Laying the foundations for high-quality mortality surveillance in Sierra Leone: Early learnings from the Child Health and Mortality Prevention Surveillance (CHAMPS) Network

Selection of CHAMPS site and catchment area

Apart from its geographical diversity and having variations in risk for major diseases like malaria, HIV, and TB^20,21, SL was selected as a CHAMPS site because it has one of the highest U5 mortality rates in the subregion⁹. With the two chiefdoms in Bombali as the catchment area, Bombali District has the second highest U5 mortality rate in the province after Port Loko District²². Bombali District also has a recurrent history of deadly disease outbreaks, including vaccine-preventable diseases and the 2014–2015 Ebola outbreak^23,24. A 2015–2016 study that evaluated multiple death reporting streams with a focus on U5 and stillbirth revealed that stillbirths and deaths in U5 were underreported in the Bombali Sebora chiefdom¹⁶.

CHAMPS defined its catchment area (In Figure 1) using the chiefdom boundaries (Bombali Sebora and Bombali Siari) of the SL Population and Housing Geographic Information²⁵. The CHAMPS SL catchment area has urban, peri-urban, and rural characteristics; population statistics were generated from the SL 2015 Census^25,26. Additional health statistics variables were available from the SL Multiple Indicator Cluster Survey (MICS) report²², demographic health survey (DHS) results²⁷, and other surveys from Statistics SL, Ministry of Health (MoH) and UNICEF⁷.

Figure 1. SL's CHAMPS Site Location.

Site research capacity

The SL's capacity for health research has grown over the years due to the 2014 Ebola outbreak, which exposed the weaknesses in the healthcare system²⁸. Before the implementation of CHAMPS activities, postmortem specimen collection and laboratory capacity at the site were limited. Although clinicians receive training for lumbar puncture and other procedures at the only medical school in the country, these are rarely conducted at hospitals. The only pathology lab in the entire country was at the Connaught Hospital (a tertiary hospital) in Freetown. However, it was not in use until CHAMPS began its MITS collection and laboratory procedures in 2019.

Several laboratory-driven studies have been conducted over the past decade in SL, including an investigation during which the Bombali virus was discovered²⁹, an Ebola vaccine trials through the STRIVE study³⁰ and University of SL Ebovac study in Kambia³¹, and Lassa fever study³². In 2019, following the CHAMPS MITS launch, the Countrywide Mortality Surveillance for Action (COMSA)³³ was implemented. COMSA has produced annual data on mortality and causes of death at national and subnational levels for use by the government and health policymakers. It uses the 2016 WHO verbal autopsy (VA) tool to track causes of death but does not conduct any diagnostics on postmortem biological samples³³.

The CHAMPS site in SL is implemented by a consortium of non-governmental organizations (NGO), unlike other CHAMPS network sites implemented by research institutions or university teaching hospitals and training institutions (Table 1). Staff are hired locally and internationally by the implementing organizations, and they were trained by the CHAMPS Program Office (at Emory University) and Barcelona Institute for Global Health (ISGlobal) to adhere to CHAMPS protocols.

SL's Ministry of Health (MoH) provides supervisory oversight and a policy framework for CHAMPS SL. Between 2017 and 2018, the CDC provided technical oversight, eHealth Africa provided fiscal management, while ICAP provided technical leadership on laboratory, histopathology, data and informatics until 2022. Since 2018, Crown Agents, a not-for-profit international development company has progressively taken over eHealth Africa’s role (2018) and ICAP's technical roles (2022). Crown Agents is therefore the lead fiscal and technical management partner for the three implementing partners: FOCUS1000 remains the lead for socio-behavioral sciences (SBS) from the outset; while World Hope International (WHI) also remains responsible for all mortality surveillance (MS) activities from the outset.

Country and catchment area characteristics

SL is located on the western shoreline of Africa, bordering Guinea and Liberia. As of 2015, the country had a population of 7,092,113⁴¹ spread across 71,740 km² (27,699 sq mi—Table 2). Although English is the official language, the most widely spoken languages are Krio, Themne, Mende, Limba, and Kono^42,43. The majority of the population (78.6%) is Muslim, with Christians making up 20.8% and traditional religion less than 1% of the population^42,44.

The CHAMPS SL site is located in Bombali District, comprising two chiefdoms—Bombali Sebora and Bombali Siari—which were initially one chiefdom before re-districting in 2017^25,42. The catchment area includes 96 communities, spread across an estimated area of 281 km², with a population of 162,383⁴¹ as of 2017. The site comprises parts of the district capital, Makeni City (population 125,970)⁴¹, with an estimated driving distance of 185 km from Freetown (Connaught Hospital), the country’s capital city (population density: 7,397/km²)⁵⁴.

The site's healthcare services are delivered at three levels: (1) primary healthcare services at the community level are provided through peripheral health units (PHUs). PHU health services are provided by trained nurses, midwives, medical technicians called Community Health Officers (CHOs), etc.; (2) secondary health care services at the district level are provided through district hospitals, which act as referral hospitals for PHUs and provide more advanced healthcare services by nurses, midwives, lab technicians, CHOs, medical officers and medical specialists, etc. (3) Tertiary-level facilities are all located in the capital, Freetown⁵⁵. The Makeni Regional Hospital, which hosts the mortuary and the main CHAMPS lab, is a district and regional referral hospital (secondary level)⁵⁵. The site catchment area comprises 28 health facilities, including the Makeni Regional Hospital (MRH) five other hospitals and the associated catchment PHUs as shown in Figure 2.

Figure 2. Spatial distribution of laboratories and health facilities in SL.

Themne is the primary language spoken within the CHAMPS catchment area, followed by Krio, then Limba.

Study design

CHAMPS is a population-based surveillance study design that combines active and passive data collection methods from multiple sources to obtain comprehensive data on the causes of child deaths.

Study population

The inclusion (eligibility) criteria for enrolment in CHAMPS are residence in the catchment area within the last four months, confirmed stillbirth or U5 death (Non-MITS), and body available within 24 hours of death for MITS.

CHAMPS implementation process

As previously described by CHAMPS` manual⁵⁶, Salzberg, et al.⁴⁷, and other CHAMPS Sites^19,37, CHAMPS implementation occurs in a stepwise process (Figure 3). CHAMPS starts with community engagement to establish trust and ensure the acceptability and feasibility of CHAMPS activities. This is followed by mortality surveillance to identify, report, and investigate all U5 deaths and stillbirths. This process includes an eligibility assessment and informed consent process with the parents or guardian of the child. Upon consent, the body is taken to the mortuary for specimen collection using a minimally invasive tissue sampling (MITS) technique; MITS is a postmortem sample collection technique that uses biopsy needles to collect fluids and key organ samples⁵⁶. These samples are taken to the lab for extensive testing. The surveillance team also visits family members and health facilities to conduct verbal autopsies using the WHO Verbal Autopsy tool⁵⁷. The team also visits all health facilities the child or mother visited during illness and conducts a standardized clinical record abstraction of the child or mother. Once all the data and lab results are collated, an expert panel is convened and reviews all the information to determine the cause of death (DeCoDe). The DeCoDe findings are communicated to families, communities, and healthcare workers as feedback. Data may also be used to inform other interventions at all levels of the health care system or nationally through policy changes that can decrease U5 deaths and stillbirths. This process is defined as "data-to-action"⁵⁶.

Figure 3. Outline of CHAMPS Stepwise Implementation Process.

Phased implementation strategy in SL

The CHAMPS SL site opted for a phased introduction of MITS to ensure sufficient time to engage with the community and address concerns adequately. The CHAMPS SL site adapted the standardized CHAMPS network protocol, procedures, and data collection tools described in Figure 3^58–60. This also involved developing a detailed CHAMPS SL protocol and Standard Operating Procedures (SOPs) and submitting them to the SL Institutional Review Board (IRB) for approval before commencing CHAMPS processes^61,62. The following phases were deployed incrementally (Phase 1 to Phase 4) to build community confidence, establish local research capacity and build laboratory and surveillance infrastructure to absorb the entire CHAMPS methods from notification, recruitment, investigation and family/community follow-up processes.

I. Formative research

Formative research was used to examine and assess critical aspects of the viability and perceptions of CHAMPS surveillance approach and factors that might hinder CHAMPS activities^64,65. Emerging themes included 1) religious beliefs, 2) cultural norms, 3) political and economic conditions, 4) disease frequency and incidence, and 6) environmental factors. Formative research addressed the difficulties caused by the underlying community stigmatization surrounding deaths during the Ebola epidemic. Before any formative research study, participants provided verbal consent after the approved standard script was read to them. Semi-structured interviews, key informant interviews (KII) and in-depth interviews, and focus group discussions (FGD) were implemented for data collection and the results were shared with the wider implementation team to inform the next phase of CHAMPS activities and processes.

II. Community Engagement (CE)

The CE team conducted weekly routine community meetings through the Community Liaison Officers (CLO) to cover all communities in each section within the catchment area⁶⁶, with a target of visiting each community at least once a month. During these visits, CLOs addressed community concerns, looked out for rumors, addressed rumors if they arose, and gave community feedback regarding CHAMPS activities. The communities regularly invited CHAMPS staff during special occasions like celebrations, thanksgiving, sporting activities, or large community meetings and gatherings in their assigned communities, which they used as platforms to discuss CHAMPS activities and answer any questions. The collaboration between community members and the CE team during community events significantly contributed to the acceptance of CHAMPS research study and the MITS procedure.

The SBS team also partnered with the district-level MOH social mobilization pillar to promote maternal and child health activities in the CHAMPS catchment area. The partnership involved collaboration during community outreach events and health promotion campaigns, such as immunization, bed net distribution, nutrition drives, and health talks. These collaborations greatly improved the positive perception of the CHAMPS program.

The use of electronic media also helped increase acceptability, especially radio discussion programs led by trusted critical district stakeholders as panelists to address MITS concerns, promote child health, and give feedback. The radio discussions were routinely aired and simulcast across five radio stations in Makeni to reach people in the entire catchment area.

The SBS team trained religious leaders using scriptural references to promote maternal and child health messaging and promote acceptance, endorsement, and dissemination of CHAMPS messages.

To approach specialized communities and populations, such as military and police personnel, who often prohibit citizens from visiting their barracks until a permit is provided. The SBS adhered to their standard operating protocols, such as sending notice letters to heads to schedule meetings. The team organized separate engagement sessions for each of these specific groups and their dependents. These trainings and continuous engagement increased confidence between the soldiers and the SBS team, allowing unlimited access to the barracks for CHAMPS activities, including MITS. The team also attended muster parades to inform personnel who are regularly rotated or moved (to other duty station), as well as to provide CHAMPS information to newly transferred personnel.

III. Rumor Surveillance

The process involves collecting and analyzing unverified information or doubtful explanation that is circulating from various sources, including community members, health workers, and media reports about CHAMPS activities⁶⁷. Each community engagement team member (CLO) conducted rumor surveillance in the section of the assigned catchment area. While doing routine community visits, a CLO, who also serves as a Rumor Surveillance Officer (RSO), gets information from Community Reporters (CRs) and other sources. CRs are approached and asked to share whatever information they have heard since their last visit. Whenever the rumor and its source were shared, the RSO followed up by talking to the individual or going to where the rumor originated to gain better understanding. Next, the RSO visited various locations in the community where people gathered, including "Ataya bases” (These are places where people gather to drink green tea and discuss burning community and national issues), "cookery baffers" (A traditional eatery or meal sale point) and the marketplace, among others. If any rumors were encountered, they were documented and addressed right away. CR and Facility Reporters (FRs) refresher training and stakeholder feedback workshops were also platforms that the team used to gather rumors. All the rumors were documented and shared with the core research team. High-priority rumors were immediately communicated to the site leadership, while other rumors were shared monthly with the team leads. Both the community engagement and core research teams prioritized the rumors-based on potential impact, and the research team created tools for investigation. Within 24 to 48 hours, the RSO and the corresponding research team member conducted a community follow-up on high-priority rumors. Low-priority rumors were addressed during the next routine visit. Examples of priority rumors from the community included the following: (1) the MITS process involved the draining all the blood from the body of a deceased child; (2) CHAMPS did MITS to remove body parts and organs from the dead children to sell for profit abroad; (3) community reporters were paid for every child that died and is reported; (4) CHAMPS was sent to the community to spread another dangerous disease, like Ebola, etc.

I. Clinical Data Abstraction (CA)

Mortality surveillance activities involved using standardized procedures and forms for clinical data abstraction. Following parental consent for MITS or Non-MITS, relevant clinical and demographic data were abstracted from maternal and child clinical records, if available at the facility. All data were entered electronically into REDCap. Data abstractors from the MS team visited all health facilities within or outside the CHAMPS catchment area where the deceased may have been managed during the illness to compile as much information as possible⁵⁶.

II. Verbal Autopsy (VA) and Social Autopsy

This was a standard MS process that collected information about the circumstances that led to the child's death. CHAMPS surveillance officers used the 2016 WHO Child and Neonatal VA tool⁵⁷, translated into two local languages, Themne and Krio. Unlike other sites, CHAMPS SL also collected social and demographic information from the deceased's household, using specially designed Social Autopsy (SA) questionnaires developed as part of an associated study on the social factors affecting stillbirths and U5 deaths⁶⁹. As suggested by formative research findings, CHAMPS conducted all VA & SA interviews within 3–14 days from death notification.

Minimally Invasive Tissue Sampling (MITS) and Diagnostics (Labs) methods

As detailed in the CHAMPS Network manual^56,74, three MITS staff members typically perform the MITS autopsy procedure (a MITS technician and two MITS assistants). This process includes preparation of the body (cleaning and sterilizing), gross examination (preparing the sample for diagnosis), anthropometric measurements (measuring the length of arm, head, leg, foot, etc.), photography, sample collection, and documenting MITS collection using the MITS collection forms. Samples such as cerebrospinal fluid, blood, nasopharyngeal secretions, liver tissue samples, lung samples, brain, placenta tissue, etc. are obtained⁵⁶. Upon request, a family member may observe the MITS procedure.

Samples collected were transported to the MRH lab, where they are accessioned and distributed to the various labs for intensive testing. These include clinical, microbiological, and molecular testing at the MRH and histopathology testing at the Connaught Hospital Histopathology Lab in Freetown. Histopathological samples were also sent to the CDC Central Pathology Laboratory (CPL) in Atlanta, for advanced testing and quality control and quality assurance. Histopathology testing was performed at Connaught Lab and the CDC Lab on formalin-fixed, paraffin-embedded tissue specimens. Findings from both labs are discussed through telepathology sessions, during which findings and diagnoses are compared among site and CPL pathologists.

At the MRH, laboratory diagnosis was made for diseases such as malaria, HIV, TB, Sickle Cell Disease, etc., using SD Bioline (rapid test), a microscope, GeneXpert, etc. At the same time, microbiology (blood and cerebrospinal fluid culture) was done using BD Bactec, and molecular testing by TaqMan array cards (TAC)⁵⁶ using the QS7 machine.

The expert panel to Determine Cause of Death (DeCoDe)

The CHAMPS SL DeCoDe panel was constituted and trained in October 2019. It consists of national and international medical experts, including a pediatrician, a neonatologist, a pathologist, an epidemiologist, a microbiologist, and an obstetrician. Local representation from the Bombali District, MRH hospital management, and clinicians who cared for the deceased were also present. All panelists underwent standard research, confidentiality, and COD determination training.

A DeCoDe case manager generates case packets from the CHAMPS database, which includes all data collected on case demographics, antemortem clinical summaries, VA, MITS data, and laboratory diagnostics. Each panel member reviews the packets before the meeting and assigns a preliminary cause of death (CoD) diagnosis, utilizing the CHAMPS Diagnostics Standards⁵⁹ and the WHO Medical Certification of Cause of Death Guidelines (ICD-PM)⁷⁵. These preliminary adjudications are discussed during the panel session to reach a consensus⁵⁹. The expert panel also determines whether the death was preventable and makes recommendations to the MOH to inform needed interventions (‘Data-to-Action') to prevent similar deaths in the future. Some actions may be at an individual or family level, while others may be at a higher policy level. The final CoD and DeCoDe panel recommendations are summarized in preparation for family and community feedback.

The site introduced case mapping to nearby health facilities, as part of the SL site DeCoDe packets. This is a unique activity of the SL site in the CHAMPS network that provides the DeCoDe panelists with a concise summary of whether there were potential delays in reaching the health facility, when reviewed in conjunction with VA summaries. The data team develops distance maps from the deceased child's home to the nearest health facility and the main referral facility (MRH) using Quantum GIS – a geographic information system software. QGIS allows the team to analyze and edit spatial information with a standardized estimated time and distance to each health facility - walking and by motorbike. The travel times applied standard estimates of the distance and time of walking (12 to 14 minutes per kilometer) and using a motorcycle (20 to 30 kilometers per hour) to develop DeCoDe distance maps.

Family and community feedback

Feedback to the family may happen at various stages during the process. This may happen within a few days of death notification for notifiable infectious diseases, or within a few months for the final DeCoDe results. The MS surveillance team member(s) who conducted the verbal autopsy are responsible for returning results to the family. Periodically, an MS supervisor and an SBS team member may accompany the team in delivering the results. They provide the cause-of-death certificate that includes findings and recommended preventive measures to the family. The team also document the responses from family members upon communicating the DeCoDe result.

CHAMPS Data-to-Action

Data-to-Action, the last stage of the CHAMPS process, included using learnings from the findings in the preceding Phases or observations from DeCoDe findings to inform actions and policies to prevent future deaths (Figure 4 and Figure 5). Actions could be on an individual or family level or at a community or national level through policy changes. The national CHAMPS Advisory Committee and the local Data-to-Action Working Group were formed with MOH, MRH, and CHAMPS staff to examine findings and recommend immediate actions to reduce U5 deaths.

CHAMPS SL explored multiple opportunities to improve the uptake and use of CHAMPS data for policy and action at community, local and national levels. The CHAMPS National Advisory Committee and Local Data-to-Action Working Group were established in 2020 to spearhead these activities. The two groups reviewed and prioritized possible implementation ideas using the DeCoDe Panel recommendations for reducing stillbirths and U5 deaths. Examples of Data to Action activities, have included clinical review meetings with the scope of sharing findings of the CHAMPS DeCoDe panel with clinicians in the CHAMPS catchment. Often led by one of the DeCoDe panel experts, the clinical review meetings have been a critical forum to share findings and recommendations and jointly explore opportunities for improvement of clinical services – from the ground up. The local hospital and PHU clinicians are best placed to help identify additional gaps and use the CHAMPS findings to consider changes in practices or interventions that could prevent future under-five deaths and stillbirths. The Healthcare Workers Forum was an opportunity to publicly present the project's findings to clinical and public health students during their training at the local university. Child death audits at the MRH have become a weekly avenue to review two or three deaths and identify recurrent preventable causes of child death at the child emergency room and neonatal care units. These avenues have resulted in more open discussion of systemic and individual failures, with a focus on finding solutions rather than distributing blame. This has been a major cultural shift within the CHAMPS catchment area.

Site data management and informatics methods

By design, the CHAMPS data is publicly and immediately available to researchers and the public at the network website⁷⁶. CHAMPS data management system uses approved tools, systems, and secure servers to maintain data integrity and preserve patient confidentiality. The REDCap⁷⁷ version 12 and SurveyCTO⁷⁸ version 2 (Open Data Kit⁷⁹ can also be used as an alternative to REDCap and SurveyCTO, since it is a free and open source tool) secure platforms are used to collect and manage the data quantitative and qualitative data respectively. Data are reviewed at multiple levels, and data quality issues are resolved using REDCap, Microsoft Excel, Stata, and other data systems. Deidentified data are further managed and backed up at the CHAMPS Program Office at Emory University in Atlanta in Lab-Key. Data outputs, including dashboards, monthly bulletins, presentations, newsletters, manuscript tables, etc., are regularly produced by the site for dissemination to key stakeholders across the country. GIS is also critical in the enrollment of cases during eligibility as well as for the DeCoDe process. GIS app, Map.me is a GIS tool installed on the tablets, along with catchment area maps, to assist surveillance officers during eligibility screening and catchment verification in enrolling eligible cases. QGIS is used to develop the DeCoDe map. Geocoordinates used are collected during the return of the body and VA procedures.

Ethics approval

Before the initiation of CHAMPS SL in 2017, a comprehensive research protocol was submitted to the Sierra Leone Ethics and Scientific Review Committee (SLESRC) at the Directorate of Training and Research for Institutional Review Board (IRB) approval. The protocol included the CHAMPS's objectives, community engagement activities, surveillance processes, specimen collection, laboratory methods and procedures, DeCoDe processes, data collection tools, consent processes, and other required documents.

In adherence to the protocol, every CHAMPS staff member completed standardized research ethics training before starting work. Caregivers of potentially eligible participants were requested to provide verbal consent before screening for eligibility processes commenced. Furthermore, parents or guardians of deceased children were approached to provide written informed consent before the MITS and non-MITS procedures. All consent discussions took place in the parent or guardian's native language.

The CHAMPS research protocol was approved by the SLESRC on 3^rd May, 2017, and subsequent CHAMPS annual progress reports are submitted to the SLESRC for annual ethics renewal. The most recent SLESRC approval number is 002/05/2024 with approved date 02/05/2024.

Phased strategy

The use of a phased implementation approach at the Sierra Leone site was essential to the success of the CHAMPS site. It allowed time to engage well with the community, reinforce surveillance systems, and build laboratory capacity and infrastructure and properly embed the study within the Ministry of Health as part of a holistic health system approach to find and resolve systemic issues. Further, the slower implementation allowed the team to integrate lessons learned at each stage of the scale-up.

Local capacity building

Throughout its implementation, CHAMPS has capitalized on its resources to increase the ability of communities, health facilities, and staff to report deaths and conduct investigations, including enhancing laboratory capacity and establishing local cause of death certification in Sierra Leone. To help strengthen the district surveillance system, CHAMPS has trained over 500 community and facility reporters on timely death reporting procedures, reporting and investigation of notifiable diseases, and reporting and investigation of other unusual events. As an important side benefit, open avenues and for a to discuss health system challenges has provided community members a space and the boldness to jointly disclose and codesign strategies to improve their community health, with support from CHAMPS field staff.

Similar to other workstreams, the local laboratory team has undergone frequent refresher trainings with each update of the standard operating procedures and were kept up to date on the latest equipment, software, and procedures. The CHAMPS research laboratory now ranks is the most sophisticated in the country, with the laboratory staff providing training and support to other district and national laboratories as part of our corporate social responsibility. The new ultra-modern laboratory facility at the MRH (commissioned in July 2022) currently houses the CHAMPS and clinical labs, providing an opportunity to upgrade the laboratory capacity of the regional hospital and the clinical care of patients. This co-location has encouraged enhanced capacity building for district laboratory staff and university students, several of whom use the CHAMPS lab for their internships. Similarly, the recently revitalized histopathology lab at Connaught Hospital remains the only functional pathology laboratory in Sierra Leone. In addition to postmortem testing of CHAMPS MITS samples, the Connaught lab has enabled the MOH to reinstate diagnostic histopathological services, now serving as the sole hispathology laboratory for the entire country. Since 2022, clinicians across all 16 districts are able to send formalin-fixed samples to Connaught for testing. Prior to making this available, such samples were typically shipped abroad for testing (if patients could afford the costs); and due to shipping delays, histopathology results sometimes were returned after a patient had passed on or their cancer had metastasized.

Enhanced innovative mortality surveillance processes

Through its continuous surveillance capabilities, CHAMPS techniques and procedures provide evidence-based information for action at the local and national levels. The continued engagement of mortuary attendants at the MRH as part of the facility reporters has enhanced sensitivity of death reporting. To further support this effort, the hospital management implemented guidelines mandating all deaths to be conveyed to the mortuary in the first instance, even if relatives prefer immediate burial due to cultural norms. All mortuary workers were required to report all these deaths to CHAMPS for potential screening. In addition, the surveillance team collaborated with the SBS team to create radio messages and jingles about death reporting. These were constantly aired on the local radio stations at peak listening times, further encouraging community reporting of deaths.

All MITS cases are initially tested for notifiable diseases, such as HIV, TB, Ebola, COVID-19, and any other priority diseases, conditions, and events. All notifiable diseases identified are immediately reported through the CHAMPS leadership and surveillance teams to the appropriate MoH structures at the district and national levels (surveillance, M&E, HIV, TB teams, etc.,). This immediate reporting, prior to full testing of each case, has helped to facilitate an integrated response action between CHAMPS and MoH. CHAMPS often supports the MOH to conduct further investigation or contact tracing, as required by national protocols.

Integrating Geographic Information System (GIS) in CHAMPS surveillance and DeCoDe processes has become a critical component of our unique processes compared to the rest of the CHAMPS network. Use of Map.me for eligibility catchment verification and QGIS for quality management and maps were utilized to effectively reduce enrolling errors in the CHAMPS eligibility process, as they aided in confirming that all enrolled cases were within the catchment boundaries. During data analysis, this has also aided in geolocating cases based on residency, such as those from various chiefdoms, by determining whether they were urban or rural. The maps integrated into the DeCoDe packages have proven extremely valuable to the DeCoDe panels by providing the information to evaluate any delays or barriers that can hinder early reporting to health facilities. The display of the distance and time required to reach the nearest health facility and regional referral hospital (MRH) has aided in the accurate recommendations to the MOH and other partners in addressing the delays that may have led to the child's death.

Family, community, facility and stakeholder feedback

The families of deceased children appreciated the one-to-one feedback sessions. These were private sessions held at their family homes, and provided opportunities to resolve misunderstandings, discussed how to improve the care of other children and sometimes advice for further testing, depending on the cause of death (e.g. for sickle cell disease, HIV, communicable diseases, etc.). At the hospital level, attendance of the management and staff at the CHAMPS DeCoDe panel meetings resulted in more in-depth discussions about mortality prevention and needed systemic changes to improve survival. As deaths were not systematically counted nor reviewed prior to CHAMPS, this has been one of the critical levers for change. The hospital matron's specific quick actions have included minor adaptations with a huge impact on neonatal and child survival, such as providing findings to the wards in charge and adapting the process for transferring neonates from the theater to the special care for baby unit (SCBU). The significant reduction in transfer times has guaranteed that those crucial minutes resulted in prompt interventions to reduce neonatal mortality.

The annual stakeholders’ summit, CHAMPS Advisory Forum, and feedback to the national and district stakeholders on CHAMPS activities have served as critical platforms to elevate CHAMPS findings to the appropriate levels, resulting in evidence-informed actions at various levels. These platforms have both enlightened and inspired representatives, including political leaders at the district and national levels, to take responsibility for preventing U5 deaths.

Dissemination of CHAMPS data to enhance public engagement or inform action, such as the clinical review meetings, healthcare worker forums, CHAMPS national stakeholder summaries, etc., were used as platforms to disseminate CHAMPS data and findings. Positive audience responses demonstrated the success of data sharing via presentations in forums, meetings, and conferences. A participatory community feedback method was established to improve participation, dramatically changing data sharing and consumption. This technique explored unique options, such as personal-interest stories, graphical representations of topics, radio discussions, and community drama plays, to vary the distribution approach and assure more extensive community participation.

Since December 2022, over 50 CHAMPS bulletins with the latest findings have been distributed to local and national leaders. In 2020, the Site designed a widely distributed online newsletter to periodically share its actions and activities with local, national, and international partners. All these CHAMPS information outlets have been used as references for evidence-based child mortality discussion at various MOH meetings and platforms, including monthly in-charge meetings, national directorate meetings and during donor discussions on future healthcare priorities.

At the national level, CHAMPS is a recognized member of key technical working groups and regularly share data with national managers and development partners. Examples include the reproductive and child health quarterly technical working group meetings and the health NGO network.

Data-to-Action activities

Data from the site has been used by CHAMPS and the MoH to achieve several quick wins by addressing immediate issues. CHAMPS SL's local Data-to-Action Committee has been more agile in addressing "low-hanging fruits," such as interventions that can be carried out locally with more focused effort and a significant impact. Such actions are implemented at multiple system levels—hospitals, communities, and districts.

Clinical review sessions for doctors and nurses as clinical feedback, facilitated by DeCoDe subject matter experts, have offered a chance for health workers to reflect on their actions throughout the management of the sick child or mother. It is used for continuous professional growth to improve clinical abilities and to remain updated on medical knowledge and standard operating procedures. It enables the identification of areas in patient care that require improvement. These meetings also improved communication and coordination by bridging communication gaps between peripheral health units and MRHs during referrals. They have been successful in supporting healthcare workers in understanding the importance of their commitments to these deaths and the impact that their level of effort will have. As a result, the MRH, notably the pediatric ward, recently commenced a quality assurance committee that meets regularly to review quality concerns affecting healthcare delivery at the hospital.

The MRH also benefited from CHAMPS efforts to upgrade the medical records management system and records storage facilities. The upgraded and furnished records room now provides better access to clinical records for both CHAMPS and clinical cases. Similarly, clinicians' documentation has anecdotally improved significantly since CHAMPS was launched, due to constant engagements, training, and regular review of patient folders for information. Other vital interventions included enhancing the daily briefing, updating, and ensuring adherence to policy guidelines and standard operating procedures, and improving infection prevention and control (IPC) at the wards.

Enhancing outbreak preparedness and response – CHAMPS Covid-19 outbreak response

During the COVID-19 pandemic, it was demonstrated how a well-funded mortality surveillance system, like CHAMPS, played a crucial role in Bombali District’s COVID-19 response. CHAMPS’s existing infrastructure provided a framework that was swiftly modified into a Bombali District preparedness and response strategy.

CHAMPS' strengths were evident across multiple response pillars: the CHAMPS SBS team was leading the mobilization and engagement on the dos and don'ts of preventing COVID-19; support to the district leadership for coordination of response activities; disease surveillance, with CHAMPS surveillance officers spearheading efforts to identify, notify and investigate all COVID-19 alerts; and logistical support through use of CHAMPS vehicles, bikes, and a conference hall for daily surveillance response meetings.

CHAMPS platform as a reliable local “Health Research Institute”

With its infrastructure and well-trained technical personnel on site, CHAMPS SL is exploring additional research opportunities to leverage these investments and advance goals linked to supporting children and better calibrating mortality rates. In 2020, CHAMPS received a supplement to collaborate with the University of Toronto to conduct minimally invasive tissue sampling (MITS) on 200 U5 deaths for the Countrywide Mortality Surveillance for Action (COMSA) project in Bo District. This project has now completed enrolment (as of February 2022) and in the process of using MITS DeCoDe information to calibrate and validate verbal autopsy information obtained by the COMSA project in their national surveillance effort. Secondly, CHAMPS is collaborating with WHO and the Fleming Fund project on enhancing clinical antimicrobial resistance testing at the MRH. The goal is to improve clinical sample testing from U5 children so as to better inform antimicrobial prescription and stewardship by local clinicians. Thirdly, the Sierra Leone site has commenced the first health and demographic surveillance system in Sierra Leone. Baseline enumeration of the entire catchment area was concluded; this will provide more credible denominators for calculate rates and act as a platform for future studies. Fourthly, the site has established pregnancy surveillance to fully monitor pregnancy outcomes prior to 20 weeks of gestation. It is hoped that this will improve early linkage to antenatal care and reduce perinatal mortality.