Community stakeholder engagement during a vaccine demonstration project in Nigeria: lessons on implementation of the good participatory practice guidelines

Introduction To report on the successes and challenges with implementing the good participatory practice guidelines for the Nigerian Canadian Collaboration on AIDS Vaccine (NICCAV) project. Methods An open and close ended questionnaire was administered to 25 randomly selected community stakeholders on the project. The questions sought information on perception about the community entry, constitution and function of the community advisory board (CAB) and community based organization (CBO), media engagement process, and research literacy programmes. The quantitative and qualitative data were analysed and findings triangulated. Results The project exceeded its targets on CBO engagement and community members reached. Stakeholders had significant improvement in knowledge about HIV vaccine research design and implementation (p=0.004). All respondents felt satisfied with the community entry, CAB constitution process, function and level of media engagement; 40% were satisfied with the financial support provided; 70% felt the community awareness and education coverage was satisfactory; and 40% raised concerns about the study site selection with implications for study participants' recruitment. Conclusion The NICCAV community stakeholder engagement model produced satisfactory outcomes for both researchers and community stakeholders. The inclusion of an advocacy and monitoring plan enabled it to identify important challenges that were of ethical concerns for the study.


Introduction
The field of HIV vaccine development was reinvigorated by the results of the ALVAC-HIV (vCP1521) (Sanofi Pasteur)/AIDSVAX B/E (Global Solutions for Infectious Diseases) trial that demonstrated 31.2% protection [1], and lower risk of HIV infection for the study population when compared with populations in other trials [2][3][4][5][6]. A number of studies with new products to boost the potency of this and others with new vaccine concepts are ongoing [7][8][9]. These studies require large numbers of volunteers required for the implementation of these studies. Nigeria could provide an ideal setting for conducting HIV vaccine trials since it has the second largest burden of HIV infection in Africa [10]. Also has the unique HIV subtype G, CRF02 A/G along with other recombinant forms that accounts for about 40% of its HIV infection [11]. The Nigerian Canadian Collaboration on AIDS Vaccine (NICCAV) project was developed to address this need. Its focus was to build local capacity to conduct HIV vaccine trials [12] through a four years "mock" clinical trial [13]. The project had multiple objectives one of which was to identify how best to mobilise and effectively engage stakeholders in a HIV vaccine research project, and to ensure the community benefits by having improved knowledge and understanding of HIV vaccine research and clinical trials. Community and stakeholder engagement was prioritised for this project because it was considered an ethical imperative [14,15] as it enhances the quality and outcome of research [15,16], facilitates social change emanating from individual and community empowerment in support of health and development [17] and enhances the science, the sense of joint ownership of a study by both researchers and research stakeholder [17]. In effect, it improves research implementation, procedures, and outcomes; and helps to build effective and sustainable research collaborations [18].
Engagement with stakeholders critical to HIV treatment and prevention research has a long history [19]. Day et al. [18] conducted a systematic review and identified multiple strategies for facilitating stakeholder engagement most of which were to inform protocol development and trial participants' recruitment. There were also few literatures reporting on stakeholder engagement with HIV prevention research conducted in low and middle income countries, countries where stakeholder engagement is considered critically important [20].
Community members themselves identify no justification for omitting community engagement plans during the implementation of research [21]; and if strategically engaged, can actually make critical inputs into decisions on study designs [22]. In view of these, the NICCAV project developed a community stakeholder engagement plan. The community stakeholder engagement programme was contracted to an independent civil society organisation (CSO) that had   Table 1 is a summary of stakeholder engagement plan developed and submitted for ethics approval. We report on the method used for implementing the community stakeholder engagement model, the mid-term evaluation process, and the outcomes of the mid-term evaluation. We then discuss the successes and challenges with implementing the GPP. An end-of-term assessment could not be conducted due to a last quarter cut-back in funding. We therefore cannot objectively discuss the impact of the project.

Methods
NICCAV project study design: the NICCAV study enrolled 534 HIV-1 exposed sero-negative study participants who were 18 years old and above [23] from Jos and its environs, from October 2011 till December 2012 and followed-up each study participants for two years [23].  Table 1). The questionnaire was pilot tested for content clarity. The 25 survey participants (10 CBO representatives, four CAB members, seven journalists and four NICCAV project staff) were randomly selected by balloting, from the lists provided by NHVMAS. An external evaluator conducted the evaluation. The proposed end-of-term evaluation planned for the project could not be conducted due to cut-back of the grant three months to the expiration of the project.
Data analysis: a descriptive analysis was conducted for the quantitative data. The likert-like scores were re-categorised into three.
Scores 1 (excellent) and 2 (good) where re-categorised and coded as satisfactory, and scores 3 (fair) and 4 (poor) were categorised as unsatisfactory. Score 5 remained as unknown. An inductive thematic analysis was conducted to identify salient themes after reading and re-reading the hand written notes from the telephone interviews. The findings were triangulated with the findings from the quantitative data.
Ethical consideration: the main NICCAV research study received ethical approvals for the study [12]. NHVMAS obtained ethics approval for its stakeholders' engagement programme from the Plateau State Specialist Hospital, Jos (NHREC/05/01/2010B). All data were handled confidentially. Verbal consent was obtained from all participants during the mid-term evaluation.  Despite the enthusiasm on engagement of the media, the limited budget allocated for the media engagement programme also meant journalists had to look for free slots to discuss the NICCAV project rather than create an active ongoing programme that specifically focused on the NICCAV project. methods. This often involves formal social science methods such as in-depth interviews or focus group discussions. It is unclear how effective these methods are for fostering meaningful partnerships and continuous dialogue as the GPP guidelines recommend [3].

Discussion
Additionally, the extent to which top-down engagement methods can inform the design and conduct of HIV clinical trials depends entirely on trial researchers. Thus, while the GPP guidelines recommend that trial researchers carefully consider and select from the range of possible advisory mechanisms [3]. We felt that the media engagement programme was also a strong component of this model. Medeossi et al. [26] highlighted the importance and benefits of using multiple media outlets to engage the community for biomedical HIV prevention research. Although we could not demonstrate the impact of the media engagement programme, the active dissemination of information about the project was a crucial step in creating awareness and generating interest about the project. The study had its limitations.
First, it could not develop an issue management plan as prescribed by the GPP. The issue management plan should describe how research teams intend to manage any unexpected developments that may emerge before, during, or after the trial [18]. While the community engagement project recognised the need for this, the project was unable to develop these plans as it required the active engagement of the research team. Attempts made to develop this plan failed. We felt this was because the attempt was made by an organisation external to the research team. Though engaging an independent organisation to manage the community engagement process is commendable, this made it challenging to actively engage the research team in the development of this plan. The autonomy of the community engagement program however, came with its advantages. It enabled the project develop a sustainable community engagement programme rather than focusing on promoting study participants recruitment and retention [27][28][29].
Community members were also able to identify their need for the research [30], helped the project identify ways of minimising associated internal and external risks [31], and built research literacy competency of the community stakeholders. It also facilitated the constitution of a CAB whose members' allegiance was to their primary constituencies thereby addressing some of the concerns research projects had about CABs [32]. Secondly, the stakeholder engagement programme commenced after study participants recruitment had commenced. This is not in line with the GPP requirement: community stakeholder engagement programs should commence at conceptualisation of the research question design of the study [18,33] and continue throughout the research life-cycle [33]. The failure to engage stakeholders early in the NICCAV project's life-cycle made the project fail to design a study that was sensitive to the religious inclination of the community. Although the study met its recruitment target [19], the likelihood was that the distribution of the recruited study participants' may have shown religious bias. Third, the limited financial support for the project came with challenges. The community engagement program's budget was cut back to 37.5% of its initial budget because of cut-backs in grant allocation by sponsors. Yet, the scope of the program could not be reduced and the project engaged 14 extra CBO. Although the programme leveraged on many existing project to conduct its activities [34], this approach limited the extent to which the community research literacy and the media engagement programmes was executed. Prior programs had identified the constant threat to the funds allocated for community engagement when the research project faces financial constraints [35,36]. We recommend that future community stakeholder engagement programme budget appropriately for its programmes. Despite these limitations, this study showed how studies can innovatively achieve its goals through engagement of local, state and national political stakeholders' in a research project in line with the recommendations of the GPP.

Conclusion
The NICCAV study community stakeholder engagement model was an • Very sparse evidence on the process, outcomes and impact of community engagement in biomedical HIV prevention research.

What this study adds
• This study information on the 'how' of community engagement in research; • This is a report of an indigenous process of facilitated community engagement with a biomedical HIV prevention research howbeit a demonstration project; • It also provides evidence on the outcomes of a community engagement programmes based on monitoring and evaluation indicators put in place at the designing of the project, and adds unique information to the field of GPP implementation.

Competing interests
The authors declare no competing interests.

Authors' contributions
Morenike Oluwatoyin Folayan conceptualise the study. All the authors were actively engaged in finalising the design of the study and its implementation. Morenike Oluwatoyin Folayan and Florita Durueke designed the monitoring and evaluation plan for the study. Morenike Oluwatoyin Folayan developed the first draft of the manuscript. All the authors contributed equally to revising and finalising the manuscript.
They agreed to the final version of the manuscript and to its submission.

Acknowledgments
We acknowledge the contributions of the stakeholders who invested a lot of time to develop and implement this project. Table 1: telephone interview guide for the formative evaluation of the NICCAV project Table 2: some media publication on the NICCAV project