Thecosystem: A Systemwide Health and Social Care Decision Analytical Model to Assess Population-Based Interventions for Older People’s Care in the UK
Author(s)
Tarun K. George, MD, MHS,MPH1, Harry Hill, BSc, MSc, PhD2, Chris Bojke, BA, MSc, PhD3.
1SCHARR, School of Health and Related Research, University of Sheffield, Sheffield, United Kingdom, 2University of Sheffield, Sheffield, United Kingdom, 3Lumanity, Leeds, United Kingdom.
1SCHARR, School of Health and Related Research, University of Sheffield, Sheffield, United Kingdom, 2University of Sheffield, Sheffield, United Kingdom, 3Lumanity, Leeds, United Kingdom.
OBJECTIVES: With an increasing proportion of older people and rising health and social care costs, it's essential to make holistic decisions that incorporate relevant costs and resources used in various health & social care sectors (such as GP care, hospitalisation and social services). The objective of this study was to develop a system-wide economic model to study population based interventions for older people’s care in the UK.
METHODS: A conceptual Markov model (The Elder Care Ecosystem) was iteratively developed after a detailed review of literature and diverse stakeholder involvement. Frailty, a cumulative deficit based clinical state was used as the spine of the model which linked to services used, costs and QALYs. Inputs such as frailty transitions, and service utilisation rates were obtained from relevant UK based sources and a societal perspective was taken. The model was used to study 3 population subgroups and four population based interventions.
RESULTS: The basecase projected the life years, QALYS, services used and costs for the population. A scenario analysis for a cohort of ethnic minorities and social deprivation demonstrated the quantitative changes and provided valuable insights to the changes in costs, QALYs and services used. The deprivation subgroup had 12% lower QALYs than the basecase and the asian ethnicity accounted for 26% higher costs. Among the intervention types (ITs) frailty reduction programs strongly dominated being more cost effective and less resource utilising than fall prevention, hospital at home or health financing programs.
CONCLUSIONS: This novel model is a valuable tool that provides a quantitative projection of scenarios that can help priority setting, resource allocation and policymaking for communities. The versatility of the model enables assessment of a range of ITs with appropriate assumptions. Subsequent iterations could involve microsimulation based models which would be more data heavy but also realistic.
METHODS: A conceptual Markov model (The Elder Care Ecosystem) was iteratively developed after a detailed review of literature and diverse stakeholder involvement. Frailty, a cumulative deficit based clinical state was used as the spine of the model which linked to services used, costs and QALYs. Inputs such as frailty transitions, and service utilisation rates were obtained from relevant UK based sources and a societal perspective was taken. The model was used to study 3 population subgroups and four population based interventions.
RESULTS: The basecase projected the life years, QALYS, services used and costs for the population. A scenario analysis for a cohort of ethnic minorities and social deprivation demonstrated the quantitative changes and provided valuable insights to the changes in costs, QALYs and services used. The deprivation subgroup had 12% lower QALYs than the basecase and the asian ethnicity accounted for 26% higher costs. Among the intervention types (ITs) frailty reduction programs strongly dominated being more cost effective and less resource utilising than fall prevention, hospital at home or health financing programs.
CONCLUSIONS: This novel model is a valuable tool that provides a quantitative projection of scenarios that can help priority setting, resource allocation and policymaking for communities. The versatility of the model enables assessment of a range of ITs with appropriate assumptions. Subsequent iterations could involve microsimulation based models which would be more data heavy but also realistic.
Conference/Value in Health Info
2025-11, ISPOR Europe 2025, Glasgow, Scotland
Value in Health, Volume 28, Issue S2
Code
EE730
Topic
Economic Evaluation
Topic Subcategory
Novel & Social Elements of Value
Disease
No Additional Disease & Conditions/Specialized Treatment Areas