INFLUENCES ON INFERENCES - EFFECT OF ERRORS IN DATA ON STATISTICAL EVALUATION

Background. Inadvertent random and systemic errors introduced into dat a sets and manipulation of data are well-defined sources of discrepanc ies in statistical evaluation of clinical trials. In this study, the a uthors show the influence of errors on the widely used statistical res ult, P values. Methods. Using data from a retrospective study of patie nts with Hodgkin disease treated at the University of Minnesota betwee n 1970 and 1984 and observed to 1988, we introduced various errors int o the data to study the impact on results. Results. Inadvertent random and systemic errors affect statistical results. Data entry and transc ription errors, vague definitions of endpoints and prognostic factors, and the omission and selection of patients are examples of frequent e rrors that affect statistical evaluation. Conclusion. The results and inferences of many studies are sensitive to systemic errors and data m anipulation. Great care must be given to the clear definitions of term s, exclusion and inclusion criteria, group assignments, treatment prot ocols, and the subgroups on which statistical analysis is performed. C linicians and statisticians must work together to improve the performa nce and interpretation of clinical trials.