PICO Framework

PICO breaks a clinical question into four parts: Population/Patient, Intervention (or exposure), Comparator, and Outcome. It is sometimes extended with Time and Study design (PICOT / PICOS). A well-formed PICO sharpens both the literature search and the study-selection that follow.

A scenario (“an adult with a suspected PE”) is decomposed into its four PICO components and then reassembled into a single, answerable clinical question. Each box names one element: the patient group, the intervention being considered, what it is compared against, and the outcome that actually matters.

Read it top-down: start with the patient in front of you, fill each of P, I, C and O, then read the assembled question back. If any box is vague or empty — especially the Comparator or a patient-important Outcome — the question (and therefore the search) will be unfocused. PICOT adds a time horizon; PICOS adds the study design you want.

A vague, unfocused question yields an unfocused search and irrelevant evidence. PICO forces you to state the comparator (an effect is always relative to something) and to pick an outcome patients care about — not a surrogate. It is the entry point examiners expect for evidence-based-practice and search-strategy questions.

• Population · Intervention/exposure · Comparator · Outcome
• Extensions: +Time (PICOT) · +Study design (PICOS)
• Always name the comparator and a patient-important outcome

Framing an ED question with PICO — a patient presents with low-risk chest pain. The vague question “is troponin useful?” becomes searchable as PICO: P adults with low-risk chest pain, I a high-sensitivity troponin pathway, C standard serial troponin, O 30-day major adverse cardiac events (a patient-important outcome). That structure maps directly onto trials such as the high-sensitivity troponin rule-out pathways. Swap the outcome for a surrogate (e.g. “troponin level” rather than 30-day MACE) and the question looks answerable but no longer reflects what matters to the patient — a classic framing error.

• Omitting the comparator — without one there is no interpretable effect.
• Choosing a surrogate outcome instead of a patient-important one.
• A population that is too broad (irrelevant hits) or too narrow (no evidence found).