Condition Specific Applications for the Valuation of Health

The research included in this dissertation focused on exploiting different approaches to apply and to improve the use of generic health-related quality of life (HRQL) measures in disease- or condition-specific situations. Four separate projects were undertaken to evaluate the measurement properties, strengths and limitations of both generic and disease-specific measures of HRQL, and to estimate patient (experience)-based value sets for a health classifier system. The first of study evaluated the extent of unique content covered by disease-specific preference-based measures when contrasted with the EQ-5D, and identified potential dimensional extensions that may be justified to be added to EQ-5D for various disease-specific applications. The second study examined the properties of several recently developed generic HRQL measures, the Patient Reported Outcomes Measurement Information System 43-item short form (PROMIS-43) and five-level EQ-5D (EQ-5D-5L), in patients with chronic obstructive pulmonary disease (COPD), and found evidence to support their validity and reliability. The third study utilized health systems level data to examine the relationship between EQ-5D index and EQ-5D Visual Analogue Scale (EQ-VAS), and Oxford Knee Score (OKS) in patients undergoing knee replacement. We found that EQ-VAS systematically differs from the OKS and EQ-index as an indicator of surgical outcome, but captures unique but valid information about how patients feel about their health. The final study derived patient-based EQ-5D value sets using VAS from patients who experienced knee replacement surgery. Both before and after surgery, larger preference weights were assigned to anxiety/depression and self-care. Greater value decrements were given for health problems after knee replacement compared to before surgery. This time-related shift in values could be a consequence of a gap between experience and expectations of surgery and requires further research, particularly if patient-based value sets are used to inform medical decision making about surgery.