Over The Counter HIV Testing Estimating The Public Health Benefit Recommendations for Required Accur

1
Over The Counter HIV TestingEstimating The
Public Health Benefit Recommendations for
Required Accuracy
Freya Spielberg, MD MPH Freya_at_u.washington.edu
2
Public Health Risk vs. Benefit

• Risks
• False negatives undiagnosed HIV
• False positives negative emotional reactions.
• Benefits
• Greater knowledge of HIV status decreased HIV
  incidence
• Early treatment Decreased morbidity, mortality,
  negative economic and psychological consequences.

3
Weighing the Risks and Benefits

• Surveys of Acceptability of OTC Testing1,2
• Experience with Home Specimen Collection3
• Self-Testing Study Positive Perspective4,5
• Effective Sensitivity6

4
Public Health Risk vs. Benefit

• If OTC results in more people overall with HIV
  learning their status then the benefits will
  likely outweigh the risks.

5
Will OTC Reach The Untested?

• Two survey studies among high risk (N460,
  N354), Seattle and San Francisco -OTC preferred
  by 20 to 241,2
• OTC significantly preferred by people who had
  never tested. 1
• In self-testing study among 240 people with HIV,
  61 preferred to learn their results with OTC
  over clinic testing.3,4
• Home Specimen Collection (HSC) post marketing
  study - 49 of 1494 HIV positive users had never
  tested before. 5
• OTC preferred 20 to 24 times more than HSC1,2
• Conclusion It is likely that an OTC will reach
  a substantial number of people with HIV who have
  never tested before.

6
How Accurate Does an OTC Need to be for the
Public Health Benefits to Outweigh the Risks?

• Effective Sensitivity The of true positives
  identified in a population when testing is
  available Assume performance sensitivity is 99
  for staff-testing and 95 for OT C.
• If staff-based rapid testing is accepted by 50
  of the people with unidentified HIV,
• and adding an OTC results in an additional 20 of
  those infected testing
• then the Effective Sensitivity of the staff test
  would be 49.5 whereas the effective sensitivity
  in a population where OTC tests were also
  available would be 66.5.

7
A Case for Lowering the Accuracy Bar for FDA
approval of an OTC

• It is unlikely that people at highest risk will
  have as high performance accuracy as trained
  staff.
• In our pilot (N240) of self testing we found 95
  self-testing sensitivity 3
• Requiring equivalent accuracy to lab-based tests
  (98) will discourage companies from performing
  clinical trials.
• Although the test performance sensitivity of an
  OTC may be somewhat lower than a staff-based
  test, the overall effective sensitivity of OTC
  testing will be comparably much higher.

8
Recommendations for Accuracy Criteria for OTC
Approval

• Standard Clinical Trails Require performance
  sensitivity/specificity of gt 95.
• Or require clinical trials that measure effective
  sensitivity.

9
References

• Spielberg F, Branson B, Goldbaum G, Lockhart D,
  Kurth A, Celum C, Rossini A, Critchlow C, Wood R.
  Overcoming Barriers to HIV Testing Preferences
  for New Strategies Among Clients of a Needle
  Exchange, an STD Clinic and Sex Venues for MSM.
  JAIDS 2003 32(3)318-327.
• Skolnik HS, Phillips KA, Binson D, Dilley JW.
  Deciding where and how to be tested for HIV what
  matters most? J Acquir Immune Defic Syndr 2001
  27 292300.
• Spielberg F, Levine RO, Weaver M. Self-testing
  for HIV a new option for HIV prevention? Lancet
  Infect Dis. 2004 Oct4(10)640-6. Review.
• Spielberg F, Camp S, Ramachandra E. HIV Home
  Self-Testing Can It Work? 2003 National HIV
  Prevention Conference, Atlanta, July 27-30,
  Abstract 1007.
• Branson B. Home sample collection tests for HIV
  infection. JAMA 1998 280 1699701.