The five steps of evidence based practice are commonly summarised as ‘Ask, Acquire, Appraise, Apply and Assess’. The first one of these – just asking a question – can prove terribly time consuming and difficult, but with a bit of dissection can be made much easier. The first step when deconstructing the anatomy of inquiry is to ask ‘What sort of question is being asked?’. If it’s about a clinical topic (not directions to the Pharmacy), then the questions can be grossly categorised as ‘foreground’ and ‘background’. Foreground questions are specific and pointed, and can be fitted into a ‘PICO’ frame :Patient-problem, Intervention, Comparision, Outcome. An example might be ‘in bronchilitis [problem], is ipratropium [intervention] or salbutamol [comparison] better for improving respiratory distress? [outcome]’ (The answer, for those who are interested, is ‘Neither’.) Background questions are much much broader, and often of the ‘what is’ or ‘what causes’ type: ‘what is a pilomyxoid astrocytoma?’, ‘what causes asthma in childhood?’.
There are another group of questions that aren’t quite either, though. Take ‘when should an LP not be performed as part of a septic screen in an infant?’. It seems quite specific, at first glance, but can’t be fitted into the PICO frame … and if you dissect it further, it’s probably an amalgam of a few different questions: In a febrile infant[P] does age [I], max temperature [C], clinical assessment of ‘unwellness’ [C], white cell count [C] or CRP [C] predict meningitis[O]? It may also have elements of clinical/societal judgement too: ‘How much risk am I/these parents/this society prepared to take of undiagnosed meningitis?’ (If it’s “zero”, then you’ll be needing to LP every child, for every febrile illness, and probably twice.)
It’s this last group of questions – the ‘not quite in focus’ that are the stumbling blocks for many in their search for evidence-based answers. If you hit one of these – your own or one of your learner’s – take a step back and rephrase, dissect and reconstruct. It may be one, two, or more individual questions. Then with a truly foreground focussed question you can move on to acquire the answer.
 Bob Phillips. Practicing evidence based acute paediatrics.
Current Pediatric Reviews, 2005, 1, 51-54