EES:Methods, S:Methods, GPH:Methods
Statistics and Mathematics.
Use of formal statistical procedures for the evaluation of evidence in biomedical research or in decision making is becoming ever more important. Researchers or policy makers who do not have basic biostatistics skills are at a serious disadvantage – to either use or interpret existing study results for their own purposes – or when setting up their own studies. For example, methods such as linear and logistic regression, are now ubiquitous in the scientific medical literature. A complete understanding of results presented in journals or scientific reports will often require basic proficiency in the statistical methods used. Evidence-based approaches to medicine or management of public health services typically involve statistical approaches. Funding agencies demand justification of statistical design and analysis methods when setting up studies.
The purpose of this course is to acquaint you with some of the basic statistical methods and concepts used in either research or management for medical practice and public health. We will focus on concepts and the reasons for the methods proposed. Methods will be illustrated and introduced from practical motivating examples. Emphasis will be on the motivation for the methods discussed and the interpretation of summary statistics that may be derived from them. Applicability of methodology for specific practical studies will be reviewed. We will also critically evaluate suitability of methodology for specific study designs and identify possible misuses or incorrect interpretations of statistical measures. You will learn to apply these methods for the analysis of real data in biomedical studies examples using statistical software. You will learn to identify appropriate methodology for specific practical applications and formulate conclusions based on evaluation of results generated from practical data analysis using the methods discussed in the course.
After this course you will:
Know the different study designs and its pros and cons.
Be able to apply T-tests, ANOVAs and Chi-square tests
Know when and how to apply relative risks and odds ratios
Be able to interpret and apply multivariable linear and logistic regression analysis.
Know when and how to apply prediction models
Understand the concept of power and type I and II errors
Once available, timetables will be published here.
Mode of instruction
The course will consist of lectures and practicals (computer sessions) to exercise the concepts and methods.
In the third week a short questionnaire will be given to assess you understand the main concepts (20%).
In the middle of the course you get a practical. Your written analysis report will be graded (40%)
The course will finish with an open book exam consisting of open questions (40%).
There will be a Blackboard site available for this course. Students will be enrolled at least one week before the start of classes.
For main course book we will use:
- Regression Methods in Biostatistics: Linear, Logistic, Survival, and Repeated Measures Models (Statistics for Biology and Health) (Springer) 2012 Vittinghoff, Glidden, Shiboski and McCulloch
The following are optional additional reading materials for people with an interest in medical statistics:
Essentials of Medical Statistics, by Betty Kirkwood [this book is really a classic]
Regression Models as a Tool in Medical Research, by Werner Vach [Provides much more detail on issues concerned with practical application of regression methods than many other texts do, so well worth the read – but it focuses on STATA as the statistical package.)
Statistical methods in cancer research. Volume 1 – The analysis of case-control studies. WHO. International agency for research on cancer. 1980.
Statistical methods in cancer research. Volume 2 – The design and analysis of cohort studies. WHO. International agency for research on cancer. 1987. [The above two volumes are in some sense the bible of basic biostatistics and epidemiology in applied biomedical research. Written on behalf of one of the leading international agencies on medical research.]
Clinical trials. A practical approach. Stuart J. Pocock. [Bit older and more introductory, but that could be an advantage.]
Statistical issues in drug development. Stephen Senn [probably a must-read on the topic.]
This course is open to LUC students and LUC exchange students. Registration is coordinated by the Curriculum Coordinator. Interested non-LUC students should contact email@example.com.
Bart J. A. Mertens