Anesth Analg 2007; 105:1711-1713
© 2007 International Anesthesia Research Society
doi: 10.1213/01.ane.0000286150.99529.f0
ECONOMICS, EDUCATION, AND POLICY
Changes in Utilization of Intraoperative Laboratory Testing Associated with the Introduction of Point-of-Care Testing Devices in an Academic Department
David B. Wax, MD, and
David L. Reich, MD
From the Department of Anesthesiology, Mount Sinai School of Medicine, New York City, New York.
Address correspondence and reprint requests to David Wax, MD, Department of Anesthesiology, Mount Sinai School of Medicine, 1 Gustave L. Levy Place, Box 1010, New York, NY 10029. Address e-mail to david.wax{at}mssm.edu.
Abstract
BACKGROUND: Availability of point-of-care testing (POCT) technology may lead to unnecessary testing and expense without improving outcomes. We tested the hypothesis that frequency of intraoperative blood testing (IBT) would increase in association with installation of POCT devices in our surgical suites.
METHODS: We performed a retrospective analysis of 38,115 electronic anesthesia records for cases performed in the 1 yr before and 1 yr after POCT installation. For each case, the frequency of IBT was tabulated and the change in frequency of IBT between the study periods was calculated for individual anesthesiologists, for the department as a whole, and for clusters of anesthetizing locations.
RESULTS: For the department as a whole, there was no significant change between the before and after study periods in the 13% proportion of cases in which IBT was obtained. For cases in which IBT was used, there was no significant increase in the number of IBTs per case.
CONCLUSIONS: We found no significant increase in the overall utilization of IBT associated with POCT presence in noncardiothoracic operating rooms.
Point-of-care testing (POCT) technology has made it possible to perform sophisticated diagnostic testing in perioperative areas, with more rapid results than are available from traditional, centralized laboratories. POCT, however, is generally more costly than central laboratory testing. If cost savings are to be achieved, therefore, savings must be realized indirectly from related benefits, such as prevention of unnecessary care, improvements in patient outcomes, or increases in efficiency of health care resource utilization (1).
There are limited data to suggest that POCT availability improves patient outcomes compared with central laboratory processing. Conversely, it is possible that the convenience of POCT leads to unnecessary testing and expense, consistent with the observed phenomenon that increased supply of health care services results in an increased utilization when the cost of the service is not considered by the ordering clinician. This "Roemer Effect" has been used to explain the increased demand for expensive diagnostic testing (e.g., magnetic resonance imaging) when the availability of such testing increases in a geographic area (2).
The present study characterizes the changes in intraoperative utilization of blood testing by anesthesiologists in a large academic medical center before and after the introduction of POCT. We hypothesized that frequency of intraoperative blood testing (IBT) use would increase.
METHODS
Our medical center operates a centralized laboratory that provides around-the-clock blood testing. Such testing requires that a blood sample vial and request form be sent to the laboratory in a plastic bag through our tube-transport system. Results are typically available by telephone or in the electronic medical record after approximately 20 min.
Since 1998, POCT of whole blood chemistry, hematocrit, and blood gas analysis has been available on a limited basis in our hospital's surgical suites. Testing requires approximately 0.3 mL of blood and 1 min for results to be printed on a paper slip. Before 2004, the POCT devices were only located in the cardiothoracic operating rooms (OR) and were operated primarily by perfusionists. In July 2004, two additional POCT stations (GEM Premier 3000; Instrumentation Labs; Lexington, MA) were added for general use to be shared among our approximately 40 noncardiothoracic anesthetizing locations. All anesthesia providers (faculty, residents, and certified registered nurse anesthetists) were provided with training and allowed unrestricted use of the devices. At all times before and after POCT installation, testing in the hospital's central laboratory remained available to clinicians. Regardless of where the blood sample analysis was done, test results were manually transcribed into our anesthesia information management system (AIMS) (CompuRecord, Philips Medical, Andover, MA). The testing location (i.e., POCT versus central laboratory) was not recorded in our AIMS.
With IRB approval, we retrospectively reviewed all noncardiothoracic electronic anesthesia records stored in our AIMS for cases performed in the 12 mo before (PRE) and after (POST) July 2004. For each case, the number of recorded test panels was determined. We obtained cost data as well as administrative data regarding the volume of tests done with the POCT devices to estimate the distribution of testing between POCT and central laboratory.
We calculated the change in frequency of use of IBT between the PRE and POST periods. This was calculated for the department as a whole, as well as for clusters of anesthetizing locations with consistent case mix. Changes in testing rates for individual attending anesthesiologists were also calculated.
RESULTS
By the latter half of the POST period, 81% of all IBT was being performed using the POCT devices rather than the central laboratory. Table 1 shows the utilization of IBT between the study periods. For the department as a whole, there was no significant change in the proportion of cases in which any IBT was used (13% in both periods). In cases where IBT was used, there was no significant increase in the total number of IBTs per case. Only one surgical cluster (with mostly vascular and gastrointestinal cases) demonstrated a statistically significant change in use, a 12% relative increase in use of any IBT as well as a trend towards more IBTs per case.
View this table:
[in this window]
[in a new window]
|
Table 1. Quantity of Intraoperative Blood Tests (IBT) in Noncardiothoracic Anesthesia Cases for Entire Department and in Six Anesthetizing Location Clusters Pre- and Postinstallation of Point-of-Care Testing (POCT) Devices
|
|
Frequency of testing was compared between the study periods for 52 individual attending anesthesiologists (working with or without a resident or certified registered nurse anesthetist assistant) who were present in both periods and performed a median of 316 and 366 anesthetics each in the PRE and POST periods, respectively. IBT rates for individual clinicians showed a decrease in median IBT rate from 13% to 12% (Wilcoxon's ranked sum test P < 0.01) between the PRE and POST periods, respectively. Changes in usage by individual practitioners ranged from –23% to +18%, with 65% of them showing a decreased frequency of IBT.
DISCUSSION
Within the first year of installation, POCT replaced our central laboratory as the predominant method for performing IBT. Residual use of the central laboratory is generally in cases in which an attending anesthesiologist is working alone (i.e., without another care team member who is able to leave the OR to perform POCT).
Overall, there was no evidence to suggest a major Roemer Effect associated with availability of POCT. Our data suggest that the addition of POCT did not have a major impact on the use of IBT for the department as a whole or for individual attending anesthesiologists present both before and after POCT deployment. In contrast to the expected Roemer Effect, our analysis of individual clinicians showed a small but statistically significant decrease in use. One surgical cluster in close proximity to a shared POCT workstation showed a small increase in IBT, but that cluster is typically used for more complex patients and cases that may warrant more IBT. Clusters in proximity to the other shared POCT workstation, however, did not show a significant increase in IBT, suggesting that proximity alone is not sufficient to increase usage. Nevertheless, we speculate that further increases in the number of workstations may reach some critical number (perhaps one in every anesthetizing location) that could add visual cues and convenience that result in increased usage.
Although there was no large increase in the use of IBT with the introduction of POCT, there was a large increase in testing expenditures due to the over fourfold higher cost per sample for the 81% of IBT performed by POCT that would otherwise have been performed in the central laboratory at its lower per sample cost. The ranges of cost per sample for IBT in our central laboratory versus POCT are $0.07–$1.13 vs $4.30–$5.05, respectively. These costs are established by contractual agreements with the testing device vendors, and pricing varies based on monthly testing volume and includes the workstation, disposables, and maintenance. Our data do not enable us to quantify the savings, if any, resulting from improved outcomes or operating efficiency attributable to POCT that might offset this increased cost.
We are unaware of any previous studies of utilization of POCT in the intraoperative noncardiac setting. One study of POCT in an emergency department reported that clinicians felt the speed of POCT was of benefit to some patients and prevented some unnecessary care, but there was no overall change in measured outcomes (3).
In conclusion, we found negligible changes in the overall utilization of IBT associated with the installation of POCT devices in noncardiothoracic ORs.
Footnotes
Accepted for publication August 3, 2007.
There is no financial relationship between the authors and any commercial interests with a vested interest in the subject matter herein. No outside funding was received in support of this work.
REFERENCES
- Gutierres SL, Welty TE. Point-of-care testing: an introduction. Ann Pharmacother 2004;38:119–25[Abstract/Free Full Text]
- Baker L, Birnbaum H, Geppert J, Mishol D, Moyneur E. The relationship between technology availability and health care spending. Health Aff (Millwood) 2003;W3-537-51
- Kendall J, Reeves B, Clancy M. Point of care testing: randomised controlled trial of clinical outcome. BMJ 1998;316:1052–7[Abstract/Free Full Text]
|
