UTILITY AND DISUTILITY VALUES FOR STROKE, MYOCARDIAL INFARCTION AND RELATED CARDIOVASCULAR HEALTH STATES: A SYSTEMATIC LITERATURE REVIEW
Author(s)
Beata Smela, PhD1, Beata Lukiewicz, MSc1, Kinga Kuzdak, MSc1, Aurélie Millier, PhD2, Luke Bamber, MSc3.
1Clever-Access, Krakow, Poland, 2Clever-Access, Paris, France, 3Bayer AG, Wuppertal, Germany.
1Clever-Access, Krakow, Poland, 2Clever-Access, Paris, France, 3Bayer AG, Wuppertal, Germany.
OBJECTIVES: To identify and synthesise utility and disutility values for key health states and events relevant to health economic modelling of secondary stroke prevention, including non-disabling stroke (NDS), disabling stroke (DS), modified Rankin Scale (mRS 0-5), transient ischaemic attack (TIA), myocardial infarction (MI), bleeding events and dementia.
METHODS: A systematic literature review (SLR) was conducted in MEDLINE, Embase (January 2025) and the International Health Technology Assessment Database, supplemented by targeted searches of health technology assessment agency websites. A complementary rapid review (RR) of MEDLINE covering the previous 5 years captured recent primary studies. The SLR focused on economic models, with a linking analysis to identify underlying primary utility sources. Dual independent screening and quality assessment were performed in line with the NICE reference case.
RESULTS: A total of 44 primary studies and 20 economic models were included. Utility and disutility estimates were identified across all prespecified health states. EQ‑5D was the predominant elicitation method (33/44 studies). Mean utility values from primary studies ranged from 0.64-0.86 for NDS, 0.12-0.66 for DS, 0.82-0.89 for TIA and 0.35-0.99 for MI. Utilities declined consistently with increasing disability by mRS (0: 0.89-1.00; 3: 0.37-0.70; 5: −0.19 to 0.38). Estimates for other events included major bleeding (0.67-0.84), minor bleeding (0.68-0.69) and dementia (0.19-1.00). Economic models showed broadly consistent patterns but tended to report lower utility values for stroke (NDS: 0.40-0.79; DS: 0.22-0.45). Substantial heterogeneity was observed across studies, driven by differences in populations, study design and country-specific EQ‑5D value sets.
CONCLUSIONS: This review provides a comprehensive and structured evidence base of utility and disutility inputs for secondary stroke prevention modelling. While consistent patterns are observed across sources, substantial heterogeneity highlights the need for transparent selection, justification and sensitivity testing of utility inputs in economic evaluations.
METHODS: A systematic literature review (SLR) was conducted in MEDLINE, Embase (January 2025) and the International Health Technology Assessment Database, supplemented by targeted searches of health technology assessment agency websites. A complementary rapid review (RR) of MEDLINE covering the previous 5 years captured recent primary studies. The SLR focused on economic models, with a linking analysis to identify underlying primary utility sources. Dual independent screening and quality assessment were performed in line with the NICE reference case.
RESULTS: A total of 44 primary studies and 20 economic models were included. Utility and disutility estimates were identified across all prespecified health states. EQ‑5D was the predominant elicitation method (33/44 studies). Mean utility values from primary studies ranged from 0.64-0.86 for NDS, 0.12-0.66 for DS, 0.82-0.89 for TIA and 0.35-0.99 for MI. Utilities declined consistently with increasing disability by mRS (0: 0.89-1.00; 3: 0.37-0.70; 5: −0.19 to 0.38). Estimates for other events included major bleeding (0.67-0.84), minor bleeding (0.68-0.69) and dementia (0.19-1.00). Economic models showed broadly consistent patterns but tended to report lower utility values for stroke (NDS: 0.40-0.79; DS: 0.22-0.45). Substantial heterogeneity was observed across studies, driven by differences in populations, study design and country-specific EQ‑5D value sets.
CONCLUSIONS: This review provides a comprehensive and structured evidence base of utility and disutility inputs for secondary stroke prevention modelling. While consistent patterns are observed across sources, substantial heterogeneity highlights the need for transparent selection, justification and sensitivity testing of utility inputs in economic evaluations.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
EE621
Topic
Economic Evaluation
Disease
Cardiovascular Disorders (including MI, Stroke, Circulatory)