TY - JOUR
T1 - Common medical and statistical problems
T2 - the dilemma of the sample size calculation for sensitivity and specificity estimation
AU - Oliveira, M. Rosário
AU - Subtil, Ana
AU - Gonçalves, Luzia
PY - 2020/8/1
Y1 - 2020/8/1
N2 - Sample size calculation in biomedical practice is typically based on the problematicWald method for a binomial proportion, with potentially dangerous consequences. This work highlights the need of incorporating the concept of conditional probability in sample size determination to avoid reduced sample sizes that lead to inadequate confidence intervals. Therefore, new definitions are proposed for coverage probability and expected length of confidence intervals for conditional probabilities, like sensitivity and specificity. The new definitions were used to assess seven confidence interval estimation methods. In order to determine the sample size, two procedures-an optimal one, based on the new definitions, and an approximation-were developed for each estimation method. Our findings confirm the similarity of the approximated sample sizes to the optimal ones. R code is provided to disseminate these methodological advances and translate them into biomedical practice.
AB - Sample size calculation in biomedical practice is typically based on the problematicWald method for a binomial proportion, with potentially dangerous consequences. This work highlights the need of incorporating the concept of conditional probability in sample size determination to avoid reduced sample sizes that lead to inadequate confidence intervals. Therefore, new definitions are proposed for coverage probability and expected length of confidence intervals for conditional probabilities, like sensitivity and specificity. The new definitions were used to assess seven confidence interval estimation methods. In order to determine the sample size, two procedures-an optimal one, based on the new definitions, and an approximation-were developed for each estimation method. Our findings confirm the similarity of the approximated sample sizes to the optimal ones. R code is provided to disseminate these methodological advances and translate them into biomedical practice.
KW - Conditional probability
KW - Coverage probability
KW - Sample size
KW - Sensitivity
KW - Specificity
UR - http://www.scopus.com/inward/record.url?scp=85089841299&partnerID=8YFLogxK
UR - https://www.mdpi.com/2227-7390/8/8/1258
U2 - 10.3390/MATH8081258
DO - 10.3390/MATH8081258
M3 - Article
AN - SCOPUS:85089841299
VL - Vol. 8
SP - 1258
EP - 1275
JO - Mathematics
JF - Mathematics
IS - n.º 8
ER -