What motivates or demotivates injecting drug users to participate in hypothetical HIV vaccine efficacy trials? A qualitative study from urban Tanzania

Background: HIV vaccine efficacy trials require the active participation of volunteers who are committed and adherent to the study protocol. However, information about the influence of Injecting Drug Users (IDUs) to participate in HIV vaccine efficacy trials in low-income countries is inadequate. The present study explored the factors that motivate or hinder IDUs from participating in HIV vaccine efficacy trials in Dar es Salaam, Tanzania. Methods: A qualitative descriptive study design was employed among IDUs at Muhimbili National Hospital (MNH). A purposeful sampling technique was used to recruit the participants. Three (3) focus group discussions (FGDs) and 10 In-Depth Interviews (IDIs) were used to collect the data. The data from participants were audio-recorded, transcribed, and analysed using the content analysis approach. Findings: The participants reported that altruism and the desire to reduce risks of HIV infection were the motivators to participate in hypothetical HIV vaccine trials. In addition, participants reported to consult close relatives towards motivation to participate in the vaccine trial. In contrast, the perceived fear of vaccine side effects, lack of information about HIV vaccine studies, and HIV-related stigma towards participants were described as barriers to participate in the HIV vaccine trials. Conclusion: Participation in a hypothetical HIV vaccine trial among IDUs is influenced by positive and negative factors. Actual recruitment plans could be made through a better explanation of HIV vaccine trials, the expected individual and collective benefits associated with the trials. Community involvement and sensitisation is likely to enhance participation in future HIV vaccine trials in Tanzania.

decreased from 5.1% in 2012 to 4.7% in 2016 6 , the prevalence of HIV among IDUs is still high at 42%. 7 Factors associated with an increase in HIV infection among IDUs include age, sharing of syringe and needles, low level of education, and history of drug overdose. [8][9][10] Various harm reduction programs such as methadone assisted treatments have been developed as one of the strategies for combating HIV infection among IDUs. 11,12 Additionally, the development/ deployment of an Integrated Methadone and Antiretroviral Therapy (IMAT) have shown to be effective in decreasing HIV prevalence among IDUs in Tanzania. 13 However, a multi-lateral approach including behavioural interventions, Treatment as p-BACKGROUND S ince the beginning of the epidemic, injecting drug behaviours are the key facilitators of HIV transmission 1 . In 2017, 15.6 million Injecting Drug Users (IDUs) aged 15to 64 years were identified worldwide, of whom 17.8% were living with HIV 2 . A substantial geographic variation in the prevalence of HIV infection across all countries has been noted. The number of IDUs in South-East Asia and East Asia has increased 3 . Evidence shows an increase in injection drug use and associated HIV infections in Sub-Saharan Africa, the region with the highest global HIV prevalence 4,5 .
Although the overall HIV prevalence in Tanzania has revention (TasP), pre-exposure prophylaxis, and vaccines is more likely to lead to effective infection control than deployment of a single approach. [14][15][16] The efforts dedicated to discovering an HIV vaccine have a long history since HIV was first described in the early 1980s. The potential to slow the HIV epidemic and save lives fuels the development of an effective vaccine. 17 Clinical trials require the active participation of volunteers who are committed and adhere to the study protocols. Vaccine efficacy trials take years to monitor the effectiveness and side effects of any new vaccine. The process is more challenging for HIV vaccine efficacy trials because it involves socio-behavioural issues that affect the volunteers' participation. 18,19 The successful development and implementation of a safe and efficacious vaccine greatly depend on the extent to which the at-risk populations are motivated to participate in HIV vaccine research.
Studies in high-income countries report financial reimbursement, reduction of risk behaviours, and support from researchers as the motivating factors among men who have sex with men (MSM) and female sex workers (FSW). [20][21][22] However, the social pressure of significant others, perceived lack of vaccine safety, and logistical concerns have also been identified as barriers to actual participation in the trials. 23 Also, misconceptions about HIV vaccine trials, personal and social risks, and costs are reported to hinder participation in hypothetical HIV vaccine trials in many high-income countries. 24 However, there is a paucity of information about motivations and barriers for participating in HIV vaccine trials among injecting drug users in low-income countries.
A recent study in Kenya reported a desire to receive healthcare and information about HIV were the motivating factors to participate in an HIV vaccine trial among MSM and FSW 25 . Furthermore, a review of the literature reported that retention and sexual disinhibition were the main socio-behavioural challenges for HIV vaccine efficacy trials in Sub-Saharan Africa. 26 However, none of these studies reported motivations and barriers to participating in HIV vaccine trials among IDUs in lowincome countries. Thus, this study describes the reasons behind the willingness of IDUs to participate in the HIV vaccine trials and provides recommendations for future studies.

MATERIALS AND METHODS Design
A qualitative descriptive study design was employed. We applied this approach to better understand the fundamental motives influencing participation in HIV vaccine trials among IDUs. This design was also used to raise awareness and increase insight into the best ways of conducting HIV vaccine trials in the study population. 27 In addition, the approach allowed interaction between the authors and the participants which increased trust among each other.

Setting
The study was conducted at the Methadone Clinic in Muhimbili National Hospital (MNH), Dar es Salaam. The site is 1 of the 3 Medication-Assisted Treatment clinics (MAT) for IDUs who are trying to stop the use of heroin in the region. The MAT clinic at MNH was the second to be opened in Sub-Saharan Africa after Mauritius 28 . The clinic is well staffed throughout the year to accommodate IDUs to come to the clinic daily to receive a Directly Observed Dose (DOT) of liquid methadone. The maximum duration of treatment to recovery for most IDUs was 5 years, although some finished treatment in 3 years, depending on medication adherence and effective use of counselling services. The authors were not part of the MAT clinic staff and had no affiliation with the hospital where the participants were receiving treatment services.

Population and Participants
This study was based on a high-risk population attending the MAT clinic. We selected the IDUs for their relatively high risk of HIV infection compared to other key populations in Tanzania. 29 The risk of IDUs was largely due to a history of needle and syringe sharing practices and that was suitable for the HIV vaccine efficacy trial.

Inclusion and Exclusion Criteria
The study included participants who were injecting drugs, physically and mentally stable, aged 18 and above, and HIV negative. Both males and females were included. IDUs who were physically and mentally unstable and those that could not communicate appropriately were excluded.

Sampling Procedure
We used purposive sampling to recruit the participants as proposed by Palinkas et al. 30 The decision to use this sampling technique was based on information-rich participants who could be willing to share their concepts and views about HIV vaccine studies. This sampling technique allowed the authors to obtain adequate information related to the phenomena of interest. A trained research assistant selected the participants from the MAT clinic.

Sample Size
The sample size was determined using the principles of saturation. That is, we terminated sampling when no new information was obtained as proposed by Hennink et al. 31

Data Collection Briefing Sessions
Before the commencement of data collection, we provided a brief overview of HIV vaccine efficacy trials to the participants, including the nature of vaccine material, how it would be administered, what to expect if the participant is enrolled in the study, and the issues related to vaccine-induced seropositivity and how it could be handled. This information promoted awareness of HIV vaccine efficacy trials, since some participants had never heard about these trials before.

Focus Group Discussions (FGDs)
FGDs were used as the main data collection method. We first conducted FGDs to identify the main recurrent ideas. This method provided an opportunity to interact with the participants and explore their understanding of HIV vacc-ine trials. The group discussions were conducted immediately after the participants had taken their daily methadone doses to ensure maximum cooperation. The group size ranged between 6 and 8 participants.
We used homogeneous FGDs in the sense that male and female groups were conducted separately (2 groups and 1 group of males and females respectively). This approach allowed the free expression of ideas and views among the participants. A discussion guide was used to collect the data. The questions in the guide comprised of core questions from previous studies. 32-34 which were improved further through pilot testing. 35 The questions elicited the general views about the risk of HIV infection, the level of motivation to adhere to treatment protocols, and any perceived barriers toward HIV vaccine trials.
The following questions were asked during the group discussion: 1. What are your views on the risk of HIV infection due to injecting drugs? 2. What are your views if you are asked to participate in HIV vaccine efficacy trials? 3. What would motivate you to participate in an HIV vaccine efficacy trial? 4. How would the people you live with influence your decision to participate in HIV vaccine efficacy trials? 5. What would hinder you from participating in HIV vaccine trials? These were followed by specific probing questions to obtain additional information or clarification. The first and second authors reviewed the discussion guide after the first FGD to include more emerging themes. The FGDs lasted between 45 and 60 minutes. During the group discussion, the first author moderated the discussion while a research assistant took notes and controlled the external environment. Data saturation was reached when no new information was obtained after a new group was added as guided by Hennink et al 31 resulting in 3 focus group discussions.

In-Depth Interviews (IDIs)
To complement the data from FGDs, we conducted IDIs with IDUs who participated and who did not participate in the FGDs. The individual interviews were conducted in a quiet, well-lit room in the hospital premises away from the MAT Clinic. This was important to ensure safety and maximum cooperation from the participant as well as to making the environment natural. 36 The IDIs elicited more descriptive information on HIV vaccine trial participation. The first author conducted all interviews. The IDIs used the same questions that were applied in the FGDs. However, some specific probing questions were geared towards individual participants such as: How would your schedule fit in with a proposed HIV vaccine trial? What are your views on the availability of an effective preventive HIV vaccine? After the 10 th IDI, we reached information saturation. Of these 10 interviews, 4 participants (3 males and 1 female) had participated in FGDs. These participants were invited to take part in the IDIs because during FGDs, the first author observed that they were hesitant to share their views; however, they appeared to have some ideas. The intervi-ews lasted between 30 to 40 minutes. Both FGDs and IDIs were conducted in Kiswahili, the language spoken by most people in Tanzania, and was well understood by the participants. Data were audio-recorded.

Data Analysis Focus Group Discussions
Data analysis started as soon as the first FGD was completed. This guided the subsequent levels of questions and probes in the discussion guide. The research assistant (a nurse) transcribed verbatim of all the audio-recorded data in the Kiswahili language and typed it into the Microsoft Word computer program. The first and second authors checked the transcripts against the audiorecorded data to ensure the correctness of the transcribed data. The unit of analysis was a whole transcript. All files of transcripts were transferred to NVivo 11.0 software (QSR International, Melbourne, Australia) for coding and organisation. The texts were analysed in the native language of the participants. The first 2 authors read the transcripts iteratively and thoroughly to immerse themselves in the data. Interesting content areas were coded, as guided by content analysis principles. 37 We used inductive coding whereby codes were developed from the data using phrases or terms utilised by the participants themselves. In this way, we were able to stay close to the data, mirroring what is actually in them. The coding of the contents continued throughout the rest of the documents. Reflecting on the objective of the present study, code classifications were created containing defined attributes related to the topic of interest. After we had coded all information and organised it into a manageable format, all codes were shared between the first 2 authors for discussion, and the consensus was reached on the coded information. The process of sharing the codes helped to improve the credibility of the coding system and organisation 38 . We then continued reading and abstracting the contents into more specific ideas that were mutually exclusive of each other. In other words, the text was divided into meaning units that were condensed, abstracted, and eventually labelled with codes. Coding continued for all transcripts to form categories and themes. (Table 1).

In-Depth Interviews
The same analysis process was carried out on the IDIs data. Following the analysis, we checked the FGDs' themes and categories to ascertain the new information obtained from the individual interviews. This contributed to an enhanced understanding of the participants' perspectives on HIV vaccine trial participation. Representative ideas and quotes from all IDIs were identified for each FGD theme and category. A new category emerged from IDIs and was reported in addition to the FGDs' themes. The whole text was translated into English. The translation of the text was conducted according to Brislin 39 .

Ethical Consideration
Ethical clearance was obtained from the Institutional Review Board at Muhimbili University of Health and Allied Sciences (MUHAS) with Ref. No. 2017-06-028/ AEC/Vol.XII/85. A permission letter was obtained from the Executive Director of Muhimbili National Hospital (MNH). The first author reviewed the informed consent form and explained to the potential participants the principles of voluntary participation, anonymity, and the right to withdraw from the study at any time without losing any benefit from the health services at the clinic. Written consent was obtained from all participants before data collection. All potential participants consented to the audio-recording during the discussions and interviews. We ensured the anonymity of the information provided by using codes instead of their names in all documents. Participants were reimbursed Tshs 4,000 (equivalent to 1.76 USD) for transportation and their time.

FINDINGS
Characteristics of Participants 28 participants participated in the study as follows: 18 participated in FGDs only, 6 participated in IDIs only and 4 participated in both FGDs and IDIs. The following section reports the socio-demographic characteristics of the FGDs and IDIs' participants separately.

Focus group discussion
The ages of the 22 participants ranged from 19 to 50 years, with a mean age of 37.2 (SD=7.8). Of these 22 participants, 16 were males. Most of the participants had primary education levels. 10 of the participants were self-employed, performing activities that enabled them to obtain an income. 18 of the participants were single (Table 2a).

In-depth interview:
The ages of the 10 participants ranged from 25 to 44 years, with a mean age of 32.6 (SD=5.8). Of these 10 participants, 6 were males. Most of the participants had primary education and half were self-employed. 8 participants were single (Table 2b).

Themes and Categories
Motivations and barriers to participating in HIV vaccine trials were the 2 themes identified in this study. Both themes were derived from the FGDs. 3 categories are reported as motivators and 3 categories as barriers. Among the 6 categories from the themes, 5 categories were derived from FGDs and IDIs while 1 category emanated from IDIs only. The findings are presented together for both FGDs and IDIs (Table 3).

Motivation to Participate in HIV Vaccine Trials
Participants reported different factors that would drive them to participate in the trials. Altruism, the desire to reduce the risk of HIV infection, and social support were the main motivators for IDUs to participate in HIV vaccine trials as described in the following;

Altruism
Participants expressed a desire to participate in HIV vaccine trials with the hope that a successful vaccine would benefit many people in Tanzania and other countries. Also, they hoped that their participation would not only result in an effective HIV vaccine but might also encourage others to be vaccinated against HIV infection as expressed below: "  if their community could see the results of their participation, it might influence others to participate in future vaccine trials study as expressed below: "If we reach the community, we will inform them that we were involved in the vaccine trials and that the trials have provided the best answers; the trial has been achieved, and its advantages are the same as you see us. These trials are both positive and harmless, and we volunteered as pioneers." (FGD2 participant 12, male, age 38) Participants expressed interest in being in a vaccine trial to obtain the positive results of a potential vaccine, knowing that they contributed to its development. Their main satisfaction was to know that other people might also be motivated to join the study. This factor was one of the strategies they thought they could use to attract their colleagues as described below:

Lack of information about HIV vaccine studies
Participants expressed concern about the lack of knowledge about HIV vaccine trials. Lack of information was a hindrance to participants to volunteer in a vaccine trial if one was available. Some participants stated that they had heard about an HIV vaccine trial before this study, but did not understand what it was all about. This lack of knowledge discouraged them from participating in the trial. They asserted that they would allow to be recruited if they understood more about the nature of the vaccine and how it works.  protect from infection infection trials is will give information that helps to protect me from infection"

Stigma towards HIV vaccine trial participants
Stigma towards HIV vaccine trial participation was prominent during IDIs. This concept did not emerge during the FGDs. In IDIs, many participants reported fears of being labelled and criticised by community members and people around them, including relatives. They stated that people might point fingers at them if they participate in the HIV vaccine efficacy trial because they believed that the participants are infected with HIV, as verbalised by one of the participants below. "…so, everyone will be pointing his/her finger at you because you participated in the HIV vaccine trials. The community can discriminate against you because they think you already have HIV, which has no treatment" (IDI, participant 1, female, age 27) Other participants were worried about social isolation in the community when participating in the HIV vaccine trial. They asserted that the community members would Other participants argued that drug users had difficulty understanding information about HIV vaccine trials. They referred to the hardship experienced during the recruitment of participants for the methadone clinic. They reported that drug users could not understand the importance of methadone and continued to inject illicit drugs even after educating them. One peer educator who used to recruit drug users from the street and educate them about the importance of attending the clinic and using the methadone treatment remarked: "We followed and educated them about the importance of using methadone at the clinic. We told them that the drug is free but they were so difficult to understand. …I'm not sure if they can understand and be motivated to engage in the HIV vaccine trial." (FGD3, participant 7, female, age 24) Another participant opposed the ideas and added: "…we drug users are not a problem but when awareness is given, then people will understand. What is needed is just information and education. The only thing we ask for is education. People will be motivated to participate if given appropriate education" (FGD3 Participant 6, Female, shun and even isolate them if the vaccine would not be effective: "I would like to volunteer to participate and be given the vaccine material to see if it works. However, if I get the vaccine material and it does not work, this will hurt me. …and the community will look at me negatively and even isolate me." (IDI, participant 1, female, age 36) Overall, the participants reported HIV related stigma towards HIV vaccine trial participants as an obstacle that needed to be eliminated through community involvement and education. They expressed that community sensitisation using various education materials such as fliers could help to promote community awareness and thus decreasing stigma towards HIV vaccine trial participants.

Comparison of findings from FGDs and IDIs
The  19,40 This indicates that altruistic reasoning plays an essential role in motivating participants to join HIV vaccine trials in Tanzania. Similar reasoning is also reported in Kenya whereby the willingness to participate in HIV vaccine efficacy trials was driven by various forms of altruism. 41 Further evidence to support altruism as the motivating factor has been reported in the USA, the Netherland, and Canada 42,43 . Given the participants' responses in this study, our findings suggest that the part-cipants' lives might have meaning and purpose because of participation in an HIV vaccine trial, particularly if it yields a positive outcome. The desire to reduce the risk of HIV infection as a motivator for participation in HIV vaccine trials can be attributed to the high-risk behaviours that participants had experienced before joining the study. The participants are greatly affected by their memories of sharing contaminated needles/syringes and unsafe sexual behaviours as described in the Health Belief Model. 44,45 This may have motivated them to participate in HIV vaccine trials to reduce HIV infection among themselves and the community at large. A multi-site study in the US, Canada, and the Netherlands revealed similar findings that volunteers were motivated to participate in the HIV vaccine trial to reduce risk behaviour. 22 Such findings were also reported in Philadelphia where protection from HIV infection was the motivator to participate in the HIV vaccine trial. 46 The desire to reduce the risk of HIV infection may also be accounted for by the knowledge of harm reduction program that participants were involved in before the current study. 12,28 Thus, participants perceived HIV vaccine trials as one of such programs for HIV risk behaviours reduction. Therefore, intensive training is needed to differentiate between HIV vaccine trials and other risk behaviour reduction programs during the implementation of actual vaccine trials. Likewise, the motivation to participate in an HIV vaccine trial for reducing HIV infection was reported in a phase I/II HIV vaccine trials study among police officers. 40 Given the experience of participants in our study, motivation to participate in vaccine trials to reduce high-risk behaviours is an important factor to consider when planning for future HIV vaccine trials among IDUs.
In the context of social support, IDUs demonstrate the key abilities needed to make meaningful decisions about HIV vaccine trial participation. Similarly, previous HIV vaccine studies in Tanzania show that social support plays an essential role in HIV vaccine trials. 18,47 In Tanzania, the reported importance of involving close people when making decisions may be described by the socio-cultural experience of household decisions among couples 48 and the type of family patterns. 49 This is similar to the study conducted in the United States which reported consultation with other people was one of the factors in the decision-making process among adolescents. 50 This reinforces our understanding that information sharing is important for informed decision-making. It also implies that participants have a meaningful relationship with other people and value their input when making difficult decisions. The findings in our study also correlate with findings from South Africa, whereby the ultimate decision to engage children in HIV vaccine trial participation rested on their mothers after they had shared information with their significant others. 51 However, further research is needed in this area to explore the social and behavioural characteristics of IDUs who can be motivated to participate in HIV vaccine trials based on consensus from significant others.

Barriers to participating in HIV vaccine trials
The reported fears of vaccine side effects as a barrier to participation in HIV vaccine trials may be contributed by the lack of proper information about the nature of vaccine materials. Similarly, the phase I/II HIV vaccine trials among police officers in Dar es Salaam, Tanzania, reported fears of vaccine side effects as one of the reasons to decline from participating in an HIV vaccine trial. 52 Likewise, a study in India showed that participants feared vaccine-induced HIV infection. 53 Thorough and accurate information related to the vaccine is needed for potential HIV vaccine trial participants. Expanding education about HIV vaccine trials may help to decrease misperception and misinformation. Promoting awareness and comprehensive education for participants about what to expect during the trial is crucial for effective HIV vaccine trial participation.
Lack of information about HIV vaccine trials can be described by the fact that research findings have not been adequately disseminated among the population of interest. Dissemination of HIV vaccine-related information is important for raising awareness in the participating community. 54 A previous study in Uganda reported improved communication between participants and research staff that created a sense of community ownership among participants. 55 Nevertheless, a study among transwomen in 4 cities of the USA revealed that having either no exposure or limited exposure towards HIV vaccine trials which was translated as receiving inaccurate information from the laypeople 56 is a barrier to participation. The findings in our study indicate that the recruitment of prospective participants in an HIV vaccine efficacy trial requires sufficient education to address misperceptions. Such education may potentially decrease barriers towards participation in the vaccine trials. In other words, for effective HIV vaccine efficacy trial participation among IDUs, participants must have a broader understanding of the nature and procedures of the HIV vaccine trials.
As revealed in the present study, HIV-related stigma may prevent participants from volunteering for HIV vaccine efficacy trials. The negative reactions from their communities have greater impacts on the decision to participate. Such negative reactions and their impacts on participation in HIV vaccine studies have also been reported in Kenya. 57 Participants in our study believed that their participation in an HIV vaccine trial would expose them to prejudicial and discriminatory practices similar to those directed at HIV positive people. Several studies have reported similar findings in other countries. [58][59][60][61] These barriers may be reduced by providing the correct information about the HIV vaccine program. In HIV vaccine efficacy trials, high-risk populations are required for participation. Based on the findings of our study, IDUs represent a good vaccine trial population, as they have been involved in many health promotion programs. Researchers must provide educational materials and ensure that all behavioural and social needs are met before, during, and after the vaccine trials.

Limitation
This study is not without limitations. First, the study sample was recruited from the methadone clinic which might be different from IDUs in the general population. However, the risk and behavioural characteristics of the participants validate the information. The findings of 0ur study are valuable for planning future HIV vaccine efficacy trials. Second, although the findings of this study should not be generalised beyond the studied sample, the information obtained is important when formulating an HIV research study in a similar setting.
The use of the qualitative method allowed the authors to examine a study sample that had not been previously investigated in Tanzania. Finally, the integration of FGDs and IDIs data as a form of triangulation has been challenged in establishing rigour 62,63 and therefore might have affected the integrity of findings. However, the current study used data from both methods to complement each other. For example, the 4 participants who were reluctant to express their ideas during FGDs appeared more interactive during the IDIs.
Complementing is important to the qualitative inquiry as it allows for the recognition of multiple realities. In this case, the IDIs added additional information that was not recognised in the FGDs. The combination of 2 sources of data increased the richness of the information obtained, thus making the findings more valuable.

CONCLUSIONS
Participation in a hypothetical HIV vaccine trial among IDUs is influenced by positive and negative factors. Actual recruitment plans could be made through a better explanation of HIV vaccine trials, the expected individual and collective benefits associated with the trials. Correct information about the HIV vaccine studies and community sensitisation is likely to enhance participation in future HIV vaccine trials.