Formulating your Question
This page should be useful to everyone, in particular to anyone attending the SMACC Dublin workshop on evidence based emergency medicine and critical care. Watch the video on formulating answerable questions. Notes on question construction are shown below.
Asking the right question
Evidence based medicine works best when you can identify a specific problem and phrase it in such a way to allow it be answered. If you are doing a BET or any other EBM review and get this bit wrong you can waste a hell of a lot of time. In our journal club we encourage new authors to bring their three part questions along even before they start searching so that they get it right at the beginning. It is terribly frustrating when people come along with a purportedly finished BET only to find out that it was critically flawed from the outset.
We use the well known “three part question” system, which, unsurprisingly has three parts:
- Patient group
- Intervention or defining question
- Relevant outcomes
These three elements combine to form the question which then, and this is why it is so important, go on to define the search strategy that you will use to retrieve the evidence.
The patient group
Defining the patient group is essential and often takes a fair bit of thought. Making this completely specific to a small group (e.g. 35 year old men called Simon living in Virchester) means that if you find evidence it will be very applicable, but it does mean that you probably won’t find any papers at all. One way to get this right is to write down your patient group in such a way so that someone working in the same field as yourself would be able to instantly imagine that type of patient in front of them. Try not to include terms that are irrelevant. A common error is to specify a particular sex or narrow age range when it is unlikely to make any difference to the eventual outcome.
- Good: Adults with undisplaced Weber A ankle fractures to the lateral malleolus
- Not so good: Patients with ankle fractures (too broad)
- Worse: Inuit men aged 18-28 with a Jones fracture to the 5th metatarsal following a fall from a horse (too narrow)
Intervention or defining question
Having decided which group of patients we are talking about we then need to think about what we are going to do to them. This will vary slightly depending on whether we are considering therapeutic questions, diagnostic questions, prognosis questions etc. but the basic design will remain the same. In therapy, diagnosis or intervention questions we typically compare one strategy against another, for example in a trial of a drug we may compare treatment A against treatment B or in diagnostics we may compare a gold standard against a new imaging modality or blood test. In prognostic studies we can usually define a factor that we suspect the outcome that affects the patient group.
What is essential, is that this section of the three part question defines what we think is influencing the outcome that we will define in the final part of the question.
For example, in our previous case of the patient with an ankle fracture a number of interventions could be considered.
- Are NSAIDs better than paracetamol at…
- Is tubigrip better than plaster of paris at…
- Is early mobilisation better than standard 4 week immobilisation at…
- Is MR scan better than X-ray for…
- Is physiotherapy better than simple advice at…
- Are active sportspeople more likely than sedentary people to…
Oddly enough this is the element of the three part question that often causes the most problems. If we are to practice in an evidence based way then it is vital that we are able to define what is an important clinical outcome. We see many questions that define an outcome as something rather vague like “making it better”. It’s difficult to measure “better” and it does not help with the searching.
Outcomes for the three part question should ideally be:
- Clinically relevant
- Patient orientated
In many cases there will be more than one relevant outcome, that’s fine, put them all down. You may not always find the evidence for every outcome but it is important to define what you and your patient think are important. Going back to our previous example we could define many different outcomes depending on which aspect of care was needed. Pain, disability (short and long term), return to work are all potential important outcomes for Mr Jones. An example of a complete three part question that might answer a clinical conundrum for Mr Jones doctor could be.
[In adult patients with an undisplaced Weber A fracture of the ankle] is [Tubigrip better than immobilisation in plaster of paris] at [decreasing pain, decreasing time to return of normal activities and/or improving eventual range of movement]
By convention we place square brackets around each part of the question, this helps us get everything in the right place.
If you have got this far, that’s great. We would suggest that you now show your question to a colleague and ask them if it makes sense to them. Ideally show it to someone who has written their own three part questions in the past.
Once you have done this you are ready to register your question on the BestBETs website. This will allow the question to be registered under your name for a period of 6 months. You should also search on the database using the search function to make sure that no-one else has registered the question already. We will have a brief look at the question at this time and give feedback if possible.
If you are coming to the workshop.
Firstly, fantastic, you are most welcome and we are really looking forward to seeing you. We would lo0ve you to bring along a three part question and a related clinical scenario on the day. You can even post it on Twitter before the conference using the hashtag #smaccEBM so that we can give feedback in advance.
If you are not coming to the workshop or you missed out.
No worries, we’re glad you dropped in to read and learn with us. You can still post a question though, use the hashtag and you never know someone might pick it up and answer it for you.
- Mackway-Jones K, Carley S. bestbets.org: odds on favourite for evidence in emergency medicine reaches the world wide web. Journal of Accident & Emergency Medicine 2000; 17(4):235-236.
- Carley SD, Mackway-Jones K, Jones A, Morton RJ, Dollery W, Maurice S et al. Moving towards evidence based emergency medicine: use of a structured critical appraisal journal club.[see comment]. Journal of Accident & Emergency Medicine 1998; 15(4):220-222.
- Mackway-Jones K, Carley SD, Morton RJ, Donnan S. The best evidence topic report: a modified CAT for summarising the available evidence in emergency medicine. Journal of Accident & Emergency Medicine 1998; 15(4):222-226.
- Crombie IK. The pocket guide to Critical Appraisal. London: BMJ Publishing, 1996.
- Sackett DL HRGGTP. Clinical Epidemiology: A basic science for clinical medicine. Boston: Little Brown, 1991.
- Sackett D. How to teach and practice evidence based medicine. 2nd ed. London: Churchill, 2000.
- Greenhalgh T. How to read a paper. The basics of Evidence Based Medicine. 2nd Ed ed. London: BMJ, 2001.