Random error units, extension of a novel method to express random error in epidemiological studies

Imre Janszky,1 Johan Håkon Bjørngaard,1 Pål Romundstad,1 Lars Vatten,1 Nicola Orsini2 1Deparment of Public Health, Faculty of Medicine and Health, Norwegian University of Science and Technology, Trondheim, Norway; 2Department of Public Health Sciences, Karolinska Insitutet, Stockholm, Sweden Abstract: Currently used methods to express random error are often misinterpreted and consequently misused by biomedical researchers. Previously we proposed a simple approach to quantify the amount of random error in epidemiological studies using OR for binary exposures. Expressing random error with the number of random error units (REU) does not require solid background in statistics for a proper interpretation and cannot be misused for making oversimplistic interpretations relying on statistical significance. We now expand the use of REU to the most common measures of associations in epidemiology and to continuous variables, and we have developed a Stata program, which greatly facilitates the calculation of REU. Keywords: statistical significance, confidence intervals, P Value, random error, random error units