Evaluating Past Site Performance: A Key to Smarter Clinical Trial Feasibility

Introduction: Why Historical Site Performance Matters

In the competitive landscape of clinical trials, choosing the right sites can make or break a study. One of the most predictive indicators of future success is a site’s historical performance in prior trials. Regulators like the FDA and EMA expect sponsors and CROs to use past performance as part of risk-based site selection under ICH E6(R2) guidelines.

Evaluating site performance isn’t simply about how fast a site can enroll. It includes understanding past enrollment trends, protocol deviation rates, audit findings, data quality issues, and patient retention patterns. This article provides a detailed methodology for assessing historical site performance as part of a robust feasibility process, supported by real-world examples and performance dashboards.

Key Performance Indicators (KPIs) for Site History Evaluation

To evaluate a site’s past performance, sponsors should examine a mix of quantitative and qualitative KPIs. These include:

• Actual vs. projected enrollment rates
• Screen failure ratios and dropout rates
• Frequency and severity of protocol deviations
• Query resolution timelines and data quality metrics
• Audit findings (internal, sponsor, and regulatory)
• Inspection outcomes (e.g., FDA 483s, Warning Letters)
• Timeliness of regulatory and EC submissions
• Monitoring burden (e.g., number of follow-ups required)

These metrics should be reviewed for at least 3–5 previous trials, ideally within the same therapeutic area and trial phase.

Sources of Historical Site Performance Data

Collecting past performance data requires a blend of internal systems, external databases, and direct site engagement. Typical sources include:

• CTMS (Clinical Trial Management System): Site visit logs, enrollment data, deviation reports
• EDC Systems: Query logs, data entry timelines, SDV delays
• Monitoring Reports: CRA visit notes, risk indicators
• Trial Master File (TMF): Inspection reports, CAPAs, and audit summaries
• Regulatory Databases: Publicly available inspection databases like [FDA 483 Database](https://www.fda.gov/inspections-compliance-enforcement-and-criminal-investigations/inspection-technical-guides/fda-inspection-database)
• WHO ICTRP or [ClinicalTrials.gov](https://clinicaltrials.gov): Used to identify prior studies at the site or by the PI

Sample Performance Scorecard Template

A standardized scorecard helps quantify site performance for comparative analysis.

Such tools allow sponsors to adopt objective, data-driven site selection methodologies.

Case Study: Impact of Historical Performance on Site Choice

In a global oncology trial, Sponsor X was selecting 40 sites across Europe and Asia. Site X1 had responded quickly to feasibility and had solid infrastructure. However, their CTMS record showed:

• 8 major protocol deviations in the last study
• 2 instances of delayed AE reporting
• 5 subject dropouts within the first 4 weeks

Despite strong initial feasibility responses, these historical indicators led the sponsor to deselect the site. Another site with moderate infrastructure but better historical KPIs was chosen instead, reducing overall trial risk.

How to Score and Benchmark Sites

Organizations can develop internal scoring systems based on historical metrics. A basic example includes:

• Enrollment performance: 30 points
• Protocol compliance: 30 points
• Data quality: 20 points
• Inspection/audit history: 20 points

Sites scoring above 80 may be pre-qualified. Those under 60 should be considered only with additional oversight or justification.

Integrating Performance Data into Feasibility Systems

To make site history actionable, integration into planning systems is essential:

• Link CTMS and feasibility dashboards for real-time performance scoring
• Use machine learning to predict high-risk sites based on historical patterns
• Tag underperforming sites with audit flags or CAPA requirements
• Centralize all prior audit and deviation data into the site master profile

Organizations using integrated platforms report faster site selection, improved regulatory compliance, and better patient retention.

Regulatory Expectations for Documenting Site Selection

Per ICH E6(R2), sponsors must “select qualified investigators and sites” and provide documentation to justify their selection. Key expectations include:

• Documented rationale for site inclusion or exclusion
• Evidence of performance metrics and monitoring trends
• Identification and mitigation of prior compliance issues
• Storage of evaluations in the TMF for inspection purposes

EMA inspectors, for example, may request justification for selecting a site with prior inspection findings or underperformance, especially if not mitigated by CAPAs.

Best Practices for Historical Site Review

• Review minimum 3 prior trials within the last 5 years
• Include PI-specific metrics as well as site-wide data
• Engage QA to review audit and CAPA history
• Cross-check with public databases (e.g., FDA 483s, EU CTR)
• Use scorecards to support selection meetings and approvals
• Archive all scoring and rationale documents in the TMF

Conclusion

Evaluating a site’s past performance is a critical component of modern, risk-based clinical trial feasibility. It ensures that decisions are informed, justified, and aligned with regulatory expectations. Sponsors and CROs that adopt structured performance reviews—integrated with feasibility workflows and planning systems—can reduce trial risks, enhance subject safety, and accelerate startup timelines. As trials become more complex and globalized, historical data will remain a core strategic asset in clinical operations planning.

Designing Feasibility Questionnaires Using Historical Site Data

The Importance of Historical Site Data in Feasibility Planning

Feasibility questionnaires are foundational tools in clinical trial planning. They help sponsors and CROs identify and select high-performing sites based on several factors like patient pool, investigator experience, infrastructure, and regulatory track record. However, when these questionnaires are designed without historical context, they can result in overly optimistic or inaccurate site responses. That’s where leveraging historical site data becomes critical.

Historical site data includes past enrollment rates, protocol deviation frequencies, screen failure rates, regulatory inspection outcomes, and adherence to visit schedules. Sponsors that fail to incorporate this data often face recruitment delays, budget overruns, and poor site compliance. Regulatory bodies including the FDA, EMA, and MHRA emphasize the use of evidence-based feasibility strategies during sponsor inspections.

In this article, we explore how to use historical site data to design smarter, more predictive feasibility questionnaires that improve site selection and study startup efficiency.

Types of Historical Data Relevant to Questionnaire Design

Historical site data spans multiple domains. The most useful categories include:

• Enrollment History: Number of subjects enrolled in similar trials within a specific timeframe
• Protocol Adherence: Frequency of deviations and their root causes
• Screen Failure Rates: Percentage of screened patients not meeting inclusion criteria
• Site Activation Timelines: Average time from contract finalization to first patient in (FPI)
• Regulatory Inspection Outcomes: FDA 483 observations, MHRA findings, or internal QA audits

Below is an example data summary from three sites in a cardiovascular trial:

From this table, it’s evident that Site A outperformed others in all key areas. Integrating this insight into a questionnaire helps to focus future feasibility assessments on parameters that matter.

Integrating Data into Feasibility Questionnaire Logic

Feasibility tools often consist of static checklists or self-reported site capabilities. When these are integrated with historical performance data, they become much more predictive. Here’s how historical data can enhance questionnaire sections:

• Recruitment Potential Section: Pre-fill enrollment numbers from past studies and ask the site to explain any changes
• Protocol Adherence Section: Highlight deviation patterns from previous trials and assess current mitigation measures
• Timeline Commitments: Use actual past activation data to validate new timeline estimates

For example, a dynamic form might display: “In your last three trials in this therapeutic area, your average enrollment was 20 patients over 6 months. What has changed to support your estimate of 60 patients in this protocol?”

This approach discourages over-promising and helps differentiate high-performing, realistic sites from aspirational responders.

Sources of Historical Site Data

Historical site data can be gathered from several internal and public sources:

• Clinical Trial Management Systems (CTMS): Capture site-level metrics from previous studies
• Electronic Data Capture (EDC) Platforms: Document protocol adherence and visit windows
• Trial Registries: Data from Be Part of Research (NIHR) and other registries to validate enrollment timelines
• Quality Management Systems (QMS): Archive audit outcomes, CAPA timelines, and deviations

Sponsors that maintain a structured site master file with past feasibility, audit reports, and performance summaries can extract this data with minimal effort. It’s also beneficial to include CRO partner databases and publicly available performance scores (e.g., from the TransCelerate Shared Investigator Platform).

Feasibility Questionnaire Elements That Benefit from Data Integration

Not all parts of a feasibility questionnaire require historical data, but certain sections benefit significantly from it:

By designing forms with auto-filled historical fields, sponsors can reduce bias and increase transparency. Some tools even allow scoring systems based on prior performance benchmarks.

Case Study: Data-Driven Feasibility Yields Better Enrollment

In a 2023 Phase II neurology study, the sponsor used historical site performance data to filter out low-recruiting sites from a previous epilepsy trial. By incorporating metrics such as “patients enrolled per FTE” and “visit adherence rate,” they excluded 30% of sites that had previously delayed timelines. The remaining sites achieved 95% of the recruitment target three months ahead of schedule.

This outcome illustrates how applying historical metrics during feasibility tool design directly impacts enrollment, cost, and data integrity.

Tools and Platforms That Support Data-Driven Questionnaire Design

Sponsors can use various platforms to operationalize this approach:

• CTMS Platforms: Veeva Vault CTMS, Medidata RAVE
• Feasibility Tools: SiteIQ, Clinscape Feasibility Module
• Analytics Dashboards: Tableau, Power BI connected to CTMS/EDC sources
• Risk-Based Monitoring Tools: RBM dashboards that include performance trend lines

These systems allow sponsors to design adaptive questionnaires, conduct real-time validation of site claims, and score site responses against benchmarks.

Challenges and Considerations

Despite the advantages, there are challenges to using historical data:

• Data inconsistency across CROs and systems
• Lack of access to complete legacy data for global sites
• Privacy and data protection regulations (e.g., GDPR)
• Misinterpretation of context (e.g., poor performance due to protocol flaws, not site issues)

Therefore, sponsors must contextualize historical data and allow sites to provide explanations for deviations or poor performance. Data should be used to initiate dialogue, not penalize sites without cause.

Conclusion

Designing feasibility questionnaires using historical site data enables evidence-based site selection, reduces trial risk, and improves regulatory compliance. Sponsors should move away from static, self-reported surveys and adopt dynamic, data-informed tools that consider past performance. Platforms such as CTMS, QMS, and analytics dashboards can help integrate these insights into feasibility tools, creating a predictive framework for identifying high-performing, inspection-ready sites. In doing so, the industry takes a meaningful step toward smarter, faster, and more reliable clinical trial execution.