National interoperability, the ability for systems and software to exchange information across the United States is a goal sought by various stakeholders in the health care field. The dramatic increase in the amounts of data available due to electronic health records and health information technology adoption, creates opportunities for greater access to patient data, availability of more comprehensive health records, and improved care coordination among various providers (1–3). Despite the rise, health information exchange is still siloed in health systems, hospitals, and within state lines. The Patient Centered Data Home (PCDH) is an initiative to connect these silos using a network of health information exchange (HIE) organizations. This study aimed to characterize the PCDH initiative through the lens of the Heartland Region pilot.

Patient-centered care seeks to strengthen patient-clinician relationships, promote communication, and empower patient involvement in health care (4). To reach effective patient-centered care, access to accurate and timely information for informed health decision-making is pertinent (5). In addition to the lack of information at the beginning of a visit, studies indicate only half of patients in the emergency rooms receive timely follow-up after discharge. Patients who do not receive follow-up care are more likely to have worsening conditions and return to the emergency department (6–13). Additionally, because all health care is not local, patches in patient information are inadequate for proper patient-centered care. The wide adoption of electronic health records improves the availability of patient health information to inform decision-making during a patient visit. Follow-up care coordination also benefits from this adoption (2). Although advances in governance and patient consent have improved the exchange of EHR data, it is still frequently siloed within hospitals and health systems.

Silos of health information exchange are characterized as data systems that do not exchange data with other similar systems (14). Challenges in maturity such as, the lack of adequate infrastructure, electronic health record data integration, technical support, security features, and training for staff, add to the disconnection and lack of exchange between health systems (15). The lack of supportive infrastructure through HIEs is especially elevated in rural areas (16). Lack of clarity in the return on investment, leadership investment, readiness and ability of HIEs, and privacy and security concerns also hinders the exchange of health information (15, 17).

Health information exchange on a global scale was examined in China, England, India, Scotland, Switzerland, and United States. Most of the countries had a level of implementation of electronic records, exchange of those records were fully or partially implemented at some levels (19). The authors indicated that the implementation of health information exchange needs a desire to optimize the quality and outcomes of healthcare with relevant, timely, and accurate patient information at the point of care. As these countries are mostly developed countries, more research is needed to characterize health information exchange in developing countries. The authors also concluded that supportive structures, such as economic, political, and cultural, are needed to combat challenges for health information exchange (19).

Challenges with health information exchange has seen relief through Health Information Exchange (HIEs) organizations that serve as secure, comprehensive storage and exchange centers for patient data from various health and community organizations. HIEs use a variety of technological approaches to improve provider access to patient information both collected and maintained by other organizations. Exchange of information by HIEs has been timely, comprehensive, and addresses threats to quality, safety, and efficiency (20). HIE benefits include reduced duplicate procedures, improved patient safety, and lower costs (12, 20). Regional HIEs expanded the scope of exchanging health data electronically but are still limited to regions and states.

In order to facilitate electronic health exchange, accurate patient matching is necessary. Lack of consistent evidence based guidelines for patient matching has been a continuous barrier to improving interoperability (21). Patient matching algorithms use patient demographics and patient identifiers to link records (22–25). Independent entity agreement, cohesion on data governance, and the appropriate process of patient matching itself has been a source of challenge. Nicknames, married names, security, merging and unmerging data, updates from hospitals and HIES, and strengths and limitations to understand each HIE, can slow the patient matching process. Patient matching has the potential to enhance patient care and ultimately patient satisfaction, reduce costs from duplicated testing, inform innovation through patient outcomes, and identify fraudulent activity. However, the challenge with quality and effective patient matching is standardization (25). At a national scale, patient matching can increase continuity of care among multiple providers.

In 2009, Healthbridge, a Health Information Exchange Network in Cincinnati, Ohio, and the Indiana Health Information Exchange (IHIE), in Indianapolis, Indiana, developed an “electronic postman” approach reducing the use of fax to exchange clinical health data between Indiana and Ohio (26). In 2012, The Sequoia Project began management of a nationwide health information exchange, eHealth Exchange, which focused on governmental agencies' exchange of health information (27). Also directed by The Sequoia Project, Carequality expanded interoperability nationwide including various non-governmental agencies, such as private care providers, hospitals, and clinics (28). Recently, the National Coordinator for Health Information Technology began the Trusted Exchange Framework and Common Agreement to link electronic data sharing islands (29). Vendors, such as EPIC and Cerner, have enlarged the scope of health data exchange beyond their systems. Although there is successful data exchange on these larger known scales, there are many data exchange initiatives that operate in silos without connection to others and are not widely known.

The purpose of this study was to characterize the PCDH initiative from the perspective of Indiana Health Information Exchange's participation in the Heartland Region Pilot (described below). This manuscript describes one approach toward national interoperability using an interconnected network of HIEs to exchange patient health information across state lines while accommodating individual HIE governance and standards. Although in its beginning stages, PCDH has shown operating potential as a pathway toward national interoperability. This manuscript explored the exchange and data generated through this beginning process. Although there has been some discussion on PCDH, this is the first published evaluation of the PCDH initiative. Additionally, conducting an overlap analysis and patient matching to our characterization of this data are unique and key features of this study.

The structure of the remainder of this paper includes an origin of PCDH and the Heartland Pilot, the process of PCDH, explanations and results of data collection, zip code validation, overlap analysis, chief complaint classification, patient matching, and analyses of these data.

The Strategic Health Information Exchange Collaborative (SHIEC) “is a national collaborative representing health information exchanges (HIEs) (18)”. With over 70 HIEs, SHIEC covers more than 200 million people across the United States (18). SHIEC having this coverage, implemented three PCDH pilots: Western Region, Central Region, and Heartland Region. The Western Pilot includes Arizona Health-e Connection, Quality Health Network in Western Colorado, and Utah Health Information Network. The Central Hub Pilot includes MyHealth Access Network in Oklahoma and Texas, and Arkansas State Health Alliance for Records Exchange. The Heartland Region includes 7 HIEs in 5 states: East Tennessee Health Information Network in Knoxville, Tennessee; Great Lakes Health Connect in Grand Rapids, Michigan; HealthLINC in Bloomington, Indiana; Indiana Health Information Exchange in Indianapolis, Indiana; Kentucky Health Information Exchange in Frankfort, Kentucky; Michiana Health Information Network in South Bend, Indiana; and The Health Collaborative in Cincinnati, Ohio (Figure 1). The pilot began in September 2016, by supporting alerting and sharing summary information among these seven HIEs. In January 2018, all three regions were connected, demonstrating a next step toward national interoperability (18).

In the PCDH process, the HIE that services the patient's residential zip code is designated as the storage center and “home” HIE. PCDH's goal is to store a comprehensive, longitudinal patient record in the “home” HIE; in where all data sent and received becomes part of the home record. At the beginning of the PCDH pilot, zip codes from the HIE service area were shared among the HIEs in the Heartland region. The PCDH process is triggered when a patient visits another medical facility not in their “home” HIE service area, called an “away” HIE (Figure 2). The facility in the “away” HIE queries the “home” HIE requesting access to available patient information. If available, the “home” HIE generates an HL7 version 2, Admission, Discharge, Transfer (ADT) message that contains pertinent medical information. Once the encounter is completed, the “away” HIE sends a summary ADT to alert the home HIE that there are new records available. The PCDH process allows for comprehensive medical histories to follow a patient wherever treatment is sought. In the future, the care documents will be stored in the home HIE for care providers to see and select as appropriate.

The study received protocol approval from the Indiana University Institutional Review Board and the Indiana Health Information Exchange securely released the data for analysis. The sample for this exploratory study was collected December 2016 through December 2017. Using HL7 version 2 ADT messages, summary data was both sent and received by the Heartland Region HIEs. The HL7 version 2 messages were de-duplicated to identify unique encounters for the PCDH initiative. Once de-duplicated, specific fields were parsed for analysis: sending facility, receiving facility, message date, patient date of birth (year, month, day), patient gender, patient social security number, patient state of residence, chief complaint or reason for visit, and patient class (referred to as visit type). Transactions missing the parsed fields were excluded from analysis due to the inability to definitively identify them as an encounter.

Using the deduplicated and parsed ADT data, a zip code validation and overlap analysis was conducted to determine the coverage areas of the HIEs. A chief complaint classification to determine the reasons for the visit, and deterministic probabilistic patient matching were applied to the data for further characterization of the data (Figure 3). Further explanation of each can be found below.

The results indicate a robust exchange rate between December 2016 and 2017. The findings show the messages sent and received between the Heartland Region, and demographics, such as, ages, sex, facility types, chief complaints, and patient matching.

The PCDH Heartland Region pilot generated 1,492,367 ADT messages. IHIE sent an estimated 805,994 ADT messages and received 686,373 ADT messages. Table 3 indicates the direction and volume of ADT messages sent and received by IHIE during the Heartland Region PCDH pilot. The majority of the messages were sent from MHIN, KHIE, and HL from IHIE, while IHIE received the most messages from MHIN, HL, and THC.

In characterizing the PCDH initiative from the perspective of Indiana Health Information Exchange's (IHIE) participation, findings indicate that the PCDH framework is practical and feasible to exchange data. Building upon past attempts and knowledge from failed initiatives, the PCDH initiative has found a path to make national interoperability actually work.

The value of this data extends beyond clinical practitioners to the public health workforce for improved interventions, increased surveillance, and greater awareness of gaps in health for needs assessments. We expect that SHIEC's existing interconnection of HIEs will provide a sustainable path toward national interoperability. In January 2018, all three regions were connected, demonstrating a next step toward national interoperability (35). In June of 2020, North Carolina celebrated being one of the newest HIEs to join the PCDH national initiative in the Midwest region, which includes Alabama, Missouri, Kansas, and Wisconsin. Between April 2020 and June 2020 the HIE exchanged over 11,000 ADT notifications with “18 HIEs for out-of-state patient visits” (36).

PCDH uses a “bottom–up” approach to interoperability. Future examination of the impact of PCDH and use in the clinical setting as well as for population health will be necessary. Future work research could include integration of other systems, such as those with social determinants of health features, studying other pilots across the nation, including various medical, dental, and pharmaceutical areas, and ways to continue to improve the process toward sustainable methods of national interoperability.

At the completed phases, PCDH plans benefits of centralization through a national ADT alerting and identifying assurance, accurate gap analyses, precise quality measures, option for centralized patient consent management, and patient access to their entire record in one location. With low costs and scalability PCDH seeks to continue to expand and include all levels of HIEs. Each HIE's policies, technology and values will be honored by preserved governance, sustained management processes, honored data use agreements, maintained privacy and consent models, unchanged business models, and conserved technical architecture. During this study, ADTs were the only alerts being transferred but the process had begun to send and receive continuity of care documents for a more comprehensive record in individuals' home record.

The Heartland Region found success in having a peer-to-peer networking, having one gateway hub in the region to route to other gateway hubs, and utilizing a tiered process to achieve national interoperability. PCDH has moved away from the idea that national interoperability will happen one way only and accepts the possibility of all ways. Despite challenges, PCDH has found actual success toward national interoperability. Currently, SHEIC has reported 45 HIEs in PCDH, with over 177 million total event notifications exchanged by HIEs (37). Future research should include an evaluation of the current PCDH and outcomes, and potential implications for multinational interoperability.

The datasets presented in this article are not readily available because Data was given by Indiana Health Information Exchange and would need permission to access. Requests to access the datasets should be directed to https://www.ihie.org/contact-us/.

KW created the first draft of the manuscript. Both authors contributed to manuscript revision, read, and approved the submitted version. Both authors contributed to the conception, design of the study, and performed the statistical analyses.

KW was supported by a training grant (Award Number T15NLM012502) from the National Library of Medicine of the National Institutes of Health.

The content is solely the responsibility of the authors and does not necessarily represent the official views of the CDC, NIH or HHS.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.