Among disabled individuals in the first six months of Medicare eligibility, comorbidities and health care utilization differ greatly across categories of State support.

Beneficiaries who qualify for some assistance but not full Medicaid benefits are particularly high users of Medicare services.

The percentage of beneficiaries who receive any support or full Medicaid benefits varies widely across States.

METHODS

We used Medicare enrollment data for the period 2007 to 2009.

Subjects include newly eligible disabled Medicare beneficiaries in 2007 and 2008 identified by their Medicare start date from the Beneficiary Annual Summary File and Medicare Status Code = 20. We extracted all Medicare claims for the initial six months of eligibility for these individuals.

We categorized beneficiaries who received any State support into three groups: Full Duals had full Medicaid coverage including prescription drugs for the entire six-month period or until their death (State Reported Dual Eligible Status Code 02, 04, or 08 all six months). Partial Duals received State support for all six months in the form of premium or copayment assistance (i.e., Specified Low-Income Medicare Beneficiaries [SLMB] or Qualified Medicare Beneficiary [QMB] only, State Reported Dual Eligible Status Code 01, 03, 05, or 06 all six months) or a mix of this support and full Medicaid coverage as defined above. Incomplete Duals received at least one month of some form of State support (either full Medicaid or SLMB/QMB) but less than six months of support (at least one month of State Reported Dual Eligible Code NA and at least one month of State Reported Dual Eligible Status Code 01, 02, 03, 04, 05, 06, or 08). Finally, we classified beneficiaries who received no State assistance as Nonduals (State Reported Dual Eligible Status Code NA for all six months).

We used the monthly “State buy-in” variables to identify whether new beneficiaries had Medicare Part A (hospital and institutional coverage, 1, 3, A, C) and Part B (physician and outpatient services coverage, 2, 3, B, C) for all six months. People without both Parts A and B coverage do not have comprehensive care from the Medicare program. In addition, we had no access to health care usage information for persons enrolled in Medicare managed care plans (HMO indicator not 0 or 4).¹⁰ Typically, people with equal months of Parts A and B coverage with no managed care enrollment are considered “likely to have complete claims” and thus are the appropriate focus of a Medicare claims-based analysis.¹¹ Therefore, we restricted our analysis of health care use and diagnoses to persons likely to have complete claims. Because differences in mortality would complicate assessment of utilization, and because our report specifically examines whether early Medicare experience differs, we excluded persons who died during their initial six months in the Medicare program.

Using the Chronic Condition Warehouse algorithms, we estimated the frequency of six chronic conditions: cancer, Alzheimer’s, chronic obstructive pulmonary disease (COPD), depression, diabetes, and ischemic heart disease.¹² Cancer is indicated if one or more of the following cancers are classified: female breast, colorectal, lung, or prostate. If the first claim for any of these chronic conditions occurred in the initial six months of Medicare eligibility, the chronic condition flag is indicated.

We report utilization and reimbursement separately by type of service and provider. “Acute Inpatient” hospitalizations and days are defined as hospitalizations in acute care hospitals that do not include services in a rehabilitation unit. In Acute Inpatient claims, the third digit of the provider number is 0 or the third and fourth digits are 13.

“Other Inpatient” includes both rehabilitation care and long-stay facilities, including psychiatric hospitals, none of which are included in the Acute Inpatient category. “Hospital Outpatient” includes care provided in hospital outpatient departments, including emergency, radiology, and day surgery. “Skilled Nursing Facility” (SNF) includes care provided by such facilities. Likewise, the “Home Health Care” category includes all care provided by home health agencies and the “Hospice” category includes care provided by Medicare-certified hospices under the Hospice program.

Utilization of Part B services is defined by the unique combination of line item claim procedure code and Berenson-Eggers type of service (BETOS) for the procedure.¹³ The Centers for Medicare and Medicaid Services developed BETOS codes to provide clinically meaningful groupings of procedures for the purpose of analyzing growth in Medicare expenditures. These codes are added to each line item during processing. Evaluation and Management codes (E&M) are identified by BETOS codes beginning with M. Procedures are identified by BETOS codes beginning with P. We classified procedure codes not beginning with P or M as “Other Part B Services.”

We calculated average covered days as the average number of days spent over the initial six-month period receiving each type of care, restricted to those beneficiaries receiving any care of that type. We calculated Part B events (E&M, procedures, other Part B services) and Durable Medical Equipment (DME) events as the number of distinct dates that services of each type were received. Average Medicare payment amount is the average among users of each type of service and summed across all use in the initial six-month period.

RESULTS

From 2007 to 2008, 1,351,446 new beneficiaries enrolled in the Medicare program through the disability benefit. More than 40 percent of these new enrollees received some form of State support (Table 1). Across categories of State support, few beneficiaries were partial dual (QMB/SLMB only) for the entire six-month period (4 percent). Almost 21 percent of beneficiaries had both full Medicaid and full Medicare benefits for all six months, and 16 percent had mixed State support over the same duration.

Table 1

Characteristics of newly Medicare-eligible disabled beneficiaries by demography, 2007–2008.

Sixteen percent of newly Medicare-eligible persons with disability had unequal months of coverage for Part A (hospitalization and institutional care) and Part B (physician and outpatient care). This likely reflects either the challenges of transitioning into the Medicare program or in meeting the cost of the Medicare Part B premium ($105.80 per month in 2007 for beneficiaries earning less than $80,000 per year).¹⁴

Females were more likely to receive some form of State assistance than males (46 percent vs. 36 percent). Receipt of State assistance varied widely by race; 31 percent of whites received assistance in contrast to 48 percent of African Americans, 44 percent of Asian/Pacific Islanders, 44 percent of Hispanics, 51 percent of American Indian/Alaska Natives, and 60 percent of those classified as “Other.” Even within the restricted range examined, older persons (ages 55–64) were much less likely to receive State assistance than younger persons. We found the highest rate of full dual eligibility among newly Medicare eligible under age 45 (37 percent) and the lowest rate for those 55–64 (12 percent).

States vary considerably in terms of the percentage of newly Medicare-eligible disabled persons who received some form of State assistance in the first six months; however, geographic patterns were generally similar between full Medicare/Medicaid assistance and any State assistance (Figure 3, Figure 4, Appendix A).

Figure 3

Newly Medicare-eligible who receive full Medicaid benefits.

Figure 4

Newly Medicare-eligible receiving some form of State support.

The mortality rate in the first six months was similar across the four analytic groups at between 1.06 percent and 1.69 percent (Table 2). Managed care enrollment among newly Medicare-eligible disabled beneficiaries varied by level of State support. While approximately 15 percent of partial and nonduals enrolled in Medicare managed care, only seven percent of full duals did so.

Table 2

Characteristics of newly Medicare-eligible disabled beneficiaries by level of State support, 2007–2008.

The prevalence of selected comorbidities differed across groups of disabled (Table 3). Among those likely to have complete claims, depression and diabetes were the most common comorbidities. We found that partial duals had the highest prevalence of all conditions except Alzheimer’s. Between 31 percent and 52 percent of newly eligible Medicare beneficiaries had one of the six included conditions.

Table 3

Prevalence of selected chronic conditions in initial six months of Medicare eligibility.

In general, patterns of health care utilization varied in consistent ways across beneficiary categories (Table 4). The most striking patterns were related to the higher use by partial duals compared to full duals, nonduals, and incomplete duals. The partial duals had the highest rates of hospitalization and use of hospital outpatient services, home health care, E & M Part B services, Part B procedures, DME, and all other Part B services. The nonduals tended to use the fewest services across categories, and more than 35 percent of nonduals used no services covered by Medicare in their first six months in the program.

Table 4

Utilization of services by newly Medicare-eligible disabled beneficiaries in initial six months, by level of State support and type of service, 2007– 2008.

Usage rates varied greatly between selected services, but patterns of use across categories of beneficiaries varied less. Acute inpatient stays were of similar length for all groups except nonduals (range 7.2–7.8 vs. 5.7 days) (Table 5). Other inpatient stays were of similar length across groups but SNF stays were considerably longer for full duals (average 39 days) and incomplete duals (36.4 days) than for partial and nonduals (28 and 23.2 days, respectively). Partial duals had the most home health care visits (an average 32.3 visits vs. 21.6–24.5 visits). Nonduals had the fewest E&M visits (6.2) while partial duals had the most (8.0).

Table 5

Intensity of service use by newly Medicare-eligible disabled beneficiaries in initial six months of Medicare coverage, by level of Statesupport and type of service, 2007–2008.

Partial duals were most likely and nonduals were least likely to use services, but average payment per user did not vary as strongly across categories (Table 6). In most cases, reimbursements were approximately equal across beneficiary categories. Acute inpatient stays are an exception to this pattern, costing most for incomplete duals ($14,505) and least for nonduals ($11,450). A second exception was for SNF stays, for which average payment ranged from a high of $12,987 for full duals to a low of $8,926 for nonduals. These findings are consistent with the relative lengths of stay reported above.

Table 6

Average Medicare payment per user ($) for disabled Medicare beneficiaries in initial six months of Medicare coverage, by type of service, 2007–2008.

DISCUSSION

Our analysis underscores the need for policy and research to focus beyond disabled Medicare beneficiaries who qualify for full Medicaid benefits to include those who receive some assistance (SLMB/QMB) and those whose assistance status changes over the first six months of Medicare enrollment. These four groups differ significantly in their demographic and health care profiles.

It is safe to assume that nearly all full duals are already covered by Medicaid when they become Medicare eligible. Thus, although this group tends to be very poor, its members do not enter the Medicare program after an extended period without health insurance. In contrast, disabled persons with no Medicaid benefits during the Medicare waiting period are often completely uninsured, and thus may delay needed care or not fill all prescriptions due to cost concerns. The relative usage of Medicare benefits across categories of assistance is consistent with these patterns. If higher usage by those with partial and incomplete assistance levels reflects pent up demand, then experience in the first six months will not necessarily correlate with later usage patterns. Two studies have examined the effect of eliminating the waiting period, and both concluded that doing so would increase Medicare expenditures to an extent not completely offset by longer term consequences of delayed care seeking.^15,16

Based on our data, we can neither use insurance status to categorize individuals entering Medicare, nor accurately determine the impact of prior insurance status on care use during the first six months in the program. To discern how to best direct programs aimed at appropriate use, further research should examine usage over longer time periods and seek to determine whether usage levels stabilize and whether categories of use continue to differ across groups. Our analysis does not distinguish beneficiaries who are institutionalized from those who are community dwelling, or those whose disabilities are developmental from those whose disabilities are acquired. These important distinctions likely correlate with both health care consumption and level of State assistance.

Most estimates suggest that disability rates among working-aged adults are rising. ^17,18,19 The growing number of people who receive Medicare benefits under the Social Security Administration disability program probably does reflect an actual increase in the population of disabled persons under age 65. However, there is no simple way to determine whether there is a change in the percentage of persons with disabilities who receive health care through the Medicare program.

Pezzin and others suggest that the generosity of State Medicaid programs plays an important role in dual eligibility.²⁰ Such examinations are beyond the scope of this report. We do, however, show considerable State variation in the percentage of newly disabled who receive State assistance. State Medicaid policy could affect Medicare disability enrollment in multiple ways. For example, increased generosity could include assistance with completing applications and thus lead to an increase in Medicare disability applications.

Alternatively, reduced State generosity would lead to an increased number of disabled people turning to the Social Security Administration and Medicare for support. Monitoring trends in disability and program enrollment will be necessary to determine the impact of State generosity in these realms and to identify differences in how States and individuals use State support for Medicare benefits.

Our findings suggest that health care usage in the first six months of Medicare enrollment varies significantly across categories of State support. In particular, new enrollees who qualify for assistance with copayments and/or deductibles but who do not receive full Medicaid benefits for at least part of the six-month period are particularly high users of health care. This pattern is consistent with the reality of pent up demand for health care in this group. We were intrigued to find that the greatest health care need is not among full Medicaid enrollees—who presumably have the greatest health problems—but rather among poor individuals who receive financial assistance for Medicare premiums and copayments. This analysis serves as a reminder that broad groupings of disabled persons obscure important distinctions. Future research is needed to examine the nature and persistence of the patterns we have identified. If these initial distinctions among groups persist beyond the first six months, they would point to potential opportunities for focused outreach during the early enrollment period in Medicare.

REFERENCES

1.

Medicare: history of provisions. Social Security Bulletin Annual Statistical Supplement. 2000. [Accessed February 27, 2012]. Available at: www​.ssa.gov/history/pdf/hlth_care.pdf.

2.

Medicare enrollment - disabled beneficiaries: as of July 2010. Centers for Medicare & Medicaid Services (CMS) Denominator File. [Accessed February 28, 2012]. Available at: www​.cms.gov/MedicareEnRpts​/Downloads/10Disabled.pdf.

3.

2010 CMS Statistics: Medicaid beneficiaries/State buy-ins for Medicare: Table I.19. Centers for Medicare & Medicaid Services; 2010. [Accessed February 26, 2012]. Available at: https://www​.cms.gov/Research-Statistics-Data-and-Systems​/Research​/ResearchGenInfo/CMSStatistics.html.

4.

Medicare: a timeline of key developments. Menlo Park, CA: Kaiser Family Foundation; [Accessed February 27, 2012]. Available at: www​.kff.org/medicare/timeline/pf_90.htm.

5.

O’Leary JF, Sloss EM, Melnick G. Disabled Medicare beneficiaries by dual eligible status: California, 1996–2001. Health Care Financ Rev. 2007;28(4):57–67. [PMC free article: PMC4194998] [PubMed: 17722751]

6.

Rowland D. Medicare at forty. Health Care Financ Rev. 2005;27(2):63–77. [PMC free article: PMC4194923] [PubMed: 17290638]

7.

Pezzin LE, Kasper JD. Medicaid enrollment among elderly Medicare beneficiaries: individual determinants, effects of state policy and impact on service use. health Serv Res. 2002;37(4):827–46. [PMC free article: PMC1464012] [PubMed: 12236387]

8.

Foote SM, Hogan C. Disability profile and health care costs of Medicare beneficiaries under age sixty-five. Health Aff (Millwood). 2001;20(6):242–53. [PubMed: 11816665]

9.

Coughlin T, Waidmann T, Watts MO. Medicaid and the uninsured: where does the burden lie? Kaiser Commission Issue Paper. Menlo Park, CA: Kaiser Family Foundation; 2009. [Accessed February 27, 2012]. Available at: www​.kff.org/medicaid/upload/7895-2.pdf.

10.

Asper FM, Mann EM. Medicare managed care enrollees and the Medicare utilization files. Minneapolis, MN: Research Data Assistance Center; 2006. [Accessed February 28, 2012]. Updated June 2011. Publication No. TN-009. Available at: www​.resdac.org/tools​/TBs/TN-009_MedicareManagedCareEnrolleesandUtilFiles_508.pdf.

11.

Nattinger AB, Shapira MM, Warren JL, et al. Methodological issues in the case of administrative claims data to study surveillance after cancer treatment. Med Care. 2002;40(suppl):IV-69–74. [PubMed: 12187171]

12.

Chronic Condition Data Warehouse. [Accessed February 28, 2012]. Available at: www​.ccwdata.org/chronicconditions/index​.htm.

13.

Berenson-Eggers Type of Service (BETOS). Baltimore, MD: CMS; [Accessed February 28, 2012]. Available at: https://www​.cms.gov/HCPCSReleaseCodeSets/20_BETOS.asp.

14.

Frequently Asked Questions: What are the Medicare Part B Premiums for 2007? Austin TX: mymedigap; Available at: www​.mymedicaresupplementinsurance​.com/2007​_Medicare_Part_B_Premiums.html.

15.

Riley GF. Health insurance and access to care among Social Security Disability Insurance beneficiaries during the Medicare waiting period. Inquiry. 2006;43:222–30. [PubMed: 17176966]

16.

Dale SB, Verdier JM. Elimination of Medicare’s waiting period for seriously disabled adults: impact on coverage and costs, Task Force on the Future of Health Insurance Issue Brief. New York, NY: The Commonwealth Fund; Jun, 2003. [Accessed February 28, 2012]. Available at: www​.commonwealthfund​.org/Content/Publications​/Issue-Briefs/2003​/Jul/Elimination-of-Medicares-Waiting-Period-for-Seriously-Disabled-Adults--Impact-on-Coverage-and-Costs.aspx. [PubMed: 12868457]

17.

DeJong G, Palsbo SE, Beatty PW, et al. The organization and financing of health services for persons with disabilities. Milbank Q. 2002;80(2):261–301. [PMC free article: PMC2690107] [PubMed: 12101873]

18.

Lakdawalla DN, Bhattacharya J, Goldman DP. Are the young becoming more disabled? Health Aff (Millwood). 2004;23(1):168–76. [PubMed: 15002639]

19.

Martin LG, Freedman VA, Schoeni RF, et al. Trends in disability and related chronic conditions among people ages fifty to sixty-four. Health Aff (Millwood). 2010;29(4):725–31. [PMC free article: PMC2874878] [PubMed: 20368601]

20.

Pezzin LE, Kasper JD. Medicaid enrollment among elderly Medicare beneficiaries: individual determinants, effects of state policy and impact on service use. Health Serv Res. 2002;37(4):827–46. [PMC free article: PMC1464012] [PubMed: 12236387]

Acknowledgments

The authors wish to thank Jessica Zeglin for her graphic design expertise and Mary A. Leonard for original Data Points report design.

This project was funded under Contract No. HHSA29020100013I from the Agency for Healthcare Research and Quality, U.S. Department of Health and Human Services, as part of the Developing Evidence to Inform Decisions about Effectiveness (DEcIDE) program. The authors of this report are responsible for its content. Statements in the report should not be construed as endorsement by the Agency for Healthcare Research and Quality or the U.S. Department of Health and Human Services. This project has been approved by the University of Minnesota Institutional Review Board.

APPENDICES

Suggested Citation: Virnig BA, Skellan D, O’Donnell B, et al. Newly Medicare-Eligible Disabled: Comparison of Duals and Non-Duals. Data Points # 11 (prepared by the University of Minnesota DEcIDE Center, under Contract No. HHSA29020100013I ). Rockville, MD: Agency for Healthcare Research and Quality. June 2012. AHRQ Publication No. 12-EHC028-EF.