More than two decades ago, the Institute of Medicine defined “lab stewardship” as a management tool to guide clinical decisions through care that is respectful of and responsible to patient preferences, values, and needs (1). Interpretation of this definition evolved to include patient-focused considerations, such as access to the correct lab tests at a time during care that addresses the most salient clinical requirements. 

During the COVID-19 pandemic, U.S. patients experienced unprecedented access to lab testing that was made possible through smart-phone assisted conveniences of telemedicine and self/mobile-collection of various sample types. This was an inflection point in access-to-care that served clinical needs and mitigated infection exposures (2). 

However, increased access to testing without concordant clinical laboratory management (CLM) guidance will distract from patient-focused care. Test reports with corresponding reference ranges are equally important as ordering correct tests. For example, a reference range (RR) for some assays affected by third-spacing of fluids may differ between a non-pregnant and a pregnant female. To further illustrate this point, consider that the RR indicates test values corresponding to studies performed in a cohort of “normal” individuals with an absence of clinical conditions, including medications that may affect the test results. In the case of establishing an INR range for patients taking warfarin, for instance, such confounding variables would invalidate a RR study for a normal population. Therefore, an INR reference range of 2.0–3.0s is not normal because it’s the range for individuals on standard-intensity warfarin therapy. The INR RR for individuals on high-intensity warfarin therapy may be closer to 2.5–3.5s, and 0.9–1.1s for an apparently healthy population not taking anticoagulants.

In addition to maintaining clinical actionable lab reports and robust lab stewardship, CLM teams and individual providers must continuously ensure the right tests are ordered at the right time for the right patients. Furthermore, CLM teams must ensure that laboratory stewardship is not devalued through testing described as: Aberrant, Excessive, Incorrect, Obsolete, and/or Useless (AEIOU). To be successful in providing superior clinical services and to mitigate patient harm, lab stewardship operations must proactively target test-use through CLM that focuses on patient-centered guidance and directs efforts to eliminate “AEIOU” testing and asks WHY the test is being ordered. If the “why” cannot be answered, then ask, “What alternate course of action should be taken?” The illustrations below demonstrate the “AEIOU and always Why” in navigating lab stewardship operations.

A = Aberrant: Aberrant ordering is usually associated with high-volume or costly genetic testing in patients with low pretest probabilities (LPP). For example, for a patient being tested for hereditary hemochromatosis, ordering mutation panels for HFE gene variants is usually a common reactionary practice if iron studies, hemoglobin, and liver enzymes are minimally abnormal. Instead of rushing to genetic testing, transferrin saturation, a test usually performed in the hospital lab, should be ordered first to assess iron overload. Genetic testing should be performed only if an elevated transferrin saturation can’t be attributed to a reason for iron overload (eg, chronic transfusion). CLM teams could also use a SOFT approach when deciding to approve requests for expensive genetic testing. This approach explores:

S: Screening options (alternate targeted testing)

O: potential Outcomes (will management change based on test result)

F: curtailing ‘Fishing’ expeditions (ie, endless testing)

T: evaluating the request in the context of available Treatment (3)

Ordering approaches must also be considered in high pretest probability (HPP) cases. D-dimer testing is a good example of a test that should not be ordered in patients with a HPP for a venous thromboembolism (VTE). This is because the test depends on negative predictive value (based on low pre-test probability [LPP]), which makes its use inappropriate in groups with a high prevalence of VTE. Many results would incorrectly rule-out VTE in HPP patients. D-dimer testing should be used to rule-out VTE in patients with LPP, otherwise patient safety may be compromised. 

E = Excessive: Excessive ordering can be due to Fear-Of-Missing-Out (FOMO). Ordering serial cardiac troponin (cTn) based on FOMO of the peak and/or trough level contradicts American College of Cardiology guidelines. Knowing cTn peak or trough levels provides no additional value in assessing acute myocardial infarctions (AMI) once the rise-and-fall pattern is identified using a high-sensitive (hs) cTn test. Other FOMO examples include selecting multipanel allergy testing that usually provides useless additional information and potentially confusing false-positive results. A more patient-focused approach could include using guidelines in the setting of a HPP related to allergen exposure or referral to an allergist. FOMO-induced excessive testing can lead to unwarranted blood draws, incorrect order times, and patient harm, exacerbated in the context of delayed diagnoses.

I = Incorrect: Incorrect testing orders can result from failures to follow or implement processes. Process failures may occur if a routine parathyroid hormone (PTH) is ordered when the intraoperative (i.o.) PTH is required for a patient undergoing parathyroidectomy. Both tests are analytically identical. However, ordering routine PTH for a patient who requires the i.o-PTH is in complete disregard for lab preparation, which is essential for timely testing, reporting i.o-PTH results, having a successful parathyroidectomy, and ensuring patient safety. Incorrect orders can also be due to misunderstanding of analyte names. For example, ordering total testosterone to evaluate testosterone status would not account for free (ie, active) testosterone. Furthermore, for the most accurate and clinically actionable results (eg, for diagnosing “low-T”), the Endocrine Society recommends that testosterone testing should be performed on samples drawn from patients who are fasting and within 3-hours after waking, but testing is not restricted for patients who are receiving testosterone replacement therapy or androgen deprivation therapy. Another case in which the analyte name is often confused is vitamin D. Total 25-hydroxyvitamin D (25-OHD) levels, the sum of 25-OH-vitamin D2 and 25-OH-vitamin D3, is the appropriate indicator of vitamin D body stores and is available in most clinical labs. The most biologically active form, 1,25-dihydroxy-vitamin D, should not be used to assess vitamin D status because it can vary with calcium status. Therefore, measurement of 1,25-di-OHD is not useful in screening nutritional vitamin D deficiency. It can be misleading with normal values in patients with potentially severe vitamin D deficiency (4); 25-OHD should be used instead. The 1,25-di-OHD is a much more expensive send-out test that measures bioactive vitamin D and is in the differential of hypocalcemia, renal osteodystrophy, or chronic renal failure. The expensive 1,25-di-OHD is neither suitable for diagnosis of vitamin D deficiency nor for monitoring supplementation in most patients. Therefore, the combination of inappropriate testing and increased cost diminishes the quality of care when 1,25-di-OHD is ordered instead of 25-OHD. To reduce confusion between these tests, orders for 1,25-di-OHD could being changed to “Calcitriol” and restricted to orders from endocrinologists or through a lab approval process.

O = Obsolete: Obsolete tests are often antiquated and have been replaced with tests that are more technically designed to manage care. No one would consider ordering a pregnancy test in which hCG in the patient’s urine would induce egg production in a frog that received an injection of the patient’s urine. However, this was a common pregnancy test from the 1940s–1960s, boasting a turn-around-time of < 2 days (5). The bleeding time test (BTT), a “newer” obsolete assay, was/is used to evaluate platelet function and vascular integrity that would be better evaluated using coagulation tests (PT and aPTT, platelet count, and ruling out von- Willebrand’s disease). If the results of these tests are negative, a platelet function disorder can be investigated with platelet aggregation testing. Furthermore, the relationship between bleeding time and the risk of a patient’s actually bleeding has not been established and is not supported by the American Society for Clinical Pathology; therefore, the BTT should be removed from lab menus. It is highly operator–dependent, lacks reproducibility, and is confounded by technical variables such as incision location, pressure applied, operator experience, and patient factors (eg, age, gender, diet, hematocrit, skin laxity, medications). Some cardiac markers have also become obsolete in assessing AMI. Specifically, cTn has replaced the antiquated lactate dehydrogenase, creatinine kinase-MB (CK-MB), and myoglobin assays (6). The American College of Cardiology recommends hs-cTn as the only lab analyte for assessing acute chest pain.

U = Useless: Useless orders can be defined as those tests whose results will not change management and play no part in patient care. Ordering against guidelines, such as requesting a PSA on a 95-year-old male, has no clinical utility. Other cases may not be as obvious. For example, there is minimal justification for routine orders of reverse T3 (rT3) in thyroid function assessment. Perhaps the main use of rT3 is a prognostic marker of non-thyroidal illness or to monitor patients receiving amiodarone (7). Another example of a useless order is for methylmalonic acid (MMA) to assess B12 deficiency. Some providers use MMA results to identify a cause of low vitamin B12 levels. However, treatment for mild deficiencies has little value because disease progression in severity is rare. Furthermore, while elevated MMA can indicate vitamin B12 deficiency, it is not specific, does not implicate vitamin B12 deficiency in elderly patients, and provides no information on severity or likelihood of progression. Contributing to the low utility of MMA testing, HIBCH gene polymorphisms in people of European ancestry demonstrate elevated MMA levels, irrespective of vitamin B12 status (8). Perhaps MMA testing should be reserved for cases of suspected methylmalonic acidemia, a rare cause of devastating intellectual disability.

Summary points to remember

• Lab information computerization can be used to inform providers of appropriate reference ranges, to prevent duplicate orders, and to guide testing based on evidence-based medicine.
• Managing genetic testing approval based on prevalence, pretest probability, and clinical outcome supports laboratory stewardship and patient care.
• A multi-pronged approach, required to maximize value of lab use, must focus on test menus, reduce AEIOU orders, and implement processes to guide all in-house and send-out testing.
• To maintain the right test orders at the right time for the right patient, CLM teams must continuously maintain SOFT approaches in communicating with providers to mitigate AEIOU testing. After all, the most significant errors in lab medicine are those that compromise patient safety due to neglected stewardship when ordering tests and those that lead to failures related to inaction or misinterpretation of the results.

References    

1. Institute of Medicine: Crossing the quality chasm: A new health system for the 21st century. Washington DC: National Academy Press 2001. 
2. Roberts AJ, Malik F, Pihoker C, et al : Adapting to telemedicine in the COVID-19 era: Feasibility of dried blood spot testing for hemoglobin A1c. Diabetes Metab Syndrome 15:433–437, 2021. 
3. Abadie JM: Laboratory perspectives on test ordering and interpretation: Or when is a cholesterol test better than a $5,400 send-out? LabMedicine July 2007 Volume 38 No. 7; 401-403. 2007. doi: 10.1309/R9NEJWUP74KQ4A8V
4. Amrein, K., Scherkl, M., Hoffmann, M. et al: Vitamin D deficiency 2.0: an update on the current status worldwide. Eur J Clin Nutr 74:1498–1513, 2020. https://doi.org/10.1038/s41430-020-0558-y 
5. Sulman F, Sulman E: Pregnancy test with the male frog (Rana ridibunda). J Clin Endocrinol Metab 10 (8):933–938, 1950. https://doi.org/10.1210/jcem-10-8-933 
6. Abadie JM: Cardiac injury markers and a failed algorithm: can accurate assessment of acute myocardial infarction be cost effective? Mil Med 167(8):683–687, 2002. doi: 10.1093/milmed/167.8.683 
7. Gomes-Lima C, Burman KD: Reverse T(3) or perverse T(3)? Still puzzling after 40 years. Cleve Clin J Med 85:450–455, 2018. 
8. Molloy AM, Pangilinan F, Mills JL, et al: A common polymorphism in HIBCH influences methylmalonic acid concentrations in blood independently of cobalamin. Am J Hum Genet 98(5):869–882, 2016. doi: 10.1016/j.ajhg.2016.03.005