How Does Patients' Overall Assessment of Their Health Vary Across and Within Different Disease Groups?

Overview

EQ-5D is one of the most commonly employed patient-reported outcome (PRO) measures. It is included in many of the Swedish National Quality Registers (NQRs). EQ-5D health states are usually summarized using 'values' obtained from healthy members of the general public. However an alternative - which remains to be studied in detail - is the potential to use patients' self-reported overall health on the visual analogue scale as a means of capturing experience-based values. The overall aim of this project is to increase knowledge on the potential applicability of EQ VAS as a health state valuation method through assessment of its variability across and within patient groups and compared with that of the general population in Sweden. Data on nearly 700,000 patients from 12 NQRs covering a variety of diseases/conditions and from the general population will be analysed. Longitudinal studies of PROs among different patient groups will be conducted at baseline/first visit and 1-year follow-up. Descriptive analyses comparing EQ-5D health states and observed self-assessed EQ VAS within and across registers will be performed. Comparisons of the change in health state and observed EQ VAS values over one year will also be made. Regression models will be used to assess whether EQ-5D dimensions predict observed EQ VAS values to investigate patient value sets in each NQR. These will be compared across the patient groups and with the existing Swedish experience-based VAS and time trade-off (TTO) value sets obtained from the general population. This research project will provide information on the variation among different patient groups in terms of self-reported health status through EQ VAS and comparison with the general population. Knowledge on the relative importance of different dimensions of the EQ-5D to different patient groups as well as the general population will be gained in this project. The possibility of getting value sets based on patients' self-reported EQ VAS values and their comparison with value sets from experience-based general population studies will be discussed.

Research questions: 1. How do EQ-5D health states and self-assessed EQ VAS values vary across and within patient groups, and over follow-up stages, and in comparison to the general population data? 2. To what extent do EQ-5D health states predict EQ VAS values, and how do the resulting experience-based patient value sets differ when estimated from patients' data at baseline and 1-year follow-up and how do the EQ VAS values predicted for EQ-5D health states differ between different patient groups? 3. How do these patient value sets modelled using data from the registers compare with the Swedish VAS and TTO experience-based EQ-5D value sets obtained from the general population? 4. How do value sets for EQ-5D-3L, predicted from EQ VAS differ from value sets predicted for the EQ-5D-5L, predicted from its EQ VAS? Data: In this project 12 National Quality Registries holding records of PROs, on the EQ-5D instrument, records of approximately 700,000 patients will be included. Clinical data (age, sex, BMI, diagnosis (es) and interventions) and PROs data (EQ-5D-3L and condition-specific) will be retrieved from the registries. Data from cross-sectional population surveys in Sweden will be included for comparison; a total of approximately 45,000 records used in developing the Swedish TTO and VAS value sets. The quality registries to be part of the study include; 1. Better management of patients with osteoarthritis (BOA) 2. The Swedish Ankle Registry (Swedankle) 3. The Swedish Fracture Register (SFR) 4. The Swedish Heart Failure Registry (SwedHF) 5. The Swedish Hip Arthroplasty Register (SHAR) 6. The Swedish Knee Arthroplasty Register (SKAR) 7. The Swedish National Anterior Cruciate Ligament Register (XBase) 8. The Swedish National Quality Register for Bipolar Disorder (BipoläR) 9. The Swedish National Registry for Respiratory Failure (Swedevox) 10. The Swedish Registry for Systematic Psoriasis Treatment (PsoReg) 11. The Swedish Rheumatology Quality Register (SRQ) 12. The Swedish Spine Register (Swespine) Data analysis: Records of patients in the registries with baseline and follow-up data on PROs will be included for the different analyses to be conducted. The analyses will focus on the three main data components coming from the EQ-5D instrument. Data collected by the EQ-5D descriptive system on the five dimensions; the EQ-VAS score and EQ-5D index resulting from transforming the EQ-5D health profile into a single index using value sets. Hence, analyses looking into how the EQ-5D profile varies within and across patient groups and how it varies over time will form one component of the analyses to be conducted. Similarly, the pattern of EQ-VAS scores across patient groups and follow-ups will be analysed. Similar comparisons of EQ-5D index within and across patient groups, based on current Swedish value sets and patient VAS value sets to be elicited from the dataset, will be performed.