_6632_(8754693417).jpg){kind=link}
- Published
- Updated
The target difference entered into a sample size calculation is not the same as an expected result. Assuming a very large effect can produce a conveniently small study while limiting information about smaller effects that may still matter clinically.
Define the question and assumptions
Examine clinical importance, previous evidence and feasibility separately. Distinguish the desired difference from the difference considered plausible. Check whether variability, event frequency and follow-up assumptions refer to comparable populations. A statistician can show how the proposed design behaves under several defensible target-effect scenarios rather than one optimistic input.
Hypothetical study example
In a hypothetical trial, the high estimate from a small earlier study is the only planning source. The team evaluates its uncertainty and potential selection effects. It calculates information needs under more moderate effects, making the scientific consequences of funding and scheduling decisions visible to decision makers.
Design review checks
- Document clinical and empirical rationales for the target effect separately.
- Include uncertainty around the source estimate in the planning discussion.
- Evaluate the design under smaller effects that could still be clinically important.
Record the decision and its limits
An accepted decision does not disguise the affordable sample as the scientifically ideal target. If feasibility constrains size, explain the inferential limitation. State which effect sizes the trial can inform and which findings may remain uncertain, without promising success or asserting adequacy independently of the underlying assumptions.
Frequently asked questions
- Is the target effect the expected true result?
- No. It is a design assumption requiring its own supporting rationale.
- Should the largest previous effect be used?
- Not automatically. Assess uncertainty and contextual relevance first.