Research question formulation

Research Question - FINER is Finer

A well-formulated research question is the foundation of a study. Use the FINER criteria to evaluate its quality.

📌 FINER Mnemonic:

• Feasible: Can it be done? Considers sample size, resources (time, money), and technical expertise.
• Interesting: Is it intriguing to the investigator and the scientific community?
• Novel: Does it confirm, refute, or extend prior findings? Avoids unnecessary duplication.
• Ethical: Can it be safely conducted? Must be approvable by an Institutional Review Board (IRB).
• Relevant: Will the results impact scientific knowledge, clinical practice, or health policy?

⭐ A strong research question often evolves from a PICO formulation: Patient/Population, Intervention, Comparison, Outcome.

Question Structure - The PICO Punch

A well-structured research question is the cornerstone of evidence-based medicine, guiding study design and literature searches. The 📌 PICO framework provides a mnemonic for formulating answerable clinical questions.

• P - Patient, Population, or Problem
  • Defines the specific group of interest. (e.g., adult patients with type 2 diabetes)
• I - Intervention or Exposure
  • The main treatment, diagnostic test, or exposure being considered. (e.g., treatment with metformin)
• C - Comparison or Control
  • The alternative to the intervention. (e.g., placebo, standard care like diet and exercise, or another drug)
  • Note: Not always necessary for all question types (e.g., prognosis questions).
• O - Outcome
  • The result or effect being measured. (e.g., change in HbA1c levels, incidence of cardiovascular events)

⭐ A precisely formulated PICO question is the most critical step for a high-quality literature search (like in PubMed) and for defining the scope of systematic reviews and meta-analyses.

Hypothesis - Null's the Word

• Null Hypothesis ($H_0$): Assumes no relationship or difference between groups. It's the baseline or "status quo" assumption. E.g., "New drug has no effect."
• Alternative Hypothesis ($H_1$): Contradicts the null. States that a relationship does exist. This is the researcher's claim. E.g., "New drug is better than placebo."
  • Can be one-sided (specifies direction, e.g., >) or two-sided (specifies a difference, e.g., ≠).
• Goal of Testing: Not to prove $H_1$, but to gather enough evidence to reject $H_0$. 📌 Think: "Innocent ($H_0$) until proven guilty."

⭐ The p-value is calculated assuming $H_0$ is true. It's the probability of getting your study results (or more extreme ones) purely by random chance if the null were correct.

High‑Yield Points - ⚡ Biggest Takeaways

• A strong research question is the essential foundation for a valid study.
• Use the FINER criteria to develop a question: Feasible, Interesting, Novel, Ethical, and Relevant.
• The PICOT framework is key for structuring clinical questions.
• Population, Intervention, Comparison, Outcome, and Timeframe.
• A well-defined question dictates the appropriate study design and prevents ambiguity in results.