Challenge contest
The challenge contest was hosted by SESH (Social Entrepreneurship to Spur Health) and TDR and called the “Women Leaders in Global Health Challenge” to complement the Women Leaders in Global Health Conference organized by the London School of Hygiene and Tropical Medicine [
13]. We followed the TDR “Crowdsourcing in Health and Health Research Practical Guide,” and used the following steps: organizing a multisectoral steering committee; engaging the community to contribute; evaluating ideas based on pre-specified criteria; recognizing exceptional finalists through commendation at the Women’s Leaders in Global Health Conference; and implementing selected ideas to adjust TDR fellowship policy [
9].
The challenge steering committee organized the challenge contest. A multisectoral steering committee included key stakeholders from diverse geographic regions. These included women who previously received this TDR fellowship, gender research experts, and public health experts. Countries represented included China (1), Germany (1), India (1), Kenya (2), Malaysia (1), South Africa (2), Switzerland (1), Tanzania (1), Tunisia (1), United Kingdom (4), United States (7), and Zambia (1). The steering committee met via Zoom teleconference on a monthly basis to guide challenge contest design, promotion and evaluation.
An open call was disseminated that explained the purpose of the contest and solicited short descriptions of solutions to promote women’s participation in this TDR fellowship [
14]. We included quotes and photographs from women who had joined the fellowship before in order to provide context. A short video was posted online to clarify contest rules. Confidentiality was maintained throughout the challenge. The contest was open to anyone from any location, but ideas from women in low- and middle-income countries were particularly encouraged. We used the TDR website, social media platforms (Twitter, Facebook, YouTube, and WeChat), email listservs and networks of collaborating organizations represented by the steering committee to disseminate the call for entries.
All submitted ideas were screened for eligibility and then judged. Each idea was screened for eligibility by at least two independent individuals (EL, SW). Eligible ideas were relevant to the topic, written in English, and under 500 words. Each eligible idea was then independently evaluated by three members of the steering committee. They rated ideas according to pre-specified criteria: capacity to increase the number of women who apply and receive TDR fellowships; feasibility; and innovation, defined as different from the current practice used in the fellowship. Ideas were scored on a single 1–10 scale. Steering committee members who had a conflict of interest recused themselves accordingly. Conflicts were defined as collaborating, co-authoring, helping, receiving or providing monetary or other support, or anything that could be perceived as a conflict of interest. Mean scores of the three scores were calculated for each idea.
A subset of individuals who submitted ideas were recognized by commendations (pre-specified as ideas with mean scores greater than 7.0) and selected finalists from this group were supported to join the Women Leaders in Global Health Conference. Finalists were selected by the steering committee based on the same criteria. Then, we created a working group composed of finalists, TDR representatives, and fellow alumni to refine ideas and present them at the 2018 Women Leaders in Global Health Conference in London. All of the finalist ideas were also presented to TDR for consideration and they ultimately selected three ideas to modify the mid-career fellowship.
Quantitative and qualitative methods
Summary statistics described gender identity, nationality, WHO region, and mean scores associated with ideas. In addition, we undertook a textual analysis of the ideas using qualitative methods. First, we developed a thematic codebook to classify and analyze themes. Themes were organized using Excel (Office 365, Microsoft). Two independent individuals with training in qualitative methods examined the top 50 ideas in order to identify potential themes. Themes were discussed with two qualitative researchers in order to finalize the code book. All eligible submissions were coded by three trained individuals. Discrepancies were brought to two additional study team members for discussion and resolution. Higher order themes were then iteratively identified and further characterized by example quotations.
We collected information from TDR about the total number of eligible men and women applicants to the mid-career fellowship before (2014–7) and after (2019) implementing the top three ideas. SPSS (version 26.0, IBM SPSS Statistics) was used for quantitative analysis.