Evidence Based Medicine - Asking an answerable question

We are regularly confronted with clinical situations where we do not have complete knowledge of the subject.  Part of the process of improving our knowledge involves being able to ask an answerable question.  A good clinical question has four components (PICO):
1.  Patient
     -  What are the important characteristics?
2.  Intervention (or exposure)
     -  Drug/test/procedure/environmental exposure
3.  Comparison
     -  What are you comparing the intervention to?
4.  Outcome
     -  What are you hoping to measure?

Clinical questions are commonly assessing therapy, harm, diagnosis, or prognosis.

Clinical scenarios

1. I was recently called to see a patient with an exacerbation of fluid overload in the setting of severe dilated cardiomyopathy (EF 18%, normal right heart).  The medical registrar felt that the patient needed a period of cardiac monitoring in the HDU.  I noted he had previously been considered for an AICD.  This begs the question (assessing a therapy):
• In adult patients with dilated cardiomyopathy and low ejection fraction, does the insertion of an AICD reduce the incidence of arrhythmogenic sudden cardiac death, compared with medical management?
2.  I recently attended the ED to see a patient transferred from a small community via RFDS where he had been diagnosed with NSTEMI.  Despite improvement in his (non-thrombolysable) ECG changes and adjunctive medical management (thrombin inhibition with Heparin, anti-platelet agents, GTN infusion, and analgesia) he had ongoing ischaemic-sounding chest pain.  I considered GP IIb/IIIa inhibitors and wondered (assessing harm):
• In adult patients with unstable acute coronary syndrome, does the addition of a GP IIb/IIIa inhibitor to standard medical treatment result in increased risk of major bleeding events or death?

Hopefully at some point I'll find the time to answer these questions!