AtlasSession 4L10: Sample size calculationsp.10
Binary outcomes
- 8 earlier topics
- CIs and p-values
- Beta error and power
- Why sample size matters
- Continuous outcomes
- Binary outcomes
- Next: No previous study
Fig. 1
Haemodialysis access: grafts blocked within 12 months (%), ePTFE graft vs tissue-engineered vessel, patients per group
Fig. 1. Two groups of 142 patients: ePTFE graft and tissue-engineered vessel. Filled dots show blocked within 12 months: about 23 in 100 and about 10 in 100.
- Patients per group
- 142
- Blocked within 12 months: ePTFE graft
- about 23 in 100
- Blocked within 12 months: tissue-engineered vessel
- about 10 in 100
Trial Design Board · First look
Level 1 · First look · Lecture 10 · p.10
Explain how sizing a study works for a yes/no outcome and why no spread needs to be looked up.
List the inputs for a binary-outcome sample size, explain why no SD is needed, and reproduce the 23% → 10% → 142 per group example.
Yes or no (the step you are on)
Each graft either blocks within a year or stays open; you count, not measure.
Each graft either blocks within a year or stays open, so you count, not measure: about 23 in 100 grafts block with the usual graft.