PhD defence A.J.L.M. (Sandra) Geraerds

Dissertation in short:

Health care spending comprises approximately 9.9% of the general domestic product of European countries. Due to scarcity, health care spending cannot rise endlessly, and
therefore choices have to be made in allocating health care resources. Policy decisions with respect to reimbursement of health care can be informed by economic evaluations,
taking both costs and effects of health care into account. The effect of health care is often measured with a generic health-related quality of life instrument (HRQoL), such as the
EQ-5D. However, there are some concerns with respect to the sensitivity of the EQ-5D for changes in health. Possibly, the five dimensions that are used in the EQ-5D are too limited
to measure all change in health. Adding an additional dimension, a so-called ‘bolt-on’, could improve the sensitivity of the EQ-5D in the general population and specific patient
populations.

Apart from health outcomes, insight in costs of illness is also crucial in making reimbursement decisions. Costs of illness or interventions can be studied relative to the associated health outcomes, or on a more descriptive base to gain insight in costly groups in a patient population. Costs relative to the associated health outcomes are often studied using a cost-utility analysis, in which the health outcomes are expressed in utilities, calculated using the outcome of for example the EQ-5D. In case of no difference in health outcomes between two groups, a cost-minimization analysis, including a cost comparison only, is sufficient.

The main aim of this thesis was to gain more insight in the methods to assess HRQoL and methods to perform cost studies, and to apply these methods in different patient populations. This thesis consists of eleven studies, divided in three parts that each addressed a research question: 

1. What is the current state of practice in measuring HRQoL in patient populations and which measurement traits are important to control for?
2. Does the addition of dimensions to the EQ-5D improve the psychometric performance in different patient populations, and how can we determine the added value of additional dimensions?
3. What are the health care (and productivity) costs of different types of patient populations, and which determinants affect the amount of costs?