Pilot and feasibility studies

Pilot Studies - Testing the Waters

• Small-scale, preliminary studies conducted before a large-scale trial to assess feasibility.
• Key Objectives:
  • Test recruitment strategies & consent processes.
  • Evaluate and refine study protocols & interventions.
  • Assess data collection methods & instruments.
  • Estimate sample size parameters (e.g., standard deviation).
  • Identify potential logistical problems.

⭐ Pilot studies are primarily for assessing feasibility and are not powered for formal hypothesis testing. Their results should be interpreted with caution.

Core Objectives - The Pre-Flight Checklist

• Assess Feasibility: Evaluates recruitment potential, participant retention rates, and overall willingness to participate. Can the study actually be done?
• Refine Protocol: Tests and improves study instruments, intervention acceptability, and data collection methods before the main investment.
• Estimate Key Parameters: Provides preliminary estimates of effect size and standard deviation to inform the definitive sample size calculation ($n$) for the main trial.
• Logistical Dry-Run: Tests the project's budget, resource allocation, and proposed timelines to identify potential operational issues early.

⭐ Pilot studies are not designed for hypothesis testing. Being underpowered, a non-significant result doesn't rule out a true effect. Their primary goal is to inform the design of the main trial.

Design & Analysis - Sizing It Up

• Core Objective: To assess feasibility and estimate parameters for a future, larger study, not for hypothesis testing.
• Sample Size:
  • Not based on formal power calculations.
  • Aims to reliably estimate key variables like standard deviation or recruitment/dropout rates.
  • General guideline: 10-12 participants per group.
• Statistical Analysis:
  • Emphasizes descriptive statistics and confidence intervals over p-values.
  • Focus is on precision of estimates, not statistical significance.
  • Uses pre-defined progression criteria (e.g., meet 80% of recruitment goal) to determine if the main study is viable.

⭐ A key pilot study outcome is estimating the outcome's standard deviation, which is essential for performing an accurate power calculation for the main trial.

• Pilot studies are small-scale versions of a main study, testing the entire research process (e.g., recruitment, randomization, data collection).
• Feasibility studies ask if a future study can be done, questioning its overall practicality and viability.
• The primary goal is to identify problems before committing to a large, expensive trial, allowing for protocol refinement.
• Data can inform the sample size calculation for the main study.
• They are not for hypothesis testing due to being statistically underpowered.