Revealing Hidden Patient Populations to Optimize Targeting Strategies

A pharmaceutical manufacturer used NorstellaLinQ to uncover hidden patient populations that traditional structured data could not identify. By combining open claims, structured EHR, and unstructured EHR data with advanced data curation, Norstella identified moderate-to-severe atopic dermatitis patients, revealed biologic adoption patterns across healthcare organizations, and delivered actionable insights that improved account targeting, disease-state education, and patient identification. Norstella:

Identified hidden patient populations using structured and unstructured real-world data
Stratified patients by disease severity to improve targeting precision
Uncovered biologic adoption trends to optimize account-level engagement and education

Key Results

Improved Account Targeting with Disease Severity Intelligence

Challenge

Across the U.S., many healthcare organizations (HCOs) are falling behind recommended treatment guidelines for atopic dermatitis (AD)—especially when it comes to adopting biologic therapies for moderate-to-severe patients. To support improved disease-state education and guide more effective account-level targeting, one manufacturer wanted to identify moderate-to-severe AD patients within a large and heterogeneous national population, and then assess biologic utilization at the HCO level.

But isolating the right patient cohort was far from straightforward. Unlike other conditions with clear diagnostic markers or severity indicators, AD lacks standardized, structured identifiers for severity. There is no ICD-10 code for “moderate-to-severe AD,” and physicians use a variety of clinical tests to evaluate severity. Even these results can vary from one HCP to another, with one HCP labeling a score “moderate-to-severe” and another labeling the same score “mild-to-moderate.” These assessments often live only within narrative clinical notes, making the target population essentially invisible in structured claims or EHR fields.

Solution

Norstella, powered by NorstellaLinQ, combined open claims, structured EHR, and unstructured EHR data to precisely identify and stratify AD patients by severity.

First, the team established an initial AD cohort using open claims and structured EHR. Then, Norstella applied advanced data curation to extract clinical assessment scores buried within unstructured EHR notes, capturing severity indicators that would otherwise be inaccessible. This enabled Norstella to classify patients as mild-to-moderate or moderate-to-severe based on actual clinical evaluations, not proxy markers or assumptions.

Patient-level findings were aggregated to HCOs, enabling an operational view of treatment patterns and guideline alignment. With clear visibility into severity, Norstella could then analyze biologic use specifically among the biologiceligible population, rather than across all AD patients.

Outcome

By stratifying patients by disease severity and linking those cohorts back to HCOs, Norstella delivered a comprehensive assessment of biologic use, treatment adoption, and guideline concordance across the manufacturer’s priority accounts.

The findings also surfaced unexpected, but highly valuable, insights. For example, some HCOs that the manufacturer expected to be far along in biologic adoption were actually lagging, while a handful of others demonstrated adoption rates close to double the observed average percentages. This helped the manufacturer learn from high-adoption organizations and apply those insights to slower-adopting accounts.

The manufacturer could then focus outreach and education on the right HCOs based on true moderate-to-severe patient volumes. It also increased the potential for improved patient outcomes by enabling more accurate identification of biologic-eligible patients to support pathways to more timely and appropriate treatment.

Ultimately, the manufacturer achieved a level of visibility that had previously been impossible, transforming unstructured EHR data into strategic intelligence that informed both account targeting and patient-centric decision-making.

Get Started

Get Closer to the Patient Than Ever Before

NorstellaLinQ links claims, EHR, lab, and electronic prescription data into a single patient view — decision-grade and AI-ready. See what it can do for your team.

Frequently asked questions

Everything You Need to Know

How can real-world data uncover hidden patient populations?

By combining open claims, structured EHR, and unstructured EHR data, real-world data can identify patients who are difficult to find using structured datasets alone. Advanced data curation captures clinical details from physician notes, providing a more complete view of disease severity and patient populations.

Why is unstructured EHR data important for identifying disease severity?

Many disease severity indicators are documented only in narrative clinical notes rather than structured fields. Extracting these assessments from unstructured EHR data enables more accurate patient stratification and identification of patients eligible for advanced therapies.

How does disease severity intelligence improve account targeting?

By stratifying patients according to actual clinical severity and linking those insights to healthcare organizations (HCOs), pharmaceutical companies can identify priority accounts, tailor disease-state education, and focus engagement where biologic adoption opportunities are greatest.

What challenge did Norstella solve in this case study?

The manufacturer needed to identify moderate-to-severe atopic dermatitis patients and evaluate biologic adoption across healthcare organizations, but the absence of standardized severity codes made these patients difficult to identify using traditional claims or EHR data alone.

What business outcomes did the project deliver?

Norstella uncovered hidden patient populations, revealed biologic adoption trends, improved account-level targeting, enabled more effective disease-state education, and helped identify biologic-eligible patients for more timely and appropriate treatment.