Method comparison studies of telomere length measurement using qPCR approaches: A critical appraisal of the literature
Fig 5
A. The sample size required to test effect sizes of 150, 200 and 300 bp with a t-test with a power of 0.9, as a function of measurement error as expressed in the ICC. Calculations assumed a realistic (true) standard deviation of 650 bp and power analysis was done using G*Power [57]. N is the combined n of the two groups to be compared and was assumed to be equally distributed among the two groups. B, C. Power to detect a 33% change of telomere shortening rate, up or down, with p<0.05 relative to a baseline shortening rate of 25 bp/year. D. Four-year follow-up period. E. Eight-year follow-up period. Power was calculated for sample sizes as shown (200–2800), equally divided over the two levels of telomere shortening rate. Baseline telomere shortening was simulated assuming a Poisson distribution with mean/variance of 25, and population SD of telomere length was maintained at 0.65 kb at both time points.