Intended for healthcare professionals

Editor's Choice

Balancing benefits and harms

BMJ 2013; 346 doi: (Published 05 June 2013) Cite this as: BMJ 2013;346:f3666
  1. Fiona Godlee, editor, BMJ
  1. fgodlee{at}

It is a basic principle of pharmacotherapy that all drugs have beneficial and harmful effects. And as Risto Huupponen and Jorma Viikari say in their editorial this week, it is the doctor’s job to find a justified balance between them (doi:10.1136/bmj.f3156). When it comes to statins, the balancing act must now include the risk of diabetes. But is this a class effect or is the risk higher with some statins than with others?

A meta-analysis in the Lancet three years ago identified an increased risk of diabetes in people taking statins. Now, Aleesa A Carter and colleagues have analysed data from nearly 500 000 new users of statins. They found a greater risk of diabetes with the more potent statins and with higher doses (doi:10.1136/bmj.f2610). Huupponen and Viikari conclude that the overall benefits of statins still outweigh the risks but recommend maintaining the lowest possible potency and dose.

Unfortunately in the balance between benefits and risks, it is an uncomfortable truth that most drugs do not work in most patients. On the positive side, Andrew Moore and colleagues say that if we can embrace the fact that most treatments fail, we will deliver better and safer care. But this needs a radical shift in the way we evaluate and use drugs (doi:10.1136/bmj.f2690).

Using data from systematic reviews in pain medication, the authors found that fewer than half of patients achieved at least a 50% reduction in pain intensity, with failure rates highest over the longer term in patients with chronic pain. But, importantly, hidden within this sobering picture are a minority of patients who respond well to treatment.

The authors recommend a different approach to the data—responder analysis—which reports the proportion of patients achieving outcomes that patients consider worthwhile. Clinical trials that don’t take this approach will underestimate the effectiveness of treatments, they say. They call for a change in the way regulators decide on which drugs to license. “Regulators need to recognise that failure is the norm,” they say.

As for clinicians, if you expect failure in individual patients and act swiftly when it occurs, your patients are more likely to get the best and safest treatment. If a drug fails it should be stopped, avoiding its adverse effects and opening the door to other treatment options. “Only effective drugs should continue to be prescribed,” they say.

This may sound obvious, but the key here is to move away from “a slavish reliance on the average.” And because success rates are low, a wide range of drugs is needed to do the best for most patients, especially in complex chronic conditions. Practice guidelines therefore need to reflect the realities of drug failure. Instead of restricting treatment options to one or two drugs, less restrictive guidance centred on the interaction between patient and clinician may do better. The authors hold up as a good example of such guidance the guidelines on osteoarthritis from the National Institute for Health and Care Excellence.

The challenge for doctors is to find what works for whom in what circumstances. As well as evidence, the authors prescribe “a large dose of clinical wisdom.”


Cite this as: BMJ 2013;346:f3666