Original research•

Health research mentorship in low-income and middle-income countries: a global qualitative evidence synthesis of data from a crowdsourcing open call and scoping review

Abstract

Introduction Research mentorship is critical for advancing science, but there are few practical strategies for cultivating mentorship in health research resource-limited settings. WHO/TDR Global commissioned a group to develop a practical guide on research mentorship. This global qualitative evidence synthesis included data from a crowdsourcing open call and scoping review to identify and propose strategies to enhance research mentorship in low/middle-income country (LMIC) institutions.

Methods The crowdsourcing open call used methods recommended by WHO/TDR and solicited descriptions of strategies to enhance research mentorship in LMICs. The scoping review used the Cochrane Handbook and predefined the approach in a protocol. We extracted studies focused on enhancing health research mentorship in LMICs. Textual data describing research mentorship strategies from the open call and studies from the scoping review were coded into themes. The quality of evidence supporting themes was assessed using the Confidence in the Evidence from Reviews of Qualitative research approach.

Results The open call solicited 46 practical strategies and the scoping review identified 77 studies. We identified the following strategies to enhance research mentorship: recognising mentorship as an institutional responsibility that should be provided and expected from all team members (8 strategies, 15 studies; moderate confidence); leveraging existing research and training resources to enhance research mentorship (15 strategies, 49 studies; moderate confidence); digital tools to match mentors and mentees and sustain mentorship relations over time (14 strategies, 11 studies; low confidence); nurturing a culture of generosity so that people who receive mentorship then become mentors to others (7 strategies, 7 studies; low confidence); peer mentorship defined as informal and formal support from one researcher to another who is at a similar career stage (16 strategies, 12 studies; low confidence).

Interpretation Research mentorship is a collective institutional responsibility, and it can be strengthened in resource-limited institutions by leveraging already existing resources. The evidence from the crowdsourcing open call and scoping review informed a WHO/TDR practical guide. There is a need for more formal research mentorship programmes in LMIC institutions.

What is already known on this topic

  • Mentorship is key to improving overall research impact and outcomes. Existing strategies have focused on improving mentor and mentee roles, predominantly in high-income settings.

  • Embedding research mentorship within institutions will enhance research impact and sustainability in the long term.

What this study adds

  • This global qualitative evidence synthesis highlights practical strategies through which low-income and middle-income country institutions can build and institutionalise research mentorship activities.

  • These include recognising research mentorship is a collective institutional responsibility to be provided by all involved and leveraging already existing resources.

How this study might affect research, practice or policy

  • Our findings suggest that research mentorship programmes should be embedded within institutions and provided to all researchers for more impact.

  • There is a need for supportive policies and further research to implement the proposed strategies with monitoring and evaluation to sustain success.

Introduction

Mentorship is fundamental to global health research.1 Mentorship is often catalytic in launching individual research careers, building research teams at the group level and sustaining research institutions over time.2 Although the individual and institutional pillars of research mentorship are widely recognised, there are fewer resources focused on enhancing research mentorship at the institutional level.3 Institutions (eg, universities, research institutes and other groups) establish and shape mentorship expectations, programmes, incentives and policies.

Research mentorship tools have been mainly designed for high-income research institutions, neglecting low/middle-income countries (LMICs).4 LMIC institutions may have different traditions, structures, cultures and capacities related to research mentorship.5 Earlier research mentorship models for LMICs have focused on mentor and mentee roles. Specific toolkits and strategies are needed to enhance health research mentorship in LMICs. For example, LMIC institutions often have fewer training grants focused on building research mentorship, comparatively fewer senior mentors per mentee and less institutional support.6 7 At the same time, there are many indigenous research mentorship strategies that suggest LMIC-centred approaches are feasible and effective.2 This suggested an opportunity to identify practical strategies to enhance research mentorship at LMIC organisations.

In response, we organised a crowdsourcing open call and scoping review to identify strategies to enhance health research mentorship in LMICs. A crowdsourcing open call has a group of people to solve a problem and then implement selected solutions.8 The open call and scoping review were commissioned by the UNICEF/UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Diseases, TDR. TDR Global is a global group of scientists and experts passionate about building capacity for research on neglected infectious diseases.

The aim of this global qualitative evidence synthesis was to identify strategies to enhance health research mentorship in LMIC institutions and included data from the crowdsourcing open call and scoping review.

Methods

We used a qualitative evidence synthesis approach with findings from a crowdsourcing open call and a scoping review to better understand research mentorship in LMIC organisations.

Crowdsourcing open call

The crowdsourcing open call on research mentorship was organised by Social Entrepreneurship to Spur Health (SESH), TDR Global and the Armauer Hansen Research Institute (AHRI). The open call was implemented using stages outlined in the TDR/SESH/Social Innovation in Health Initiative (SIHI) practical guide on crowdsourcing,9 10 that included the following steps: (1) organised a multisectoral steering committee; (2) engaged the community to contribute; (3) independently evaluated submissions; (4) recognised finalist participants; and (5) refined ideas and implemented (develop practical guide) selected ideas (table 1). The open call was launched on 21 October 2021 and closed on 15 February 2022.

Table 1
|
Outline of the main steps of the crowdsourcing open call

We organised a multisectoral steering committee from key stakeholders across diverse geographic regions. Steering committee members were recruited based on previous experience in research mentorship projects, their country of residence (prioritising LMIC researchers), and their previous experience with TDR Global. The 24-person steering committee members included public health experts (n=6), government leaders (n=5), training directors (n=5), social media and communications experts (n=4) and clinical physicians (n=5).

The open call was announced on the SESH website, social media channels and through partner organisation networks. Participants were invited to submit ideas for strategies that strengthened existing initiatives, established new mentorship programmes, and aimed to create and sustain strong cultures of research mentorship. The website included a translation widget to facilitate non-English speaker participation. The website included the purpose of the open call, background information, judging criteria, eligibility criteria, formatting details and deadlines. All participants were also asked to complete a brief online survey that gathered demographic data including age, gender, country of residence and education level.

The open call was promoted through digital networks (table 1) including social media (Twitter, Facebook and LinkedIn), email listservs and networks of collaborating organisations represented by the steering committee. We used real-time social media analytics, website visiting frequency and submissions received to assess regional engagement and re-direct promotion efforts. All promotional materials, social media cards, and emails were translated into Spanish and French.

A total of 12 independent judges rated each submission 1–10 scale (1 as low and 10 as high) in five categories: (1) clear description; (2) potential for enhancing research mentorship in LMICs; (3) innovation; (4) potential for transferability in diverse LMIC settings; (5) promotion of equity and fairness. In addition, each judge gave an overall score of 5–50. Judges with a conflict of interest recused themselves from reviewing that entry. Conflicts were defined as collaborating, coauthoring, helping, receiving, or providing monetary or other support, or anything that could be perceived as a conflict of interest. The non-English entries were evaluated for initial eligibility with the use of translation software and then were judged by those proficient in the language of the entry. An overall score that averaged the means for each of the five subcomponents (clear description, potential for enhancing research mentorship in LMICs, innovation, potential for transferability in diverse LMIC settings, promotion of equity and fairness) was calculated.

An overall score of greater than 35 (a threshold prespecified by the steering committee,) on the 5–50 scale received a commendation of excellence on behalf of TDR Global, SESH and AHRI. Five finalists were ultimately selected based on the judging criteria. The five open call finalists were invited to join a WHO/TDR virtual working group to contribute to the development of a WHO/TDR practical guide to institutionalise research mentorship in LMIC. Three finalists were also invited to attend a TDR Global/AHRI conference on research mentorship in Addis Ababa, Ethiopia on 23 June and 24 June 2022.

Scoping review

A concurrent scoping review was conducted according to the Joanna Briggs Institute proposed methodology and reported using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews.11 The purpose was to summarise available practices, lessons and gaps/needs of health-related research mentorship programmes in LMICs. Keyword search was conducted in PubMed, EMBASE, Cochrane Database of Systematic Reviews, JBI Evidence Synthesis, EBSCO, SciELO (Scientific Electronic Library Online) and AJOL (African Journals Online). Grey literature was also searched using included ProQuest Dissertations and Theses, Google Scholar, and institutional websites (National Institute of Health) (online supplemental appendix 1). The search results were screened using a predefined set of inclusion and exclusion criteria, and eligible articles for inclusion in the review were collated for data extraction. Inclusion criteria were focused on quantitative, qualitative and mixed methods studies on both formal and informal health research mentorship programmes in LMIC as defined by the World Bank.12 Studies on mentorship of non-research health activities (eg, clinical and administrative) were excluded (online supplemental appendix 2).

Data analysis

From the open call submissions, we received both quantitative data and qualitative data and we used a parallel mixed methods approach to analyse and present the data.13 The quantitative data, included the participants demographics and submission characteristics. These data were analysed and presented using basic descriptive frequencies. Quality of studies contributing to the review findings was assessed using the Critical Appraisal Skills Programme (online supplemental appendix 3). The data included articles and the content of the submissions were mostly textual qualitative data, and this was analysed thematically using the framework approach including familiarisation with the data, coding, charting and mapping out themes for interpretation.14 15 The themes identified from the open call were further strengthened with findings from the scoping review (figure 1). To ensure validity and reliability in presenting findings, the eligible submissions and included articles were coded separately by two independent reviewers (EEK and KM) and discrepancies were reviewed by a third team member (JDT). We used the GRADE-CERQual (Confidence in the Evidence from Reviews of Qualitative research) approach to assess confidence in the certainty of the review findings.16 This includes an assessment based on methodological rigour, coherence of the review finding, adequacy of the data and relevance of the included studies to the review question.16–21 Each of these components was assessed and made into an overall judgement on the confidence in each review finding (very low, low, moderate, high).

Figure 1
Request permissionrequest permission icon
Figure 1

Stages in the identification of key themes.

Patient and public involvement

This study was carried out as a crowdsourcing open call for ideas and a scoping review. No patients were involved. The scoping review used publicly available research on research mentorship. Our crowdsourcing open call solicited ideas for institutionalising researcher mentorship. A practical guide was developed by the same authors with the findings for mentorship in health research.22 This guide is open access and available online, to the public and provides practical advice for research mentorship practice in LMICs.

Results

The open call received a total of 60 submissions and 46 submissions were eligible for judging. Twelve submissions met the prespecified criteria for excellence (figure 2).

Figure 2
Request permissionrequest permission icon
Figure 2

Distribution of mean scores on a 5–50 scale.

Submission characteristics and participant demographics

We received submissions from a total of 33 different countries including 24 submissions from low-income countries, 18 submissions from middle-income countries and 4 submissions from high-income countries. Top countries include Nigeria (10 submissions), Ethiopia (6 submissions) and Malawi (3 submissions) (online supplemental appendix 4). Of the eligible submissions, 40 were in English language and six were submitted in Spanish. We had submissions from 26 male and 20 female participants, most of them between 28 and 47 years of age (32 vs 14 participants), and with a doctoral or professional degree (36 vs 14 participants) (table 2).

Table 2
|
Characteristics of global research mentorship open call participants

Strategies to enhance mentorship

Themes identified from the analysis of the open call submissions have been broadly categorised into strategies that can enhance research mentorship within LMIC institutions. A total of 10 themes were identified from the results. The working group met and examined the evidence supporting each theme and five of these were prioritised as key. Key findings include recognising mentorship as an institutional responsibility, leveraging existing research and training resources, digital tools to kickstart and sustain mentorship relations over time, nurturing a culture of generosity and peer mentorship (figure 3).

Figure 3
Request permissionrequest permission icon
Figure 3

Overview of strategies to enhance research mentorship in low/middle-income countries (top five themes in bold).

Mentorship as an institutional responsibility

Eight strategies and 15 studies23–37 identified research mentorship as an institutional and collective responsibility which should be expected from and provided by all team members. To support this, some submissions highlighted the development of a quick reference guide or policies to ensure that everyone is engaged in mentorship (five submissions). Specific strategies to ensure institution-wide coverage include building research mentorship training into routine onboarding procedures, requiring research ethics committees to consider mentorship as part of the review criteria and requiring grant applications to support research mentorship.

Digital tools to support mentorship

A total of 14 strategies and 11 studies31 35 38–46 demonstrated that digital tools enhanced research mentorship in LMICs. Digital tools including apps, websites and other web-based platforms to aid in mentor/mentee matching, communication, and to establish and sustain mentorship relationships over time (14 submissions). While some strategies required internet access and sufficient bandwidth, there were many low-tech solutions that would be relevant in resource-constrained settings. These included mobile instant text messaging apps, social media groups such as Facebook and WhatsApp. However, several studies mentioned that digital approaches may exacerbate inequalities (eight studies) and be less relevant in a large number of LMIC institutions with limited digital infrastructure.47–54 Additional challenges with digital tools in LMICs include power cuts and constant interruptions to internet access.55

Leveraging existing research and training resources

Fifteen open call strategies and 49 studies7 23 24 26–36 39 41 42 44 46 56–86 showed that leveraging existing research and training programmes facilitated research mentorship. Already existing training programmes with databases of staff, contacts/institutions that can be partnered with to network on mentorship and leverage other institution’s resources is beneficial for facilitating sustainability research mentorship. Leveraging established relationships with other institutions to create inter-institutional mentor/mentee relationships (six submissions). Another example is the use of formal research project supervision with mentors acting as supervisors and mentees acting as researchers (seven submissions).

Culture of generosity

Seven open call strategies and seven studies27 30 33–35 77 87 included in the review emphasised the need for a culture of generosity and holistic approach in mentorship relationships. Mentors are to consider social life, be respectful of diversity and protect mentees from negative critics and racial/gender bias. This warm and inclusive culture of generosity is thought to facilitate and develop motivation in mentees to become future mentors. Mentorship should purposely cultivate a culture of cooperation rather than competition within mentor/mentee programmes (five submissions).

Peer and group mentoring

A total of 16 open call submissions and 12 studies27 31 33–35 44 79 83 87–90 recognised the role of peer mentorship in enhancing research mentorship in LMIC settings. Peer mentorship is defined as informal and formal support from one researcher to another who is at a similar career stage. Introducing junior scientists and PhDs into mentoring early on to create a culture of mentorship that is sustainable over time (four submissions). Creation of a formal or organised mentorship club, group or other form of community to aid in creating lasting mentor/mentee relationships (14 submissions). However, two studies highlighted hesitancy and barriers in peer mentorship that include fear, embarrassment, lack of knowledge and awareness.51 52

A summary of these key findings, the contributions strategies and studies and assessment of the certainty of the evidence of the finding are presented in the GRADE CERQual table (table 3).

Table 3
|
Evidence profile and assessment of confidence in the review findings as per GRADE-CERQual methodology

Discussion

Our data provide evidence-based strategies to improve health research mentorship at LMIC institutions. Institutions need to recognise that research mentorship is a collective responsibility that should be expected and provided to all members. Promoting a culture of generosity increases the sense of collective responsibility for research mentorship. Ongoing research and training resources can be leveraged to spur research mentorship at the institutional level. This manuscript extends the literature2 by centring evidence and strategies from LMIC researchers, including data from a global crowdsourcing open call, and assessing the strength of the evidence using the CERQual approach.

Our data suggest that institutions should be responsible for ensuring that research mentorship is provided to and expected from all members. This contrasts the practice of many LMIC research mentorship programmes that are offered to a subset of researchers.30 91 There are several strong reasons to consider research mentorship as a fundamental right of being in a university or research institute. Research mentorship can enhance recruitment and retention of promising research talents, build a sense of common purpose and enhance research outcomes. Greater institutional support for mentorship have been correlated with positive mentorship experiences.92 Unfortunately, under-represented racial/ethnic minorities less often have research mentors compared with other researchers.93 94 Policies that make research mentorship available to all members could decrease mentoring disparities.

Our work suggests that nurturing a culture of generosity within research institutions can increase the likelihood of current mentees becoming subsequent mentors. This is consistent with research showing that people who receive mentorship are more likely to serve as mentors for other people.95 Mentorship within research institutions could create virtuous cycles that spur further kindness between researchers. Positive mentorship experiences provide examples of behaviours to emulate; negative mentorship experiences could be useful as a reminder of what not to do when one is a mentor.95

LMIC institutions have existing research and training resources that can be leveraged to enhance research mentorship. This finding aligns with previous mentoring toolkits96 and mentorship guidance.1 Having research in an institution allows opportunities to interact with external scientists and to embed mentorship activities into funded research or conferences. From a training perspective, using open access learning materials (eg, massive open online courses), developing peer mentorship groups and organising university elective credit for research can formalise research mentorship.

This study has limitations. First, although, we received some non-English submissions, our relatively English-focused promotion materials likely limited contributions from some LMICs. At the same time, we accepted submissions in each of the five official United Nations languages and translated the call for submissions. Second, anticipating the complexity of health research mentorship, we decided to accept text and non-text submissions. Third, published literature on research mentorship in LMICs is limited. However, our use of a global crowdsourcing open call allowed us to elevate the voices of LMIC researchers and learn from indigenous strategies that are not part of the published literature.

The data from this qualitative evidence synthesis have implications for health research mentorship programmes and policy. From a programme perspective, these data suggest that research mentorship programmes should be embedded within institutions and provided to all researchers. Expanding the scope of research mentorship could help decrease disparities in mentoring and build a sense of collective solidarity. From a policy perspective, the data from this qualitative evidence synthesis directly informed a WHO/TDR practical guide called HERMES (HEalth Research MEntorship in Low and Middle-Income CountrieS.

Research mentorship is a critical component of developing vibrant research institutions in LMICs. The evidence identified through this global open call and scoping review provide specific strategies and guiding principles for research mentorship. Research on implementation strategies to enhance mentorship at LMIC institutions is needed to advance this field. Monitoring and evaluation of research mentorship are critical for sustained success.

Article metrics
Altmetricopen-url
Dimensionsopen-url