Allergy/immunology fellow at University of California San Francisco

Chang Su, MD

The Case for Delabeling NNAALs in Obstetric Care, With Chang Su, MD

Non-narcotic analgesics allergy labels (NNAAL) significantly elevated rates of eclampsia (aRR, 1.5; 95% CI, 1.06–2.12), preterm birth (aRR, 1.21; 95% CI, 1.14–1.28), NICU admission (aRR, 1.17; 95% CI, 1.10–1.25), and neonatal withdrawal syndrome (aRR, 1.51; 95% CI, 1.24–1.84) among labeled mothers across 2,244,210 California singleton livebirths — point toward a clear intervention target: the family planning stage, before pregnancy makes formal drug challenge testing contraindicated.

American Academy of Allergy, Asthma & Immunology (AAAAI)

 2026 Annual Meeting held in Philadelphia, Pennsylvania, from February 27 to March 2, Chang Su, MD, allergy/immunology fellow at University of California San Francisco, outlined why timing matters enormously for NNAAL de-labeling efforts. Evaluating a prior ibuprofen reaction, for example, ideally requires a provocation challenge — which is contraindicated during pregnancy. Waiting until a patient is already pregnant to initiate drug allergy evaluation therefore forecloses the most definitive diagnostic pathway. Su's preferred model is pre-conception referral, integrated into routine family planning screening by OB-GYN and cardiology colleagues, so that allergy status is clarified before it becomes clinically urgent. She described active efforts to work NSAID allergy screening into standard documentation templates and electronic health record workflows used by APPs and physicians in OB practices at her institution — a structural intervention designed to normalize proactive referral rather than relying on reactive, crisis-driven evaluation.

Su also used the final portion of the conversation to reframe how the field should think about drug allergy labels more broadly. NNAALs affect approximately 2% of the population — rarer than beta-lactam labels, which appear in roughly 10% of charts, and therefore harder to study at scale. The study does not overturn existing paradigms around allergy labeling, Su acknowledged, but it extends the known harms of inappropriate labels from beta-lactams into the NSAID class, and makes it easier to extrapolate that other inaccurate allergy labels likely carry similar outcome consequences. Her broader call to action was cultural as much as clinical: rather than entering a blank medication allergy into the chart, clinicians should document the specific reaction type, severity, and any confirmed tolerability thresholds — information that allows downstream providers to make nuanced decisions rather than defaulting to broad avoidance. As Su put it, allergy labels are not a single entity, and the field is increasingly recognizing that indiscriminate labeling carries real repercussions, particularly when it walls patients off from first-line therapies tied to the best outcomes.

Su has no relevant disclosures to report.

1. Su C, Baer R, Otani I, Schatz M, Chambers C. The relationship between maternal non-narcotic analgesics allergy labels and maternal and fetal outcomes: results from a large, administrative cohort. Presented at: AAAAI 2026 Annual Meeting, February 27-March 2, Philadelphia, Pennsylvania. Abstract #L24

2. Non-Narcotic Analgesics Allergy Labels Linked to Adverse Maternal and Fetal Outcomes. News release. AAAAI. February 26, 2026. 

https://www.aaaai.org/about/news/news/2026/narcotic

Videos

Su emphasized how care should be shifted to reflect known impacts of NNAALs on obstetric care.

Why De-Labeling NNAALs Should Be Routine Obstetrics Care, With Chang Su, MD

With maternal non-narcotic analgesic allergy labels significantly associated with eclampsia (aRR, 1.5; 95% CI, 1.06–2.12), preterm birth (aRR, 1.21; 95% CI, 1.14–1.28), NICU admission (aRR, 1.17; 95% CI, 1.10–1.25), and neonatal withdrawal syndrome (aRR, 1.51; 95% CI, 1.24–1.84) across a California birth cohort of 2,244,210 singleton livebirths, the case for integrating allergy de-labeling into routine obstetric care is compelling.

 2026 Annual Meeting held in Philadelphia, Pennsylvania, from February 27 to March 2, with Chang Su, MD,  allergy/immunology fellow at University of California San Francisco, she situated the NNAAL findings within a broader de-labeling effort already underway in obstetrics. Her group has had significant success with beta-lactam allergy de-labeling in collaboration with OB and primary care colleagues — driven in part by prior data showing that penicillin allergy labels, like NNAALs, divert patients away from first-line therapy, in that case for Group B streptococcal infection around the time of delivery. Approximately 90% of patients with a beta-lactam allergy label do not have a true allergy on formal evaluation, with emerging data suggesting the true negative rate may be even higher. A similar dynamic appears to apply to NNAALs: one prior study found that approximately 85% of patients carrying such a label can ultimately tolerate NSAIDs, suggesting a substantial proportion of the 10,460 NNAAL-carrying mothers in the study may have been unnecessarily restricted from first-line analgesics and aspirin-based preeclampsia prevention.

Critically, Su emphasized that de-labeling is not an all-or-nothing proposition. For patients who cannot be fully cleared — due to the nature of their prior reaction — a formal allergy evaluation can still produce a concrete, individualized action plan specifying dose thresholds and tolerable agents. In the context of preeclampsia prevention, for example, a patient who cannot tolerate the recommended 162 mg aspirin dose may still safely receive 81 mg, a clinically meaningful alternative that would otherwise go unprescribed without specialist input. Su described ongoing efforts to expand this model beyond beta-lactams to other drug classes — including NNAALs — and stressed that the referral pipeline from OB-GYN colleagues is central to making de-labeling scalable as a component of routine prenatal care.

“For example, for preeclampsia prevention, aspirin 162 milligrams is often recommended. But for certain patients, perhaps they can tolerate 81 milligrams based on the types of reaction that they had. And if we can prove that, then perhaps we can't get them the most recommended therapy, but we can get the second best of them,” Su said.

Su discussed the role of delabeling non-narcotic analgesic allergy labels in obstetrics care.

Non-Narcotic Analgesic Allergy Labels in Pregnancy Drive Suboptimal Care and Adverse Outcomes, With Chang Su, MD

Unverified drug allergy labels are a well-documented driver of suboptimal prescribing — but a large population-based study presented at the 

 2026 Annual Meeting held in Philadelphia, Pennsylvania, from February 27 to March 2, now extends that harm into the perinatal setting, showing that mothers carrying non-narcotic analgesic allergy labels face significantly elevated risks of eclampsia, preterm birth, and adverse neonatal outcomes compared to those without such labels.

To discuss the findings, we spoke with Chang Su, MD, allergy/immunology fellow at University of California San Francisco. The study drew on the Study of Outcomes in Mothers and Infants, a population-based cohort of all singleton livebirths in California between 2016 and 2021 (N = 2,244,210), of which 10,460 were born to mothers carrying non-narcotic analgesic allergy labels (NNAALs). After adjustment for maternal characteristics, NNAALs were significantly associated with increased rates of eclampsia (aRR, 1.5, 95% CI, 1.06-2.12), preterm birth (aRR, 1.21, 95% CI, 1.14-1.28), NICU admission (aRR, 1.17, 95% CI, 1.10-1.25), and neonatal withdrawal syndrome (aRR, 1.51, 95% CI, 1.24-1.84), as well as longer mean infant hospital stays (

 <.0001). A decreased rate of large-for-gestational-age infants was also observed (aRR, 0.92, 95% CI, 0.87-0.98). No significant associations were found with maternal preeclampsia, small-for-gestational-age infants, major structural birth defects, or 5-minute APGAR score below 7.

Su highlighted 2 primary mechanisms through which NNAALs appear to be shaping obstetric management — and outcomes — in ways that diverge from clinical guidelines. First, aspirin is recommended for preeclampsia prevention in up to 85% of pregnant patients with relevant risk factors, yet patients carrying an NSAID or aspirin allergy label are substantially less likely to receive it, even when indicated. The significantly elevated eclampsia risk in the NNAAL cohort — a condition that prophylactic aspirin is specifically intended to prevent — is consistent with that gap in guideline-concordant care, though Su was careful to note the association cannot establish causality with certainty. Second, non-narcotic analgesics are the recommended first-line treatment for postpartum pain, and prior studies have shown that patients with NNAALs receive opioids at higher rates and higher doses postpartum — a prescribing pattern with downstream implications in the context of the ongoing opioid epidemic. Together, Su argued, these findings make a strong case that proactive allergy de-labeling should be considered as part of routine prenatal care, not reserved for specialist referral after an adverse event has already occurred.

“People with NNAALs have a higher risk of eclampsia, whether that's because they're not started on aspirin when they need to or due to other reasons… but it does seem like when patients are less likely to receive the recommended preventative care, that that leads to the adverse outcome that it's aiming to prevent,” Su said.

Su described how NNAALs drive management decisions in pregnancy and post partum care.

Chang Su, MD

Articles