7 Data Signals Pharma Teams Can Use to Research a New Therapeutic Market

A new therapeutic market usually enters a pharma planning deck as a prevalence figure. That number belongs on a board slide, and it does very little to tell a launch team where to spend its first year.

Prevalence counts people with a condition. It cannot show which clinicians diagnose it, where those clinicians practice, which health systems control their referrals, or whether anyone nearby is running trials in the indication.

The gap widens as pipelines move toward smaller populations. In 2025, half of the FDA’s 46 novel drug approvals carried orphan drug designation, so a large share of new products now launch into markets where a few hundred specialists may manage most treated patients. In a market that thin, a national estimate can be correct and still send a field team to the wrong cities.

Pharma market research holds up better when the single estimate gives way to observable signals. Healthcare claims data shows diagnosis and procedure activity at the provider level.

Organizational data shows who controls access, while publication and clinical trial data shows who shapes treatment norms before a product arrives. None of these signals is sufficient alone, so the analytical work lies in reading them against each other.

Where Relevant HCPs Are Concentrated Across Specialties and Markets

The first three signals describe the provider side of a market before patient volume enters the analysis. Together they map the clinicians who treat the condition along with the geographic and organizational structures that determine how a commercial team can reach them.

1. Specialty Concentration Defines Who Actually Manages the Patient

Specialty is the first cut most teams make, and it is usually too coarse. A condition named on a drug label rarely maps onto one specialty.

Psoriatic disease sits with dermatology and rheumatology, and many milder cases never leave primary care. Advanced practice clinicians widen the picture further.

AANP reported in November 2025 that the US now has more than 461,000 licensed nurse practitioners, a workforce large enough to change the denominator in any market sizing that counts physicians alone.

The useful output here is a distribution. It shows the share of treating clinicians in each specialty and how concentrated activity is within each group. If a small tier of rheumatologists sees most diagnosed patients, the right field model looks very different from one built for volume spread across thousands of primary care practices.

HCP data for pharma becomes far more valuable once it is filtered by NPI taxonomy and then weighted by observed clinical activity, because specialty labels alone overstate how many clinicians matter.

2. Geographic Concentration of HCPs and Clinical Activity

Where clinicians practice and where patients live are two different maps, and therapeutic area analysis needs both. Cardiology offers a clear example.

An analysis of all 3,143 US counties found that 46.3% had no practicing cardiologist, and residents of those counties traveled an average of 87.1 miles round trip to reach one versus 16.3 miles elsewhere.

Specialist density and claims activity therefore cluster in metropolitan hubs that draw patients from wide catchment areas. A region with a heavy disease burden and few specialists usually sends its patients somewhere else, and the receiving practices carry volume far beyond their local population.

Mapping both layers separates true opportunity clusters from places where demand exists but care happens elsewhere. It also informs territory design, since a territory drawn around population counts will misallocate reps whenever patients travel for care.

3. Healthcare Organizations and Physician Affiliations

Affiliation data shows how many independent decisions a commercial team actually has to win, and that number has been shrinking.

The share of group-practice physicians affiliated with vertically integrated health systems rose from 56.9% to 61.9% between 2018 and 2022, while system-affiliated groups grew from 54 to 66 doctors on average.

In a heavily affiliated market, formulary decisions and clinical pathways are often set at the system level. Ten high-volume oncologists inside one integrated delivery network may amount to a single account decision.

Pharma market intelligence should therefore group HCPs by their organizational parent and measure how much of the treated population sits within the largest systems.

When five organizations hold most of the diagnosed patients, account-based planning and early market access work deserve priority, whereas fragmented markets reward broader field coverage.

Diagnosis and Claims Activity in the Therapeutic Area

Healthcare claims data records what clinicians bill for, which makes it the most direct evidence of clinical activity available at national scale. Diagnosis codes and procedure codes capture different aspects of care, and a credible pharmaceutical market analysis draws on both.

4. Diagnosis Volume Measures the Population Already in Care

ICD-10 diagnosis codes give claims-based research its granularity. The ICD-10-CM classification contains approximately 70,000 codes, which is why most analyses roll them into broader groupings such as CCSR categories.

The roll-up works well for trend reporting. For a specific indication it can blur the distinction a launch depends on, such as a disease subtype or a complication captured in a separate code. Working at the individual code level keeps the treated population aligned with the label.

Diagnosis volume also carries a built-in limit. Claims capture people who have been diagnosed and seen.

CDC estimates that 27.6% of US adults with diabetes are undiagnosed, about 11.0 million people. Diagnosis counts will never reflect that group, and the gap between prevalence and diagnosed volume is itself a market signal.

A wide gap points to a market where screening and disease awareness may matter as much as product positioning.

Payer mix belongs in this signal too. Medicare-only data skews toward patients over 65, so a condition concentrated in working-age adults or children will look smaller and more geographically distorted than it is.

All-payor claims spanning commercial and Medicaid coverage give a truer count for those indications, and the payer split itself previews the access work a launch will need.

Tracked over several years, diagnosis volume by provider also shows which practices are growing their share of the treated population. That shift is often the earliest sign of an emerging center of care.

5. Procedure Volume and Relevant Clinical Activity

Procedure codes matter to pharma more than many teams assume. CPT and HCPCS codes record drug administration and diagnostic testing, including the tests that determine whether a patient is eligible for a therapy.

Biomarker testing volume, for instance, shows which practices are already identifying the patients a targeted therapy needs.

For physician-administered products, procedure data also reveals the site of care. Fee-for-service Medicare spent about $54 billion on separately paid Part B drugs in 2023, with physician offices accounting for 57% of that spending and hospital outpatient departments for 39%.

That split shapes contracting and reimbursement support, and it varies sharply by therapeutic area and region.

Infusion markets centered on community oncology practices behave differently from those dominated by hospital outpatient departments, even when diagnosis volumes look identical.

Procedure activity is also a practical proxy for clinical readiness. Clinicians who already perform an adjacent procedure or administer a related biologic have the infrastructure and experience a new product will require.

Publications and Clinical Trial Activity

Claims describe current practice, while publications and trials point to where practice is heading. They also identify the clinicians whose views carry beyond their own patient panel, which matters most in the year before and after approval.

6. Publication Records Show Where Scientific Influence Sits

PubMed contains more than 40 million citations and abstracts of biomedical literature, so publication activity in almost any therapeutic area is measurable down to the individual author. Recent publications in the indication, linked through co-authorship, turn that record into an influence layer.

Topic specificity matters here. An author with two hundred papers across cardiology and three on the target indication carries less weight in that market than a younger investigator whose last five years of work sit squarely inside it.

This signal often surfaces a different set of names than claims data does. Some of the most cited authors in a field see relatively few patients, and some of the highest-volume treaters rarely publish.

Clinicians who appear in both datasets tend to be the most valuable early relationships, because they combine practical credibility with reach. Structured KOL identification depends on exactly this overlap.

7. Trial Site Activity Reveals Research Capacity Before Launch

Clinical trial data shows which institutions and investigators are already working in the indication, often years before a product reaches market. Registry records are reasonably complete because the publication system enforces them.

The ICMJE recommends that medical journals require trial registration in a public registry at or before first patient enrollment as a condition of publication, which ties trial records closely to the literature in signal six.

Trial activity is also highly concentrated. ASCO found that 70% of US counties lack a single active cancer treatment trial, and only 10% of oncologists practice in rural areas. The trial phase adds nuance.

Early-phase sites tend to sit in academic centers with dedicated research infrastructure, while late-phase studies reach into larger community networks, so the phase mix in an indication shows how far research activity has already spread beyond major institutions.

Investigator sites mark the centers where clinicians gain early hands-on experience with new mechanisms, and those centers frequently become the first adopters after approval. Competitor trials matter just as much.

A cluster of rival studies in one region shows where another sponsor is building investigator relationships ahead of its own launch.

Combining Multiple Data Signals to Build a Market View

Each signal answers one question well and misleads when used alone. Diagnosis volume without affiliation data overstates how many independent decisions a team must win, and publication counts without claims data surface researchers who treat few patients.

Combining them lifts therapeutic market analysis from a sizing exercise into a pharma market opportunity a commercial team can act on.

Layered Signals Separate Market Size From Market Accessibility

A practical sequence starts with diagnosis volume at the ICD-10 code level to define the treated population. Specialty and taxonomy filters then show which clinician groups manage it, and geographic overlays show where that activity clusters relative to where patients live.

Affiliation data converts individual HCPs into accounts. Every layer needs a shared key for this to work, because author names in PubMed and investigator names in trial registries rarely match the format in claims, and the NPI is the most stable anchor available for linking them.

Procedure activity adds site-of-care context, and publication and trial records assign each clinician an influence tier.

The output is a segmented, ranked market. It shows how large the treated population is and how much of it sits within reach of a realistic field footprint and account strategy. Cohort analysis adds the time dimension by comparing provider segments across periods, which shows whether the opportunity is consolidating or shifting between sites of care.

Alpha Sophia Puts the Seven Signals on One NPI-Anchored Record

Most of these signals live in separate public datasets with different identifiers, which is why assembling healthcare market data for one indication by hand can take weeks.

Alpha Sophia, a pharma commercial intelligence platform, provides them on a single provider record anchored to the NPI. Its all-payor claims data spans Medicare, Medicaid, and commercial payors across more than 4 million providers, with procedures captured as CPT and HCPCS codes and diagnoses available as individual ICD-10 codes and CCSR categories.

Specialty taxonomy, organizational affiliations, publications, and clinical trial activity sit on the same record, so a team can filter by diagnosis and immediately see affiliation and influence data for the resulting providers.

A team evaluating a new immunology indication, for example, can start from the relevant ICD-10 codes, narrow to rheumatologists and dermatologists with meaningful volume, and see which of them are affiliated with the same systems or already running trials in the area.

For the geographic signal, the Territory Manager supports heat map analysis and shows opportunity size alongside territory boundaries. KOL AI draws on publishing activity, clinical trial leadership, and citation networks to identify the clinicians shaping a field.

Teams that work inside their own AI tools can connect Claude, ChatGPT, or Cursor to Alpha Sophia over MCP, use the in-app Assistant, or run autonomous workflows that size a market and pass the result to other connected systems. Developers can pull the same reference data into internal tools through the Alpha Sophia API.

Conclusion

Prevalence confirms that a market exists, and the seven signals explain how it is organized, which is the information launch planning actually runs on. Specialty and geographic concentration define the clinicians and places that matter.

Affiliations convert those clinicians into accounts, claims activity measures the treated population and its sites of care, and publications and trials identify who will shape opinion once a product arrives.

The practical discipline is to keep every signal tied to the same provider identity, so a diagnosis count, an affiliation, and a trial site can be read as one picture of the same clinician.

Book a demo with Alpha Sophia to see these signals applied to a therapeutic area you are evaluating.

FAQs

What data should pharma companies use for market research?
Pharma companies get the most reliable picture by combining healthcare claims data with organizational, publication, and clinical trial data. Claims show diagnosis and procedure activity by provider, while the other sources show who controls access and who shapes treatment norms. Prevalence estimates remain useful context but cannot show where care actually happens.

What is therapeutic market analysis?
Therapeutic market analysis is the process of sizing and structuring the market for a specific disease area or indication. It covers how many patients are diagnosed and treated, which clinicians and organizations manage them, and how that activity is distributed geographically. The result guides launch sequencing, field deployment, and market access planning.

How can HCP data support pharma market research?
HCP data identifies the clinicians who treat a condition, along with their specialty, location, and affiliations. Linked to claims activity, it shows how concentrated the treated population is among a small group of providers. That concentration drives decisions about field force size and account strategy.

How can claims data help research a therapeutic market?
Claims data records the diagnoses and procedures clinicians bill for, which makes it a direct measure of clinical activity at the provider level. Diagnosis codes size the treated population, and procedure codes reveal testing, drug administration, and site of care. Tracked over time, claims also show which practices are gaining share.

Why are clinical trials useful for pharmaceutical market research?
Clinical trial records show which investigators and institutions are already working on an indication before a product launches. Those sites often become early adopters because their clinicians have direct experience with new mechanisms. Competitor trials also reveal where rival sponsors are building relationships.

How does Alpha Sophia help pharma teams research new markets?
Alpha Sophia provides all-payor claims data across more than 3.9 million providers on a single NPI-anchored record that also carries specialty, affiliations, publications, and clinical trial activity. Teams can filter by ICD-10 or CPT code, view results on territory heat maps, and identify KOLs with KOL AI. The data is accessible in the app, through connected AI tools over MCP, or through the API.