Realist Evaluation of the Early-Stage Implementation of a Smartphone-Based Disease Surveillance Project in Two Armed-Conflict Communities in Nigeria

"The advantage of involving local community members in security-constrained areas who understand the terrain and various routes to avoid direct encounters with insurgents has been crucial to the project's success."

A growing body of literature focuses on how mobile phones can help eradicate some of the challenges that hamper quality healthcare services in communities experiencing armed conflicts in developing economies. This paper investigates the early-stage implementation of a smartphone-based acute flaccid paralysis (AFP) surveillance project in two rural communities experiencing armed conflicts in Northern Nigeria. AFP is a significant indicator of the existence of the wild poliovirus.

The smartphone-based surveillance project is a short message service (SMS)-based innovation developed to detect and report suspected cases of AFP within local communities. The project became necessary to avoid AFP recurrence after the World Health Organization (WHO) declared Nigeria polio free in August 2020. Security challenges in the local communities in places like Borno state make leveraging mHealth technology essential to ease the challenges for primary healthcare centres, governments, and inter-governmental organisations to effectively detect and report suspected cases to avoid possible outbreaks. Engaging local community members who are both familiar with the insecure terrains and who understand local traditions is a key strategy for strengthening AFP surveillance. As part of the project, training was provided to local community volunteers to manage problems such as mistrust and lack of security as they carried out AFP surveillance, based on a house-to-house search, through the smartphone-based surveillance application installed on an android phone. A 30-second video clip is embedded in the app that prompts the informants as a weekly reminder to submit their report.

In this study, the stakeholder groups involved in the project comprised the communities, community health workers (CHWs), governments, inter-governmental organisations, community health informants (CHIs), and disease surveillance and notification officers (DSNOs). Studies in the information systems (IS) literature highlight how the role of multiple and diverse stakeholders in IS implementation projects impacts the set project outcomes and can pose challenges. To understand the perspectives of these stakeholders, the researchers used the following: a qualitative case study method, realist philosophy, semi-structured interviews with 33 research participants, participatory observation, and thematic data analysis.

The findings reveal mechanisms (sense of connectedness, ownership, fear) and contextual conditions (infrastructural challenges, healthcare workforce, security constraints) influencing early-stage implementation of the AFP surveillance project. For example:

• Some remote settlements were inaccessible during the peak of the Boko Haram insurgency. However, the project seems to have thrived against these odds. Once communication is established with the CHI, the informant brings the suspected case to a safer location in that settlement. The CHIs are familiar with the terrain and routes, enabling them to avoid a likely confrontation with insurgents. That said, there has been a growing concern about local bandits in Adamawa state who are known to physically assault individuals and rob them of their mobile phones and other valuable possessions. One of the research participants noted that "it affects our performance because the yan shila cause us to fear,...but there is nothing we can do. It is in the hands of law enforcement agency."
• Armed conflict has caused massive displacement of people, especially in rural communities; smartphones offer opportunities to bridge the gap between community members and foster an improved sense of connectedness. In the smartphone-based surveillance project, the deployment of smartphones, free monthly airtime, and closed user group (CUG) services increased communication and connectedness among informants. The improvement of communication among CHWs and CHIs has led a timelier response to suspected cases and more efforts to follow up on each case until the end. However, good network coverage is essential to experience these benefits.
• Disease surveillance of AFP cases used to be predominantly paper-based, which is characterised by numerous logistical challenges that have contributed to missing some suspected cases of AFP, which might have been true cases. Implementing the project helped address most of these challenges because of the involvement of CHIs assisting CHWs in identifying and reporting suspected AFP cases. The adoption of the smartphone technology to alleviate this process has resulted in more timely data reporting with relative ease for the CHWs.
• Active case search within communities to find and report suspected AFP cases and to create awareness among community members has created an opportunity for the informants to build and maintain relationships, primarily through meetings. Informants themselves become more aware of prevailing diseases as they are being discussed in these meetings. They also share knowledge and ideas, as well as discuss and address some prevailing challenges they face. In addition, the DSNO at the state and local government level became more familiar with the informants in local settlements.
• Community sensitisation is usually organised by the CHWs and CHIs periodically as part of their voluntary endeavour. The translation of the surveillance application into different local languages has been an essential driver for improved community sensitisation. The video playback embedded in the application showing a child with AFP symptoms has helped informants create awareness and educate community members about the effects of AFP. The video is translated into different local languages, including Hausa, Yoruba, Igbo, Kanuri, Fulfulde, French, and Arabic. Thus, community members become more receptive to the project and health workers.
• Despite the high unemployment rate, the project implementation provides idle youth with opportunities to volunteer and receive cash, smartphone, and airtime incentives. The provision of smartphones has also resulted in a sense of fulfilment and belonging. The CHIs, in particular, have reported a sense of increased prestige in their respective communities based on their involvement in the surveillance project. Community members have developed trust in the selected CHIs in the communities, and they become the "go-to" persons for health emergencies or health-related information.
• The implementation of the smartphone-based surveillance project was not without challenges, especially regarding the aspect of logistics (e.g., poor network reception or downtime). Despite the poor state of infrastructure, especially in rural areas, the use of smartphones and other related technologies for smartphone-based surveillance has raised enthusiasm among the CHWs and CHIs involved. The simplicity of the smartphone surveillance application has encouraged users in the uptake and utilisation of the application to submit reports efficiently.

In short, more surveillance coverage has been achieved with this project by recruiting key community members to report suspected cases through smartphone use. This engagement has resulted in achieving near real-time updates from the community level to the national level of the health system. The relevant stakeholders involved receive notification updates and closely monitor the status of the investigation. This communication enables the superior health officers to follow up and ensure that all cases have been investigated and closed in the shortest possible time. This process is essential to ensure that an outbreak does not occur.

Based on the findings, the researchers postulate a model that explains the factors influencing the early-stage implementation of the smartphone-based AFP surveillance project in security-constrained communities. The model includes these propositions:

1. The availability of relevant infrastructures (C = context), such as good network quality and functional smartphones, fosters a sense of connectedness (M = mechanism), as well as accountability and ownership (M) among health workers. This results in improved community engagement and mobilisation (O = outcomes) among the health workers and impacts the cost of maintenance and replacement of smartphones (O).
2. In the context of high security challenges (C), there is a heightened sense of fear and apprehension (M) among health workers during active case surveillance. This results in technology improvisation to avoid known conflict areas and achieve high surveillance coverage with timeliness and uncompromised detection rates (O).
3. In the context of a high unemployment rate (C), the implementation of the mHealth surveillance project offers opportunities to community members and volunteers, which fosters a sense of fulfilment and social prestige (M) within their communities. This reduces health workers' workload (O), thereby leading to improved job satisfaction and efficiency.

The paper concludes with recommendations for the project designers of disease-specific surveillance interventions in armed conflict communities:

• Ensure that CHWs and CHIs are involved early in the project planning.
• Identify and manage conflicting stakeholder needs and interests from the onset.
• Integrate the traditional or informal healthcare system.
• Involve DSNOs in the selection and recruitment of community members to ensure the right people are selected and facilitate workforce efficiency.
• Keep in mind that incentives play an essential role to motivate CHWs and CHIs for better work performance.