Analysis of Medicaid/SCHIP Eligibility

From our model results, we find a small reduction in consumer financial risk after a Medicaid expansion, but those newly on Medicaid/SCHIP would have substantial reductions in financial risk:

• We find a 2 to 12 percent reduction in the median proportion of income spent on health care by the non–elderly overall. Read more below
• For those newly on Medicaid/SCHIP, our model shows reductions of 51 to 64 percent in the median percentage of income spent on health care for the newly insured and even larger reductions among those previously on group insurance. Read more below
• We estimate that the proportion of people living in families who spend more than 10 percent of their income on health will decline 31 to 47 percent among those newly insured; reductions among those who were previously on group insurance are much larger. Read more below

We find a 2 to 12 percent reduction in the median proportion of income spent on health care by the non–elderly overall.

As shown in Table 1, a Medicaid/SCHIP expansion will reduce the median proportion of income spent on health care by the non–elderly from 6 to between 5.3 and 5.9 percent, depending on the new income eligibility. A reduction in the median percentage of income spent on health care of 10 percent or greater is meaningful.

Table 2 shows that the proportion of non-elderly who will pay 10 percent or more of their income on health care is reduced 3 to 15 percent, a notable difference at the population level. For a family of four at 100 percent of the federal poverty level, 10 percent of income is about $2,120. This is consistent with the modeling result showing a decrease in out-of-pocket spending (see Spending results).

For those newly on Medicaid/SCHIP, our model shows reductions of 51 to 64 percent in the median percentage of income spent on health care for the newly insured and even larger reductions among those previously on group insurance.

In Table 3, we show that the median percentage of income spent on health care is substantially reduced under a Medicaid/SCHIP expansion, both among those who become newly insured (51 to 64 percent reduction) and among those who switch from another source of insurance (94 percent reduction). About half of the newly insured were spending considerably less on health care than the insured population (about 3 to 4 percent compared to 6 percent for the population) before the policy change, and after the policy change about half will spend just 1 to 2 percent on health care. The median percentage of income spent on health care among those who had group insurance before the policy change is higher than the median for the non-elderly population and then declines after the policy change to near zero.

We estimate that the proportion of people living in families who spend more than 10 percent of their income on health will decline 31 to 47 percent among those newly insured; reductions among those who were previously on group insurance are much larger.

In Table 4, we show that a Medicaid/SCHIP expansion will substantially reduce the proportion of people in families who spend more than 10 percent of their income on health care. Among those who are newly insured, the proportion is reduced 31 to 47 percent; this change makes the newly insured less subject overall to significant health care costs. Among those who switched from group insurance to Medicaid/SCHIP, the reduction is 95 percent. A higher proportion of people in group insurance before the policy change were spending more than 10 percent of their income on health than were the uninsured; by enrolling in Medicaid/SCHIP, this group has almost no one with expenditures at this level.

Expanding Medicaid/SCHIP eligibility is unlikely to affect the reliability of care:

• No studies directly examine changes in the reliability of care when individuals enroll in Medicaid or SCHIP. Read more below
• Multiple studies suggest that the reliability of care does not vary significantly with insurance status; when uninsured individuals have access to health services, the reliability of these services is the same for both the insured and the uninsured. Read more below

No studies directly examine changes in the reliability of care when individuals enroll in Medicaid or SCHIP.

Reliability refers to the consistency with which the health care system delivers evidence based care to everyone in the population. A reliable health care entity is one that provides the right care, to the right patient, at the right time, every time. Increasing the number of individuals with coverage by expanding Medicaid/SCHIP eligibility would not be expected to directly affect the reliability of the health care system as a whole. That is, the likelihood of the health care system delivering reliable care would not be expected to change simply because people gain coverage.

No studies have examined the relationship between expanding Medicaid/SCHIP eligibility and reliability of care. However, there is some literature on the relationship between the type of insurance coverage people have and the reliability of care they receive.

Multiple studies suggest that the reliability of care does not vary significantly with insurance status; when uninsured individuals have access to health services, the reliability of these services is the same for both the insured and the uninsured.

Uninsured people have less access to health services than insured people, they receive lower levels of care, and they are more likely to seek nonurgent care in the emergency department. In theory, reliability could improve for these patients when they gain coverage if they receive care in a primary care setting in which they can receive regular screening, earlier diagnosis, and timely treatment. However, there are significant deficiencies in the reliability of care throughout the health care system, and the evidence thus far does not show differences in reliability of care between the insured and the uninsured. Most evidence shows that, once the uninsured are receiving services, the reliability of their care is similar to the level experienced by insured people. In fact, the gap between actual and optimal overall reliability for all people is much larger than the gap between reliability for the uninsured and that for the insured.

The available evidence does not show a strong link between insurance coverage and reliability. RAND studies (McGlynn et al., 2003; Asch et al., 2006) have provided the most comprehensive assessment of the reliability of the health care delivery system and have found few differences in the reliability of care for the uninsured versus the insured. There were no large differences, even when looking at specific types of care (acute, chronic, and preventive) or functions of care (screening, diagnosis, treatment, and follow-up). A limitation of these studies is that everyone in the study population had at least one contact with the health care system in two years, which may not reflect the experience of individuals who receive no care at all.

Several other studies also found no differences in reliability between insured and uninsured people. Leape and colleagues (1999) found no overall differences in underuse of coronary revascularization between uninsured and insured people. Young and colleagues (2001) found that "insurance and income had no effect on receipt of appropriate care for depressive and anxiety disorders." Harman, Edlund, and Fortney (2004) reported disparities in initiation of depression treatment between the uninsured and the insured, but no differences in quality of care once treatment was initiated. Of note, these studies do not necessarily examine the overuse of care, which might result from the uninsured receiving duplicate or discontinuous services at multiple locations.

On the other hand, some studies using data from surveys on self-reported health and health care use have found differences in quality measures between uninsured and insured populations. Ayanian et al. (2000) evaluated results from the Behavioral Risk Factor Surveillance System (BRFSS) and found that uninsured adults were less likely to have had cancer screening, diabetes care, or cardiovascular risk factor reduction and were less likely to have had a routine checkup in the preceding two years. Ross, Bradley, and Busch (2006) used the BRFSS survey to compare preventive service use between high and low income uninsured individuals. For both groups, rates of cancer screening, diabetes management, and cardiovascular risk reduction were lower than those for insured patients. Busch and Duchovny (2005) examined changes in access to and quality of care that resulted from Medicaid expansions in the 1990s. These expansions allowed the enrollment of low income parents at higher income levels. The study found that, among previously uninsured mothers who had not been receiving cancer screening, 29 percent received screening following enrollment into the program. These studies may be limited in that they rely on self-reports of care received.

Although we do not expect any system-level changes in delivery that would contribute to significant improvements in reliability as a result of Medicaid expansions, some changes may be observed for individuals who become newly covered. We know relatively little about how the patterns of quality change among people who move from being uninsured to being covered by Medicaid/SCHIP, and this policy option results in a large proportion of people switching from other sources of insurance to Medicaid.

From our model, we estimate a small increase in life years associated with a Medicaid/SCHIP expansion:

• We estimate that a Medicaid/SCHIP expansion would result in 410,000 to 1.55 million additional life years. Read more below
• Theory and published studies suggest that, if expanding Medicaid/SCHIP increases rates of coverage, the health of some groups should improve. Read more below
• The magnitude of the effect on health may depend on the health of the individual when he or she gains insurance and other socioeconomic factors, as well as on changes in access afforded by health insurance and the reliability of care received. Read more below

We estimate that a Medicaid/SCHIP expansion would result in 410,000 to 1.55 million additional life years.

In our analysis, we used projections from the RAND Future Elderly Model to determine the increase in life expectancy attributable to a change in insurance status (Goldman et al., 2004). We assumed that individuals who obtain coverage under this policy option would retain that coverage continuously until they become eligible for Medicare at age 65. We further assumed that those who obtain insurance change from the expected mortality rate observed for the uninsured to the expected mortality rate observed for the insured. We observed no difference in mortality among the population over age 65 based on their insurance status prior to becoming eligible for Medicare; therefore, we assumed that all of the effect of becoming newly insured occurs between becoming insured and age 65, a mortality difference of 9.06 percent. We assumed that there are no changes in rates of treatment or in the effectiveness with which medical care is delivered as a result of this policy change.

In the table, we show the expected gain in years of life for all Americans under a Medicaid/SCHIP expansion that affects individuals with incomes below 100, 200, and 300 percent of the federal poverty level (FPL). Depending on the new threshold for Medicaid/SCHIP eligibility, total years of life gained could range from 410,000 to 1.55 million.

Theory and published studies suggest that, if expanding Medicaid/SCHIP increases rates of coverage, the health of some groups should improve.

Acquiring health insurance coverage should improve access to medical care and result in improved health. The literature suggests that there may be a modest relationship between insurance status and health outcomes, but methodological issues have made it difficult to make a precise estimate. The RAND Health Insurance Experiment randomly assigned families into health plans, providing an opportunity to assess how benefit generosity affected health outcomes in a setting in which health was unrelated to insurance choice (Newhouse and Insurance Experiment Group, 1993). Overall, the study found that benefit generosity had a negligible influence on health outcomes for the general population, although there were some benefits for low income participants. However, since everyone in the RAND Health Insurance Experiment had at least minimal health coverage, its results cannot necessarily be used to gauge the effect of becoming insured.

Levy and Meltzer (2008) reviewed the literature on the relationship between health insurance and health and drew several conclusions. First, they found that most studies are not able to establish a causal relationship between health insurance and health because the studies do not account for the multiple other factors that affect these two variables or the fact that health itself affects insurance status. Second, they found substantial evidence that health insurance improves health in vulnerable populations, such as infants, children, individuals with HIV, and some low income adults, but there is less evidence of this relationship for other groups. Finally, they suggest that it may be difficult to generalize the results of the studies thus far.

Hadley (2003) reviewed the literature from the past 25 years on this subject and found general consensus among studies that providing the uninsured with health insurance would result in improved health. Hadley's "best guess" on the size of the effect was a 4 to 25 percent reduction in mortality for previously uninsured people. Our estimate (discussed above) is contained in this range. Although Hadley cites substantial methodological difficulties within these studies, he acknowledges the substantial consistency in the findings: improved health with insurance in many populations and disease states.

The magnitude of the effect on health may depend on the health of the individual when he or she gains insurance and other socioeconomic factors, as well as on changes in access afforded by health insurance and the reliability of care received.

Over the longer term, it is possible that Medicaid/SCHIP expansions will improve health; however, studies from previous insurance expansions show mixed results on the effects of health. Some subgroups of children show improvements in health when they enroll in Medicaid, but the results are not consistent across all populations (Holl et al., 2000; Lykens and Jargowsky, 2002; Stevens, Seid, and Halfon, 2006). Note that Child Health Plus (CHPlus) was a child health insurance program implemented in New York State in 1991 to provide health insurance to children age 13 or under. CHPlus later became a model for the nationwide SCHIP. Studies of Medicaid expansions to pregnant women also seem to suggest improved infant mortality rates (Currie and Grogger, 2002).

Because medical care represents only one of the many determinants of health (others include behavior, socioeconomic status, education, and genetics), improved access to medical care via insurance changes may have only a modest effect on health. McGinnis, Williams-Russo, and Knickman (2002) suggest that only 10 percent of preventable deaths may be due to problems with medical care.

Estimating the effect of insurance on health poses significant methodological challenges. In particular, insurance status and health are not independent; that is, health can directly affect the ability or desire to obtain coverage. Healthy people may be less likely to enroll in insurance because they anticipate having minimal health expenditures. On the other hand, sick individuals may be unable to purchase individual policies; however, when these individuals become sick enough, they may qualify for a government sponsored program, such as Medicaid/SCHIP. Other unobservable characteristics may also influence both health and insurance status. Thus, when researchers attempt to study differences in health based on health insurance status, it can be difficult to discern whether the provision of insurance really causes improved health.

Expanding Medicaid/SCHIP eligibility will not have a direct effect on capacity:

• Given our model results, we do not estimate increased utilization of services and thus do not expect that expanding the eligibility of Medicaid/SCHIP will directly affect the capacity of the health care system. Read more below
• Prior studies consistently show that utilization of health care services increases when people gain insurance coverage; the health of new enrollees will affect the level of demand. Read more below

Given our model results, we do not estimate increased utilization of services and thus do not expect that expanding the eligibility of Medicaid/SCHIP will directly affect the capacity of the health care system.

Capacity refers to the human resources (personnel and their productivity) and capital (medical equipment, hospitals, etc.) of the health care system. There is no evidence that expanding Medicaid/SCHIP eligibility will change the capacity of the U.S. health care system. In theory, there is no direct connection between the capacity of the health system and the change in the number of persons becoming covered under a Medicaid/SCHIP expansion. Our model estimates project a 6 to 30 percent increase in the net number of newly insured, from about 4 to 14 million, depending on the new minimum level of eligibility; however, even at the highest level, the increased demand is likely to be small. Our model also estimates that spending will likely not change in the aggregate, suggesting that there will not be an increased demand for services. We estimate that 33 to 61 percent of those newly enrolling in Medicaid/SCHIP who previously had other insurance will actually decrease their use of health services.

Prior studies consistently show that utilization of health care services increases when people gain insurance coverage; the health of new enrollees will affect the level of demand.

Many studies have shown that uninsured people use fewer health care services than insured people, and that changes in insurance coverage are associated with changes in health care utilization. This finding would suggest that capacity may be affected in some areas as coverage levels increase. Buchmueller and colleagues' (2005) review of relevant studies found consensus that expanding health insurance coverage would increase utilization among the newly insured population.

Several studies of SCHIP in various states show that use of primary and specialty care increases with enrollment in the program. Holl et al. (2000) and Szilagyi et al. (2000) suggest that the increase in utilization is due at some level to previously unmet need, but the extent of utilization post-enrollment that is medically necessary is unknown. Results from Banthin and Selden's (2003) simulation of individuals who would have been eligible for Medicaid before and after its implementation showed that Medicaid has a statistically significant effect on the utilization of physician visits: Between 1987 and 1996, the proportion of Medicaid eligible children who had a physician visit increased from 57 percent to 64 percent. The study did not find any statistically significant changes in the proportion of Medicaid eligible children who visited the dentist or the emergency department.

Studies of Medicaid/SCHIP expansions show that these programs also increase health care utilization among adults. In a study examining the effects of a public coverage expansion in Tennessee, researchers found that enrolled adults used a greater number of most of the measured health services than did the uninsured, including preventive services (Moreno and Hoag, 2001). A study of adults enrolled in the Oregon Health Plan, a Medicaid waiver program, found that enrollment increased the use of preventive services, including Pap smears and blood pressure checks, compared with matched low income insured and uninsured individuals. Enrollees were also more likely to have had a physician's visit, a visit with a specialist, a dental visit, an emergency department visit, and a hospitalization (Mitchell et al., 2002).

An increase in utilization may not be as large as expected under a Medicaid/SCHIP expansion, because newly eligible beneficiaries are likely to be healthier than current beneficiaries. In a study that modeled incremental health care spending related to Medicaid expansions that occurred between 1984 and 1994, Gordon and Selden (2001) found that children who enrolled in past Medicaid expansions had lower incremental costs per enrollee compared with average Medicaid per capita expenditures (spending estimates in this study do not account for a reduction in government spending that is already being directed toward the uninsured). The authors credit a healthier and older population of newly eligible enrollees for the lower incremental spending per enrollee, compared with existing enrollees, because the majority of children with substantial health problems already would have been enrolled under Medicaid's provisions for the medically needy. Increases in utilization may also not be as large under a Medicaid expansion if new enrollees are unable to find physicians who are willing to treat Medicaid patients because of low reimbursement rates.

It is difficult to project how reductions in uncompensated care that result from Medicaid/SCHIP expansions may affect overall capacity, particularly given current rates of reimbursement in this program. In their study of the relationship between public sector health policy changes on for-profit and nonprofit hospitals' provision of uncompensated care, Davidoff and colleagues (2000) found that uncompensated care provided by for-profit hospitals declined relative to Medicaid/SCHIP eligibility expansions. They did not, however, find the same to be true among nonprofit hospitals.