A clinical trial can struggle before the first site opens because its protocol is built around an inaccurate picture of the patient population.

Proposed eligibility criteria may exclude more patients than expected. Required treatment histories may not reflect current practice. Selected sites may appear to have access to eligible patients but lack the capacity or referral pathways to recruit them.

Real-world data allows study teams to test these assumptions while the protocol and operational strategy can still be changed. Claims data, electronic health records, registries, laboratory results, and other sources can help teams:

• Estimate the effect of eligibility criteria on the available population
• Compare protocol assumptions with current treatment patterns
• Identify where potentially eligible patients receive care
• Evaluate whether site and country strategies reflect the disease population
• Find recruitment and participation risks before startup

This article explains how to apply real-world data during protocol development and feasibility planning, where it provides the greatest value, and what it cannot reliably answer on its own.

What Is Real-World Data in Clinical Trials?

Real-world data is information generated through healthcare delivery and patients’ daily lives rather than collected specifically for a traditional clinical trial.

Common sources include:

• Electronic health records
• Medical and pharmacy claims
• Disease and product registries
• Laboratory and diagnostic data
• Patient-generated health data
• Data from wearable devices and digital health technologies

These sources can show how frequently a disease occurs, which treatments patients receive, how care pathways differ across locations, and how many people might meet proposed eligibility requirements.

No single source provides a complete account of a patient’s experience. Claims data may show that a service was provided but offer limited clinical detail. Electronic health records can contain richer medical information, although relevant data may be missing when patients receive care across different health systems.

Understanding those strengths and limitations is essential. The goal is to use the right data to answer a defined study-planning question, rather than assuming that more data will automatically produce a better decision.

How Can Real-World Data Improve Protocol Design?

Every protocol contains assumptions about the patients, treatment pathways, and healthcare settings involved in the study. Real-world data gives teams an opportunity to examine those assumptions before they become fixed requirements.

A study team may develop eligibility criteria based on scientific literature, previous trials, regulatory considerations, and clinical expertise. Each criterion may make sense individually, but the combined effect can create a much smaller patient population than expected.

Real-world data can help teams estimate how many patients remain potentially eligible as individual criteria are applied. It can also show which requirements reduce that population most sharply.

For example, teams can investigate:

• How many patients have the required diagnosis, disease stage, or biomarker
• How common the permitted and excluded comorbidities are
• Which previous treatments patients typically receive
• Whether the required treatment sequence reflects current practice
• How laboratory thresholds affect the potential population
• Whether certain criteria exclude some demographic groups at higher rates

This analysis does not determine which criteria belong in the protocol. Scientific, safety, ethical, and regulatory considerations still guide those decisions. It gives the study team a clearer view of the operational consequences attached to each choice.

Can Real-World Data Make Eligibility Criteria More Practical?

Eligibility criteria influence how many patients sites must screen, who can participate, and how long enrollment may take. Restrictive requirements can be necessary, particularly when they protect participants or define the population needed to answer the research question.

Other requirements may be based on assumptions that deserve closer examination.

Consider a study requiring patients to have received a specific sequence of previous treatments. Real-world data may show that the sequence is uncommon in certain countries or that clinical practice has changed since the study concept was developed.

A laboratory threshold may also appear reasonable until patient records indicate that it would exclude a large percentage of the intended population. Washout periods, age limits, comorbidity restrictions, and narrowly defined disease characteristics can produce similar effects.

Finding these issues before the protocol is finalized gives teams several options. Depending on the scientific and regulatory context, they may be able to:

• Reconsider criteria that add limited value
• Clarify requirements that sites may interpret inconsistently
• Refine country and site strategies
• Adjust recruitment projections
• Prepare sites for likely screening challenges
• Develop targeted referral or outreach plans

Even when criteria cannot be changed, understanding their likely impact leads to more realistic planning.

Does the Protocol Reflect Current Clinical Practice?

Clinical practice changes continuously. New therapies enter the market, diagnostic standards evolve, and patients move through increasingly complex care pathways. A protocol based on outdated assumptions may struggle even when the underlying scientific concept remains sound.

Real-world data can show which treatments patients currently receive, how long they remain on therapy, when treatment changes typically occur, and which tests and procedures are part of routine care.

This comparison can uncover important gaps between the protocol and the real world. A required procedure may not be routinely available outside major academic centers. A treatment sequence may be common in one country but unusual in another. A proposed visit schedule may require patients to attend far more appointments than they would during standard care.

These differences do not automatically mean the protocol should change. Some additional procedures and visits are necessary to conduct reliable research and protect participants. Identifying them early allows teams to determine which requirements are essential and what support patients and sites will need to complete them.

Can Real-World Data Predict Clinical Trial Recruitment?

Real-world data can improve estimates of patient availability, but it cannot predict recruitment on its own.

A patient represented in a dataset may meet the protocol’s clinical requirements without being realistically recruitable. The patient may no longer receive care at the same organization, may be participating in another study, or may be unable or unwilling to take on the demands of the trial.

Recruitment also depends on factors that are difficult to observe in structured healthcare data, including:

• Patient interest and trust
• Provider awareness of the trial
• Site staffing and capacity
• Competing studies
• Travel and transportation requirements
• Caregiver availability
• Financial and time demands
• Willingness to complete study procedures

A dataset might identify 2,000 potentially eligible patients in a region. That number does not reveal how many can be contacted, referred, screened, or enrolled.

Real-world data should therefore establish a more informed starting point for recruitment planning. Direct input from sites, patients, caregivers, and healthcare providers is still needed to understand whether the population can realistically be reached.

How Does Real-World Data Support Site Selection?

Site selection has traditionally relied on investigator experience, previous enrollment performance, internal relationships, and feasibility questionnaires. Real-world data can add an independent view of where relevant patients receive care.

This can help sponsors identify established research centers with strong patient access as well as community practices and regional networks that may be overlooked by familiar site lists. It can also reveal differences in patient characteristics and treatment patterns across locations.

Patient volume alone does not make a site suitable for a study. Teams must also consider:

• Current staffing and study capacity
• Investigator interest
• Competing trials
• Research infrastructure
• Required equipment and procedures
• Referral relationships
• Experience with the intended population

Historical data also requires context. A site that performed well in a previous study may now have limited capacity, different staff, or several competing trials.

The strongest feasibility assessments combine population evidence with current, study-specific site intelligence. Real-world data indicates where potentially eligible patients may receive care. Direct engagement establishes whether a site can reach those patients and execute the protocol successfully.

Can Real-World Data Support More Representative Trials?

Real-world data can help teams compare the proposed study population with the broader population affected by a condition.

Analyzing eligibility criteria across demographic groups may identify requirements that disproportionately exclude certain populations. Geographic analysis can show whether the proposed site strategy provides reasonable access to communities carrying a significant share of the disease burden.

This information can prompt practical questions early in planning:

• Are the sites located near the intended patient populations?
• Do eligibility criteria create avoidable barriers?
• Are language and communication needs reflected in the recruitment plan?
• Will transportation or digital requirements limit participation?
• Are community providers included in the referral strategy?
• Does the protocol create greater burden for some populations than others?

Real-world data has limitations in this area. Some populations may be underrepresented or incompletely described, and healthcare records can reflect existing disparities in access, diagnosis, and treatment.

Used carefully, however, the data can help teams identify where additional research and community input are needed. It provides an opportunity to consider participation barriers during study design rather than attempting to correct them after enrollment falls short.

What Can Real-World Data Not Tell Study Teams?

Real-world data can make protocol and feasibility decisions more informed, but the precision of an analysis should not be confused with certainty.

Different datasets capture different portions of the patient journey. Important information may be missing, outdated, inconsistently recorded, or difficult to compare. A large dataset can still provide a misleading picture if it does not represent the countries, care settings, or populations relevant to the study.

Before applying an analysis, teams should ask:

• What population does the dataset represent?
• Which relevant information is missing?
• How current is the data?
• Were the variables collected consistently?
• Does the source reflect the intended countries and care settings?
• Can the data answer the specific question being asked?
• What additional context is needed from patients, sites, or clinicians?

Real-world data also cannot fully explain human behavior. It may show that patients discontinued a treatment or missed appointments without revealing why. It cannot reliably measure trust, motivation, caregiver support, current site capacity, or willingness to participate in a particular trial.

Those gaps are where human insight becomes most valuable.

How Should Study Teams Apply Real-World Data?

Real-world data creates the greatest value when it is connected to a specific decision. A practical process can keep the analysis focused.

1. Define the assumption

Identify what the study team currently believes about the patient population, treatment pathway, or research setting.

2. Ask a decision-ready question

Determine what evidence would confirm, challenge, or refine that assumption. “How many patients are there?” is often less useful than asking how a particular eligibility criterion affects the potential population in each proposed country.

3. Select an appropriate data source

Evaluate whether the data represents the intended population and contains the variables needed to answer the question.

4. Compare alternative scenarios

Model different criteria, countries, sites, or protocol options instead of relying on a single estimate. Comparing scenarios helps teams understand the tradeoffs associated with each decision.

5. Add human context

Review the findings with patients, caregivers, sites, clinicians, and operational specialists who understand what the data may not capture.

6. Document the decision

Record how the evidence influenced the protocol, feasibility strategy, or operational plan. This creates a stronger foundation for future study planning and prevents useful lessons from being lost.

When Should Real-World Data Be Used?

The greatest opportunity often comes before the protocol is finalized, when study teams can still evaluate eligibility criteria, procedures, geographic assumptions, and participation requirements.

Real-world data can continue to provide value throughout development:

• Concept development: Understand the disease population and current treatment landscape.
• Protocol development: Test eligibility criteria, procedures, and treatment assumptions.
• Feasibility planning: Estimate patient populations and compare potential countries.
• Site selection: Identify where relevant patients receive care.
• Recruitment planning: Focus outreach and referral strategies.
• Study execution: Compare actual enrollment and screening patterns with initial projections.
• Post-study review: Apply operational lessons to future protocols.

Using evidence across this lifecycle also creates a learning loop. Actual study performance can improve the assumptions used to design the next trial.

Better Studies Begin With Better Assumptions

Real-world data cannot eliminate uncertainty from clinical research. It can reveal where a protocol or operational strategy depends on assumptions that should be examined more closely.

By testing eligibility criteria, comparing protocol requirements with current practice, identifying potential patient populations, and examining geographic access, teams can find important risks earlier. They can then focus patient, site, and clinician engagement on the questions that data alone cannot answer.

This creates a stronger starting point for clinical trial design. Scientific objectives remain at the center, supported by a more realistic understanding of the patients, care pathways, and research settings involved in delivering the study.

Continue the Conversation at SCOPE Summit Europe

Real-world data, clinical trial feasibility, protocol optimization, and patient-centered study design will be important areas of discussion at SCOPE Summit Europe. Leaders from across clinical research will explore how better evidence and cross-functional insight can support stronger decisions throughout study planning and execution.

Learn more and register here.