Andrew S. Parsons, MD, MPH, Joesph R. Wiencek, PhD
Low-value services are a major contributor to healthcare waste in the US. Commonly defined as patient care that could be eliminated without reducing health quality, these services carry an estimated cost of around $800 billion per year.
When ordered in a clinically stable hospitalized patient, repetitive laboratory testing (aka daily labs) qualifies as one of these low-value services. Repetitive testing can put patients at risk: it can lead to iatrogenic anemia in an estimated 20 percent of hospitalized patients, and ordering providers may act on clinically irrelevant abnormals or trigger the testing cascade, subjecting patients to further unnecessary and potentially harmful diagnostic and therapeutic interventions.
Repetitive laboratory testing is therefore an ideal target for quality improvement interventions. Such interventions have been highlighted by multiple medical societies as part of the Choosing Wisely campaign, an initiative led by the American Board of Internal Medicine Foundation to educate providers and patients on potential areas of overuse. Effective initiatives to reduce unnecessary laboratory utilization bend the cost curve and fulfill the physician's first oath to do no harm. In this article, we aim to highlight effective approaches to reducing overutilization of daily labs.
Audit. Individualized and group feedback on laboratory ordering practices is an effective way to reduce unnecessary utilization. Studies have found that emails detailing provider-specific test ordering information sent at routine intervals or easily accessible clinician dashboards reduce the number of common laboratory tests ordered and laboratory costs per patient day. Both anonymous comparison data and individualized reports appear to be effective. These reports can be sent to all ordering providers or they can be targeted to high utilizers. The limitation with this approach, if used alone, is sustainability, because laboratory ordering rates tend to return toward baseline rates after feedback reports are discontinued.
Buy-in. Institutional buy-in is vital to transitioning from a culture of waste to one of high-value care. To promote long-lasting change, the buy-in must have a multidisciplinary approach. First and foremost, buy-in must come from the institutional leadership and key chief executives, commonly referred to as the C-suite. Discussions with these key individuals must be strategically aligned with the institution's immediate and long-term goals. Support from the C-suite can often provide the leverage needed to facilitate pilot studies as well as resources such as project managers who could help tackle larger initiatives. It is important that these key members are updated on a regular basis about current and future projects, projected cost savings, and overall impact on patient care. Second, outside the C-suite, it is important to engage colleagues about various stewardship efforts and assure them that the primary focus is on appropriate care of the patient rather than on reducing costs. Sharing data and having open discussions in the form of departmental grand rounds, other small presentations, or informal one-on-one conversations can help spread the message of high-value care.
Competition. Even when done anonymously, comparing ordering practices among clinicians provides an incentive for change. At the University of Virginia Medical Center, we have an ongoing chief resident–led initiative in which internal medicine residents receive a weekly email containing individualized and team daily lab-ordering practices in comparison with those of their peers. This intervention, based on a similar initiative at Vanderbilt University Medical Center and paired with an educational online module, has shown an immediate and sustainable reduction in repetitive laboratory utilization.
Decision support. The electronic health record (EHR) may be an underexplored resource to reduce unnecessary lab tests. Through EHRs, clinical care teams can gather information from a patient's visit in real time and track lab tests with a click of a button. EHRs can also provide decision support at the point of entry and restrict tests to specific locations, patient populations, or ordering clinicians. The support tools can include rules on frequency (e.g., daily labs), sex-specific tests (e.g., prostate-specific antigen), costly send-out tests (e.g., genetics), and many more. They are typically vetted by decision support committees that determine the best approach to implementation. There are also additional third-party decision support systems that can be incorporated into EHRs. However, these systems are in the early stages of development; it is therefore important that organizations evaluate and validate support tools before full adoption. Regardless of internal or external options, a well-designed decision support system may help clinicians make better use of laboratory tests, but more data is needed.
Education. Interventions focused solely on education have found mixed results. Educating clinicians commonly leads to an initial decrease in laboratory ordering practices, but the approach lacks sustainability. There is evidence that resident trainees are particularly prone to overutilization of common laboratory tests, which is not surprising given the known association between diagnostic uncertainty and increased laboratory ordering. Additionally, a recent study found that nurses and advanced-practice clinicians may be more prone to overutilization of common laboratory tests compared with residents and attending physicians. We recommend educational initiatives at all levels. Medical students should be taught principles of high-value care, and attending physicians should role-model high-value practices for residents. At the University of Virginia School of Medicine, we have developed a case-based introduction to laboratory medicine curriculum for first- and second-year medical students and are using the SOAP-V (Subjective, Objective, Assessment, Plan, Value) framework as an assessment tool for Entrustable Professional Activity 3 (EPA3) on clerkship. The High Value Practice Academic Alliance (http://hvpaa.org) provides a number of resources to guide teaching in addition to the Choosing Wisely guidelines.
Formulary. Many institutions have developed and implemented a laboratory formulary committee to eliminate unnecessary laboratory testing. Laboratory formularies are typically modeled after the Pharmacy and Therapeutics committee (P&T committee), which is often tasked with determining whether a specific medication or treatment should be added to the pharmacy formulary. Similar to the P&T committee, laboratory formularies carefully review diagnostic laboratory tests offered and supportive literature to determine the appropriateness or inappropriateness in their respective patient populations.
The University of Rochester has demonstrated great success with their formulary, which relies on a tiered testing approach (see table below).
Tiered Test Features of the University of Rochester’s Laboratory Formulary
|Tier 1||Tier 2||Tier 3|
|Description||Common tests with clear and well-proven clinical utility||More specialized tests with narrow clinical indication||Unclear, controversial tests, poorly proven or with very limited clinical indications|
|Ordering Requirements||No restrictions||Restricted to subspecialties||Restricted to subspecialties; Laboratory Diagnostic Committee approval required|
|Examples||Cystic fibrosis 32 mutations; thyroid-stimulating hormone receptor antibody||1,25-Dihydroxy Vitamin D; fecal pancreatic elastase||Hepatitis C recombinant immunoblot assay; β-Thalassemia gene|
In the first year of implementation of this approach, the University of Rochester was able to reduce send-out test volumes with initial savings of approximately half a million dollars. Other institutions have adopted a similar tiered approach, including Vanderbilt University Medical Center, which has reported institutional savings of more than $1 million. The success of these two programs relied heavily on institutional buy-in as well as a widespread membership that includes representatives from inside and outside the laboratory.
The most effective interventions for reducing repetitive laboratory testing are multimodal, combining a number of the approaches described above in an effort to produce meaningful and sustainable advances. Targeting academic institutions for improvement interventions would seem to be the most effective and efficient means of effecting change, but all institutions should seek high-level buy-in to begin or enhance improvement initiatives specific to their own needs. We hope clinicians and laboratorians will find the above approaches useful in these endeavors.