What should the study update say?
Separate verified outcomes from outcomes missing at the cutoff.
The question
An internal update describes a fictional study’s results at one cutoff. The biostatistician owns the analysis. The clinical lead owns interpretation.
| Records | |
|---|---|
| Due for verification | 80 |
| Recorded as meeting the study outcome | 44 |
The draft uses the full 80-record group as its denominator.
“What can you put in the brief, and what needs specialist review?”
Not saved by this site. Your browser may restore text when you navigate back.
A strong answer, step by step
Open each step in turn. This is one strong answer, not the only one.
Step 1: Name the draft denominator
What would you do here? Open the move you’d make.
Check how many outcomes are verified at the cutoff.
The strong move. The denominator can include records with a known result and records still awaiting verification.
Describe 55% as a confirmed study result.
Tempting, but it misses something. The draft gives no evidence that all 80 outcomes are confirmed.
See how a strong candidate says it
“The draft counts 44 of the 80 due records. I’d verify how many of the 80 have confirmed outcomes before describing a rate.”
Why it works: The denominator can include records with a known result and records still awaiting verification.
Step 2: Ask for the cutoff counts
What would you do here? Open the move you’d make.
Ask the analysis owner for verified and unresolved counts.
The strong move. Using one cohort and window lets the reader see exactly what the numbers cover.
Use the counts from a different window if they are complete.
Tempting, but it misses something. A different cutoff can change who is counted. The committee asked for this one.
See how a strong candidate says it
“I’d ask the biostatistician for the verified counts and confirm the cohort and window. I’d keep both in the update.”
Why it works: Using one cohort and window lets the reader see exactly what the numbers cover.
Step 3: Separate the missing outcomes
The biostatistician confirms 64 verified outcomes. Of those, 44 meet the measure. For 16 of the 80 records there is no verified outcome. The draft counted them as below the measure.
What would you do here? Open the move you’d make.
Report 44 of 64 verified outcomes and give the missing count.
The strong move. The verified subset and the fixed 80-record group answer different questions. The limits show what the missing outcomes could change.
Use 68.75% as the confirmed result for all 80.
Tempting, but it misses something. The 16 unresolved outcomes are outside the verified denominator.
See how a strong candidate says it
“44 of 64 verified outcomes meet the measure, or 68.75%. There are 16 unresolved records. Across the 80, the arithmetic limits are 55% if none of the unresolved meet the measure and 75% if they all meet it.”
Why it works: The verified subset and the fixed 80-record group answer different questions. The limits show what the missing outcomes could change.
Step 4: Give a bounded update
What would you do here? Open the move you’d make.
State the evidence limits and ask both owners for their review.
The strong move. The committee gets the observed count and sees which decisions belong to the specialists.
Use the arithmetic range to make a clinical conclusion.
Tempting, but it misses something. The range accounts only for unresolved outcomes. The study analysis and clinical interpretation require their owners.
See how a strong candidate says it
“I’d say 44 of 64 verified outcomes meet the measure and 16 are unresolved at this cutoff. I’d ask the biostatistician for the reviewed analysis and the clinical lead for its interpretation.”
Why it works: The committee gets the observed count and sees which decisions belong to the specialists.
The answer, in short
- Decision
- Give the committee the verified count and the unresolved count.
- Because
- The draft mixed missing outcomes with verified outcomes below the measure.
- What would change it
- A reviewed analysis and the clinical lead’s interpretation could support a fuller update.
Common traps
- Treating an unresolved outcome as a verified result.
- Using arithmetic limits as a clinical conclusion.
Check your answer
- I named the cutoff and group.
- I checked both counts.
- I kept the denominators separate.
- I named the specialist owners.
Last reviewed Nov 11, 2026