The VA faces many challenges in rebalancing its long-term care between institutional nursing home care and non-institutional long-term services and supports in order to meet Veterans' growing demands. It must make policy determinations concerning which Veterans it will provide NH care to in the future and the mix of short-stay and long-stay services it will offer. VHA-enrolled Veterans can receive NH care provided in VA Community Living Centers (CLCs), contract NHs (CNHs), and State Veterans' Homes (SVHs), Medicare Skilled Nursing Facilities (SNF) or Medicaid.
This study used existing VA and CMS data to describe: (A) Veterans newly admitted to NHs; (B) a cross-sectional panel of Veterans residing in NHs; and (C)collected primary data on a sample of VAMCs regarding their NH referral and contracting processes. The specific aims were:
A.Among a cohort of Veterans newly admitted to NHs:
Aim I. Examine factors related to Veterans' initial NH placement setting type (CLC, CNH, SVH, SNF, Medicaid-paid NH or other-paid NH) given their alternative choices/
Aim II. Examine factors related to transfers between different NH settings both directly from one setting to another, and indirectly through a hospitalization following initial admission to NH.
Aim III. Examine factors related to newly admitted Veterans remaining in a NH setting for extended care.
Aim IV. Compare initial type of NH utilization, transitions, and extended-care NH utilization among mandatory (P1A) and non-mandatory Veterans. In particular, examine potential demographic, geographic, and low-acuity casemix differences across VAMCs.
B.Among a prevalence cohort of Veterans:
Aim V. Compare the use of long-term NH services by Veterans with low-acuity case-mix in priority status P1A and in the next highest priority status, P1B.
Aim VI. Compare P1A Veterans who use VA-paid NH care and non-VA-paid NH care.
C.Among a sample of VAMCs:
Aim VII. Conduct qualitative interviews with key informants (VAMC Employees) to better understand the referral and contracting process
VHA administrative data was obtained for calendar years 2003-2010 for all Veterans in the VIREC finder file which included Veteran eligibility, vital status, distance from home to VHA facilities, utilizations, and NH MDS assessments for both VA-provided and VA-paid care. We also obtained Medicare claims (part A and Carrier files), national MDS, Medicaid Personal Summary and Long-Term Care files. NH level data was obtained from CMS OSCAR, and other VA sources
All data were combined to create a VA version of the Residential History File (RHF) with, , longitudinal daily histories of Veterans' health service locations. We validated the RHF site of death, comparing its location on the day of death with location noted on the California Death Statistical Master File on a cohort of Veterans who used NHs and died in California. The validation study showed >90% concordance of RHF and California sites of death.
We developed a new admission cohort of Veterans and cross-tabulated payer source and location of the first admitting NH and various resident characteristics to compare Veterans newly admitted to those NHs.
We conducted 36 semi-structured interviews with key informants at 12 VAMCs chosen to ensure variation in catchment area (rural, urban), available NH options (CLCs, CNHs, SVHs), and four geographic regions. Interviewees included VAMC personnel responsible for NH contracting and discharge planning, including 20 social workers, 12 nurses, and 4 physicians. Interviews were recorded, transcribed, and coded to identify recurring themes and patterns in responses.
We identified 845,257 Veterans who were newly admitted to NHs between the years 2005-2009 with no NH use in the 2 prior years. Of these Veterans, 11% were admitted to CLCs, 2% to CNHs, 64% to SNFs and 21% to community NHs. Veterans who were newly admitted to all NH types in 2009 were more functionally impaired than other Veterans newly admitted in 2005. Veterans admitted to CLCs or CNHs were younger, more likely to be African-American and less likely to be physically impaired than those admitted to SNFs or other non-VA paid NHs. Veterans admitted to CLCs were more likely to have cancer, a terminal diagnosis, and/or serious mental illness.
Almost 21% of 20,962 Veterans newly admitted to CLC s in 2009 were P1As. P1A Veterans were somewhat younger (53% <65 year old vs. 40%), more likely to be married (49% vs 36%), and were more likely to have serious mental illness (15.6% vs. 5.6%). On the other hand, non P1As wre more likely to have cancer (27.7% vs. 22.6%) or end stage disease (19.3% vs. 12.2%). P1A Veterans were more likely to be admitted to CLC or CNH compared to non-P1A (45.6% vs. 16.1%).
The NH contracting process largely mirrors VACO's contracting regulations, but with significant variation in how it has evolved across VAMCs due to differences in local needs, resources, and practices. NH placement decisions are largely contingent upon Veterans' priority status and available NHs (CLCs, SVHs, other) for that priority status. However, large variations continue to exist by regional markets and local policies/practices.
This study provides information about where Veterans receive nursing home (NH) care and what factors relate to admissions, transitions, and length of stay (findings not presented for the latter topics in this summary). This information is critical to informing VHA's planning efforts to meet the LTC needs of an older, increasingly frail Veteran population, especially with the onslaught of Vietnam-era Veterans.
This study resulted in the development of the RHF, an important data infrastructure for studying Veterans' LTC use. The RHF permits systematic examination of Veterans' transfers among VA services and across Medicare- and Medicaid -paid utilization. This study's data infrastructure provided the foundation for the Providence VAMC HSR&D funded Center of Innovation (COIN) on Long-Term Services & Supports, and CREATE Project on LTC for America's Veterans. Importantly, our partner, VACO Office of GEC continues to support this infrastructure through the Geriatrics & Extended Care Data & Analyses Center (GEC DAC) which it established to provide data infrastructure, develop measures and tools, and conduct program evaluations.
External Links for this Project
- Lima JC, Intrator O, Karuza J, Wetle T, Mor V, Katz P. Nursing home medical staff organization and 30-day rehospitalizations. Journal of The American Medical Directors Association. 2012 Jul 1; 13(6):552-7. [view]
- Katz PR, Karuza J, Lima J, Intrator O. Nursing home medical staff organization: correlates with quality indicators. Journal of The American Medical Directors Association. 2011 Nov 1; 12(9):655-9. [view]
- Miller EA, Intrator O. Veterans use of non-VHA services: implications for policy and planning. Social work in public health. 2012 Aug 8; 27(4):379-91. [view]
- Intrator O, Gidmark SI, Burton D. Nursing Home Referrals within the Veterans Health Administration: Practice Variation According to Facility Type and Payment Source. Poster session presented at: AcademyHealth Annual Research Meeting; 2013 Jun 24; Baltimore, MD. [view]
- Intrator O, Cai S, Gidmark S, Meucci M, Kinosian B, Phibbs C, Massey K, Shay K. The Geriatrics and Extended Care Data and Analyses Center: evaluation of the VHA Office of Geriatrics and Extended Care’s transformational pilot projects. Paper presented at: AcademyHealth Annual Research Meeting; 2013 Jun 24; Baltimore, MD. [view]
- Intrator O. Unintended consequences of the VHA Millennium Act on Veterans Requiring Nursing Home Care. Poster session presented at: AcademyHealth Annual Research Meeting; 2013 Jun 24; Baltimore, MD. [view]
- Intrator O, Miller E, Miller S, Guihan M, Meucci M, Hojlo C, Kinosian B. Who are Veterans newly admitted to nursing homes? Poster session presented at: AcademyHealth Annual Research Meeting; 2013 Jun 25; Baltimore, MD. [view]