Peer-Reviewed and Non–Peer-Reviewed Publications Included in the Systematic Literature Review
| Author (Reference) | Publication Type | Summary/Conclusions |
|---|---|---|
| Frisse & Holmes (2) | Economic impact model using original data | The authors estimated ED cost savings attributable to an HIE and concluded that the combination of reduced duplication of laboratory tests, reduced duplication of radiology tests, lower ED expenditures, lower costs for communication distribution, and reduced inpatient hospitalizations resulted in a total estimated savings of $20.2 million over 5 years. Cost estimate for HIE implementation and maintenance over the same period was $14.2 million. |
| Walker et al (3) | Cost-benefit analysis | The authors reviewed the literature and interviewed experts to create conceptual and analytic (software) models of varying levels of HIE across multiple stakeholders. Costs included creating interfaces (reconciling diverse codes, data structures, and terminologies) and acquisition, training, and maintenance costs. Projected benefits included reduced delays and costs for ordering tests and managing results, prescribing medication, handling chart requests, and referrals. Conclusion: a fully standardized national HIE could yield a net value of $77.8 billion per year. Although national standards have not be defined or adopted, the business case for universal standards is strong. |
| Overhage et al (6) | Research (qualitative) | One hundred thirty-four RHIOs provided data about their HIE initiatives in 2004. Eighty percent of respondents involved outpatient, primary care, and specialty care in their HIEs. The authors identified challenges including standards, architecture, and sustainable business models. |
| Afilalo et al* (7) | Research (cluster-RCT of 23 family medicine practices) | ED notes accessed via secure, web-based, electronic information system showed advantages over ED notes sent by regular mail; physicians received information more often, found it more useful, and initiated actions more often. No significant difference in the patient follow-up rate. |
| Lang et al (8) | Research (cluster-RCT of 23 family medicine practices) | The study evaluated the impact of an electronic link for clinical data transfer between ED and family physicians’ offices. Physicians were assigned to get electronic ED notes over the internet versus usual procedure (mailed information). There was no reduction in repeat visits to the ED, length of stay, or other admissions. |
| Grossman et al (9) | Research (qualitative) | The Community Tracking Study conducted more than 1000 semistructured interviews in 12 US communities. Findings: The majority of hospitals (70%) are implementing portals for clinical data sharing with affiliated physicians. Hospital executives see portals as a competitive strategy that is appealing to physicians. HIE among unaffiliated providers is rare. Competition, cost, and control of data are barriers. Sustainable business models are a major challenge for RHIOs. |
| Adler-Milstein et al (10) | Research (survey of 145 RHIOs) | Few RHIOs were financially self-sustaining. Of those known to exist in July 2006, nearly 1 in 4 was defunct by early 2007. Early successes involved exchange of test results. Sources of support included fees from participants and grants. |
| Burton et al (11) | Review | Although the focus of the review is EHRs for coordinating care of people with multiple chronic conditions, it also deals with HIEs as “coordination of care across settings” in communities. Barriers: (1) no common format or standard for recording clinical information, (2) high costs of implementation and maintenance, (3) no demonstrated clinical and/or financial benefits for ambulatory care physicians participation, (4) patients’ concerns about information sharing and possible loss of privacy, (5) physicians’ concerns about legal liability. The authors offer 3 recommendations essential for HIE: (1) physicians and their professional leaders should agree on a common health record, such as the Continuity of Care Record; (2) regional governance structures that encourage the exchange of clinical data should be established; and (3) insurers and managed care plans should pay for completing EHRs. |
| McGowan et al (12) | Research (qualitative) | Interviews with IT leaders developing 2 health information networks in largely rural Vermont. Early and broad collaboration is key. Challenges include ownership, governance, costs versus value, and a fear the state might assume control because of their initial investment. Few rural physicians have health IT training and most are not fully aware of HIE potential. |
| Solomon (13) | Descriptive | Case studies of 3 RHIOs (in Massachusetts, Indiana, and California). RHIOs can be catalysts for change by assuring privacy, engaging purchasers of care, and creating incentives for clinicians. Barriers: ownership and privacy concerns and physician resistance and worry about how information will be reused to measure clinical performance. |
| Middleton et al (14) | Web publication | Provides guidance about the critical issues required for communities to address to establish a successful HIE. |
| Anonymous (15) | Article; non–peer-reviewed | To make RHIOs work, provider stakeholders must be using EHRs, and HIEs must create practical value by focusing on workflow integration and impact. |
| Frolich et al (16) | Editorial/commentary | The Santa Barbara project was funded in 1999 by the California HealthCare Foundation to establish a peer-to-peer, community-wide HIE. It closed in 2006. Lessons learned included: deal with privacy and liability concerns and develop a sustainable business model; an incremental approach is preferred; start with small goals that are valued by providers, eg, laboratory test result sharing; legacy systems are built to enter and display data (not share with other systems) and there is a lack of standardization of existing data, so technology development is slow; and alternative methods such as local hospital portals and direct laboratory feeds to providers are quicker to establish. |
| Ash & Guappone (17) | Review (qualitative research methodologies) | Because most HIE initiatives are in the early stages of development, formative evaluation is recommended. The authors describe the way qualitative research methods (interviews and focus groups) can elucidate the motivations for HIE participation and assist in HIE development efforts. |
| Bates & Gawande (18) | Review | This review focuses on Health IT's potential to improve safety. HIE can decrease errors that result from inadequate access to clinical data, can rapidly communicate dangerous results, monitor adverse events, and provide decision support. Barriers to adoption include high costs, lack of standards/privately licensed standards, and cultural barriers (eg, clinicians and policymakers tend to see IT as unimportant). |
| Simon et al (19) | Research (qualitative) | Cross-sectional comparisons of physicians in the Massachusetts e-health collaborative (n = 355) and others randomly selected in the state (n = 794). Benefits: physicians in the HIE demonstration site were more likely to report positive effects from health IT and were more likely to expect health IT to control costs. Barriers: time, costs, productivity loss, and lack of standards and technical support. |
| Simon et al (20) | Research (qualitative) | Cross-sectional comparison of EHR adopters and nonadopters. Eighty percent of physicians can view laboratory reports, 47% can order laboratory tests, 45% can transmit prescriptions to a pharmacy electronically. Adopters were significantly more likely to perceive benefits such as controlling costs, improving quality, and having access to information. Fewer adopters are demoralized with the state of medical practice in general. Small practices are less likely to use the advanced functions available in EHRs. It is important to move beyond EHR adoption into meaningful use and HIE. |
| Grossman et al (21) | Web publication | This addresses barriers to various stakeholder participation, including data providers, data user, and data funders, and strategies to address the identified barriers. |
| Marchibroda (22) | Review | The eHealth Initiative has helped more than 20 states create more than 250 HIEs. The author concludes that the biggest problems are assessing the value of HIE services and converting those values into business plans for stakeholders. More HIE is being supported by grants and contracts from states. |
| Congressional Budget Office (23) | Web publication | This addresses the evidence of costs and benefits of health IT, barriers to broader health IT use, and possible options for the federal government to promote health IT use. |
| Center for Community Health Leadership (24) | Web publication | This identifies and discusses 10 critical issues for HIE success, from identifying a value proposition to measuring success and ROI. |
| First Consulting Group (25) | Web publication | Ten HIE challenges are identified, including governance, leadership, and stakeholder participation. |
| Rosenfeld et al (26) | Web publication | Summarizes the then-current state of HIE activity in the United States. |
| Hagland (27) | Article; non–peer-reviewed | Discussion of the role of nursing homes as key stakeholders in RHIOs. |
| Kern & Kaushal (28) | Descriptive | Description of New York state's HEAL NY program of health IT and HIE. First phase involves adoption of EHRs, electronic prescribing, and community-wide clinical data exchange. There will be systematic evaluation of outcomes and the financial return on investment. |
| Mattocks et al (29) | Research (survey) | Connecticut eHealth surveyed physicians in the Connecticut State Medical Society-Independent Practice Association and found that 467 practices (61%) were connected to at least one hospital information system, sharing laboratory (42%) and radiology (42%) results, discharge summaries (37%), and pharmacy data (15%). Most common barriers were cost (71%) and time required (39%). Primary care providers and small practices (≤2 physicians) were more likely to cite cost than specialists or larger practices. |
| Doebbeling et al (30) | Review | Based on literature synthesis and expert opinions, the authors present potential strategies for IT management and HIE implementation. Key dimensions are people, process, and programs (the majority of implementation is social engineering and only 20% to 25% is technical implementation). Benefits: HIEs could eliminate unnecessary testing, improve safety, facilitate efforts to improve quality. Barriers: structural, financial, policy-related, cultural, and organizational. Financial burdens include implementation costs, slow and uncertain payoffs, and disruption of clinical practices. |
| Basch (31) | Letter | The author contends that in the Walker model (3), cost savings to providers are erroneously overestimated, especially for small practices. For example, the model uses activity-based costing that fails to consider how personnel in small practices typically multitask, and that administrative savings in small practices would require downsizing to “less than zero” staff. |
| Anonymous (32) | Article; non–peer-reviewed | This is a report about Dr. David Brailer's speech at the 2006 Annual Health Information Management and Systems Society conference. |
| Terry (33) | Article; non–peer-reviewed | Focuses on the costs of RHIOs and who should be paying for them. |
| Anderson (34) | Review (literature review and secondary analysis of surveys) | Physicians believe there are benefits to electronic health IT; however, overcoming barriers will require subsidies and performance incentives. Main barriers are lack of capital, complexity of systems, need for privacy safeguards, and lack of data standards for HIE. |
| Reed & Grossman (35) | Web publication | Discusses the wide variation of health IT adoption by physicians and identifies characteristics associated with adoption. |
| Hersh (36) | Editorial/commentary | The author describes HIE as “the anytime, anywhere access to clinical care information across traditional business boundaries.” Benefits: improved health care quality and patient experience, an estimated financial savings of $87 billion per year after initial investment. Barriers: lack of financial incentives, especially for early adopters. Privacy is a perceived barrier, although in reality paper records are no more secure than electronic. |
| Holmquest (37) | Editorial/commentary | Reflecting on experience of the Santa Barbara HIE, the author believes the key issues are who should pay for HIE and what model will support initial costs of constructing it? |
| Ghosh & Marquard (38) | Review | RHIOs can be collaborative knowledge networks, serving as forums for regional exchange of evidence-based knowledge and not just data. Primary barriers are fear of change among physicians, cost, the proliferation of proprietary technologies that cannot exchange data, and lack of standards. |
| Frisse (39) | Descriptive | Describes the formation of the MidSouth eHealth Alliance in Tennessee and summarizes lessons learned: creating a broad coalition can be challenging if it impinges on existing local HIEs, so governmental involvement and oversight may be necessary; some groups may push their own connectivity efforts to create business advantages for themselves or their vendors; the added technology demands of HIE can stress an organization's internal IT resources; it is important to identify “quick wins” such as e-prescribing; it is essential to align laws and financial incentives toward harmonization rather than fragmentation. |
| Hripcsak et al (40) | Expert opinion (conference proceedings) | This is a summary of a 2006 meeting where experts discussed how to evaluate HIEs and developed a series of ordered steps for the evaluation process, including (1) platform evaluation, (2) usage studies, (3) assessment of the immediate business case, (4) assessment of clinical and administrative impact, (5) unintended consequences, (6) comprehensive return on investment, and (7) program evaluation. |
| eHealth Initiative (41) | Web publication | This discusses the findings from the eHealth Initiative's 2008 survey of US HIEs. |
| Terry (42) | Article; non–peer-reviewed | Discusses the benefits experienced by physicians using the Indiana HIE. |
| Foldy (43) | Research (survey of 16 HIE projects in Wisconsin) | Seven HIEs deliver data only to centralized registries for public health or quality surveillance, 1 delivers information only to clinical care providers, and 8 do both. Two also deliver information to patients. Fewer than half use data standards for interoperability. Internal barriers were funding, organizational and staff issues, governance, and technology. External barriers were marketing, enlisting participants, regulatory issues, and sustainability. |
| Drazen et al (44) | Web publication | This is a review of published experiences with HIEs that also examines case studies from 2 Massachusetts’ HIEs. |
| Marchibroda (45) | Review | This review focuses on the ways in which health IT and HIE can improve care of patients with chronic disease toward the goal of “timely, effective, equitable, patient-centered care.” Clinical data resides with local organizations and collaborations require trust and building of “social capital”; thus a national network will need to be built from the ground up. HIE has been provider-centric but needs to involve health plans, employers, and others involved in chronic care management. Improved patient engagement can be achieved through personal health records and HIE. |
| Kaelber & Bates (46) | Review | HIEs have potential to improve patient safety through improved processing of medication information such as allergies, doses, drug–drug interactions, drug–diagnosis match; timely notification and appropriate follow-up for laboratory and radiology information; improved communication from provider to provider and provider to patient; improved public health surveillance. A few studies show HIEs can actually jeopardize patient safety (incorrect information made available to providers, wrong patient, errors in translating from one system to another). |
| Kilbridge & Classen (47) | Review | This review focuses on the opportunities and challenges with health IT applications for patient safety. It provides 7 major recommendations for interventions at the federal level. |
| Noblin (48) | Research (secondary analysis of surveys) | Privacy concerns are a key deterrent to widespread acceptance of RHIOs. National Consumer Health Privacy Survey of 2005 (2000 people) found 76% are somewhat or very concerned about privacy of their PHI and feel EHRs are less secure than paper records. Specific concerns include unauthorized sharing or redisclosure of protected health information, medical identity theft, and fear of discrimination based on health-related conditions. Federal HIPAA guidelines are limited and state confidentiality laws vary greatly, potential problems when RHIOs cross state lines. Markle Foundation Survey 2006 found 75% of American feel the government should establish rules to provide protection. |
| Halamka et al (49) | Editorial/commentary | The authors propose universal adherence to a basic framework of policies and standards for information sharing among regional networks, the Common Framework, which will address (1) consistent data standards, (2) privacy and security, (3) pluralism (all local systems and all levels of sophistication must be accommodated), (4) accuracy, and (5) flexibility. The authors suggest a decentralized and federated model for HIE, where PHI remains in the hands of patients and their providers. |
| Kuhn et al (50) | Expert opinion (conference proceedings) | This record of a planning meeting among representatives from several nations highlights the universal potential of HIE. It provides valuable details regarding opportunities in common as well as problems stemming from national differences. |
| Anonymous (51) | Article; non–peer-reviewed | |
| Burstin & Clancy (52) | Editorial/commentary | HIE across providers and settings could help automate frustrating and time-consuming tasks, eg, locating records and reports. |
| Carter et al (53) | Expert opinion (practice guideline) | The guideline defines HIE models: federated model with shared repositories, federated model with peer-to-peer network, nonfederated peer-to-peer network, and centralized database or data warehouse. Guiding privacy principles are openness and transparency; purpose specification and minimization; collection limitation; use limitation; individual participation and control; data integrity and quality; security safeguards and controls; accountability and oversight; remedies. Legal and operational issues include the need for HIE participation agreements, HIPAA regulations. |
| Christoper & Jensen (54) | Web publication | |
| Corey & Grossman (55) | Web publication | |
| Deloitte Center for Health Solutions (56) | Web publication | |
| Diamond & Ricciardi (57) | Review | Connecting for Health, a public–private collaborative, proposes a common framework of rules for Internet-based HIE. Model policy documents address issues such as patient consent, privacy, security, control, and access. |
| eHealth Initiative (58) | Web publication | |
| Grossman & Reed (59) | Web publication | |
| Heinold & Albritton (60) | Web publication | |
| Johnson & Gadd (61) | Review (evaluation methodology) | The authors describe evaluation measures that align with stages of development of an HIE. HIE initiatives must be integrated into the workflow of clinical practice to achieve benefits. |
| Lawrence (62) | Article; non–peer-reviewed | |
| Metzger & Zywiak (63) | Web publication | |
| Miller & Miller (64) | Research (qualitative) | More than 40 interviews of participants in the Santa Barbara, California, HIE project. This pioneering RHIO failed after 8 years when its perceived value did not match the cost. Technology delays and liability concerns constrained progress. A combination of grants, incentives, and mandates may be necessary to develop and sustain RHIOS in providing unprofitable but socially valued services. |
| Office of the National Coordinator for Health Information Technology (65) | Web publication | |
| Reed & Grossman (66) | Web publication | |
| Zafar & Dixon (67) | Descriptive | The Indiana Network for Patient Care has provided consistent, secure clinical data exchange for more than 10 years. The authors describe the security policies, federated data-sharing model, standards, and component-based architecture. Organizational factors associated with success were incremental evolution, knowledgeable human staff, and leadership's ability to make a clear, evidence-based business case. |
* Lang et al (8) discusses same study.
RHIO, regional health information organization; ED, emergency department; IT, information technology; HIE, health information exchange; EHR, electronic health record; HIPAA, Health Insurance Portability and Accountability Act; ROI, return on investment; PHI, personal health information; RCT, randomized controlled trial.