How to calculate the appropriate sample size for the development of a new clinical test? The test would either be a quantitative test if the accuracy is good enough (i.e., test would predict a quantitative value similar to blood concentration) or be a semi-quantitative test if not accurate enough for quantitative testing (i.e., indicating statuses such as "healthy", "at risk", "disease"). The accuracy of the test would be evaluated by comparison with a quantitative gold standard.
Any help is appreciated.
EDIT: If the outcome of the test is binary (e.g. healthy/disease) I could find enough information in the literature regarding calculations of the sample size. But as far as quantitative data are concerned as well as "semi-quantitative" (more than two groups, like "healthy", "risk", "disease") I did not find any proper information on my own.
To calculate the appropriate sample size for the development of a new clinical test, you need to consider the type of accuracy you want to achieve, the effect size, the alpha level, the power, and other factors that may affect your sample size calculation.
Thank you Dinesh Kumar . Do you have any documentation on this topic? If the outcome of the test is binary (e.g. healthy/disease) I could find enough information in the literature regarding calculations of the sample size. But as far as quantitative data are concerned as well as "semi-quantitative" (more than two groups, like "healthy", "risk", "disease") I did not find any proper information on my own.
Pauline Delhez Please refer to Röhrig B, du Prel JB, Wachtlin D, Kwiecien R, Blettner M. Sample size calculation in clinical trials: part 13 of a series on evaluation of scientific publications. Dtsch Arztebl Int. 2010 Aug;107(31-32):552-6. doi: 10.3238/arztebl.2010.0552. Epub 2010 Aug 9. PMID: 20827353; PMCID: PMC2933537.
It is depend on population comes under your research area. The sample size should be 10% of the total population. I don't have right now reference for this. But For clinical, all the available sample will be first priority as per the requirements of research work selected by researcher.... Thank you.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis