Usable Health IT for Physicians
Executive Summary
A physician-specific point-of-care system that continuously adapts to practice patterns can potentially result in dramatic improvements to the quality and efficiency of healthcare delivery.
The major objective of health IT should be to subtract work, not to add work or make it harder. Most health professionals do not use available health IT systems because current systems fail to offer value. In 2001, the Institute of Medicine issued a landmark report that stated “to improve quality in health care, health care professionals needed to interact effectively and efficiently with the health IT systems.”
Unfortunately, the report's corollary, that health IT systems effectively and efficiently support clinical users, has not necessarily been the case. Clinical IT systems have adoption rates that are typically less than 15 percent. A recent HIMSS report, released by its EHR Usability Task Force in June 2009, concluded: “Electronic medical record (EMR) adoption rates have been slower than expected in the United States…. A key reason is lack of efficiency and usability of EMRs currently available.” Others have concluded that the current health IT efforts may even set back the vision of 21st century healthcare.
Underlying the conclusions of the above reports is the fundamental relationship between perceived value of an IT system and both the usability and utility to its intended users. A recent commercial for the Apple iPhone provides an important exemplar. A couple is finishing dinner and decides to go to the movies. With a few quick finger flicks, the couple identifies a movie they want to see, locates a nearby theater, and buys the tickets, all in less than a minute. The iPhone delivers immediately useful information-utility-and requires minimal effort by the user with almost no training-usability.
This kind of innovative IT is needed in health care. The iPhone's rapid and pervasive success is a clear example of how utility and usability are inextricably intertwined to provide value to the user. This goal can be met in healthcare with a Smart Point of Care (POC) concept that is designed to specifically address value for clinical users by providing immediate utility within an easy-to-use pervasive computing environment.
The Smart Point-of-Care System
Imagine a physician support system at the point of care (POC) that:
knows and uses the physician's context-where he or she is, what patient is being seen, what set of tasks need to be performed-based on locally relevant outcomes and measures;
supports all the coordination and scheduling tasks that the physician must “orchestrate” the patients;
is customized based on what information is entered, what the physician needs to see and what he does and closely replicates the way he thinks;
moves from device to device-installing automatically on whatever device is being used, scaling to the display, hardware capacity and operating system; and
connects securely to whatever source of information is required, whether electronic records, results or reference literature.
Clinicians are avid adopters of useful information technology and just want a product that can help them do everything they need to do at the point of care.
The fusion of efficient, best clinical practices and patient information at the point of care will directly support improved quality of care, and produce cost savings that have not been realized by current health IT systems. Savings and improved quality of care can never be realized if clinicians won't put the data in. Contrary to some conventional wisdom, clinicians are avid adopters of useful information technology and just want a product that can help them do everything they need to do at the point of care. Thus a clinician's work environment that actively supports intelligent provision of clinical data and information to and from its clinical users could become a highly leveraged interface to health and other systems that are needed to support the full spectrum of health services delivery.
Such a system could transform the healthcare sector and realize the true potential of useful IT. This work environment would support its clinical users by reducing the time it takes for purely administrative tasks while providing relevant clinical information and knowledge to the point of care, thus increasing the time available for a clinician to think about all the data and information about a patient and then, thoughtfully, address the patient's problem. It is not intended to increase the number of patients a physician or nurse can manage per hour, per shift or other measure of “productivity.” Although the overall timeline is shortened, the time saved is now available for listening to and talking with the patient.
The Smart POC system described above has other key attributes. It anticipates the clinician's needs and has data and information available before it is needed. It understands the clinician's context-dependent workflow. It also hides all the complexity of underlying health IT systems with simplicity-“magical” IT.
This context-aware Smart POC implements a systems engineering approach for the collection, distribution and maintenance of best practices, clinical data and system performance.
A conceptual architecture for Smart POC support is shown in the figure. The three components operate within a service-oriented architecture and exchange data within the Smart POC and to external information sources, such as local hospital information systems, health information exchanges (HIE) and knowledge sources, using standardized messages.
The heart of the architecture is the context/task manager (C/TM). It monitors user's activity to determine context, uses models of user's tasks and current/expected context to anticipate activities, tasks and necessary data exchanges with the user interface manager (UIM) and the information broker (IB) components, and maps user activities and tasks to the most appropriate support application for a true extensible software-as-a-service framework.
The IB component is the data/information cache for its users as well as the connection point to external systems. The set of services required by the IB is available in many commercial HIE or service-oriented architecture offerings from vendors. Both the C/TM and IB have analytic engines that monitor the efficacy and efficiency of user and system tasks versus outcomes to continuously enhance best practices and system performance. The UIM component presents relevant data, information and medical knowledge to clinicians and gathers data from them. It has presentation strategies to achieve communication goals that depend upon current context, criticality of message and device being used, adapts to the unique style of the clinician and provides a consistent set of metaphors regardless of the clinician's location. The Smart POC system is designed to automatically present relevant clinical data and information via pre-filled clinical widgets; offer executable patient care plans; unobtrusively collect patient data; and generate relevant charge or billing information as a byproduct.
This context-aware Smart POC implements a systems engineering approach for the collection, distribution and maintenance of best practices, clinical data and system performance. It uses clinician-specific and continuously-adapting practice patterns that have the potential to dramatically enhance the quality and efficiency of health service delivery. The systems approach addresses the very thorny and expensive issue of how to make practice guidelines/best practices relevant to local context and, at the same time, solves the “How can we maintain and evolve the practices that we have implemented?” question.
A well-designed medical home, and its underlying IT systems, will support physicians doing their job and reap the benefits of more effective delivery of care, and satisfied and healthy patients.
The built-in business intelligence and analytic tools provide clinicians and managers with a new mechanism that changes the focus from “what's been done” to “what should be done,” based on context and outcomes. This near real-time feedback loop simultaneously provides analyses for informed decisions about: what is best for my patients; what is best for our community, our state and our (population-level) nation; and best practices. Thus, the Smart POC enables an integrated evaluation framework that supports continuous feedback of outcomes, cost and benefit directly to the point of care-where treatment decisions are made by the clinician and the health care consumer.
User Expectations
Today's technology-aware users expect their devices to be nearly magical-like the iPhone described above. In many cases, fairly robust versions of software products are free, like Google Analytics and many web conferencing systems, or the international phone service Skype. As a result, today's users expect to access whatever information one needs, wherever and however one needs it, and execute useful transactions with no learning curve. The gold standard is instant access to online banking services, from an iPhone, in a seamless interface that just works to pay one's bills and manage one's finances.
Furthermore, a number of technologies have transformed business sectors other than health. Agile companies are realizing reduced costs, increased revenue, faster time-to-market, and increased customer satisfaction. They are capitalizing on Internet-enabled applications such as dynamic supply chains, customer relationship management, and emerging web services opportunities. Concurrent with that trend, cloud computing, vast grids of always-on computing resources, is fundamentally changing how companies purchase these IT components and services. No longer are businesses being held captive by expensive and proprietary hardware and software. They can access inexpensive, best-of-breed systems and services that are priced as commodities in the fast pace of Internet time.
New Practice Models
The U.S. health care system is a very large, $2+ trillion enterprise with many diverse business units. There is significant pressure to continue focusing on “sick care,” or treating disease. New approaches to delivering “healthy care,” such as the medical home (MH) efforts, are focused on coordinated, patient-centric care that integrates all the health services for their clients. A well-designed MH, and its underlying IT systems, will support physicians doing their job and reap the benefits of more effective delivery of care, and satisfied and healthy patients. Most of these capabilities are not available in any existing EHR system.
In addition to supporting the clinicians, there is a strong expectation that the MH will be the connection point for all interactions between the health consumer and his or her required medications, specialists and diagnostic studies. These users, like the iPhone consumers, have come to expect a high level of transparency and access to critical information in other aspects of their lives-particularly financial information. The health consumer in a MH environment will transition from a passive patient told what to do to a fully engaged and active partner in his or her care.
Thus, a successful MH must engage and reward all participants, including the end user, the health care consumer. Capitalizing on both of these trends through an intuitive Smart POC for consumers, will support the desired outcomes, with healthier people receiving efficient, targeted and appropriate health services. The consumer version of the Smart POC will actively engage users to access their own information, understand the choices they have for care, actively manage their own chronic conditions, and participate as a partner in the healthcare system, rather than being at the mercy of it. They will have immediate access to what are the best practices, what practices are not effective or not safe, and what practices are more expensive without added value.
The value of the Smart POC is precisely enabling value-added information exchanges for all participants wherever and whenever they make decisions.
Importantly, the Smart POC system and its IT infrastructure will enable the MH to collect appropriate clinical, administrative, and patient outcome information as a by-product of providing and orchestrating health services. The customer portion of the MH web collects relevant information from individual participants as well as patient monitoring data from instrumentation in the consumer's home. As a result, best practices, local clinical guidelines and clinical decisions will be linked directly to patient outcomes.
To leapfrog today's inefficient practices, the MH must include a world-class set of analysis tools so that each practice can not only provide the most effective and efficient health services to its members, but can also continuously improve those health services based on immediate feedback from its outcomes. Measuring progress and assessing successful or unsuccessful outcomes across all dimensions-consumers, the MH practice and the IT infrastructure-is fundamental to this revolutionary approach. Results will be fed back frequently to all involved parties for their evaluation; successful ones will be accelerated and MH sites could adopt best clinical and management practices while avoiding IT solutions or practices that do not work.
These data, the IT infrastructure and associated clinical and business intelligence tools come together as an innovative technology platform. The resultant positive and negative feedback loops define a true cybernetic system in which the outputs of the health system-outcomes, appropriateness and costs-can directly and immediately affect the health system's performance. Thus, MH clinical staff, management and consumers will know at the point of care-whether that is in the office, the car, a place of business or at home-what are the best practices, what practices are not effective or not safe, and what practices are more expensive without added value.
This potentially innovative technology platform and cybernetic system is the missing link in comparative effectiveness research (CER). Since the diffusion of clinical practice guidelines into clinical practice has been notoriously slow and updating and evolving practice guidelines expensive and very difficult, CER functions must be embedded into the daily functioning of a MH at the point of care in a way that it is used continuously by all participants. The value of the Smart POC is precisely enabling value-added information exchanges for all participants wherever and whenever they make decisions. This information continuously informs decisions by all participants so they can adjust their local practices and behaviors to improve their MH performance. As a result, MH with the Smart POC can become a key model for implementing local and national CER data at the point of care when and where decisions are made.
This approach for the medical home is designed to address the maintainability and sustainability of guidelines. Guidelines are implemented within the Smart POC system, then continuously adapted, evolved and communicated to the local practice setting by feeding back the MH's outcomes, costs and utilization data and new biomedical knowledge onto the guideline itself. It should be a fascinating story for the science of CER to observe and analyze the time-oriented adaption and evolution of guidelines, both within and across communities and special populations. After all, as the famous Canadian physician Sir William Osler once stated: “It is much more important to know what sort of patient has a disease than what sort of disease a patient has.”
We believe the Smart POC system and CER form the basis of true meaningful use that may transform healthcare. We would like to thank Dr. Joseph Jasinski, Director, Healthcare and Life Sciences at IBM Research, for his support of the Smart POC efforts.
John Silva is president of Silva Consulting Services, Eldersburg, Md.; Nancy Seybold is owner of Quartermark Consulting Washington, D.C., and Marion Ball is senior advisor, Healthcare and Life Sciences Institute, IBM Research and Professor Emerita, Johns Hopkins University, Baltimore, Md. Healthcare Informatics 2010 July;27(7):40-43
