A step towards quality medical research

Only well planned and conducted research can answer questions that arise in day-to-day clinical practice. Unfortunately, most studies in India are def icient in design or execution, The results thus obtained are biased and can not be extrapolated to the population studied. Some important questions whi ch arise during the statistical design of investigations are selection of s ubjects on which observations are to be made, number of observations fallin g into different groups, allocation of subjects among different groups, etc . Besides planning and execution, appropriate methods need to be chosen for data analysis beforehand, If the study has been well designed and correctl y analysed, the interpretation of results can be fairly simple. The present communication focuses on some issues relating to the planning of investiga tions in medical research with emphasis on statistical design.