Attracting, equipping and retaining young medical doctors in HIV vaccine science in South Africa

Background HIV remains a significant health problem in South Africa (SA). The development of a preventive vaccine offers promise as a means of addressing the epidemic, yet development of the human resource capacity to facilitate such research in SA is not being sustained. The HIV Vaccine Trials Network (HVTN) has responded by establishing South African/HVTN AIDS Early Stage Investigator Programme (SHAPe), a programme to identify, train and retain clinician scientists in HIV vaccine research in SA. Objectives The present study sought to identify factors influencing the attraction and retention of South African medical doctors in HIV vaccine research; to understand the support needed to ensure their success; and to inform further development of clinician research programmes, including SHAPe. Methods Individual interviews and focus groups were held and audio-recorded with 18 senior and junior research investigators, and medical doctors not involved in research. Recordings were transcribed, and data were coded and analysed. Results Findings highlighted the need for: (1) medical training programmes to include a greater focus on fostering interest and developing research skills, (2) a more clearly defined career pathway for individuals interested in clinical research, (3) an increase in programmes that coordinate and fund research, training and mentorship opportunities and (4) access to academic resources such as courses and libraries. Unstable funding sources and inadequate local funding support were identified as barriers to promoting HIV research careers. Conclusion Expanding programmes that provide young investigators with funded research opportunities, mentoring, targeted training and professional development may help to build and sustain SA's next generation of HIV vaccine and prevention scientists.


Introduction
South Africa (SA) has one of the highest HIV prevalence rates globally. 1 Despite promising new developments in non-vaccine prevention modalities, a preventive vaccine against HIV offers the best hope to end the epidemic. 2 Several large-scale HIV vaccine, and other prevention, trials will begin in SA in 2016. 3,4 These trials will utilise the strengths of South African investigative teams, requiring a wide breadth of research capacity. Whilst senior-level South African HIV vaccine researchers have developed enormous expertise, the number of junior clinician investigators entering the field has lagged. Since the early 1990s, SA has seen a decrease in the number of clinical researchers and an ageing of publishing scientists. 5,6 A concerted effort is required to increase the number of young medical doctors entering the HIV prevention and vaccine field to ensure the country's ongoing contribution to this critical effort.
In 2010, responding to the challenge, the HIV Vaccine Trials Network (HVTN), with support from the National Institute of Allergy and Infectious Diseases (NIAID) and the Fogarty International Center (FIC), established the South African/HVTN AIDS Early Stage Investigator Programme (SHAPe), a peer-reviewed medical doctors/PhD programme that recruits and supports young clinician investigators as they become independent investigators. Components of the programme, coordinated by the Desmond Tutu HIV Centre at the University of Cape Town, include 3-year salaried appointments at HVTN clinical trial sites; financial support for a mentored research project and research-related costs; targeted training and tuition for a concurrent PhD programme; travel to HVTN meetings and international HIV conferences; and facilitated integration into the HVTN scientific community via participation on scientific and governance committees.
To inform the SHAPe programme and future efforts, formative research was conducted to: (1) understand facilitators and barriers to attracting, equipping and retaining young South African medical doctors in careers in HIV vaccine research, (2) develop recommendations to address identified challenges and (3) inform the design of clinician investigator development programmes, including SHAPe.

Sampling
During 2011, participants were recruited from clinical research sites and via referrals throughout SA. Target participants comprised a small group of specialised individuals and therefore purposive sampling with specific criteria was used. These groups were defined as: (1) senior investigators (SI)those who led research teams and held the role of 'principal investigator' (PI) for at least 5 years, (2) junior investigators (JI)those who had medical degrees and worked at clinical research sites for 10 years or less without assuming the role of PI and (3) young medical doctors (MD) -those who had received medical degrees no more than 12 years ago and expressed interest in research but had not pursued research careers.

Data collection
We used a qualitative approach and conducted semistructured interviews and focus groups. K.B. underwent comprehensive training and utilised interview and focus group schedules to guide the sessions. Fifteen one-on-one, semi-structured, confidential telephone interviews were conducted, lasting approximately 1 hour and included five JIs, five SIs, and five MDs. Data collected during the interviews informed semi-structured questions used for three faceto-face focus groups, lasting approximately 1.5 hours and included four SIs, three JIs and four MDs in separate groups. All interviews and focus groups were conducted in English, recorded using a hand-held audio recorder, and transcribed in English. Demographics questionnaires were completed by those who participated in focus groups.

Data analysis
Three investigators independently coded five interview transcripts and then met as a team to review codes, resolve differences in coding by consensus, and create a codebook. The codebook was used to analyse subsequent interview and focus group transcripts, and generate new codes. Subsequent transcripts were assigned to three of the authors who independently cleaned and coded them using the codebook. The three investigators reconvened again to resolve differences in coding by consensus, analyse the transcripts, discuss patterns in the data and organise identifiable themes in ATLAS.ti.6.2.25 (GmbH Berlin). Data and investigator triangulation were utilised to enhance validity of the findings. 7

Ethical considerations
All participants were over 18 years old and provided written informed consent. The study was approved by the University of Cape Town's Human Research Ethics Committee (IR File #128/2011) and the Fred Hutchison Cancer Research Center Institutional Review Board (IR File #7448).
The following themes were identified for each research objective and were endorsed by participants from all groups (JI, SI, MD) unless otherwise stated. Suggestions for supporting these activities included providing students with small mentored research projects, offering stipends to support projects, and combining a medical degree with a master's or PhD degree.

Attracting young doctors to research
Several participants indicated that better integration of researchers into university academic departments would provide them with greater access to funding support,

Equipping and retaining doctors for careers in research
There was considerable overlap in themes identified as important for equipping and retaining doctors; therefore they are combined here. Some participants suggested seeking alternative funding sources, particularly from the South African government, to establish a more stable funding base to supplement grants: 'If the government is proactive in leading research … the private sector will also move in and pour in funds when they see there is commitment from the government.' (JI)

The South African/HVTN AIDS Early Stage Investigator Programme
Participants were asked to reflect on the design of the SHAPe programme and comment on the strengths and weaknesses of this model, and provide input on ways to improve the programme.
Most participants were enthusiastic about the SHAPe programme and agreed that it served an important role in providing an opportunity not previously available: to recruit and train young doctors in research. Participants identified SHAPe's strengths as providing a mentored research project and full-time salaries, incorporating a PhD degree programme, offering experience as a clinical trial doctor, and opportunities for travel to research meetings and conferences. Lack of awareness of the programme by the medical community was often identified as the most significant weakness: Participants suggested several ways to improve SHAPe. Most advocated greater involvement from SIs and JIs in publicising the programme and presenting research findings at medical schools and hospitals to increase interest in HIV vaccine research.
Most participants agreed that the best recruitment approach is multi-pronged and includes 'personal outreach', 'national publicity' and advertising through the 'provincial Departments of Health', 'newspapers', 'major medical journals' and 'medical association websites'. A few participants also advocated use of social media such as Facebook (SI), direct SMS (MD) and email (MD).
Some participants suggested expanding the programme to support independent research for mid-career HIV vaccine investigators. A few suggested increasing the number and location of sites where scholars can work.

Discussion
The present study identified factors influencing the attraction and retention of South African medical doctors into HIV prevention research; increasing the understanding of the resources and support needed to ensure their success; and eliciting suggestions to inform design of clinical research development programmes, such as SHAPe.
Of note is that findings were largely applicable to HIV prevention research generally and clinical research more broadly, increasing their value and impact in informing recommendations and future practice. Overwhelmingly, there was recognition from all groups that research was not adequately represented in the medical training curriculum and that this limited exposure discouraged interest in, understanding of, and entrance into this field. The overall shortage of doctors in primary healthcare is critical in SA 8,9,10 and requires urgent redress. 11 As a result, medical education has renewed its focus on primary healthcare. 12

Study limitations
There are several limitations to the present study. The sample size was small, comprising 18 participants, most of whom took part in both one-on-one interviews and focus groups. Whilst this is considered an adequate sample size for qualitative research, 14 it possibly limits the generalisability of the present results. However, notably two of the three groups included in this study (SIs and JIs) are, by definition, small in size. HIV vaccine research is a specialised field and it is partly because of the small number of doctors entering this field that the present research was conducted. Consequently, particularly for SIs, our sample included a significant proportion of the total SI population involved in HIV vaccine research in SA.
In addition, given the small size of this research community in SA, it is possible that participants did not feel comfortable speaking freely and honestly, out of concern that they would be recognised, despite removal of identifiers. This risk might have been an issue particularly during focus groups, as many investigators are colleagues and collaborators. Nevertheless, all participants expressed an interest in and recognition of the importance of the topic, and were aware of the value of the research process, and it is hoped that this encouraged openness and honesty throughout.
Finally, factors identified by participants as significant in attracting, equipping and retaining doctors in HIV prevention science research were largely representative of their own needs rather than a broader impression of what might be required overall. Whilst inevitable, this may be viewed as a limitation.
which may have inappropriately influenced them in writing this article.