Longitudinal Changes in Health-Related Quality of Life for Chronic Diseases: an Example from the Hemophilia Utilization Group Study Part Va (hugs Va)
J. L. Poon,J. Doctor,F. Gwadry-Sridhar,M. Ullman,B. Riske,J. Baker,X. Niu,M. Lou,M. B. Nichol
DOI: https://doi.org/10.1016/j.jval.2013.03.590
IF: 5.156
2013-01-01
Value in Health
Abstract:In chronic diseases, health-related quality of life (HRQoL), a key health outcome measure, varies not just between- but also within-individual patients over time due to acute changes in clinical or environmental circumstances. Cross-sectional assessments of HRQoL consider only between-patient differences and do not reflect within-patient changes, which require multiple assessments over time. This study aims to disaggregate the between- and within-individual factors influencing HRQoL over time in persons with hemophilia A. In a two-year observational prospective study of health care utilization and burden of illness in US hemophilia A patients (HUGS Va), the SF-12 (adults) and PedsQL (children aged <18 years) was used to assess HRQoL at baseline, 6, 12, 18 and 24 months. Information was also collected about disease severity, sociodemographic variables, annual bleeding and emergency room visits frequencies and missed days from work/school. Only small variations in HRQoL were observed over time in the entire study population of 157 adults and 164 children, of which 64.6% have severe disease. At baseline, 6, 12, 18 and 24 months, adult mean mental component scores were 50.9±10.1, 49.8±9.7, 50.2±10.2, 49.5±10.6 and 50.4±9.5 respectively, while mean physical component scores were 43.4±10.7, 43.3±10.8, 50.2±10.2, 49.5±10.6 and 50.4±9.5 respectively. Among children, mean psychosocial functioning scores at each follow-up were 84.1±15.3, 83.3±13.3, 80.3±15.6, 83.5±16.0 and 82.9±13.2 respectively, while mean physical functioning scores were 89.5±15.2, 85.5±17.8, 82.7±22.3, 85.7±18.0 and 84.7±17.3 respectively. After plotting the HRQoL scores of individuals over time, however, greater variations in the magnitude and pattern of fluctuation can be observed, reflecting acute within-patient changes. HRQoL fluctuates over time on the individual level, but not at the study population level as aggregation of scores to the mean masks individual variation. In order to identify factors influencing changes in HRQoL across time, multivariate multilevel modelling will be employed to disaggregate between- and within-individual differences in HRQoL.