Intended for healthcare professionals

Rapid response to:

Education And Debate

Risk factor thresholds: their existence under scrutiny

BMJ 2002; 324 doi: https://doi.org/10.1136/bmj.324.7353.1570 (Published 29 June 2002) Cite this as: BMJ 2002;324:1570

Rapid Response:

Risk factor modification must be evidence-based

EDITOR-Wald and Law1 explore the log-linear association between
certain physiological variables, such as serum cholesterol and bone
mineral density (BMD), and their related diseases. Proportional risk
increments, they argue, should translate into similar risk reductions if
risk factor levels can be reversed. Moreover, the risk of disease ‘can be
judged…reversible’ across the entire distribution of a risk factor,
provided randomised controlled trials confirm benefit in subjects with
high levels of that risk factor. They conclude that treatment should
reflect absolute rather than relative risk, and that threshold-based
protocols are inappropriate. However these recommendations must be viewed
with caution, because the assumptions on which they rest are frequently
undermined by clinical evidence. For example, drug-induced improvements
in BMD neither fully explain2 nor guarantee reductions in recurrent
osteoporotic fractures. In primary prevention, alendronate reduces
absolute and relative risks of first vertebral fracture, but benefit is
inversely related to baseline bone mass, and absent at T scores > -
2.0.3 Whether such thresholds exist for reductase inhibitors in coronary
heart disease (CHD) is unclear.1,4 Secondary prevention trials have
confirmed that simvastatin (4S) and pravastatin (CARE and LIPID) lower LDL
cholesterol and reduce CHD outcomes in patients with average lipid
profiles.4 Subgroup analyses of the pravastatin studies suggest
attenuation or loss of CHD benefit if baseline LDL concentrations are
<3.2 mmol/l, while 4S excluded this group. Contemporary guidelines4
and clinical practice have reflected this uncertainty, although fresh
evidence suggests that simvastatin is effective in reducing vascular
events well below this threshold.5

There are various reasons why interventions may produce unpredicted
results. Drugs designed to reverse risk factors may have other important
pharmacodynamic effects, which may be unmeasured or unknown. Some risk
factors are difficult to measure and surrogates may be unreliable, while
other alleged risk factors may in reality be risk markers. BMD reliably
predicts native bone strength, but may not fully capture biomechanical
changes (positive or negative) in response to treatment. The
cholesterol/CHD hypothesis is persuasive, but statins do more than lower
cholesterol, and benefit may relate to plaque stabilisation and improved
endothelial function. This may help explain why the risk of vascular
events in the Heart Protection Study5 was positively correlated with
baseline levels of LDL cholesterol in the placebo group, but unrelated to
percentage reductions in LDL cholesterol in the treatment group. Wald and
Law rightly focus attention on absolute rather than relative risk, but
therapeutic decisions should still be based on empirical evidence rather
than theory.

Mark Garton consultant physician
Medicine and cardiovascular directorate,
Perth Royal Infirmary,
Perth PH1 1NX

1. Law MR, Wald NJ. Risk factor thresholds: their existence under
scrutiny. BMJ 2002:324;1570-6.
2. Cummings SR, Karpf DB, Harris F, Genant HK, Ensrud K, LaCroix AZ, et
al. Improvement in spine bone density and reduction in risk of vertebral
fractures during treatment with antiresorptive drugs. Am J Med
2002:112;281-9.
3. Cummings SR, Black DM, Thompson DE, Applegate WB, Barrett-Connor E,
Musliner TA, et al. Effect of alendronate on risk of fracture in women
with low bone density but without vertebral fractures. JAMA 1998;280:2077
-2082.
4. Scottish Intercollegiate Guidelines Network (SIGN). Secondary
prevention of coronary heart disease following myocardial infarction.
Edinburgh:SIGN;2000.
5. Heart Protection Study Collaborative Group. MRC/BHF Heart Protection
Study of cholesterol lowering with simvastatin in 20,536 high risk
individuals. Lancet 2002;360:7-22

Competing interests: No competing interests

07 July 2002
Mark J Garton
Consultant Physician and Rheumatologist
Perth Royal Infirmary, Perth PH1 1NX