gms | German Medical Science

Introduction: Medical biometry is part of the cross-sectional subject Q1, composed of the subjects epidemiology, medical informatics, and medical biometry, of the curriculum of the study programme human medicine in Germany. A basic understanding of the central aspects of study planning, conduction, and analysis as well as interpretation of the findings enables physicians to independently assess the value of novel research and its contentual relevance. This is eventually a basic requirement to be compliant with the directive of good scientific practice.

Objectives: Despite the apparent relevance of the subject for medical research, there is still the potential to improve the reputation of medical biometry among the students. Thus, it is all the more important to demonstrate the practical relevance of medical statistics by means of appropriate examples which are ideally flanked with activating teaching methods. Therefore, the "NaNa"-example study has been developed and will be applied in the course of the PC-based seminar Medical Biometry in the present winter term.

Material and methods: The two-arm NaNa study will include 4 groups of students, each of a maximum size of 24. Seminar contents (exercises and short exams) are the same for all groups, though two of them will be taught as usual using real observational data from a finalized study in pediatrics and two groups will work with self-assessed observational data regarding their habits in consuming snacks (either candy (called "Naschkatzen" in Germany) or salty snacks (called "Nagetiere" in Germany)). Primary endpoints to evaluate the effect of working with self-assessed research data are the cumulative score over all short exams and an evaluation score.

Results: The results of the NaNa study will be available in January 2017 and can be presented during the RIME conference in March. Data will be analyzed by means of descriptive as well as inferential statistical methods. Specifically, it could be expected that the cumulative exam score will be normally distributed which is why an unjoined t-test will be used to compare the interventional groups. Since the evaluation score is ordinally scaled, a non-parametric Mann-Whitney-U test will be used for this endpoint. Moreover, the authors will share their experience on general aspects in conducting an interventional study in educational research (e.g. willingness of students to participate, etc.)

Conclusion: This educational research study will evaluate the potential of further improvements in a basic course of medial statistics among students of human medicine. In the past a PC-based seminar using statistical software for the analysis of research data has been developed in order to enable students to learn the principles of medical statistics more practically. However, there are still many students who are not completely convinced about the practical relevance for their own work. Self-assessment of research data which are processed subsequently might improve the acceptance among students, leading in consequence to an increase in examination success.