The Author's reply: We appreciate Dr Kawada's1 interest in our article entitled ‘Impaired endothelial function in persons with obstructive sleep apnoea: impact of obesity’ recently published in Heart.2 Dr Kawada raises several questions concerning the statistical methods used in our article and the appropriateness of using these methods.

Dr Kawada's first question concerns the distribution of variables and the use of linear regression analysis. As reported in our article,2 we checked the distribution of all variables by normal quantile-quantile plots and histograms. Relevant to the linear regression model, age and apnoea-hypopnoea index (AHI) were not normally distributed. However, the inclusion of these non-normally distributed variables …