Bias in healthcare is often discussed in terms of outcomes, access, or patient experiences. But one of the most overlooked places where bias quietly shapes care is in the language clinicians use. A newly published study in JAMA Network Open sheds light on how negative language appears in electronic health records and what that means for patients living with sickle cell disease.
This blog explores the findings, why they matter, and what they reveal about the intersection of race, pain, and opioid treatment in modern healthcare.
Sickle cell disease is a chronic condition that primarily affects Black patients in the United States and other populations of African descent globally. It is characterized by severe pain episodes, frequent hospital visits, and long-term complications. Because of these factors, patients often require opioid medications for pain management.
Unfortunately, these same characteristics can expose patients to stigma. Race, chronic pain, and opioid use are each independently associated with bias in healthcare. When combined, they may amplify negative perceptions among clinicians.
This study takes a closer look at how those perceptions show up in clinical documentation.
Researchers analyzed nearly 40,000 clinician notes from over 18,000 patients. Using natural language processing and machine learning, they identified instances of negative descriptors in medical records.
The focus was on seven specific terms:
Patients with sickle cell disease were compared with four groups:
Patients with sickle cell disease had significantly higher odds of being described using negative language compared to:
This suggests that the bias goes beyond race or pain alone.
Interestingly, the level of negative descriptors for sickle cell patients was similar to that of patients with opioid use disorder.
This is a critical finding. It suggests that stigma associated with opioid use may play a major role in how patients with sickle cell disease are perceived, even when they do not have an addiction.
Patients who had all three factors, sickle cell disease, chronic pain, and opioid use disorder, experienced the highest rates of negative language in their records.
This highlights the concept of intersectionality. When multiple stigmatized identities overlap, the impact is not simply additive, it is amplified.
The most frequently used descriptors for sickle cell patients were:
These terms often imply patient behavior is the problem, rather than exploring underlying reasons such as inadequate pain control, mistrust, or prior negative experiences.
Clinical notes are not just documentation. They influence:
A single negative descriptor can shape how future clinicians perceive a patient before even meeting them.
For example, labeling a patient as “noncompliant” may lead to less aggressive treatment, reduced empathy, or assumptions about behavior.
Pain is subjective and difficult to measure. Patients with sickle cell disease often experience severe pain episodes that require urgent treatment.
However, repeated visits and requests for pain relief can be misinterpreted as drug-seeking behavior, especially in environments already influenced by opioid-related stigma.
The opioid crisis has led to increased caution among clinicians. While caution is necessary, it can also result in bias.
This study suggests that opioid-related stigma may overshadow other factors, even surpassing race or chronic pain in influencing clinician language.
Although the study found opioid use to be a stronger driver of negative language, race still plays a significant role.
Black patients overall had more negative descriptors than non-Black patients, reinforcing existing evidence of racial disparities in healthcare.
One of the most concerning insights is how bias can become self-reinforcing.
Breaking this cycle requires awareness and intentional change.
Instead of labeling behavior, clinicians can describe context:
This approach promotes understanding rather than judgment.
Training programs should focus on recognizing and mitigating bias, especially in high-risk areas like pain management and opioid prescribing.
Standardized approaches to managing sickle cell pain can reduce variability and minimize subjective judgments.
Clinicians should consider how their words may influence future care. Documentation should aim to inform, not stigmatize.
While this study focuses on sickle cell disease, the findings likely extend to other groups:
It highlights a broader issue within healthcare systems, how language reflects and reinforces inequality.
Like all research, this study has limitations:
Despite these limitations, the large sample size and robust methodology provide valuable insights.
This study offers a powerful reminder that bias in healthcare is not always overt. Sometimes, it is embedded in everyday language.
For patients with sickle cell disease, the combination of race, chronic pain, and opioid treatment creates a perfect storm of stigma. The words used in medical records can either reinforce that stigma or help dismantle it.
Improving healthcare equity requires more than policy changes. It requires attention to the small details, including how we describe and understand patients.
Language matters. And in healthcare, it can shape lives.
Wesevich A, Vangelatos A, Sun M, et al. Negative Descriptors of Patients With Sickle Cell Disease in the Electronic Health Record. JAMA Network Open. Published April 13, 2026. doi:10.1001/jamanetworkopen.2026.6458
This blog is for informational and educational purposes only. It is not intended as medical advice, diagnosis, or treatment. Always seek the advice of qualified healthcare professionals regarding any medical condition or healthcare decisions. The views summarized here are based on a single research study and should be interpreted within the broader context of ongoing research and clinical practice.
Most Accurate Healthcare AI designed for everything from admin workflows to clinical decision support.