The experience of Partners HealthCare offers some unique insights into the process of electronic medical record adoption across a large, diverse health system.
Originally posted to AJMC.com
In 2002, Partners HealthCare (Partners) launched a strategy to accelerate the use of ambulatory electronic medical records across its network of 6000 physicians. Through focus on quality software products, creation of a system of financial incentives, and active engagement of health system leadership, Partners reached high levels of physician adoption by late 2006. Partners eventually introduced a mandate that made ambulatory electronic medical record use a requirement for all of its physicians. During this multi-year initiative, Partners also focused on the effective use of electronic medical records and introduced a series of tactics designed to optimize the use of these systems. With introduction of the meaningful-use concepts in the Health Information Technology for Economic and Clinical Health (HITECH) Act, Partners will transition its efforts toward this important national priority. Partners' experience offers some unique insights into the process of electronic medical record adoption across a large, diverse health system.
(Am J Manag Care. 2010;16(12 Spec No.):SP9-SP12)
In October 2002, the leadership of Partners Community Healthcare, Inc. (PCHI) met to discuss the future of its physician network. Over the prior 8 years, education and communication had been the primary tools used by the network to drive its quality and utilization improvement efforts. Believing that these techniques had reached their natural limits, the leadership was looking for a “game changer”—a new approach that would leapfrog over current efforts. They considered an aggressive electronic health record initiative as the next step.
Partners Community Healthcare was accustomed to taking on big, audacious goals. In 1994, the organization was founded after the creation of Partners HealthCare (Partners), the integrated delivery system formed by the merger of Massachusetts General Hospital and Brigham and Women’s Hospital. The role of PCHI in this new delivery system was to create a physician network in eastern Massachusetts and, once formed, to support those physicians as they sought to improve clinical care and manage utilization. By 2000, PCHI achieved its first major goal by building a network of 6000 physicians, including more than 1000 primary care physicians.
The network created from these efforts was anything but homogeneous. The physician groups varied from private practices to independent practice associations, physician hospital organizations, and large, academic physician organizations. Some had fewer than 10 physicians, and some had hundreds of physicians. Each had a strong primary care presence in its local community, and some also came to the network with a number of specialists. Many were employed, but affiliation grew to become the dominant physician relationship model.
Could this heterogeneous physician network achieve systemwide adoption of electronic medical records? Was this the game changer that would drive significant improvements in care quality and utilization? Partners had a history of success with healthcare information technology. As one of the earliest adopters of computerized provider order entry, Partners’ hospitals experienced firsthand the value of information technology for patient safety. The leadership of PCHI considered whether it could achieve similar results with information technology in the ambulatory environment.
Implementing ambulatory electronic medical records throughout this network would be a big undertaking. Although approximately two-thirds of the physicians at the founding hospitals used electronic medical records in 2002, use of electronic medical records in community practices was minimal. Only 9% of community primary care physicians and barely 1% of community specialists used electronic medical records. On top of this, the community physicians were not organized in a way that facilitated any large-scale roll-out. These physicians were spread out across eastern Massachusetts in small practices. Although a handful had 10 or more physicians, the vast majority of these community practices were solo practices or groups of 2 or 3 physicians.
Through discussion and debate, PCHI leadership came to understand the challenges of this ambulatory electronic medical record effort and resolved that “successful deployment of electronic medical records will be critical to differentiate the PCHI Network in terms of quality and efficiency.” Thus, PCHI’s electronic medical record initiative was launched.
Getting It Done
Following the decision to pursue this strategy, a task force was created to spell out the tactics. This task force returned with its recommendations in February 2003. These recommendations were organized along 3 dimensions: product features, financial incentives, and leadership.
1. Product Features. The task force felt strongly that interoperable electronic records were essential to the organization’s success. Also, they realized that the current state of electronic medical record software was such that interoperability was difficult and perhaps impossible. Therefore, the task force recommended that the network limit the choice of electronic medical records. This approach would ensure the quality of the software selected, preserve some degree of choice for private physicians, and offer the hope of some level of interoperability.
2. Financial Incentives. The most frequently mentioned obstacle to adoption of electronic medical records was cost, with emphasis on the up-front capital costs to acquire the software, hardware, and services to complete the transition from paper to electronic records. The task force recommended that PCHI work with the region’s payers and leverage the existing pay-for-performance contracting platform to create financial incentives for this effort.
3. Leadership. Although the task force’s recommendations for leadership were less specific, the message was clear: all levels of management must emphasize the importance of electronic medical records. From the top Partners’ executives to the smallest practice’s leadership, the organization had to communicate the value of electronic medical records and the importance of this effort to the network’s future success.
With the task force’s recommendations in hand, PCHI launched its effort in 2003 with an initial focus on primary care physicians. Physicians were given information on the selected software products, were offered practice-specific analysis of the return on investment, and were encouraged (although some might describe it as “hounded”) to make a decision. The results of the effort were dramatic. A year into the effort, the number of community-based primary care physicians using electronic medical records had doubled. The following year, it doubled again. It doubled yet again the year after that. Within 3 years, community primary care physician use of electronic medical records had gone from 9% to 65%. At that point, the benefits of another tactic became evident.
Early on, the task force recognized that electronic medical records were well suited to primary care practice, but were a more difficult fit for specialists. For that reason, the task force recommended that the rollout strategy focus first on primary care physicians. Once primary care physicians were among the converted—literally and figuratively—PCHI should leverage the primary care physician base to attract and motivate specialists. Thus, in 2006, with roughly two-thirds of the community primary care physicians using electronic medical records, the number of community specialists showing interest in this effort started to rise. For many, the motivation was the opportunity to implement a new technology that would improve patient care. For some, it was the fear of losing referrals from primary care physicians who might find it easier to refer to a colleague who could share information electronically. Whether the motivation was the opportunity for improved care or the fear of business loss, the specialist adoption levels started to rise. By the end of 2006, 17% of the community specialists had adopted electronic medical records and another 30% had selected their software product and were starting the implementation process.
Throughout this initiative, senior leadership had often wondered whether a mandate for electronic medical records was an option. However, in the early years when adoption levels were low, such a mandate might appear too out of step with the physician network. The discussion became focused on how long the current system of incentives should continue and when the optimal time would be to transition from these“carrots” to “sticks.” This moment seemed to arrive in early 2007.
Creating the Mandate
With the 2006 year-end results in hand, the PCHI network had much to be proud of. Among primary care physicians, including those in the community as well as those at the founding hospitals, 76% had completed the transition to electronic medical records, and more than 60% of specialists had done the same. In March 2007, PCHI leadership considered whether the time had come to switch from carrots to sticks. The leadership drew on Everett Rogers’ Diffusion of Innovations.1 In this work, Rogers describes diffusion as “the process by which an innovation is communicated through certain channels over time among the members of a social system.”
To PCHI, the spread of electronic medical records represented such diffusion. According to Rogers’ theory, members of the social system accept the innovation at varying rates. The latest group to adopt often is referred to as “laggards,” those individuals who resist change, respect tradition, and often change only when forced to do so. The PCHI leadershiphad reached the point where a mandate would be necessary to achieve complete implementation of electronic medical records. In March 2007, PCHI leadership set a series of deadlines by which physicians would need to choose their preferred electronic medical record and complete its implementation. The goal was to complete the implementation of ambulatory electronic medical records by December 2009.
The remaining 2 years of the initiative moved swiftly. Although the mandate presented a few challenges, it seemed that most of the network understood the reasons behind the move and made the transition. For roughly 3% of the network’s membership, the mandate forced a choice that led these physicians to leave the network.
Adoption Is More Than Acquisition
With all of the activity associated with acquiring a system, converting paper records, training staff, and going “live,” the initial focus was not on how these systems were being used or whether they were collecting data needed to support quality improvements. However, midway through this 7-year process, PCHI started to look at these issues and, not surprisingly, found that there was plenty of room for improvement. Leveraging the tactics that achieved so much success with adoption, PCHI turned its attention to effective use of its electronic medical record systems.
The leadership team knew that achievement of effective use was an important goal. After all, didn’t they buy into this promise of electronic medical records years ago? With leadership support, the discussion focused on which elements of electronic medical record use should be emphasized. To address this issue, PCHI looked at 2 dimensions:
1. Clinical Relevance. Across all of its initiatives, PCHI had learned that in order to be successful in changing physician behavior, the desired change had to be clinically relevant and meaningful to patients. Demanding change in the absence of these 2 conditions rarely produced a sustainable change in physician actions. So if PCHI was to focus use of electronic medical records on specific activities, these activities had best be ones closely associated with improvements in patient care.
2. Frequency. Any reporting effort that identifies lapses in quality, safety, or uses of technology must happen often enough to produce reliable measures. The rare medical record documentation event, while clinically important, would not generate enough opportunities for appropriate electronic documentation to produce a meaningful measure of any physician’s performance.
With these 2 dimensions in mind, PCHI identified a series of high-frequency, clinically relevant events that require specific actions in the electronic medical record: computer-based prescribing, accuracy of problem lists, and smoking documentation. PCHI began to measure these events at the practice and individual-physician levels. Later, as with electronic medical record adoption, PCHI leveraged its pay-for-performance platform to create incentives for improvement in these areas. Through this process, PCHI leadership drove significant improvements in key areas of electronic medical record use.
Transitioning to Meaningful Use
With the release of the final rule for meaningful use, the Centers for Medicare & Medicaid Services have provided PCHI with the next stage in its electronic medical record evolution. What started with acquisition and migrated to a focus on effective use of healthcare information technology will now transition to meaningful use of these systems. With the federal focus on terminology and exchange standards, the promise of interoperable electronic medical records will be realized.
The physicians of the PCHI network are well positioned to achieve meaningful use. With PCHI’s early start on acquisition in 2003 and an evolving focus on effective use of electronic medical records beginning in 2006, PCHI physicians have all the building blocks for meaningful use. However, they will need to master more than just a few elements of electronic medical record use to be successful. The 3 or 4 areas of emphasis will grow to 20 or more.
As it has through the prior 2 phases of electronic medical record adoption, PCHI will help its physicians achieve meaningful use by communicating the significance of these efforts, providing assistance to physician practices, and, where appropriate, identifying mandates to ensure the highest possible levels of performance.
Health System Leadership Regardless of Size
Throughout this experience, Partners has learned a great deal about what it takes to manage a large-scale health information technology project and to work with a diverse group of clinicians to transform work flows to take advantage of this technology. These lessons include providing high-quality products, creating financial incentives that are aligned with the cost of implementation, and offering services and support to assist with the migration.
The nation’s population of community physicians is a health system of tremendous scale. However, some of the same principles that guide a small system like Partners also will guide the nation. The Office of the National Coordinator for Healh Information Technology (ONC) has developed a strategy that begins with certified interoperable electronic health records and couples those products with meaningful financial incentives to accelerate the adoption of healthcare information technology. Through the leadership of the ONC and its regional extension centers, the nation’s physicians also will have access to information and services to support them through this transition. These tactics, proven at the smaller health system level, are certain to have a similar impact at the federal level.
What remains unclear is whether the notion of a mandate will emerge from the federal program. A mandate was necessary for Partners to achieve full adoption of electronic records. The hardened core of resistant physicians would never have made the decision to adopt (or leave) without that dictate. As the nation pursues meaningful use, it too will face a moment when it has to decide whether to introduce some form of mandate. Without that mandate, it is possible the nation will never truly meet its health information management goals.
Author Affiliations: From Partners HealthCare (CLB), Needham, MA; and Partners HealthCare (THL), Boston, MA.
Funding Source: The authors report no external funding for this article.
Author Disclosures: The authors (CLB, THL) report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.
Authorship Information: Concept and design (CLB, THL); analysis and interpretation of data (THL); drafting of the manuscript (CLB, THL); critical revision of the manuscript for important intellectual content (CLB, THL); and administrative, technical, or logistic support (CLB).
Address correspondence to: Cynthia L. Bero, MPH, Partners HealthCare, 115 Fourth Ave, Needham, MA 02494. E-mail: email@example.com.
1. Rogers EM. Diffusion of Innovations. New York, NY: Free Press; 1962.