‘Something Is Wrong’: Don’t Dismiss Pregnant Patients

I remember the moment clearly. She was on the operating table for an urgent cesarean. She was alone. Her voice, steady but strained, cut through the drape: “Something doesn’t feel right.”

Her vital signs were within normal limits. Surgical bleeding was minimal. In many ways the room felt calm. Someone offered faint reassurance. Another instructed her to try to calm down. The conversation moved on.

But she said it again. Then she vomited.

In obstetrics, we are trained to recognize patterns, the expected progression of labor, the thresholds that trigger intervention, the vital signs that signal danger. We are taught to rely on data, protocols, and on what can be measured and documented. But there is another form of data we too often overlook: what patients tell us about their own bodies.

In obstetrics, patient-reported concern should function as a trigger for reassessment, no different from an abnormal vital sign. Yet, we have no formal mechanism to capture it, document it, or respond to it systematically.

For Black women, that omission can be deadly.

By now, every health professional should be well aware of the stark inequities in maternal outcomes in the U.S. Black women are significantly more likely to experience severe maternal morbidity and are several times more likely to die from pregnancy-related causes than their white counterparts. These disparities persist across income and education levels, suggesting that access alone does not explain them. Something else is happening within the walls of our hospitals, in the moment-to-moment decisions that shape care.

One of those moments is when a pregnant patient says, “Something is wrong,” and we do not fully hear her.

Dismissal rarely announces itself as bias. It is quiet. It shows up in decisions to wait rather than reassess, in assumptions that symptoms are exaggerated, in subtle reframing of concern as anxiety or noncompliance. These small acts accumulate. They delay escalation. They narrow our differential diagnoses. They create space for preventable complications to unfold.

We see the consequences: hemorrhage recognized too late, postpartum warning signs dismissed as routine recovery, hypertensive disorders minimized until they become emergencies. In each of these scenarios, there is often a point — sometimes early, sometimes fleeting — when the patient tried to tell us that something was not right.

And we did not listen.

As a Black ob/gyn, I navigate these moments from a complicated position. I am both a clinician within the system and a witness to its failures. I have felt the subtle shift in tone when a Black patient expresses pain or concern. I have seen how quickly credibility can be questioned, how easily urgency can be downgraded. I have also felt the weight of deciding when to intervene, when to push harder, speak louder, or reframe a concern in language that will be taken seriously by others in the room.

Listening in these moments is not passive. It is an active clinical skill. It requires us to treat patient-reported symptoms as meaningful data, even when they do not immediately align with our expectations. It requires humility, an acknowledgment that our training does not make us infallible, and recognition that patients often perceive changes in their bodies before those changes become measurable. And it requires awareness of the ways bias can shape whose voices we prioritize and whose we question.

If we are serious about reducing maternal morbidity and mortality, we need to be equally serious about changing how we listen.

First, we must recognize patient concern as a clinical indicator, not a distraction. When a patient says something is wrong, that statement should trigger reassessment, not reassurance alone. It should prompt us to pause, to ask more questions, and to re-evaluate what we might be missing.

Second, we need to build listening into our systems of care. Just as we have checklists for hemorrhage and protocols for hypertension, we can create structured moments for reassessment that explicitly center the patient’s perspective. Documenting patient concerns, revisiting them, and treating them as data points would move listening from intention to practice.

Third, we must train clinicians to recognize dismissal for what it is. Bias in medicine is often discussed in abstract terms, but it manifests in concrete behaviors: interrupting patients, minimizing symptoms, delaying action. Naming these behaviors and addressing them directly in training is essential.

None of these changes require new technology or complex interventions. They require a shift in how we define clinical data and clinical responsibility.

I still think about that patient — how easy it would have been to pause, to reassess, to take her words as seriously as we would a change in vital signs. How often that moment is missed, not because clinicians do not care, but because we have not been trained to hear certain voices with the urgency they deserve.

Until we treat patient concerns as clinical data to document, revisit, and act upon, we will continue to miss some of the earliest warning signs of maternal deterioration. And Black women will continue to speak, and too often, they will not be heard.

Debra Eluobaju, MD, MPH, is a fourth year ob/gyn resident at Johns Hopkins Hospital in Baltimore.