Table 3.

Challenges and mitigation strategies during screening and enrollment.

Challenge	Solution/mitigation strategy
Multiple regulatory authorities’ approval was required prior to trial initiation. Regulatory approval was required from Uganda Virus Research Institute Research Ethics Committee (UVRI REC), the National Drugs Authority (NDA), the Uganda National Council for Science and Technology (UNCST), the London School of Hygiene and Tropical Medicine Research Ethics Committee (LSHTM REC), and Imperial College London Research Ethics Committee (ICL REC)	Prior engagements were made by the trial investigators with these authorities to make them aware that plans were under way to conduct this trial, and to expect study submissions. The site initially made application to the UVRI REC, followed by the NDA. On submission to NDA, request for joint review from the Ugandan authorities was made, Unfortunately, NDA notified the site that joint review was no longer possible since UVRI REC review had commenced already. Joint review would require that submissions are made to all Ugandan authorities at the same time, so that review could be done at the same time. This was a learning experience for the site.
Resentment of COVID-19 vaccines from the community. While this trial was able to enroll adequate numbers as had been planned, the experience from the community engagement activities showed that various potential participants who came for pre-screening sessions, did not return for subsequent visits including screening. The numbers may not have been captured for this publication, but research staff noticed that potential participants had more questions than usually asked for other trials, and a considerable number of attendees was not able to come for subsequent sessions. Many of these questions were derived from information received via social media.	The team endeavored to offer as much information as possible to inform participant decisions, addressed all questions asked as informed by research data including that from the UK version of this trial.11 The site has also established a very comprehensive community engagement strategy at the center of which is a competent Community Advisory Board (CAB), that supports conduct of clinical trials at the unit. The functions and operations of this community advisory board have been described elsewhere.24
Evolution of the outbreak, with the emergency of variants of concern. As described in the text, the emergency of the Omicron variant toward the end of 2021 and beginning of 2022 made it extremely challenging to enroll SARS-CoV-2 antibody negative participants as the positivity rate increased rapidly (60.8% to 84.4%, based on data from this study).	The site had experience in successful community engagement prior to initiation of screening. When the challenge surfaced, it was facile to shift focus from the urban settings where initial screening had focused, to the rural settings where it was assumed that positivity rate would be lower given reduced mixing of people.
High screen outs due to grade 1 laboratory parameters. As already noted, the trial used the FDA toxicity grading scale, which had a number of parameters, especially hematological parameters graded as grade 1, that otherwise were in the normal range if assessed using local reference ranges.	Although this was not done for this trial, the solution to this would have been designing the protocol to allow for consideration of clinical significance in the assessment, especially for low grade laboratory parameters. The more useful and durable recommendation is for toxicity grading tables to be modified or created to grade laboratory parameters based on the populations in context.
Conflicting confirmation of serological results for participants screened. As indicated in the methods section, serological tests involved use of two kits, i.e., Multi G (MGFT3), and Standard Q (Standard Q COVID-19 IgM/IgG Plus). Unfortunately, 23(10.9%) of participants returned discordant results. By the time the trial was conducted, these were the only two kits that had been assessed and validated to be used in Uganda. The other challenge ensuing from this was communicating to participants with discordant results if they had been exposed to SARS-CoV-2 before or not.	There was no practicable solution for this during the trial, apart from screening more participants to cover for the discordant results. The ideal solution would be encouraging more quick manufacture and testing of kits, including from local manufacturers where possible. Concerning communication of discordant results, the participants were informed that due to the challenge of only having two validated kits, we could not conclusively establish the exposure status of these participants. These participants were not enrolled in the trial, but were encouraged to go for routine vaccination at the government facilities where vaccines were available.