Statistical power in UK genetic syndrome research; evidence from studies of Down syndrome, Fragile X syndrome and Williams syndrome as model syndrome groups.

Abstract

Research on genetic syndromes is vital to our understanding of how development unfolds, but the rarity of genetic syndromes can mean that studies are carried out with small sample sizes. Small sample sizes can reduce the statistical power of a study to produce reliable and replicable results. Here, we review all UK journal articles on three target genetic syndromes published from 2013 to 2022. There were 368 eligible articles. The median sample size of genetic syndrome groups was N = 30, and only 6.5% of articles reported a power analysis. Power analysis was performed on the 123 articles classed as 'Cognitive' research, as a test case. This demonstrated an average power of only 54% for a medium effect size and an alpha of 0.05. This is well below the recommended threshold of 80% power. The low power of UK genetic syndrome research has consequences for the replicability of the field due to the risk of Type II errors and reduced precision in effect size estimates, as well as implications for the communities that this research seeks to serve. We provide suggestions for researchers, journal editors and funders for improving the replicability of the field of genetic syndrome research.