This Article Abstract Full Text (PDF) All Versions of this Article: M2671v1 15/3/267    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Google Scholar Articles by Glaser, J. P. PubMed PubMed Citation Articles by Glaser, J. P.

Seven Durable Ideas

Department of Information Systems, Partners Healthcare, Boston, MA.

^_* Correspondence: John Glaser, PhD, Partners Healthcare, 800 Boylston Street, Boston, MA 02199 (Email: jglaser{at}partners.org).

Received for publication: 11/26/07; accepted for publication: 02/13/08.

	Abstract

Top Abstract Introduction Durable Ideas Centrality of Processes Organizational Partnership Progressive Incrementalism Agility Architecture Embedded Research Engage the Field Conclusions References

 
Partners Healthcare, and its affiliated hospitals, have a long track record of accomplishments in clinical information systems implementations and research. Seven ideas have shaped the information systems strategies and tactics at Partners; centrality of processes, organizational partnerships, progressive incrementalism, agility, architecture, embedded research, and engage the field. This article reviews the ideas and discusses the rationale and steps taken to put the ideas into practice.

	Introduction

Top Abstract Introduction Durable Ideas Centrality of Processes Organizational Partnership Progressive Incrementalism Agility Architecture Embedded Research Engage the Field Conclusions References

 
I have had the privilege of being a Chief Information Officer at Partners Healthcare for 20 years; 12 years at Partners and, prior to that, 8 years at Brigham and Women's Hospital. (Partners Healthcare is an integrated delivery system founded by the Brigham and Women's Hospital and the Massachusetts General Hospital in 1994. Partners has since grown to encompass ten hospitals and 7,000 physicians. In fiscal year 2007, Partners had revenues of $6.4 billion, and had 4,000,000 outpatient visits and 160,000 admissions.) Over that period of time, Partners and its affiliated hospitals have made significant advances in implementing a broad array of clinical information systems. Inpatient computerized provider order entry has been implemented in all ten of its hospitals, and its outpatient electronic medical record is being used by 3,000 attending physicians. Partners is one of the world's leaders in the application of clinical decision support.

Partners has established highly regarded medical informatics, medical imaging, bioinformatics and telemedicine research groups. Partners Information Systems (IS) staff have served on a wide range of boards of industry associations and academic and government committees. Application software from Brigham and Women's Hospital formed the basis of Eclipsys, and significant partnerships with HP and Siemens are at the center of our current clinical information systems agenda.

Over the course of these two decades, the evolution and accomplishments of the Partners IS group have been guided by a small number of durable ideas. These efforts follow and leverage a rich tradition of information technology innovation at the Brigham and Women's and Massachusetts General Hospitals.^1–3

	Durable Ideas

Top Abstract Introduction Durable Ideas Centrality of Processes Organizational Partnership Progressive Incrementalism Agility Architecture Embedded Research Engage the Field Conclusions References

 
A durable idea can be a philosophy, a belief, or a value. A specific idea guides a wide range of management decisions. For example, the idea of agility is based on the belief that it is important to be able to quickly and efficiently respond to change. This idea can guide decisions regarding architecture, application sourcing strategies, and the design of the IS organization.

At a high level, the ideas have undergone little change over the course of 20 years. The strategies and tactics adopted to implement these ideas have changed, at times significantly. In addition, the organization's understanding of the power and nuance of the ideas has deepened.

The perspective of durable ideas can be found in management literature that has examined factors that enable sustained information technology excellence by organizations.^4,5

Seven ideas have provided the management framework for the work done by Partners IS; centrality of processes, organizational partnerships, progressive incrementalism, agility, architecture, embedded research, and engage the field.

	Centrality of Processes

Top Abstract Introduction Durable Ideas Centrality of Processes Organizational Partnership Progressive Incrementalism Agility Architecture Embedded Research Engage the Field Conclusions References

 
An imperative of any IS group is defining, designing, implementing, and supporting applications that improve core organizational processes. Regardless of condition, processes will be imperfect and the objectives of a process will change as care practice, reimbursement, and regulatory demands and service expectations change.

At times the IS department views its primary role as acquiring, implementing, and supporting applications. Applications are only relevant to the degree that they enable improvements in core processes. Processes are central. Applications are not. This perspective has had several ramifications.

Application Project and Support
The ability of any application to support a process is clearly dependent on the organization's understanding of the target process, how well it has engaged application users in the design of an application's process support, the capabilities of the application, and the quality of training and ongoing support. Application project teams must be composed of IS and user staff who have a very good understanding of the target processes and are competent at process change. High-performance application development/acquisition, implementation, and support staff and practices form the backbone of any IS organization. Efforts to improve this performance must be a permanent item on the IS management agenda.

Intelligence
In health care, clinical processes have an unusually high dependence on information and knowledge. Effective medication ordering can be very dependent on knowledge of medication interactions and generic substitution opportunities.

A critical objective of virtually all clinical applications is the delivery of decision support that enable processes to be intelligent. This support requires infrastructure such as event and rules engines. It requires well-structured data for critical data types such as medications. It also requires organizational mechanisms to identify new decision support logic, understand the fit between workflow and decision support interventions, evaluate the effectiveness of existing logic, and maintain logic currency.

Assessment and Measurement of Processes
Many IS goups, especially those with strong academic credentials in health services research, can be significant contributors to the assessment and measurement of the impact of information technology interventions on processes.

Assessment and measurement of process have greater power to persuade providers of need for process change and information technology (IT) interventions if they are conducted with good scientific design and evaluation rigor. In addition, process assessment can help determine whether the IT intervention accomplished its goals or whether further work is needed.

	Organizational Partnership

Top Abstract Introduction Durable Ideas Centrality of Processes Organizational Partnership Progressive Incrementalism Agility Architecture Embedded Research Engage the Field Conclusions References

 
Applying information technology to improve care requires a multifaceted partnership between the IS group and the rest of the organization. There is an abundance of literature, and our own experience that suggests that the quality of the partnership has an impact that is greater than any other variable on the ability of an organization to be successful in applying information technology. This partnership is centered on two areas.

Agenda Development
The IS agenda must be formed. This agenda identifies which initiatives will be pursued and why. The agenda formation also identified budgets, relative initiative importance, initiative governance, and implementation timing. The organization must consider agenda development to be legitimate, thoughtful, competent, and transparent. There is no "correct" mechanism. Individual hospitals have different mechanisms ranging from IS Steering Committees to informal conversations between members of hospital leadership.

The leadership must appreciate the messiness of agenda development. Ideally this agenda is unambiguously linked to an organizational strategy and set of goals. In practice the linkage is partial and imperfect; there are often initiatives undertaken because of nothing more than experienced management instinct. The effectiveness of agenda development is based on superior working relationships between organizational leaders who trust and respect each other.

Project Management
To be successful, a specific project requires a solid partnership. The IS staff must bring project management, technology, workflow assessment and re-engineering, application design, training, and support prowess to the table. Project members from the clinical community must bring an understanding of their accountability for the success of the project, thoughtful guidance of major project decisions, and commitments to undergoing the process changes that are rarely easy. Clearly clinicians can also be very competent at project management.

Ongoing efforts to improve project management prowess are necessary, and there should be means to do "post mortem" reviews of projects that got into trouble. In addition, IS management attention is often focused on the health of the project and the health of the relationships between project team members.

Although the IS staff must be skilled, seasoned, and at times patient, their ability to contribute to projects is helped significantly if they have had solid exposure to clinical practice and operations. Staff who have been nurses, physicians, medical technologists and technicians bring an understanding and experience that is of significant value.

	Progressive Incrementalism

Top Abstract Introduction Durable Ideas Centrality of Processes Organizational Partnership Progressive Incrementalism Agility Architecture Embedded Research Engage the Field Conclusions References

 
Progressive incrementalism recognizes that material organizational IT-based progress may be best achieved through a series of modest steps that occur over years. This approach does not preclude an occasional "big bang" implementation, but such implementations should occur only when there is no practical way to reduce the big bang to a series of modest steps.

This idea is based on two considerations. First, big bang implementations are inherently risky; the risk of an IT project being determined by its size, breadth and depth of operational change and the newness of the technology. Although the organization might survive the implementation, it has a high likelihood of seeing nontrivial budget and timetable overruns. A series of smaller projects reduces risk.

Second, it is very difficult to fully understand the important features of an application or the nature of the process changes needed by large-scale projects. A series of smaller projects enables the organization to progressively learn. The organization understands the impact of Phase N and then can alter its course before it embarks on Phase N+1.

Two major strategies have been used to achieve this idea.

Break It into Pieces
The development of project plans for large-scale initiatives, such as CPOE implementation across multiple hospitals, should have discussions devoted to developing answers to the following question, "How do we break this initiative into pieces such that each piece is of modest size, the sequence of pieces can be arbitrary, and an indefinite period of time can separate pieces?" No large project can be perfectly broken apart, but most (but not all, e.g., a new payroll system) can be decomposed to modest-sized phases.

Identifying Readiness
Progressive incrementalism can refer to the decomposition of large projects. It can also refer to moving a broader agenda to the next step. For example, the organization can believe that it should move toward standard clinical systems across the enterprise. However, the organization may not be ready for complete standardization. Nonetheless, can it take steps toward standardization today? Further steps can be taken later as environmental changes and organizational comfort leads the organization to be ready for the next step.

This progressive implementation is a form of organizational evolution. It requires very good instincts about readiness and the ability to frame an initiative that takes steps that are not too far ahead of the organization and are not too timid. At times, the IS group can sponsor pilots or demonstrations of new technologies to help the organization determine whether it is ready or not. Lack of readiness can often be due to ignorance.

	Agility

Top Abstract Introduction Durable Ideas Centrality of Processes Organizational Partnership Progressive Incrementalism Agility Architecture Embedded Research Engage the Field Conclusions References

 
As the world changes, the capabilities of applications must change and new applications and technologies must be implemented. The application and infrastructure response to change needs to be as efficient and rapid as possible and to enable reasonably quick recovery from responses that were mistakes. Improving the agility of the infrastructure, applications, and IS organizations is an important and persistent goal.

Infrastructure agility is largely achieved through two tactics.

Leverage Standards
The adoption of industry standards enables the organization to have a degree of "plug and play" and to take advantage of the offerings of often large numbers of companies that support the standard.

Core Technology Assessment
The IS organization needs a function and approaches to experimenting with new infrastructure technologies, e.g., Radio Frequency Identification, in an effort to understand the maturity, capabilities, and potential value of the technology.

Application agility is achieved through three tactics.

Internal Development
Application agility traditionally has been achieved through internal development. Although internal software development has its own set of challenges and limitations, it can provide the ability to quickly and efficiently change the application to meet specific demands of the organization. Internal development can be particularly important when the mature form of the application is not well understood and the organization believes that application excellence will be critical to the organization's ability to achieve an important objective; hence control over software direction is important.

Rapid Feedback
Application agility can be enhanced through obtaining rapid user feedback on application design and early efforts to use the applications. Application prototypes and the implementation of preliminary versions of an application into a clinic setting can provide guidance on needed changes. Rather than wait the year needed for complete design and development, early versions of an application are released within months of project initiation.

Service-oriented Architecture
More recently, the growth in the capabilities of service-oriented architecture infrastructure and applications has led to the prospect of composite applications. Composite applications will introduce agility because the significant re-use of services enables the quicker and more efficient crafting of new applications.

Any IS strategy that shortens the time required to deliver an application or reduces the cost of application acquisition and implementation enhances agility. To the degree that an organization can implement more applications per unit of time or per dollar spent, the application throughput potential has been increased. Increased throughput enables the organization to execute a response to a larger number of changes in its environment.

Organizational agility is achieved using two tactics.

Streamline Decision Processes
Academic health centers and large integrated delivery systems can have management styles that are rich with committees and emphasize consensus. Nonetheless, the organization can be made more agile through efforts to streamline committees, shorten the budget development process, and ensure clarity of decision-making rights. For example, organizations that approve capital budgets quarterly have greater agility than organizations that approve annually.

Chunks
To the degree possible, all projects should be structured such that they have reasonably independent phases that can each be implemented within relatively short time intervals. This enables the organization to shift direction at the end of any particular phase. For example, the implementation of an Electronic Medical Record can be done in phases that could be accomplished within two years or over the course of ten years. The pace difference can result from decisions to pause EMR implementation while IS efforts are directed to some other set of projects.

	Architecture

Top Abstract Introduction Durable Ideas Centrality of Processes Organizational Partnership Progressive Incrementalism Agility Architecture Embedded Research Engage the Field Conclusions References

 
Information technology architecture is the set of organizational, management, and technical strategies and tactics used to ensure that systems have important characteristics, e.g., high reliability, and capabilities, e.g., are extensible into a patient's home.

Poorly architected systems can be the great disabler. If systems are inflexible, prone to obsolescence, difficult to integrate, or too expensive, they can damage the organization. On the other hand, systems that are fast, reliable, efficient, and capable of reasonably gracefully capitalizing on technology advances can be great enablers. Creating and managing well-architected systems requires three approaches.

Great Architects
There are some IS positions for which the difference between a B player and an A player is very significant. IS staff who oversee infrastructure and clinical systems architecture are some of these. These staff possess a command of high-level architectural concepts and a ground-level understanding of the capabilities of current technologies. A premium is placed on attracting and retaining architects.

Processes, Rules, and Standards
Architecture requires that standards be set for data, communications, operating systems, application vendors, and other components of the infrastructure. Some of these standards are suggestions, whereas others are mandates. Processes are needed to arrive at and evolve standards and to review proposed new technologies and applications to determine standard conformance.

Ability to Recognize and Leverage Major Technology Advances
From time to time, major advances in technology occur. These advances enable applications that may not have been practical (or imaginable) before or significantly improve the potency of an application. Networked personal computers, clinical decision support, the Web, and service-oriented architectures are examples of major technology advances.

Not all advances are technology per se. At times, the advances are management understandings and processes. For example, knowledge management has become critical in recent years.

An indication that an advance is at hand is when the concepts and language used to describe the advance are different than concepts and language that have been used before. For example, applications as a term is being replaced by the phrase "composites of services." Clinical decision support discussions often include the phrase "the computer knows"—a phrase that may not have been used before in discussions of clinical systems.

	Embedded Research

Top Abstract Introduction Durable Ideas Centrality of Processes Organizational Partnership Progressive Incrementalism Agility Architecture Embedded Research Engage the Field Conclusions References

 
Vibrant and well-integrated medical informatics research programs have been critical contributors to the accomplishments at Partners. Programs have been established, over the years, in core medical informatics, bioinformatics, connected care (also known as telemedicine), evaluation, and medical imaging.

These programs have provided a thoughtful and rigorous means to understand how to most effectively apply information technology, e.g., assessing approaches to clinical decision support based on genetic test results.

The programs have served as a talent nucleus for IT for areas that are important and often emerging. For example, it is clear that personalized medicine will have a significant impact on the practice of medicine and will increasingly be an area of fertile medical research. However, there is much to be learned about how best to apply the technology to these areas. The program can bring together a critical mass of talent that will not only further the organization's understanding of the area, but also lead the development of organizational competency in the area.

The research programs serve as a terrific staff recruitment tool. These programs not only attract individuals who will be core members of the program, but also will attract IS staff that want to do work that supports these programs. Moreover, virtually all employees want to be proud of the organization they serve. Knowing that their IS group does cutting-edge work is an intangible but potent recruitment and retention mechanism.

Although there are differences in knowledge domain and expertise across these groups, there are two characteristics that are common.

Physician Leadership
These groups are led by physicians with varying degrees of formal and informal information technology training. These physicians all practice part time, and most of them have non-IS administrative responsibilities, e.g., managing a division or responsibility for a specialty training program. Leadership by a practicing physician helps to ensure the internal credibility of the program with the medical staff and that the program is attuned to the realities of clinical practice, education, and research.

Embedding Research Programs in Operations
These programs have research and operational responsibilities. Although they have obligations to obtain grants and publish, they also have some level of responsibility for ongoing operations, i.e., they have responsibilities for managing areas such as supporting telemedicine operations or implementing new Picture Archival and Communication System technologies. The research is strengthened by the understanding of operational realities, and the operations are strengthened by the insight and experimentation of the research.

These programs have budgets that are partially funded by organizational operating budgets and partially funded by grant and contract support. This diversity of funding sources helps to ensure the financial permanence of the group; shortfalls in one source do not inherently imperil the group. The duality of responsibilities also helps to ensure that the organization perceives that IT research is relevant to its core care responsibilities, which eases the challenge of ensuring that the research programs will receive appropriate operating and capital budgets.

Embedding these groups has required that their leadership and staff have formal titles in the IS organization and are integrated into routine IS processes and management conversations; they are an integral part of the IS team. These managers and staff are involved in discussions of budgets, implementations of financial systems, major infrastructure changes, and changes in personnel policies.

	Engage the Field

Top Abstract Introduction Durable Ideas Centrality of Processes Organizational Partnership Progressive Incrementalism Agility Architecture Embedded Research Engage the Field Conclusions References

 
The IS group overall, but in particular the research groups, are expected to engage the broader health care industry. Engaging the field means that they are encouraged and expected to publish, present at conferences, serve in leadership roles in professional associations, and become members of commissions and task forces that are addressing critical topics in the field. Engaging the field is intended to achieve several objectives.

The staff will learn from their industry colleagues and develop deeper and more sophisticated understandings of both the industry's challenges and approaches to addressing those challenges. This knowledge will serve the organization well. Moreover, they will develop relationships that will invariably be beneficial. Effective external working relationships are often a predicate for getting something done of internal importance.

Many staff enjoy the opportunity to engage the field. By providing staff with the time, encouragement, and resources needed, the organization has helped to increase the likelihood of staff retention. In addition, our staff are often terrific, although unintentional, recruiters. It is common for a person applying for a job to note that they heard one of our staff give a speech that impressed them and led them to decide to seek employment with us.

As a management exercise, creating the ability to engage the field does not require significant effort. Two areas must be addressed.

Create Capacity
Budgets for travel and memberships must be established. In addition, staff must be given time to pursue these activities. From time to time, some staff must be reminded that, while engaging the field is important, they do have a day job back home.

Incentives
Many staff will naturally gravitate to these opportunities. In an academic health care setting, these opportunities are often needed if staff are to achieve academic promotion. For staff who aspire to climb the management ladder, these opportunities are often necessary preparation for additional management responsibilities. For some staff members, engaging the field is often a component of their performance review.

	Conclusions

Top Abstract Introduction Durable Ideas Centrality of Processes Organizational Partnership Progressive Incrementalism Agility Architecture Embedded Research Engage the Field Conclusions References

 
These seven durable ideas have had a significant influence on the progress made at Partners Healthcare. The strategies and tactics used to implement the ideas have changed as the organization developed a more sophisticated understanding of each idea.

The ideas have not prevented us from making mistakes. In some cases, we were too aggressive with centralization of some IS functions, leading to service degradation and increased expense. At times we have been too tolerant of individual hospitals pursuing different solutions for common problems, leading to unnecessary heterogeneity. We have been guilty of letting research initiatives dominate too much of an application's development agenda, and at other times, we have let operational demands inhibit our ability to do important research.

In addition to realizing that ideas will not prevent mistakes, I have come to appreciate that the IS leadership, particularly the role of CIO, must understand four things, one of which is the value (and limitations) of durable ideas.

Vision
The IS leadership must have the vision and perspective to see broad opportunities and challenges. They need not see the specifics, but they should see the major implications of areas such as reimbursement based on care performance, the progressive advent of personalized medicine, and the potentials of interorganizational interoperability. Once the implications are visualized, they must be communicated in ways that educate and inspire others and enable the formation of the initiatives and projects necessary to achieve that vision.

Engineering
The IS leadership should understand that improving care through the use of IT is a different form of engineering. Often one views engineering as the design and construction of tangible, highly specifiable outputs. Engineering complex organizations is different and is similar to engineering the path that water takes down a valley. You cannot engineer the actions of each water molecule, but you can engineer the contours of the valley such that the water, in aggregate, behaves a certain way. Similarly, engineered organizational surroundings will result in the actions of many individuals generally moving in the desired direction.

Staff
The IS leadership must appreciate that the most significant asset that they have is talented, motivated, organized, and resourced staff. These staff will accomplish amazing things if they are set up properly, nurtured, and protected from some of the harsher aspects of organizational life. IS leadership must always remember that it is never about "you," it is always about "them."

	References

Top Abstract Introduction Durable Ideas Centrality of Processes Organizational Partnership Progressive Incrementalism Agility Architecture Embedded Research Engage the Field Conclusions References

 

1. Slack WV, Bleich HL. The CCC systems in two teaching hospitals Int J Med Inf 1999;54:183-196.[CrossRef][Medline]
2. Barnett O. Computers in medicine JAMA 1990;263:2631-2633.[CrossRef][Medline]
3. Bates DW, Lo HG. Patients, doctors, and information technology: clinical decision support at Brigham and Women's Hospital and Partners HealthcareIn: Greenes RA, editor. Clinical Decision Support The Road Ahead. London: Elsevier; 2007. pp. 127-142.
4. McKenney J, Copeland D, Mason R. Waves of Change: Business Evolution Through Information TechnologyBoston, MA: Harvard Business School Press; 1995.
5. Weill P, Broadbent M. Leveraging the New InfrastructureBoston: Harvard Business School Press; 1998.

This Article Abstract Full Text (PDF) All Versions of this Article: M2671v1 15/3/267    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Google Scholar Articles by Glaser, J. P. PubMed PubMed Citation Articles by Glaser, J. P.