BACKGROUND/RATIONALE:
The Veterans Health Administration (VHA) and Indian Health Service (IHS) entered into a Memorandum of Understanding (MOU) in 2010 to improve access to healthcare and clinical collaborations for American Indian (AI) and Alaska Native (AN) Veterans who are served by both IHS and VHA. The MOU cited specific examples of clinical collaborations, including Home Based Primary Care (HBPC), which was about to expand to rural AI reservations and Tribal communities through special funding to 14 VHA medical centers (VAMC). HBPC is a VHA non-institutional long term care (NILTC) benefit that is provided in the home and is designed to meet the healthcare needs of Veterans with complex chronic conditions. IHS, which provides healthcare to federally-recognized Tribes, is managed by the Department of Health and Human Services (DHHS) or by Tribes in agreement with DHHS. The proposed expansion of HBPC anticipated three potential areas of contributions by IHS: in-kind clinical support, working closely with HBPC staff, and providing office space and equipment; the VHA would deliver HBPC services, disseminate this NILTC model and facilitate outreach and enrollment. Our aims were to use the "natural laboratory" of these innovative programs to identify emerging models of collaboration or co-management between VHA and IHS and to identify barriers and facilitators to broader implementation.

OBJECTIVE(S):
The immediate aims were: a) Identify and characterize variation in VHA-IHS/Tribe models for delivery of HBPC services; b) Describe variation in organizational relationships between VHA and IHS/Tribe partners; c) Identify key barriers and facilitators for implementation of HBPC with IHS/Tribe partners. The long-range objectives of this study are to implement best practices in coordinated healthcare delivery systems in rural areas between VHA and IHS/Tribe partners.

METHODS:
We used a qualitative observational design to study the HBPC innovation programs, collecting data at the program-level and at the organizational-level.

Program level data were collected through one hour telephone interviews with key respondents who were knowledgeable about these HBPC programs. Up to 6 potential participants were identified by each VAMC Chief of Staff, or a designee, and then a purposeful sample of VHA employees was constructed by the Principle Investigator. The sample represented each of the 14 VAMC, various HBPC-related roles at the program, facility and regional Veterans Integrated Service Networks and a range of clinical and non-clinical disciplines. We developed a semi-structured open-ended interview guide that was informed by the Consolidated Framework for Implementation Research and by past surveys conducted by the VA Center for Implementation Practice and Research Support. Respondents were asked to describe the structure and processes of the HBPC programs and the experience of collaborating with IHS/Tribal healthcare organizations. Additionally, sites were asked to share program documents and formal agreements between VHA and IHS and/or Tribes for case comparison. Interview transcripts and program documents were analyzed using qualitative summary and matrix methods for cross case comparisons.

Organizational level data were collected from 1) supervisors' responses on the VA All Employee Survey (AES) items on organizational culture for the years immediately prior to the implementation of the program (FY2008-FY2009) and immediately following implementation (FY2010-FY2012) and 2) publically available descriptions of the medical facilities.

FINDINGS/RESULTS:
From a potential pool of 84 respondents, 53 prospective respondents were identified; 48 were assigned to the study sample and invited to participate as study volunteers; 37 volunteered. The final sample included at least one respondent from each of the 14 VAMC and 51% of respondents represented HBPC at the program-level.

Twelve of the VAMC established HBPC programs with IHS or Tribes. Planning and roll-out efforts varied across these programs; some VAMC proceeded with local informal agreements with IHS/Tribes, while others developed site-specific MOU to detail the obligations of each healthcare organization. This variation was not associated with the organizational structure of IHS as federal direct healthcare or as Tribal healthcare.

In addition to the "traditional" HBPC model, (i.e., meeting the HBPC Handbook 1141.01 description of program operation and staffing with interdisciplinary teams), "non-traditional" models (i.e., conforming to the Handbook's opportunity for innovative expansions to improve access) also developed. "Non-traditional" models included: 1) streamlined staffing, often with a nurse practitioner working from his/her own residence in more remote locations, and 2) integrated partnership with joint privileging of key medical staff. VAMC that implemented "traditional" models were characterized by significantly higher ratings on the AES organizational culture items of entrepreneurial and team dimensions than VAMC that implemented "non-traditional" models. The types of HBPC program that were implemented by VAMC was not associated with organizational structure in IHS as federal direct or as Tribe mangement healthcare.

At all sites, VHA HBPC staff used formal (i.e., electronic health record, FAXed reports) and/or informal (e.g., telephone) communications with IHS/Tribe about mutual patients to coordinate care and optimize access to healthcare resources. IHS/Tribe collaborative efforts varied by location and included identifying Veterans as potentially eligible for HBPC through informal recommendations and/or formal referral process, providing medications at no-cost or with greater convenience to patients unable to use the VHA mail-order pharmacy, and providing clinical services (e.g., physical therapy) that would not be available through VHA staff or contracts in isolated rural areas. Office space was rarely provided by IHS/Tribes as an in-kind or a leased option. Challenges to the innovation projects included the expected issues of timely and effective clinical coordination in the absence of a shared medical record, difficulty recruiting and staffing HBPC positions in rural areas and increased operational costs related to rural areas. Additional challenges were overcoming mistrust in federal institutions and lack of awareness about VA benefits by AI Veterans and/or Tribes, and lack of cultural awareness and competencies by VHA staff. The single most important lesson across all settings was the importance of taking sufficient time to build a trusting personal and institutional relationship before offering a new service. In many cases implementation was also facilitated by active involvement of a VHA leader or by VHA staff who were experienced in working with Native communities.

IMPACT:
These evolving collaborations demonstrate opportunities to coordinate clinical care between federal healthcare organizations and are models for delivery of patient-centered care in rural areas. Future dissemination will require building relationships between VAMC and local IHS/Tribe partners, accounting for characteristics of VAMC organizational culture and available resources.

---

External Links for this Project

NIH Reporter

Grant Number: I21HX001034-01
Link: https://reporter.nih.gov/project-details/8467881

---

PUBLICATIONS:

Journal Articles

1. Kramer BJ, Cote SD, Lee DI, Creekmur B, Saliba D. Barriers and facilitators to implementation of VA home-based primary care on American Indian reservations: a qualitative multi-case study. Implementation science : IS. 2017 Sep 2; 12(1):109. [view]
2. Leff B, Carlson CM, Saliba D, Ritchie C. The invisible homebound: setting quality-of-care standards for home-based primary and palliative care. Health affairs (Project Hope). 2015 Jan 1; 34(1):21-9. [view]

HSR&D or QUERI Publications

1. Kramer BJ, Creekmur B, Saliba D. VHA-Indian Health Service Collaborations Reveal Opportunities to Improve Rural Health Care. Forum: Research Highlight, Summer 2017 Issue. 2017 Aug 1. [view]

DRA: Health Systems
DRE: Treatment - Observational
Keywords: none
MeSH Terms: none