Comparative Effectiveness Research: Past Failings and Future Directions

1
Comparative Effectiveness Research Past Failings
and Future Directions
Randall S. Stafford, MD, PhD Stanford Prevention
Research Center Program on Prevention Outcomes
and Practices Stanford University, Stanford CA

• Alexander GC, Stafford RS. Does Comparative
  Effectiveness Have a Comparative Edge? JAMA 2009
  301 2488-90. PMID 19531789.
• Stafford RS, Wagner TH, Lavori PW. New, but Not
  Improved? Incorporating Comparative-Effectiveness
  Information into FDA Labeling. NEJM 2009 361
  1230-3. PMID 19675326.

2
Past Failings

• Evidence assessments available only after
  widespread adoption of new practices
• Lack of explicit mechanisms to modify practice in
  response to CER evidence
• Focus on drugs neglects array of other clinical
  strategies
• Regulatory environment is mismatched with concept
  of comparison
• Selection of CER topics haphazard
• Costs seen as a secondary consideration

3
Future Directions

• Generate data much more rapidly
• Link evidence to strategies proven to modify
  clinical practice
• Broaden agenda beyond drugs and devices
• Alter regulatory environment to better match need
  for comparisons
• Expand use of data on real world practices
• Explicitly consider the cost implications of
  practice alternatives

4
More Rapid andMore Relevant CER

• Generate primary evidence more quickly
• Anticipate need for comparisons prior to
  widespread dissemination
• Real-time observational/surveillance data
• Pragmatic trials as part of clinical practice
• Better studies that represent equipoise
• Facilitate active comparator RCTs
• Analyze and aggregate data more rapidly

5
Change Practices Based on Evidence

• Link CER evidence to strategies proven to modify
  practice
• Examples of slow adoption of evidence
• Use of warfarin in atrial fibrillation
• Examples of slow discontinuation of harmful
  therapies
• Persistent use of phen/fen after reports of harm
• Need for new models of dissemination, including
  use of EHR and self-management

6
Beyond Drugs and Devices

• CER has focused on medications and medical
  devices
• Need to routinely compare different treatment
  strategies and systems of care
• Surgical and other invasive procedures
• Behavioral changes e.g., physical activity
• Complementary and alternative medicine
• Systems of care e.g., electronic health records,
  nurse case-managers

7
Regulatory Environment

• Alter regulatory environment to facilitate the
  generation of comparative data
• Change labeling requirement to state whether
  comparative evidence exists
• While this drug works better than placebo, there
  is no evidence to suggest its superiority to
  other existing treatments
• Anticipate future off-label uses in initial New
  Drug Application approval
• Better linkage between FDA and CMS

8
Expand Use of Real World Data

• Improved Selection of CER Topics
• Select topics critical to actual practice, not
  based on controversy and presence of RCTs
• Topic refinement and analysis
• Appropriate definition of research questions
• Practice variation as indicator of uncertainty
• Judging applicability of RCT evidence
• Quantify healthier case-mix and better outcomes
  in RCTs vs. real world
• Derive information on cost of therapies

9
Include Cost Implications

• Consider the cost implications of alternatives
• Consideration of value cannot be separated from
  consideration of costs
• Menu without prices Garber
• Comparing auto acceleration
• Viper 89K ? 0-60 mph 4 s, 600 HP, 12 mpg
• Yaris 15K ? 0-60 mph 11 s, 106 HP, 29 mpg
• May need to evaluate multiple cost metrics,
  beyond simply comparing cost per QALY

10
Conclusions

• Despite accomplishments, need to focus on past
  failings of CER
• Re-engineer CER with proper linkages to
• Patients
• Providers
• Payers
• Regulators
• Industry
• Ignoring costs is politically expedient
• Similar to expanding insurance coverage without
  mandating participation