Health-related quality of life of patients with a recent fracture attending a fracture liaison service: a 3-year follow-up study

Abstract

This study explored the course of health state utility value over 3 years in patients with a recent fracture attending a Fracture Liaison Service and suggested that the overall change in health-related quality of life was not significant, although significant improvements were observed at 6 and 12 months compared to baseline.

Introduction: To estimate the 3-year health-related quality of life (HRQoL) of patients with a recent fracture presenting at a Fracture Liaison Service (FLS) and to explore factors associated with health state utility value (HSUV).

Methods: Patients' HSUVs were derived from the EQ-5D-5L and SF-6D and calculated at six time points. Multiple imputation was applied for missing data. Linear mixed-effects regression analysis with random intercept and slope was applied to explore the course of HSUV over 3 years. The impact of subsequent fracture and the length of time between FLS visit and patients' index fracture on HSUV were also investigated. A backward stepwise elimination was applied to identify factors associated with HSUV.

Results: A total of 499 patients were included. The change of EQ-5D HSUV was not significant over 3-year follow-up (P = 0.52), although slightly but significantly higher HSUV was captured at 6 months (mean difference (MD): 0.015, P = 0.02) and 12 months (MD: 0.018, P = 0.01). There was no significant difference in the course of EQ-5D HSUV between fracture locations (P = 0.86). A significant increase in HSUV was only captured for patients had shorter time period (< 107 days) between FLS visit and their index fracture. Suffering a subsequent fracture was associated with significant QoL loss (MD: - 0.078, P < 0.001). Subsequent fracture, previous treatment with anti-osteoporosis medication, a prevalent vertebral fracture (grade 2 or 3), use of a walking aid, previous falls, and higher BMI were negatively associated with mean EQ-5D HSUV over 3 years. Comparable results were found using SF-6D HSUV. The lack of HRQoL data immediately after fracture and selection bias were two main limitations.

Conclusion: The 3-year change in HSUV was not statistically significant, although significant improvements were observed at 6 and 12 months in comparison with baseline. Six factors were negatively associated with EQ-5D HSUV.

Keywords: EQ-5D-5L; Health utility; Longitudinal analysis; SF-6D.