Skip to main content Skip to navigation

Risk - what your clinician should tell you

Open Door - November 2007 pages 9-10

Richard Thomson, Professor of Epidemiology and Public Health
Institute of Health and Society, Newcastle upon Tyne Medical School

What should you expect from your clinician when discussing the course of your disease or the effectiveness of treatment? Increasingly today you should expect your clinician to advise you of both the risks and benefits of treatments. This is risk communication - an open two way exchange of information and opinion about risk, leading to better understanding and more informed decisions. Good communication in this area is based on evidence that patients who are better informed and more engaged in decisions about their own care are more knowledgeable, more likely to take their treatment, and may have better outcomes.

Whenever we take treatments there is a need to balance risks and benefits. If you take painkillers for arthritis, there is a very small risk that you might get a bleeding stomach ulcer. But you should be able to decide whether the risk of side effects is overridden by your wish for pain relief.

You might also be prepared to take greater risks for greater benefits. Thus, you might well take very toxic drugs to treat cancer, and accept a risk of serious side effects that would be totally unacceptable with a treatment for arthritis.

Consider also that we measure the effectiveness of most treatments not in absolute terms - this will work or it won't - but on the balance of probability. Thus we might treat a 1000 people with drug A, and 800 will get better and 200 won't. This is a good result overall, but not if you are one of the 200! So how can you understand these issues?

The language of risk

In our day to day life we may describe risk as low, moderate or high, or use terms such as probable, unlikely, and rare. However, interpretation of such terms varies widely from person to person. If your clinician uses them you might want to ask what he or she means by them.

Is there a better way that your clinician can present risks? Whilst there is no single correct way to do this, it is clear that some are better than others. For example consider the two statements below:

  • You have a 5% chance of having a stroke in the next year
  • Taking this treatment doubles your risk of having a heart attack in the next 10 years.

The first statement is difficult to interpret. Either you will have a stroke or you won't – you cannot have 5% of a stroke. A better way of presenting this information may be to say "Of 100 people like you, 5 will have a stroke in the next year".

The second statement gives a relative risk ie how many times more or less likely you are to have an event compared to others. This can be misleading; if I tell you that you have double the risk of suffering a heart attack compared to other people, that sounds like bad news. But if the risk for others is one in a million, this makes your risk two in a million – still a very low risk and an overall increase in risk for you of one in a million. However, if the risk for others is 10 in 100, then your risk would be 20 in 100 and this might be much more worrying for you. We saw this in the response to the presentation of risks of oral contraceptives in 1995, when it was announced that they doubled the risk of DVT and pulmonary embolus in women. This doubling of risk represented a very low additional risk of 15 episodes per 100,000 'women years', but led to many women stopping their contraceptive, with subsequent unplanned pregnancies which in themselves represented a greater risk to women than the pill itself.

Time frames

Presentation of risk should also give a time frame eg of a 100 people like you, three would suffer a heart attack over the next 10 years. If your clinician gives you a risk but doesn't say over what time period, ask for clarification.

Positive or negative?

The same outcome can be presented as a 97% chance of surviving (positive) or a 3% chance of dying (negative). Focusing only on the positives could lead you towards a choice that you might regret later if you feel a full and balanced picture was not provided. The obvious solution in this example is to present both the chances of survival and of death.

Conveying uncertainty

When we convey risk, we are communicating uncertainty. When discussing treatment you would like to know for certain whether you will recover or suffer an adverse effect, but it is rarely possible to provide such certainty. The use of phrases such as 100 people like you as well as such terms as "our best guess is..." may help convey this.

Visual presentation

Presentation in charts or graphs can improve risk communication, but this needs care. If your clinician presents risk using charts and graphs, and you don't understand, ask for an explanation – she may be able to present it in a different way. Indeed, there is value in providing several ways of presenting data to stand a better chance of meeting the varying needs of different patients.

Individualised risk presentation

A major challenge lies in conveying personal risk, rather than an average figure for a larger group containing dissimilar people. For example, it is not appropriate to say to someone with atrial fibrillation (irregular heart beat) that you have a five fold increase in the risk of stroke, not only because this breaks one of the cardinal rules above, but also because it disregards your individual risk factors, such as age, blood pressure, etc. Your clinician should, as far as possible, try and give you a feel for the risks for people like you, whenever such information is available.

Summary

Effective risk communication can improve your knowledge of your likely disease course and the effectiveness and risks of treatments, and involve you in better decisions about your treatment. In contrast, poor communication can make you more anxious, lose confidence in your clinicians, and have other adverse effects. You can expect better risk communication from your clinicians today, but don't be afraid to ask for clarification.

Return