1- Health Metrics Research Center, Institute for Health Sciences Research, ACECR, Tehran, Iran
Abstract: (1327 Views)
Dear Editor,
In “before-after" intervention studies with qualitative outcome(s), we should not use the same formula when the outcome(s) are quantitative since the appropriate variance in this formula are not used and thus the sample size will not have adequate confidence and power. A formula for such studies is proposed to overcome problems highlighted above.
Subject:
Letter to editor Received: 2023/10/29 | Accepted: 2023/11/1 | ePublished ahead of print: 2023/11/13 | Published: 2023/12/27
References
1. Haeri-Mehrizi a A, Sadighi J. Why should we report the correlation coefficient or the variance of the score differences in "before-after" intervention studies when the dependent variable is continuous? Payesh 2023; 22: 653-655 [Persian] [
DOI:10.61186/payesh.22.5.653]
2. Connor R. J. Sample size for testing differences in proportions for the paired-sample design. Biometrics 1987; 43:207-211 [
DOI:10.2307/2531961]
3. Chow S, Shao J, Wang H. Sample Size Calculations in Clinical Research. 2nd Edition, Chapman & Hall/CRC Biostatistics Series: USA, 2008