Standards Watch | The Electronic Health Record: A Recent HL7 Standardization Effort

The terms Electronic Medical Record (EMR), Computerized Patient Record (CPR) and Electronic Health Record (EHR), are used interchangeably when people speak of an electronic patient health record, however, they have different meanings representing their evolution.

The Computerized Patient Record (CPR) systems really started to be implemented in the early 1990s, even though the concept dates back to the 1980s. The major push for implementation was a 1991 publication by the Institute of Medicine (IOM). The CPR was basically a replacement of paper-based records in an electronic format. Some institutions were actually scanning the documents when data entry was not available or applicable.


The Electronic Medical Record (EMR) captures and manages the patient information that was initially acquired in electronic format. It allows for better management because its contents can be represented in different manners, such as conducting a search. Access and presentation is much easier as well. The EMR is the state-of-the art patient file.

The Electronic Health Record (EHR), on the other hand, is the true, lifetime health record, which spans different institutions, and is basically owned by the patient. It can be exchanged among different providers.

Standards for CPRs are not widely implemented, and most systems are incompatible. Of course, this creates problems when this information needs to be exchanged between different institutions. There also is no dominant vendor in the field, and therefore, no default standard is available.

One of the requirements for exchanging medical records is a uniform medical vocabulary. For example, the word hypertension has several common terms used to describe it. The most extensive

common medical vocabulary is managed by the College of American Pathologists (CAP), called SNOMED - CT (Systemized Nomenclature of Medicine - Clinical Terms), and is made available by the National Library of Medicine (NLM), to facilitate the sharing of medical records.


The Institute of Medicine (IOM) has been charged with developing a standard definition of the EHR. The HL7 standards organization is performing the work, allowing it to become an ANSI-based standard. This standard is not a fully detailed specification, but rather a so-called DSTU: Draft Standard for Trial Use, which contains a functional definition of the EHR. A second ballot was just sent out, and results can be expected to be discussed at the upcoming HL7 meeting in San Antonio. One of the major drivers behind the EHR is the U.S. government, promising a different reimbursement rate for those institutions that support an EHR.

The Institute of Medicine specifies the requirements for the EHR based on five criteria:

  1. Improve patient safety: Tens of thousands of people die each year in the U.S. as a result of preventable, adverse events due to improper healthcare.
  2. Support the delivery of effective patient care: Provide it to the ones who need it, and not to the ones who do not. Approximately half of all Americans receive recommended healthcare consistent with evidence-based medicine.
  3. Facilitate the management of chronic conditions: Persons with chronic healthcare problems account for more than 75 percent of all healthcare spending, more than half of them have three or more different providers who send out conflicting information, order duplicate tests, and seem to work totally uncoordinated.
  4. Improve efficiency: Healthcare costs are rising; patients are paying more out-of-pocket costs, and receiving fewer benefits.
  5. Implementation Feasibility: This takes into account the available software as well as the time-to-market for new functions.

The requirements of the EHR include eight "core functionalities" as follows:

  • Health Information & Data Examples include information about patient allergies to prevent adverse drug reactions, previous lab test results to prevent duplication, alerts and reminders for drug administration, and abnormal test results at the point of care. It was noted that "information overload" is not a good thing, therefore, the user interface should be balanced, i.e. not too much, and not too little. 
  • Results Management Electronic availability, for example, of lab tests and radiology reports, including previous results will improve efficiency and decrease costs. 
  • Order entry & Management Lost orders and ambiguous handwriting have a major impact on workflow. Simple medication order entry has shown to reduce the number of medication errors by more than 80 percent. 
  • Decision Support Computer reminders and prompts have shown to dramatically improve preventive practices. In addition, drug selection and potential drug interaction can be easily facilitated. Quality problems can be easily detected as well. 
  • Electronic Communication & Connectivity This includes telemedicine, home tele-monitoring, e-mail, and web messaging. 
  • Patient Support Computer-based patient education has proven to be very beneficial, especially for common diseases such as diabetes. 
  • Administrative Processes Scheduling systems for admissions, procedures, and visits increase efficiency for the provider, and also provide a better service to the patients. 
  • Reporting and Population Health Management Patient safety and quality, as well as public health issues can be facilitated, especially when combined with a standard terminology.

The EHR does not only apply to one particular institution. For example, a family practitioner might be able to access a hospital discharge record on-line. In addition, by using an electronic medical record system that properly documents visits and examinations, the same practitioners can appropriately be reimbursed for their services.


With regard to the EHR/HL7 standards effort, the EHR functional specification is divided into three different parts:

  • Direct Care Services related to direct patient care.
  • Supportive Functions Admitting, discharge, transfer, billing and registration.
  • Information infrastructure Providing the infrastructure for this standard.

The EHR effort is a truly international effort; and will therefore have different specifications applicable to specific countries. Each will identify the functions as "Immediate," "Essential future," "Optional," or "Not Applicable."

Full-scale implementations of an EMR are relatively rare. It is available in about 10 percent of all institutions, versus more than 20 percent with an installed PACS. The impact on patient care can be huge: adverse drug reactions that are "caught" by a computerized physician order entry system thus preventing consequent hospitalizations can be drastically reduced. Reductions in medication processing time, assuming a physician can prescribe drugs at the bedside using a handheld wireless device, can be up to 70 percent. The EMR should ideally also provide radiology results, and preferably the images that support the diagnosis. This could be done relatively easily, by having the EMR software issuing a query to the PACS for the applicable images. This requires some level of integration. Most systems seem to have the images on-line for a certain period, e.g. 3 months, in a web-friendly format such as a JPEG compressed image, which can easily be accessed using a url. If the image is not available from this intermediate server, a query to the main PACS archive can retrieve the images.

One should also note that there is an initiative from the National Health Service (NHS) in England, to provide an electronic integrated care records service (ICRS), a nationally accessible core repository for medial data, an electronic ordering system, and the electronic transmission of prescriptions. Of interest is the requirement that this will all be based on the HL7 version 3.0 standard. This activity, together with the U.S. initiative, will significantly speed up the development and implementation of the EHR. 

Herman Oosterwijk is president of OTech Inc., specializing in training on PACS, see for details.