Studies indicate that certain populations are still at substantial risk of infection with HIV-1(1-6) and emphasize the critical need for more effective prevention strategies, including vaccines. Phase I and II clinical trials are being conducted to test the safety and immunogenicity of various candidate vaccines(7,8), and preparations are underway for placebo-controlled trials to test the efficacy of suitable vaccine candidates(9-13). Efficacy trials of candidate HIV-1 vaccines require study populations who are at high risk of infection despite other risk-reduction interventions, who adhere to study protocols, and who are willing to participate.
Project ACHIEVE (AIDS Community Health Initiative Enroute to the Vaccine Effort) was established to collect background data needed to prepare for HIV-1 vaccine efficacy trials in New York City among men who have sex with men. We have previously reported the study's enrollment strategies, level of risk of the population, and the population's adherence to study protocols(9). In this paper, we address several other questions that are relevant to preparing for efficacy trials. First, to what extent are gay men willing to participate in vaccine efficacy trials and are certain subgroups more or less willing? Second, to what degree is willingness sustained in the presence of presentations in the media of scientific progress and setbacks? In particular, in June 1994, the AIDS Research Advisory Committee of the National Institutes of Allergy and Infectious Diseases (NIAID) recommended not proceeding with vaccine efficacy trials of recombinant gp120 vaccine. Around the same time, infections among several vaccine recipients (i.e., "break-through" infections) in phase I and II trials were reported in the media with some confusion about whether the vaccine might have caused the infections. These events were widely reported in the media(14,15). Finally, what are the factors influencing willingness to participate in vaccine efficacy trials? To address these questions, we present data from the baseline visit of a cohort of HIV-1 antibody-negative, homosexually active men recruited in New York City.
Subjects and Screening Procedures
From October 1993 through January 1996, men at least 18 years old who reported sex with other men and were HIV-1 antibody negative were recruited into an HIV-1 vaccine preparedness study(9). Recruitment strategies were similar to those that may be used for a vaccine trial and included advertising in gay publications and other print media; presentations on cable television stations, radio stations, and computer bulletin boards; street outreach, and referrals from community-based organizations, health clinics, other research studies, and from participants themselves.
At an initial screening visit, pretest counseling was given, a blood specimen was drawn for HIV-1 antibody testing, and participants completed a brief self-administered questionnaire on demographics, history of sexually transmitted diseases (STDs), and risk behaviors. Two weeks later, after post-test and risk reduction counseling, all HIV-1 antibody-negative men were asked to enroll in follow-up, and those who enrolled completed a more extensive self-administered questionnaire. Men indicated the number of male sex partners in the previous 3 months, and to capture the potentially highest risk situations, they were asked to report the frequency of specific sexual behaviors in the previous 3 months with HIV-1 antibody-positive male sex partners or partners of unknown antibody status.
Questionnaires on vaccine trial willingness were completed at the enrollment visit. The questionnaires were developed by investigators involved in the Centers for Disease Control and Prevention Collaborative HIV Seroincidence Study conducted among gay men in other cities(11,16,17). Several areas related to vaccine trial willingness were addressed in these questionnaires. First, overall willingness to participate in a vaccine trial using a recombinant gp 120-type vaccine was assessed. Using an interviewer-administered format, a description of an HIV-1 vaccine efficacy trial was presented that included explanations of a recombinant gp120-type vaccine, placebos, blinding, follow-up needed to determine efficacy, and the possibility of vaccine-induced seropositivity. Participants were asked to rate their willingness to participate in the described vaccine trial on a four-point scale: definitely, probably, might be, or definitely not. Second, participants were asked to give reasons for their willingness or lack of willingness to determine potential motivators and barriers to participation in vaccine trials. These open-ended answers were coded by one of the investigators(B.A.K.) as motivators and barriers to vaccine trial participation. Third, participants were asked in an open-ended fashion about what other information they would need before making a decision about participating. These data were collected to identify needs for future education regarding preventive HIV-1 vaccines and vaccine efficacy trials. Fourth, to help understand participant decision-making, participants were asked whether they would seek outside advice about participation and from whom.
Analyses of categorical variables were conducted using Chi-square test of significance. The potential impact on participant willingness of the June 1994 media events about the decision not to proceed with vaccine efficacy trials and the breakthrough infections was assessed by comparing participants enrolled before the media event (October 1993 through May 1994), those enrolled during this period (June 1994 through August 1994), and those enrolled after the media event period(September 1994 through January 1996). Multivariate analysis of the association of willingness with selected covariates was conducted with multiple logistic regression. The covariates included were those significantly associated with willingness in univariate analyses (i.e., age, HIV-1 antibody testing history, advertising as a recruitment strategy, exchanging sex for money or drugs, enrollment period) and those found to be significantly different by enrollment period (i.e., age, race or ethnicity, self-perceived risk of HIV-1 infection, recreational drug use).
As of January 31, 1996, 698 men were enrolled in Project ACHIEVE. Demographic and behavioral characteristics of the men are presented in Table 1. The average age of the cohort was 31.5 years, and 32.1% were men of color. Most participants heard about the study through advertising or through participant referral. Most partners in the 3 months before the first visit were of unknown antibody status. Almost one half of the men reported receptive anal sex, with 33.7% of these men reporting the highest risk behavior, receptive anal sex without their partner using a condom or while having a condom break or slip off. Almost 60% of the men reported insertive anal sex, with 42.6% of these men reporting unprotected insertive anal sex.
Willingness to Participate in an HIV-1 Vaccine Efficacy Trial
At enrollment, 211 (30.3%) men indicated that they would definitely be willing to participate in an HIV-1 vaccine efficacy trial, 265 (38.0%) were probably willing, 162 (23.3%) might be willing, and 59 (8.5%) were definitely not willing. Men who were older, previously tested for HIV-1 antibody before enrolling in this study, recruited through advertising, and exchanged sex for money or drugs in the previous 3 months were significantly more likely to be definitely or probably willing compared with men without these characteristics; the definitely and probably willing categories were combined because associations with other characteristics did not differ (Table 2). Willingness was not associated with race or ethnicity, education, other modes of recruitment, self-perceived risk of HIV-1 infection, having an HIV-1 antibody-positive partner, engaging in unprotected receptive or insertive anal sex, history of an STD in the past year, or use of recreational drugs. In multivariate analysis, older age, previous HIV-1 antibody testing, and exchanging sex for money or drugs remained significantly associated with willingness (see Table 2).
Willingness and Media Events
Willingness to participate in vaccine efficacy trials was significantly lower among men enrolled during the media event period (57.0% definitely or probably willing) compared with those enrolled before (73.5%) or after the media event period (68.8%). We also found that study population characteristics changed over the three time periods. The study enrolled an increasing proportion of younger men, men of color, less educated men, and men reporting lower perceived risk of HIV-1 infection and less recent use of recreational drugs (data not shown). However, no differences were found by enrollment period in sexual risk behaviors, having an HIV-1 antibody-positive partner, or reporting of STDs in the past year. Controlling for population changes over the enrollment period in multivariate analysis, the men enrolled before and after the media event were significantly more likely to be willing to participate compared with men enrolled during the media event period (see Table 2).
Other Factors Influencing Willingness to Participate
Motivators to participate in vaccine efficacy trials were grouped into five major categories: altruism or helping the community, protection from infection, perceived minimal vaccine risk or burden, help in the research effort, and personal benefits. Typical statements for each of these categories can be obtained from the lead author. About 70% of men reported at least one motivator. Of the 30.2% not citing any motivators, most of them(69.1%) were definitely not willing or only might be willing to participant in vaccine trials. More than one half (56.6%) of the men cited altruistic reasons for their willingness to participate in vaccine trials, but only 11.5% of men gave protection from infection as a motivator. Perceived minimal vaccine risk or burden of participation was cited by 10.0%, to help in the research effort by 5.3%, and personal benefits by only 3.6% of the men. Men who were definitely or probably willing to participate in trials were significantly more likely to cite altruism, protection from infection, perceived minimal risk, and helping the research effort as reasons to participate compared with men who might be willing or were definitely not willing (Table 3). The proportion of men citing possible personal benefits did not differ between the two groups. No significant differences were found in the percentages of men citing any motivator by enrollment period (see Table 3).
The barriers to participation in vaccine efficacy trials were grouped into seven major categories: safety of the vaccine, vaccine-induced seropositivity, misperception that blinding means not knowing if a participant is infected, fear or mistrust of research or government, expectations that a vaccine trial would encourage high-risk exposures to test the vaccine, social risk, and doubt about vaccine research. Almost 60% of the men reported at least one barrier. Of the 41.3% of men not reporting any barriers, 88.2% indicated that they were probably or definitely willing to participate in vaccine trials. Concern about safety of the vaccine was cited by 36.2% of the men as a potential barrier to participation. Vaccine-induced seropositivity was a concern reported by 18.6%, misperception about the meaning of blinding by 10.5%, fear or mistrust of research or government by 6.6%, encouragement of high-risk exposures by 5.6%, social risk by 4.9%, and doubts about vaccine research by only 3.2%. Men who were not willing to participate were significantly more likely to cite concerns about safety, vaccine-induced seropositivity, fear or mistrust of research or government, expectations that a vaccine trial would encourage high-risk exposures, and doubts about the vaccine compared with men who were willing (see Table 3). Men who were enrolled during the media event period were significantly more likely to mention concerns about safety, fear or mistrust, and social risks compared with men enrolled before or after the media event period (see Table 3).
Information Needs and Sources of Advice
The most frequently mentioned information that participants indicated they would need before making a decision about vaccine trials was about safety of the vaccine(cited by 16.8% of participants), including information on side effects, the risk of getting HIV from the vaccine, interactions with other medications, effect if infected with other sexually transmitted diseases, and effect on partners. A similar percent (15.9%) of participants wanted more information about the vaccine itself, including results from previous animal studies and phase I and II trials and information on composition of the placebo, studies with other vaccines (particularly hepatitis B vaccine), ownership of the vaccine, and cost. The next most frequent request (9.3%) was for more information on the methods to differentiate a vaccine response from natural infection. Eight percent of participants wanted more information on the study design and logistics, including the effect of moving outside the study site area, timing of code breaking, explanations on importance of blinding, time commitment, and diversity of participants in the trial.
Most participants (73.0%) indicated that they would seek advice from someone else about making a decision to participate in an HIV-1 vaccine efficacy trial. Of those seeking advice, the person most frequently indicated was a friend (78.9%), followed by a physician (54.9%), partner(43.1%), counselor (32.7%), and family member(30.5%). Forty-one percent of participants reported that they would go to a community group or organization for information about vaccine efficacy trials. Of those, 71.1% would go to a community-based or service organization, 18.8% to an activist group, 12.5% to research studies, and 4.9% to health care centers.
Trials of vaccines to prevent HIV-1 infection require populations with a high incidence of infection who are capable of adhering to study protocols and who are willing to participate. In this vaccine preparedness study among gay men in New York City, 68% indicated willingness to participate in future efficacy trials. Other studies have found similar results (75%) among gay men recruited in other cities in the United States(16) and a range (44% to 52%) of willing participants among injection drug users(12,18).
Recruitment for a vaccine efficacy trial would be most efficient if the highest-risk individuals were also more likely to be willing compared with those at lower risk. Men reporting factors related to risk of HIV-1 infection in this population (i.e., unprotected receptive anal sex and lower educational level ) were, however, not significantly more likely to be willing than those not reporting these factors. Other studies have found that those who reported HIV-1 risk behaviors were significantly more likely to be willing compared with those not reporting risk behaviors(12,16). The reason for the differences in these findings is unknown. If a sizable proportion of willing participants are lower risk, significant resources will be needed for vaccine efficacy trials to screen enough potential participants to enroll those who are at high risk and willing.
We identified factors that are important to participants' willingness to join a vaccine efficacy trial. Altruism is the overwhelming motivator to participate in a trial. Only 12% cited possible protection as a motivator. Even though this number is relatively low, it is of concern, because the vaccine in a trial will be of unknown efficacy and one half of the participants will be given placebo, reinforcing the need for participants to understand that they may not be protected even though they are participating in a vaccine trial.
Although we did not have direct measures of the effect of the June 1994 events on willingness or data on whether or not participants were actually exposed to the media events, the NIAID decision not to proceed with a phase III efficacy trial of gp 120 vaccine and the confused reporting on the significance of the breakthrough infections among vaccine recipients in phase I and II trials were associated in time with a significant, although temporary, negative effect on willingness. The breakthrough infections may have contributed more to this effect, because men cited vaccine safety (overall, the most frequently cited barrier), fear or mistrust of research or government, and social risks (e.g., impact of participation on insurance or travel) more often during the media event period compared than in the other periods. For example, several questions arose among participants during this time. Did the vaccines cause infection? Did the vaccines cause some individuals to be more susceptible to infection when exposed? Did the infections indicate that the vaccine did not work at all? As future trials are planned(a phase II one is being considered using a prime with a canarypox virus vector vaccine and boost with a recombinant gp120 subunit vaccine ), substantial participant and community education will be needed concerning what is known and not known about the safety of the vaccines. For example, it is known that subunit and recombinant vector vaccines cannot cause HIV-1 infection, but it is unknown whether vaccination could enhance susceptibility to infection or severity of disease if infection occurs. Assurances of absolute safety can never be given for any vaccine, and somewhat limited safety data from phase I and II studies would be available at the start of an efficacy trial.
Investigators and staff involved in vaccine trials must also anticipate that breakthrough infections in phase I or II trials may have an impact on participant willingness, emphasizing the need to relay what specifically can and cannot be learned from breakthrough infections occurring in small-scale phase II trials. For example, phase II trials should expect breakthrough infections if they involve individuals at high risk, because no vaccine is 100% effective and even highly effective vaccines are not immediately protective. Moreover, the contribution of breakthrough infections in phase II studies to the understanding of failures to protect, of immunologic enhancement, and of correlates of protection will be limited given the small sample size and the absence of information on efficacy.
The second most frequently cited barrier was concern about vaccine-induced seropositivity, cited by almost one of five men. Tests to differentiate natural infection from vaccine response will be available for participants at study sites involved in vaccine trials. However, these tests may not generally be available outside of the study. One study, for example, reported that 9% of gay men participating in a vaccine preparedness study in three cities had been required to undergo HIV-1 antibody testing some time before enrolling in the study(20). The testing occurred most commonly in conjunction with insurers and the military, but also with immigration agencies, private employers, and the criminal justice system. These data indicate that outside testing should be expected during a trial. Study sites need to have adequate staff and time to help participants resolve the potential misinterpretation of test results and to help eliminate this barrier to vaccine trial participation.
The need for widespread community education before and during a vaccine trial is also underscored by the large percentage of men who would seek outside advice about their decision to participate in a vaccine trial from friends, partners, family members, and members of the medical and social service communities. Educational efforts need to include, for example, information about the vaccine, trial design concepts, methods to differentiate vaccine response from natural infection, implications of vaccine-induced seropositivity, what is known and unknown about safety, and other risks and benefits of participation.
The results of this study are encouraging but are somewhat limited in their ability to predict eventual success in recruiting volunteers for trials. The vaccine trial scenario presented to the participants was hypothetical and concerned a trial of gp120-type vaccine, a vaccine that is not likely to be tested by itself in efficacy trials in the United States. Although this study provides an estimate of the percentage of persons willing to participate and the factors influencing willingness, we do not know the proportion of men who may be willing to participate in a trial of another type of vaccine or who may volunteer when presented with an actual trial. As new vaccine concepts are identified that have the potential to advance to efficacy trials, assessments will need to be made to identify factors specific to the new vaccine concepts that may affect willingness. For this study, we did not measure the level of understanding of specific vaccine trial concepts, a factor that may be important in determining willingness and the effect of media events on willingness.
This study of almost 700 gay men in New York City indicates that a substantial proportion are willing to participate in future trials of vaccines to prevent HIV-1 infection. Several concerns raised by participants, particularly about safety of the vaccine, vaccine-induced seropositivity, and possible protection from the vaccine, should be the focus of participant and community education and must be addressed in the design and implementation of future trials.
Acknowledgments: The authors thank the participants, the staff of the Laboratory of Epidemiology, and the Project ACHIEVE Community Advisory Board for their efforts and contribution to this study, and we are grateful to Kathleen MacQueen, Ph.D., for her review and comments. This work was supported by grants R29MH49968 and R01AI3806 from the National Institutes of Health and grants from the New York Community Trust.
1. Van Griensven GJP, Koblin BA, Osmond D. Risk behaviour and HIV infection among younger homosexual men. AIDS
2. Dean L, Meyer I. HIV prevalence and sexual behavior in a cohort of New York City Gay Men (aged 18-24). J Acquir Immune Defic Syndr
3. Hays RB, Kegeles SM, Coates TJ. High HIV risk-taking among young gay men. AIDS
4. Stall R, Barrett D, Bye L, et al. A comparison of younger and older gay men's HIV risk taking behaviors: the Communication Technologies 1989 cross-sectional survey. J Acquir Immune Defic Syndr
5. Lemp GF, Hirozawa AM, Givertz D, et al. Seroprevalence of HIV and risk behaviors among young homosexual and bisexual men: the San Francisco/Berkeley Young Men's Survey. JAMA
6. Osmond DH, Page K, Wiley J, et al. HIV infection in homosexual and bisexual men 18 to 29 years of age: the San Francisco Young Men's Health Study. Am J Public Health
7. Walker MC, Fast PE, Graham BS, Belshe R, Dolin R. Phase I/II preventive vaccine trials: conference summary. AIDS Res Hum Retroviruses
8. Belshe RB, Graham BS, Keefer MC, et al. Neutralizing antibodies to HIV-1 in seronegative volunteers immunized with recombinant gp 120 from the MN strain of HIV-1. JAMA
9. Koblin BA, Taylor PE, Avrett S, Stevens CE. The feasibility of HIV-1 vaccine efficacy trials among gay/bisexual men in New York City: Project ACHIEVE. AIDS
10. Hoff R. Progress towards readiness for efficacy trials in the United States: conference summary. AIDS Res Hum Retroviruses
11. Buchbinder SP, Douglas JM, McKirnan DJ, Judson FN, Katz MH, MacQueen K. Feasibility of human immunodeficiency virus vaccine trials in homosexual men in the U.S.: risk behaviors, seroincidence and willingness to participate. J Infect Dis
12. Meyers K, Metzger DS, Navaline H, Woody GE, McLellan AT. HIV vaccine trials: will intravenous drug users enroll? Am J Public Health
13. Nelson KE, Vlahov D, Galai N, Astemborski J, Solomon L. Incident human immunodeficiency virus (HIV) infections in a cohort of injection drug users in Baltimore, Maryland. AIDS Res Hum Retroviruses
14. Five in big AIDS vaccine test develop virus. The New York Times
May 29, 1994, Section 1, p. 24.
15. Panel rejects wider testing to develop AIDS vaccine. The New York Times
June 17, 1994, Section 1, p.8.
16. Bartholow BN, MacQueen K, Douglas J, Buchbinder S, McKirnan D, Judson F. The decision to participate in HIV vaccine efficacy trials: an assessment of changing willingness among homosexual men. Abstracts of the XI International Conference on AIDS, Vancouver, Canada, July 7-12, 1996.
17. MacQueen KM, Buchbinder S, Douglas JM, Judson FN, McKirnan DJ, Bartholow B. The decision to enroll in HIV vaccine efficacy trials: concerns elicited from gay men at increased risk of HIV infection. AIDS Res Hum Retroviruses
18. Vlahov D, Astemborski J, Solomon L, Galai N, Basarab L, Nelson K. Interest in HIV vaccines among injection drug users in Baltimore, Maryland. AIDS Res Hum Retroviruses
19. Fauci A. Special plenary session. Eighth annual meeting of the National Cooperative Vaccine Development Group for AIDS. Bethesda, MD, February 11-15, 1996.
20. MacQueen KM, Buchbinder SP, Douglas JM, Judson FN, McKirnan DJ, Bartholow BN. Required HIV antibody testing, social risk and HIV vaccine efficacy trials. AIDS Public Policy J