Using Electronic Medical Records To Improve Patient Care: The St. Jude Children’s Research Hospital Case


Mark N. Frolick
Xavier University


Information technology is proving to be a vital element in the administration of healthcare. Specifically, most healthcare institutions in the United States are adopting information systems that provide more accurate and timely information regarding patient care. This paper explores the implementation of an electronic medical record system at St. Jude Children’s Research Hospital in Memphis, Tennessee. St. Jude Hospital is primarily dedicated to the treatment of children with catastrophic illnesses and the performance of research studies to improve the clinical outcomes of such diseases. Appropriate patient care requires the use of clinical procedures as well as applied research protocols. Information must be accurate and immediately available to individuals involved in the care of patients. An electronic medical record system was introduced as a way to facilitate a centralized patient information repository. Benefits realized by this system included improvements in patient care, clinical research, and patient service and satisfaction. The ultimate goal of this project was to provide a paperless patient medical record that linked research and clinical data.


Although technological advancements in science have greatly improved medical care in recent decades, improvements in the management of patient information have been languid. Many healthcare institutions continue to rely on paper-based medical records as the primary source of patient medical and demographic information. Medical care decisions are based on the information stored in these charts.

While most health care institutions employ information systems to manage some aspects of patient care, the systems are often disjointed. Many times each department in a hospital has its own information system. In such cases, communication between departments is reduced to printing the information from one system and sending that output to the other department. Ultimately, these paper records are transferred to the patient medical record. Delayed or inappropriate patient treatment is often due to miscommunication, lost or destroyed records, and the overall inefficiency of the paper system.

Substantial improvements in patient care can be realized through the use of electronic medical records. These electronic medical records provide the ability to capture, organize, and present relevant clinical information in a manner superior to the physical records currently in use. In addition, unlike the paper-based record, electronic medical records allow all caregivers to access the patient record at the same time.

St. Jude Children’s Research Hospital (SJCRH) in Memphis, Tennessee is dedicated to the treatment of children with catastrophic illnesses. Defined protocols exist for the treatment of leukemia, solid tumors, specific genetic disorders, and some infectious diseases. The criteria for each protocol of treatment are carefully defined. Patients who meet the criteria of a protocol and are enrolled on that protocol are monitored closely for both clinical and research purposes. In order to accurately assess the effects of treatment protocols, patients must be followed from referral to the end of treatment. Many patients are followed for several years. Complete and accurate documentation of all clinical details and variances in treatment must be maintained. The ability to link protocol-defined treatments with the effects of those treatments on patients improves medical care for future patients.

While undergoing treatment at SJCRH, most patients receive treatment in the outpatient clinic. The fact that the majority of patients are ambulatory introduces another factor into patient care. In order to provide the best patient care, hospital visits must be well-managed so that patients receive all the care needed but are not required to spend excessive amounts of time at the hospital due to an inefficiency of institutional processes. Patient scheduling must be well coordinated. Orders for treatment, medications, and diagnostic testing must be legible, complete, and accessible. In addition, test results must be reported accurately in order to provide the best and most efficient patient care.

This case study examines the implementation of an electronic medical records system at St. Jude Children’s Research Hospital. It addresses the problems that existed with the paper-based medical record and explores the advantages of electronic medical records. In addition, this paper explains the issues that SJCRH explored before choosing the Cerner HNA Millenium system. As a result of this implementation effort, all active patients at the hospital have a paperless medical record that provides both demographic and medical information. This electronic medical record provides a single point of access for all employees seeking information both for clinical treatment of patients and for research data collection.


Providing excellent medical care for a patient requires an accurate and organized record of the patient’s medical history. While SJCRH had previously implemented a computerized information system to manage a limited amount of patient information such as demographics, clinical laboratory results, diagnostic imaging results, and pharmacy records, the records in this system still lacked a vast amount of information. All manually generated paperwork and printouts generated by individual department’s information systems were ultimately stored in the paper medical record. Thus, the paper medical record represented the only complete source of a patient’s medical history.

The Medical Record as a Source of Information

The typical paper medical record contained nine sections including information on admissions, discharge summaries, progress notes, protocols, laboratory results, radiology results, surgical and pathology reports, orders for treatment, and nursing notes (I.T.S. steering committee, 1999). Most documentation regarding treatment of a patient was written directly in the patient’s medical chart. However, some additional paperwork items such as consent forms and transfusion records were also stored in the chart.

Appointment scheduling, treatment plans, and record keeping were performed manually in departments such as physical therapy, respiratory therapy, social services, and psychology. Patient care activities in these departments produced another set of paperwork that had to be stored in the patients record.

Patient medical records were stored in the Health Information Management Services (HIMS) department and were retrieved for use when a patient was admitted for inpatient care or returned to the outpatient clinic for treatment.

Computerized Information

While all patient information was stored in the paper record, SJCRH had previously installed a computer information system in order to automate some of the record-keeping processes required for patient care. This original information system at SJCRH consisted of two separate systems. One system was used to manage patient scheduling and registration. Information collected in this system was transferred to a second system that provided record-keeping and test-resulting capabilities for the clinical laboratory, pharmacy, and diagnostic imaging departments (Rogers, October, 1999). While this record-keeping system was used to order, enter, and review clinical test results, printouts of all test results were ultimately placed in the paper medical record.

Accessibility of the Medical Record

On any given day, physicians, nurses, medical residents, health information management personnel, data managers, and hospital unit coordinators needed to access the information in the medical record. Typically, charts were pulled from HIMS the day before a patient’s scheduled return, transported to the outpatient clinic where the patient would be seen, and then returned to HIMS at the end of the day to be filed again (Milli-our system, 1999). In addition, data managers, who collected information for clinical protocol review and research studies, retrieved the medical charts from HIMS and manually extracted data several times during a patient’s enrollment on a protocol (Milli-our system, 1999).

It was necessary for the medical record to follow the patient throughout their visit. Thus, if the patient was seen in one clinic where orders were written, it was necessary to physically transport the record when the patient moved to the medicine room for treatment. Inpatient charts were kept at the appropriate nursing station until the patient was discharged.

While the record-keeping system was available to review test results, the information available in this system was limited. Information such as physicians’ notes, transfusion records, and dietary notes could not be accessed through the computer system. Thus, it was necessary, under most circumstances, to have access to the patient’s medical chart in addition to the computerized records.

The Medical Record as a Form of Communication

The information added to a patient’s medical record was regularly reviewed to determine adherence to protocol standards and to monitor a patient’s progress. In addition, attending physicians used the charts to review the previous attending physician’s notes in order to provide the best care for their patients. For example, notes that detailed an adverse reaction to a specific medication could be used to prevent that medication from being prescribed in the future. Nurses beginning their shifts reviewed their patients’ charts to determine how the patients had progressed through the day as well as to follow up on any orders that had not been filled. Dietary personnel and psychology staff used the chart to determine if particular medications had been prescribed that would alter a patient’s eating habits or behavior patterns. Needless to say, caregivers relied on the information in the record to be accurate and complete.

In addition, departments such as the blood bank and pharmacy were obligated to have a physician’s signature attached to any order they filled. Thus, a copy of the original order for medication or transfusion was faxed to the appropriate department. Unfortunately, faxes were not always legible nor were they always sent to the correct department. Patients’ wait times were unnecessarily lengthened because orders were illegible or sent to the wrong department and had to be faxed again when the mistake was discovered.

Disadvantages of the Paper Medical Record

While the tasks of retrieving patient records, extracting and adding information, and returning these records to their appropriate place of storage each day was burdensome, managing the information in the records was even more tedious. Patients who had been followed for years often had medical records consisting of several large volumes. Any of these volumes might have contained the information necessary to properly care for the patient. As a result, all of the volumes had to be accessible to caregivers.

Since the medical record consisted of many individual pages from many different sources, pages could easily be misfiled or even misplaced entirely. In some cases, if information could not be located in the record, tests and procedures were simply repeated. For example, patients or their guardians were required to sign medical consent forms that were then placed in the medical record. If a signed consent form could not be found, a new form was completed and added to the chart. Of course, such actions led to redundant information being filed in the patient record. Most institutions find that such extensive documentation is often incomplete, inaccurate, illegible, and difficult to access when needed (Protti, et al., 1998). The risk of improper patient care or misinterpretation of research protocols increases when data is illegible or inaccurate.


While the first attempt at computerizing medical records helped with records management, it was little more than a storage mechanism for a very limited amount of patient information. As a result, SJCRH undertook the development of an electronic medical records system.

The introduction of electronic medical records at SJCRH afforded the possibility for records to be created, processed, stored, retrieved, and cross-referenced more efficiently. The electronic records of SJCRH patients include all of the information stored in the paper-based medical record. It allows caregivers with appropriate security clearance to access patient’s electronic medical records from any personal computer in the hospital. This feature eliminates having to locate the patient chart in order to obtain information necessary for treatment.

In addition, information can be easily sorted or grouped according to certain criteria such as the date on which the test was performed. The electronic medical record also allows the user to graph a set of results over time. For example, a patient’s blood glucose level can be graphed over a time period of ten days, thus allowing the caregiver to notice trends that might be important for proper patient care.

Not only does the electronic medical record system provide a central location for storing patient information, it provides several powerful functions that allow for better patient care. These functions were facilitated by the various components of the electronic medical records system.

The Components of the Electronic Record: OCF, PowerChart, and Discern Expert

The electronic medical record system at SJCRH has several components. The Open Clinical Foundation (OCF) is the repository at the center of the new electronic medical record. The OCF is an Oracle database that stores clinical and administrative information. This new database functions as a data warehouse and has the ability to group information based on any one particular patient parameter. For example, patient clinical outcomes with regard to a particular protocol can easily be grouped and presented by the electronic medical records system, thus eliminating the need for data managers to collect the information manually (Rogers, October, 1999).

PowerChart is the graphical user interface that caregivers access at the clinical workstation. It is composed of two parts: the Organizer and the Chart. The Organizer allows the user to quickly check for new patient test results immediately after logging onto the system. In addition, it allows the user to indicate which test results they have reviewed.

The Chart is the electronic form of the patient medical record. It is through this interface that the user can review clinical lab results, nurses’ notes, physicians’ notes, and patient demographics. Features that are available in the Chart such as the problem list, visit list, and growth chart also allow caregivers to track a patient’s medical progress. This electronic medical record makes gathering patient information more efficient because it automatically groups similar data together.

PowerChart will act as the foundation for more advanced patient care applications to be installed in the future (Milli project scope document, 1998).

Discern Expert is a program that evaluates best clinical practice criteria and monitors events in the system for compliance. It is part of a decision support system that assists healthcare providers at the point of care by linking historical patient data with current clinical data and assessing that data based on built-in clinical rules. Historically, work in clinical decision support systems has been concentrated on designing alerts and reminders for physicians, however more recent systems are focused on overall compliance with patient care plans (Broverman, 1999). A “starter set” of rules was developed in Discern Expert for the lab, radiology, pharmacy, PowerChart, admitting/registration, and the Chart modules (Milli project scope document, 1998). Managing patient care through the use of decision support systems ultimately means that a patient’s quality of care improves.

Benefits of the Electronic Medical Record

The benefits of the electronic medical record system at SJCRH includes information that is more accessible and improved communications.

Information that is more accessible. All clinical laboratory and patient demographic information is a part of the electronic medical record and can be viewed at any computer terminal in the hospital as long as the employee has security clearance. Data retrieval is more accurate and efficient due to automated clinical documentation and protocol information being electronically linked to clinical data reports (Milli project scope document, 1998).

Information systems that collect and deliver information efficiently are only the forerunners to more sophisticated systems. Ultimately, electronic medical records systems will not only deliver the correct information, they will actively assist in the treatment of patients through the use of that information. The electronic medical records system will act as an assistant to the healthcare provider in addition to its role as a storage device for patient information. The ultimate success of these systems, however, ultimately depend on healthcare providers’ willingness to “change the way they have traditionally recorded, retrieved, and utilized clinical data” (Anderson, 1999, p. 62).

Improved communication. Communication between departments in a healthcare organization can also be improved when all healthcare providers have access to the same information. For example, an electronic medical record that provides ready access to patient scheduling, test ordering confirmation, and completed test results allows healthcare providers in every department to clarify orders and assess patients more efficiently (Barsukiewicz, 1998). Essentially, the quality of patient care depends on the collection of detailed clinical information and the timely delivery of that information to the appropriate healthcare provider (Chan, et al., 1999).

Clinical decision support systems can improve patient care, reduce costs, support clinical diagnosis and treatment-plan processes, manage patients on research and chemotherapy protocols, monitor the need for follow-up care, and manage administrative details (Perreault, et al., 1999). Essentially, SJCRH’s new electronic medical record provides a link between clinical treatment and adherence to research protocols.

The benefits realized by the implementation of an electronic medical record include “increased efficiency in managing clinical information and improved quality of care and cost savings through decision support and management of patient care” (Anderson, 1999, p. 62). In addition, orders placed through the use of an information system are legible, free of transcription errors and can be directly routed to all departments that need access to the order (Teich, 1999). Improvements in the manner used to capture and store patient information and then relate the information to clinical treatment and research protocols has lead to much better patient care.


While SJCRH had previously installed an information system to manage some aspects of patient care, it was not a substitute for the paper medical record. Most businesses have determined that information systems can greatly reduce errors, increase customer satisfaction, and provide better employee time management. Likewise, healthcare providers have discovered that information systems can improve patient satisfaction as well as employee performance. In an effort to improve patient care and satisfaction, SJCRH implemented their electronic medical records system. While implementing this system, SJCRH discovered several important issues that should be considered prior to implementing an electronic medical records system.

Determine Objectives

In an effort to support institutional objectives, Information Technology Services at SJCRH identified the implementation of an advanced clinical information system as one of its goals. The mission of the electronic medical records software implementation was “to provide immediate access to clinical information, to enhance patient care and accelerate clinical research” (Milli project scope document, 1998, p. 1). The implementation of the new system resulted in improvements in the following areas:
patient care, patient service and satisfaction, clinical research, and access to both clinical and research data (Milli project scope document, 1998).

Determine Who Should Evaluate the System

A system selection committee was formed that consisted of a representative from each hospital department including physicians, nurses, pharmacists, and clinical lab personnel (Rogers, December, 1999). In addition, process-improvement teams were formed that consisted of a process improvement consultants, a SJCRH application analyst, an application analyst, and a SJCRH departmental subject matter expert (Milli project scope document, 1998). Process improvement teams were responsible for documenting and analyzing the former processes and recommending the goals to be achieved by the new system. Leadership teams comprised of SJCRH senior management representatives were assigned to each of the process improvement teams (Milli project scope document, 1998).

Once these process and leadership teams made their recommendations, a Steering Committee that was comprised of senior management representatives reviewed the recommendations of the process improvement teams. The proposal was then presented to the SJCRH board of directors for approval (Rogers, December, 1999).

Determine Which Software To Implement

Several clinical software packages were reviewed. Three packages appeared to meet the desired criteria. Final vendor selection was based on the clinical functionality of the software, the stability and market position of the vendor company, the product strategy, the implementation and follow-up support provided by the vendor, and the cost and business terms of the agreement (Rogers, December, 1999). Since the Cerner platform was already in place at SJCRH, the system selection committee felt that conversion to a newer Cerner system would be easier than introducing an unfamiliar system. In the end, the Cerner HNA Millennium (Milli) package was chosen.

An Oracle database that housed research data was already in place at the hospital. The Cerner system also provided an integrated system in which all information would be stored in a single database (Oracle) thus eliminating redundancy (Rogers, December, 1999). In addition, data from the research database could be easily transferred to the new system.

Determine Risks and Advantages of the New System

Risks involved with choosing this particular software package, however, included the fact that very few clinical products had been written to this particular platform. In an effort to influence the final product design, SJCRH entered into a developmental partnership with Cerner. This agreement ensured that the specific criteria defined by SJCRH would be incorporated into any new software that Cerner developed (Rogers, December, 1999).


Electronic medical records provide healthcare workers with an extensive option for improving patient care. Not only can electronic records not be misplaced as paper records can, but the information in the record is more accurate as well as legible. St. Jude Children’s Research Hospital has implemented a paperless medical record in order to improve care for both current patients and future patients. Successfully linking clinical treatment plans with research protocol information provides valuable information when assessing the effectiveness of particular treatment plans. Research efforts can move forward more rapidly when data is collected in an efficient manner as is provided by the electronic record. In addition, human error is less of a factor than it was when data was collected manually from paper charts.

St. Jude Children’s Research Hospital’s goal was to implement a system that not only met the organization’s current needs, but would also provide support for a growing number of protocols and clinical treatment plans. Not only does the new electronic medical records system maintain and organize all the information necessary to support clinical and research efforts, but it also guides healthcare providers in their daily treatment of patients. All in all the new system is at the center of providing the best possible patient care and advancing treatment for future patients.


Anderson, J. D. (1999). Increasing the Acceptance of Clinical Information Systems. MD Computing, 16 (1), 62-65.

Barsukiewicz, C. K. (1998). Managing care Through Managing Information:
Are We Headed in the Right Direction. Journal of Healthcare Information Management,
12 (4), 67-77.

Broverman, C. A. (1999). Standards for Clinical Decision Support Systems. Journal of Healthcare Information Management. 13 (2), 23-31.

Chan, W., Centiu, C., & Morris, J. A. (1999). Uniform Data Standards for Capturing Patient Medical Record Information at the Point of Care. Journal of Healthcare Information Management, 13 (3), 85-95.

I.T.S. Steering Committee Strategic Planning [Slide Presentation]. (1999, June). (Available from St. Jude Children’s Research Hospital, 332 N. Lauderdale, Memphis, TN 38105).

Milli — Our System for Advanced Patient Care and Research [Slide Presentation]. (1999). (Available from St. Jude Children’s Research Hospital, 332 N. Lauderdale, Memphis, TN 38105).

Perreault, L. E. & Metzger, J. B. (1999). A Pragmatic Framework for Understanding Clinical Decision Support. Journal of Healthcare Information Management. 13 (2), 5-21.

Protti, D., & Peel, V. (1998). Critical success Factors for Evolving a Hospital toward an electronic Patient Record System: A Case Study of Two Different Sites. Journal of Healthcare Information Management, 12 (4), 29-38.

Rogers, J. (Director of Patient Care Systems). (1999, October). [Interview]. St. Jude Children’s Research Hospital, 332 N. Lauderdale, Memphis, TN 38105.

Rogers, J. (Director of Patient Care Systems). (1999, December). [Interview]. St. Jude Children’s Research Hospital, 332 N. Lauderdale, Memphis, TN 38105.

St. Jude Children’s Research Hospital. (1998, January). Milli Project Scope Document (Document No. unknown). Memphis: No Author.

Teich, J. M. (1999). Inpatient Order Management. Journal of Healthcare Information Management, 13 (2), 97-110.

  Related Posts
  • No related posts found.