Managing information from bedside testing.

Electronic transfer and consolidation of bedside testing data enhance quality assurance, access to test results, and monitoring of test utilization.

IMPROVING THE MANAGEMENT of information is one of the greatest challenges facing bedside testing programs today. Over the years, the technical capabilities of bedside testing devices and bedside test menus have undergone rapid expansion. Reports indicate similar expansion of test volume conducted at the bedside.[1-5] Unfortunately, development of laboratory information systems (LISs) for such testing has not kept up the same pace.[6]

In many facilities, bedside testing is one of the highest volume sections of the lab. In our hospital, for example, an average of 600 specimens a day are analyzed using bedside testing devices; only central chemistry exceeds this number. The magnitude of this testing means collecting and analyzing data typically require electronic transfer of this information to central computer systems.

* Negative consequences. Failure to transfer bedside information can have a number of negative consequences. Among them:

* Physicians have poor access to test results, which interferes with patient care and results in test repeats.

* Quality control (QC), equipment maintenance, competency of testing personnel, and other quality assurance (QA) issues become difficult to track, threatening the quality of testing.

* Information on test utilization is unavailable.

For labs to provide the high quality patient care required by the Clinical Laboratory Improvement Amendments (CLIA) and the Joint Commission on Accreditation of Healthcare Organizations (JCAHO), as well as to achieve optimal cost-effectiveness, improved standards are needed for bedside testing with regard to result reporting, QA, and utilization review.

* The benefits. In many bedside testing programs, the lab collects only enough data to monitor QC. While this is certainly important, much more information should be acquired and entered into central information systems. For example, information regarding personnel training and competency, instruments and reagents, and laboratory results with patient identification, collection time, operator identification, and comment codes are vital to providing effective bedside testing.

While compiling these data is a major undertaking, significant benefits can be derived. In some settings, such as in emergency departments or outpatient clinics, these records can be used to generate bills for reimbursement. Integrating data from various sources into one system also provides a valuable resource that can be described as a value-added network.(6)

* Two systems, one goal. The goal of our bedside testing program is straightforward: centralized electronic collection of data through our Office of Bedside Testing and result reporting through a central LIS. As a result, two bedside testing information systems have been implemented. One system collects data from analyzers used in five intensive care units (ICUs). Test results (for the last 50 specimens) for whole blood analyses of glucose, potassium, sodium, chloride, BUN, and hematocrit are stored in each analyzer's memory and downloaded each shift via infrared data ports to a central data station on each unit.

Results are transferred automatically by telephone lines to the Office of Bedside Testing and then to the LIS. Bedside testing results are printed in daily lab reports; results are accessible via the hospital information system at computer terminals throughout the hospital.

* Tracking test use. Before our current bedside testing information program, no one in our facility was able to accurately determine the number of bedside glucose tests being performed in each nursing unit. Test volume was guessed from the number of glucose test strips ordered. Though we estimated about 200,000 tests per year, we didn't know how many of these strips were being used for QC, discarded, or sent home with patients.

Thanks to our Office of Bedside Testing, we are now able to review the number of QC and patient results from each floor. This, in turn, enabled us to identify the nursing units that perform a low volume of testing and to discontinue testing in those areas. Additionally, these utilization data help us to budget for testing and provide records of nursing units' workloads.

* Problem reporting practices. Archived laboratory data and mechanisms designed to retrieve results are as important for bedside testing as they are for testing performed in the central lab. It is simply unacceptable for a central laboratory to report results over the phone to a health care practitioner without providing a written laboratory report. Yet where bedside testing is concerned, reporting results verbally to a physician and then handwriting them into nurses' progress notes or patient charts is common practice. Consequently, results are often difficult to locate and compare with results from the main lab.

When physicians have limited access to test results, patient care suffers and unnecessary repeat testing runs rampant, producing considerable added expense. Worse yet, misplaced test results increase liability risks. What would happen if physicians who were being sued for a patient's death claimed they had treated the patient appropriately for a low glucose or potassium level taken at the bedside but couldn't produce the test results to back up their assertions?

* Result reporting. A portion of one of our daily laboratory reports, including bedside testing results, is shown in Figure 3. Bedside results for each measured analyte are listed in columns designated "WB" (whole blood) to distinguish them from serum analyses performed in the central chemistry lab (for instance, "potassium" vs. "WB potassium"). We also chose to separate this information since reference ranges for whole blood and serum for selected analytes can differ,(7) although during our evaluation, we observed no significant differences.

Upon the recommendation of a JCAHO inspector during a recent survey, we are currently in the process of making slight modifications to our report format. We plan to indicate the lab site (bedside) where the testing is being done.

ICU physicians explicitly requested that bedside test results be placed adjacent to corresponding analyses from the central laboratory. Aligning these results, as shown in Figure 3, enables doctors to follow changes in an analyte over time, regardless of where results have been obtained (at the bedside or in the main lab). Side-by-side comparison of sequential values also provides ongoing delta checking, an important mechanism for assuring the quality of testing on a daily basis and for identifying patients with specific interferences.

* Data review. Our Office of Bedside Testing reviews test results daily to identify operator problems, error codes, and data omissions such as patient or operator identification numbers. Operators who make frequent mistakes are retrained. Data from each unit are compiled into monthly utilization and QA reports.

When bedside analyzers detect a potential problem with a test cartridge or specimen, error codes, rather than test results, are produced and reviewed carefully. A summary of repeat testing for two ICUs is shown in Figure 5.

* System costs. Data systems for on-line transfer of information to an LIS require a capital investment, just as adding electronic interfaces to other laboratory instrumentation incurs capital expense. A central data station (a personal computer plus software) lists for about $3,500; telephone lines, moderns, and cables are usually additional. A more sophisticated interface with an LIS can cost up to $10,000.

In our system, five central data stations service four ICUs. Total capital expense per ICU is less than typical costs for interfacing a laboratory instrument to the LIS.

Adding data systems to bedside testing instruments increases analyzer costs, too. These expenses may be overlooked as a separate component, however, and included in the total cost of the instrument.

* An extension of central labs. Improving the transfer and centralization of data generated from bedside testing instrumentation provides the communication link necessary for this new technology to serve as a true extension of the main laboratory. Improved information systems provide the means for bringing bedside test review, result reporting, QA, and review of test utilization up to the standards of central labs.

Figure 5

Monthly review of cartridge use and test repeats for two ICUs

                                      Neonatal     Neurosurgical

Cartridge use

6+                                       341            654
EC4+                                     758            252
Glucose                                  509            266
Errors (test failures)                   301             97

Total                                  1,909          1,269

Error codes

(20/77) Calibrator bubbles                12              5
(40/78) Chip crosstalk                     7              0
(24/42) Prerupture                         2              1
(31) Seal failure                         45             14
Other                                      1              1

System failure total                      67             21

(21) Prerupture at use                     6              3
(44/45) Sample bubbles                    34             13
(38/39) Bubbles/short sample              37             26
(35/36) Short sample                      49              7
(30/37) Sample overfill                   28             14
(34/76) Sample clot                       80             13

Sample/user error total                  234             76

Wastage/error rate                        15.8%           7.6%

Repeat rates were about 16% for specimens collected from newborns and 8% for
those collected from adults. Error codes identified that repeats were due to a
problem in the neonatal ICU with expelling blood from capillary tubes into a
cartridge, causing bubbles to form. Repeats were decreased by switching to
tubes with a plunger.

Achieving high standards for bedside testing is imperative when one considers that it is often used to treat critically ill patients and, therefore, influences life-threatening medical decisions. Manufacturers of these devices must carefully consider data storage and transfer capabilities in designing these products, and users must carefully evaluate these systems before bringing them on board.

References

1. Zaloga GP. Evaluation of bedside testing options for the critical care unit. Chest. 1990; 97: 1855-1905.

2. Kost GJ. New whole blood analyzers and their impact on cardiac and critical care. Crit Rev Clin Lab Sci. 1993; 30: 153-202.

3. Woo J, McCabe JB, Chauncey D, Schug T, Henry JB. The evaluation of a portable clinical analyzer in the emergency department. Am J Clin Pathol. 1993; 100: 599-605.

4. Travers EM, Wolke JC, Johnson R, et al. Changing the way lab medicine is practiced at the point of care. MLO. 1994; 26(7): 33-40.

5. Geyer SJ. Joining the technological evolution in health care. MLO. 1992; 24(9S): 2-7.

6. Friedman BA, Mitchell W. Integrating information from decentralized laboratory testing sites: The creation of a value-added network. Am J Clin Pathol. 1993; 99: 637-642.

7. Ladenson JH, Tsai LMB, Michael JM, Kessler G, Joist JH. Serum versus heparinized plasma for eighteen common chemistry tests. Am J Clin Pathol. 1974; 62: 545-552.

Figure 1

Benefits of bedside testing data systems

* Provide better access to patient data

* Help consolidate laboratory records

* Track test utilization (per unit, patient, and physician)

* Provide for billing of tests

* Help track personnel training and competency

* Improve monitoring of QC, QA, and instrument maintenance

* Offer record retrieval to avoid liability issues

* Allow ongoing comparison (delta checking) of bedside vs. central lab results * Facilitate workload recording

* Enable cost accounting/budgeting

* Acquire documentation required by lab regulations

Hortin is associate professor of pathology at the University of Alabama at Birmingham and medical consultant for the Office of Bedside Testing at the University of Alabama Hospital. Utz and Gibson are bedside testing coordinators there.