IRB Reviews of MultiSite, Low Risk, Health Services Research

1
IRB Reviews of Multi-Site, Low Risk,Health
Services Research
2
Two Studies

• Custom-Fit Approach to Implementation of Diabetes
  Hypertension Guidelines
  
• Best Precedents in Human Subjects Protection for
  Health Services Research
  

3
Guidelines Study

• 43 sites
• One-page survey of physicians
• Use of secondary databases to obtain data on
  prescriptions of participating physicians
  

4
Guidelines Study (contd)

• IRB applications included
• Written informed consent for physicians
• Waiver of informed consent for patients
• All application material was completed by staff
  at lead site.
  

5
Guidelines Study (contd)

• 19 months and 4680 hours later
• (Duration and cost of study doubled from
  original budget.)
  

6
Guidelines Study (contd)

• 1 Exemption
• 10 Expedited Reviews
• 31 Full Reviews
• 1 Disapproval

7
Guidelines Study (contd)

• Delays
• Obtaining information from site RD offices
• Finding and retaining a site PI
• Getting signatures from site PI
• Correspondence with committees and staff
  regarding requests for revisions
  
• Obtaining notification and documentation of
  review results
  

8
Guidelines Study (contd)

• Length of Time (Days) for Submission and Approval
  Process
  
• 
  Med Min Max
  
• Submission to first notification
• from committee 49
  14 196
  
• Submission to approval 133 30
  506
  
• Identification of site PI to
• receipt of approval letter 286
  52 798
  

9
Guidelines Study (contd)

• Interesting Anomalies
• Disapproval by one IRB who found the study was
  too risky.
  
• Review by one IRB for two different sites
  approved as minimal risk for one site, rejected
  pending revision for second site.
  
• Request for names of participating physicians at
  annual review, despite assurances of
  confidentiality in application and consent.
  

10
IRB Best Precedents Study

• 20 multi-site, health services research studies
  funded during 2001-2003
  
• 15 reviewed thus far
• 78 sites (mean 5.2 sites/study)
• 7 intervention, 8 observational
• 
  

11
IRB Best Precedents contd

• Number of studies with discrepant types of
  reviews (i.e., full review vs. expedited vs.
  exempt) 11
  
• Lots of requests for changes to consent forms
  that dont seem to have anything to do with local
  environment/culture
  
• 
  

12
Conclusion

• We need to find a way to reduce the time and
  costs associated with review and approval of
  low-risk, multi-site HSR studies.
  
• Balance risk of these studies with cost of
  impeding the research process.
  
• Reduce variability in IRB reviews and responses.
• Remove or revise site PI requirement in cases
  where site PI has no responsibility other than
  signing forms.
  
•