What is the Evidence that VCT Works? David Holtgrave, PhD Professor & Chair Department of Health, Behavior & Society Johns Hopkins Bloomberg School of Public Health Baltimore, MD
Overview of Presentation What is the impact of the “C” (counseling) in VCT? Does anyone include C&T in an evidence-based bundle of prevention services? What is the impact for persons testing HIV+? What is the impact for persons testing HIV-? What if “typical” counseling is not up to standard of care?
Overview of HIV Prevention Interventions Sexual transmission –Small groups –Counseling & testing –Community-level –Structural-level –STI diagnosis and treatment Perinatal transmission –AZT; nevirapine –Breast-milk supplementation Parenteral transmission –Blood safety –Occupational setting precautions –IDU programs Behavior change Drug treatment Access to sterile injection equipment Valdiserri RO et al. Nat Med. 2003;9:881-6.
Effective Prevention Tools Teacher training and peer education Male and female condoms Condom promotion and social marketing Treatment of STIs Voluntary counseling and testing Workplace programs Transfusion screening Prevention of MTCT Mass media campaigns Harm reduction programs Peer counseling Schwartlander B et al. Science. 2001;292:2434-2436.
Impact of “C” in VCT for Persons Testing HIV Seropositive Inherent confound of counseling and testing when test result is positive –Virtually no one recommends testing without counseling for PLWH/A Impact of counseling and testing for persons living with HIV is widely agreed upon in literature Meta-analysis of 11 studies of the impact of counseling and testing for PLWH/A* –68% reduction in high risk sexual behaviors with partners not already HIV+ (95% CI: 59% - 76%) –Very similar findings for men and women *Marks G et al. JAIDS 2005;39:446-453.
Sexually Transmitted HIV Incidence by Awareness of Serostatus Persons aware of HIV infection account for 30% to 46% of new sexual transmissions Persons unaware of HIV infection account for 54% to 70% of new sexual transmissions –That is, the majority of the 32,000 sexually transmitted HIV infections in the US in a given year are from persons unaware they are living with HIV Marks G et al. AIDS 2006;20:1447-1450.
Transmission Rates by Knowledge of Serostatus 1-3 Unaware of HIV seropositivity –Transmission rate estimated at 8.8 to 10.8% Aware of HIV seropositivity –Transmission rate estimated at 1.7 to 2.4% Hence, counseling and testing can have major impact for PLWH/A 1. Holtgrave DR et al. Int J STD AIDS. 2004;15:789-792. 2. Marks G et al. AIDS. 2006;20:1447-1450. 3. Holtgrave, Pinkerton. JAIDS. In press.
Impact of “C” in VCT for Persons Testing HIV Seronegative CDC’s Project RESPECT* 4 arms in RCT (all arms included HIV testing) –4 session counseling –2 session (pre- and post-test) counseling –Didactic information –Didactic information and no follow up until final time point (to control for impact of repeated surveys) STD clinics in Baltimore, Denver, Long Beach, Newark, San Francisco (5,758 HIV- clients) *Kamb ML et al. JAMA. 1998;280:1161-1167.
Project RESPECT* (continued) No major difference between 2 and 4 session counseling Pre- and post-test counseling better than simple provision of didactic information –Increased highly consistent condom use through 6 month follow-up –Decreased incident STDs by 30% at 6 month follow-up –Decreased incident STDs by 20% at 12 month follow-up –Findings consistent across sites *Kamb ML et al. JAMA. 1998;280:1161-1167.
Project RESPECT Press Release from CDC, October 1998 “This study showed that it’s not how much you talk to people about HIV prevention that matters most – but how you talk to them….,” Dr. Helene Gayle “According to CDC, the brief sessions used in this study…are feasible to implement in busy health care settings.” “In this study, the approach was implemented with existing clinic staff, in not much more time than that required for didactic messages, and cost only $8 additional dollars per client to implement.” “Far too often, prevention programs found to be ideal in research are too difficult and expensive to implement in the real world,” said Dr. Mary Kamb. “With this program the ideal can be real, with few additional resources.”
Meta-analysis of Impact of Counseling and Testing on Behavior of HIV- Persons Examining pool of 27 studies, meta-analysis found no significant impact of “counseling and testing” bundle on behavior relative to the untested Study did not tease out impact of CT vs testing alone, nor impact of various types of counseling (due to literature limitations on what actually was included under label “counseling”) Weinhardt LS et al. Am J Public Health. 1999;89:1397-1405.
CDC Responded to Weinhardt et al. CDC published rejoinder noting Weinhardt et al. paper was not an indictment of impact of client-centered counseling in clinic settings –23 of 27 studies were published before counseling and testing guidelines issued by CDC in 1993 –CDC noted strong evidence from Project RESPECT –“We worry that implementation will not take place if readers are unaware that studies published after 1997 have identified counseling approaches that work for persons at increased risk for HIV.” Kamb ML et al. Am J Public Health. 2000;90:1152-1153.
Subsequent Data Sources VCT shown to have significant behavioral benefits over health information (and lagged offering of VCT) in large RCT in Kenya, Tanzania and Trinidad –Also shown cost-effective –Voluntary HIV-1 Counseling and Testing Efficacy Study Group* Virtually impossible to have subsequent studies that are like Project RESPECT, which may have changed the outlook of IRBs as to standard of care Voluntary HIV-1 Counseling and Testing Efficacy Study Group. Lancet. 2000;356:103-112.
What if “Typical” Counseling in the Field is Now Less than Standard of Care? Should we toss out counseling, or bring it up to the “client-centered” standard of care (previously asserted by CDC to be effective, efficient and practical)? –Clearly the second option is to be preferred –To do less could conceivably be construed as a harmful act –Important to explore who in health care system (or who among community-based partners) can deliver the counseling services if clinicians do not
Summary What is the impact of the “C” (counseling) in VCT? –Does anyone include C&T in an evidence-based bundle of prevention services? Yes (and in outstanding journals) –What is the impact for persons testing HIV+? VCT bundle impacts risk behavior and transmission rates –What is the impact for persons testing HIV-? Pre- and post-test counseling impacts risk behaviors and STD incidence rates in “real world” settings What if “typical” counseling is not up to standard of care? –Fix it or get another service partner to provide it, but don’t toss it