Frequent Mistakes to Avoid When Designing Feasibility Surveys

Why Survey Design Matters in Clinical Feasibility

Feasibility surveys are the first checkpoint in clinical trial execution. Sponsors, CROs, and clinical teams rely on these tools to determine which investigator sites are capable of enrolling, complying, and delivering high-quality data. However, flawed survey design can compromise this entire process—leading to site underperformance, protocol deviations, missed enrollment targets, and costly delays. Regulatory authorities like the FDA and EMA have highlighted the importance of accurate feasibility assessments in multiple inspection reports.

Designing an effective feasibility questionnaire is not just about gathering information—it’s about ensuring the **quality, clarity, and relevance** of the data collected. Poor design choices can introduce bias, reduce response rates, and provide misleading inputs, ultimately affecting trial success.

This article explores common pitfalls in feasibility survey design and provides corrective strategies to improve accuracy, efficiency, and regulatory alignment.

1. Over-Reliance on Generic Questionnaires

One of the most frequent mistakes is using generic, one-size-fits-all surveys across all therapeutic areas and trial phases. This ignores the unique requirements of different indications. For example, asking only “Do you have imaging capabilities?” in an oncology trial overlooks critical aspects like:

• Type of imaging (CT, MRI, PET)
• RECIST or iRECIST measurement familiarity
• Archiving and transfer compliance (DICOM format)

Similarly, a vaccine trial might need cold chain logistics and mass-screening capacity questions, which may be completely irrelevant to a rare disease gene therapy study. Lack of customization can cause misaligned expectations and downstream failures.

2. Ambiguous or Leading Questions

Vague phrasing leads to inconsistent interpretation and invalid data. For instance, asking “Can you enroll patients quickly?” is subjective. What qualifies as “quick” for one site may differ from another. Instead, a better version would be:

“How many patients fitting the protocol inclusion criteria did your site enroll in the last 12 months for similar Phase II studies?”

Leading questions also bias the respondent. “You have successfully conducted previous trials, correct?” might trigger social desirability bias. Neutral, fact-based phrasing is key.

3. Excessive Length and Complexity

Lengthy surveys with poor flow reduce completion rates and frustrate site staff. In multi-center trials, sites often have limited staff availability, especially during active study periods. Surveys that take over 45 minutes are less likely to be completed accurately. Issues include:

• Redundant questions across sections
• Poor section organization (e.g., mixing regulatory and infrastructure questions)
• Lack of autosave or ability to pause and resume digital forms

As a best practice, limit questionnaires to 25–30 well-structured questions for initial feasibility, followed by site-specific deep dives as needed. Use digital platforms that allow intuitive navigation and validation.

4. Lack of Data Validation or Documentation Fields

Another flaw is the absence of cross-checks or request for supporting documentation. For example, if a site claims it can enroll 100 patients over 6 months, sponsors should request either:

• Patient registry screenshots
• De-identified electronic health record reports
• Recruitment logs from similar studies

Without these, responses are based solely on memory or estimates, increasing risk of over-promising. Platforms should include fields for file uploads, comment boxes for clarification, and warning prompts for unusual entries.

5. Ignoring Historical Site Performance Data

Failing to consider a site’s previous trial history is a major oversight. Historical data helps contextualize feasibility answers and filter out consistently underperforming sites. For example:

Integrating such data-driven benchmarking within the survey design significantly improves reliability and transparency.

Transition to Solutions and Best Practices

Now that we’ve identified major pitfalls in feasibility survey design, the next part will offer regulatory-aligned solutions, practical templates, and technology integrations to improve the quality of your feasibility assessments.

6. Neglecting to Capture PI and Sub-Investigator Details

Many feasibility surveys focus primarily on site-level infrastructure while ignoring investigator qualifications. Yet, the PI’s past experience, availability, and regulatory track record are critical success factors. A well-designed survey should capture:

• Number of trials conducted in the last 5 years
• Therapeutic area alignment with the protocol
• GCP training validity and inspection history
• Ratio of PI to concurrent active trials

Neglecting to gather such details could lead to site activation delays due to regulatory rejection of PI credentials or unavailability.

7. Overlooking Regional and Regulatory Context

Global feasibility surveys often ignore country-specific regulations and operational limitations. For example:

• In India, the CDSCO has specific rules regarding ethics approvals and compensation
• In Japan, feasibility surveys must include PMDA-specific compliance sections
• In the EU, surveys must align with EU Clinical Trial Regulation (CTR) timelines and document requirements

Not including such country-specific sections can result in inaccurate site feasibility outcomes. For global trials, it’s critical to tailor questions by region or include branching logic that triggers local regulatory queries based on country selection.

8. No Mechanism to Capture Feasibility Risk Flags

A robust feasibility survey should include logic or scoring that auto-generates red flags based on site responses. For instance:

Response	Flag
PI is involved in 7 concurrent studies	Investigator overload
Site has no GCP training in last 3 years	Non-compliance risk
Startup timeline > 90 days	High activation risk

Such automated risk indicators help feasibility managers prioritize follow-ups and site visits.

9. Lack of Digital Integration and Data Traceability

Paper-based or email surveys still exist in some trials, resulting in data loss, miscommunication, and lack of audit trails. Regulatory inspectors, including from the FDA, expect version control, date/time stamps, and investigator signatures on feasibility forms.

Surveys should be integrated into platforms like Veeva CTMS, Clario Feasibility, or other compliant digital tools. This enables audit-ready documentation and seamless comparison across protocols and regions.

10. Ignoring Site Feedback for Continuous Improvement

Finally, many sponsors and CROs fail to review site feedback post-survey. Sites may provide comments such as:

• “Questions are repetitive or unclear”
• “Form is too long for busy clinics”
• “Unable to attach required documents easily”

Incorporating this feedback into subsequent versions ensures higher response rates, better data, and improved sponsor-site collaboration. Sponsors should conduct post-survey reviews or pilot testing to optimize forms continuously.

Best Practice Recommendations

• Limit initial surveys to 25–30 critical questions
• Use digital tools with conditional logic and data upload fields
• Benchmark recruitment estimates with historical performance
• Customize by therapeutic area and regulatory region
• Include risk scoring and auto-flagging mechanisms
• Maintain an audit-ready record with version control and timestamps

Tools like Clinscape, TrialHub, and Medidata can help structure and automate these best practices into scalable survey systems.

Conclusion

Feasibility surveys are the foundation of successful clinical trials. Yet, poor design introduces risk, waste, and non-compliance. Sponsors and CROs must recognize and avoid the common pitfalls outlined above—generic questions, ambiguous wording, missing validations, and absence of risk flagging. By adopting best practices, leveraging digital platforms, and integrating historical data, sponsors can build robust, regulatory-aligned feasibility tools that drive accurate site selection and successful trial execution.

Leveraging Simulation Modeling to Optimize Adaptive Protocols in Rare Disease Trials

Introduction: Why Simulation Is Crucial in Rare Disease Clinical Trials

Designing clinical trials for rare and orphan diseases is inherently complex due to small sample sizes, high inter-patient variability, and uncertain natural history data. Adaptive trial designs—while flexible and efficient—require rigorous planning to ensure statistical validity and regulatory compliance.

To bridge this gap, simulation modeling has become an essential tool for testing, validating, and optimizing adaptive protocols before implementation. Simulation allows sponsors to visualize trial trajectories, assess risks, and fine-tune design parameters long before the first patient is enrolled.

This article outlines how simulation modeling supports rare disease trial planning, detailing statistical techniques, regulatory expectations, and examples of real-world applications.

What Is Simulation Modeling in Adaptive Trials?

Simulation modeling involves running multiple virtual trials using hypothetical patient data, varying key parameters to observe outcomes such as power, error rates, recruitment needs, and endpoint sensitivity.

Common simulation objectives include:

• Evaluating performance of adaptive rules (e.g., sample size re-estimation, early stopping)
• Estimating Type I and Type II error under various assumptions
• Optimizing timing and frequency of interim analyses
• Assessing robustness to deviations (e.g., delayed effects, missing data)

For example, in a Bayesian adaptive design for a gene therapy in spinal muscular atrophy (SMA), simulations can predict when predictive probability thresholds are likely to be met for early stopping, helping sponsors balance risk and trial efficiency.

Tools and Techniques Used in Simulation Modeling

Simulation models range in complexity from basic Excel-based calculations to sophisticated software capable of Monte Carlo simulations. Some popular tools include:

• R and SAS: For customizable simulations using statistical packages like simtrial or gsDesign
• East® (Cytel): Commercial software offering a GUI for adaptive design simulations and FDA-validated models
• FACTS® (Berry Consultants): Bayesian modeling and simulation platform tailored to adaptive designs
• Simulx®: Part of the Monolix suite for longitudinal simulation in pharmacometrics

These tools allow sponsors to test assumptions, such as recruitment delays or endpoint variability, and adjust trial architecture accordingly.

Modeling Endpoint Behavior and Variability

In rare disease trials, endpoints are often novel or under-validated. Simulation helps understand how changes in endpoint distribution affect study outcomes. For instance:

• For SMA, time to respiratory failure is a variable endpoint—modeling helps set realistic detection thresholds.
• In Fabry disease, simulations help decide whether changes in plasma Gb3 levels over 6 months are significant enough for interim analysis triggers.

Endpoint simulation supports selection of optimal primary endpoints, refinement of composite measures, and identification of early biomarkers predictive of long-term benefit.

Use Case: Modeling Seamless Phase II/III Trials in a Genetic Disorder

In a trial for a new treatment in a genetic lysosomal storage disorder, the sponsor planned a seamless Phase II/III adaptive design. Simulation modeling was used to:

• Determine when to trigger transition from dose-finding to confirmatory phase
• Validate operating characteristics under multiple dose-response curves
• Estimate likelihood of reaching success criteria for accelerated approval

Based on 10,000 virtual trial runs using Bayesian priors from natural history data, the design was approved by the FDA under the Orphan Drug pathway. The simulation saved 12 months in development time.

You can explore similar adaptive trials in rare diseases on the Japan Registry of Clinical Trials.

“`html

Simulating Recruitment and Operational Risks

One of the most unpredictable variables in rare disease trials is patient enrollment rate. Simulations can model recruitment under different assumptions, considering factors such as:

• Geographic distribution of eligible patients
• Competing trials for the same population
• Site initiation delays or protocol complexity

Using simulation, sponsors can test timelines and create mitigation strategies, such as site expansion triggers or remote monitoring protocols. This is particularly useful in global studies involving ultra-rare diseases with a prevalence of 1–5 per 100,000.

Regulatory Expectations for Simulation-Based Protocols

Regulators increasingly expect simulations to accompany adaptive protocol submissions, especially under orphan or accelerated pathways. Key requirements include:

• Detailed Simulation Reports: Including all assumptions, algorithms, and summary of outcomes
• Design Operating Characteristics (DOC): Showing probability of trial success under various scenarios
• Alignment with SAP: Simulations must mirror actual planned analyses
• Discussion in Scientific Advice/Pre-IND: Agencies prefer early engagement to evaluate simulation methodology

Both EMA and FDA have welcomed simulations in submissions, especially in pediatric rare disease protocols where empirical data may be limited.

Challenges and Limitations of Simulation Modeling

While powerful, simulation modeling has limitations:

• Garbage in, garbage out: Poor input assumptions lead to misleading outputs
• Computational complexity: Advanced models may require high-performance computing
• Uncertainty quantification: Probabilistic modeling needs robust sensitivity analyses
• Oversimplification risk: Simulations may fail to capture real-world deviations or rare safety signals

Hence, simulation results must be interpreted as decision-support tools, not predictive certainties. Regular model validation and alignment with empirical data remain crucial.

Integrating Simulations into Clinical Development Strategy

Simulation modeling should not be a one-time protocol design activity—it should be integrated into the broader clinical development strategy. Applications include:

• Portfolio planning: Modeling outcomes across multiple compounds
• Health economics: Estimating long-term benefit-risk ratios
• Manufacturing planning: Forecasting product needs based on trial success scenarios

This holistic use enhances not just trial design but business decisions in the rare disease space, where every resource counts.

Conclusion: Modeling Innovation for Adaptive Success

Simulation modeling empowers sponsors to build smarter, more resilient adaptive trials tailored to the complexities of rare diseases. From protocol optimization to regulatory strategy, simulations reduce uncertainty and facilitate data-driven design decisions.

When aligned with regulatory expectations and grounded in real-world assumptions, simulations serve as a critical bridge between scientific ambition and clinical feasibility in rare disease development.