Point-of-care testing versus central lab costs.
In the past 10 years, a number of papers have been written that analyze the cost of point-of-care testing (POCT) as compared with the cost of performing the same tests in the central laboratory. Approximately half of these papers reported that point-of-care testing reduced costs compared with central laboratory testing, whereas other authors concluded that point-of-care testing was more expensive (see annotated bibliography). Two papers looked at the effect of POCT on length of stay: First, Zaloga (1988) wrote a report without detailed information about his study design, which found that POCT shortened the length of stay in critical care units. Parvin (1996), on the other hand, concluded that POCT did not reduce the length of stay in an emergency care unit. However, if one examines a subset of Parvin's patients, the opposite conclusion can be drawn.
In addition to the opposing viewpoints noted above, almost all of the papers listed in the bibliography that follows this article on p. 52 analyze laboratory costs without considering the total cost of the episode of care. This article examines the different characteristics of POCT versus central lab testing costs, discusses which costs should be compared, and offers conclusions as to why comparison is difficult.
Laboratory costs. Whether performed in the central lab or at the point-of-care, all laboratory testing costs can be divided into fixed costs and incremental or variable costs (see Table 1). Fixed costs are those that are not related to the number of tests performed. These expenses include costs related to the facility itself: direction of the laboratory, capital equipment acquisition, maintenance and cleaning, and much of the labor involved in running the laboratory. Incremental costs, on the other hand, are directly related to the number of tests performed, and include the cost of supplies and reagents and the direct labor cost of test performance.
Episode of care costs. A single episode of care, on the other hand, entails a larger picture of costs to the hospital or other care facility for the treatment of a specific health problem. Costs related to the episode of care include the length of stay in the special care unit or hospital, nursing care costs in the hospital, professional and technical services, physician services, respiratory therapy, pharmaceuticals, IV therapy, and collateral diseases and complications. Zaloga reported that in cases of diabetic ketoacidosis, the average cost of hospitalization was $3,925, while the cost of glucose testing was $126. Figure 1 is a graphical representation of laboratory costs as compared with the total cost of the episode of care in the hospital. These costs, however, are dwarfed when compared to the long-term follow-up cost.
Long-term follow-up costs. Beyond a single episode of care, the long-term health of an individual might be influenced by the kind of lab testing he or she receives. Costs of long-term medical care include the ongoing expenses incurred by the individual, family, and insurance carrier as a disease progresses or has complications. Long-term medical care may also include both direct costs of continued medical care or chronic care and indirect costs of lost income and productivity. These costs can sometimes be avoided if appropriate medical care is given early in the disease or throughout the person's illness. Using the example of a patient admitted with diabetes: the use of glucose monitoring promotes optimal glucose regulation and has been shown to reduce long-term complications of diabetes such as renal failure, cardiovascular disease, and blindness.
Point-of-care testing and central laboratory testing are vastly different in the allocation of costs. In a central laboratory compared with POCT, fixed costs are higher and the incremental costs of test performance, including labor and consumables, are lower. On the other hand, in POCT, fixed costs are relatively low and labor and consumables increase significantly with increased testing. Illustrated in Table 1 are the fixed and incremental costs for central lab versus POC testing.
In the central laboratory, test volume has only a small relationship to the total cost because fixed costs are such a large part of the total cost. When test volume falls, most fixed costs remain, including the labor for administration, specimen collection, processing, and quality management. Incremental costs of labor and consumables are relatively insignificant. In contrast, in POCT, fixed costs are low compared to incremental costs (see Table 1). Consumables and labor represent a large portion of the total cost because more expensive consumables are used and testing is more labor intensive (in the central lab, automation keeps labor per test low). As the test volume increases, incremental costs rise, exerting a major influence on the total testing cost.
POCT cost analysis
Several groups have recognized that methods for performing POCT cost analyses are not standardized and that these studies cannot be compared. If meaningful data are to be collected and analyzed, standard rules for data collection must be developed. The College of American Pathologists had named a committee to develop standard procedures for POCT cost analysis, but this committee took no action and was subsequently disbanded without making recommendations. Instead, CAP set up a study on the costs of POCT glucose analysis as part of its 1997 Q-Probe program (Howanitz and Jones, 1998). In addition, the National Committee on Clinical Laboratory Standards is in the process of developing standards for POCT cost analysis studies.
Problems with cost studies. Current studies of POCT generally only consider the incremental costs of testing labor and consumables (Felder, 1995; Fleisher, 1995; Laposata, 1995; Tsai, 1994). Many assumptions about the costs of non-laboratory personnel have also been made. Frequently, these studies suggest that if individuals were not performing tests, they would be idle. Thus, no labor cost is assigned to these individuals. Second, these studies do not consider the different roles of fixed and incremental costs on the central laboratory as compared with POCT. Third, almost none of the studies has considered the costs of the episode of care or the patient's long-term follow-up costs.
Value versus cost. As previously mentioned, the cost of laboratory work is small compared with the cost of the episode of care or the total lifetime medical costs to the patient. Many who have studied the costs and benefits of POCT have failed to recognize or report that the value of laboratory tests may be much greater than the cost of testing. Savings should be measured by considering:
* shortened episodes of care
* lowered long-term follow-up costs
* improved long-term quality of life.
[TABULAR DATA FOR TABLE 1 OMITTED]
In Figure 2, the comparative size of the boxes illustrate that, even with increased laboratory testing costs, it is possible to lower the overall cost of the episode of the care. This potential cost saving is, however, just speculation because only a few studies have considered costs beyond the immediate testing expenses (Trundle, 1986; Zaloga, 1988).
Future cost comparisons. In future studies that compare central laboratory testing with POCT, investigators need to consider the global cost of the episode of care rather than only the laboratory costs. These studies should use automated cost analysis systems that capture all of the components of care that are delivered to a patient. Some medical centers are currently using such systems to collect cost and utilization information and perform cost analyses.
Can this be done? Perhaps, but it will be difficult and complex. Part of the reason for this is that every service provided to the patient must be entered into the computer. Many POC tests, such as stool occult blood and capillary blood glucose measurements, are not recorded in a computerized record but only on a flow chart or in a progress note; and no charges are made because in an environment of diagnosis-related groups (DRGs), reimbursement is made on the basis of the diagnosis rather than the services provided, so there is no financial incentive to capture these expenses. Frequently, the cost of data capture exceeds the cost of providing this service itself. For this reason, information about services provided to a patient is incomplete.
Evaluating current cost comparison studies. Each cost study is specific to the test system used, the volume of tests, and the specific institution. Most are not comprehensive or specific about all the test-related costs that were compared, and few measure all of the fixed costs. It is both difficult and hazardous to generalize information from an individual study to different settings. Because the episode of care costs much more than the laboratory tests, one should give little credence to studies that do not measure the total cost of an episode of care to determine whether the change in location of testing made any therapeutic difference for the patient, such as if POCT provided more rapid results that in turn made treatment available sooner and shortened the patient's length of stay. Although long-term follow-up costs are the most important costs to society, it is virtually impossible to measure the influence of testing on this.
Making a decision
If you need to choose between POCT versus central lab testing, do not wait for hard data. As discussed, reliable data have not yet been published. Moreover, it is not likely that we will see reliable studies of POCT costs until a consensus on standards for collecting and analyzing data can be determined. At present, it is best to judge the appropriateness of using POCT based on your particular situation. Place the greatest weight on reducing the cost of the episode of care and preventing the progression of illnesses that will incur costs in the future.
The following articles include the major cost studies and commentaries on POCT cost analysis.
1. De Cresce R, Phillips DL, Howanitz PJ. Financial justification of alternate site testing. Arch Path Lab Med. 1995;119:898-901. Opinion only. Comments: Authors review literature and analyze problems with previous studies that did not permit comparisons. They recommend a standardized approach to cost analysis.
2. Felder RA. Robotics and automated workstations for rapid response testing. Am J Clin Path. 1995;104(Suppl 1):S26-S32. Study site: University-affiliated hospital nursing units. Test category: Blood gases. Includes lab costs data. Effect: Using RALS for 1 year reduced costs for couriers by $19,000, nursing costs by $22,750, and supply use by $3,900. However, the on-site maintenance of RALS will require 20% more labor than blood gas systems in the central laboratory. Comments: A decentralized robotic system reduces blood gas cost compared with central lab testing.
3. Felder RA, et al. Development of a robotic near patient testing laboratory. Arch Pathol Lab Med, 1995;119:948-951. Study site: University hospital nursing units. Test category: Blood gas, electrolytes, glucose, hemoglobin. Includes lab costs data. Effect: POCT reduced lab costs by lowering lab labor and transportation costs. It also reduced TAT. Comments: Study shows reduced costs with use of the robotic unit on nursing units. No data are available on changes in patient care or length of stay.
4. Fleisher M, Schwartz MK. Automated approaches to rapid-response testing. Am J Clin Path. 1995;104(Suppl 1):S18-S25. Study site: Cancer center special care unit satellite lab. Test category: Chemistry tests. Includes lab costs data. Effect: POC whole blood glucose test is $5.54 versus $1.96 in the central lab. POC TAT is less than 10 minutes versus TAT of less than 20 minutes central lab. After installation of a pneumatic tube system, central lab TAT is shorter than POCT. Comments: Authors state that with an effective automated transport system, central lab testing TAT and costs are lower than POCT.
5. Frankel HL, et al. Minimizing admission laboratory testing in trauma patients: Use of a microanalyzer. J Trauma. 1994;37:728-736. Study site: Trauma unit. Test category: NOVA SP5 compared to standard admission panel. Includes lab costs data. Effect: POCT costs less due to reduction of tests performed. TAT is greatly shortened with POCT. Comments: Standard central lab test results were usually available only after therapeutic decisions had already been made. Shorter TAT with POC tests gave better pre-operative information even though fewer tests were performed.
6. Fuhrman SA, Travers EM, Handorf CR. The mobile laboratory in alternative site testing. Arch Path Lab Med. 1995;119:939-947. Study site: VA medical center. Test category: Blood gases, electrolytes, chemistry, hematology. Includes lab costs data. Effect: Lower costs with use of mobile lab. Central lab blood gas cost $9.45. Mobile lab blood gas costs: $10.94 if run stat by technologist: $7.71 if run routinely by technologist; $7.18 if run by respiratory therapist. Comments: In practice, this system has not been generally used and where used, costs are higher.
7. Green M. Successful alternatives to alternate site testing: Use of a pneumatic tube system to the central laboratory. Arch Pathol Lab Med. 1995;119:943-947. Study site: Emergency department and operating room. Test category: Blood gases, electrolytes, glucose. Includes lab costs data. Effect: Reduced staff by 16 FTEs and improved TAT to 7 minutes by closing satellite labs operated by non-lab personnel. Comments: Author recommends central lab processing with good pneumatic tube transport system.
8. Greendyke RM. Cost analysis; Bedside blood glucose testing. Am J Clin Path. 1992;97:106-107. Study site: VA long-term care and psychiatric: hospital. Test category: Blood glucose. Includes lab costs data. Effect: POCT is more than 3 times as expensive as central lab testing for blood glucose. Comments: POCT capillary blood glucose cost $11.50 per test versus $3.19 in the central lab. Direct and indirect costs including labor, supplies, quality management, and training were analyzed.
9. Handorf CR. POC testing: Must quality cost more? MLO Point-of-care Supplement. 1993. Opinion only. Comments: Discusses methods of cost analysis and failure of studies to produce meaningful data.
10. Howanitz PJ. College of American Pathologists Conference XXVIII on alternate site testing: What must we do? Arch Pathol Lab Med. 1995;119:979-983. Opinion only. Comments: Conclusion of conference was that there is a need to develop a fiscal understanding of the resources needed for POCT and the benefits derived. Cost analysis of POCT is extremely complex. Author provides a chart proposing cost-accounting elements to be considered.
11. Howanitz PJ and Jones BA. Q-Probes 97-05. Cost of bedside versus laboratory glucose testing: Data analysis and critique. Northfield, IL: CAP; April 1998. Publication is a summary of data from 445 institutions.
12. Innanen VT, Barquiera-de Campos F. Point-of-care glucose testing: Cost savings and ease of use with the Ames Glucometer Elite. Clin Chem. 1995;41:1537-1538. Study site: General hospital nursing units. Test category: Capillary blood glucose diabetes monitoring. Includes lab costs data. Effect: Number of strips used and cost for strips decreased by about 30% when the Elite was used instead of the Glucometer M+. Reduction was due to fewer repeat tests. Comments: Study did not compare POCT versus central lab testing, but shows that costs with different systems can be variable.
13. Keffer JH. Economic considerations of point-of-care testing. Am J Clin Path. 1995;104(Suppl 1):S107-S110. Opinion only. Comments: Author reviews and comments on cost analysis literature regarding POCT. He comments on the inadequacies of previous studies.
14. Keffer JH. Point-of-care testing and length of stay. Clin Chem. 1997;43:859. Opinion only. Comments: Letter to editor critical of a paper by Parvin, et al. Impact of POCT on length of stay in a large emergency department. Clin Chem. 1996;42:711-717. Letter identifies problems with conclusions in the paper.
15. Kost GJ. Point-of-care testing in intensive care. In: Tobin MJ, Principles and Practice of Intensive Care Monitoring. New York: McGraw-Hill; 1997: 1267-1296. Comments: Review of some literature regarding economics of POCT. Kost concludes that there is no adequate generalization regarding cost. Analysis must take into account other non-lab costs related to prolonged length of stay in high-cost care units.
16. Laposata M, Lewandrowski KB. Near patient blood glucose monitoring. Arch Path Lab Med. 1995;119:926-928. Study site: University-affiliated municipal hospital Test category: Capillary blood glucose. Includes lab costs data. Effect: Lower cost for POC glucose if volume is greater than 5 tests per day. Central lab glucose cost was $3.84 per test. One or two bedside tests per day cost $7.50 per test. Five or more per day per site cost $3.50 each. If fewer than 1 test per day was run, the cost was $18.00 each. Comments: Bedside glucose testing is not inherently more expensive than testing performed in the central laboratory; the increased cost of bedside glucometry over laboratory testing can be significantly minimized by involving a limited number of healthcare workers and performing assays only on clinical units where glucose testing is required more than five times per day.
17. Lee-Lewandrowski E, et at. Utilization and cost analysis of bedside capillary glucose testing in a large teaching hospital: Implications for managing point-of-care testing Am J Med 1994;97:222-230. Letter to editor. 1995;99:976-977. Study site: University-affiliated hospitals. Test category: Capillary blood glucose. Includes lab costs data. Effect: Average cost of POCT was $4.19 (range: $3.08-$48.16 depending on test volume) compared with $3.84 in central lab. Much of the cost of POCT can be attributed to the high costs of quality control, quality assurance, training, and documentation.
18. Leverone JP. Alternate site testing: Financial considerations. Arch Pathol Lab Med. 1995;119:961. Opinion only. Comments: Report of break-out group discussion at CAP conference on alternate site testing. Discusses the complexity of issues to be considered in a global cost analysis.
19. Nosanchuk JS, Keefner R. Cost analysis of point-of-care lab testing in a community hospital. Am J Clin Path. 1995;103:240-243. Study site: Community hospital. Test category: Stat chemistry panel. Includes lab costs data. Effect: POCT costs exceeded central lab testing by 1.1 to 4.6 times. Comments: Acquisition of automated transport system and improved computer was less costly than use of POCT.
20. Parvin CA, et al. Impact of point-of-care testing on patients' length of stay in a large emergency department. Clin Chem. 1996;42:711-717. Study site: University-affiliated emergency department. Test category: I-STAT EC6+. Includes lab costs data and length of stay data. Effect: Use of I-STAT did not reduce ER length of stay. Comments: Study was inconclusive because many patients also had central lab testing performed and remained in the ER pending receipt of those lab results. If a subset of patients who had only POCT is considered, the length of stay was actually shortened.
21. Santrach PJ, Burritt MF. Point-of-care testing, Mayo Clin Proc. 1995;70:493-494. Opinion only. Comments: Published and unpublished cost analyses show few consistent trends. Authors list lab-related costs to be considered in a cost analysis.
22. Singal BM. Point-of-care testing and cost-effective emergency medicine. Academic Emerg Med. 1995;2:163-164. Opinion only. Comments: Review and discussion of previous papers by the author. They indicated that stat POC electrolyte panels had low yield of abnormal results. These papers discussed use of electrolytes in decision making.
23. Thiebe L, Vinci K, Gardner J. Point-of-care testing: Improving day-stay services. Nursing Management, OR/Ambulatory Surgery Edition. 1993;24(12):54-56. Opinion only. Comments: Authors state opinion that POCT saves time and improves work flow.
24. Trundle DS, Weizenecker RA. Capillary glucose testing: A cost-saving bedside system. Laboratory Management. May 1986;59-62. Study site: Community hospital. Test category: Capillary blood glucose. Includes lab costs data. Effect: POCT cost $1.69 per test, hospital length of stay was 7.1 days. Central lab testing cost $3.53 per test, length of stay 8.9 days. Comments: Costs restricted to direct labor and reagents.
25. Tsai WW, Nash DB, Seamonds B, Weir GJ. Point-of-care versus central laboratory testing: An economic analysis in an academic medical center. Clinical Therapeutics. 1994; 16:898-910. Study site: Emergency department. Test category: Chem 7 and CBC. Includes lab costs data. Effect: Central lab testing is less expensive, but therapeutic TAT much shorter with POCT. Comments: Authors speculate that global costs would be, lower and efficiency of emergency department would be improved.
26. Winkleman JW, Wybenga DR. Quantification of medical and operational factors determining central versus satellite laboratory testing of blood gases. Am J Clin Path. 1994;102:7-10. Study site: University-affiliated hospital neonatal ICU. Test category: Blood gases. Includes lab costs data. Effect: Central lab tests cost $3.52 each. POC satellite lab cost $8.98. Pneumatic tube used to transport specimens to lab. TAT 1.5 minutes longer for specimens sent to central lab. Comments: Authors conclude that if an efficient transport system and protocols that give priority to urgent tests from high-impact areas are in place, central lab testing is more cost efficient and no slower than POCT.
27. Winkleman JW, Wybenga DR, Tanasijevec MJ. The fiscal consequences of central versus distributed testing of glucose. Clin Chem. 1994;40:1628-1630. Study site: University-affiliated hospital nursing units. Test category: Capillary blood glucose. Includes lab costs data. Effect: POCT is twice as expensive as central lab testing ($6.62 vs $3.30 per test). Pneumatic tubes are used to transport specimens to the lab. TAT is 1-2 minutes longer for specimens sent to the central lab. Comments: Authors conclude that if an efficient transport system and protocols that give priority to urgent specimens from high-impact areas are in place, central lab testing is more cost efficient and not significantly slower than POCT.
28. Woo J, et al. The evaluation of a portable clinical analyzer in the emergency department. Am J Clin Path. 1993;100:599-605. Study site: University-affiliated emergency department. Test category: I-STAT. Includes lab costs data. Effect: I-STAT cost $10/panel versus central lab costs of $3/panel. TAT for the I-STAT is, 2 minutes. Comments: Cost analysis is restricted to direct labor and reagent costs.
29. Zaloga GP. Bedside reagent testing: Blood, CSF, and bacterial cultures. J Critical Illness. 1988;3:85-94. Study site: University-affiliated hospital ICU. Test category: Various chemistry tests. Includes lab costs data and length of stay data. Effect: Bedside glucose testing for patients in ICU for diabetic ketoacidosis leads to a global cost of $2,380 per admission versus $3,92, per admission when central lab testing is performed. ICU days reduced by 0.9 days and hospital stay reduced by 3 days. Bedside oximetry for postoperative patients lead to global costs of $2,400 versus $3,150 when central lab testing was done. ICU and hospital days were reduced by 1 day. Electrolyte testing and therapeutic action using the central lab averaged 150 minutes versus almost immediate action when using POCT Comments: Cost savings are due to rapid therapeutic action, allowing more rapid therapeutic response and discharge from ICU and hospital.
Authors reporting that POCT reduces costs Felder (2)
Felder et al (3) Frankel et al (5) Fuhrman et al (6) Laposata; Lewandrowski (16) Trundle and Weizenecker (24) Zaloga (29)
Authors reporting that POCT increases costs
Fleisher and Schwartz (4) Green (7) Greendyke (8) Lee-Lewandrowski et al (17) Nosanchuk and Keefner (19) Tsai et al (25) Winkelman and Wybenga (26) Winkelman et at (27) Woo et al (28)
Daniel M. Baer is professor emeritus of laboratory medicine, Oregon Health Sciences University, Portland, OR.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||includes bibliography|
|Author:||Baer, Daniel M.|
|Publication:||Medical Laboratory Observer|
|Date:||Sep 1, 1998|
|Previous Article:||Optimizing your instrument training experience.|
|Next Article:||Risks and recommendations for HIV occupational exposures.|