Free Declaration - District Court of Arizona - Arizona

EXHIBIT B TO SECOND DECLARATION OF LEIGHTON KU

Case 2:03-cv-02506-EHC

Document 137-3

Filed 03/10/2008

Page 1 of 22

820 First Street, NE, Suite 510, Washington, DC 20002 Tel: 202-408-1080 Fax: 202-408-1056 [email protected] www.cbpp.org

May 7, 2003

CHARGING THE POOR MORE FOR HEALTH CARE: COST-SHARING IN MEDICAID By Leighton Ku Executive Summary Concerns about rising health care costs are leading state policy officials to consider increasing the amount that low-income Medicaid beneficiaries have to pay for health services by raising copayments or imposing premiums. The Kaiser Commission on Medicaid and the Uninsured found that 17 states planned to increase Medicaid cost-sharing during 2003. The costsharing increases being considered range from small to large, and some would require special waivers. Higher cost-sharing is not without significant risks. In particular, it risks making Medicaid less effective, because health services could become less affordable for low-income families, seniors, and people with disabilities and higher cost-sharing can cause some beneficiaries to avoid or delay essential medical care or drop insurance coverage altogether. While cost-sharing is commonplace among middle-class people with private health insurance, the consequences are more serious for the low-income beneficiaries served by Medicaid, who have limited means and already are bearing out-of-pocket medical costs that consume a larger percentage of their incomes. Effects on Health Care Utilization and Health Status. A substantial and rigorous body of research has demonstrated that low-income individuals are more vulnerable to the adverse effects of cost-sharing than other groups are. Cost-sharing policies that cause only modest reductions in health care use among middle-class individuals can result in more substantial reductions in health care use and lead to significant adverse health consequences among poorer individuals, especially those with chronic health problems.
·

Higher copayments for medical services or prescription drugs cause low-income people to use substantially fewer essential and effective medical services or medications. The RAND health insurance experiment, one of the most rigorous and important health policy studies ever conducted, found that low-income adults and children reduced their use of effective medical care services by as much as 44 percent when they were forced to make copayments, a much deeper reduction than occurred among those with higher incomes. This study also found, as have other studies conducted since then, that costsharing leads people to reduce their use of both more effective care and less effective care. Some proponents of cost-sharing argue that copayments encourage people to be more cost-conscious so that they make "smarter" health care

·

Case 2:03-cv-02506-EHC

Document 137-3

Filed 03/10/2008

Page 2 of 22

decisions and avoid unnecessary care. The research indicates, however, that most consumers, whether low-income or not, do not selectively reduce their use of less effective care; consumers typically lack the clinical knowledge to make informed choices. For example, patients might choose to go to the doctor because of coughs or discomfort that are symptoms of common colds that will pass in a day but avoid seeking care for chest pains that could signal heart disease.
·

The RAND study also found that cost-sharing led to poorer health among lowincome adults -- including worse blood pressure and vision -- than those who were not subject to copayments. Similarly, low-income children with costsharing obligations were more likely to be anemic and to have more untreated dental problems than children who received free care. In contrast, the research found no significant differences in the health status of higher-income people between those who had cost-sharing obligations and those who did not. The reduction in medical access caused by cost-sharing can impair health, aggravate illness and precipitate unnecessary health care emergencies. A large study in Quebec found that after copayments for prescription drugs were imposed, poor adults had 88 percent more emergency room visits and experienced a 78 percent increase in medical events like hospitalization or institutionalization, as a result of the medical problems experienced when these low-income people went without essential medications. The study concluded that stricter copayments could increase the toll of avoidable illnesses in Canada and the United States. The Quebec study also indicated that some of the reduced expenditures for prescription drugs were offset by increased use of costly services received in emergency rooms, inpatient hospital care, or institutionalization. Much of the savings in prescription drug costs may therefore be counteracted by increases in other health care costs. Other studies have demonstrated the difficulties that low-income people encounter in accessing medical services when copayments are required in Medicaid. For example, a study in Tennessee found that about one-fifth of enrollees were unable to afford the copayment when they visited their doctor's office. One-fifth were unable to pay copayments to pharmacists, and two-thirds of those unable to pay simply went without their prescribed medications. Research has also shown that charging monthly premiums for coverage lowers enrollment of eligible, low-income people, even if the premiums charged are relatively small. An Urban Institute study estimated declines in enrollment of 16 percent when participants are charged premiums that equal one percent of family income, enrollment declines of about 49 percent if premiums equal three percent of family income and enrollment declines of about 74 percent if premiums are set at five percent of family income. Similarly, the General Accounting Office found participation in state pharmacy assistance programs fell sharply in states that increased premiums for those programs.

·

·

·

·

Case 2:03-cv-02506-EHC

2 Document 137-3

Filed 03/10/2008

Page 3 of 22

·

Other studies have examined the use of premiums in children's health insurance programs. Research in Florida found that premiums encourage healthier participants to drop off the program, causing average health care costs to climb since those who remain on the program are less healthy, on average.

How Much Do People Pay Now? Some contend that higher Medicaid copayments would foster individual responsibility by making beneficiaries more cost-conscious; they also note that many middle-income workers in private health plans face rising cost-sharing obligations. However, Medicaid beneficiaries already spend a substantial share of their incomes on medical expenses.
·

Non-elderly, non-disabled adult Medicaid beneficiaries with incomes below the poverty line spent an average of 2.3 percent of their incomes on out-of-pocket medical expenses, based on analyses of the 1999 Medical Expenditure Panel Survey. (This figure may be higher today, since a number of states have increased Medicaid copayments since 1999.) This is more than four times the share of income spent by middle-income adults who have private health insurance; these adults spent an average of 0.5 percent of income on out-of-pocket medical costs in 1999. In dollar terms, privately-insured middle-income adults have higher out-of-pocket outlays than poor Medicaid beneficiaries, but because their average incomes are far higher than those on Medicaid, the out-of-pocket medical expenditures of middle-income adults with private insurance constitute a much smaller percentage of their incomes.

·

Another reason sometimes advanced for increasing cost-sharing is to help Medicaid "keep up with the times," since cost-sharing is rising for a large number of people with private health insurance coverage. Many low-income families, however, have less income available to spend on health expenses now than in earlier years. In many areas, rapidly rising housing costs are claiming an increasing share of poor families' incomes. The median cost of housing for a renter with income below the poverty line rose from 53 percent of income in 1989 to 65 percent of income in 2001, according to the Census Bureau's American Housing Surveys. In many areas, child care expenses also are rising. Studies show that many families with incomes below the poverty line already experience hardships such as running out of food or having difficulty paying rent or utility bills. Elevated cost-sharing in Medicaid could force many low-income individuals to choose between health care and other basic needs, such as housing and child care. Simply put, a large number of poor Medicaid beneficiaries have little or no disposable income to spend on medical care unless they stint on some other basic need. Prescription Drugs. Some states would like to use higher Medicaid copayments to encourage beneficiaries to shift to lower-cost prescription drugs, such as generic rather than brand-name drugs. Data show, however, that Medicaid patients already are more likely to use generic drugs than people with private insurance. More important, new research suggests that some copayment arrangements could even reduce patients' use of generic drugs, not promote it, by causing beneficiaries to buy fewer medications overall.

Case 2:03-cv-02506-EHC

3 Document 137-3

Filed 03/10/2008

Page 4 of 22

Health care providers have expressed strong reservations about cost-sharing, fearing they will have to cover the cost of copayments when beneficiaries are unable to. On the other hand, the only published study in this area -- a survey of pharmacists -- suggests that the primary burden of cost-sharing falls upon beneficiaries rather than providers. Who Is Most Vulnerable? The potential risks of higher cost-sharing are most acute for seniors and people with disabilities. Since these individuals use the most health services and medications, their out-of-pocket costs for copayments would be highest and they are the people most likely to avoid or delay needed health care because of cost problems. Researchers at the University of Maryland have shown that the adverse effects of Medicaid cost-sharing are the greatest for those with the worst health. Cost-sharing could also create barriers to the use of preventive and primary health care by children and pregnant women, which could have longer term health consequences. Most of those that states could ask to make higher copayments in Medicaid are individuals or families with incomes below the poverty line ($15,260 for a family of three). Federal law prohibits copayments from being required of children and pregnant women, the two groups of Medicaid beneficiaries who are most likely to have incomes above the poverty line, and most other state expansions of Medicaid to people with incomes greater than 100 percent of the poverty line have been authorized under federal waivers that already permit cost-sharing for those with higher incomes. State Savings. States stand to lose a substantial portion of the savings generated when they institute copayments or premiums, because these approaches reduce total Medicaid costs and therefore reduce federal matching funds. For example, if a prescription drug costs $60 and the federal matching rate in a given state is 60 percent, the federal share of the cost for that drug is $36, while the state pays $24. If there is a $10 copayment, the total cost to Medicaid for the drug becomes $50, so the state share will be $20 (40 percent of $50) and the federal government will pay $30. Even though a poor state resident has paid $10 more for that prescription, the state government saves only $4. This is equivalent to imposing a tax or a user fee on a poor state resident with the state government gaining only a fraction of that tax or user fee. If, instead, the state increased some other tax or user fee on higher-income people, the costs would be experienced by people who can better afford to bear them and the state would gain the full value of the tax or user fee. Cost-sharing in Medicaid is fiscally inefficient for states since it effectively shifts revenue away from state residents. Conclusions. Expanded cost-sharing is likely to harm low-income beneficiaries. In many cases, it could be counterproductive because beneficiaries' health could worsen, leading to increases in other medical costs. Those considering increases in Medicaid cost-sharing, despite its risks, should consider two principles to curb unnecessary and unintended negative consequences:
·

States should avoid raising cost-sharing for individuals with incomes below the poverty line and should be cautious in considering increases for those whose

Case 2:03-cv-02506-EHC

4 Document 137-3

Filed 03/10/2008

Page 5 of 22

incomes are close to the poverty line. For example, while copayments are not permissible for children in Medicaid, SCHIP regulations that apply to children with incomes between 100 percent (or 133 percent) of the poverty line and 150 percent of the poverty line establish a moderately higher ($5) limit on copayments than the $3 maximum copayment that applies in Medicaid.
·

States should cap monthly copayment obligations to curb the harshest effects that increased cost-sharing would impose on those who are the sickest and who need a higher level of medical care or medications.

Current Cost-sharing Policies in Medicaid and SCHIP Because Medicaid and the State Children's Health Insurance Program (SCHIP) are designed to provide affordable health care to low-income Americans, federal statutes and regulations protect beneficiaries from being charged excessive amounts for program services. (The box on page 6 describes the current rules.) In establishing these rules, policy-makers recognized that beneficiaries have limited disposable income to pay for health care and might not be able to obtain it if they are charged too much. The philosophy underlying federal policy regarding cost-sharing has been that poor beneficiaries -- a group that accounts for the vast bulk of Medicaid beneficiaries -- should face no or very limited cost-sharing but that moderate cost-sharing is permissible for those with higher incomes, such as those enrolled in SCHIP (copayments are limited to $5 for those with incomes below 150 percent of the poverty line and there is an cap on overall cost-sharing of five percent of family income for all SCHIP families). Cost-sharing is generally less of a burden for higher-income families not only because their incomes are more ample but also because they are often in better health than poor families and may need fewer health care services. Some states are currently increasing cost-sharing in Medicaid, largely to help address state budget shortfalls. A recent survey conducted for the Kaiser Commission on Medicaid and the Uninsured found that 17 state Medicaid programs planned to raise copayment charges in 2003.1 In some cases, states plan to increase Medicaid copayments or to impose them for the first time, but expect to stay within the regulatory limits on "nominal" copayments, which are typically limited to $3 per visit or prescription. Other states, however, are considering much heftier increases in cost-sharing and presume that the federal government will approve waivers that let them exceed the current federal limits. Most Section 1115 waivers approved since 2001 have permitted copayments or premiums well beyond standard Medicaid limits. On the other hand, such increases have generally been permitted only for "expansion populations" that have incomes greater than 100 percent of the poverty line, although CMS has in a few instances permitted cost-sharing for selected people even though their incomes fall below the poverty line.
1

Vernon Smith, et al., Medicaid Spending Growth: A 50 State Update for FY 2003, Kaiser Commission on Medicaid and the Uninsured, January 2003.

Case 2:03-cv-02506-EHC

5 Document 137-3

Filed 03/10/2008

Page 6 of 22

A Summary of Current Cost-sharing Rules in Medicaid and SCHIP Medicaid Existing federal rules give states the flexibility to establish "nominal" cost-sharing in Medicaid: · Copayments are fixed dollar payments that beneficiaries must pay out-of-pocket when they use health services or obtain prescription drugs. Federal legislation specifies that copayments must be "nominal," the definition of which is established by regulation. For non-institutional care, states may set copayments of from 50 cents to $3 per visit or per prescription; for institutional care (e.g., a hospital or nursing home admission), copayments may not exceed 50 percent of the Medicaid payment for the first day of care. · Deductibles are amounts a beneficiary must spend before insurance begins to cover any health services. Federal legislation requires that Medicaid deductibles must be nominal and may not exceed $2 per family per month. · Coinsurance is similar to copayments, but the amount the beneficiary must pay out-of-pocket is not fixed but is set as a percentage of the amount paid to the health care provider. The nominal maximum established by Medicaid regulations is five percent of the Medicaid payment level for a given service. For example, if a state pays a provider $500 for a given health service, the maximum coinsurance amount is $25. · Only one type of charge (copayment, deductible or coinsurance) may be required for any service. · Premiums are amounts that an individual must pay each month (or each year) to enroll in Medicaid. Under federal law, most Medicaid beneficiaries may not be charged premiums. In cases where premiums are allowed, such as the medically needy, federal regulations establish a sliding-scale schedule with a maximum of $19 per person per month. Federal law establishes a number of exceptions to the above rules: Children, pregnant women (for pregnancy-related services), those who are institutionalized (e.g., in a hospital or nursing home), and those in hospices may not be charged copayments, deductibles, or coinsurance. · States may not assess copayments, deductibles, or coinsurance for health maintenance organization services, family planning services, or emergency medical services (although copayments may be assessed for nonemergency services provided in emergency rooms). · No health care provider may deny services to a Medicaid beneficiary because he or she is unable to pay the cost-sharing amount. The rationale for this policy is that, since Medicaid cost-sharing is nominal, it is inappropriate to deny services to an indigent patient because he or she cannot pay a very small copayment; the provider will still receive the bulk of the Medicaid reimbursement
·

SCHIP The general rule for SCHIP, established by legislation, is that the cumulative amount of cost-sharing -- including premiums, copayments, and the like -- may not exceed five percent of a family's income. Further, cost-sharing may not be imposed for preventive health services. For children with incomes below 150 percent of the poverty line ($22,890 for a family of three), lower cost-sharing limits apply; these amounts are set by SCHIP regulations, as follows: · Copayments may not exceed $1 to $5 per visit or prescription. · Deductibles may not exceed $3 per family per month. · Coinsurance may not exceed five percent of the Medicaid payment level for a given service. · Premiums may not exceed the Medicaid sliding premium schedule for the medically needy, which has a maximum of $19 per month. Medicaid and SCHIP Waivers Under Section 1115 waivers, the federal government may permit states to modify these cost-sharing rules. Although these modifications commonly apply to groups that become newly eligible for coverage under the waiver (particularly individuals with incomes above the poverty line), the Centers for Medicare and Medicaid Services (CMS) at the U.S. Department of Health and Human Services (HHS) has been willing to let states modify cost-sharing for those already eligible for benefits, particularly through Health Insurance Flexibility and Accountability (HIFA) waivers. In its HIFA waiver guidance, CMS has not set a limit on total cost-sharing, except for a limit of five percent of family income for children. Federal law establishes criteria concerning when CMS may waive cost-sharing limits in Medicaid, but CMS has not necessarily followed these criteria in approving waivers.

Case 2:03-cv-02506-EHC

6 Document 137-3

Filed 03/10/2008

Page 7 of 22

Some states expect still greater flexibility in cost-sharing to be permitted in the future. Gov. Jeb Bush has proposed that prescription drug copayments in Florida's Medicaid program be increased to as much as $15 for a brand name drug that is not included in the state's preferred drug list. A federal waiver would be needed to implement such a proposal. Last year, Texas proposed requiring copayments for all Medicaid beneficiaries, including children and pregnant women, although that plan was shelved after CMS signaled that this would not be permissible. The federal government has signaled that it is willing to permit expanded cost-sharing in Medicaid. The Medicaid proposal reflected in the Administration's FY 2004 budget proposal would give states almost complete flexibility in establishing policies for "optional" beneficiaries or services, including cost-sharing policies applied to these beneficiaries or services.2 In addition, CMS announced last year that it planned to propose new regulations to lift existing Medicaid copayment limits. If either of these proposals is implemented, Medicaid cost-sharing obligations for low-income beneficiaries could become substantially greater.

Research Shows Cost-sharing Can Jeopardize Health Care for Low-income Families and Individuals Rigorous research has shown that cost-sharing creates serious barriers for low-income individuals' access to health services and, as a result, can adversely affect their health.3 Key findings from this research include:
·

When charged higher copayments or other forms of cost-sharing, low-income adults and children obtain substantially fewer health services and/or prescription drugs. Of particular importance, cost-sharing causes low-income individuals to use fewer essential health services and/or prescription drugs, which can lead to significant health problems. For example, a recent study found that when cost-sharing for low-income adults' prescription drugs was increased, these individuals used emergency rooms more and had more adverse health consequences, such as hospitalization, institutionalization, and even death. The burden of cost-sharing falls most heavily on the sickest individuals. People with disabilities, seniors, and those with chronic health problems use the most health services and medications and therefore are most heavily affected by copayments.

·

·

2

A forthcoming paper will discuss the Administration's Medicaid proposal and cost-sharing in more depth.

Julie Hudman and Molly O'Malley, Health Insurance Premiums and Cost-Sharing: Findings from the Research on Low-Income Populations, Kaiser Commission on Medicaid and the Uninsured, April 2003, provides an excellent review of the research in this area.

3

Case 2:03-cv-02506-EHC

7 Document 137-3

Filed 03/10/2008

Page 8 of 22

·

Charging monthly premiums to join Medicaid or SCHIP discourages eligible individuals from signing up for the program and also causes beneficiaries to drop off the program prematurely. It may also lead to "adverse selection," in which healthier individuals drop coverage while the sicker beneficiaries remain enrolled, which causes medical costs per enrollee to spiral upward.

The RAND Health Insurance Experiment. The RAND study, carried out in the 1970s and 1980s, is one of the most rigorous and important health policy studies ever conducted. Families were randomly assigned to receive health care for free or at various levels of cost-sharing.4 The researchers observed that the level of cost-sharing mattered less than the presence of cost-sharing and, thus, usually grouped all cost-sharing together. The study found that low-income adults made 41 percent fewer medical visits for "more effective" care (services the researchers judged to be more clinically effective in improving health outcomes) when they had to make copayments than when they received free care. Health care utilization also fell among higherincome adults when they had to make copayments, but not nearly as much as among low-income adults, who had less disposable income to pay for health services. Similarly, low-income children received 44 percent fewer clinically effective health care when care was not free, a decline almost three times as great as among higher-income children. (See Figure 1). The RAND study -- and other subsequent studies -- also contradict the claim that costsharing makes health services more efficient by encouraging responsible health care choices. These studies found that while individuals used fewer services when they had to make copayments, they did not become "smarter" health care consumers who stopped obtaining unnecessary care but continued to get effective care. Among low-income and higher-income individuals alike, copayments led to a reduction in both "more effective" and "less effective" care, as defined by clinical criteria. This may reflect the fact that Figure 1. patients typically lack the clinical Cost-Sharing Reduces the Use of Effective understanding to differentiate Medical Care & Affects the Low-income More between more and less essential Percent Reduction in Episodes Low-income people were care. For example, patients might of Effective Care in the lowest third of choose to go to the doctor because income, which is just 44% 41% above 200% of poverty. of coughs, headaches or other 29% discomfort that are symptoms of common colds that will pass in a 15% day but avoid seeking care for chest pains that could signal heart disease. Low-income Not LowLow-income Not Lowincome income

In the end, the RAND study found that copayments harmed the health of low-income individuals. Poor adults with free health care
4

Adults
Source: J. Newhouse, RAND Health Insurance Expt.

Children

Joseph Newhouse, Free For All? Lessons from the Rand Health Insurance Experiment, Cambridge: Harvard University Press, 1996.

Case 2:03-cv-02506-EHC

8 Document 137-3

Filed 03/10/2008

Page 9 of 22

had better blood pressure, better vision, and fewer dental cavities than those with cost-sharing; they also experienced fewer serious symptoms than those with cost-sharing. The results were similar for children: poor children receiving free care were less likely to be anemic and had fewer dental problems than poor children with cost-sharing. By contrast, cost-sharing had no significant effect on the health of higher-income individuals.

Figure 2. Reduction in Use of Essential Drugs Led to More Adverse Medical Events & More Emergency Dept. Visits Among Adults on Welfare
# Events per 10,000 person-months

124

78% increase in adverse events

88% increase in emergency dept. visits

70

28 15
Serious Adverse Events* Before copayments Emergency Dept. Visits After copayments

*Includes hospitalizations,

institutionalizations, and deaths The Quebec Study. A recent study published in the Journal of the American Medical Association examined the consequences of Quebec's 1996 decision to require copayments from 120,000 adult welfare recipients.5 To help protect those with greater medical needs, copayments were capped at $200 per year ($16.67 Canadian per month or about $12 U.S. per month based on currency conversion rates at that time). Even so, the use of "essential" medications by low-income adults fell by about one-seventh after copayments were imposed.

Source: Tamblyn et al. 2001

The Quebec study also tracked the health consequences of the reduction in the use of these medications. It found that emergency room use climbed by 88 percent after copayments were imposed, while adverse health problems (including institutionalization, hospitalization, and death) rose 78 percent (Figure 2). Dr. Robyn Tamblyn, the principal investigator, concluded that "Increased cost-sharing for prescription drugs had the desired effect of reducing the use of less essential drugs, but also the unintended effect of reducing the use of drugs that are essential for disease management and prevention. As a result, in the post-policy period, there was an increase in the rate of adverse events and ED [emergency department] visits related to reductions in essential drug use." He concluded, "More stringent cost-sharing pharmaceutical policies in other parts of Canada and the United States may contribute to avoidable illnesses." Studies of Copayments in Medicaid. Other studies document that the relatively modest levels of copayments that already exist in Medicaid have reduced access to health care. Researchers from the University of Maryland compared the use of prescription drugs among elderly and disabled Medicaid enrollees who live in states that impose copayments with those who live in states without copayments.6 As seen in Figure 3, copayments had a minimal effect
Robyn Tamblyn, et al., "Adverse Events Associated with Prescription Drug Cost-Sharing among Poor and Elderly Persons," Journal of the American Medical Association, 285(4): 421-429, January 2001. In addition to imposing cost-sharing on adult welfare recipients, Quebec increased cost-sharing for higher-income seniors. The effects for seniors were similar to (but somewhat less severe than) the consequences for adult welfare recipients. Bruce Stuart and Christopher Zacker, "Who Bears the Burden of Medicaid Drug Copayment Policies?" Health Affairs, 18(2):201-12, 1999.
6 5

Case 2:03-cv-02506-EHC

9 Document 137-3

Filed 03/10/2008

Page 10 of 22

on patients who were in excellent or very good health but reduced medication use significantly among those in fair or poor health. Those in poorer health use more medications and, thus, must spend a much larger share of their incomes for copayments. Thus, cost-sharing particularly affects those with the greatest health needs.

Figure 3. Medicaid Copayments Reduce Drug Utilization, Affecting Those With Chronic Illnesses The Most
Average Prescriptions Per Year for Elderly & Disabled Enrollees, by State Policy
23.4 17.4 12.5 12.2 21.6 36.0 30.2 28.4

In another example, a survey of Self-reported Health Status Tennessee's Medicaid waiver program States with Copayments States without Copayments (TennCare), which imposed copayments Source: B. Stuart & C. Zacker, 1999. on low-income beneficiaries in 1994, found that 20 percent of those who were assessed copayments could not afford to make the copayment when they visited their doctor's office. Twenty-two percent were unable to make the copayment for their medications at the pharmacist; two-thirds of those who could not afford to pay simply went without the medication as a result.7 Also, a 1970s study found that when California's Medicaid program added a $1 copayment for physician services, it reduced the demand for ambulatory services by 6 percent but increased the use of inpatient hospital services by 17 percent. The net effect was that overall Medicaid program costs appeared to rise, although the increase was not statistically significant.8 Last year, the state of Montana raised Medicaid copayments sharply. Within a few months, the state modified its copayment schedule because of reports of problems being expressed by health care providers and serious health consequences reported by patients.9

Excellent/Very Good

Good

Fair

Poor

Research Shows Premiums Lower Participation and Can Raise Costs per Person Another form of cost-sharing is to charge monthly premiums for participation in Medicaid waiver programs or SCHIP. Urban Institute researchers examined participation in programs that charged low-income individuals sliding-scale premiums for insurance and found
C. Larson, "TennCare and Enrollee Cost-Sharing: A Survey of the Previously Uninsured and Uninsurable Enrollees in Davidson County," survey conducted by the Metropolitan Health Dept. of Nashville and Davidson County, September 1996. J. Helms, J. Newhouse, and C. Phelps, "Copayments and the Demand for Medical Care: The California Medicaid Experience," Bell Journal of Economics, 9:192-209, 1978. One person who recently had a kidney transplant stopped taking his anti-rejection medications, knowing that this would lead to tissue rejection and eventual death, because he could not afford the new charges. Another person was reported to have committed suicide when she learned what her medications would cost. Ericka Smith, "Activists Say Cuts to Program Harm Poor, Disabled," Missoulian, May 18, 2002.
9 8 7

Case 2:03-cv-02506-EHC

10 Document 137-3

Filed 03/10/2008

Page 11 of 22

that the higher the premium, the Figure 4. As Premiums Rise, Fewer Low-income Uninsured lower the participation rate among People Participate eligible, uninsured individuals 10 (Figure 4). Even small premiums 70% 67% 57% discouraged some participation. For 60% example the data indicate that a 45% 50% premium that costs one percent of a 35% 40% family's income (or $13 per month 25% 30% 18% for a family of three at the poverty 20% 12% 8% 5% line) would lower enrollment by 10% about 16 percent. A premium equal 0% to three percent of family income 0 1 2 3 4 5 6 7 8 ($38 per month for a family of three Premiums as % of Income at the poverty line) would reduce Source: Ku and Coughlin, Urban Institute, Inquiry, 1999/2000 enrollment by 49 percent. Some have suggested that it would be acceptable to let cost-sharing reach as high as five percent of a family's income ($64 per month for a family of three at the poverty line); if premiums were set this high, participation would fall by almost three-quarters (74 percent). A study conducted for the Kaiser Commission on the Future of Medicaid yielded similar estimates.11 A study of Oregon's Medicaid waiver program found that when participating parents were asked to pay higher premiums, a substantial number dropped off the program, and the rate of "churning" on and off the program rose by 50 percent. Moreover, when parents left the program, their children often lost coverage as well.12 Similarly, studies of Florida's Healthy Kids program have found that monthly premiums are associated with higher disenrollment rates among children; when the program lowered its premiums, the disenrollment rate declined. Moreover, the researchers found that healthier children were more likely to leave the program, which caused an increase in the average cost of health care services due to adverse selection.13 (In another recent report, we have discussed how
Estimated % Participation Rate

10

Leighton Ku and Teresa Coughlin, "Sliding-Scale Premium Health Insurance Programs: Four States' Experiences," Inquiry 36: 471-480 (Winter 1999-2000). Judith Feder and Larry Levitt, Choices Under the New State Health Insurance Program: What Factors Shape Cost and Coverage? Policy Brief for Kaiser Commission on the Future of Medicaid, 1998.

11

12

Susan Haber, Janet Mitchell, and A. McNeill, "Using Premiums to Finance Care for the Uninsured: Lessons from the Oregon Health Plan" presented at Academy for Health Services Research and Health Policy Annual Meeting, 2000. Elizabeth Shenkman, et al., "Premium Subsidies and `Adverse Retention' in Children's Managed Care," Institute for Child Health Policy, October 1996. Elizabeth Shenkman, et al., "Disenrollment and Re-enrollment Patterns in a SCHIP," Health Care Financing Review, Spring 2002.

13

Case 2:03-cv-02506-EHC

11 Document 137-3

Filed 03/10/2008

Page 12 of 22

monthly premiums can increase disenrollment and churning in Medicaid and SCHIP and how states can minimize these effects.14) Information about the effects of premiums is also available from the experience of statefunded pharmacy assistance programs for seniors. The General Accounting Office reported that enrollment fees or premiums impose barriers to participation in these programs.15 Minnesota found that a $120 annual fee limited participation and decided to eliminate the fee. Similarly, New York's pharmacy assistance program reduced its premiums after determining that participation levels were too low. When Connecticut's program shifted from a one-time fee of $15 to an annual $25 fee, enrollment dropped by half.

Will Higher Cost-sharing Promote Responsibility among Medicaid Beneficiaries and Keep the Program Up-to-date? Some proponents of expanded cost-sharing in Medicaid argue it will foster personal responsibility among Medicaid beneficiaries. Since middle-class individuals with private insurance are being required to pay higher levels of cost-sharing, they contend, low-income individuals on Medicaid should bear equivalent responsibilities. Data show, however, that Medicaid beneficiaries already have substantial out-of-pocket medical care expenditures. On average, Medicaid beneficiaries pay a larger share of their incomes in out-of-pocket medical expenses than do higher-income individuals with private insurance. By this measure, those on Medicaid are already more "responsible" than middle-class people with private insurance. Data from the federal Medical Expenditure Panel Survey for 1999 show that poor adult Medicaid beneficiaries who were not elderly or disabled paid an average of 2.3 percent of their family incomes in out-of-pocket health expenses that year (Table 1). This is more than four times as great a share of their incomes as the out-of-pocket expenses of similar middle-income adults (i.e., of adults with incomes greater than twice the poverty line) who had private health insurance. Prescription drugs impose a particularly heavy burden on Medicaid patients' pocketbooks. Those on Medicaid paid, on average, more than seven times as much of their incomes (1.5 percent) for prescription drugs as higher-income, privately insured adults (0.2 percent of income). While poor Medicaid beneficiaries' out-of-pocket expenditures were lower in dollar terms than those of middle-income adults with private health insurance ($163 per year, compared

14

Leighton Ku and Donna Cohen Ross, Staying Covered: The Importance of Retaining Health Insurance for Lowincome Families, New York: The Commonwealth Fund, December 2002. U.S. General Accounting Office, State Pharmacy Programs: Assistance Designed to Target Coverage and Stretch Budgets, HEHS-00-162, September 2000.

15

Case 2:03-cv-02506-EHC

12 Document 137-3

Filed 03/10/2008

Page 13 of 22

Table 1. Comparison of Average Annual Out-of-Pocket Medical Expenditures for Non-elderly, Non-disabled Adults with Medicaid and Private Insurance In Dollars per Year As Percent of Family Income Adults with Incomes Below the Poverty Line, with Medicaid for Majority of Year Total Out-of-Pocket Expenses Prescription Drugs Primary and Related Care Inpatient Hospital Vision/Dental Care Average Family Income $163 $108 $23 $8 $24 $7,096 2.3% 1.5% 0.3% 0.1% 0.3% 100.0%

Adults with Incomes Above 200 Percent of the Poverty Line, with Private Insurance for Majority of Year Total Out-of-Pocket Expenses Prescription Drugs Primary and Related Care Inpatient Hospital Vision/Dental Care $406 $124 $122 $19 $141 0.5% 0.2% 0.2% 0.0% 0.2% 100.0%

$73,788 Average Family Income Source: 1999 Medical Expenditure Panel Survey, as analyzed by CBPP

to $406 for the latter group), Medicaid beneficiaries' average incomes were only one-tenth as large as those of middle-income adults, so medical expenses consumed more of their incomes.16 Given that Medicaid cost-sharing is usually quite limited, why do Medicaid beneficiaries bear higher proportionate out-of-pocket costs than the privately insured? Three reasons apply:
·

While Medicaid cost-sharing limits are nominal, cost-sharing is common. For example, about two-thirds of the states (31) imposed copayments for prescription drugs in Medicaid in 2001; that number has almost certainly grown since then.17 Most Medicaid beneficiaries have very low incomes, so even small copayments can represent significant shares of their income. A $3 copayment from a person who makes $5,000 a year is equivalent to a $30 copayment from someone making $50,000 per year or a $60 copayment from someone making $100,000. (Moreover, since the poor individual has far less disposable income, the $3 copayment may impose a greater hardship for a poor person than a $30 copayment for someone making ten times as much.) Medicaid beneficiaries sometimes find that their Medicaid coverage will not pay for certain medications or services and must pay the whole cost themselves. For

·

·

16

The survey does not include the cost of over-the-counter drugs.

National Pharmaceutical Council, Pharmaceutical Benefits Under State Medical Assistance Programs, 2001, July 2002.

17

Case 2:03-cv-02506-EHC

13 Document 137-3

Filed 03/10/2008

Page 14 of 22

example, if a pharmacist informs a patient that Medicaid will not pay for a certain medication prescribed by her doctor (because, for example, it is not a preferred drug under the state's Medicaid policies), the patient might pay for it out-ofpocket rather than experience the delays and uncertainties of asking her doctor to change her prescription or seeking special authorization from Medicaid for the original prescription. In such a case, the low-income Medicaid patient pays the entire cost of the drug, not just the Medicaid copayment. Similarly, some state Medicaid programs limit the number of prescription drugs they will cover (e.g., to three or four prescriptions per month), so those who need more medications must pay for the additional prescriptions out-of-pocket. Poor families must stretch their limited incomes to meet competing demands for rent, food, child care and other expenses. Research indicates that, despite the presence of public assistance programs like Medicaid, food stamps and the like, poor families often have difficulty meeting basic needs. A recent study found that about one-third of non-elderly families below the poverty line suffer "critical" hardships such as often or sometimes running out of food, being evicted from their homes in the past 12 months or missing needed medical care in the past year. Three-quarters of such families encounter "serious" hardships such as having their utilities disconnected in the past 12 months or being unable to obtain child care.18 If Medicaid beneficiaries must pay more for medical care or prescription drugs, they will have a harder time meeting other basic needs.

"Keeping Up with the Times" Some contend that Medicaid cost-sharing should be increased to keep up with the times since the limits have not been raised for a number of years and consequently have fallen in inflation-adjusted terms. There are countervailing arguments. As noted above, poor Medicaid patients already pay significantly higher share of their incomes for medical care than middleincome people with private health insurance. Other expenses -- particularly housing and child care costs -- have increased over the past decades, so many poor families may actually have less disposable income to contribute to medical care. According to data from the Census Bureau's American Housing Surveys, the median housing cost for renters with incomes below the poverty line rose from 53 percent of their income in 1989 to 65 percent of their income in 2001.19 Increases like these leave poor individuals with less remaining income for health care. Certain other costs that low-income families bear have risen as well. For example, in almost half the states, families' copayments for subsidized child care rose between 1995 to 2000. Since January 2001, almost half the states have reduced access to child care for low-income
18

Heather Boushey and Bethney Gundersen, When Work Just Isn't Enough: Measuring Hardships Faced by Families Moving from Welfare to Work, Economic Policy Institute, June 2001. This study analyzed national survey data from the Survey of Income and Program Participation and the National Survey of America's Families. U.S. Census Bureau, American Housing Survey for the United States: 2001, Current Housing Report H150/01, October 2002, and prior versions from earlier years.

19

Case 2:03-cv-02506-EHC

14 Document 137-3

Filed 03/10/2008

Page 15 of 22

families by raising copayments or limiting slots, according to the General Accounting Office. 20 Increasing cost-sharing in Medicaid could force many low-income individuals to choose between adequate health care and other basic needs, such as housing and child care.

Will Higher Cost-sharing Encourage Patients to Use Lower-cost Drugs? Much of the debate about cost-sharing concerns whether it can help restrain the cost of prescription drugs, the most rapidly growing area of Medicaid expenditures. State officials are eager to hold prescription drug expenditures down and are interested in shifting patients from more expensive medications -- such as brand-name drugs heavily advertised on television -- to less expensive but effective drugs, particularly generic drugs. A number of states have increased Medicaid drug copayments. One recent trend is the use of "tiered" copayments, in which generic drugs have no copayment or a low copayment level (such as 50 cents or $1) while brand-name drugs or drugs that are not part of a "preferred drug list" have higher copayments, such as $2 or $3. In some Medicaid waiver programs, copayment levels are far higher. For example, Florida's pharmacy assistance waiver program (SilverCare) for seniors with incomes close to the poverty line charges a copayment of $2 for generic drugs, $5 for non-generic drugs on the Medicaid preferred drug list, and $15 for medications not on the preferred drug list. States hope that staggering the copayment levels will steer patients toward lower-cost generic drugs. State Medicaid agencies are turning to these cost-sharing approaches in large measure because they have been used by private insurers. But is the private model applicable to Medicaid? The research cited earlier in this paper demonstrates that setting cost-sharing levels too high can harm the health of low-income individuals. In addition, Medicaid has alternative methods of containing prescription drug costs, some of which are similar to those used by private insurers and some of which are tighter in Medicaid than those used in the private market. As of 2001, two-thirds (32) of state Medicaid programs required pharmacists to substitute generic drugs when their patients were prescribed brand-name drugs for which generic alternatives are available, and 42 states required that certain drugs not be given without prior authorization from the state Medicaid agency.21 A recent survey conducted for the Kaiser Commission on Medicaid and the Uninsured found that 45 states were planning additional efforts to contain drug costs in 2003, so such policies are spreading .22 States have a number of tools to limit prescription drug costs in Medicaid without increasing cost-sharing.23
D. Ewen, et al., State Developments in Child Care, Early Education, and School-Age Care, Children's Defense Fund, 2001. General Accounting Office, Child Care: Recent State Policy Changes Affecting the Availability of Assistance for Low-income Families, GAO-03-588, May 2003.
21 20

National Pharmaceutical Council, op cit. Vernon Smith, et al. op cit.

22

23

Kaiser Commission on Medicaid and the Uninsured, Medicaid and the Prescription Drug Benefit, September 2002.

Case 2:03-cv-02506-EHC

15 Document 137-3

Filed 03/10/2008

Page 16 of 22

One result of Medicaid's existing cost-containment policies is that Medicaid beneficiaries are more likely to use low-cost generic drugs than privately insured individuals are. Researchers at the federal Agency for Healthcare Research and Quality found that 54 percent of Medicaid beneficiaries used generic drugs, compared to 42 percent of the privately insured.24 In some cases, strict Medicaid policies, including copayments and other policies, can make it difficult for Medicaid beneficiaries to obtain needed drugs. A recent study found that 41 percent of Medicaid patients with two or more chronic health conditions reported problems obtaining medications prescribed to them, as have 16 percent of those with no chronic health problems. These rates were about three times higher than those found for privately insured patients.25 While Medicaid patients had better access to prescription drugs than the uninsured, they did not fare as well as the privately insured. Finally, although there is still relatively little research about the impact of tiered copayments, a major study of employer-sponsored health plans suggests they do not cause an increase in the use of generic drugs. A recent RAND study found that, among patients in employer-sponsored plans, tiered copayments lowered health plans' medication costs not by leading patients to switch to generic drugs but rather by shifting costs to patients and by causing patients to use fewer medications overall.26 Generic drugs actually made up a smaller share of all drugs used under a tiered copayment schedule (33 percent) than under the approach that state Medicaid programs typically employ, which uses nominal uniform copayments and requires the use of generic drugs (39 percent). One plausible explanation is that when patients are faced with higher out-of-pocket costs due to higher copayments, they reduce their spending for both generic and brand-name drugs. In sum, tiered copayment systems may lead patients to use fewer drugs in general, which could pose threats to their health. (Note: This study applied to privately insured patients, not those on Medicaid. We are not aware of any comparable Medicaid-specific research.)

The Effects of Cost-sharing on Health Care Providers Some of the strongest complaints against cost-sharing come from health care providers, particularly pharmacists.27 Medicaid payments to providers are reduced by the amount of the
24

G.E. Miller and J.F. Moeller, "Outpatient prescription drug prices and insurance coverage: An analysis of therapeutic drug class and user characteristics from the 1996 Medical Expenditure Panel Survey," in Investing in Health: The Social and Economic Benefits of Health Care Innovation, 2001, pp. 23-57.

25

Peter Cunningham, "Prescription Drug Access: Not Just a Medicare Problem," Issue Brief 51, Center for Studying Health System Change, April 2002.

26

Geoffrey Joyce, et al., "Employer Drug Benefit Plans and Spending on Prescription Drugs," JAMA, 288(14):173339, October 9, 2002.

For example, when the Texas Medicaid agency increased copayments for prescription drugs, the state pharmacists' association brought a lawsuit that has resulted in a temporary restraining order blocking the copayments from being implemented.

27

Case 2:03-cv-02506-EHC

16 Document 137-3

Filed 03/10/2008

Page 17 of 22

copayment, and in cases where patients cannot afford to make the copayment, that cost is effectively shifted to providers since federal rules stipulate that health services may not be denied if a person is unable to afford the copayment. In addition, collecting copayments imposes new administrative costs on providers. Because of concerns like these, the National Association of Counties has warned that substantial increases in Medicaid cost-sharing would increase the uncompensated care burdens of county-funded hospitals and clinics, shifting costs and responsibilities from states to counties.28 If Medicaid beneficiaries drop coverage because of premiums, county and local health facilities will consume county or local funds to provide uncompensated care for many of the same patients, who will now be uninsured. For example, Oregon recently increased premiums and about 15,000 Oregon Health Plan beneficiaries were going to lose coverage due to non-payment. Some local clinics determined it was more cost-effective to use local funds to pay the premiums for a few thousand of their patients to maintain their insurance coverage, reasoning that they would otherwise have to care for many of them as uncompensated care patients.29 There are surprisingly few data on the extent to which Medicaid beneficiaries actually make copayments, how much providers lose, and how often beneficiaries are denied services. Some have claimed that health care providers receive only 50 percent of the copayments they are due, but we are not aware of any good documentation for claims this high. We know of only one published study of this subject, based on a survey of pharmacists in three states conducted by researchers at the University of Maryland. In the survey, most pharmacists said that most Medicaid recipients made their copayments. Half of the pharmacists said they always collected copayments, while only six percent said they had to waive copayments for more than ten percent of their Medicaid patients. In addition, while pharmacists were generally aware that federal rules prevented them from denying services to Medicaid beneficiaries who could not pay, many admitted that they did not waive copayments and either denied the medications or forced patients to pay. 30 This study suggests that the burden of copayments falls primarily upon Medicaid beneficiaries rather than health care providers and that, despite federal rules to the contrary, low-income patients are sometimes denied services when they cannot pay. A study of TennCare found that a small minority of people could not make their prescription drug copayments and two-thirds of those unable to pay failed to receive their medications.31

28

National Association of Counties, "Resolution On HIFA Waiver And Services To Low Income Individuals," Adopted July 16, 2002.

Don Colburn, "10,000 Low-Income Oregonians Will Be Cut From State Health Plan," Oregonian, April 30, 2003 and personal communication with Jim Edge of the Office of Medical Assistance Programs, May 1, 2003.
30

29

Cheryl Fahlman, Bruce Stuart, and Christopher Zacker, "Community pharmacist knowledge and behavior in collecting drug copayments from Medicaid recipients," American Journal of Health System Pharmacies. 58(5):38995, March 2001. C. Larson, op cit.

31

Case 2:03-cv-02506-EHC

17 Document 137-3

Filed 03/10/2008

Page 18 of 22

Poor and Sick Individuals Would Bear Most of the Burden of Cost-Sharing; Children and Pregnant Women Could Be Harmed Some supporters of increased cost-sharing in Medicaid have argued that the program has evolved over time and a large number beneficiaries now have incomes above the poverty line. Thus, it is argued, Medicaid beneficiaries can afford copayments or premiums. It is important to realize, however, that aside from children and pregnant women, relatively few Medicaid beneficiaries have incomes above the poverty line. A modest number of states expanded Medicaid eligibility for low-income parents, seniors or people with disabilities, but the number of persons enrolled with these higher incomes remains relatively small. Moreover, most of the states that have expanded eligibility beyond the poverty line have done so using Section 1115 waivers and those waivers already permit higher cost-sharing for individuals with incomes above the poverty line.32 The most serious public health threats posed by Medicaid cost-sharing exist for seniors and people with disabilities, the great bulk of whom have incomes below the Figure 5. poverty line. Because these groups Medicaid Prescription Drug Use Highest Among Elderly have more serious health conditions, and Disabled; They Are Most Affected by Copayments they tend to use far more medical care than children or non-elderly adults. In Average Prescriptions 31.0 29.4 Per Year for turn, they would have the highest out25.8 25.5 Medicaid Beneficiaries of-pocket health expenditures when being assessed copayments. As Figure 15.7 14.4 5 shows, the average number of 10.8 9.0 prescriptions filled among Medicaid 4.4 2.5 1.8 2.9 beneficiaries is far higher for the elderly and for those with the poorest CHILDREN ADULTS ELDERLY health (e.g., people with chronic health Self-reported Health Status Source: 1997 Medical problems or disabilities). For example, Expenditure Panel Survey, Overall Exc./V. Good Good Fair/Poor CBPP analysis adults in fair or poor health use seven times as many prescriptions as adults in excellent or very good health and thus would be charged about seven times as much in copayments. An increase in copayment rates could jeopardize their use of health services or medications. Currently, federal law exempts pregnant women and children from being charged copayments. If these exemptions are cancelled -- as the state of Texas proposed last year -- that could reduce the use of preventive and primary health care services and endanger the long-term health of infants or children. Copayments could cause pregnant women to begin receiving
Only two states (Maine and the District of Columbia) have expanded Medicaid eligibility for parents with incomes above 100 percent of the poverty line without using Section 1115 waivers. (Connecticut had such a program but has acted to eliminate coverage for parents with incomes above the poverty line, although a court order has temporarily delayed implementation of the change.) Only one state -- Mississippi -- has expanded coverage for seniors or the disabled with incomes above the poverty line without using a waiver.
32

Case 2:03-cv-02506-EHC

18 Document 137-3

Filed 03/10/2008

Page 19 of 22

prenatal care later or to receive fewer prenatal visits on average, or could reduce the likelihood that children are immunized or screened for developmental health problems.

State Savings through Cost-Sharing Will Be Less than the Costs Borne by Poor Residents A paradox in cost-sharing in Medicaid is that it shifts substantial costs onto some of the poorest of state residents, but the state itself would not recoup the full amount in Medicaid savings. If a state charges a copayment or a premium, it reduces the total amount of Medicaid expenditures, which is the basis for computing the federal matching share in Medicaid. The actual state savings will always be half or less of the total copayments or premiums. Consider an example for a state that has a 60 percent federal matching rate. If Medicaid pays $60 for a prescription drug, the state's share of the cost is $24 (40 percent of $60) while the federal share is $36 (60 percent of $60). When a $10 copayment is imposed, the patient pays $10 and the net cost to Medicaid is $50. The state now pays $20 and the federal government pays $30. Even though a low-income state resident is now paying $10 more, the state government saves only $4. The situation would be similar if a premium was imposed instead.33 Copayments and premiums can be viewed as being like taxes or user fees imposed selectively on some of the poorest individuals in a state. But only a fraction of those taxes or user fees collected actually benefit the state government, and at least half of the savings are passed on to the federal government. In contrast, if a state increased taxes or user fees (not involving Medicaid) on higher-income people, the costs would be borne by those who are better able to afford the additional costs than poor Medicaid beneficiaries, and the state government would gain the full use of those additional revenues.

Conclusions The analyses and literature review provided in this paper demonstrate that:
·

Medicaid beneficiaries already shoulder substantial out-of-pocket medical expenses and contribute a larger share of their incomes to meet these expenses than middle-class, privately insured individuals. Research shows that when copayments are increased, low-income individuals respond by reducing their use of effective health services and medications, which can impair their health and lead to adverse consequences such as greater use of emergency rooms.

·

33

States could have some additional savings through reductions in health care utilization or program participation, but at least half of such savings would accrue to the federal government, not to the states. Almost all of the state and federal savings would also be felt as reductions in Medicaid revenue paid to health care providers in the state. Some of the savings would be offset by increased uncompensated care burdens at public clinics and hospitals.

Case 2:03-cv-02506-EHC

19 Document 137-3

Filed 03/10/2008

Page 20 of 22

·

Charging premiums to participate in Medicaid would lower participation, increase disenrollment rates, and potentially create "adverse selection," which would cause costs per enrollee to rise. State Medicaid programs have other ways to contain the costs of prescription drugs without asking patients to pay more, such as requiring the use of generic drugs when they are appropriate substitutes for other, more expensive medications. Medicaid patients are already more likely to use generic drugs than those with private insurance. The burden of increased Medicaid cost-sharing would fall primarily upon individuals whose incomes are below the poverty line, especially seniors and people with disabilities, because most of those who can be asked to provide copayments under current federal law have incomes below the poverty line. Although cost-sharing is equivalent to a tax or user fee imposed selectively on low-income state residents, the actual savings received by the state government constitute half or less of the amounts collected. The majority of savings accrue to the federal government. In net, resources are drawn away from the state.

·

·

·

"Nominal" cost-sharing is already common in state Medicaid programs -- particularly for prescription drugs -- and federal policies permit somewhat higher levels of cost-sharing for higher-income beneficiaries in SCHIP and in Medicaid waiver programs. Nonetheless, as state policy makers consider elevating cost-sharing burdens in Medicaid, they should consider two principles that could be applied to help mitigate the adverse consequences. First, policy officials should avoid increasing cost-sharing for beneficiaries whose incomes are below the poverty line. This would protect those with the lowest incomes, who are least able to pay larger out-of-pocket costs for health care. It is especially counterproductive to increase cost-sharing for Medicaid beneficiaries who are receiving cash assistance under TANF or SSI. In such cases, higher cost-sharing is tantamount to back-door benefit reductions. Yet while TANF or SSI benefit reductions would be partly offset by increases in food stamp benefits (because the amount of food stamp benefits is based on a recipient's cash income and rises when welfare benefits are reduced), Medicaid costsharing simply reduces the beneficiary's resources without any offsetting increase in benefits.34 State officials should also be cautious in considering cost-sharing for low-income people whose incomes are close to the poverty line. For example, SCHIP regulations, issued in 2001,
34

Higher Medicaid cost-sharing could result in slightly higher food stamp benefits because of the Food Stamp Program's excess medical deduction, but this would apply in only a small number of cases. This deduction is available only for the elderly and disabled and does not aid non-elderly adults or children. It only covers medical expenses to the extent that they exceed $35 per month. Because documenting these expenses is difficult for both recipients and caseworkers, only a tiny percentage of food stamp households claim this deduction. In fiscal year 2002, the most recent year for which data are available, only 4.1 percent of food stamp households received any medical deduction at all.

Case 2:03-cv-02506-EHC

20 Document 137-3

Filed 03/10/2008

Page 21 of 22

permit copayments as high as $5 for children whose incomes are below 150 percent of the poverty line. Since Medicaid covers children under the age of six with incomes up to 133 percent of the poverty line and children six to nineteen with incomes up to 100 percent of the poverty line, the SCHIP policy essentially establishes a maximum $5 copayment for those with incomes between 100 percent (or 133 percent) of the poverty line and 150 percent. While this limit is higher than Medicaid's $3 cap on copayments, it continues to assure protection for beneficiaries with incomes near the poverty line.35 In addition, even when SCHIP families have incomes that exceed 150 percent of the poverty line, the law limits total cost-sharing to five percent of family income. Second, to protect those who are the sickest, maximum monthly copayment obligations should be capped. For example, when the Texas Medicaid program recently proposed to increase copayments for prescription drugs, it also proposed capping copayments at $8 per month to avoid harming those with chronic diseases who require multiple medications. It should be noted that the Quebec study cited earlier found serious adverse effects from copayments for prescription drugs even though the level of drug copayments was capped at less than $17 Canadian per month or about $12 in U.S. dollars. States that establish monthly caps need to establish mechanisms to inform patients and health care providers when the caps have been reached. So