Working toward fairness in peripartum drug screening and beyond

Lab professionals bring a deep expertise in quality assurance throughout the entire testing cycle, from the preanalytic to postanalytic phases. They also have a nuanced understanding of how well-intentioned testing can sometimes cause unintended harm. By leveraging these strengths, laboratorians play a powerful role in advancing health equity.

Peripartum drug screening is a case in point. Drug use occurs in approximately 5% of pregnancies (1), and the American College of Obstetricians and Gynecologists (ACOG) recommends universal verbal screening throughout pregnancy. In this approach, clinicians engage in structured, nonjudgmental conversations with pregnant individuals about their substance use, using validated tools such as the 4P’s Plus or the NIDA Quick Screen. The goal is to foster open dialogue, provide anticipatory guidance, and facilitate timely referrals for treatment when needed (2).

Despite ACOG’s emphasis on verbal screening, laboratory-based urine drug screening (UDS) is frequently performed in the peripartum setting, particularly when a pregnant person presents with signs of substance use or historical risk factors (3). While the affect of UDS is to guide clinical management, the screening also introduces complexities that can impact equity. In this article, we describe a data-driven intervention that eliminated the UDS indications most associated with low-value and inequitable testing.

Cannabis use and pregnancy

Because most standard drug panels include cannabis, pregnant people who have used this substance are subject to routine testing, even when the clinical significance of peripartum cannabis use is uncertain (4). Historically, at our institution — Washington University School of Medicine/Barnes-Jewish Hospital in St. Louis — a “history of isolated cannabis use” alone was considered sufficient justification for peripartum UDS.

Although observational studies have suggested an association between cannabis use during pregnancy and adverse fetal or neurodevelopmental outcomes, no definitive causal link has been established (5). ACOG discourages cannabis use during pregnancy but maintains that individuals should not face criminal or civil penalties for its use (4). Additionally, in the peripartum setting, a positive test for isolated cannabis use typically does not change clinical management, which raises questions about the necessity and utility of such testing (4).

The issue is further complicated in mandatory reporting states such as Missouri, where any positive UDS result, including one for isolated cannabis use, must be reported to Child Protective Services (CPS) (6). In this context, UDS can alienate patients, erode trust in healthcare providers, reduce vital rooming-in time between mothers and babies, and expose families to legal consequences (7).

Lauren Nacke, a former social worker at Barnes-Jewish Hospital, describes this impact firsthand:

“Collecting a UDS felt like a ‘gotcha’ moment on labor and delivery. While helping this first-time mom to the bathroom just hours after childbirth, the nurse informed her that her urine would be collected due to her history of marijuana use. This felt like a failure to me. Moments after birthing her first child, we inserted panic and fear with an accusation of child abuse. Without any knowledge of our urine drug screen policy or the hotline process, this young mother could assume the worst. Would her child be removed? Would the police be involved? We caused harm at a vulnerable and memorable time.”

Impact on equity

Racial minority patient groups disproportionately bear the burden of this harm. A multidisciplinary study at at the Washington University School of Medicine that included providers from obstetrics and gynecology, pediatrics, neonatology, nursing, and social work found that Black mothers presenting to the labor and delivery unit at Barnes-Jewish Hospital were subjected to peripartum UDS at more than twice the rate of White mothers.

As a result, 1 in 4 Black mothers underwent UDS, and about 1 in 10 were reported to CPS. Most referrals were for isolated cannabis positivity — a result with low clinical utility. These findings highlight a stark inequity: Black pregnant patients are more likely to experience the harms of peripartum UDS without receiving any meaningful clinical benefit (8).

A data-driven intervention

The research team collaborated with the pathology informatics team in the department of laboratory and genomic medicine at their institution to address this disparity through a data-driven intervention (9). By integrating education, policy changes, and electronic clinical decision support (CDS), the multidisciplinary team eliminated the UDS indications most associated with low-value and inequitable testing — specifically, “history of cannabis use” and “limited prenatal care.” Following the intervention, the pathology informatics team analyzed UDS and CPS reporting practices, comparing the 16 months before with the 16 months after implementation. They found that screening rates for Black mothers dropped five-fold, CPS reporting rates were cut in half, and the disparity in screening between Black and White mothers was eliminated.

Crucially, the rates of detection of illicit drug use did not change following the intervention, suggesting that it did not compromise sensitivity for clinically actionable UDS results.

UDS and CPS reporting rates for White mothers also declined following the intervention. This result challenges the notion that health equity is a zero-sum game that inherently disadvantages one group to benefit another. Instead, it illustrates that equitable policies can improve care for all patients.

Why labs are equity leaders

These studies reveal an important insight: While laboratories are often viewed as behind-the-scenes entities, they are well positioned to advance health equity by improving fairness in laboratory testing. In other words, lab professionals help ensure that all patients can reasonably expect to realize the benefits of testing without having to shoulder a disproportionate share of the harms. As we describe below, several key characteristics make laboratories ideally suited for this role.

Labs oversee the entire testing cycle, from the preanalytic to postanalytic phases (10). Within the UDS project, the lab team’s knowledge about testing and computerized order entry helped them identify changes to the workflow that could facilitate mindful UDS ordering and auditing.

Lab professionals have an opportunity to build on these successes. By expanding existing laboratory quality-assurance frameworks, they can identify and address unfairness in laboratory testing that could arise at any point. For example, in the preanalytics phase, they might explore ways to expand patient access to testing and improve provider ordering behaviors.

Labs have made progress with fairness metrics. A first step in this direction is operationalizing fairness metrics, such as those described in the field of algorithmic fairness (11), for reporting on laboratory quality. For example, demographic parity, equalized odds, and predictive parity could be applied to laboratory data to screen for unfair laboratory practices.

Implementing and scaling this approach will require new tools to complement existing laboratory quality analytics. This was the topic of FairLabs, the 2024 ADLM Data Analytics Challenge (12). The success of the FairLabs competition underscores that members of the laboratory medicine community are ready, willing, and able to fill this gap in our analytical toolbox.

Lab professionals also should broaden the use of their valuable laboratory and clinical data streams, which provide insights into patterns of test utilization (13), disparities in access to diagnostic services (14), and variations in results across patient populations (15). The data can be used to uncover systemic barriers to care and inform interventions targeted to patient demographic and socioeconomic characteristics.

In the UDS project, laboratory informatics experts were central to identifying problematic order indications, implementing CDS to circumvent the usage of those indications, and monitoring the effectiveness of the intervention.

Labs are well integrated across healthcare. Finally, laboratories are deeply integrated across service lines, enabling them to influence equitable care across multiple disciplines. This positioning allows them to collaborate with providers, administrators, and policymakers to ensure diagnostic testing is accessible, appropriate, and effectively utilized for all patient populations. Kim Hamlin, MD, a pediatric hospitalist at St. Louis Children’s Hospital, describes the effect that laboratory stewardship had on obstetric and pediatric clinicians:

“The policy of UDS for isolated cannabis use was one that brought significant dissatisfaction for clinicians. They knew the results offered little benefit for the patient or the baby, but held potential for great harm, and felt powerless that they were perpetuating an inequitable system. Changing the policy was met with significant clinician support as this barrier to relationship and trust-building was removed for a large proportion of patients and has allowed clinicians to focus their resources and support to target patients in need.”

Changing clinical practice patterns is difficult, and we attribute the success of this intervention to the multifaceted approach that leveraged interdisciplinary expertise and collaboration, education, policy change, and electronic CDS. Although we cannot disentangle these components’ contributions to the study’s outcomes, we can attest that combining policy change and CDS helped to create and reinforce accountability for guideline-concordant care. For healthcare institutions with limited resources to replicate the entire intervention, replicating the CDS component may be a practical starting point.

Equity matters

There are many reasons for why achieving health equity should be a priority for all healthcare professionals. First, it is a moral and ethical imperative. Indeed, the core principles of medical ethics, such as nonmaleficence and justice, require medical professionals to avoid harming vulnerable patient subgroups. Second, achieving health equity can effectively improve patient outcomes by alleviating the detrimental effects of health inequity. Third, addressing health inequity reduces healthcare costs by preventing avoidable testing and complications due to poorly managed chronic diseases.

Of course, pursuing health equity can directly benefit laboratorians. By providing empathetic care that considers patients' lived experiences, clinical lab professionals can foster a stronger and more meaningful connection with those we serve. To get there, we must work together to elevate fairness to the same plane as other core quality domains in laboratory testing.

Vahid Azimi, MD, is laboratory director at Alameda Health System in Alameda County, California. +Email: [email protected]

Mark A. Zaydman, MD, PhD, is an assistant professor of pathology and immunology at Washington University School of Medicine in St. Louis. +Email: [email protected]

References

1. Polak K, Kelpin S, Terplan M. Screening for substance use in pregnancy and the newborn. Semin Fetal Neonatal Med 2019; doi: 10.1016/j.siny.2019.01.007.
2. Sanjanwala AR, Lim G, Krans EE. Opioids and opioid use disorder in pregnancy. Obstet Gynecol Clin North Am 2023; doi: 10.1016/j.ogc.2022.10.015.
3. Price HR, Collier AC, Wright TE. Screening pregnant women and their neonates for illicit drug use: Consideration of the integrated technical, medical, ethical, legal, and social issues. 2018; doi: 10.3389/fphar.2018.00961.
4. Marijuana use during pregnancy and lactation. Committee Opinion No. 722. Obstet Gynecol 2017; doi: 10.1097/AOG.0000000000002354.
5. Badowski S, Smith G. Cannabis use during pregnancy and postpartum. Can Fam Physician 2020;66:98-103.
6. Prenatal Drug Exposure: CAPTA Reporting Requirements for Medical Professionals. https://ifwhenhow.org/wp-content/uploads/2024/02/2024-Prenatal-Drug-Exposure-CAPTA-Reporting-Requirements-for-Medical-Professionals.pdf (Accessed August 2024).
7. Siegel MR, Cohen SJ, Koenigs K, et al. Assessing the clinical utility of toxicology testing in the peripartum period. Am J Obstet Gynecol MFM 2023; doi: 10.1016/j.ajogmf.2023.100963.
8. Rubin A, Zhong L, Nacke L, et al. Urine drug screening for isolated marijuana use in labor and delivery units. Obstet Gynecol 2022; doi: 10.1097/AOG.0000000000004930.
9. Azimi V, Trammel C, Nacke L, et al. Racial equity in urine drug screening policies in labor and delivery. JAMA Netw Open 2025; doi:10.1001/jamanetworkopen.2025.0908.
10. Wians FH. Clinical laboratory tests: Which, why, and what do the results mean? Laboratory Medicine 2009; doi.org/10.1309/LM4O4L0HHUTWWUDD.
11. Azimi V, Zaydman MA. Optimizing equity: Working towards fair machine learning algorithms in laboratory medicine. J Appl Lab Med 2023; doi: 10.1093/jalm/jfac085. PMID: 36610413.
12. Merrill A. All is not fair in health and data: Understanding and ensuring algorithmic fairness. CLN Daily 2024; https://myadlm.org/cln/cln-daily/2024/all-is-not-fair-in-health-and-data-understanding-and-ensuring-algorithmic-fairness
13. Baird G. The laboratory test utilization management toolbox. Biochemia Medica 2014; doi: 10.11613/BM.2014.025.
14. Khan A, Culbreath K, Goldstein Z, et al. Access barriers in testing for sexually transmitted infections across gender and sexual identities. Clinics in Laboratory Medicine 2024; doi: 10.1016/j.cll.2024.07.003.
15. Azimi V, Fiala MA, Zaydman MA. Use of the current standard of practice serum free light chains (sFLC) reference interval puts Black patients at 5 times higher odds for light chain monoclonal gammopathy of undetermined significance (LC-MGUS). Clin Chem 2023; doi: 10.1093/clinchem/hvad083.