2.1 Laying the Groundwork for RTS,S/AS01 E Readiness

The capacity-building strategy, implemented as part of the phase IV RTS,S/AS01E post-approval studies, was informed by insights from phase III clinical trials and previous pharmacovigilance training initiatives [17, 18, 20]. The phase IV studies employed active safety surveillance, incorporating predefined AEFIs and AESIs within a comprehensive monitoring framework. These predefined AEFIs and AESIs included events that have historically been associated with other vaccines, as well as meningitis and cerebral malaria, which had emerged as potential safety signals in earlier trials. The aim was to enhance and harmonize the recognition, documentation, and reporting of such events, both by study site staff and community-based healthcare professionals.

Before starting this initiative, it was essential to have a comprehensive understanding of the capacities of the study sites, particularly concerning case detection and study endpoint ascertainment. To support this, preparatory visits were conducted to all study locations by an external partner (Agence de Médecine Préventive [AMP]). During these visits, AMP engaged with Principal Investigators (PIs), community healthcare professionals, volunteers, and health district representatives. In addition to assessing existing reporting systems, these interactions served multiple purposes: introducing the study, preparing for training implementation, supporting the cascading training strategy, identifying potential challenges, and fostering collaboration with local and national stakeholders.

2.2 Practical Strategies and Approaches to Building Capacity2.2.1 In-Person Training

Two in-person training packages were developed: one for medical professionals and one for non-medical professionals. Medical professionals (i.e., doctors, pharmacists, nurses, nursing assistants, and midwives) working at both the facility and community level were trained on diagnosis, case definitions, clinical management, and reporting procedures for AESIs and AEFIs. Non-medical professionals included volunteers, traditional midwives, and other healthcare personnel involved in patient care or surveillance activities at both facility and community level. This group also included community health workers, who supported the collection of data through standardized questionnaires, identification of symptoms and danger signs, referral of patients, reporting of AEs, and verification of vaccination cards during home visits. Each training package comprised three modules (foundational concepts, specific topics, and practical exercises) tailored to these respective roles (Table 1). An example of tailored content, specifically the definition of AEFI, is provided in Fig. 1.

Table 1 Overview of in-person training modulesFig. 1

Tailored definition of an adverse event following immunization (AEFI) for a medical and b non-medical professionals in training materials

The training sessions were organized in each geographical study cluster involved in the GSK-led evaluation in the pilot countries, as defined by the WHO and Ministries of Health [8, 21]. The sessions targeted both study staff and community healthcare professionals within the study areas and followed a cascading ‘train-the-trainer’ model, in which external experts first educated designated trainers for both groups, who were then responsible for educating other professionals within their regions. For each cluster, one trainer was assigned to study staff and four trainers to community healthcare professionals, who were selected by the PI. The training-of-trainers session spanned 2.5 days and covered content tailored to both medical and non-medical professionals, as well as teaching methods. Emphasis was placed on andragogical principles, to ensure that adult learners would recognize the relevance of the training, feel prepared to apply the knowledge to real-life scenarios, and have their prior experience acknowledged. A significant focus was placed on participatory learning approaches, encouraging active engagement through questions and sharing insights, and role-plays to reinforce practical training application.

The standard training sessions given by the trained trainers usually lasted 2 days and included 15–25 participants. Everyone received regular refresher courses; for study staff, this was approximately every 6 months. The trainings of community healthcare professionals were followed by a ‘supportive supervision’ system. In this system, those who did not pass a knowledge assessment quiz several months after the training received targeted supervision from the trainers, including personalized feedback on the professional’s quiz responses and a supervised healthcare consultation.

2.2.2 Job Aids

Practical job aids were provided to support professionals in AESI identification, with two tailored versions: one for medical and another for non-medical professionals. These quick-reference materials outlined key signs and symptoms relevant to the study-defined endpoints or AESIs. The version for medical professionals included diagnostic guidance aligned with the AESI case definitions. For non-medical professionals, the job aids served as visual tools to help recognize concerning symptoms and to understand the procedures to follow. In some cases, a combination of symptoms needed to be recognized before determining the action to follow. Action icons indicated the recommended steps, that is, patient referral to a health center, hospital, or in case of a life-threatening emergency, taking urgent action. The job aids were introduced during the in-person trainings to ensure effective use. An example of the job aids related to intussusception is shown in Fig. S1 (medical professionals) and Fig. S2 (non-medical professionals) in the electronic supplementary material (ESM).

2.2.3 Telecourses

At study start, a series of courses was offered to PIs and medical staff in the form of live presentations by experts from the Réseau en Afrique Francophone pour la Télémédecine (RAFT) network [22]. These sessions were provided in both English and French, recorded, and made available via an online learning management platform, along with a knowledge quiz. Additionally, multiple rounds of live Q&A sessions covering various course topics were held throughout the study period. The courses covered mostly AESIs and included extracts from the medical job aid (ESM).

2.2.4 Tele-Expertise System

A web-based tele-expertise platform, named Bogou, was developed to support medical healthcare professionals in accurate diagnosis and treatment of diseases listed as AESIs. The store-and-forward system allowed healthcare professionals to post clinical case questions to central experts from the RAFT network [22]. To submit a case, the healthcare professionals were required to provide the study participant’s identification number, birth date, and gender. They were also required to provide a detailed explanation of the case and had the option to attach relevant documents. The platform was introduced during the investigators’ meeting at study start. In addition, online workshops were conducted, and user tutorials were made available on demand, with assistance provided as needed.

2.2.5 Mobile Healthcare Alert System

To enhance reporting of meningitis, cerebral malaria, and other AESIs, a mobile healthcare alert system was developed to enable field staff to send real-time alerts to the PI of their region. In addition to facilitating close monitoring, the system aimed to ensure that all procedures and tests outlined in the study protocol were completed, supporting proper diagnosis and participant safety. The system was designed as a complementary tool and did not replace the reporting through electronic case report forms.

An introductory workshop was held, and an online video tutorial and user manual were provided; a 24/7 phone helpdesk was also available for technical support. Field staff were equipped with mobile devices that had the application installed and a phone list of key contact persons. Mandatory fields included the study participant’s identification number, date of the event, type of health alert (selected from a drop-down menu), staff initials, and the identification number of the healthcare facility. Internet connection was required to submit the health alert; alerts submitted without connection were queued and automatically transmitted once connection was acquired. Upon receipt of health alert data, the PI received an email notification and could access a web application to review the case details. The system then allowed the PI to confirm review of the alert and to add comments.