Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
Don't throw the baby out with the bathwater
- Kevin Pearce (9 July 2007)
-
Risky morality
- Anthony T J O'Brien, Anthony O'Brien (24 July 2007)
-
Re-thinking cardiovascular risk table interpretation.
- Thierry C Christiaens (27 July 2007)
|
|
|||
|
Kevin Pearce, GP Harrow HA2 6HL
Send response to journal:
|
Which would you prefer to have ? A drug that reduces your risk of dying from a heart attack by 17% or one that reduces your risk by 1.5% ? Similarly wouldn't you be scared to take a drug that increased your risk of a heart attack by 43% ? I know I would, but I am not so sure that I would worry about one that increased my risk from 17 in 10000 to 25 in 10000 ? The answer of course is that the drugs concerned are the same in each instance. The first example is from the Heart Protection study where simvastatin reduced the absolute risk of dying from heart disease from 9.1% to 7.6% over 5 years. A reduction of 1.5% in absolute terms but 17% in relative terms. The second comes from a recent metanalysis of rosiglitazone where the risk of heart disease went up from 0.17 to 0.25%. The difference illustrates the dramatically distorting effect of relative risk. In my own practice I use (at least in terms of cardiovascular risk) two measures. I give the patient the absolute risk -without which they cannot decide how bad things are. I then show them the risk that an "ideal" person would have of the same age which gives them the "extra" risk that their blood pressure or cholesterol confers upon them. I then leave it to them to decide if this is a risk worth worrying about. Absolute risk has its faults (and it is right that Dr Bonneux has highlighted these) but please let us not return to the days when only relative risk or the equally impenetrable odds ratios were the only risks presented. Competing interests: None declared |
|||
|
|
|||
|
Anthony T J O'Brien, General Practitioner Wyndham House Surgery, Silverton, Devon EX5 4HZ, Anthony O'Brien
Send response to journal:
|
Luc Bonneaux makes an excellent summary of the dilemmas concerning who, when, and how we should treat raised cardiovascular risks. His “red wine versus statin” example being the most notable illustration. However if we choose to address the moral questions he poses regarding the utilisation of our limited resources we might find it extremely uncomfortable. How much risk are we willing to tolerate before intervening? Maybe this is a question we should leave to individuals and concentrate our resources on those who are actually sick. After all if the healthy had to make their own choices regarding screening programmes, statins and other preventative measures they might pay more attention to how much (or how little) these actually alter their risk profiles. If we went further and suggested they should fund these choices themselves then many might prefer to invest their money elsewhere (eg. in a better bottle of red wine every week). The problem for the medical profession is that we have a conflict of interests. Many of our jobs depend on healthy people choosing medical interventions. Keeping them publicly funded is extremely beneficial to us. The moral questions may be less of a public or political issue than we would like to think. Cheers. Competing interests: None declared |
|||
|
|
|||
|
Thierry C Christiaens, Professor, Dept. of General Practice and PHC & Heymans Institute of Pharmacology, Ghent University UZG 1K3 De Pintelaan 185 B 9000 Gent Belgium
Send response to journal:
|
Editor, The cardiovascular risk concept may provoke a wide scale medicalisation of large parts of healthy populations in western countries (1). Therefore, the editorial of Luc Bonneux concerning the use of cardiovascular risk tabels(2) is not only interesting in the view of the new QRISK model (3). Yet, it is disappointing that the analysis of L. Bonneux stops at the point where it starts really to matter for every clinician: when to start medication? The use of an absolute risk score as a starting point for medication is dangerous and is not evidence based. Age is such an important risk factor that all risk tables are misleading. Implementing Framingham, ASSIGN, SCORE or QRISK gives the same strange reality: the older people are, the more statins you need to give them. But most studies with statins included few persons older than 70 years old. The PROSPER trial in this elderly population gives only a 15% lowering for the primary composite end point (Number Needed to Treat for 3 years of 48)and a marginally significant gain for cardiovascular death ( RR 0.76 with a 95% CI 0.58-0.99; NNT 112)and no effect at all on total mortality(4). A exemplary non-smoking men of 70 with a nice systolic tension of 140 mm Hg and a total cholesterol of 5 mmol (far lower than the median cholesterol in many countries)is at “high risk” (more then 20% risk in the Framingham table). In contrast, a smoking men of 50 years with a systolic tension of 180 mm Hg and a total cholesterol of 6 mmol/l is only at “median risk”. A better way of using risk tables should be by comparing the risk of an individual patient with te minimal risk of a person of the same gender and age. Treatment should be considered when he/she has -for example- the triple of that minimal risk for his/her age and gender. All attempts to make risk tables more accurate, as done in the QRISK algorithm (3), are necessary and should be welcomed. However, this is not the key problem. We have to fundamentally re-think how to use these risk tables in clinical practice. (1) Getz L, Sigurdsson JA, Hetlevik I, Kirkengen AL, Romundstad S, Holmen J.Estimating the high risk group for cardiovascular disease in the Norwegian HUNT 2 population according to the 2003 European guidelines: modelling study. BMJ 2005;331: 551-4. (2) L.Bonneux. Cardiovascular risk models. Moral Implications of models based on absolute risk could be better understood. BMJ 2007 335: 107-108. (3) Hippisley-Cox J, Coupland C, Vinogradova Y, Robson J, May M, Brindle P. Derivation and validation of QRISK. A new cardiovascular disease risk score for the UK.BMJ 2007;335:136-41. (4) Shepherd J, Blauw GJ, Murphy MB, Bollen EL, Buckley BM, Cobbe SM, et al. Pravastatin in elderly individuals at risk of vascular disease (PROSPER): a randomised controlled trial. Lancet 2002;360: 1623-30. Competing interests: None declared |
|||