Authors’ response

Sir,

I appreciate the interest shown by Dr Raina on our article on epidemiological trends of RF/RHD in school children of Shimla in north India1. He has raised certain concerns about validity of statistical tests used in our study. The objective of the study was to document the change in prevalence of RF/RHD over a period of 15 years in the context of changes in socio-economic state in defined geographical area in a specified target population using similar screening protocol and tools. It is evident from the epidemiological studies that distributions of RF/RHD reflect the status of human development index globally234. Thus the variables related to poverty, overcrowding, and health care facilities and indicators of public health that existed at the time of these two survey studies have been recorded and revealed substantial change. Demographic characteristics of the study population of these two survey studies were also similar. Our interest was not to document the relative contribution of changes in different indices of socio-economic and health care services (that are important determinants of risk of RF/RHD) on prevalence of RF/RHD, therefore, use of regression modelling was not required to adjust for effect of different variables related to socio-economic state. Regarding the comment about “inappropriate” use of Z test in evaluating the statistical significance of change in the prevalence of RF/RHD documented after 15 years time we believe that the Z test is an accepted statistical test for evaluating the change in prevalence of RF/RHD estimated at two points. Chi square and Fisher exact tests (if numbers of observations are less than 5) are accepted and valid statistical tests to assess the significance of differences in the distribution of severity of valvular dysfunction observed in study population evaluated in 1992-1993 and 2007-2008. Therefore, concerns raised by reader in our opinion are not appropriate.