How can we use group data to treat individual patients?

This course features lecture recordings from the first part of the 2015 symposium on Using Group Data to Treat Individuals presented by the Predictive Analytics and Comparative Effectiveness (PACE) center at Tufts Medical Center.

Learning Objectives & Content Outline

1. Risk-Based Heterogeneity of Treatment Effect (HTE) in 30 Large RCTs

David Kent, MD, MSc

After completing this lecture, you will be able to:

• Explain why clinically significant risk heterogeneity is common in clinical trials
• Describe why the typical patient is generally at lower risk than reflected by the trial summary results
• Explain why a risk stratified approach to trial analysis may be the most clinically informative
• Describe the implications of the study.

Lecture Topics

• Risk-Based HTE in 32 Large RCTs -PCORI Pilot Example (5:26)
• Summary Results of Risk-Based HTE (5:49)
• Findings -Risk-Based HTE (1:19)
• Summary Results for 18 Positive Treatment Comparisons (3:26)

2. Person-Level HTE

Issa J. Dahabreh, MD, MS

After completing this lecture, you will be able to:

• Explain person-level HTE in distinction to group-level HTE
• Describe repeated period crossover studies
• Discuss the results and the implications of the study.

Lecture Topics

• Person-Level HTE (4:10)
• Person-Level HTE Methods and Results (3:22)
• Person-Level HTE Study Examples (4:28)
• Person-Level HTE Conclusions (1:41)

3. A Proposed Guideline for Reporting HTE in Large Randomized Clinical Trials

Rod Hayward, MD

After completing this lecture, you will be able to:

• Discuss the general credibility of subgroup analysis
• Explain when individual patient characteristics should be part of a primary HTE analysis
• Discuss when risk-based HTE should be part of a primary HTE analysis.

Lecture Topics

• Proposed Guideline for Reporting HTE (5:43)
• Prior Published HTE Work (6:46)
• Suggested Guidelines for HTE (7:13)