Complete Case or Full Dataset? Choosing the Right Analysis Approach for Missing Data

Handling missing data is a critical decision in clinical trial analysis. Two commonly considered approaches are Complete Case Analysis (CCA) and Full Dataset Modeling (e.g., MMRM or Multiple Imputation). Choosing between them requires understanding the underlying assumptions, data structure, regulatory expectations, and impact on validity.

This guide explores when it is appropriate to use complete case analysis versus full dataset methods in biostatistical evaluations. We’ll also discuss the regulatory context from agencies like the USFDA and EMA, and offer practical recommendations to guide your decision-making process.

Understanding Complete Case Analysis (CCA)

Complete Case Analysis involves analyzing only those subjects for whom all relevant data are available. Any patient with missing data on the outcome or a key covariate is excluded from the analysis.

Advantages of CCA:

• Simple to implement and interpret
• Works with standard statistical tools
• No modeling assumptions about the missing data

Limitations of CCA:

• Leads to loss of sample size and statistical power
• Results may be biased if data are not Missing Completely at Random (MCAR)
• Cannot be used when missingness is high or systematic

When to Use CCA:

• When the proportion of missing data is low (<5%)
• When data are MCAR (i.e., probability of missingness is unrelated to both observed and unobserved data)
• When conducting exploratory or supportive analyses

CCA may be acceptable under specific circumstances, but its limitations must be clearly stated in the trial documentation.

Understanding Full Dataset Analysis

Full Dataset Analysis refers to techniques that incorporate all available data, including cases with partial information. Examples include:

• MMRM (Mixed Models for Repeated Measures): Accommodates MAR (Missing at Random) data
• Multiple Imputation: Uses observed data to predict and fill in missing values
• Maximum Likelihood Estimation: Accounts for partial data without explicit imputation

Advantages of Full Dataset Methods:

• Preserves statistical power by using all available information
• Yields unbiased estimates under MAR assumptions
• Widely accepted by regulatory agencies

Limitations:

• Requires correct specification of the model
• May be computationally intensive
• Assumptions (like MAR) must be justified

These methods are favored in regulatory reviews, especially for primary endpoints. Their inclusion in the Statistical Analysis Plan reflects best practice in handling missing data.

Regulatory Guidance: CCA vs Full Dataset

Regulators discourage CCA as a primary analysis method unless MCAR can be assumed and justified. For pivotal trials, agencies like the FDA and EMA recommend full dataset approaches with appropriate sensitivity analyses.

Key Guidelines:

• FDA Guidance on Missing Data (2010): Emphasizes pre-specification and avoidance of CCA
• ICH E9(R1): Introduces estimands that define the role of intercurrent events like dropout
• EMA Guideline on Missing Data: Encourages model-based analyses with sensitivity checks

Documentation of methods and justification of assumptions is critical for regulatory compliance.

Practical Comparison: When to Choose What

Best Practices for Implementation

• Include both CCA and full dataset methods in SAP as primary and supportive analyses
• Clearly define assumptions about missing data mechanisms
• Perform and report sensitivity analyses (e.g., tipping point, delta adjustment)
• Use statistical software with validated imputation modules
• Document rationale and results per SOPs and in the CSR

Conclusion

The decision to use complete case analysis or full dataset modeling should be driven by data characteristics, missingness mechanisms, and regulatory requirements. While CCA is easy to apply, it is limited to rare MCAR situations and should only be used as supportive analysis. Full dataset approaches like MMRM and multiple imputation offer robust solutions under MAR and are preferred in regulatory submissions. Incorporating both strategies—alongside transparent assumptions and sensitivity analyses—ensures your trial results remain valid and defensible.

How to Adjust Sample Size for Dropouts and Noncompliance in Clinical Trials

One of the most overlooked yet critical steps in clinical trial planning is adjusting the calculated sample size to account for patient dropouts and noncompliance. These real-world challenges can significantly reduce the effective power of a study, increasing the risk of inconclusive or biased results. Proactively planning for attrition and protocol deviations ensures the integrity and regulatory acceptability of trial outcomes.

This guide walks through the rationale, formulas, and best practices for adjusting sample sizes for expected dropouts and noncompliance, aligned with expectations from regulatory authorities such as the USFDA and CDSCO.

Why Adjust for Dropouts and Noncompliance?

The ideal number of subjects calculated from a power analysis assumes perfect retention and compliance. However, in real trials:

• Participants may withdraw due to side effects, relocation, or personal reasons
• Subjects may not follow the protocol (miss doses, skip visits)
• Data may be incomplete or lost

These issues compromise the **intention-to-treat (ITT)** and **per-protocol (PP)** populations, reducing power and introducing bias. Adjusting for this anticipated loss ensures that the trial meets its original objectives.

Understanding Dropouts vs. Noncompliance

Dropouts

Participants who discontinue the study prematurely and do not provide complete endpoint data. This affects both ITT and PP analyses.

Noncompliance

Subjects who remain in the study but deviate from the treatment protocol. Their inclusion/exclusion may affect only PP analyses.

Step-by-Step: Adjusting the Sample Size

Step 1: Calculate Initial Sample Size

Use standard formulas based on effect size, alpha, power, and variability, assuming 100% compliance and no attrition.

Step 2: Estimate Dropout and Noncompliance Rates

Base your assumptions on:

• Previous trials in similar indications
• Pilot studies or feasibility assessments
• Therapy burden, follow-up duration, and patient population

Typical dropout rates:

• Short-duration trials: 5–10%
• Chronic conditions: 15–25%
• Oncology or long-term follow-up: ≥30%

Step 3: Inflate Sample Size

The adjusted sample size (n_adjusted) can be calculated using:

  nadjusted = n / (1 − d)
  

Where:

• n = Initial sample size per group
• d = Anticipated proportion of dropouts/noncompliant subjects (e.g., 0.15 for 15%)

Example:

Initial sample size = 120 subjects

Expected dropout = 20%

Adjusted sample size = 120 / (1 − 0.20) = 150 subjects

Handling Multiple Attrition Risks

In some studies, dropout and noncompliance are treated separately. A conservative approach is to add buffers sequentially:

  n′ = n / (1 − dropout) × (1 − noncompliance)
  

Example:

Dropout = 15%, Noncompliance = 10%

n′ = n / (0.85 × 0.90) = n / 0.765

→ Inflate by ~30.7%

Regulatory Perspective on Adjustments

Both regulatory agencies and ethics committees expect realistic planning for attrition. Key expectations:

• Justification of dropout and noncompliance estimates
• Impact assessment on statistical power and endpoint interpretation
• Clear documentation in the SAP and clinical protocol
• Plans for patient engagement and retention strategies

Best Practices for Managing Dropout Impact

1. Historical Data: Use dropout rates from comparable studies as a baseline
2. Protocol Design: Reduce patient burden to minimize attrition
3. Patient Engagement: Incorporate reminders, follow-ups, and retention campaigns
4. Monitoring: Track dropout trends throughout the study for early correction
5. Analysis Populations: Plan ITT, PP, and as-treated analysis sets in advance

Example in Practice: Phase 3 Diabetes Trial

• Initial calculated sample: 180 subjects per arm
• Expected dropout: 15%
• Expected noncompliance: 10%
• n_adjusted = 180 / (0.85 × 0.90) = 235 subjects

The team would plan to recruit 470 subjects total to ensure 360 compliant completers for final analysis.

Tools and Resources

• Sample size calculators with dropout adjustment modules (e.g., G*Power, nQuery)
• Statistical programming in R (e.g., pwr and epiDisplay packages)
• Validation of calculations through pharmaceutical validation processes

Common Mistakes to Avoid

• Using generic dropout rates without context
• Failing to document adjustments in SAP
• Over-recruiting without power recalculation
• Ignoring compliance monitoring plans
• Assuming retention efforts alone will suffice

Conclusion: Proactive Adjustment Ensures Trial Integrity

Failing to account for dropouts and noncompliance can jeopardize an otherwise sound clinical trial. Adjusting the sample size with realistic estimates helps maintain statistical power and aligns with ethical and regulatory expectations. This essential step should be incorporated early during the SAP and protocol development phases, ideally with involvement from a biostatistics and quality assurance team.

Explore More:

• Stability testing protocols
• Pharma SOP checklist
• GMP documentation