I have been involved in nursing informatics for almost 30 years and have seen many developments in this field. Back in the early 1980s nursing informatics was defined as “...the application of computer technology to all fields of nursing--nursing services, nurse education, and nursing research” (Scholes & Barber, 1980, p. 73). Since then we have moved through the seminal definition provided by Graves and Corcoran (1989) who defined informatics as "a combination of computer science, information science, and nursing science designed to assist in the management and processing of nursing data, information, and knowledge to support the practice of nursing and the delivery of nursing care" (p. 227), and on to the point where we have identified three threads important in nursing informatics: basic computer skills, informatics knowledge, and information literacy (Hart, 2011).
Along with the continuing development of definitions has come development in the use of computers in the clinical area. Computerized provider order entry (CPOE) has freed nurses from order transcription and related, order-clarifying phone calls. Additionally test results are now available at the point of care shortly after the tests are completed, allowing interventions to be initiated in a much more timely manner. Computerized clinical documentation is also finally moving forward.
The first commercial electronic medical record (EMR), ‘Technicon’s Management Information System’ (MIS), implemented in 1971, was designed by Lockheed engineers in close collaboration with clinicians at El Camino Hospital in central California (Buchanan, 1984). Ann Farrell, who was at El Camino when the system was used, has recently described the system thus:
The system was an inpatient-oriented, hospital-wide system. Key to its longevity and success were its true integration, clinician workflow support, and ease of use. Physician, nurse and pharmacy clinical processes were tightly integrated, with a single version of orders as well as an eMar. Ease of use was a guiding principle with guaranteed sub-second response time, use of clinical English words and phrases (not computer codes like other systems), and creation of the Informatics Technology (IT) industry’s first pointer device, a light pen. The collaborative culture and respect for the role of RNs was reflected in system design. MIS provided more functionality and better integration for nurses than we see in most systems today; requirements of the healthcare providers drove the technical innovations (A. Farrell, Principal, Farrell Associates, personal communication, September 2011).
The early 1980s were exciting for those of us interested in informatics! We enjoyed great conferences listening to the early pioneers such as Margo Cook (El Camino) and Judy Ronald (University of Buffalo, State University of New York). Networking at these conferences gave rise to great hopes. We imagined that within the next 10 years informatics would provide great benefits to clinical nurses, other healthcare practitioners, and patients. We envisioned such things as minimal time spent in documentation, working together with patients to document past history and care received, a lifetime healthcare record, and the use of aggregated data to improve nursing practice. Informatics, we believed, would free nurses and other healthcare professionals to spend more time with patients and minimize the pain of documentation.
As I look back on those early days full of hope and see where we are today, I still have those same dreams. We have come a long way since the 1980s; yet there is still a long road ahead. There has been some progress; for example care delivery organizations are now moving from using information technology in financial and ancillary department systems to implementing and using this technology in full clinical systems. Also the United States government is providing financial carrots (incentives) for what is termed ‘meaningful use’ of electronic medical records (EMRs). Meaningful use, however, cannot be attained without information technology (Medicity, 2010) and solid informatics principles.
Most nursing schools now teach at least one of the threads of informatics. Additionally both the National League for Nursing’s and the American Association of Colleges of Nursing’s accrediting agencies have added beginning informatics as a curriculum requirement. This is in line with the Institute of Medicine’s (IOM) specification that education for the healthcare professions includes content that enables the use of informatics in the clinical area (IOM, 2003). Many nursing and other healthcare-related journals include technology columns and articles about electronic records. Furthermore, graduate degrees are now offered in both nursing informatics and healthcare informatics to prepare personnel who are highly knowledgeable regarding informatics, who recognize that systems need to serve the clinician rather than the clinician serving the system, and who are able to design and implement clinically driven systems rather than information technology (IT) projects.
Unfortunately, today there are still many barriers to overcome. These barriers include the lack of system integration, failure to implement informatics principles, inadequate screen designs, and probably most importantly, “...difficulty imagining new ways made possible by tools that enable things... never dreamed of” (Schulman, Kuperman, Kharbanda, & Kaushal, 2007, p. 539), as described below.
Too many healthcare agencies still lack systems that can share data. These agencies are often using ‘best of breed’ systems, i.e. systems that provide superior automation for one department, but are not interfaced with any other agency department or system. In an ideal system, all data from all departments will be integrated. The use of these non-interfacing systems, in which different healthcare professionals need to each enter the same data, hinders quality control efforts (Whelchel, 2011). In addition, it creates errors. For example, recently I observed the results of an error caused by such a system when I found the name on a discharge medication list different than that of the patient to whom the list was given. In such systems, a nurse selects a patient’s name from a closely spaced list which is then placed on a document. Next the nurse manually retypes the medication list from a paper list printed from another system onto this document to create a computerized discharge medication list for the patient. On two different occasions, I observed this process causing an extra 15 to 20 minutes of additional work for each patient discharged, to say nothing of setting the stage for errors. Ideally, both clinical and medication data would be available together; and the nurse would be able to locate the patient’s name, immediately see that patient’s medication list, check it, and print it for the patient’s discharge.
A cardinal rule in informatics is one entry of a piece of data, many uses. Thus within a given patient’s record, any entered data is available for use any place in the record that it is needed. Every time a piece of data is re-entered, the likelihood of errors increases. Ideally, in cases where different healthcare professions use the same data, the first professional to ask the question would enter the data, and the data would be seen automatically on the screen of any other professional who subsequently needed this data. This ‘subsequent’ professional would then recheck to see if the information was still current or whether it needed to be updated. If two different healthcare professionals have differing opinions as to the current/correct information, they would need to consult with each other. Systems that require multiple entries of the same data invite not only error, but also work-arounds and non-use of the system.
Another feature that requires thoughtful attention is screen design. Thoughtful screen design by both the vendor and those implementing a system for professional users can save user time by eliminating unneeded information and making the system easier and more efficient to use (Zopf-Herling, 2011). The use of human-interface principles in screen design can prevent data entry errors.
The scanning of patient healthcare records into the computer, with the belief that an electronic record is being created, sidesteps informatics principles and demonstrates a failure to imagine new ways. Too often this practice creates a digital record that is more difficult to use than the paper record because the information scanned in is not easily retrievable in a useful format. If one must scan the information into the record, forms that allow the data to be scanned into a structured format should be used. Hand-written forms, which are difficult to read in the original, may be even more difficult to read once they have been scanned into a computer. If the attempt is made to make them more legible by converting them to text, extensive editing is required to be sure that they are accurate.
Today it is clear that if our earlier dreams of EMRs are to be realized we need to make greater use of informatics principles in system design and implementation. The dream of totally integrated systems may not be realistic in our current environment. However, we do need to invest time, money, and effort in pushing to implement integrated processes involving patient care responsibilities that are shared by a variety of disciplines. We also need to attend more closely to the usability of the system. This type of progress will only occur when we employ principles of human factors to influence the functioning of technological systems and require that clinician workflow factors drive system design and implementation.
One thing that we early dreamers failed to consider was the cultural changes that would be needed to reach our dreams. One of the biggest of these needs is a move from a silo mentality to a multidisciplinary perspective. Silo mentality creates a patient situation analogous to that of three blind men describing an elephant, with each man examining a different appendage and making decisions without considering the nature of the ‘whole’ elephant. If a holistic EMR that assists healthcare providers to co-ordinate and manage patient care is to be created, more than lip service needs to be given to the patient as a holistic being. This requires that all healthcare disciplines develop an appreciation of the contributions of all members of the healthcare team.
What we did realize, however, was the potential to improve care that electronic patient care records can provide. This will only be realized, however, when the end goal of implementing a system is improved patient care and safety and there is greater use of informatics principles including human factors in system design and implementation. Achieving this requires an understanding of and appreciation for clinician workflow, abandonment of a paper chart mentality, and acceptance of the need to implement or update systems that are clinically focused rather than ‘IT focused.’ If we can change our thinking in this manner, eventually our clinical systems will be as integrated, patient centric, useful, and usable as the original El Camino System.
Linda Thede, PhD, RN-BC
Article published January 23, 2012