ESTIMATING THE IMPACT OF VALUE-BASED INSURANCE DESIGN ON U.S. LIFE EXPECTANCY, HEALTH CARE COSTS, AND INCREMENTAL COSTS OF PROVIDING UNIVERSAL HEALTH INSURANCE

Purpose: Evidence is accumulating to suggest that cost-sharing decreases health expenditures but also reduces high-value care. Value-based insurance design (VBID) has been proposed to mitigate this effect while controlling health expenditures. Our objective was to estimate the impact of VBID on U.S. life expectancy, health care costs, and incremental costs of providing universal health insurance.

Methods: We used a published simulation of costs and life expectancy gains from U.S. health care to estimate the impact of promoting VBID across the U.S. health care system. We assumed that cost-sharing would be eliminated for high-value services (incremental cost-effectiveness ratio [ICER} <$100,000 per life-year), would remain unchanged for intermediate- or indeterminate-value services (ICER $100,000- $300,000 per life-year or unknown), and would be increased for low-value services (ICER >$300,000 per life-year). Sensitivity analyses included varying the level of cost-sharing for low-value services through a plausible range (20%-50%). Elasticity of health care demand was based on RAND data. All costs were estimated from societal perspective in 2003 US$.

Results: With prevailing patterns of cost-sharing and health insurance, health care conferred 4.70 life-years at a per-capita annual expenditure of $5700. VBID increased the benefit conferred by health care to between 4.91 life-years (with cost-sharing for low-value services set at 50%) and 4.96 life-years (with cost-sharing for low-value services set at 20%). Health expenditures were decreased below current levels (i.e., the reduction in spending on low-value services outweighed the increase in spending on high-value services) if cost-sharing for low-value services was >21%. If VBID was combined with universal health insurance coverage, benefit from health care would increase further to between 5.11 life-years (with cost-sharing for low-value services at 50%) and 5.16 life-years (with cost-sharing for low-value services at 20%). Health expenditures were decreased below current levels (i.e. the reduction in spending on low-value services outweighed the increase in spending on high-value services and more health care for the previously uninsured) if cost-sharing for low value services was >30%. Sensitivity analyses suggest that life-expectancy increases from VBID were robust

Conclusion: VBID may offset the incremental cost of universal health insurance if cost-sharing for low-value services is ≥30%, and may increase life expectancy of the U.S. population by nearly half a year.