WARNING: the interactive features of this website use CSS3, which your browser does not support. To use the full features of this website, please update your browser.
When we collect data such as blood pressure in research studies, there is always some variability in the measurements due to error. This can be caused by things like day-to-day fluctuations in the values or by imprecision in the equipment used to take the measurements. It is possible to get an idea of how large the measurement error is and what effect it has on our conclusions by re-measuring some of the participants in a repeatability sub-study and calculating correction factors for the associations seen between error-prone measures and outcomes such as mortality. Researchers have to re-measure enough people to estimate the correction factors precisely enough, but not so many that it becomes too expensive or time consuming. We have written a paper explaining to researchers how they can calculate how many people to re-measure in their repeatability sub-studies. We ve illustrated this using the International Project on Cardiovascular Disease in Russia as an example, and have provided estimates of several correction factors from the UK Biobank study to help researchers plan their own studies, in our paper Reflection on modern methods: calculating a sample size for a repeatability sub-study to correct for measurement error in a single continuous exposure published in the International Journal of Epidemiology