91828 Chap.1 Evidence Questions and PICO
Evidence Questions and PICO
Evidence based practice begins with a decision that is specific enough to search and appraise. This chapter moves through the evidence cycle and shows how population, intervention, comparator and outcome turn a broad concern into a foreground PICO question. It also explains how to translate that question into searchable concepts without making the search so restrictive that relevant evidence disappears.
What this chapter covers
- 01
Evidence based medicine cycle
- 02
Foreground and background questions
- 03
Population definition
- 04
Intervention and comparator
- 05
Patient-important outcomes
- 06
Search concept groups
Worked example · free
Turn a clinic concern into PICO
- +1Population: adults newly prescribed antihypertensive therapy; intervention: text-message reminders.
- +1Comparator: usual care; outcome: medication adherence at twelve weeks.
- +1Build one concept group for hypertension or antihypertensive therapy and one for text messages or SMS.
- +1Connect synonyms with OR and concept groups with AND; test outcome terms only if retrieval remains broad enough.
Key terms
- Foreground Question
- A foreground question asks for evidence to guide a specific clinical decision.
- Search Concept
- A search concept is one idea represented by keywords and controlled vocabulary joined with OR.
- Comparator
- A comparator is the alternative against which an intervention is evaluated.
- Outcome
- An outcome is the patient or clinical result used to judge an intervention.
Evidence Questions and PICO FAQ
Does every question need all four PICO elements?
PICO is a useful structure for intervention questions, but some questions use adapted frameworks. Even within PICO, comparator or outcome terms may be omitted from the search when they make retrieval too narrow. The clinical question should still state the intended decision clearly.
Should I search the full PICO sentence?
No. Convert the question into core concept groups, list synonyms and controlled vocabulary for each, connect alternatives with OR, then connect groups with AND. Search systems retrieve indexed terms rather than understanding a full clinical sentence in the way a human reader does.
How narrow should the population be?
Specify characteristics that change the decision, such as condition, care setting or disease stage, without adding convenient exclusions that make evidence impossible to find. Refine after a scoping search and document the eligibility boundary used in appraisal.
When should I add a time point?
Add time when it changes which outcome or evidence is relevant, such as thirty-day readmission versus one-year survival. Timing can sit beside PICO as T. Do not add it mechanically if studies use several defensible follow-ups; instead prespecify how different time windows will be grouped.
How do I know whether the search is too narrow?
Test the strategy against a small set of known relevant studies and inspect which concept prevents retrieval. A precise clinical question does not require every element in the database string. Remove poorly indexed outcome or comparator terms before applying language, date or design limits that can hide relevant evidence.
What should I record when the question changes?
Keep the original wording, revised wording, date, reason and effect on search or eligibility. A small population or outcome change can create a different evidence question. A version record lets the team explain why studies entered and prevents a later appraisal from silently answering the earlier question.
Assessment move
Build mastery by moving from an untidy clinical concern to a reproducible evidence question several times, not by memorising the four PICO letters. Start with five situations from different settings and write the decision each clinician or patient must make. Convert each decision into population, intervention, comparator and patient-important outcome, then test whether two readers would apply the same eligibility boundary.
Add timing only when it changes the clinical meaning. Next, create a search translation table with one row per concept: natural-language phrase, synonyms, spelling variants, abbreviations and a likely subject heading. Connect alternatives within a row with OR and join the indispensable rows with AND. Predict which relevant paper a narrow term might exclude.
Run a paper-based sensitivity test by removing the comparator and outcome concepts, because those elements are often poorly indexed. For each version, explain the likely trade-off between missed studies and screening burden. Practise background questions separately so you can recognise when a mechanism or definition needs a direct reference rather than a PICO search.
Keep a search log containing the database, date, complete string, limits and result count; that record makes later selection decisions defensible. When reviewing results, label false positives by the concept that failed and refine vocabulary rather than adding arbitrary limits. End the session by writing a one-sentence evidence need: what estimate, in which population and over what time would change the decision?
For AT1 preparation, create a PICO approval checklist that tests relevance, feasibility, ethics and the availability of a meta-analysis, randomised trial and lower-level study. Discuss any change in wording with the team because a small shift in population or outcome can silently create a different question. Finally, explain the question aloud without the acronym.
If the clinical choice, alternatives and consequence remain clear, the structure is doing useful work rather than acting as a form to fill.
