Using AI to Reduce Racial and Ethnic Disparities in AATD Diagnosis
This scientific poster presents the development and validation of a machine-learning model that identifies likely undiagnosed patients with Alpha-1 Antitrypsin Deficiency (AATD) from large-scale US claims data. The analysis reveals major disparities in diagnosis rates across racial and ethnic groups, especially for Black, Hispanic, and Asian or Pacific Islander populations.
Key findings include:
Underdiagnosis of AATD in non-White populations compared to US demographic benchmarks
Earlier detection of AATD among flagged patients vs. diagnosed ones (median age difference >10 years)
AI model performance with 0.90 precision in predicting likely AATD cases
High prevalence of asthma and respiratory comorbidities in likely undiagnosed patients
Accelerate early identification
Our model flags likely undiagnosed AATD patients up to 10 years earlier than traditional methods. This enables earlier cohort selection and reduces bias in trial populations.
- Enhance cohort diversity and reduce screening failures
- Improve recruitment timelines and eligibility accuracy
- Identify real-world gaps in diagnostic coverage
Support equitable communication
The model reveals gaps in diagnostic access across race and ethnicity, empowering your team with validated, stratified insights to shape impactful narratives.
- Highlight disparities with real-world AATD data
- Back education campaigns with clinical evidence
- Promote access equity across underdiagnosed populations
Bridge market and access gaps
By identifying patients otherwise overlooked, your commercial and strategy teams can build more inclusive and effective campaigns.
- Unlock hidden segments within diverse populations
- Drive precision targeting for under-addressed needs
- Quantify missed opportunities using prediction models
Why It Matters
Underdiagnosed patients from underrepresented groups not only delay trial timeline they skew scientific assumptions, treatment patterns, and medical strategy.
By identifying these patients earlier, pharma teams can:
Enhance cohort diversity for clinical development
Strengthen evidence used in medical affairs communication
Refine disease understanding for future pipeline opportunities
Improve Clinical Precision with Better Patient Identification
Current diagnostic models often overlook underrepresented patients especially in rare diseases like AATD. This gap skews patient cohorts, delays diagnosis, and risks compromising trial validity.
VOLV helps pharmaceutical teams surface high-risk, likely undiagnosed patients through advanced AI analysis of real-world data.
For Clinical Development: Increase cohort diversity and accelerate trial recruitment
For Medical Affairs: Improve scientific storytelling with stronger epidemiological insights
For Strategic Pharma Leaders: Build smarter pipelines by uncovering overlooked populations
What the Model Uncovered
16,431 likely undiagnosed AATD patients identified
48% of flagged patients were non-White
Median age at flagging: 10 years younger than diagnosed patients
Significant underrepresentation of Black, Hispanic, and Asian populations in current diagnosed cohorts
Higher asthma prevalence in undiagnosed flagged patients
+42% earlier identification compared to traditional approaches
Better diagnosis starts with better data
Learn how AI-powered patient identification can reveal gaps in current diagnostic pipelines — especially for underdiagnosed populations across racial and ethnic groups.
Why it matters for your team:
Uncovers diagnostic disparities that could skew trial populations
Improves representation of real-world patients, including underserved groups
Informs evidence generation and medical strategy
Provides actionable insights for more equitable trial design