Will Anyone See This Test Result?

We are all aware that there is substantial waste in testing. The mantra of utilization management is “the right test for the right patient at the right time.” This month, I want to focus on the right time. It turns out that many test results are never seen because they arrive after the patient has been discharged. This occurs for both routine and send-out testing. I will examine both.

Turnaround times for send-out testing are generally longer than those for tests performed in house. This means that results for tests ordered toward the end of a hospital stay are likely to be received after the patient has been discharged. Sendout tests are often expensive and, unlike tests performed in house, reducing sendout testing saves the hospital the full charge of the test. The savings can be substantial.

How do you prevent this? A recent article by Fang et al. shows one approach.[1] In this study, conducted at Stanford University, researcher displayed the cost and turnaround time of sendout tests in the computerized provider order entry (CPOE) system and achieved a 26% reduction in orders. I am aware of another hospital that restricts orders of sendout tests when the expected turnaround time is close to the expected remaining length of stay. Consider the graph in Figure 1. The upper panel shows the expected length of stay for a particular patient. The lower panel shows the expected turnaround time for a sendout test. In this case, there is a 62% chance that the test result will arrive after the patient has left the hospital.  Expected discharge dates are routinely kept and it is relatively easy to maintain a database of turnaround times. A hospital could combine these data and set a threshold for orders based on the probability that the result will arrive in time.

Standing orders are another source of waste.  I recently performed an analysis of the test rate as a function of the time until discharge (Figure 2). The test rate was 249 tests per hour for patients who were within 12 hours of discharge and 349 tests per hour for all other patients. It seems odd to me the testing rate in the final 12 hours is 70% of the “normal” testing rate. Further, the distribution of tests in both groups (those about to be discharged vs. all other patients) is very similar (Table 1). The main tests are basic metabolic panels and complete blood counts.  I suspect the majority of the testing within 12 hours of discharge is due to standing orders and the results were not needed for patient care.  The best intervention is less clear in this case because some peri-discharge testing is appropriate and it is difficult to distinguish the appropriate testing from the inappropriate testing. Education is one option. Perhaps the CPOE could raise a flag on orders for patients who are about to be discharged; however, this could be cumbersome and clinicians object to flags and popups that interfere with their workflow. I would be interested in readers’ thoughts on methods to reduce inappropriate peri-discharge testing.

In summary, some results do not reach clinicians in time to affect patient care. This is a source of waste. It is relatively easy to create an intervention to reduce inappropriate sendout testing but more difficult to reduce unnecessary peri-discharge testing.

 

Reference

  1. Fang DZ, Sran G, Gessner D, Loftus PD, Folkins A, Christopher JY, III, Shieh L: Cost and turn-around time display decreases inpatient ordering of reference laboratory tests: A time series. BMJ Quality and Safety 2014, 23(12):994-1000.

 

8-2017-fig-1
Figure 1: Comparison of expected length of stay (upper) and turnaround time (lower) for a sendout test.
8-2017-fig-2
Figure 2: Peri-discharge testing
8-2017-tab-1
Table 1: Test patterns stratified by time to discharge. The table shows the percentage of total testing accounted for each group. For example, BMP represents 15% of the total test volume among patients who are within 12 hours of discharge.

Schmidt-small

-Robert Schmidt, MD, PhD, MBA, MS is a clinical pathologist who specializes in the economic evaluation of medical tests. He is currently an Associate Professor at the University of Utah where he is Medical Director of the clinical laboratory at the Huntsman Cancer Institute and Director of the Center for Effective Medical Testing at ARUP Laboratories.

 

Does Price Transparency Improve Lab Utilization?

Physicians often have poor awareness of costs. For that reason, many believe that providing cost information to physicians would increase awareness that, in turn, could improve laboratory utilization. For example, costs of lab tests could be displayed as a field in the computerized provider order entry system. Interventions of this type are attractive because they are relatively inexpensive to implement and do not disrupt workflow with popups. Further, unlike other interventions, cost display is sustainable. Some interventions require constant training and followup whereas cost display is a one-time intervention. For these reasons, organizations are experimenting to see the effect of cost display on laboratory utilization.

Does cost display reduce lab utilization? Studies have shown wide variation in impact. Most studies have focused on orders for laboratory testing and imaging; however, a few studies have looked a pharmaceuticals.  A recent systematic review concluded that cost display is associated with a modest reduction in laboratory utilization.(1) The review included twelve studies on lab utilization and all of these showed improvement.(2-13) However, a more recent study by Sedrak et al. found that cost-display had no impact on utilization.(14) Similarly, two imaging studies found that cost-display had no effect on orders.(4, 15). There was a wide variation in impact: test utilization reduction ranged from 0% to over 30% in some cases. Overall, it appears that cost display tends to reduce utilization; however, it sometimes has no effect as shown in the Sedrak study. So far, cost display has never been associated with an increase in utilization. We have experimented with cost display at University of Utah and, like the Sedrak study, found no effect.

Why is there such a range of effects? Can we predict which organizations are likely to benefit? The short answer is that nobody knows.  The twelve studies on lab utilization where conducted in a wide range of settings (community, academic and pediatric hospitals), included different numbers of tests, or had other differences that could affect results. The way in which costs are displayed also varies. Some sites use the Medicare Maximum Allowable Reimbursement Rate, some use a series of dollar signs to indicate cost categories, and others use charges. It is not clear whether these differences matter.

There are a number of factors that might affect the impact of cost display. For example, cost display might have less impact at an institution that has an effective utilization management program in place because there is less opportunity for improvement. Or, the number of tests with costs displayed may have an impact. For example, some studies have displayed costs for a relatively few number of tests whereas other studies showed costs for a large number of tests.  Cost display for a few tests may send a different signal to providers than providing costs for all tests. Also, we don’t know how long the intervention works. Is there an initial effect that wears off? If so, how long does it last? These questions will need to be resolved by future studies.

In the meantime, should you provide cost feedback at your institution? It is hard to predict what will happen but most evidence suggests that you will see some improvement in utilization. It is not expensive to implement and some organizations have seen a significant impact. At worst, the evidence suggests that you will see no effect on testing behavior.  On balance, cost-display seems like a low-risk intervention.

 

References

  1. Silvestri MT, Bongiovanni TR, Glover JG, Gross CP. Impact of price display on provider ordering: A systematic review. Journal of Hospital Medicine 2016;11:65-76.
  1. Fang DZ, Sran G, Gessner D, et al. Cost and turn-around time display decreases inpatient ordering of reference laboratory tests: A time series. BMJ Quality and Safety 2014;23:994-1000.
  1. Nougon G, Muschart X, Gérard V, et al. Does offering pricing information to resident physicians in the emergency department potentially reduce laboratory and radiology costs? European Journal of Emergency Medicine 2015;22:247-52.
  1. Durand DJ, Feldman LS, Lewin JS, Brotman DJ. Provider cost transparency alone has no impact on inpatient imaging utilization. Journal of the American College of Radiology 2013;10:108-13.
  1. Feldman LS, Shihab HM, Thiemann D, et al. Impact of providing fee data on laboratory test ordering: A controlled clinical trial. JAMA Internal Medicine 2013;173:903-8.
  1. Horn DM, Koplan KE, Senese MD, Orav EJ, Sequist TD. The impact of cost displays on primary care physician laboratory test ordering. J Gen Intern Med 2014;29:708-14.
  1. Ellemdin S, Rheeder P, Soma P. Providing clinicians with information on laboratory test costs leads to reduction in hospital expenditure. South African Medical Journal 2011;101:746-8.
  1. Schilling UM. Cutting costs: The impact of price lists on the cost development at the emergency department. European Journal of Emergency Medicine 2010;17:337-9.
  1. Seguin P, Bleichner J, Grolier J, Guillou Y, Mallédant Y. Effects of price information on test ordering in an intensive care unit. Intensive Care Medicine 2002;28:332-5.
  1. Hampers LC, Cha S, Gutglass DJ, Krug SE, Binns HJ. The effect of price information on test-ordering behavior and patient outcomes in a pediatric emergency department. Pediatrics 1999;103:877-82.
  1. Bates DW, Kuperman GJ, Jha A, et al. Does the computerized display of charges affect inpatient ancillary test utilization? Arch Intern Med 1997;157:2501-8.
  1. Tierney WM, Miller ME, McDonald CJ. The effect on test ordering of informing physicians of the charges for outpatient diagnostic tests. N Engl J Med 1990;322:1499-504.
  1. Everett GD, Deblois CS, Chang PF. Effect of Cost Education, Cost Audits, and Faculty Chart Review on the Use of Laboratory Services. Arch Intern Med 1983;143:942-4.
  1. Sedrak MS, Myers JS, Small DS, et al. Effect of a Price Transparency Intervention in the Electronic Health Record on Clinician Ordering of Inpatient Laboratory Tests: The PRICE Randomized Clinical Trial. JAMA Internal Medicine 2017.
  1. Chien AT, Ganeshan S, Schuster MA, et al. The effect of price information on the ordering of images and procedures. Pediatrics 2017;139.

 

Schmidt-small

-Robert Schmidt, MD, PhD, MBA, MS is a clinical pathologist who specializes in the economic evaluation of medical tests. He is currently an Associate Professor at the University of Utah where he is Medical Director of the clinical laboratory at the Huntsman Cancer Institute and Director of the Center for Effective Medical Testing at ARUP Laboratories.

Test Turnover

What’s your lab’s procedure for getting rid of an obsolete test, or bringing in a new one? Any change to the test menu has some level of difficulty associated with it, however my opinion is that replacing a test with a new method that gives different results is the easiest to accomplish, followed by introducing a completely new test, and lastly removing an obsolete test. In that last case I’m specifically talking about removal of a test that is no longer the best way to diagnose a disease or monitor treatment.

In any of these cases, does your lab use the “rip the band aid off” method? For example, do you send a succinct notification that as of the first of the month this test will no longer be available in your lab, or this test will have results 33% higher than the doctors have been previously seeing? Or do you use a more gentle method, such as offering to run the old and new method together for 2-6 months to let the doctors get used to the new values? Or in the case of removing a test, do offer to try to find them an alternative lab which is still running the old test? Or do you simply leave the old test in place and hope it eventually dies a natural death from lack of use? Unfortunately, some tests never seem to die – like CKMB and bleeding time.

A lot of the difficulty, both in getting rid of old tests and in bringing in completely new ones, can probably be laid on the doorstep of human nature. Just like other humans, doctors like what they’re used to and don’t like changing their routine. Even when overwhelming evidence suggested that a new test is better (troponin), they want to use what they have always used to diagnose and treat their patients (CKMB). When the evidence for a test’s utility is not so clear cut, it’s even more difficult to introduce a new test. Examples of this include Cystatin C and fructosamine. Cystatin C is becoming more widely used and will no doubt survive as a test, but fructosamine? Part of the issue with fructosamine may have been the silly name they gave it. Fructosamine? Really? If they had called it glycated proteins/albumin it may have fared better. Fructosamine sounds too much like a fruit drink.

Maintaining a test menu that is appropriate for your population and that doesn’t include unnecessary or obsolete tests is an interesting balancing act. It definitely requires having a good rapport with your clinicians and getting their input along the way.

 

-Patti Jones PhD, DABCC, FACB, is the Clinical Director of the Chemistry and Metabolic Disease Laboratories at Children’s Medical Center in Dallas, TX and a Professor of Pathology at University of Texas Southwestern Medical Center in Dallas.