Page 41 - Contributed Paper Session (CPS) - Volume 1

CPS694 Ordak Michal
4. Discussion and Conclusion
Summing up the research carried out between 2006 and 2018, I can say that
there are significant reasons for concern. Every year, more and more errors are
observed in medical statistics. This is often connected with the lack of statistical
knowledge among medical researchers and the tendency to break statistical
rules in order to obtain a specific result. Taking account of the factors described
in this article will significantly increase the value of published medical research
results in the future. For example, the percentage of publications presenting
ambiguous results on particular topics will decrease. The purpose of this article
is to make the community around the world aware of the most common errors
made in medical statistics. These errors can often affect the most precious gift,
which is our life. The statistical consultations carried out between 2006 and 2018
allowed me to conclude why the statistical knowledge of medical students is
insufficient. This translates into many negative aspects in the future, such as the
lack of statistical knowledge among medical reviewers, problems encountered
by e.g. doctoral students and physicians when performing statistical analyses, as
well as many other problems. It seems advisable to implement the
recommendations described here in the field of medical statistics education.
Consequently, the percentage of medical employees for whom statistical analysis
will not be something completely new and who will be able to use their own
knowledge in practice will increase in the future.

References
1. Altman, D. G., Goodman, S.N. & Schroter, S. (2002). How statistical
expertise is used in medical research. JAMA. 287(21):2817-20.
2. Beckerman, P., Bi-Karchin, J., Park, A. S., Qiu, C., Dummer P. D., Soomro, I. et
al. (2017). Transgenic expression of human APOL1 risk variants in
podocytes induces kidney disease in mice. Nat. Med. 23(4): 429-38.
3. Fagerland, M. W. (2012). T-tests, non-parametric tests, and large
studies—a paradox of statistical practice? BMC. Med. Res.
Methodol.12:78.
4. Galton, F. (1986). Natural inheritance; London:Macmillan.
5. McHugh. M. L. (2011). Multiple comparison analysis testing in ANOVA.
Biochem. Med. 21:203-9.
6. Ozonoff, D. (2006). Statistics in peer review. Nature’s peer review debate.
Nature.
7. Petrovečki, M. (2009). The role of statistical reviewer in biomedical scientific
journal. Biochem. Med. 19(3):223-30.heir insufficient knowledge of
statistical analysis results from four factors, namely:
8. Zelen, M. (2006). Biostatisticians, Biostatistical Science and the Future.
Stat. Med. 25:3409-14.

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