups had already stopped smoking; sopossibly only a 'hard core' of smokers was left atthe start of the study. This socio-economic 'effect'should be kept in mind when a strategy of dietaryand anti-smoking counselling is defined.

At the end of the trial the men in both groupswere told that they might not be recalled for re­examination, but they were to abstain from smok­ing and to continue their cholesterol-lowering diet.However, they were re-examined 2 to 3 years afterthe end of the trial (figs 3 and 4). The serum chol­esterol was unchanged in the intervention groupand had fallen in the control group, probably be­cause the control subjects had been informed aboutthe results of the trial. The smokers, however, de­spite being informed of the results, had resumedsmoking more or less as much as before (fig. 4).

Thus, in our study it appeared that those whogave up smoking had to be repeatedly educated inorder to remain as non-smokers, possibly becausethe most highly motivated persons had stoppedsmoking earlier and so had not been included inthe trial.

The cholesterol-lowering eating habits, how­ever, seemed to be easier to maintain. The reasonfor this could be that there were still enough health­conscious people left who were motivated to changetheir eating habits.

83 4 5 6 7II

End of tr ial

Control : n = 628

Intervention: n = 604

1 2 Start 1 2Visit Years

:J"

~ 325.S-(5

~ 300inQ)

(5

B 275

Response to diet and anti-smoking advice turnedout to be different for each socio-economic group.The best diet response was found in the highersocio-economic groups, while the best response toanti-smoking advice was seen in the lower socio­economic groups. The reason for this could be thatthose in the higher and more educated socio-eco­nomic classes more easily picked up the rathercomplicated message about cholesterol-loweringdiet. However, with regard to smoking, we knew

Dietary and Anti-Smoking Advice: the Oslo Study

One problem for the strategy in the trial wasthat bodyweight increase in those stopping smok­ing sometimes created a rise in blood pressure inparticipants with borderline pressure. The averagebodyweight increase in all subjects who stoppedsmoking was 2 to 2.5kg during the first I to 2 years,returning to the original weight thereafter. How­ever, borderline hypertensives who also have ele­vated blood lipids have an increased bodyweightfrom the start. After smoking cessation, an addi­tional increase in bodyweight may lead to an in­crease in blood pressure, and the cholesterol-low­ering diet may be counteracted by the weightincrease. In our experience, this problem can easilybe explained to most patients .

109

References

Hjermann I, Velve Byre K, Holme I, Leren P. Effect of diet andsmoking intervention on the incidence of coronary heart dis­ease. Report from the Oslo Study group of a randomised trialin healthy men. Lancet 2: 1303-1310, 1981

Hjermann I, Holme I, Leren P. Oslo study diet and antismokingtrial. Results after 102 months. American Journal of Medicine80 (Suppl. 2A): 7-11, 1986

Leren P, Askevold EM, Foss OP, Froili A, Grymyr D, et al. TheOslo Study . Card iovascular disease in middle-aged and youngOslo men. Acta Medica Scandinavica 588 (Suppl.): 1-38, 1975

Author's address: Dr IngvarHjermann, Department of Medicine,University of Oslo Medical School, Ulleval Hospital, 0407 Oslo4 (Norway) .