What tests are available for people in my situation?

It is not always useful to have a diagnostic test – for example, if your condition is one of three possibilities, all of which have the same treatment, you may consider that it is not worth the time and effort to have tests. And the mere fact that a test exists does not mean that it is necessarily useful. For example, many people with acute low back pain are referred for X-rays, although the results of such tests do not provide a reliable guide to what treatment should entail or even to the cause of the problem.

Any test involves risks and most tests are not 100 per cent accurate. If there is a risky test for your illness but a relatively safe treatment, you might prefer trying the treatment first to see if the problem disappears. However, if there is real uncertainty about what your problem might be, and what is the best way to treat it, a diagnostic test might be valuable.

What treatments are available for people in my situation?

You need to know whether the treatments are aimed at curing the condition or simply relieving the symptoms. You may wish to consider a range of options from different types of practitioners, for example, your pharmacist, GP, specialist or homeopath.

How accurate are the tests in people like me? Could having the tests be harmful?

Diagnostic tests almost always involve some inaccuracy. As a result of this, a positive test result does not necessarily mean that you definitely have the disease, nor does a negative result mean that you definitely do not have the disease; there will always be some false-positive and some false-negative test results.

Say you decide to have a test for an infection. If the result is positive, you may have, for example, only a 60 per cent chance of having the infection – or a 90 per cent chance – depending on the test’s accuracy. (In other words 40 per cent – or 10 per cent – will be false-positive results.) If the test is negative, however, you still have a chance of having the infection – perhaps somewhere between a 5 and a 15 per cent chance. This is why it is important to find out, in terms of probabilities, exactly what the test results might mean in your case. If your practitioner simply says the result means that you are at ‘low risk’ or ‘high risk’, you may understand these terms very differently from your practitioner, whereas it is much clearer if you are told you have a ‘10 per cent’ or a ‘95 per cent’ risk.

If your practitioner suggests that you have a test, you should consider asking these questions:

• What disease are you testing for?
• What do you think my chances are of having the disease?
• If the test result is positive, what is the chance that I do have the disease? Or, if the test result is negative, what is the chance that I have the disease anyway?
• How will the test result influence treatment of my condition? (If the result will have no effect on your treatment, you may want to think twice about having the test.)
• What are the potential harms of the test?

(For further information on the accuracy (sensitivity and specificity) of tests see Chapter 16. For further approaches to deciding when a test is worthwhile, see Chapter 17 on decision thresholds.)

How effective are the treatments in people like me? What aspects of my health could be improved by the treatment? Could having the treatment be harmful?

You need to know how the benefits and harms of any treatment might affect your quality of life and your survival, rather than what it is likely to do to your red blood cell count or your blood pressure – often called surrogate measures. Surrogate measures describe the results of different biological tests. They are not necessarily an accurate reflection of what is most important for you, the quality and length of your life – these are person-centred outcomes. It’s important for you to know what aspects of your health could be improved by the treatment, not just what numbers might appear on a laboratory printout of test results.

Surrogate measures are used as markers for disease but do not describe the more important person-centred outcomes that affect how long and how comfortably you might live. For example, knowing that a treatment will lower your blood pressure, or that some diet will alter the bacteria in your bowel may be interesting, but tells you little about whether your pain will be reduced or your symptoms relieved. It also tells you nothing about whether you will live longer (on the benefit side) or whether you might feel lethargic, lose your hair or have diarrhoea (on the harm side).

Research has not, however, always examined the effects of treatments on person-centred outcomes; instead it often examines their impact on surrogate measures, such as blood counts. Sometimes, in the absence of evidence about whether your life will be happier or longer, you may have to make decisions based on the assumption that a beneficial effect on surrogate measures will translate into a beneficial effect on your health. For example, many people with HIV began taking drug cocktails because they had been shown to reduce the amount of virus in the body, but before it was known whether this would improve the length or quality of their lives. Understandably, as HIV is such a serious disease, many people were willing to take such medications because of evidence showing that they improved surrogate measures.

But such assumptions can be mistaken, as we describe in more detail later, with the drug that was widely used to treat arrhythmias – an abnormal heart rhythm that can occur after a heart attack. As people with arrhythmias are more likely to die after a heart attack than those without arrhythmia, it was assumed that giving a medication to suppress the arrhythmia would prevent deaths. When a randomised controlled trial was eventually completed, it showed that the antiarrhythmia drug suppressed arrhythmias but increased the number of people dying. Similarly, the drug mibefradil was taken off the market around the world after it became apparent that the drug, although effective at lowering blood pressure, could also cause serious problems when interacting with other drugs.

How likely are the benefits of treatment for me? How common are the harms?

You can never be certain about the outcome of any treatment; it is impossible to predict exactly what will happen to you when you have a disease or embark on treatment. What can be predicted, however, is the probability that a particular outcome will occur.

This issue is more complex than it might first seem. Consider, for example, the prediction that taking a certain drug will reduce your risk of dying from a heart attack by 20 per cent over the next 5 years. What if the drug is also known to increase the death rate from all other causes by 20 per cent over the same period? You may assume that this 20 per cent decrease in coronary deaths and 20 per cent increase in other types of death will balance out. In fact this may not be the case if fewer people die over this period from coronary heart disease than from all other causes combined.

Making an informed decision about whether to take this drug will depend on what is known about your risk of dying from a heart attack versus your risk of dying from other causes over the next 5 years. The probability that an individual will experience particular benefits or harms from the treatment is related to that person’s level of risk. A healthy adolescent, for example, will have little risk of dying from a heart attack over the next 5 years so does not stand to benefit from taking a drug to reduce death from heart disease. For him or her, the 20 per cent increase in the risk of dying from other causes clearly outweighs the benefit of the intervention. In other words, a 20 per cent reduction in a relatively uncommon cause of death is not balanced out by a 20 per cent increase in a more probable cause of death.

At the other extreme is someone at very high risk of dying from a heart attack – such as someone who has already had a heart attack or who has unstable angina. For someone in this situation, the benefits of the treatment are likely to outweigh the harms.

Another example might be to consider by how much your risk of getting or dying from bowel cancer might be reduced by having a faecal occult blood test (FOBT) every 2 years. Several large randomised controlled trials have shown that doing this will reduce your chance of dying from bowel cancer by around 23 per cent. We also know that your risk of bowel cancer increases with age, and that it is more common in men and increases if you have a family history of bowel cancer. So if you are a 60-year-old man whose mother died of bowel cancer in her 70s, your risk of dying from bowel cancer over the next 10 years is about 13 in 1000 and this is reduced to 10 in 1000 with 2-yearly FOBT. If your risk is lower to start with, then the 23 per cent reduction won’t be quite as much. The number of lives saved by screening will be much less if you are younger and don’t have a family history.

So it is important that you know not only the probability of benefits and harms from a particular intervention, but also how they relate to your situation.

When could I expect to see these benefits and harms?

Knowing when benefits and harms are likely to occur – whether they are likely to occur immediately or years down the track – can have a great bearing on their significance. For example, you may have to weigh up the immediate harms of chemotherapy for cancer – such as nausea, hair loss and discomfort – versus the chance that it will prevent future recurrence of disease and death.

How long lasting are the benefits and harms if they occur?

Knowing how long a benefit or harm is likely to last is important for evaluating its impact on your well-being. If a harm is temporary, you may consider it worth suffering in order to gain a longterm benefit. Alternatively, you may think twice if the trade-off for a short-lived benefit is a permanent harm.

An example of a temporary harm is a rash as a side effect of a treatment. When treatment stops, the rash clears. An example of a permanent harm is the loss of vision as a complication of surgery. This harm may not be reversible.

Does the available information answer my questions?

You could ask your practitioner if there are any evidence-based guidelines that cover your situation. (Evidence-based guidelines are based on a systematic examination of the best available evidence rather than on the opinions of experts, which, as we have seen, are not as reliable.) If an evidence-based guideline is not available, you could ask about the results of systematic reviews or randomised controlled trials. We discuss evidence-based guidelines briefly towards the end of this chapter and, along with the other study types, in much more detail in Chapter 10.

Many people also find it useful to speak to other people who have been in a similar situation, to get more information on their subjective experiences of the harms and benefits of a particular intervention. Remember, however, that different people experience things differently, so use their experiences only as a guide. In addition, such reports will not tell you how likely you are to experience a similar outcome. Your practitioner should be in the best situation to find out how likely you are to be helped or hurt by an intervention. If you are dissatisfied with the information provided by your practitioner, tell him or her that you still feel that you that have insufficient information and want some help with getting more.

Have I found out about all the test and treatment options that I want to consider? If not, where can I go to find out more?

You may need to think about other options if, for example, you have tried a treatment but had to stop it because you experienced a harm that was described as unlikely but none the less occurred, or if the benefit did not satisfy your needs. Or perhaps none of the options to date seems to have sufficient benefit for the harms.

If you need more information about your options, ask your practitioner, other practitioners or self-help groups. Many people find information from libraries, the internet and various electronic databases, but be aware that much information on the internet is unreliable. Later in the book we describe how to decide which information is reliable and list some good internet sites. Remember to compare new options with the best previous ones and to evaluate them with the same five important questions.

You may find it useful to seek a second opinion. There is no need to feel awkward about asking for this; most responsible practitioners will respect your right to see someone else. But be careful when selecting the number and type of practitioners whom you see; doctor shopping in order to get the answer that you are hoping for is not in your best interest. If you have difficulty making a choice, do not hesitate to discuss it with your practitioner. Remember, it’s your body, it should be your choice.

What makes a guideline evidence-based?

Evidence-based guidelines should use the accumulated high-quality evidence from research on a particular topic and recommend ways to apply this evidence to individual people who vary in their preferences and in the features of their illness. Here are some ‘ideal’ criteria to help establish whether a guideline is of high quality. They have been adapted from some internationally recognised standards from the Appraisal of Guidelines Research and Evaluation Collaboration (AGREE).⁸ A Guideline:

• should be recent, e.g. within the last 5 years. If it is not, ask if there is a more recent one. Guidelines may be updated more frequently depending on whether research has suggested changes in management.
• should be clear for whom the guidelines are intended and what they plan to address.
• should describe all the treatment options.
• should describe outcomes that are person-centred – about survival and quality of life.
• should describe both the benefits and the harms.
• should describe how the best evidence was selected and report the highest level of evidence for each recommendation. It may happen that a guideline’s supporting evidence is not ideal because no strong evidence on the topic exists, but this does not mean that such a guideline should be dismissed out of hand. It may still represent the best available information. What is important is that a guideline’s sources of evidence should be declared so that everyone knows its evidence level. The levels of evidence for evidence-based practice guidelines, from strongest to weakest, are:

  1. an evidence-based practice guideline or systematic review of all randomised controlled trials on the topic
  2. randomised controlled trials
  3. other non-randomised studies on groups of people
  4. case studies and opinions.
• Its development should have involved the main stakeholder groups across the relevant disciplines (for example, guidelines about managing a particular condition).
• It should be clear and accessible.
• It should be practical and relevant to its intended users.
• Its developers should have no conflicts of interest and it should be editorially independent of its funding body.

If all of these criteria are fulfilled, an evidence-based guideline should help you answer the five questions raised in this chapter.

Some people may still have some reservations about asking about the evidence behind medical advice for fear of implying a lack of trust. Many practitioners expect and welcome patients’ involvement in their healthcare decisions and this is increasingly the case. Health decisions have become far too complex to expect practitioners to have the correct answer to every problem at hand. Increasingly practitioners are using evidence-based practice guidelines to care for their patients, so be confident in asking your practitioner about them.

Not only is it your right to ask about the evidence, it is also in your best interest to do so. Evidence is not a substitute for clinical judgement, but should be used in conjunction with it. Clinical judgement alone, without the benefit of evidence from high-quality research, is not always reliable – especially for important decisions.

The rest of this chapter explores a minor and a major health decision, to illustrate how the five questions might help you ensure (as far as possible) that you will be making the best health choice.

Scenario I. What should Fred do about his arthritis?

Fred, 60, is a retired engineer who is in fairly good shape apart from a nagging pain in his knee caused by osteoarthritis. Although there is no cure, there are many interventions that could help his condition: weight loss, exercise and physiotherapy, and pain-killers. As Fred is thin and fairly active, the most reasonable option in this situation seems to be the pain-killers.

Scenario II. What should Pat do about her mammogram result?

Pat, 55, lives in a rural town and teaches physical education at a secondary school. She has been told that she has an abnormality on a screening mammogram that she had during a visit to her sister in another town. The initial positive mammogram was followed up by a recall for further mammograms, the results of which suggested cancer.

Let’s see how Pat answers the five questions, with the help of the consumer guideline A Guide for Women with Early Breast Cancer by the National Breast Cancer Centre (NBCC)¹⁰ and the doctor’s version Clinical Practice Guidelines: Management of Early Breast Cancer (National Health and Medical Research Council or NHMRC).¹¹ These are the Australian guidelines that she chooses to use because the UK guidelines are not due to be published until 2009 by the National Institute for Health and Clinical Excellence (NICE). (The Scottish health department issued guidelines for clinicians only, in March 2007.)

The other excellent resource that Pat finds on the internet is an evidence-based decision aid from the Canadian Cancer Society called Making Decisions about the Removal of My Breast Cancer at www.cancer.ca/ccs/internet/miniapp/0,3182,3543_16897665_19702640_langId-en,00.html