There is growing awareness that long-term and post-acute care (LTPAC) providers play a critical role in care coordination and related payment and delivery reforms intended to improve quality and reduce costs, such as accountable care organizations (ACOs), hospital and post-acute care bundling, and Medicare's hospital readmission policy. Additionally, timely electronic health information exchange (eHIE) between LTPAC providers and other providers is a promising and critical strategy for achieving these care coordination, quality improvement, and cost reduction goals.
Long-Term and Post-Acute Care Providers
LTPAC providers include a wide range of providers, such as: long-term care hospitals (LTCHs), inpatient rehabilitation facilities (IRFs), skilled and unskilled nursing facilities (NFs),4 and home health agencies (HHAs). Other providers that deliver related home and community-based services (HCBS) include hospice, assisted living facilities (ALFs), and adult day care. LTPAC and HCBS providers vary by relative emphasis on: (1) medical versus social service needs; and (2) restorative and recuperative services versus services intended to maintain functioning or slow deterioration (or in the case of hospice service the delivery of palliative care).
This project focuses on skilled and unskilled nursing facilities and HHAs to the extent possible. In addition to the request for proposal (RFP) requesting such a focus, NFs and HHAs are a major component of the LTPAC provider segment, with a relatively large number of facilities, beds, and residents/patients that are transitioning to and from other health care providers, such as hospitals. Below we provide additional information about LTPAC providers, their adoption of electronic health records (EHRs), and a conceptual framework for understanding eHIE involving LTPAC providers.
According to data from the American Health Care Association, there are 15, 632 certified NFs and 1,368,351 patients in certified beds in the United States.5 In 2012, approximately 3% of the over 65 years of age United States population resided in nursing homes (NHs), and approximately 11% of the 80 years of age United States population resides in NHs.6 Further, these segments of the United States population are growing and so NF use as well as use of alternative care settings will rise.7
Home health care is another LTPAC setting that offers a possible alternative to NF care and is a segment of the LTPAC provider market that is also growing rapidly.8 U.S. Department of Health and Human Services (HHS) Centers for Medicare and Medicaid Services (CMS) data indicates that as of 2010 there were 12,311 HHAs, and 3.4 million beneficiaries receiving home health services in the United States.9
A recent issue brief by CMS, titled Medicare Post-Acute Care Episodes and Payment Bundling,10 also provides some critical information about the volume and nature of transitions of care between hospitals and LTPAC providers. Specifically, the report notes that:
Nearly 40% of patients discharged from the hospital received post-acute care.
14.8% of those patients are readmitted to an acute hospital within 30 days.
Use of multiple post-acute care sites within 60 days is common occurs in more than half (50.5%) of post-acute care users.
Clearly, there is a major opportunity to improve the quality, safety, and efficiency of care as patients move through the continuum of care from acute to post-acute and long-term care and the various facilities and settings in which such services are provided. LTPAC provider engagement in eHIE could potentially help to achieve improvements in quality, safety and efficiency.
But what do we know about the certified EHR capabilities of LTPAC providers, particularly NFs and HHAs? There are no nationally representative data available about the current state of certified EHR adoption and eHIE by LTPAC providers.
It is very difficult to determine current rates of EHR adoption by LTPAC providers from prior studies because the best available evidence was collected before 2009.11 Additionally, our research indicates that there is not a shared definition between LTPAC providers and their trading partners of what functionality constitutes an EHR. For example, LTPAC providers appear to indicate that the ability electronic reporting of demographic and financial data is health information exchange (HIE), while their trading partners indicate that clinical data exchange and re-use as defined under Meaningful Use (MU) is the appropriate definition. Finally, other aspects of the methods and data, such as sample frames and sizes, differ substantially across studies.
Purpose of the Study
The general purpose of this project is to study and learn from early efforts to prepare for and implement eHIE between LTPAC providers and their exchange partners (e.g., hospitals, medical groups, pharmacies, and their staffs). This includes learning from the experiences of health plans (e.g., Medicare Advantage, Medicaid agencies, Medicaid managed care plans, and commercial plans), HIEs, state policy officials, and evaluators of eHIE initiatives in addition to the experiences of LTPAC providers and their exchange partners. This project also includes developing a plan for quantitatively assessing the impact of eHIE among these providers and their trading partners on key outcomes such as 30 day post-hospital discharge readmission rates, hospital admission rates from the emergency room (ER), and total Medicare resource utilization. More specifically, the project seeks to answer the following six major research questions:
What community characteristics and/or programs (e.g., service delivery and payment models, special initiatives, collaborations, etc.) enabled and continue to support the electronic exchange of health information between LTPAC providers and their HIE trading partners (e.g., physicians, hospitals, pharmacies/pharmacists, etc.)? What was/is the focus of these activities (e.g., improving coordination/continuity of care, increasing efficiencies and reducing costs, identifying information exchange needs, building trust, etc.)? Over what period of time were these activities implemented prior to and during implementation of HIE activities?
What types of health information do the LTPAC providers and their trading partners need to support continuity and coordination of care; and how were these information needs identified? What types of information do the LTPAC providers and their trading partners create and transmit? How has the type and timing of information exchange changed since implementing eHIE?
What business/organizational/quality/other factors lead to the LTPAC provider's decision to engage and invest in eHIE? What eHIE methods (i.e., what technology solutions) are used to transmit information to/from the LTPAC provider and their HIE trading partners? Does the method of exchange enable the interoperable exchange and re-use of needed clinical information? What are the costs of the technology solutions?
What activities (e.g., technological, policy, financial and human workflow) were undertaken by the LTPAC provider to prepare for and enable the provider/staff to engage in eHIE?
How has the creation, transmission, and receipt of eHIE (including interoperable exchange) at times of transitions in care and during instances of shared care impacted the clinical workflow in the LTPAC settings and that of their clinical trading partners (i.e., physicians, hospitals, and pharmacies/pharmacists)? What do the LTPAC providers and their trading partners describe as being the advantages and disadvantages of engaging in eHIE with LTPAC providers?
What is the measureable impact of eHIE on the quality, continuity, and cost of care for: (1) the LTPAC providers; and (2) their HIE trading partners? For example, how has eHIE affected 30 day post-hospital discharge readmission rates; hospital admission rates from the ER; and total Medicare resource utilization? What is the average number of eHIE message transmissions per LTPAC admission and discharge? Can the analyses being undertaken in selected communities be extended; and if so, how? Can these analyses be applied in other communities, and if so, how?12
As described further below in the methods, data, and findings sections, the scope and degree of exchange involving LTPAC providers was less than anticipated at the project's start. While our study was able to explore questions related to preparation for and implementation of eHIE involving LTPAC providers; our study was only able to partially address some of the research questions related to the impact of eHIE involving LTPAC providers on quality, continuity, cost of care, and workflow. Nonetheless, our team has developed a feasible, high-level quantitative and mixed research method plan for studying the implementation and impact of eHIE between LTPAC providers and their exchange partners that will help address other pressing questions in the near term.
The rest of this report is organized as follows. We first describe the rich methods and data sources used, specifically a review of the literature, conversations with stakeholders involved with eHIE initiatives with LTPAC providers, and in-depth case studies in two states (Minnesota and Pennsylvania). Then, we describe and discuss our findings from a structured literature review, stakeholder discussions, and case studies conducted in Minnesota and Pennsylvania (two eHIE initiatives in Minnesota and one in Pennsylvania). The sections describing the findings are organized by three topic areas: preparation, implementation, and evaluation. We also introduce each of these sections describing the specific research questions that are answered within that section. We sought to triangulate findings from the literature review, stakeholder discussions, and case studies, but in some research we obtained limited information from one data source. For example, the literature review yielded little information about the implications of eHIE involving LTPAC providers on human workflow (in terms of both preparation activities and impact). These areas are noted in the text to the extent possible. In each section we highlight where findings from the data sources were consistent with each other and where findings diverged. We close with a discussion of issues to consider in advancing eHIE involving LTPAC providers and evaluating the impact of those efforts on quality, cost, and utilization.
To address the six major research questions and guide our case studies and quantitative plan, we developed a conceptual framework (Figure 1 below) based on our stakeholder discussions and literature review. Conceptual frameworks identify concepts of importance for addressing the research questions and the hypothesized relationships between them. They also can help clarify the different levels or units of analysis (i.e., national, state, regional, organizational, individual providers, patient population or sub-populations). Although this conceptual framework was based on our findings from the stakeholder discussions and literature review, described in detail in the "Methods" below, we introduce and provide an overview of the conceptual framework here because it informed our selection of states and regions for case studies and it also provides a roadmap for our background and findings sections.
|FIGURE 1. Conceptual Framework for Studying eHIE Involving NH and Home Health Providers|
Several aspects of our conceptual framework are noteworthy. First, our conceptual framework is comprised of three major "levels." These include:
The environment or context (top row) in which specific regional eHIE efforts between LTPAC providers and their exchange partners are occurring. In general, this environment or context consists of major federal and state policies that are shaping eHIE and the behavior of LTPAC providers and their exchange partners. As Figure 1 shows, there are three major sub-components of the policy environment: American Recovery and Reinvestment Act (ARRA) Health Information Technology for Economic and Clinical Health (HITECH), Medicare Policy and Payment, and Medicaid Policy and Payment.
As described in the background and overview section and further below in our findings sections, ARRA HITECH programs have been the primary driver of eHIE efforts in various states and regions through particular programs such as the state HIE cooperative agreements and the Medicare and Medicaid EHR payment incentive programs. Although the state HIE cooperative agreements have largely ended, their experience and degree of success continues to shape whether and what kind of opportunities for eHIE that LTPAC providers and their exchange partners currently have in the region. In 2014, the HHS Office of the National Coordinator for Health Information Technology (ONC) also developed and released a ten-year vision to achieve Interoperable health information technology (HIT) infrastructure, prioritizing strategies and activities required to achieve interoperable exchange in the short and longer terms, including greater consumer and patient engagement.13
The Medicare and Medicaid EHR incentive programs defined the MU in stages and gave eligible hospitals and professionals (e.g., hospitals and medical groups but not LTPAC providers) incentives to adopt and use certified EHRs and engage in eHIE. Stage 2 MU in particular gave eligible hospitals and professionals greater incentive to engage in eHIE generally, so some began to engage in eHIE with LTPAC providers. Proposed modifications to MU in 2015-2017 and the proposed Stage 3 MU rules were released in April 2015 and the final rule was released on October 6, 2015. The 2015-2017 modifications restructures Stages 1 and 2 MU to: align them with Stage 3 in 2017 or 2018; refocus the existing program toward more advanced use of EHR technology; and align the required reporting periods for providers to support a flexible, clear framework, ensuring sustainability of the Medicare and Medicaid EHR Incentive programs. All providers will be reporting at the Stage 3 level by 2018 regardless of previous progress.14 Overall, the 2015-2017 and Stage 3 rules provide even further incentive for eHIE as the requirements related to eHIE have increased.
Finally, ONC and others at the federal and state levels have been working to further clarify, develop and strengthen privacy and security policies. Challenging issues remain in this area (e.g., trust and authentication protocols for providers, legal liability for data exchanged and accepted into an EHR, patients' ability to opt-out and/or access and control some or all of the information about their health).
Similarly, as shown to the right in the top row of Figure 1, there are major Medicare and Medicaid policy and payment changes underway that are shaping LTPAC providers and/or their trading partners incentives to engage in eHIE with one another. For example, both Medicare and Medicaid have a variety of provider payment and delivery system reform initiatives fully implemented (e.g., hospital readmission penalties) or underway (e.g., ACOs, patient-centered medical homes [PCMHs], Health Homes) that provide greater incentives to hospitals and medical groups to consider engaging in exchange with LTPAC providers. Conversely, Medicare requires LTPAC providers to collect and report assessment data such as the Outcome and Assessment Information Set (OASIS) data for home health patients and Minimum Data Set (MDS) data for NH patients and the Improving Medicare Post-Acute Care Transformation (IMPACT) Act also places additional requirements to standardize and make interoperable assessment data elements in LTPAC settings.
In some markets, private health plans with products and populations that may require NH care (e.g., Medicare Advantage Plans, Medicaid managed care organizations for certain sub-populations) are seeking to create preferred provider networks with LTPAC providers and require a willingness and ability to engage in eHIE to be in that preferred network.
Collectively, these major policy and payment areas can create greater financial and non-financial incentives for LTPAC providers and their partners to participate in eHIE.
Finally, it's important to note that in order for eHIE to occur, both parties (i.e., sender and receiver) have to be willing and able to engage in exchange. As we describe, sometimes there is a misalignment of incentives or willingness and ability between LTPAC providers and their exchange partners, so exchange does not happen at all or is constrained to particular uses and/or technological approaches. This is particularly true at this relatively early point in certified EHR adoption and use by Medicare/Medicaid EHR Incentive Program MU eligible providers, relatively limited availability of certified EHR technology (CEHRT) and use by LTPAC providers, and the early and rapidly changing nature of eHIE approaches in health care.
The characteristics of the region in which eHIE between LTPAC providers and their exchange partners are taking place (middle row). Of particular importance at this level are: (1) whether the region had developed a functioning and independent health information organization (HIO) and, if so, whether it was funded with federal and state funds or it was privately funded, what technological solutions it employs, and if it can be sustained; (2) the structure of the health plan, hospital, medical group and LTPAC markets; and (3) the financial and non-financial incentives present in the specific market, including things like provider payment and delivery reforms and or other major demonstrations, grants, or projects that are taking place.
Many of the regions in which we found early eHIE efforts between LTPAC providers had a relatively successful HIO (i.e., at least operationally, even if engagement with and involvement of LTPAC providers was still in the early phases) and/or one or more large integrated delivery system or network (IDSN) that dominated the market. As we describe further, in both of our case study states and regions, large IDSNs were either a founding member and sponsor of the public or private HIE (e.g., Geisinger helping start and support KeyHIE (Keystone HIE) in Pennsylvania) or served as an HIO and eventually become the certified HIO in the region or state (i.e., Allina in Minnesota).
When the HIO is either supported or closely aligned with an IDSN, it is important to consider the specific type or components of the IDSN (e.g., does it own and operate a health plan? Is it also operating as an ACO?) because these likely shape the systems incentives for and approach to engaging in eHIE with LTPAC providers, their ability to lead and organize an eHIE initiative in the region and sometimes state, and how their HIO or IDSNs actions are perceived.
Some IDSNs dominate the regional market and are perceived to have the financial resources and technical expertise to not only adopt and use EHRs but to either support a public HIO or serve as the HIO itself. However, in some case other providers fear that those IDSNs will use the information in the HIO to gain greater market power, for example, by strategically using the information in the HIO to assess referral patterns, performance of other organizations, and risk of various sub-populations. So, as one respondent noted, "eHIE often moves at the speed of trust."
Similarly, the structure of the LTPAC organizations that are not owned by the IDSN is important also. If the LTPAC providers are not owned by IDSNs, are they part of national chains (which are typically for-profit), smaller regional chains, or single or quite small providers (e.g., "mom-and-pops")? There is significant variation across regions in the structure of the LTPAC providers and their exchange partners and the hence the incentives (financial and non-financial) that they have to engage in exchange, the perceptions of the business case and/or return on investment (ROI) for adoption certified EHRs and engaging in eHIE in specific regions and states, as well as what technological solutions that are available and most desirable.
Finally, it important to note that when a public or private HIO is successfully operating, there is pressure to develop a funding mechanism and technological solution that is viable in the particular region. Initially, Medicaid and private health plans were thought to be an additional and longer term funding source for HIOs besides federal and state grant funds, but in some states Medicaid and private plans were less involved in eHIE efforts. More recently, Medicaid is playing a key role in some states, supporting HIOs and eHIE through 90-10 matching funds through the Medicaid EHR incentive program and related population health or Medicaid related provider payment and delivery reform efforts in the state, such as Medicaid ACOs, PCMHs, or Health Homes. Additionally, some private insurers have supported their own HIE (e.g., Blue Cross/Blue Shield in California). Many HIOs are reportedly struggling, but at least some HIOs appear to have developed viable, long-term funding models that serve as one viable mechanism and avenue for exchange moving forward. As noted, competitors are individual IDSNs and exchange with providers using the same EHR and EHR portals which we describe in greater detail in the findings sections.
The characteristics of the specific provider organizations (bottom row) in the region, including both the LTPAC providers and their exchange partners. Some key issues at this level are whether the organization has adopted a certified EHR, if so what kind, and whether they believe there is a strategic advantage and/or positive business case for engaging in exchange. If they believe there is a strategic advantage and/or positive business case for eHIE, there also is the question of what kind of data will be exchanged, for what purpose, and through which technological means.
The literature on EHR adoption and use points to a possible fourth level of analysis: that is the individual level. Specifically, individual providers (e.g., physician, nurse) and/or staff (e.g., managers, clerks) attitudes and views toward EHRs and eHIE may vary based on their own characteristics. For example, physician and other clinicians' age is negatively associated with willingness and ability to use an EHR or use all of its features, even when the practice or hospital they work in has one installed.15 Similarly, patients/residents and their family or guardians view of EHRs, eHIE, and privacy and security and related issues, such as willingness to provide consent, may vary by education, income, race/ethnicity or other individual characteristics. However, we do not include this level explicitly in our conceptual framework, as we were unable to collect much data on these issues through our case studies. Nonetheless, where our respondents reported on these issues in their own organizations, we report them.
The second noteworthy feature of our conceptual framework is that it allows for bidirectional effects of each level (e.g., environment or context) on others (e.g., region or community) and how characteristics of eHIE approaches in a specific region relate to implementation and outcomes (intermediate and ultimate) over time. For example, our stakeholder discussion, literature review, and case studies show that the federal and state environment or context has affected specific regional eHIE efforts and there is great variation across regions in both the level and types of eHIE. Conversely, promising initiatives and lessons learned about eHIE between LTPAC providers and their exchange partners can be used to help implement some ongoing HITECH programs (e.g., latest round of HIE cooperative agreement, 2015-2017 rule and Stage 3, use of Medicaid 90-10 matching funds) and other payment and delivery system reforms at the state and federal levels. As Yin has noted, one of the central contributions of case studies is to better understand what aspects of the environment or context are most important and how they affect planning, implementation, and outcomes over time.16 Others have also noted the importance of "multi-level" research and evaluation to better understand the complex environment, interactions, and outcomes of new programs or interventions.17
Finally, our conceptual framework includes both intermediate and ultimate outcomes. The ultimate outcome of interest (to the far right in the middle row) cannot be achieved until robust enough exchange occurs and this is likely to take time. That is, the volume and nature of exchange occurring clearly affects the ability to achieve improvements in care coordination and quality and reduction in total costs for the population served by the organizations in the region. Additionally, the ability to use and re-use exchanged data to achieve these ultimate outcomes requires robust but affordable technological solutions as well as workflow redesign and related education and training. While LTPAC providers and their exchange partners are in the early stages of exchange, as described in further detail in our finding section, we have identified some promising states and regions throughout the country where research and evaluation on intermediate and ultimate outcomes could potentially take place. Further information on possible outcome or impact analysis can be found in the "Impacts and Evaluation" portion of the paper.
This study sought to answer the research questions listed above by conducting a systematic review of the literature (peer-reviewed and gray) over the past five years, semi-structured discussions with key informants throughout the country (N=22), and site visits to two states (the Urban Institute team studied two initiatives in Minnesota, one in Pennsylvania) in the United States where eHIE involving LTPAC providers is more advanced in planning and early implementation activities compared to most areas of the country.
The Urban Institute team used each data source for a different purpose. The literature review was used to collect information on past and ongoing efforts to plan, implement, and evaluate eHIE involving LTPAC providers that was published over the past three years. The Urban Institute team built on this knowledge by engaging in discussions with key stakeholders in eHIE initiatives involving LTPAC providers across the United States. The Urban Institute team used these discussions to explore the latest planning, implementation, and early evaluation developments in eHIE involving LTPAC providers with an eye toward identifying two communities that could be the focus of more in-depth case studies. The Urban Institute team gained further more in-depth insight into the planning, implementation, and evaluation efforts of three eHIE initiatives involving LTPAC providers by conducting case studies of those initiatives.
The Urban Institute team used the information obtained through the targeted review of the literature and key informant discussions to also develop a conceptual framework for understanding how LTPAC providers and their trading partners prepared for the implementation of eHIE, and the impact of this exchange on clinical workflow, work force, and the quality, continuity, and cost of care the LTPAC providers and their HIE trading partners. The Urban Institute team used the information gained through the review of the literature, the stakeholder discussions as well as the newly developed conceptual framework to identify several sites of eHIE activating involving LTPAC providers that could be the subject of more in-depth case studies.
A systematic approach was used to identify and synthesize current literature (peer and non-peer-reviewed) on the planning, implementation, and impact of eHIE, particularly as they pertain to LTPAC settings, providers, and care coordination. First, in consultation with a research librarian, search terms were developed. Second, the search terms were applied to the databases EBSCOhost, Medline and Scopus to identify relevant literature. We initially focused our initial search to the last five years (2009-2014). We also conducted a targeted review of websites from government and professional associations to identify any relevant materials. Using these two literature review approaches, the research team identified a total of 2,021 articles for review.
Researchers then reviewed abstracts. Articles published prior to 2011 (the last three years of literature) that did not meet inclusion and exclusion criteria were eliminated, reducing the total to 303 articles. Finally, the 303 full articles were reviewed by members of the research team, and extraneous articles were eliminated. This process resulted in a final set of 74 peer-reviewed and non-peer-reviewed articles or materials. Appendix A contains citations for all considered (fully reviewed and included) articles.
The research team used the stakeholder discussions to explore the latest planning, implementation, and early evaluation developments in eHIE involving LTPAC providers with an eye toward identifying two communities that could be the focus of more in-depth case studies. Using information from informal discussions with federal officials overseeing eHIE programs and the non-peer-reviewed literature; the research team developed an initial list of informants. In total, 17 discussions were held with 22 individuals representing 12 regions around the country.
Using the information obtained through the literature review and key informant discussions as well as our conceptual framework (described above), the Urban Institute team identified several sites of eHIE activity involving LTPAC providers that could be the subject of more in-depth case studies. The sites selected for more in-depth case studies were chosen in part because their markets for exchange between LTPAC providers (especially NFs and HHAs) and their exchange partners were considered relatively mature. The Urban Institute team developed and used a number of criteria in conjunction with the project officer's input to select alternative case study sites. Overall, these criteria allowed the research team to assess the potential pros and cons of various regions/states, their potential complementarity, and generalizability of case study results. For example, we sought to identify a region with a relatively mature market for exchange between LTPAC providers (especially NFs and HHAs) and their trading partners so they had more experience on which to draw and perhaps some early insights into impacts. Additionally, we considered the technological approach using, including public HIEs, private HIEs, vendor networks, and portals, allowing us to capture the diverse ways eHIE is being achieved, technologies that are more likely to be scaled up and spread (rather than unique, homegrown systems), and the implications for things like workflow and impact. Finally, we sought to identify sites that would welcome the research team.
After reviewing and discussing that list, our HHS Office of the Assistant Secretary for Planning and Evaluation (ASPE) project officer selected three initiatives across two states for case studies and site visits: KeyHIE in the Northcentral/Northeast region of Pennsylvania and the Fairview-Ebenezer and Benedictine initiatives in and around Minneapolis, Minnesota.
In summer 2015, the Urban Institute team conducted 43 interviews with 47 respondents in site visits to Pennsylvania and Minnesota. In Pennsylvania, the team held 19 interviews with 21 respondents; in Minnesota, 24 interviews with 26 respondents.
After completing the site selection process, the Urban Institute team began identifying the key organizations and informants that should be targeted for interviews during the in-person site visits as well as the most appropriate points of contact for securing those interviews. The types of informants sought for interviews included:
Clinical personnel such as physicians, nurses, medical assistants, office managers, and EHR/HIT staff from providers participating in the programs in the selected sites (including NFs, HHAs, hospitals, organized delivery systems, and medical groups).
Local and state leaders.
Plan and payer representatives.
Other stakeholders with experience and knowledge of eHIE.
To identify the most appropriate organizations and respondents for site visit interviews, the Urban Institute team used several strategies. First, the team had informal discussions with key stakeholders including state staff and, in the case of KeyHIE, HIE leadership. Second, the Urban Institute selected participating providers for interviews with the goal of having some variation in the following characteristics:
- Ownership or type (e.g., hospital or system owned, non-profit or for-profit);
- Geographic setting (urban/suburban/rural);
- Size; and
- Teaching status.
Third, the Urban Institute team inquired about other people informants thought we should interview. Through this "snowball sampling" procedure, we built a more robust list of individuals and organizations potentially able to participate in interviews.
The Urban Institute team developed and received approval from the Urban Institute's institutional review board (IRB) to use four interview protocols targeting different types of respondents. The use of four protocols (rather than one) allowed interviewers to more easily direct questions to the most appropriate respondent. For example, we directed more technical questions about a provider's technology tools toward the organization's EHR/HIE lead, and workflow related questions toward that provider's clinical staff. Since the interviews were semi-structured, interviewers were able to ask other questions and probes as needed.
The Urban Team conducted the site visits in summer 2015 using two pairs of interview teams. Each interview team consisted of a lead interviewer and a note taker. Interviews were also recorded to facilitate polishing interview notes taken during the site visit and to produce interview notes for the interviews in which there were only audio recordings. The team conducted some interviews after the site visit due to scheduling conflicts and the geographic distance of some respondents. All interview notes were cleaned and analyzed at the end of each respective site visit. In addition, during the site visit, the team discussed preliminary findings, which covered key topics to describing findings from the site visits and was used to inform the site visit memoranda. The Urban Institute team also received follow-up materials from some respondents to provide additional information on the topics covered during the interview. Findings from each site visit were summarized in memoranda.
Findings from both site visit memoranda were synthesized in the final report to illustrate how the initiatives examined overlapped and where they differed in approach and outcomes in order to identify key lessons learned and potentially generalizable findings.
Description of Case Study Site: Minnesota
Our team studied two initiatives in Minnesota at varying stages of eHIE implementation and usage. The first was the Benedictine-Allina initiative. Exchange of patients' clinical information between the one participating Benedictine Health System (BHS) NF (St. Gertrude's) and the one participating Allina hospital (St. Francis) is performed by creating, sending, and receiving a Continuity of Care Document (CCD) using each facility's EHR. The two facilities are located on the same property, share many patients, but do not have the same EHR vendor. Exchange with other trading partners is still largely performed via fax.
The second initiative is the Fairview-Ebenezer initiative, which is in the planning phases and builds off previous grant-funded work. This project focuses on improving the eHIE capacity of providers serving Burnsville, Minnesota. Fairview Health Services is the lead organization of a collaboration that includes Ebenezer and Burnsville Emergency Medical Services (EMS). Fairview owns Ebenezer but not the Burnsville EMS. Fairview-Ebenezer eHIE implementation is pending award of the State Innovation Model (SIM) testing grant. In both initiatives, hospital portals are the most common way that LTPAC providers view patient data from trading partners, which only provides information about a patient's most recent acute care encounter and is only uni-directional exchange.
Minnesota was selected in part because it has a unique provider landscape. Most of the hospitals and physicians serving the region are owned or affiliated with one of several systems (e.g., Allina Health, Fairview Health Services) or multi-specialty group practices (e.g., Fairview Physicians Associates). There is also a high preponderance of LTPAC providers in Minnesota, many of which are part of senior service health systems including both initiatives studied (e.g., BHS and Fairview Health Services). Additionally, ACOs are common in Minnesota. For example, both Fairview and Allina are ACOs under CMS' Pioneer ACO Program.
Minnesota's approach to eHIE is also unique. Rather than supporting a state-wide HIE or regional public HIOs, the state has taken a market-based approach to HIE, granting funds to private organizations to stand up exchange within communities. This has created a relatively decentralized and market driven model that operates within the boundaries of an overarching state plan and regulatory framework.
Finally, Minnesota has implemented state laws to spur adoption of eHIE including a 2005 EHR mandate for most providers. There are also two active and influential associations in Minnesota for "older adult services" (LeadingAge and Care Providers) that have been instrumental in securing exemptions for post-acute care providers from the state's EHR mandate18 (described below) and will be contributing to the LTPAC Roadmap, which further defines the future of eHIE with LTPAC providers in Minnesota. Despite these favorable conditions, eHIE with LTPAC providers remains limited in Minnesota.
Description of Case Study Site: Pennsylvania
KeyHIE is a national leader in HIT. Founded in 2005. KeyHIE is one of the oldest and largest HIEs in the country. Originally under the umbrella of Geisinger Health Systems, KeyHIE is backed by decades of health care innovation and serves close to 4 million patients in the Northcentral/Northeast region of Pennsylvania.
KeyHIE currently has 18 LTPAC facilities connected, and has plans to bring on an additional 55 over the next three years as part of a grant from the HHS Health Resources and Services Administration (HRSA). KeyHIE offers participating providers three HIT solutions: KeyHIE (query-based) Transform, MyKeyCare and is now implementing Direct Secure Messaging (DSM).
KeyHIE was selected because the Northcentral/Northeast region of Pennsylvania is further advanced in implementing eHIE with LTPAC providers than most regions of the country. This is largely a result of the leadership provided by the Geisinger Health System as one of the initial sponsoring organizations for KeyHIE. Geisinger dominates the region; it currently shares data with KeyHIE and also extends its network through EpicCare Link, a portal for community providers.
Finally, KeyHIE was selected because of its development and use of the Transform tool, which takes MDS and OASIS data and converts the clinically meaningful information to a CCD. This CCD can be exchanged using KeyHIE so that the all participating provides could access the CCD. The Transform tool is inexpensive relative ($500 per year for facilities with 99 beds or below) to the cost of interfacing with an exchange, which appeals to LTPAC facilities who may otherwise not be able to afford to participate in information exchange. Use of the Transform Tool is spreading to other regions and communities (e.g., Colorado, Delaware, and Illinois).
Below, we discuss findings from the literature review, stakeholder discussions, and case studies related to
- Preparing for eHIE between LTPAC providers and their exchange partners;
- Implementing eHIE between LTPAC providers and their exchange partners; and
- Assessing the impact of eHIE between LTPAC providers and their exchange partners.