Information management's key role in today's critical care environment.
Healthcare is undergoing enormous changes driven primarily by market forces, creating clear winners and losers as the transformation takes place. Similarly, the medical laboratory is facing tremendous pressure to adapt to new requirements of the marketplace, especially for critically ill patients. Clearly those laboratories that do not adapt will not survive.
Moreover, we are entering the information age in laboratory medicine where we no longer do laboratory tests in the traditional sense; rather, due to computerization and automation, we simply deliver barcode labeled specimens to automated systems designed to extract information from specimens. Accordingly, in many respects, performing the test has become the easiest and most controlled aspect of the process of testing.
However, errors and problems abound in the information and specimen management steps of the testing process, which includes 1) the test order, 2) specimen collection or acquisition, 3) specimen logging and processing, 4) specimen analysis, 5) results reporting, and 6) billing/business management activities. Those laboratories that fail to address these errors and inefficiencies will soon be out of business and replaced by more computerized and automated laboratory competitors in this new market-driven paradigm.
Intrinsic value of laboratory data
As laboratory scientists, we tend to focus more on the analytical aspects of our laboratories and disregard the value of our information. However, laboratory data for critical care depreciates in value so rapidly with time that it essentially is worthless if not available in time to assist in the diagnosis and management of the critically ill patient.
Accordingly, better laboratory information systems (LIS) can print or automatically fax all laboratory results to critical care areas and transmit results to the hospital's electronic medical records system as soon as they are verified in the laboratory. Critical laboratory results (or potentially life-threatening laboratory values) are flagged and transmitted immediately by the LIS to the hospital information system (and also called to the physician and the patient's location).
For example, in our institution, as soon as results are verified in the laboratory, they are sent to printers in intensive care units (pediatric and adult), coronary intensive care, emergency departments (pediatric and adult), and neonatal intensive care units. Also, with new options on the LIS emerging for automatic faxing, E-mail, automatic paging, wireless communication, etc., the opportunities for reporting critical care laboratory results in real time are improving.
Critical care environment
For years now, a common problem in laboratories has been the triage of specimens from critical care areas with those tests received from nonemergency patients seen on hospital units, in ambulatory clinics, and physicians' offices. For inefficiently run laboratories with slow or unpredictable turnaround times (TATs), requests for Stat results are so frequently abused by physicians that laboratorians have difficulty identifying and handling the true Stat requests in a predictable fashion.
The solution to this problem is obvious: design the laboratory computer system and other technologies including robotics and instruments so that essentially all bottlenecks in the laboratory are eliminated. For non-point-of-care testing (POCT), with electronic test orders being received from essentially all clients, the front-end specimen receiving and computer accessioning area of the laboratory must be highly automated, otherwise specimens will accumulate as they arrive.
POCT and associated data management must be handled in a totally automated fashion as well. After all, getting data into a computer system is the rate-determining work in using any computer system in the laboratory. In addition, non-POCT instruments must be selected with 1) large menus and high throughput, along with primary tube sampling from barcode labeled specimens, and 2) a robust host-query interface between the instrument and the LIS that conforms to standards of the American Society for Testing and Materials.
POCT instruments and associated instrument controllers must conform to ASTM interface standards as well. Since all tests in such well designed laboratories essentially are performed in real-time, TAT becomes highly predictable, even for non-emergency testing. The total number of specimens in the laboratory test queue becomes smaller and thus more manageable; hence specimens and associated problems are handled more efficiently and predictably, and when problems occur, fewer customers are affected by the event.
Such highly automated laboratories (I call them "queueless" laboratories) have information systems that readily retrieve information relative to the status of the test order, test results, interpretation, test requirements, etc. Accordingly, calls from clients are handled efficiently because the computerized tracking system allows everyone in the laboratory to know where a particular test is with regard to the testing process.
Special networked workstations designed on the LIS using a multitasking graphic user interface (such as Windows 95) are required to handle inquiry in this manner. Computerized tracking and inquiries support are critical in a market-driven laboratory handling critical-care specimens since the customer's worst fear is that we have lost or mishandled his or her specimens somewhere in the process. Every time a client interacts with your laboratory is yet another opportunity for you to demonstrate that you have total control of the testing information in the entire laboratory.
What about information for POCT?
Certainly, a significant way to eliminate bottlenecks in the high-volume, queueless laboratories described previously is to decentralize a good portion of the critical care testing, especially testing routinely done on patients in critical care units and emergency departments. Combining the phenomenal improvement in biosensors; the ability to test whole blood; and powerful, miniaturized computer processors (and associated software) to produce highly sophisticated, reliable. self-contained POCT instrumentation has changed the practice of laboratory medicine irrevocably. Laboratorians and data processing professionals, however, need to focus efforts on the integration of POCT data into the electronic medical record.
Indeed, while many laboratories have employed POCT instrumentation in their hospitals, less than 10% indicate that they have automated their data handling of POCT. As community health information network (CHIN) laboratories continue to emerge,[5,6] many hospitals are relying almost entirely on POCT technology for on-site testing, while referring other less critical testing to the off-site CHIN core laboratory.
To the physicians who use the CHIN laboratories, however, the entire consolidation process must pass the test of transparency; that is to say, laboratory services should continue to be as good or better than testing prior to consolidation. In addition, CHIN laboratories must handle information so that POCT data is integrated into the electronic medical record and displayed or printed on cumulative reports along with other laboratory data.
This requires that POCT instruments gather test data in a totally automated manner and provide for electronic transmission of that data from the patient's location to the central laboratory, as well as provide for the automatic upload of POCT results to the patient's electronic record (via accession number or medical records number).
In our laboratory, we have a data gathering system for our POCT blood gas analyzers where POCT results data are:
* transmitted to a personal computer (called an instrument controller, or IC) in the core laboratory, where results are subsequently reviewed on the IC workstation and verified by a technologist
* released for upload after technologist verification to the LIS
* transmitted from the LIS to the hospital information system (HIS) for permanent storage located on the patient's electronic medical record system.
For remote POCT, we attach the POCT data gathering system to a laboratory multiplexer (a device with multiple ports), which then transmits the POCT data stream over a shared modem line to the POCT IC located in the core laboratory. Other laboratories transmit data from POCT devices to the core laboratory IC over the local area network (LAN) or use wireless technology to transmit data. As POCT data are uploaded to the LIS, software on the LIS automatically:
* identifies the patient by medical record number and automatically orders the point-of-care test(s), a process that:
* generates an accession number for each test
* appends a specimen collection date and time
* appends results to the patient's test record
* assigns a verified status to the results on the LIS
* reports POCT results in real time (if appropriate) by auto-fax and/or remote-print to clients (especially remote clients)
* transmits POCT results to the HIS, which appends them to each patient's permanent electronic record.
The POCT IC uploading the POCT data stream emulates a terminal on our LIS and thereby uses programs already present on the LIS, obviating the need to develop a unique interface and special software to handle POCT data. Billing and other statistical data are collected as well, just as if the test had been performed locally by the laboratory.
Printed POCT results reporting
POCT results are of course reported locally in real time and from our LIS via patient cumulative reports, auto-fax, and remote status printing. In our laboratory, POCT data are also available on the inquiry screens on the LIS and our HIS system. We report POCT results under a separate POCT header rather than combine these data with other non-POCT results.
Our expert LIS automatically appends a coded comment to POCT results stating that all testing was performed by nursing or respiratory care personnel. A vertical format is used where the POCT names and results are displayed from left to right with dates and times of specimen collection displayed along with appropriate reference ranges (age- and sex-matched) specific for the particular POCT instrument. Abnormal results are flagged as a boldface L (for low), H (for high), and C (for critical), and comments are appended.
Our non-POCT hematology results are reported concomitantly with POCT values and in a similar format with both test groups under the hematology header. This format allows physicians to track results visibly over time, an essential feature for treatment and prognosis.
Graphical displays of these POCT results can be generated by an interactive inquiry program as well, where the LIS system builds a realtime spreadsheet of results, both POCT and non-POCT. Using cut-and-paste options under Windows, these spreadsheets can be included into consultative reports easily.
POCT electrolyte chemistry results are reported on a patient's cumulative report. The format of this report is identical to the hematology results described above. We continue to display the nonPOCT electrolytes under the same heading so the physician can review these concurrently. Table 1 depicts reporting of whole blood pH and ionized Ca results performed on our other POCT blood gas analyzers.
Importance of data handling and networking
The laboratory's network and data handling capabilities are essential for the success of any CHIN laboratory, and especially those laboratories engaged in POCT critical care testing.[9,10] Indeed, the costs of data handling alone, if a manual data entry system was employed, would drive the costs of POCT to unaffordable levels. Hence, in a market-driven paradigm where only the most efficient laboratories will survive in communities, a total, integrated plan for handling information generated by all types of laboratory testing - POCT and non-POCT - must be in place to ensure success.
What are the attributes of a good CHIN laboratory information system? As we have pointed out, an LIS system must have 1) rapid response even under load; 2) total reliability with redundancy and little or no unplanned downtime; 3) total connectivity to the laboratory instruments and LAN, hospital and campus backbones, and to the community, region, and world (via the Internet); and 4) total adaptability to the changing demands of healthcare using high-level languages and off-the-shelf tools.
Table 1 Sample cumulative report for POCT pH and ionized calcium Chemistry Serum April 5th, 17:45 Ref. range (mmol/l) Ion Ca++ 1.14 1.12-1.32 Ph 7.63 - ICA COR NA - Pulmonary tests: point-of-care blood gases Blood April 4th April 5th April 5th April 6th Ref. range 18:50 4:31 17:17 2:17 Ph 7.65 H 7.45 H 7.33 L 7.43 7.38-7.44 Blood April 6th April 7th Ref. range 15:46 3:57 Ph 7.47 H 7.40 7.38-7.44 All results from point-of-care instrumentation. Testing performed by nursing or respiratory care personnel. Blood bank: blood bank tests Blood April 4th April 4th April 4th 15:42 15:44 15:59(*) ABO/RH o-pos o-neg AB screen neg * Test done on 10/13/96 by J.S.
The workstations on a CHIN must be intelligent, use standard graphical user interfaces, and connect to the LAN by industry standard cabling, hubs, routers, etc. Peripheral devices such as printers, modems, faxes, etc., are also connected to the LAN.
Thus, the network becomes the more permanent aspect of the CHIN laboratory system. Traditional systems then function more as servers on the network; hence, a server can be replaced as needed without replacing the more permanent and expensive LAN infrastructure.
Instruments including POCT instruments can be interfaced via the LAN, obviating the need for multiplexers and modems and point-to-point cabling. Astute laboratorians soon will appreciate the power of the Internet and Intranets for communicating with remote clients (especially for test orders), transmitting POCT data to the LIS, returning results via E-mail, and compatibility of result data strings with the client's electronic medical records system. Basically, a good CHIN laboratory has a computer system that makes the laboratory a virtual laboratory, especially to outreach clients.
Making sense out of the data generated: Interpretive reports
It is clear that the days of printing endless spreadsheets of laboratory data are behind us. In fact, due to the plethora of reported data in some laboratories, the meaningful information is difficult to find on the report.
Accordingly, the clients/physicians in this market-driven paradigm, as well as third-party payers, are requiring that laboratory tests proceed along a more diagnostic line. Thus, results reporting and testing must be handled by an expert CHIN laboratory system that has expert decision-making capability. Expert programming must be flexible and in a script language, thereby easily adaptable to the changing requirements. The present hardcoded (vendor-supplied proprietary program) solutions on many LIS systems of today will not be acceptable. Profile testing will give way to reflex testing where the expert system makes appropriate decisions about whether to order or cancel additional tests or confirm or rule out a particular diagnosis.
Some limited interpretative comments also may be generated by the expert LIS, with these being reviewed by the laboratory technologist and directors prior to reporting.
Control and ownership issues
Laboratorians need to appreciate the value of information as an essential asset for doing business. The POCT CHIN model presented here requires the laboratory to maintain control of critical-care testing by ensuring 1) that the CHIN laboratory meets all regulatory requirements; 2) that appropriate and documented training of all POCT personnel has taken place; 3) that a timely review of calibration and quality control data takes place; 4) that all POCT results are verified prior to reporting; 5) that information management mechanisms are in place for acquisition, storage, and reporting of POCT results; and 6) that the laboratory has the means for automatic charge capture, coding, and billing for POCT.
What does such a CHIN laboratory Computer system look like?
In our current CHIN laboratory computer system, all information flows automatically to and from different components of the system with the central core laboratory computer system serving as the heart of the system. In general, outreach laboratories as part of a CHIN laboratory perform only POCT reported under separate headers, as described above, with major instrumentation for testing being located in the core CHIN laboratory. Such a system will continue to evolve and will require the active involvement of laboratorians to ensure it continues to meet the needs of our clients in the community and region.
The key to a successful CHIN laboratory computer system is the automated flow of critical laboratory information throughout the testing process, thus eliminating bottlenecks and manual steps in the test process; and foremost, ensuring good business practice throughout. Essentially, all testing - POCT and non-POCT - gets logged and tracked from initial order to final reporting, electronic charting, and billing. For the system to address client issues, the system must be modular and easily adaptable to individual client needs. It is only with proper design and control by laboratorians of the CHIN laboratory system described here that true control can be returned to the laboratory for the services they provide.
1. Herzlinger R. Market driven health care: Who wins, who loses in the transformation of America's largest service industry, New York, N.Y.: Addison-Wesley Inc.; 1996:1-111.
2. Blick KE. Selecting a laboratory information system. Clin Lab News. 1996;22:44
3. ASTM Committee (1991) E-31 ASTM specifications for low-level protocol to transfer messages between clinical laboratory instruments and clinical laboratory computers, ASTM Standards, Vol 14.01. Philadelphia, Pa.: American Society for Testing and Materials; 1991.
4. Kisner HJ. Talking about technology. Clin Lab Management Rev. March/April, 1995:130-133.
5. Blick KE. Community health information networks (CHIN) and their impact on laboratory practice, Proceedings, TEPRA96. 1996:231-234.
6. Skjei E. Health care takes is on the CHIN. CAP Today. 1995;9(1):32-35.
7. Boyce N. The health care system's changing climate, Clin Lab News. 1996; 22(8):10-11.
8. Blick KE. Laboratory Computer Systems: The critical factor in the survival of your laboratory, Advance for Administrators of the Laboratory. 1997. In press.
9. Skjei E. POC data-handling efforts more than lip service. CAP Today. 1996;10 (9):22-28.
10. Boyd JC, Felder RA, Savory J. Robotics and the changing face of the clinical laboratory, Clin Chem. 1996;42:1901-1910.
11. Blick KE. Decision making laboratory computer systems as essential tools for the achievement of total quality. Clin Chem. 1997. In press.
Kenneth E. Blick, PhD, is director, Laboratory Computer System, Department of Pathology, University of Oklahoma Health Sciences Center and The University Hospitals in Oklahoma City, Okla.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Critical Issues in Critical Care|
|Author:||Blick, Kenneth E.|
|Publication:||Medical Laboratory Observer|
|Date:||Sep 1, 1997|
|Previous Article:||Blood gas testing in the laboratory and at the point of care: finding the right mix.|
|Next Article:||Benchmarking for a competitive edge.|