Thank you for inviting me to appear before the Committee this morning. I am a Principal Research Associate at The Urban Institute, and a Senior Fellow with the Center for Studying Health System Change, which are independent, nonprofit, research institutions here in Washington. Most of my research over the last two years has concentrated on economic analyses of the costs and consequences of being uninsured. Based on this research and reviews of other studies done over the last 25 years, my presentation focuses on some of the broad economic costs associated with a large uninsured population.
I have three major points. • First, as a nation we already spend a substantial amount of money to pay for care received by uninsured people. • Second, much of this money is spent inefficiently, going to hospitals for emergency room and inpatient care to treat people who probably would have been treated earlier, more cheaply, and more effectively if they had insurance. • Third, lack of insurance reduces the health of the nation, and as a result, also reduces the wealth of the nation.
1. We already spend a substantial amount of money on care to the uninsured.
In a study published earlier this year, we estimated that in 2001 the nation spent about $35 billion on uncompensated care received by the uninsured, both those who are uninsured for a full year and those who lack coverage for part of a year.
About two-thirds of uncompensated care, almost $24 billion, was provided by hospitals caring for uninsured people in emergency rooms, outpatient departments, and as inpatients.
We also estimated that a substantial portion of uncompensated care, perhaps as much as $30 billion, is already being financed by taxpayers through programs such as: Medicare and Medicaid Disproportionate Share Payments; Medicaid Upper Payment Limit payments; state and local tax appropriations, primarily to public hospitals and clinics; Federal grants to community health centers; and Federal direct care provided by the Department of Veterans Affairs and the Indian Health Service.
For people who are uninsured all year, uncompensated care covers about 60% of the care they receive. However, in spite of what appears to be a substantial subsidy, uncompensated care is not a substitute for insurance, nor are the uninsured free riders who are taking advantage of everyone else. On average, the uninsured receive about half as much care as people insured all year, roughly $1,250 compared to about $2,500 per person for someone covered by private insurance.
Moreover, in spite of receiving about half as much care as the privately insured, the uninsured actually pay about the same amount out-of-pocket for the care they receive, and what they do pay out-of-pocket represents a bigger burden on their family incomes. Being uninsured represents triple jeopardy: you receive less medical care than the insured, you pay about as much out-of-pocket, and what you pay represents a bigger burden on your family’s resources.
2. Much of the money we currently spend on uncompensated care is spent inefficiently.
Being uninsured is like playing Russian roulette with your health. Research clearly shows that compared to the insured, the uninsured are more likely to delay seeking care and to have unmet health needs. If they’re lucky, they’ll get better without any care. But if they’re not, the uncompensated care they eventually receive from the safety net can wind up costing much more than if they had been treated when symptoms first appeared or if their illness were diagnosed before symptoms become apparent.
Studies have shown that the uninsured are more likely to be hospitalized for preventable conditions, i.e., medical conditions that can be adequately treated on an outpatient basis and should not require hospitalization. One study estimated that about 12% of the uninsureds’ hospital stays were for preventable conditions, compared to about 8% for the privately insured.
Another study of avoidable hospitalization estimated that the extra cost associated with preventable stays was $105 million in only nine states.
Studies also suggest that the expansions of insurance coverage for children through Medicaid and SCHIP have led to reduced rates of avoidable hospitalizations for children, by as much as 22%. Other studies show that uninsured people with cancer are more likely to be diagnosed at an advanced disease stage, which is strongly related to reduced survival.
Numerous other studies have found that the uninsured are less likely to receive screening and diagnostic tests known to lead to early detection of cancer, heart disease, and diabetes – diseases with high mortality rates and high levels of disability and diminished activity status.
Even among people who know they have hypertension or diabetes, use of appropriate medications and routine follow-up care is lower for the uninsured compared to the insured. In sum, a large body of research provides convincing evidence that the uninsured receive less preventive and diagnostic care, receive less therapeutic care even after being diagnosed, and, as a result, die earlier and experience greater limitations than otherwise similar people with insurance coverage.
Moreover, as I noted earlier, access to the safety net is not a substitute for insurance. In an ongoing study, we compared the effects of expanding insurance coverage versus expanding the safety net on low-income people’s access to care. Simulations suggest that a 10% increase in insurance coverage, would reduce the proportions reporting an unmet medical need or putting off care by 25-30%. Spending a comparable amount of money on expanding the safety net would reduce unmet need and putting off care by one-third to half as much. If insurance coverage were universal, the percentages with an unmet need or delaying in seeking care would fall to 4-9%, roughly the same levels we observe among people with full-year coverage through their employers.
3. Reduced health is associated with lower earnings and educational attainment, and probably with higher payments from public programs.
Lack of insurance reduces health, which has adverse effects on adults’ work and earnings. Evidence also suggests that poor health in children affects their educational attainment. Unfortunately, social science studies of the relationships between health and work, earnings, and educational attainment are all hampered by the difficulty of sorting out what is causation and what is association. We don’t have the ability to randomly assign people to excellent, good, or poor health and then see how it affects their education, work, and earnings. Nevertheless, there is enough presumptive evidence, I believe, that suggests that poor health among adults leads to lower labor force participation, lower work effort if in the labor force, and lower earnings. While it is difficult to put precise numbers on these effects, it appears that a person in fair or poor health might earn from 15-20% less on an annual basis than an otherwise similar person in very good or excellent health.
Poor health of a family member also affects the ability to work. Studies have shown that family caregivers, parents caring for sick children or a spouse caring for a sick partner, work less and earn less. This lost work time and lost earnings represent foregone productive activity that would contribute to our national economy, and to tax revenues collected.
Studies of children’s health and educational achievement suggest that children in poor health have poorer school attendance and lower school achievement and cognitive development. However, a number of these studies focus on comparisons of children who were born at low birthweight. The research is more ambiguous in showing that insurance coverage improves birthweight, although it is much more clear that insurance leads to higher infant survival.
While there is still much that needs to be done to develop more precise estimates of exactly how much health would improve and what the quantitative impact would be on earnings and public program payments, I can provide more detail about a piece of the puzzle from recent studies of health insurance, health, and medical use by older middle-aged adults. These studies show that lack of insurance increases the probability of disability or major health deterioration in older middle-aged people, roughly between the ages of 50 and 65. Disability at this age leads to early coverage by the Medicare program and transfer payments made through the DI and SSI programs. A highly relevant question to this Committee’s deliberations is whether complete insurance coverage in late middle age would improve people’s health at age 65 and, if it does, what are the implications for Medicare and Medicaid spending on people after they turn 65. Another ongoing study suggests that lack of insurance during late middle age does in fact lead to significantly poorer health at age 65 – fewer people survive and those who do have an increased incidence of being in fair or poor health with a disability. Our analysis simulates how much health would improve if this cohort had complete insurance coverage and whether Medicare and Medicaid spending would increase or decrease after age 65.
We find that more people survive to age 65, and those who survive are in significantly better health. As a result of the health improvement, and in spite of the fact that more people survive, our simulation suggests that Medicare and Medicaid would save about $10 billion a year on care to 66-68 year olds. Our calculations also suggest that these savings would cover about 50% of the cost of expanding coverage to this cohort of older middle-age people.
SUMMARY
The debate on whether to expand health insurance coverage to all Americans will inevitably emphasize the cost of providing insurance. It must also include the benefits of having insurance. While more work needs to be done to develop precise quantitative estimates of the magnitude of these benefits, I believe that the research is quite clear in demonstrating that lack of insurance leads to poorer health, and that poorer health is associated with less educational attainment, lower labor force participation, and lower earnings. These consequences undoubtedly lead to lost tax revenues and higher public program payments for both medical care and income support payments. Finally, I’ve focused only on the dollar and cents issues around the question of the consequences of being uninsured. However, poor health and premature death obviously have significant subjective effects on one’s own and on family members’ sense of well-being. The total value of good health goes beyond, possibly well beyond, a narrow accounting of financial consequences.
