Using data from routine care

Course goals

Students will

• understand why it is difficult to obtain consent or anonymize (big) data in health research.
• learn to apply privacy by design strategies (such as pseudonymization) to a case.
• understand how ethical principles govern the use of Big Data in health research.
• learn to apply ethical principles to an exemplary case.
• learn how health data is captured in different health care data sources and how data can be extracted, and quality checked prior to specific research questions.
• learn about data modeling and how health data can be transformed into evidence.
• learn to define which information is required to be able to determine a specific measurement of the effectiveness of an intervention in health care or public health.
• learn which methods to use to answer causal questions about the effect of intervention on observational big data.
• be able to explain the fundamental ideas underlying linear and non-linear machine learning for classification, probabilistic prediction and survival analysis.
• be able to state several machine learning methods, such as kNN, penalized regression.
• be able to apply machine learning methods to a dataset.
• be able to evaluate whether the machine learning method is suitable to answer the research question.
• understand the bias-variance tradeoff and overfitting in high dimensions.
• be able to evaluate the predictive performance and usability of a machine learning method.

Assessment

• one group assignment during the whole of the course that covers all topics discussed per week. Counts for 70% of the final grade.
• one individual assignment, to be handed in at the end of week 10 of the course. Counting for 30% of the final grade.

The average weighted grade will be your final score for this course and the one entered into Osiris.

To qualify for a repair of the final result the mark needs to be at least a 4, or “AANV”.

Content

In this course, we will look into three main topics:

• legal and ethical considerations of gathering and using data from health registries and routine care.
• data management of data from health registries and routine care.
• data analysis of data from health registries and routine care.

Nowadays, the way the data from multiple registries are made accessible, may be constrained for legal (e.g. privacy) and ethical reasons.
We see that typically, data from registries are used to answer questions such as the effectiveness or occurrence of side effects of drugs, e.g. COVID vaccines.
These are causal questions, because we want to ascribe effectiveness or side effects to the drug. Because the prescription of drugs for a patient is based on indication by the physician, rather than randomization in an experimental study, it is important to learn how to draw causal conclusions from these observational data.

Data from routine care pose another problem to the analysis of data, depending on the purpose:
- researchers may want to make a classifier or prediction model for the health outcome of patients, that can be used to inform patients, select patients for a certain treatment, or identify patients who are at risk for a bad outcome, to refer them to more advanced care. 
- researchers may want use EHRs (Electronic Health Records) to compare patients who did receive a certain treatment with patients who received no treatment or another treatment.

To use data from routine care, you first have to know about the legal and ethical considerations.
We will then look into how to extract data from registries and manage the acquired data.
Next, we will see how we can draw conclusions on effectiveness based on the causality of the data using some practical methods that are applicable to both registry and routine care data (EHRs).
And lastly, what types of analysis should be done on EHRs, depending on the purpose.

Course form
Lectures, exercises, assignments.

Literature
Chapters and articles, listed in the course manual.

Additional information

Enrolment in this course is only available by using the pre-enrolment form that will be forwarded to all ADSM students in the first week of November.