Page 35 - Contributed Paper Session (CPS) - Volume 1

CPS694 Ordak Michal
the medical research community with my research results that are based on 13
years of experience in conducting statistical analyses and reviews in everyday
practice. My intention is to make the global medical community aware of why
there are reasons for concern when it comes to the direction in which medical
statistical analysis is going.
"Some people hate the very name of statistics but I find them full of beauty
and interest. Whenever they are not brutalised, but delicately handled by the
higher methods, and are warily interpreted, their power of dealing with
complicated phenomena is extraordinary. They are the only tools by which an
opening can be cut through the formidable thicket of difficulties that bars the
path of those who pursue the Science of man" (Galton, 1986). I often say these
words to students of various medical fields to introduce them to the
interesting world of medical statistics. Statistics plays a very important role at
every stage of scientific biomedical research. Data analysis is used not only by
scientists and physicians directly involved in clinical trials, but also by medical
employees who keep up to date with the results of new research and want to
interpret their results correctly (Zelen, 2006). Unfortunately, nowadays
knowledge of basic statistical methods is negligible in the medical
environment. Medical students unfortunately do not have full awareness of
the importance and necessity of applying appropriate statistical methods in
their research. Consequently, students are often unable to perform statistical
analysis for publications, dissertations, etc. (Altman et al., 2002). Medical
reviewers therefore assess submitted publications primarily in terms of
innovation and very often do not evaluate the correctness of statistical analysis
(Ozonoff, 2006; Petrovečki, 2009). As a long-time expert in statistics, I must
unfortunately state that a large number of medical reviewers accept submitted
works even in very high-ranked magazines despite significant statistical errors.

2. Methodology
I examined a group of 14,000 people working in the medical environment,
including doctors and researchers. The article presents the basic mistakes made
by the medical community in the performance of statistical analysis. For this
purpose, I analysed additional 30 publications from 50 medical journals. During
the last thirteen years, i.e. between 2006 and 2018, I carried out 25,300
statistical consultations in a group of students of various medical majors.

3. Results
Repeated testing of means is a growing problem today. Increasingly, more
medical researchers use t-tests for research schemes with one independent
variable on more than two levels. Making multiple comparisons by testing the
difference between two means within one dependent variable and a multilevel
independent variable leads to false conclusions. This phenomenon is called type

24 | I S I W S C 2 0 1 9