Enhancing Benefit Assessments: Insights from IQWiG Real-World Data Report

How will the IQWiG real-world data report (linked below) change drug benefit assessments in Germany?

The new IQWiG real‑world data report sets a formal framework for how real-world data (RWD) supports non-randomized comparative studies underpinning benefit assessments under §35a SGB V. By standardizing methodologies such as target‑trial emulation, systematic confounder identification, and digital patient-reported outcome (PRO) collection, the report aims to strengthen the credibility and utility of RWD in Germany’s AMNOG pharmaceutical assessment process.

Learn more about the official findings and recommendations in IQWiG’s detailed press release and full analysis.

Executive Summary

• The IQWiG real-world data report establishes clear standards for designing, collecting, and analyzing RWD in German benefit assessments, with a strong emphasis on target‑trial emulation, rigorous confounder control, and advanced handling of missing data.
• Registries are highlighted as primary data sources, while the guidance clarifies when to apply Active-Comparator New-User (ACNU) versus Prevalent-New-User (PNU) study designs.
• The framework encourages digital collection of PROs, introduces the “shifted null” concept for interpreting effect sizes, and proposes orienting sample-size approaches for studies with limited information.

Key Insights from the IQWiG Real-World Data Report

Target-Trial Emulation: Raising RWD Standards

• Core Principle: Align study design elements—eligibility, treatment assignment, outcomes, and follow-up—with a hypothetical randomized controlled trial (RCT) to minimize bias.
• Confounder Pre‑specification: Use a three-step approach: review existing literature, conduct structured clinician interviews, and apply relevance scoring to identify and prioritize covariates.

Robust Designs Fit for Real-World Practice

• Study Architecture:
  • Active-Comparator New-User (ACNU): Favored when both comparison arms start new therapies simultaneously.
  • Prevalent-New-User (PNU): Applied if one group continues standard care; methods such as time-conditional propensity scores account for prior exposure.
• Handling Treatment Switching: Conduct primary analyses per intention-to-treat and supplement with sensitivity analyses (e.g., inverse probability of censoring weights) only if confounder data are sufficiently robust.

Missing Data: Advanced Methods Required

• Data Imputation: Multiple imputation by chained equations (MICE) is required, including within-imputation propensity-score workflows, comprehensive sensitivity analyses, and the inclusion of outcomes in the imputation model.
• Guidance: Complete-case analyses are discouraged unless missingness is minimal.

PRO Collection: Digital Tools for Engagement

• Best Practices: Schedule assessments at standardized intervals with flexible tolerance windows and use digital platforms for PRO capture—facilitating feedback loops to patients and clinics.

Sample Size & Small Populations

• Orienting Sample Size: When prior data are sparse, employ patient-count-driven scenarios and the “shifted null” threshold (e.g., RR0=2 or HR0=0.5) to support robust effect estimation.
• Small Sample Solutions: Propensity-score adjustments remain possible if overlap and convergence are demonstrated; use pre-specified variable reduction and variance bootstrapping as needed.

Implications for Health Economics and Outcomes Research (HEOR)

Operationalizing the IQWiG Real‑World Data Report

1. Study Design Requirements:
   • Specify a target-trial protocol and statistical analysis plan (SAP).
   • Clearly define study “time zero,” and justify design choices (ACNU vs. PNU).
   • Pre-register confounders using the recommended three-step identification strategy.
2. Analytical Standards:
   • Match estimands (ATE, ATT, overlap) to chosen propensity-score techniques.
   • Quantify overlap and balance, perform robust missing data imputation, and transparently report sensitivity analyses.
3. PROs & Small Populations:
   • Use validated digital tools for PROs, schedule frequent early assessments, and ensure feedback to increase participation.
   • For studies with limited sample sizes, rely on weighting/adjustment methods and initiate interim sample size reviews based on actual accrual.
4. Pricing, Access, and Global Consistency:
   • The report’s standards will heighten the credibility of RWD, directly impacting AMNOG price negotiations and re-assessments.
   • These methodologies are in step with international expectations, facilitating cross-border registry studies and multinational evidence submission.

Explore more on the methodology and implications in IQWiG’s press release and report summary.

Frequently Asked Questions (FAQ)

What is the “shifted null” in the IQWiG real-world data report?
A shifted null applies a more stringent threshold for statistical significance (e.g., setting a risk ratio null at 2 or hazard ratio at 0.5) to account for potential residual confounding in non‑randomized studies. This ensures observed effects surpass a higher bar, thereby supporting more credible benefit claims.

Are registries required for studies under §35a SGB V?
Registries are the recommended data source due to their longitudinal tracking and capacity for detailed, digital PRO collection. While claims data may be used when appropriate, they often lack the granularity needed for rigorous confounder adjustment and alignment with inclusion criteria.

How should treatment switching be addressed in RWD studies?
Primary analyses should apply an intention-to-treat (ITT) approach or treatment-policy estimand. Sensitivity analyses (such as censoring at the point of switching or using IPCW) are only recommended if registries offer sufficient confounder data and assumptions can be clearly justified.

Conclusion & Next Steps

The IQWiG real-world data report establishes a gold standard for using RWD in German benefit assessments—emphasizing methodological transparency, digital innovation, and global compatibility. Organizations submitting to AMNOG or aiming for multinational HTA submissions should align protocols and analyses with this framework to reinforce evidence credibility and support positive health technology outcomes.

For deeper insights and official documentation, consult IQWiG’s comprehensive press materials.