1 / 33

Willi McFarland, MD, PhD, MPH&TM

Best Practices and Key Issues Related to Measurement Approaches for Characterizing MARPS Epidemics and Populations Overview. Willi McFarland, MD, PhD, MPH&TM Director of HIV Surveillance, San Francisco Department of Public Health

cairo-newton
Download Presentation

Willi McFarland, MD, PhD, MPH&TM

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Best Practices and Key Issues Related to Measurement Approaches for Characterizing MARPS Epidemics and Populations Overview Willi McFarland, MD, PhD, MPH&TM Director of HIV Surveillance, San Francisco Department of Public Health Associate Professor, Epidemiology and Biostatistics, University of California, San Francisco Interventions with Most-at-Risk Populations in PEPFAR Countries: Lessons Learned and Challenges Ahead, Chenai, India, February 18-20, 2009

  2. MARPs Measurement Issues Sampling Population size estimation Key measures in tracking the HIV epidemic and our response

  3. Sampling MARPS • Gold standard: Probability-based sampling • Representative • Accurate • Reproducible • Approximations of probability sampling (RDS, TLS, TS) • The best of both worlds? • Or, Purgatory? • Down to earth: Convenience and snowball sampling • Inclusive • Consistent • Standardized

  4. I RDS

  5. IDU in Tehran, Iran

  6. Equilibrium for sex with foreigners in RDS study of MSM, Uganda No foreign sex partner Had any foreign sex partner

  7. China Ukraine ? Vietnam USA Honduras Brazil Croatia Nepal Uganda Paraguay Bangladesh Kenya Thailand South Africa Zanzibar Indonesia MSM: RDS Successes from Malekinejad et al., 2008

  8. Ukraine UK Mexico Vietnam San Francisco Estonia Nepal India Russia Iran Thailand Albania Tanzania Serbia Bosnia Montenegro Indonesia IDU: RDS Successes

  9. San Francisco Tanzania Vietnam Honduras India Papua New Guinea Female sex workers: RDS Successes

  10. Keeping ahead of the curve: HIV prevalence MSM in Beijing, by RDS RDS successfully tracked the transition from low level to concentrated HIV epidemic % infected 2004 2005 2006 Courtesy of Xiaoyan Ma (China CDC Beijing) and colleagues

  11. Chicago Coupon economy Russia Iran Kosovo India IDU: RDS Failures

  12. China Brazil Caribbean India Paraguay High stigma Low trust Low incentive Not networked Low trust Serbia Too trafficked Female sex workers: RDS Failures

  13. Beijing Repro-ducibility Jinan Estonia Low incentive Anhui Low incentive Caribbean Low incentive Low trust High stigma Albania? Papua New Guinea? Kosovo Fake MSM? Low incentive penile swab Cambodia Coupon economy Guangzhou Bangladesh? Fake MSM? Ninxia MSM: RDS Failures Bad reputation

  14. Equilibrium levels in 3 RDS surveys, MSM, Beijing RDS was not as reproducible as expected

  15. I RDS

  16. I TLS

  17. Targeted Sampling and TLS Locations of IDU activity in San Francisco Physical and ethnographic mapping Targeted sampling - recruit purposely to reflect ethnography Time-location sampling (TLS) – randomize venues, days, and times to recruit

  18. San Francisco MSM Bangkok MSM Harare Alcohol users Salgueiro Mobile truck drivers Shenzhen MSM Forteleza Upper SES MSM TLS successes and lessons learned

  19. HIV prevalence among MSM in Bangkok by TLS TLS studies sounded an alarm for MSM across Asia % infected 2007 2005 2001 2003 Courtesy of Frits van Griensven (US CDC Thailand) and colleagues

  20. Many TLS Issues Dependent upon formative research Uncertain probabilities of inclusion Logistics, safety, time-consuming Ethical, exposure of safe venues

  21. Sampling Best Practices: TLS or RDS? TLS “What you see is what you get” RDS “Faith-based” Both require good formative research

  22. MARP Population Size Estimation No Census, no gold standard Definitions vary Methods vary

  23. Population Size Estimation Methods Census method Population-based surveys Components Compartmental Truncated Poisson Nomination Network scale up Capture-recapture Multiplier method • Delphi • Consensus • Conventional wisdom • Borrow from neighbor • Adjust from a neighbor

  24. MARP Measures Issues Post ART scale-up world Increasing demand for data New biological measures New behavioral measures

  25. Biological Measures HIV and STI prevalence HIV incidence (e.g., BED) Post ART scale-up era: Previously unknown infection CD4, unmet need for ART ARV resistance

  26. Behavioral Measures • Core, consistent indicators • Basic transmission-related behaviors • Knowledge, attitudes, beliefs • Post ART scale-up era • Behavior affected by knowledge of serostatus, partners’ serostatus, and ART use • New hypotheses • Sexual networks rather than individual level behavior

  27. Seroadaptation among known HIV-positive MSM, San Francisco, 2008 Any unprotected sex: 70% High HIV transmission risk: 19%

  28. Concurrent When there are 2+ partners, African American MSM are 3 x more likely to have concurrent partnerships than other MSM W Serial African American MSM Other MSM

  29. Conclusions: best practices (or Purgatory?) Methods: Appropriate to the target population, resources Triangulation: Use multiple methods, measures, sources, and types of data Measures: Keep ahead of the curve

  30. Keep ahead of the curve

More Related