Skip to main content
NCBI home page
The BMJ logoLink to The BMJ
. 2002 Oct 26;325(7370):969. doi: 10.1136/bmj.325.7370.969

Primary prevention strategies for cardiovascular disease

Gill Tyerman 1,2, Peter Tyerman 1,2, Trefor J Roscoe 1,2
PMCID: PMC1124458  PMID: 12399359

Editor—The assumptions made by Marshall and Rouse in their study are too broad and may lead to erroneous conclusions.1 Using the Framingham equation with average values is going to stratify the population only by age, sex, and diabetes status. Such stratification is simplistic and will prove only that older people and people with diabetes are more at risk. By assuming the subjects are non-smokers Marshall and Rouse exclude the most important single weighting in the Framingham calculation. Most practices have a register of smoking status for most of their patients, so this should be included.

Marshall and Rouse admit that their method may not be able to screen all of the population, so some will lose out. By concentrating on the oldest patients and patients with diabetes first, it will not detect some who have most to lose, such as 50 year old smokers with hypercholesterolaemia and hypertension. Adding a few extra years to the life of a 70 year old non-smoking normotensive diabetes patient may be laudable, but just because it is easier and cheaper does not mean that these patients should be prioritised to be assessed first, as this method seems to suggest.

Many of Marshall and Rouse's other assumptions do not fit well with the day to day realities of general practice. Few general practitioners would be able to stabilise antihypertensive treatment with only four prescriptions and two follow up consultations in a year. Government policy is to issue 30 days' supply, so Marshall and Rouse underestimate dispensing costs. Not all of this work can or should be done by practice nurses. Other diagnoses will be found requiring medical intervention, and many patients need more careful clinical assessment than a simple protocol provides. Marshall and Rouse's assumption that two 20 minute consultations by a practice nurse are all that is needed to treat is wrong, so their economic argument is flawed.

Their conclusion about statins and angiotensin converting enzyme inhibitors costing more, so strategies avoiding these may allow more disease to be prevented is odd. It would be actionable to omit a cholesterol measurement on a high risk patient or if a high level was found, negligent not to treat. Avoiding them is not an option in the real world.

This paper adds little to the subject other than confusing people into thinking there is a quick and cheap method of cutting short the hard and expensive, but necessary, work of finding all those at risk of heart disease and lengthening their lives.

References

  • 1.Marshall T, Rouse A. Resource implications and health benefits of primary prevention strategies for cardiovascular diseases in people aged 30-74: mathematical modelling study. BMJ. 2002;325:197–202. doi: 10.1136/bmj.325.7357.197. . (27 July.) [DOI] [PMC free article] [PubMed] [Google Scholar]
BMJ. 2002 Oct 26;325(7370):969.

Authors' reply

Tom Marshall 1, Andrew Rouse 1

Editor—Our principal conclusion has not been challenged: authors of guidelines should explicitly model the resource implications and health benefits of following their recommendations. The table shows the results of modelling using Tyerman et al's suggestions: 12 prescriptions a year, all assessment and follow up by doctors, patient assessment taking 1 hour, stabilisation and follow up taking 2 hours over five years. A practice that preselects 200 patients to assess prevents 82% as much cardiovascular disease at 57% of the cost.

Simvastatin costs more and is less effective than identifying and treating new patients with aspirin and antihypertensive drugs. Five years' treatment with simvastatin 10 mg costs £1175. Five years of aspirin, atenolol, and hydrochlorothiazide, is £718, including staff costs. The relative risk of cardiovascular disease with statin treatment is 0.7, with aspirin and antihypertensive drugs 0.6 (0.8×0.75=0.6). This will be explored further in a future paper.

We agree that there is an unresolved dilemma in cardiovascular prevention. A clinician acting in the best interests of an individual patient prescribes a statin; a clinician acting in the best interests of the whole practice population does not but employs a practice nurse to find and treat more patients. The former is consistent with the clinician's duty of care but is a poor prevention policy. The latter is consistent with the optimum use of public funds.

Table.

Resource implications and health consequences of cardiovascular disease prevention with Tyerman et al's suggestions

No of patients assessed
No of patients eligible for any drug treatment
Workload (h)
Benefit: CVD events prevented per 5 years
Cost per 5 years (£)
Assessment clinic
Follow up clinic
100 75 100 50 6.7 145 735
200 113 200 75 9.2 222 967
300 135 300 90 10.3 269 687
400 147 400 98 10.8 301 094
939 163 939 108 11.2 389 099
Open in a new tab

CVD=cardiovascular disease. 

We can also model the implications of Tyerman et al's preference for prioritising 50 year olds over 70 year olds. For 50 and 70 year old men life expectancy is 25 years and 11 years, respectively, and cardiovascular mortality (per 10 000) is 12 and 111.1-1 Per 10 000 people, cardiovascular disease therefore accounts for 312 and 1243 lost years of life in 50 and 70 year old men. We therefore can potentially add four times more years to life in 70 year olds than in 50 year olds. This is not necessarily irrational. Only a rational decision maker who valued the life years of 50 year olds four times more than those of 70 year olds would prioritise the younger men over the older ones.

We agree that information on smoking should be included when calculating prior risk estimates. This would improve the accuracy and efficiency of preselection. However, risk factor recording is not universally accurate,1-2 so we modelled our approach in a practice with the bare minimum of electronically recorded data.

With our preselection strategies, a woman's cardiovascular risk is equivalent to that of a man 12 years her junior. A good approximation of our preselection rankings is obtained by assigning a ranking number to every patient. For men it is their age, for women their age minus 12. Patients are prioritised for cardiovascular risk assessment by their ranking number (highest number first).

References

Articles from BMJ : British Medical Journal are provided here courtesy of BMJ Publishing Group

RESOURCES