Department of Global Public Health and Primary Care

Medical statistics

Medical statistics involves using statistical modeling, estimation methods, and inferential theory to draw conclusions from medical and health-related datasets.

Main content

Medical statistics involves utilizing statistical modeling, estimation methods, and inferential theory to draw conclusions from medical and health-related datasets. Key statistical methods are founded on mathematical modeling, data simulations, and statistical learning and prediction models. Some of the most common examples include experimental design with sample size calculation, descriptive and graphical methods for presentation, model selection and fitting, regression modeling, estimation and calculation of significance probabilities and confidence intervals, machine adaptation of prediction models, as well as meta-analyses for summarizing already published results. Typical regression models include linear, logistic, and Cox models for lifetimes. The dataset can originate from observational studies, randomized clinical trials, or laboratory data.

These methods are absolutely crucial in almost all forms of medical and health-related research. To provide evidence-based treatment, it's essential to critically evaluate the literature, and a strong statistical understanding is fundamental for such critical assessment. Our courses are offered at all levels from basic to doctoral. In cases where advanced calculations are required for teaching, we will, with few exceptions, use Stata. Stata has robust procedures for most methods in modern statistics. Additionally, its syntax is easy to both write and understand, and its graphical capabilities are of high quality.

About the Statistics Group The group consists of three full-time employees in permanent scientific positions, one permanent employee in a 50% position, two employees with 20% positions, and any PhD students and postdoctoral researchers. In addition, one permanent position from the Institute of Clinical Dentistry is part of the group, and the core facility BIOS contributes instructors. The members are primarily involved in research within medicine, epidemiology, and biostatistics, engaging in both application and development of statistical methods.

Areas of Work The group is responsible for the majority of statistics education at the Faculty of Medicine and should continually evaluate the course content, discuss and comment on topics related to the discipline, mentor students, and provide statistical guidance services (https://www.uib.no/fg/biostatistikk/123416/veiledningstjeneste-medisinsk-statistikk-v%C3%A5r-2019). A core facility for biostatistics and data analysis (BIOS: https://www.uib.no/fg/biostatistikk) is affiliated with the academic community, providing statistical project support for small and large projects.

Professional Development, Collaboration, and Network The teaching group is one of two teaching groups within the field of EPISTAT at the IGS, and both in research and teaching, there is close collaboration with the Epidemiological Teaching Group. The group members belong to the EPISTAT domain at IGS, and in both research and teaching, there is close collaboration with epidemiologists, who constitute the other major group within EPISTAT.

Every Wednesday, lunch seminars are organized, and each fall, a larger meeting is held with both internal and external speakers. Nationally, doctoral candidates affiliated with the group often participate in research schools such as EPINOR or NORBIS.