Testimony of NHeLP's Mara Youdelman Before the Congressional Asian Pacific American Caucus

Thank you, members of the Caucus, for holding this important hearing. As a staff attorney working on civil rights issues, and particularly language access issues, I welcome the opportunity to speak with you this afternoon on behalf of the National Health Law Program. I look forward to discussing measures for the Caucus and Congress to improve language services.

As my organization focuses on health care issues primarily affecting individuals of limited incomes, my testimony will concentrate on issues of language access in the public health programs. We believe special attention should focus on language services in health care for individuals with limited English proficiency (LEP). [1] These services promote access to and quality of care in health care, which is a crucial – and sometimes life-or-death – issue, affecting everyone throughout their lives.

We would ask the Caucus to consider three issues:

1. Increasing the federal "match" for language services provided to Medicaid and State Children's Health Insurance Program enrollees to 90%.

2. Authorizing and appropriating funds for HHS to establish a national toll-free 24-hour-a-day "language line" so that all CMS grantees and providers can access interpreter services for their patients/clients, including providers offering services to Medicaid, Medicare and State Children's Health Insurance enrollees.

3. Mandating the collection of racial, ethnic and primary language data throughout HHS health programs to assist in identification of and targeted assistance to address disparities in access to and quality of health care.

Background

The need for language services is well-documented, most recently by the 2000 Census. Over 21 million Americans speak English less than "very well." And 22.5 % of Asian and Pacific Islander speakers speak English "not well" or "not at all." Much research has documented the health disparities in access, treatment and outcomes that can arise due not only to race and ethnicity but also language. As documented in the Institute of Medicine's (IOM) recent report entitled "Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care", [2] language barriers pose a problem for patients where health systems lack the resources, knowledge or institutional priority to provide interpretation and translation services. According to the IOM, "[l]anguage barriers may affect the delivery of adequate care through the poor exchange of information, poor shared decision making, or ethnical compromises (e.g., difficulty in obtaining informed consent [citation omitted])." The report recommended supporting the use of interpretation services when community needs exist.

Further, a recent report from The Access Project compared the experiences of uninsured individuals receiving care at hospitals across the country in three groups: 1. those who did not need interpreters; 2. those who needed and received interpreters; and 3. those who needed but did not receive interpreters. The survey found that 27% of those who needed but did not get an interpreter said they did not understand the instructions for taking prescribed medications, compared to only 2% of those who either got an interpreter or did not need one. The report also documents that individuals needing but not receiving an interpreter were less likely to be offered information about paying for their medical care, were more likely to say they would not seek care at a particular facility because of their health care debt, and would not use the hospital even if they became insured. [3]

And a recent study documented the potential adverse consequences of using untrained interpreters. In pediatric emergency rooms, researchers evaluated usage of professionally-trained interpreters, "ad hoc" interpreters (e.g. friends, relatives, children, untrained staff or strangers from the waiting room) and no interpreter. Overall, eighteen percent of interpretation errors had potential adverse medical consequences for patients. For example, in one case a family thought they should give their child two tablespoons instead of two teaspoons of a medication and they failed to understand that a second medication was to be given. But the rate of errors of potential medical consequence was significantly lower for those encounters using a professional hospital interpreter – about 12 percent – as compared to those using an ad hoc interpreter – 22 percent – and those using no interpreter – 20 percent. [4]

Federal Funding for Language Access is Available but Underutilized

Given the documented need, one issue that often hinders the provision of language access is funding. The issue of paying for language services is cited most often by providers as the reason they are unable to hire interpreters or translate written materials. However, the costs for ensuring linguistic access need not be prohibitive – a recent report from the Office of Management and Budget estimates that language services would cost only an extra .5% of the average cost per visit. ^[5] Moreover, significant federal funding is available for language services in health care. HHS currently offers funding primarily through three sources – Bilingual/Bicultural Demonstration Project Grants through the Office of Minority Health, funding through the HIV/AIDS Bureau of the Health Resources and Services Administration (HRSA), and allowance of federal matching funds for language services provided to Medicaid and State Children's Health Insurance Program (SCHIP) enrollees. Additionally, funding through the Office of Refugee Resettlement may be available for services provided to refugees and asylees.

In a letter to Medicaid directors issued on August 31, 2000, CMS reminded states of the availability of federal matching funds for language services provided as either an administrative or covered expense to Medicaid and SCHIP enrollees. [6] Currently, however, only 9 states are drawing down available federal funds to pay for the costs of language services. Because of both a lack of knowledge and recent budget crises, only one state – Pennsylvania – is expected to begin using Medicaid and SCHIP funds to pay providers for the costs of language services.

States can obtain federal Medicaid/SCHIP funding for language services as a "covered service" (e.g. for interpretation services provided in conjunction with a Medicaid covered service) and obtain federal matching funds according to the states' FMAP. States can also seek federal funding for language services as an administrative expense (e.g. for interpretation services for application assistance) and receive 50% reimbursement. We believe that if Congress increased the federal share of cost to 90%, more states would provide language services through Medicaid/SCHIP, more providers would hire interpreters and translate materials, and both access and quality of care would improve for LEP individuals.

States already expect that providers will provide language services and the contract between the state and providers and managed care organizations requires compliance with Title VI but rarely pay the providers' full costs. Even without directly reimbursing for language services, however, many states are likely subsidizing them indirectly by paying for operating (including administrative and overhead) expenses of health providers. But including the costs of language services, for example, in the provider's payment rate, makes it unlikely that theactual costs of hiring an interpreter are available to the provider (since the payment rate itself often does not even cover the provider's actual cost of providing care). Thus, the provider is put in a bind between hiring interpreters and losing money or failing to provide language services and potentially being found in violation of Title VI. When deciding whether to hire an interpreter without reimbursement, a provider must balance the additional cost above the Medicaid/SCHIP reimbursement rate with funding needed for other expenses such as rent, equipment, or staff salaries. The result is that many providers do not hire interpreters or use untrained interpreters such as family members, friends and even minors. Additionally, clinics and other providers supported by states may provide language services – either hiring interpreters or having staff interpreters divert from their job responsibilities to informally serve as interpreters – without direct funding to do so. Either way, the quality of care to patients is sacrificed and medical errors may arise. If Congress provided increased funding to the states, the state could subsidize a greater share of the costs of ensuring language access and more providers will likely meet the expectations of federal law to provide language access.

Currently, we have identified nine states that offer direct reimbursement for the costs of language interpreters. [7] These are Idaho, Hawaii, Maine, Massachusetts, Minnesota, Montana, New Hampshire, Utah and Washington. While how they provide reimbursement varies, each state has recognized the ability of leveraging state funds to draw down additional federal funds to ensure language access. Each state has flexibility to design its own reimbursement mechanism and set a reimbursement rate, currently ranging from $7 to $50 per hour. Three states contract with language agencies to hire and coordinate interpreters. Three states allow providers to screen and hire interpreters and then reimburse providers for their outlays. Two states reimburse interpreters directly; New Hampshire requires interpreters to enroll as Medicaid providers. The varied mechanisms illustrate the flexibility each state has to design a reimbursement mechanism that fits its needs, budget and resources. The costs vary but each state draws down a minimum of 50% of its costs from the federal government through Medicaid and the State Children's Health Insurance Program (although states' administrative expenses in SCHIP are limited to 10% of their allotment). An outline of the existing reimbursement mechanisms follows.

We are able to provide specific information about any state if requested and have materials in a recently released Language Services Action Kit that explain the Medicaid reimbursement mechanism and existing state models.

Language Access is the Law

While federal funding is available, one question that may arise is why states should offer direct reimbursement since providers already have an independent duty under Title VI to provide language services. First, the state is already expected under federal law to ensure meaningful access to enrollees of Medicaid and SCHIP. Thus, in its direct operation of the program – eligibility determinations, communications with enrollees, fair hearings, etc. – states should ensure the availability of interpreters to comply with Title VI. In addition, to the extent that states requires providers – through their Medicaid contracts – to comply with Title VI, offering reimbursement for language services to providers will assist them in complying with federal law as well as demonstrate a state's commitment to ensuring the access of its residents to government funded programs.

As additional assistance for states and providers, particularly small providers with fewer resources, we suggest that HHS establish a national 24-hour-a-day toll free telephonic interpretation line that can be accessed by all providers who are unable – because of either resource constraints or available interpreters (such as in rural areas) – to utilize face-to-face interpreters. State benefit agencies could also access this service to assist LEP individuals apply for benefits, understand notices sent by state agencies, and understand their legal rights. States should still be encouraged to provide direct reimbursement to providers for in-person interpretation and other language services such as translation of written materials and signage.

While the issue of language services has garnered national attention over the past two years because of recent developments, including an Executive Order signed by President Clinton and reaffirmed by President Bush, the underlying requirements to ensure meaningful access to individuals who have no or limited English proficiency (LEP) arise from Title VI of the Civil Rights Act of 1964. [14] This federal law prohibits discrimination by any entity that receives federal financial assistance on the basis of race, color or national origin. In a 1965 case, Lau v. Nichols, [15] the Supreme Court determined that language is incorporated in national origin and thus recipients of federal financial assistance must also ensure individuals who are LEP have meaningful access as well. Thus, Title VI applies to both state agencies that receive financial assistance – such as the Medicaid agency – as well as to individual health care providers who receive funds from HHS such as Medicaid, SCHIP or research funds.

Thus, since 1964, every state has had an affirmative duty to ensure that LEP individuals have meaningful access to federally funded programs. Each state should have interpreters available to assist LEP individuals in applying for Medicaid and SCHIP. In addition, providers who offer health care services to LEP individuals should, depending on certain circumstances, also have interpreters available. The Department of Health and Human Services (HHS) Office for Civil Rights has issued guidance on how recipients of HHS funds should offer language services. The guidance outlines certain factors – including the size of the covered entity, size of eligible population, nature of program/service, program objectives, resources (including costs), and frequency of encounters (both discrete languages and LEP individuals) – that determine the extent to which recipients of federal financial assistance should offer language services. [16]

Language Access Improves Quality of Care and May Decrease Costs

In addition to the legal mandate to ensure meaningful language access, the goal of providing quality health care should also guide efforts to ensure language access for LEP individuals. Without effective communication between patient and provider, a number of issues can arise – misdiagnosis, unnecessary diagnostic testing, lack of understanding of treatment and prescription regimens, failure to obtain informed consent, and potential liability issues.

Further, individuals with unmet healthcare needs are more prone to suffering exacerbated health problems that require costly and avoidable emergency treatments which the city may have to pay for later if interpreters are not provided now. The unmet needs of individuals, in the aggregate, result in a health care system that is burdened by an increasing number of people with complex or multiple B and often preventable B chronic conditions. [17] This creates a strain on personnel, resources, and financing. According to the Robert Wood Johnson Foundation, the unmet needs of individuals with chronic conditions lead to exacerbated health problems, costly treatments, and unnecessary pain and suffering. [18]

The lack of access to care takes a serious toll on one=s health. [19] Individuals with reduced access to health care over many years, and in some cases for even relatively short periods, suffer deficiencies in their health as compared to individuals with regular access. Data on this issue is most prevalent when analyzing the lack of access of uninsured individuals. Their lack of access mirrors that of LEP individuals (both insured and uninsured) because neither group can access needed care B the uninsured because of barriers created by a lack of insurance and LEP individuals because of language barriers. For example, the uninsured are hospitalized 50%-70% more than the insured for Aavoidable hospital conditions,@ such as pneumonia, and are more than twice as likely to be hospitalized for conditions which should be treated on an outpatient basis, like diabetes and malignant hypertension. [20]

Further, the direct reimbursement of interpreters and a national language line for health care providers offers a method of providing more cost-efficient and cost-effective care. Many diseases require early treatment to alleviate further complications and costs. By providing interpreters to LEP individuals diagnosed with these diseases, the result can be a better quality of life as well as the potential for more cost-effective – and higher quality – health care. For example, early detection of cancer is crucial to obtaining life-saving treatment. Treatment of late-stage cancers yields a lower rate of remission and recovery than early intervention. And the costs of terminal care significantly exceed initial diagnosis and treatment costs. [21] For individuals diagnosed with end-stage renal disease, dialysis treatments can enable them to maintain normal functioning and hold most symptoms at bay. Without dialysis, they will suffer recurrences of renal failure, or develop hypertension, which can lead to strokes or cerebral bleeding. Strokes or cerebral bleeding in turn induce a deterioration of health status, including the possibility of paralysis or loss of functional ability B conditions which would not have resulted had an individual received sufficient interpretation to explain the need for ongoing utilization of preventive care. In the absence of such interpretation, the costs of treating the complications often outweigh the costs of providing an interpreter.

In addition, a study conducted at the Boston Medical Center documents that the use of trained medical interpreters impacts Emergency Department (ED) services and reduces charges. [22] Individuals who needed but did not receive interpretation returned to the ED more frequently and visited out-patient clinics less than those who obtained interpretation. In addition, these individuals received the fewest ED services and spent the least time in the ED, indicating that much of their treatment did not need ED attention but could have been addressed in clinics. Overall, the use of trained medical interpreters can increase the appropriate use of clinics, decrease expensive repeat ED visits, decrease the cost of care, and decrease disparities between English and non-English speakers in intensity of medical care received. [23]

For individuals whose language barriers create an inability to understand their diagnosis or treatment, serious health consequences can result. For example, a Russian speaking patient profiled by the PBS program Healthweek[24] was diagnosed with diabetes. Because the doctor did not ensure the patient understood his diagnosis and treatment, the patient left without knowing he had to change his diet to avoid further complications. A few months later, with his blood sugar levels dangerously high, and suffering bouts of dizziness and weakness, the patient returned to the doctor. With family members interpreting, he was finally able to understand his diagnosis. If he had not received treatment, he faced life-threatening complications B diabetes is the leading cause of new cases of blindness in adults age 20-74, of end-stage chronic irreversible kidney disease, and of lower-extremity amputations (not related to injury). [25] According to the Centers for Disease Control and Prevention (CDC), many of the complications from diabetes can be prevented with early detection, improved delivery of care, and better education on diabetes self-management. The CDC estimates that at least half of the new cases of diabetes-related kidney failure, half of all lower extremity amputations and 90% of blindness could be prevented each year with early detection and treatment. [26] The LEP patient=s risk of complications is easily remedied B and further costs avoided B if interpretation is provided to explain the diagnosis and treatment.

The costs of non-compliance also include those associated with legal suits and malpractice claims brought against providers and others who fail to ensure a patient understands necessary treatment. The title to an article inMedical Economics says it all – "A misinterpreted word worth $71 million." Doctors and paramedics in Florida misdiagnosed a Spanish-speaking patient's stroke, resulting in him becoming a paraplegic. Crucial to the case was the use of the Spanish work "intoxicado" – without an interpreter, the doctors and paramedics believed the patient's mother and girlfriend meant the patient was "intoxicated" rather than their intended meaning of "nauseous."

In another striking example from California, a 51 year-old mother of seven came to America as a refugee from Laos. She was diagnosed with tuberculosis. Her disease, which was not contagious, was determined to be a drug-resistant form that required long-term therapy. But Mrs. Souvannarath ceased taking the medication because the side effects led her to believe the drugs were going to kill her. County health officials B one Hmong worker and one Thai, neither of whom spoke her language B could not explain the need for treatment and a Lao interpreter was never provided. County health authorities jailed Mrs. Souvannarath for failing to take her medications. At the jail, a Hmong officer who spoke no Lao misinterpreted for her. She thought he asked if she was afraid of dying so she said Ayes@. The actual question was if she was thinking of killing herself, and her Ayes@ answer led the jail to put her on a suicide watch. She was jailed for ten months without a proper court order. Attorneys representing Mrs. Souvannarath subsequently filed a federal complaint for damages and equitable relief that was ultimately settled for $1.2 million. [27] The trauma to Mrs. Souvannarath and her family, as well as the resulting payment to her of $1.2 million, could have been easily avoided had an interpreter been provided.

Data Collection

In addition to the specific funding needs for language services, we also urge the Caucus to take a more prominent role in increasing collection of racial, ethnic and primary language data. Collection, reporting and analysis of this data are crucial to identifying and ultimately eradicating health disparities. Yet despite recognition of the need for increased data, HHS has missed significant opportunities to improve data collection. In a report NHeLP authored with the Summit Health Institute for Research and Education, we laid out a number of recommendations to improve data collection.

The main conclusion of our report is that collection of racial, ethnic and primary language data is legal and in fact encouraged by Title VI. Further, in work NHeLP is currently contracted to do for the Office of Minority Health, we have surveyed the 50 states and Washington DC to identify data collection practices. Only four states have limitations on collection of this data and all those state limitations apply to pre-enrollment collection and often in the individual or group market.

Overall, we found significant evidence at both the federal and state level pointing to the legality of data collection. Yet this data is not uniformly and comprehensively collected. For example, states are supposed to report race and ethnicity data to CMS for Medicaid enrollees. But each state can use its own race and ethnicity categories, making state-by-state and even national analysis extremely complicated. And states often have differing data for SCHIP and Medicaid enrollees – SCHIP requires the usage of governmental standards set by OMB for race and ethnicity categories while Medicaid does not. Further, Medicare race and ethnicity data is notoriously unreliable except for the categories of "White" and "Black". This is due in large part to the fact that CMS relies on the Social Security Administration for this data. SSA does not routinely collect race, ethnicity or primary language data.

With the significant federal spending on Medicaid, SCHIP and Medicare, we believe Congress has a responsibility to ensure disparities in care based on race, ethnicity and language do not exist. Congressional oversight of these programs should be heightened to ensure equal treatment for all. But only by improving data collection will the data be available to analyze access, treatment and outcome data to document any existing health disparities.

Conclusion

We strongly support efforts to increase the federal share of language services provided in Medicaid and SCHIP to 90%, establish a national toll free 24-hour-a-day language line, and improve collection of racial, ethnic and primary language data.

We applaud this committee for bringing attention to the needs of LEP individuals in accessing government services. At the same time that the Caucus recognizes a lack of access to government services, a dire need continues to grow in access to health care services. We urge the Caucus to address the needs of language services in health care where a misdiagnosis can result in the ultimate affront to an LEP individual's lack of access – permanent disability and even death. We have the means to ensure language access and federal funding is available. We need only demonstrate our desire to do so.

For the purposes of my testimony, references to "LEP" include both non-English and limited English proficient individuals.

Smedley, Brian D., Stith, Adrienne Y., and Nelson, Alan R., Editors, Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care, Institute of Medicine, National Academy Press (2002).

Andrulis, Dennis, Goodman< Nanette, and Pryor, Carol, What a Difference an Interpreter Can Make: Health Care Experiences of Uninsured with Limited English Proficiency (April 2002).

Glen Flores, et. al, Errors in Medical Interpretation and their Potential Clinical Consequences in Pediatric Encounters, Pediatr. 111-6-14 (2003).

This was based on the total number and average cost of ER visits, inpatient hospital visits, outpatient physician visits, and dental visits. Office of Management and Budget, Report To Congress, Assessment of the Total Benefits and Costs of Implementing Executive Order No.13166: Improving Access to Services for Persons with Limited English Proficiency, (March 14, 2002)(hereinafter OMB Report) available at

Dear State Health Official letter, August 31, 2000, available at http://cms.hhs.gov/states/letters/smd83100.asp.

The nine states have been identified as of July 9, 2003. NHeLP is conducting a 50 state survey of policies for interpreter reimbursement; additional states may be identified through that survey.

States can draw down Medicaid/SCHIP funding in two ways – as a "covered service" (paying for the cost of a service, such as a doctor's office visit or a hospital stay) or as an "administrative expense" (paying for the costs of administering the program). For information see How Can States Get Federal Funds to Help Pay for Language Services for Medicaid and SCHIP Enrollees? in this Action Kit.

For "covered services", the federal reimbursement rate varies from 50-85%, based on the state's per capita income. For "administrative" expenses, every state receives 50% of its costs from the federal government.

"Fee-for-service" generally refers to services not provided through a hospital, managed care organization, or community health center. Providers agree to accept a state-set "fee" for the specific "service" provided.

Language agencies are organizations that contract with and schedule interpreters. They may also oversee assessment and/or training.

Each hospital or psychiatric facilities' amount is based on a percentage of the difference between the qualifying entity's total Medicaid costs and total Medicaid payments from any source.

Providers who have staff interpreters can not submit for reimbursement.

42 U.S.C. § 4000d, 4000d-4a; see also HHS regulations, 45 C.F.R. Part 80.

414 U.S. 563 (1974).

See Policy Guidance on the Title VI Prohibition Against National Origin Discrimination As It Affects Persons With Limited English Proficiency, Office for Civil Rights, Department of Health and Human Services, Notice of Republication, 67 Federal Register 4968 (February 1, 2002).

Numerous medical studies point to the ability of regular care to lessen overall health care costs. One such study documents the reduced costs, and hospital readmissions, that result from providing comprehensive discharge planning and home care intervention for at-risk hospitalized elders. Naylor, M, et al., Comprehensive Discharge Planning and Home Follow-Up of Hospitalized Elders, Journal of the American Medical Association, Vol. 281, No. 7, 613, 617 (February 17, 1999). Post-discharge assistance has also been documented to reduce unplanned admissions, days of hospitalization upon admission, and poor long-term outcomes in patients with high-risk congestive heart failure. Simon Stewart, et al., Prolonged Beneficial Effects of a Home-Based Intervention on Unplanned Readmissions and Mortality Among Patients with Congestive Heart Failure, Arch. Intern Med., Vol. 159, 257 (February 8, 1999).

The Institute for Health and Aging, University of California, San Francisco, on behalf of the Robert Wood Johnson Foundation, Chronic Care in America: A 21^st Century Challenge, www.rwjf.org/library/chrcare/p3pg27.htm at 1 (November 1996).

John Holahan et al., Urban Institute, Health Policy for Low-Income People in New York, http://newfederalism.org/html/NYhealth.html at 7 (November, 1998).

Hoffman, Catherine & Schlobohm, Alan, The Kaiser Family Foundation Commission on Medicaid and the Uninsured, Uninsured in America: A Chart Book, 2^nd Edition (May 2000) at 56.

Herbert Schuette, et al., The Costs of Cancer Care in the United States: Implications for Action, Oncology, Vol. 9, No. 11, http://Intouch.CancerNetwork.com/journals/oncology/ qol1195b.htm at 6, (November 1995).

Bernstein, J. et. al., The Use of Trained Interpreters Affects Emergency Department Services, Reduces Charges, and Improves Follow-Up, Boston Medical Center, Boston University School of Medicine and School of Public Health.

Id.

See PBS Healthweek, Program No. 506, Medical Interpreters, http://www.pbs.org/healthweek/featurep1_506.htm.

National Center for Chronic Disease Prevention and Health Promotion, Diabetes: A Serious Public Health Problem, At-A-Glance 2000, http://www.cdc.gov/diabetes/pubs/glance.htm at 3-4.

Id. The CDC estimates that adequate care could prevent 6-12,000 cases of blindness each year, 16,500 cases of kidney failure, and 43,000 amputations.

See, The Los Angeles Times (May 31, 1999); The Fresno Bee, Woman jailed for TB will get $1.2m (April 5, 2001).