An Investigation of Interstate Variation in Medicaid Long-Term Care Use and Expenditures Across 40 States in 2006. Appendix C. State Long-term CARE Data Anomalies

07/01/2013

MAX data contain a variety of anomalies, many of which are specific to individual states. The anomalies most likely to impact this report's analyses of 2006 MAX are listed in Table C.1, by state. A full list of anomalies is available from the CMS website at https://www.cms.gov/MedicaidDataSourcesGenInfo/07_MAXGeneralInformation.asp.

As a result of these anomalies, we excluded 11 states from all analyses: Maine, because complete 2006 data were not available; Arizona, because it provides most long-term services via managed care arrangements, whereas this study analyzes services provided on an FFS basis; Minnesota, because the state was transitioning many of its LTC recipients to managed care during the study period; New Hampshire, because its expenditure data for HCBS relative to ILTC were believed to be unreliable when compared with data from CMS Form 64; and Massachusetts, Michigan, Montana, Oregon, Pennsylvania, Rhode Island, and Texas because their HCBS data potentially were unreliable. In addition, we excluded the District of Columbia, Washington, and Wisconsin from analyses of populations with physical disabilities or ID/DD because waiver data used to identify these populations were incomplete in these states.

TABLE C.1. MAX 2006 State LTC Data Anomalies
State Excluded
From Study
Anomalies
All States   Expenditures reported as service tracking claims are not included in MAX as they cannot be attributed to specific persons.
Alabama   No notes.
Alaska   Alaska had a state-operated Pioneers Home System, not included in Medicaid that provided services to many people who otherwise might be in a nursing facility. The average Medicaid Payment Amount for nursing facility claims is about two times higher than average but is consistent across years.
Arizona X Most people are enrolled in managed care and more than half of the other Medicaid enrollees are in the Indian Health Service, so FFS distributions are unusual. As a result, Arizona is excluded from the analyses presented in this report.
Arkansas   Dual enrollment may be unreliable in Arkansas from January-September 2006 because some full-benefit aged and disabled dual-eligibles were incorrectly identified as partial dual-eligibles who received only Medicare cost-sharing benefits. All partial dual-eligibles were excluded from this report. Possibly as a result, HCBS expenditures were 21% lower in MAX ($220 million) than in Form 64 ($278 million).
California   California had PACE plans, and the state's Senior Care Action Network 1115 waiver included a Medicare Special Needs plan in 2006. Expenditures for LTC services provided through these plans cannot be not identified in MAX.
Colorado   Colorado had PACE in 2006, and LTC expenditures for services provided through these plans cannot be identified in MAX.
Connecticut   MAX HCBS user counts do not correspond well with those reported in Ng et al. (2009). Ng et al. show a decline in 1915(c) enrollees in 2006, whereas MAX data do not. However, waiver expenditures in MAX correspond to those in Form 64 data.
Delaware   No relevant notes.
District of Columbia Excluded from analyses of physically disabled and ID/DD only Enrollment and claims reporting for 1915(c) waivers in MAX did not always correspond. Also, waiver expenditures were 30% higher in MAX ($45.5 million) than in Form 64 ($35.1 million), and MAX included 53% more 1915(c) enrollees (2,600) than Ng et al. (1,700). However, waiver enrollment was growing dramatically during this period, and these differences likely are explained by differences between MAX (calendar year) and Form 64 and 372 (FY) reporting periods.
Florida   Enrollment and claims reporting for 1915(c) waivers in MAX did not always correspond. Also, Florida did not report any inpatient psychiatric services for individuals under age 21, although this service is covered in the state. Finally, the state had a PACE managed care program in 2006.
Georgia   No relevant notes.
Hawaii   No relevant notes.
Idaho   No relevant notes.
Illinois   Illinois had a PACE managed care program in 2006. Expenditures for services provided through these plans cannot be identified in MAX.
Indiana   No relevant notes.
Iowa   No relevant notes.
Kansas   Kansas had PACE managed care in 2006. Expenditures for LTC services provided through these plans cannot be identified in MAX.
Kentucky   There was an error in Kentucky's claims reporting for 1915(c) services, such that some non-waiver claims for individuals enrolled in 1915(c) waivers were reported as waiver services. Some of these claims were corrected but waiver expenditures for FFS enrollees were 14% higher in MAX ($276 million) than in Form 64 ($243 million). HCBS expenditures may be somewhat overestimated.
Louisiana   MAX reported 20% more Section 1915(c) service recipients (11,000) than reported in Ng et al. (9,200). Issues related to Hurricane Katrina may have impacted service use as well as the reliability of claims and eligibility data in 2006.
Maine X Maine did not submit complete and reliable inpatient, LTC, or other claims in 2006. As a result, the state is excluded from the analyses presented in this report.
Maryland   Maryland had PACE managed care in 2006.
Massachusetts X HCBS expenditures were 32% lower in MAX ($833 million) than in Form 64 ($1.2 billion). Also, the state had PACE managed care and Senior Care Option plans, the latter being similar to PACE plans, in 2006. Expenditures for services provided through these plans cannot be identified in MAX. As a result, the state is excluded from the analyses presented in this report.
Michigan X HCBS expenditures were 68% lower in MAX than in Form 64, and MAX reported 67% fewer HCBS users (26,000) than in Ng et al. (80,200). As a result, the state is excluded from the analyses presented in this report.
Minnesota X In 2006, aged enrollees either voluntarily enrolled in the state's Minnesota Senior Health Options managed care program (which includes HCBS and 180 days of nursing facility care) or were enrolled in Senior Care Plus (which also includes HCBS and 180 days of nursing facility care). Disabled enrollees could enroll in the Minnesota Disabled Health Options program. Expenditures for LTC services, including up to 180 days of nursing facility coverage, covered through these plans cannot be identified in MAX. As a result, the state is excluded from the analyses presented in this report.
Mississippi   No relevant notes.
Missouri   Missouri had PACE managed care in 2006. Expenditures for LTC services for these plans cannot be identified in MAX.
Montana X HCBS expenditures were 53% lower in MAX ($58 million) than in Form 64 ($124 million), and expenditures for 1915(c) waiver claims were almost 70% lower in MAX ($27 million) than in Form 64 ($88 million). As a result, the state is excluded from the analyses presented in this report.
Nebraska   MAX reported fewer home health participants (5,000) than Ng et al. (7,700) for 2006. However, MAX home health expenditures corresponded well with those reported in Form 64.
Nevada   No relevant notes.
New Hampshire X Many claims could not be adjusted properly because of how adjustment claims were submitted to MSIS. There are likely to be duplicates because only the original and replacement claims were reported, and the voids were not included. As a result, the state is excluded from the analyses presented in this report.
New Jersey   Waiver expenditures were 25% lower in MAX ($630 million) than in Form 64 ($839 million). However, expenditures increased dramatically in Form 64 between 2005 and 2006.
New Mexico   New Mexico had a PACE plan but did not report this plan in MAX.
New York   New York had managed LTC and PACE in 2006, and also operates a Senior Care plan, which is reported as a comprehensive HMO in MAX. LTC expenditures provided through these plans cannot be identified in MAX.
North Carolina   No relevant notes.
North Dakota   About 40% of dual-eligibles had ILTC claims in 2006, higher than any other state. Waiver expenditures were 24% higher in MAX ($81 million) than in Form 64 ($65 million).
Ohio   Dual-eligible coding may be unreliable in 2006. Ohio had PACE managed care but did not report this plan in MAX.
Oklahoma   Oklahoma had PACE managed care but did not report this plan in MAX.
Oregon X Oregon had PACE managed care in 2006. Expenditures for LTC services provided through these plans cannot be identified in MAX. More important, waiver expenditures were more than 50% lower in MAX ($346 million) than in Form 64 ($638 million). As a result, the state is excluded from the analyses presented in this report.
Pennsylvania X Pennsylvania had managed LTC and PACE in 2006. Expenditures for LTC services provided through these plans cannot be identified in MAX. More important, waiver expenditures were 68% lower in MAX ($502 million) than in Form 64 ($1.6 billion). Also, the state undercounted enrollment in several eligibility groups from January-June 2006. As a result, the state is excluded from the analyses presented in this report.
Rhode Island X Rhode Island had PACE but did not report this plan in MAX. More important, HCBS expenditures were 67% lower in MAX ($81 million) than in Form 64 ($243 million). Also, reported use of waiver, personal care, and home health services did not correspond to counts reported in Ng et al. for 2006. As a result, the state is excluded from the analyses presented in this report.
South Carolina   South Carolina had a PACE program in 2006. Expenditures for LTC services provided through these plans cannot be identified in MAX. Waiver expenditures were 58% lower in MAX ($123 million) than in Form 64 ($293 million), but HCBS compare well overall.
South Dakota   HCBS expenditures were 26% higher in MAX ($117 million) than in Form 64 ($94 million).
Tennessee   HCBS expenditures were 25% higher in MAX ($702 million) than in Form 64 ($405 million), but reported expenditures in Form 64 increased to more than $600 million in FY 2007.
Texas X Texas had a PACE program in 2006 but did not report this plan in MAX. HCBS expenditures were 32% lower in MAX ($1.4 billion) than in Form 64 ($2.0 billion) because most state plan personal care was not identified as such on claims. As a result, the state is excluded from the analyses presented in this report.
Utah   Utah had managed LTC, and expenditures for services provided through these plans cannot be identified in MAX.
Vermont   Starting in 2006, 1915(c) waiver services were covered under Vermont's 1115 Global Commitment to Health waiver. The Global waiver puts most enrollees into a public managed care organization, but most services are reported as FFS in MAX.
Virginia   MAX reported one-third more 1915(c) service recipients (33,000) than Ng et al. (25,000). However, the problem appears to be caused by reporting of children and adults, who are excluded from this study.
Washington Excluded from analyses of physically disabled and ID/DD only Washington had a PACE program in 2006. Expenditures for LTC services provided through these plans cannot be identified in MAX. Washington did not report waiver enrollment, so people with physical disabilities and those with ID/DD could not be differentiated in the state.
West Virginia   No relevant notes.
Wisconsin Excluded from analyses of physically disabled and ID/DD only Wisconsin had managed LTC and PACE in 2006. Also, Wisconsin's iCare plan for disabled individuals included coverage for short-term nursing home stays (mostly for rehabilitation). Individuals enrolled in these plans are not included in this study. MAX reported fewer waiver service recipients than reported in Ng et al., and waiver expenditures were 12% lower in MAX ($560 million) than those reported in Form 64 ($638 million). Wisconsin did not report waiver enrollment, so people with physical disabilities and those with ID/DD could not be differentiated in the state.
Wyoming   No relevant notes.

View full report

Preview
Download

"40State.pdf" (pdf, 787.2Kb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®