July 2008

Grant Results


From 1998 to 2004, researchers at the Johns Hopkins University conducted a study of unmet needs for health and support services and barriers to accessing such services among older Medicare beneficiaries who also receive support from Medicaid ("dual enrollees").

In a survey of 2,128 dual enrollees in six states, the researchers at the university's Bloomberg School of Public Health found that 25 percent of all hospitalizations within one year were preventable for dual enrollees.

They also found that preventable hospitalizations were more common among dual enrollees with five or more chronic conditions.

Key Findings

  • About half (49.8 percent) of older Medicare beneficiaries with incomes at or below 100 percent of poverty were enrolled in Medicaid in 1996.
  • Dual enrollment of older people in both Medicaid and Medicare did not in itself appear to increase service use.
  • State policies related to spending for home and community-based services significantly influenced Medicaid enrollment among low-income, community-dwelling elderly persons.

The Robert Wood Johnson Foundation (RWJF) supported this project with two grants totaling $913,755 between May 1998 and July 2004.

 See Grant Detail & Contact Information
 Back to the Table of Contents


Medicaid has traditionally been an important source of supplementary insurance coverage for Medicare beneficiaries age 65 and older who are living in poverty. Jointly, these two programs provide older beneficiaries with a comprehensive set of acute and long-term care benefits, including physician visits, hospitalization, nursing home care, prescription drugs and home and community-based services (ranging from skilled nursing and physical therapy to help with activities of daily living such as bathing and dressing).

Older adults who are enrolled in Medicaid and Medicare, known as "dual enrollees," are vulnerable for both economic and health reasons. They are much more likely than other Medicare beneficiaries to be chronically ill or in poor health.

Although the comprehensive coverage afforded to dual enrollees has substantially reduced financial barriers to health care services, studies conducted by the federal Health Care Financing Administration (now called the Centers for Medicaid & Medicare Services [CMS]) and other researchers indicate a pattern of reduced or inappropriate use of health care services, suggesting that nonfinancial barriers may persist.

To reduce these barriers and improve the delivery of medical and supportive services, researchers need to gather more information about dual enrollees' characteristics, care needs and satisfaction with care received.

 Back to the Table of Contents


A research team at the Johns Hopkins University Bloomberg School of Public Health studied the unmet needs and barriers to accessing health and support services among older Medicare beneficiaries who also receive support from Medicaid (dual enrollees).

Under RWJF grant ID# 033667, the research team conducted a six-state survey of older dual enrollees to gather information on access and barriers to acute and long-term care and unmet care needs. The six states, which represented a range of spending levels for Medicare and Medicaid long-term care, were Georgia, Iowa, Massachusetts, New Jersey, Washington and Wisconsin.

The researchers identified 2,777 dual enrollees who were age 65 or older and lived in the community (residents of nursing homes were ineligible for the study). Through a combination of telephone (79 percent) and in-person (21 percent) interviews, 2,128 dual enrollees completed the questionnaire (for a response rate of 76.6 percent). Respondents provided information about their health status, unmet needs for care, care availability, care utilization, quality of care and difficulty in performing activities related to personal care and independent living.

Additionally, the survey questioned respondents about private health insurance coverage, out-of-pocket expenses for prescribed medicines and medical bills, delays in getting health care caused by financial problems and concerns with neighborhood crime and violence at home. Westat, a research firm based in Rockville, Md., conducted the survey under a subcontract with Johns Hopkins University.

To complement the six-state survey, the researchers also analyzed data from a national sample of approximately 10,000 older Medicare beneficiaries. The data came from the 1996 Medicare Current Beneficiary Survey, a multipurpose, longitudinal survey of a nationally representative sample of Medicare beneficiaries sponsored by the Centers for Medicaid & Medicare Services (CMS).

The purpose of this analysis was to better understand factors associated with Medicaid enrollment among Medicare beneficiaries with low incomes who lived in the community and to examine the effect of Medicaid enrollment on their use of health care services.

Under RWJF grant ID# 042637, the researchers conducted detailed analyses of the data from the six-state survey, which they supplemented with Medicare claims for 1998 and 1999.

Other Funding

The Commonwealth Fund contributed $321,104 to this project.

 Back to the Table of Contents


The researchers reported their findings from their study of older Medicare beneficiaries from the 1996 Medicare Current Beneficiary Survey in an article in Health Services Research (2002):

  • About half (49.8 percent) of older Medicare beneficiaries with incomes at or below 100 percent of poverty were enrolled in Medicaid in 1996. About 2.6 million (8.8 percent of community-dwelling Medicare beneficiaries) had incomes at or below the federal poverty level and were enrolled in Medicaid; an equal number of older beneficiaries (about 2.6 million) met eligibility criteria but were not enrolled in the program. Medicaid participation was driven primarily by poor health and disability, although Hispanic, and black beneficiaries were significantly more likely to participate in Medicaid than were white beneficiaries, and beneficiaries with higher education were less likely to participate than others.
  • Dual enrollment of older people in both Medicaid and Medicare did not in itself appear to increase service use. Dually enrolled people were more likely to make at least one visit to an outpatient department or physician's office and used more of these services than other beneficiaries. However, once the poorer health status of those who enrolled in Medicaid was considered, differences in all types of service use were modest. The number of chronic conditions was by far the most consistently significant predictor of both probability and intensity of service use. According to the researchers, these results are consistent with other studies suggesting that differences in Medicare costs between older beneficiaries who are dually enrolled and other beneficiaries are largely accounted for by differences in health status.
  • State policies related to spending for home and community-based services significantly influenced Medicaid enrollment among low-income, community-dwelling elderly persons. Older people living in states with low or medium levels of spending for Medicaid home and community-based services were less likely to be enrolled in Medicaid. However, the researchers found no evidence that states' decisions regarding allocation of resources to home and community-based services influenced the probability that beneficiaries resided in the community rather than in nursing homes.

The following is a synthesis of findings from the six-state survey of older dual enrollees in an article in Medical Care Research and Review (2005), a report to RWJF and several conference presentations, including those to AcademyHealth and the Association for Public Policy Analysis and Management:

  • Almost all (97 percent) of the older dual enrollees had a physician they regarded as a usual source of care, and the percentage who experienced financial barriers was low. The results confirm that the combined benefits of Medicare and Medicaid insurance coverage have substantially reduced financial barriers to care for older dual enrollees. However, evidence of reduced and inappropriate utilization of care by this vulnerable population suggests that other types of access barriers remain.
  • Organizational barriers outweighed geographic and financial barriers in making access to services difficult. Among dual enrollees needing ambulatory or long-term care services, 29 percent experienced organizational barriers (e.g., difficulty arranging or not knowing how to obtain service, lack of referral when service was perceived as needed, waiting list, language or communication difficulties); 14 percent experienced geographic barriers (e.g., lack of transportation or inconvenient site of care). Only 6 percent experienced financial barriers.
  • Overall, 59 percent of the study population received (or thought they needed) a referral from a physician for at least one medical or nonmedical supportive service.
  • About 33 percent of older dual enrollees who received a referral or thought they needed one experienced access barriers.
  • Physicians were more likely to refer patients to medically related than nonmedical supportive services. Of those who thought they needed care from a medical specialist, 87 percent received a referral, but only about half of people who thought they needed community services received a referral.
  • Dual enrollees who were African American were almost twice as likely to experience barriers to care as white dual enrollees. Other characteristics associated with access barriers included: financial distress, an unfavorable assessment of their physician's ability to communicate information, being in fair or poor health status and requiring assistance with activities of daily living. A few characteristics were associated with more than one type of barrier: African Americans had a greater likelihood of both organizational and geographic barriers to care; those in financial distress were more likely to confront all types of barriers.
  • Twenty-five percent of all hospitalizations within one year were preventable for dual enrollees. The most prevalent conditions associated with these hospitalizations were congestive heart failure, pneumonia, chronic obstructive pulmonary disease and kidney-urinary tract infections. Dual enrollees with out-of-pocket costs were almost one and a half times more likely than those with no out-of-pocket costs to experience a preventable hospitalization. Those with five or more chronic conditions and with high numbers of prescription medications were more likely to experience a preventable hospitalization, indicating a need for improved care management among those with the greatest need for services.
  • State policies on Medicaid enrollment and spending affected both the amount of care received and the level of unmet need among older dual enrollees. For example,
    • Greater spending on community-based versus nursing home care increased the probability of receiving supportive services; for those who received services, it increased the number of hours of care
    • Higher per capita home and community-based spending reduced the probability of older dual enrollees receiving any hours of service overall, but, among those receiving services, it increased hours (concentrating more resources on fewer people)
    • Hispanic and African-American dual enrollees were more likely to have unmet needs for assistance, despite their increased likelihood of having formal supportive services and more hours of care.


The researchers noted the following limitations:

  1. The study sample was limited to community-resident older dual enrollees in six states, so caution should be used in generalizing these findings to the national population of older enrollees. Nonetheless, these states provide information on a large sample of older dual enrollees from different areas of the country with populations that differ with respect to race, age and health status. (Medical Care Research and Review, 2005, and findings reported to RWJF)
  2. National datasets such as the Medicare Current Beneficiary Survey do not provide information that would permit a more precise characterization of people's Medicaid eligibility, particularly in view of substantial variation in state eligibility criteria and standards. (Health Services Research, 2002)

 Back to the Table of Contents


The researchers concluded that:

  • Despite the comprehensive insurance coverage provided by the Medicare and Medicaid programs to older dual enrollees, and the fact that virtually all community-dwelling dual enrollees reported that they had a usual source of care, access to care and care management problems remain. (AcademyHealth Annual Research Meeting, June 2004)
  • State policy can influence Medicaid participation. The analysis suggests that how states allocate home and community-based service resources — an important component of Medicaid policy over which states have considerable discretion — affects individual enrollment decisions. Other aspects of Medicaid policy, such as application procedures, may also influence enrollment of eligible persons. A better understanding of how "policy matters" is essential to formulating programs that meet the goal of providing access to services for vulnerable beneficiaries. (Health Services Research, 2002)
  • Improved communication and information-giving by primary care physicians may help reduce barriers to health care services for dual enrollees. Reimbursement for care coordination would allow primary care physicians or their staff to spend more time helping dual enrollees get needed supplementary services. Nonmedical support services will require added attention because barriers to receipt of these services were common. (Medical Care Research and Review, 2005)


The researchers published their findings in Health Services Research (2002) and Medical Care Research and Review (2005). Another article examining the relationship between unmet needs and Medicare and Medicaid policies is completed but not yet accepted for publication. The team is also preparing additional articles for publication. The researchers presented their findings at several national conferences, including annual meetings of the Association for Public Policy Analysis and Management and AcademyHealth. One of the team members wrote a dissertation incorporating findings from the study, which received an Association of Health Care Research and Quality Dissertation Award in April 2003. See the Bibliography for details.

 Back to the Table of Contents


  1. Researchers should anticipate delays when requesting data from the Centers for Medicare & Medicaid Services (CMS). CMS provided data to identify dual enrollees in the six-state survey. Although staff of CMS was supportive of the project, the need for approval from top administration delayed receipt of the data and slowed implementation of the survey. (Project Director)
  2. Working with high-quality research-oriented survey firms is essential. Westat is experienced in conducting large national surveys, such as the National Health and Nutrition Examination Survey, which requires it to have staff support in many states. The availability of experienced staff helped Westat achieve a very high response rate from an older, chronically ill population. (Project Director)
  3. Studies that survey the very old should anticipate that "gatekeeping" by family members will lower the participation rates. The research team and the survey firm, Westat, received many calls from family members concerned that participation in the survey would be a burden for their older parents or relatives. This "gatekeeping" by family members was a factor in lowering the response rate from the projected 86 percent to the actual 76 percent, which was still high. (Project Director)

 Back to the Table of Contents


In January 2005, the project team began a three-year, $600,000 project, funded by the National Institute on Aging, to examine the effects of alternative state and federal long-term-care policies on the availability of daily-living assistance for dual enrollees.

Project staff conducted interviews of dual enrollees in six states chosen for their varied approaches to Medicare and Medicaid spending. In a May 2008 article, "A Comparative Analysis of Medicaid Long-Term Care Policies and Their Effects On Elderly Dual Enrollees" by (Rice J, Kasper J, Pezzin L) published in Health Economics (abstract available online), researchers noted that:

  • Individuals requiring assistance received an average of 213 hours per month of formal and informal care.
  • Some 43.6 percent of disabled individuals reported having unmet needs for assistance.
  • Individuals reported fewer unmet needs when state spending focused on community-based services rather than on nursing home-based services.
  • Individuals reported fewer unmet needs when state spending focused on reaching greater numbers of people.

Researchers suggested that the most effective policy, in terms of reducing the percentage of individuals receiving insufficient levels of assistance, involved high community spending compared to nursing home spending and lower spending per recipient.

 Back to the Table of Contents



Access to Supportive Services and Unmet Needs for Care Among Community-Resident Elderly People with Medicare and Medicaid Coverage


Johns Hopkins University, Bloomberg School of Public Health (Baltimore,  MD)

  • Elderly People with Eligibility for Both Medicare and Medicaid: Needs for Care, Perceptions of Care and State Variations in Access to Services
    Amount: $ 599,991
    Dates: May 1998 to January 2001
    ID#:  033667

  • Improvements in Access to Supportive Services and Unmet Needs for Care Among Community-Resident Elderly People with Medicare and Medicaid Coverage
    Amount: $ 313,764
    Dates: August 2001 to July 2004
    ID#:  042637


Judith Ann Kasper, Ph.D.
(410) 614-4016

 Back to the Table of Contents


(Current as of date of this report; as provided by grantee organization; not verified by RWJF; items not available from RWJF.)


Komisar HL, Feder J, and Kasper JD. "Unmet Long-term Care Needs: An Analysis of Medicare-Medicaid Dual Eligibles." Inquiry, 42(2): 171–182, 2005. Abstract available online.

Niefeld MR and Kasper JD. "Access to Ambulatory and Long-Term Care Services Among Elderly Medicare and Medicaid Beneficiaries: Organizational, Financial, and Geographic Barriers." Medical Care Research and Review, 62(3): 300–319, 2005. Abstract available online.

Pezzin LE and Kasper JD. "Medicaid Enrollment among Elderly Medicare Beneficiaries: Individual Determinants, Effects of State Policy, and Impact on Service Use." Health Services Research, 37(4): 827–847, 2002. Abstract available online.


Kasper JD and Pezzin L. Descriptive Findings from the 6-State Survey of Elderly Dual Enrollees. Baltimore: Johns Hopkins University School of Hygiene and Public Health. December 2000.

Neifeld MR. Access to Ambulatory and Long-Term Care Services Among Elderly Dual Enrollees. (Ph.D. Dissertation), May 2004.

Survey Instruments

Survey of Health Care for Older Americans. Baltimore: Johns Hopkins University, Health Services Research and Development Center, fielded June–December 1999.

Data Tapes

Kasper, Judith D. Six-State Survey of Elderly Dual Enrollees in Medicare and Medicaid, 1999. Both the survey data and information on design and methodology are publicly available at the Inter-University Consortium for Political and Social Science Research online.

Presentations and Testimony

Judith D. Kasper and Liliana Pezzin. "State Medicaid Long-Term Care Policies and Unmet Needs among Elderly Dual Enrollees," at the 25th Annual Research Conference of the Association for Public Policy Analysis and Management, November 6–8, 2003, Washington.

Marlene Niefeld. "Ambulatory Care Sensitive Hospitalizations among Elderly Medicare and Medicaid Beneficiaries," at the AcademyHealth Annual Research Meeting, June 6–8, 2004, San Diego, Calif. Abstract available online.

 Back to the Table of Contents

Report prepared by: Jayme Hannay
Reviewed by: Lori De Milto
Reviewed by: Molly McKaughan
Program Officer: David Colby

Most Requested