Abstract
Dear Editor,
One of the limitations in conducting phase 1 preventive HIV vaccine trials in a resource-rich country is the difficulty in recruiting volunteers. There are currently 34 million people living with HIV and each year 2.7 million people become newly infected and 1.8 million people die of AIDS. 1 Preventive HIV vaccine investigation is a critical path in developing a long-term strategy to control the worldwide HIV epidemic. 2 Research efforts over the last two decades to develop a preventive HIV vaccine have showed that this is a particularly difficult objective. To date, over 200 phase 1–3 HIV vaccine trials have been conducted or are ongoing but only 3 have reached phase 3. Only one HIV vaccine in a phase 3 study demonstrated a modest efficacy when compared with placebo. 3 Therefore, multiple trials are needed to find an effective HIV vaccine and greater efforts for any promising immunogen to reach phase 1 clinical trials has been proposed. It is estimated that over 35,000 volunteers will be required per year for phase 1–3 HIV vaccine trials worldwide to achieve this goal. 2 Several studies have focused on the hypothetical willingness-to-participate 4 or enrollment efficiency 6 and have identified the most frequent reasons cited for declining HIV trial enrollment. Despite the growing literature on this issue, there are few data on barriers to participation in low risk volunteers willing to participate in phase 1 preventive HIV vaccine. The first phase 1 HIV vaccine clinical trial conducted in Spain, RISVAC-02, needed to recruit 30 HIV-uninfected volunteers in Spain.
RISVAC-02 was a double-blinded, randomized placebo-controlled phase 1 HIV vaccine trial conducted in two Spanish clinical centers in Madrid and Barcelona. The study was approved by the institutional ethical review board and by the Spanish Regulatory Authorities. Details of the design of the study are published elsewhere. 6,7 Inclusion criteria were age between 18 and 55 years, being at low risk of HIV-1 infection, agreeing to use an effective method of contraception from 14 days prior to the first vaccination until 4 months after the last immunization, and consent to HIV and sexually transmitted disease (STD) testing. The exclusion criteria were pregnancy or lactation, serious clinical problems, receiving a live attenuated vaccine 60 days before or any vaccine 14 days before screening, receiving any blood product or gamma-globulins 4 months before screening, participating in any other clinical trial 30 days before enrollment, history of several local or systemic vaccine reactions, and HIV/hepatitis B virus (HBV)/hepatitis C virus (HCV)/syphilis infection. We analyzed demographics, recruitment source, sexuality, motivation to participate, or discontinuation of all healthy volunteers from the first contact with us to the baseline RISVAC-02 study.
Different strategies and various approaches for recruitment of potential volunteers were used. Advertisements in mass media such as national and local newspapers, medical journals, HIV-related magazines, and TV interviews were conducted. E-mails were sent to different organizations: HIV/AIDS NGOs, gay community associations, universities, hospitals, and others social networks. Informative posters were distributed in the cities of Madrid and Barcelona. Individuals who responded and provided a telephone number or an e-mail address were contacted and given further information. If they were interested we invited them to participate in a pre-screening interview were they answered a survey that was conducted by trained nursing staff in a personal interview. Demographics, clinical history, lifestyle and risk factors, how they found out about the study and reasons for participating were obtained through an open/closed-ended questionnaire. If individuals were eligible and interested in participating, they were booked for a screening appointment. Informed consent was obtained from all the subjects. Those patients who agreed and fulfilled the inclusion criteria were enrolled.
SPSS version 17 (SPSS, Inc., Chicago, IL) was used for analysis. Descriptive statistics and cross-tabulations were generated for all the variables of interest. Continuous variables were compared between groups with the Student's t test or the non-parametric Mann-Whitney U test. Categorical variables were compared between groups with the χ2 or Fisher's exact tests. p value of ≤ 0.05 was considered significant.
A total of 356 volunteers who responded to our campaign and sent a telephone number or an e-mail address were contacted and invited to a first interview. Seventy-five subjects did not respond to this first call, 22 did not fulfill the inclusion criteria due to high-risk sexual behavior, and 88 refused to participate before the prescreening interview (67 disagreed with the compulsory use of condoms with nonregular partners and 21 argued that the trial was too time-consuming). The characteristics of refusing volunteers were compared with those who agreed to perform the prescreening (Table 1). Patients who refused to perform the prescreening were older (median age [interquartile range {IQR] 31.5 [24–42]) than those who agreed (29 [24–36]; p=0.002). There were no significant differences in gender. In addition, we analyzed if the way in which the investigation team made the first contact with the volunteer could influence in the decision to take part in the study. Half or more of the volunteers who were contacted personally or by telephone performed the prescreening, while only a 26% of those who were contacted by e-mail decided to participate (p<0.0001).
n (%): age was calculated based on data of 281 volunteers who answered the call of investigation team (n=281).
Method in which the investigation team made the first contact with the volunteer (n=299).
IQR, interquartile range.
We prescreened 171 of 356 (48%) individuals who contacted us. Eighty-eight declined to come to the hospital for the screening, 8 did not meet the inclusion criteria, and 80 stated that they were afraid of the vaccination (of being infected or the side effects). There were no differences in any characteristic between those volunteers who did not agree to perform the screening or did not fulfill criteria versus those who did agree (data not shown).
The final screening included 83 of the 356 (23%) individuals who responded to our initial campaign. Of these, 18 did not fulfill criteria (mainly due to medical conditions or blood tests anomalies), and 24 who fulfilled the inclusion criteria and agreed to participate refused at the last minute because they were worried about the side effects or being infected with the vaccine. Of the total 356 subjects who contacted us, 41 (11%) were eligible and of these we enrolled 30. The characteristics of those who refused or did not fulfill the inclusion criteria were compared with those who agreed to be vaccinated (Table 2). Refusing volunteers were more likely to be older (median age [IQR] 32 [25–39] versus 28 [22–33] in agreeing to participate volunteers, respectively, (p=0.09)], women (65% versus 41% of men, p=0.045), those who had not been to college (69% vs 41% who attended college, p=0.006), migrants (67% versus 45% Spanish, p=0.08) or participated for economic reasons (70% versus 44% of altruist volunteers, p=0.04). Occupation was not a barrier for participating. In a multivariate analysis, gender (women) and educational level (not having been to college) were the independent factors associated with not participating in the clinical trial (Table 2).
p univariate analysis.
p multivariate analysis odd ratio (95% confidence interval).
n (%).
The characteristics are collected only of those volunteers who came to screening (n=83).
IQR, interquartile range; OR, odds ratio.
Our study provides information about the efficiency of recruiting initiatives and the reasons why healthy low-risk volunteers refuse to participate in the first phase 1 preventive HIV vaccine clinical trial conducted in Spain. Previous studies that analyzed these barriers worked mainly in high risk or developing countries population. 8 –10 Regarding the recruitment strategies for our vaccine trial, the most successful were the mass media, Internet advertising, word-of-mouth and posters located in the School of Medicine and the University Hospital, while NGOs and contact by e-mail led to the fewest enrolled participants. This contrasts with other phase 1/2 trials conducted in the United Kingdom, 6 Tanzania, 10 and Thailand, 11 where HIV/AIDS NGOs, health and social networks recruited more volunteers than in our study. In reference to the analysis of barriers to recruitment in our study, the main reasons for not participating were rejecting condom use with nonregular partner (19%) and fear of the HIV vaccine (29%). While rejecting condom use was the main barrier in the first interview with the volunteers, fear of the vaccine was the main cause in the prescreening and screening period. This last limitation contrasts with clinical vaccine trials against other diseases. 4 In contrast with what is reported in high risk volunteers, 8 concerns about the time commitment was the reason for not participating in only a minority of volunteers. In the multivariate analysis, refusing volunteers were more likely to be women and those who had not been to college. In a univariate analysis, other barriers to participation were older age, migrants, and those who had an economic reason for participating. These data are in line with other clinical trials with high-risk volunteers that found that those who had been to college were less likely to refuse enrolment. 8 In addition, in this same study, women and migrants volunteers appeared more likely to refuse participation, 8 at least in the univariate analysis. Although volunteers who participated for economic reasons were more likely to refuse to be enrolled, the majority of our volunteers reported altruism and personal contact with HIV-infected people as the main reasons for participation, as has been reported in other studies. 5,8,11 In contrast, in others studies developed in resource-poor countries, there seem to be other major key motivator to participate in HIV vaccine trials. The hope of being protected from HIV and access to regular HIV counseling-testing and healthcare are the more frequent reasons to enroll in these countries. 9 In addition, it has been reported that willingness and motivation to participate in HIV treatment or prevention trials may not be uniform in different ethnic minorities. 12
Although in our study the recruitment strategies provided a sufficient number of individuals interested in participating, a high proportion declined after they received the information. This low number of volunteers agreeing to participate is similar to that observed in a similar phase 1/2 clinical trial conducted in the United Kingdom. Gray et al. 5 reported that only 16% of volunteers agreed to participate.
Considering the importance of volunteering in future HIV vaccine trials, efforts to improve recruitment and eliminating the barriers to participation in these studies should be a priority and its understanding could help to improve future enrollment in HIV vaccine trials.
Footnotes
Author Disclosure Statement
No competing financial interests exist.