Electronic Medical Records for the Department of Health Services (2003)

From Nhs It Info


The Future Lies Ahead, The Past is History

Daniel Essin, MD


According to the popular notion of how medicine will be practiced in the future, omnipresent, intelligent systems will acquire and store all available information about what is going on in the healthcare environment. Healthcare providers will simply talk to the computer to provide their input. At any point in the process the healthcare provider will be able to ask a question or make a request. In response the computer will analyze all of the relevant facts and suggest the likely diagnosis and most appropriate treatment and set the provider’s requests in motion. The same computer will monitor the status of all the patients even when the providers are otherwise engaged and will alert them if a patient requires their attention or if they have overlooked something. Such a system would be implicitly paperless. Unfortunately, such systems exist only in science fiction. The expectations that they engender are real and must be addressed for any project to be successful.

Buoyed by the technological optimism of the 1980’s, the new LAC+USC Hospital was conceived as a paperless facility. In a bold move, the architects were directed to omit from the design storage space for medical records and administrative records as well as most office space for physicians. Provisions were made, however, to allow for a paper chart of each patient's current admission. In 1989 these assumptions seemed reasonable. The industry seemed poised to deliver systems that would be marvels of technological and medical innovation; unburdening the physician, improving the efficiency and safety of patient care and, in general, conforming to the collective hope for a future that is made better through technology. There was, it seemed, more than enough time to acquire such a system and eliminate the need for long-term paper storage as well.

The gap between our expectations and what is available today is large and may not diminish any time soon. There are reasons for this gap that can be analyzed and debated at length but that does not alter the fact that the gap exists and the gap is our problem. For years now, our unrealistic expectations have stood in the way of taking practical steps to achieve a way of doing business in the new facility that does not produce paper that required long-term storage. There are a variety of pragmatic solutions that will address this requirement in isolation and a smaller number that can also deliver some of the computerized functions that physicians associate with a computerized patient record.

The Immediate Problem

The most immediate problem is that most patient care is provided in an information vacuum. Patients are treated by various specialties at multiple sites. With portions of the records stored at different facilities, unfilled and misfiled documents, charts that are unavailable and “shadow charts” that are limited to a single specialty, an individual provider rarely has access to the patient’s entire medical record. All of this information needs to be captured (either as data, text or images of documents) and stored in a way that they can be retrieved and viewed on a computer screen at any location within the healthcare network. The good news is that doing something about the paper storage problem will also have a large, positive impact on the quality and efficiency of patient care. The biggest impediments to achieving paperless operations within the replacement facility that we face in the short term are:

  • Expectation management – Some are unwilling to focus on a pragmatic solution because it is incomplete or imperfect.
  • Behavior modification – All staff at all levels will have to develop new work habits and practice styles. All will need to be held to new levels of accountability.
  • Lack of experience – Our staff has not practiced in partnership with technology and lacks the experience to make sophisticated decisions about requirements. “Experience is what you get when you don’t get what you want”
  • Lack of responsibility – It is the organization’s responsibility to craft the overall approach to these problems and take the responsibility for overseeing the integration of all the components to achieve the functional goals
  • Lack of specialized staff – The adoption of any new technology in this area will place additional importance on quality control, computer and network technology, medical terminology, knowledge engineering and clinical workflow. The time to complete any implementation will be inversely proportional to the number of qualified staff available.
  • Lack of important infrastructure – The proper functioning of an elaborate system that includes the “must have items” depends on the proper and unique identification of each patient. We would be well advised to avoid implementing too many systems in which an incorrect identification could lead to a disastrous medical outcome until we have implemented a Master Patient Index of some sort and have cleaned our existing files of duplicated entries.

Expectations are the biggest hurdle. The major expectations are that patient safety be enhanced, clinicians are able to work more efficiently and that paper storage be eliminated. While it will not be possible to fulfill every expectation completely, they all require some degree of attention in order to secure the support of the majority of clinicians that will be users of the system. Time is short and access to the existing paper charts will be required (immediately and for some time to come). This suggests that a hybrid approach will be required that can combine some of the functions normally expected from a computerized patient record system with a capability for document imaging.

The success of any scheme depends on its enthusiastic adoption by the staff. There are plenty of examples the vendor’s or programmer’s notion of what makes a usable system is out of sync with that of the intended audience – the medical staff and nursing staff.. It makes sense to introduce new technology gradually, placing the most vital components first. Every practitioner within the organization is familiar with the impending paper storage problem and should easily accept the fact that adoption of a paperless system is mandatory.

As people gain experience with new technology, their understanding and opinions evolve rapidly. Were we to perform an elaborate requirements elicitation now and then repeat it one year after the staff had been working in a paperless operating environment, their list of requirements and their priorities would be dramatically different. We should take advantage of every opportunity to provide the staff with hands-on experience as it will result in more discerning customers.

The effort to increase the utilization of computer technology in medicine is a long-term activity. The medical record itself, for example, must often be retained for more than 20 years and perhaps for a patient’s lifetime. The time span is extremely long when compared to the lifespan of an individual computer system product. It is also long when compared to the lifespan of the average vendor. It is imperative that the organization take the overall responsibility for the determination of strategy, the selection of components and the integration of those components (as they come and go) so that the clinical computing environment provides a seamless, ever-increasing level of functionality to the practitioners.

The need for highly specialized staff increases dramatically at the time that we implement systems that contain “expert system” features to provide alerts and reminders. Individuals who are knowledgeable about the medical field in question and well as about knowledge engineering must configure these features for each medical specialty. Most organizations that have that implementing knowledge-based decision support systems is a multi-year activity spanning 5 to 20 years. The short time available in which to achieve a paperless operating environment means that the timetable for implementing alerts and reminders will have to be tailored to the available resources and may not be finished for some time after the replacement facility is occupied.

When the point is reached where automated systems are critiquing or making recommendations on medical treatment, it is vital that those systems have access to ALL the information that is related to the specific patient in question and ONLY information about the patient in question. Our current systems lack appropriate technology or safeguards to insure the accurate and unique identification of each patient. It would be dangerous to attempt to implement a sophisticated decision support system until the organization has implemented an enterprise-wide Master Patient Index and has identified the majority of the duplicate entries and merged the medical records of those patients.

The Long-Term Problem

Although we must be pragmatic about rapidly implementing a Paperless Operating Environment, we must do so in the context of a long-term framework. There are Enduring Business Themes (EBTs) that originate from the business of healthcare that will remain stable for years to come, even as the technology that we use to implement actual components changes from moment to moment. EBTs are a form of design pattern. An example of an EBT is the need for accurate, non-duplicated identification and demographic information about each person that receives or provides services. This need will persist regardless of how it is implemented. There are two classes of EBTs, some address the principal business while others address the IT infrastructure needed to support the frontlist business. The following list presents examples of Healthcare Enduring Business Themes.

Principal Business Themes

  • Access to and Payment for Care (Patient Eligibility, Enrollment, Scheduling, Registration, Billing and Reimbursement)
    • Delivery of Care
      • Acute Care
      • Episodic Care
      • Coordinated Care
      • Request Management (superset of Order Entry) and Fulfillment
      • Documentation of Care
      • Retrieval of existing results and documentation
        • Creation of new documentation
      • Alerts and Reminders

IT Infrastructure Themes

  • Authentication
  • Person Identification
  • Policy and Guidelines (Superset of Permissions and Access Control)
  • Request Management (non-care related) and Fulfillment
  • Archival Storage (Text, Images, Multimedia)
  • Clinical Data Repository (primarily short-term storage and work-in-process)
  • Expert Advice (generates alerts and reminders)
  • Terminology and Nomenclature
  • Notary (Digital Signature)
  • Security, Privacy and Release of Information
  • Resource Allocation and Management (HR, Materials Management, Scheduling

This list conveys the notion that the number of Enduring Business Themes numbers in the tens, not the hundreds. Cline and Girou [1] have propounded a few principles that they have found useful in identifying EBTs:

  • Avoid looking at the details since themes are abstractions
  • Look at a business from the viewpoint of its customers, not its employees
  • Today’s requirements are not as important as future requirements
  • Avoid centralized command and control approaches

The Challenge

The perennial challenge is to address the Principal Business Themes in an optimal manner at every point in time as the day-to-day requirements of business and the state of technology change. Translating this arcane statement into a practical strategy for the replacement facility, it means that we have to adopt a way of doing business (along with the technology to support it) that is sufficient to meet our immediate needs. It also means, to the extent possible, to select and assemble the solution in such a way that individual components can be replaced in an incremental fashion as needs and technology evolve.

The key lies in modularity. The strategy is necessarily one of incremental acquisition and evolution. Over time it will become apparent that the items listed as IT Infrastructure Themes each lend themselves to being created as a “service” – a kind of black-box that receives requests, performs a particular set of internal operations and returns a result.

Thinking of Archival Storage as a service, its function would be to store information in a way that insured persistence, integrity and authenticity --- Some of the services, notably the Archive and the Master Patient Index, have few dependencies and can operate more-or-less on their own. The rest function at a higher level and depend on the existence of the core services to be able to perform meaningful work.

If, on the other hand, these services were to be specified as “requirements” of individual products rather than being acquired by the organization to be used as resources, the end-result will be the dreaded “silos” – a bunch of vertically integrated special purpose systems that contain whatever functionality they need to do their job. The end result, also called islands of information, can only be avoided if individual tasks are performed by applications tailored to the needs of each user or department that all use and share the services provided by the organization.

The creation of a Paperless Operating Environment is an incremental step toward this approach. It is a stepping-stone to establishing an Archival Service that can eventually be made available to all applications that generate information related to patients that requires storage. As the overall computerized healthcare computing environment evolves, this information will eventually be routed through the Notary that will challenge the user for their identity, verify it with the Authentication Service and then forward a signed copy to the Archive.

Each one of these services will be accessed through a well-defined set of data and communication standards that will hide the internal implementation details from its consumers. This will make it possible to upgrade any component in an orderly fashion, when it becomes necessary, without having to disturb the operation of the other services or the applications that use them.


Modularity provides a mechanism for the organizations to assert control in a consistent fashion throughout the enterprise while avoiding duplication of effort and inter-facility variations. It also provides a mechanism for orderly growth and predictability in budgeting. It minimizes the risk that is associated by making a multi-megabuck commitment to a single vendor for a single project.

In conclusion, addressing the short-term goal of providing a Paperless Operating Environment for the replacement facility is also an excellent first step toward achieving the long-term goal of a highly computerized patient care environment in which automated functions improve the quality and safety of patient care while providing the practitioners with a more satisfying and less tedious work experience.

[1] Cline, M and Girou, M. Enduring Business Themes, Communications of the ACM May 2000, Volume 43 Issue 5.

Your Ad Here
Personal tools