WK80: Clinical Prediction Models (ONLINE COURSE)
(If there is [full] to a course, please do sign up, but you will be placed on a waiting list. Once there is an open spot we will contact you. At that point you can decide whether to participate in the course.)
Because of the Corona virus, we have decided to offer the course online in January 2021. This is because there is a lot of interest from foreign students for this course and it is almost impossible to find a suitable space where sufficient distance can be kept.
|6, 7, 8 January 2021 (online course)||Tuition fee: € 750,-
Course description and topics
The purpose of a prediction model is to estimate the chance of a particular outcome as accurately as possible (prediction). Prediction models are often developed with clinical practice in mind, and involve combining information about patients to calculate an individual’s chances of illness or recovery. The model can then be presented in the form of a clinical predictive rule. General applicability – i.e. the accuracy of the prediction model when applied to new patients in the future – is another very important aspect.
The problems which can occur when developing prediction models include the difficulty of selecting the most important predictors from a large number of variables. If this is not done carefully, the quality of the prediction model can be adversely affected. Also, the prediction model will often need to be adjusted before it can be applied to new patients. All these issues are frequently overlooked or underestimated by clinicians and researchers.
The aim of the course is to provide better knowledge and understanding of the development of prediction models that are relevant to real-life practice. We will focus on the various methods for selecting variables, and the pros and cons of these different methods. Once the prediction model has been developed, it is important to assess the quality of the prediction model. For example, we will look at whether the predictions of the model are accurate and during the course, we will also consider the various ways of measuring accuracy. The question of applying the model to new (future) patients will also be addressed. An important element of this is investigating whether the performance of the prediction model deteriorates when it is applied to new patients. This component is entitled the validation of the prediction model. We will also look at various techniques for validating the prediction model.
The course consists of an intensive programme of partly interactive lectures, combined with computer-based practical work. Examples taken from clinical practice will be used for the computer-based work.
Martijn W. Heymans, PhD , coursecoordinator
Department of Epidemiology & Biostatistics. Amsterdam UMC, location VUmc
Dr. Martijn Heymans expertise is in prognostic and prediction modeling, missing data and longitudinal data analysis. He (co)-authored more than 80 scientific publications and also teaches courses in epidemiology, applied biostatistics and regression techniques and works as a statistical consultant.
The development and quality of prediction models, including:
- the characteristics of a prediction model
- the most frequently used methods for selecting variables
- the pros and cons of the various methods of selecting variables
- the different measures of quality and how to interpret them (including explained variation calibratie, discriminatie, Roc curve)
- investigating the relevance of a prediction model for real-life practice.
- Introductie R software en Frank Harrell’s rms package.
- Introduction to the Validation of prediction models
- The linear predictor (lp)
- Optimism and shrinkage
- Adjusting the intercept
- The internal and external validation of prediction models
- Internal validation (Bootstrapping techniques)
- Adjusting the slope
- External validation
- Generalizability of prediction model (Case-mix, different regression coefficients)
- Presentation formats of prediction models
- Updating of Prediction models
- Reasons for generalizability problems
- Updating the intercept and slope
- Comparing Prediction models
- Adding a new variable
- Reclassification tables
- A prediction model for survival data
- 1. The participant can recognize and identify the characteristics of a prediction model.
- 2. The participant can identify the weak points and strong points of the most commonly used methods for selecting variables.
- 3. The participant can analyse and interpret the methods that are used to determine the quality of a prediction model (including tools for discrimination such as the ROC curve, and for calibration such as the Hosmer and Lemeshow test and a calibration curve).
- 4. The participant can analyse and interpret the methods that are used to determine the value of a prediction model for real-life practice (e.g. sensitivity, specificity, positive and negative predictive abilities).
- 5. The participant can convert a prediction model into a practically useful clinical instrument.
- 6. The participant is familiar with the principles that play a role in internal validation such as over-fitting, optimism and shrinkage.
- 7. The participant can analyse and interpret the methods used in the internal validation of prediction models, such as cross-validation and boot strapping techniques.
- The participant can use methods to update the intercept and slope of the prediction The participant is able to study the added value of a new predictor variable by using reclassification tables by making use of SPSS and R software.
The participant is able to develop a prediction model and to study the quality of the model in survival data by using a Cox regression model.
11.The participant can develop prediction models, assess their quality and validate them (internally and externally) using SPSS and R software.
All course material (lectures, assignments, feedback of the assignments and computer exercises, any additional literature, any additional teaching material and information about the exam) for this course will be placed on Canvas. Canvas is EpidM's digital learning environment.
To complete the practical for this course, you must have a laptop / computer with SPSS and R. R is free to download. You will receive instructions in good time when you register for the course.
A week before the start of the course you will receive information about creating a Canvas account for this course.
Target group and pre-requisites
Target groupThe course is designed for PhD-students, practitioners and applied researchers working in the field of epidemiology, medicine, public health, psychology, human movement sciences.
The following concepts are assumed known by participants at the start of this course:
- Knowledge of basic statistical tests as t-tests and regression analyses.
- Knowledge of some basic SPSS commands.
Knowledge of R(Studio) is not a prerequisite.
Exam and accreditation
A declaration of participation is issued if the course has been followed entirety. In special cases, the course coordinator can, after prior consultation and for a valid reason, decide to issue a certificate in case of a small absence (max. 20%).
Participants who take this course as part of the Master Epidemiology always complete the course with an exam. Other participants can choose if they want to complete the course with an exam. The costs in this case are 150, - per examination or re-examination.
The exam will be in English. Only when you pass the exam you get a certificate showing the credits (study points/EC).
The examination dates can be found on the website of EpidM.
Anyone who wants to participate in the examination should apply at least four weeks before the exam to register via the website: tentamens
The examination material of reference and questions to practice can be found on the Canvas page of the course (see above).
During the examinations of EpidM the use of e-books is forbidden.
Only for Dutch students!
If you wish to be considered for accreditation points connected to this course, you must sign the attendance list on the last day of the course.
To qualify for the accreditation points, you must have been present throughout the course.