Utilities Estimated from PROMIS Scales for Cost-Effectiveness Analyses in Stroke

Area of Science:

• Health economics
• Neurology
• Outcomes research

Background:

• Established preference-based measures like EQ-5D and HUI-3 are crucial for cost-effectiveness analyses.
• The PROMIS Preference (PROPr) scoring system is a novel preference-based measure.
• Algorithms exist to map PROMIS Global Health (PROMIS-GH) to HUI-3 (HUI_LE) and EQ-5D (EQ5D_LE).

Purpose of the Study:

• To evaluate and compare health utilities derived from PROPr and PROMIS-GH in adult stroke survivors.
• To assess the distributional characteristics and correlations with stroke outcomes of different utility measures.

Main Methods:

• Retrospective cohort study of 4,159 adult stroke survivors (2015-2019).
• Calculation of a modified PROPr (mPROPr).
• Comparison of distributional characteristics and correlations with stroke outcomes for mPROPr, HUI_LE, and EQ5D_LE.

Main Results:

• Mean utility estimates varied significantly: mPROPr (0.333), EQ5D_LE (0.739), and HUI_LE (0.544).
• Correlations with the modified Rankin Scale were -0.48 for mPROPr, -0.43 for EQ5D_LE, and -0.48 for HUI_LE.
• Regression analyses suggested potential biases in mPROPr (too low for well patients) and EQ5D_LE (too high for poorly patients).

Conclusions:

• All PROMIS-based utilities correlated with stroke severity but showed distinct distributions.
• Significant differences highlight challenges in valuing health states for cost-effectiveness research.
• Mapping PROMIS-GH to HUI-3 via linear equating (HUI_LE) appears most suitable for stroke patients.