(C) PLOS One
This story was originally published by PLOS One and is unaltered.
. . . . . . . . . .
Establishing an early warning event management system at Africa CDC [1]
['Kyeng Mercy', 'Africa Centres For Disease Control', 'Prevention', 'Division Of Surveillance', 'Disease Intelligence', 'Addis Ababa', 'Stephanie J. Salyer', 'United States Centers For Disease Control', 'Division Of Global Health Protection', 'Atlanta']
Date: 2024-08
Abstract Africa is home to hotspots of disease emergence and re-emergence. To adequately detect and respond to these health threats, early warning systems inclusive of event-based surveillance (EBS) are needed. However, data systems to manage these events are not readily available. In 2020, Africa Centres for Disease Control and Prevention developed an event management system (EMS) to meet this need. The district health information software (DHIS2), which is free and open-source software was identified as the platform for the EMS because it can support data capture and analysis and monitor and report events. The EMS was created through a collaborative and iterative prototyping process that included modifying key DHIS2 applications like Tracker Capture. Africa CDC started piloting the EMS with both signal and event data entry in June 2020. By December 2022, 416 events were captured and over 140 weekly reports, including 19 COVID-19 specific reports, were generated and distributed to inform continental awareness and response efforts. Most events detected directly impacted humans (69%), were considered moderate (50%) to high (29%) risk level and reflected both emerging and endemic infectious disease outbreaks. Highly pathogenic avian influenza, specifically H5N1, was the most frequently detected animal event and storms and flooding were most frequently detected environmental events. Both data completeness and timeliness improved over time. Country-level interest and utility resulted in four African countries adapting the EMS in 2022 and two more in 2023. This system demonstrates how integrating digital technology into health systems and utilising existing digital platforms like DHIS2 can improve early warning at the continental and country level by improving EBS workflow.
Author summary Disease surveillance data is critical for outbreak response, decision making and public health planning and program evaluation. The timeliness and completeness of this data both from indicator-based surveillance and event-based surveillance plays a critical role in the promptness of containing outbreaks and health emergencies at source. Digitizing surveillance systems has shown to improve on the timeliness of surveillance data as well as the accuracy and completeness. While considerable investment has been made to digitize indicator-based surveillance data through the integrated disease surveillance and response strategy in some African countries, there still exist a significant gap in digitizing event-based surveillance systems. In addition, several countries are using the DHIS2 for their indicator-based surveillance, however this tool has not been exploited for event-based surveillance workflow. There are also challenges with interoperability as several tools deployed in countries are not interoperable with existing systems rendering data triangulation and deep analytics challenging. In this paper we show how the DHIS2 through an integration approach could be used to support event-based surveillance processes for prompt detection and reporting of health threats.
Citation: Mercy K, Salyer SJ, Mankga C, Hedberg C, Zondo P, Kebede Y (2024) Establishing an early warning event management system at Africa CDC. PLOS Digit Health 3(7): e0000546.
https://doi.org/10.1371/journal.pdig.0000546 Editor: Gilles Guillot, CSL Behring / Swiss Institute for Translational and Entrepreneurial Medicine (SITEM), SWITZERLAND Received: December 27, 2023; Accepted: June 8, 2024; Published: July 8, 2024 This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication. Data Availability: All data analysed in this manuscript are included in Table 1 to Table 6. Funding: The author(s) received no specific funding for this work. Competing interests: The authors have declared that no competing interests exist.
Introduction The African continent is home to hotspots for emerging and re-emerging diseases. In fact, all seven public health emergencies of international concern (PHEICs) declared as of 2023 have either emerged from or have severely impacted African countries. Zoonotic emerging infectious disease risk is especially elevated in forested tropical regions experiencing land-use changes and where there is high mammal species diversity [1]. With Africa losing an estimated 3.9 million hectares of forest a year between 2010 and 2020 [2] these trends in disease emergence are likely to continue. Africa is also home to 22 of the 25 most vulnerable national health systems in the world, as suggested by an analysis of infectious disease outbreak risk factors [3]. This indicates a need to strengthen early warning alert and response (EWAR) capacities in Africa at the continental, regional, country, and community level to timely and effectively detect and respond to health threats such as disease outbreaks. Event-based surveillance (EBS) supports EWAR by detecting events of public health importance early and identifying the scope and magnitude of the events needed to inform response actions [4–5]. This was seen when epidemiologists took note of the first cases of haemorrhagic fever reported in Uganda in September 2022 [6] or when the first clusters of severe pneumonia were reported from Wuhan, China health facilities just prior to the eventual discovery of SARS-CoV-2 [7]. The Africa Centres for Disease Control and Prevention (Africa CDC) used the EBS modality of media scanning to monitor the spread of COVID-19 globally as well as the initial detection and progression of COVID-19 in Africa [8]. COVID-19 is not the last pandemic, and many global initiatives focus on early detection and prevention of the next pandemic. Globally, disease surveillance has seen cycles of crisis and neglect for decades. The COVID-19 pandemic proved that low- and middle-income countries (LMICs) in Africa are at a particular disadvantage in a crisis compared to other continents, with intense competition and high/exorbitant prices for vaccines and personal protective equipment (PPE) as well as fewer resources to withstand economic and social shocks [9–10]. In addition to the EBS work Africa CDC is doing at the continental level, over 20 African countries implemented its EBS framework [11–12]. However, very few information systems exist to systematically record and manage EBS data—including the need to support critical EBS functions like monitoring the timeliness of detection, alert, and response. In response to this need, Africa CDC in collaboration with the Health Information Systems Program South Africa (HISP-SA) developed an event management system (EMS) with initial funding from the African Union to support the implementation of EBS at the continental level and also provide a solution to countries interested in improving their EBS related workflow. This article describes the system developed and its use in conducting EBS at Africa CDC, as well as its early adaptation in African countries.
Methods Prior to development, Africa CDC and HISP-SA identified the following system requirements, or objectives, for the EMS: Is open source, web-based, mobile, adaptable, and an affordable platform.
Supports standard EBS workflow, inclusive of the five key EBS steps of detection, triage, verification, risk assessment, and alert described in the Africa CDC EBS Framework [13].
inclusive of the five key EBS steps of detection, triage, verification, risk assessment, and alert described in the Africa CDC EBS Framework [13]. Allows for effective data capture and storage of EBS-related data, including the ability to capture frequently used resource data (e.g., population, pathogen characteristics, socio-economic determinants of health, health sector capacity, and relevant research data).
of EBS-related data, including the ability to capture frequently used resource data (e.g., population, pathogen characteristics, socio-economic determinants of health, health sector capacity, and relevant research data). Allows for data visualisation and analysis , including the production of automated products tailored towards specific populations (e.g., social media messaging).
, including the production of automated products tailored towards specific populations (e.g., social media messaging). Produces and stores automated reports (e.g., situation reports, briefs, etc.), inclusive of data visualisation like maps, tables, and charts.
(e.g., situation reports, briefs, etc.), inclusive of data visualisation like maps, tables, and charts. Improves timeliness of EBS , ensuring the inclusion of standardised monitoring and evaluation indicators like the Resolve to Save Lives 7-1-7 metrics [12,14].
, ensuring the inclusion of standardised monitoring and evaluation indicators like the Resolve to Save Lives 7-1-7 metrics [12,14]. Incorporates a One Health approach by including human, animal, and environmental events that have a potential impact on public health.
by including human, animal, and environmental events that have a potential impact on public health. Links to, integrates with, or can share data with other systems to improve EWAR related data sharing and collaboration with African Union (AU) Member States, global and regional bodies, and partners. Systems include: ○ WHO’s Epidemic Intelligence from Open Sources (EIOS) [15]. ○ Country-level indicator-based surveillance (IBS) digital solutions like eIDSR for notifiable diseases that exceed routine reporting thresholds. ○ Data or data systems for emergency response teams, Emergency Operation Centres, national call centres, etc. to improve linkages to response efforts.
to improve EWAR related data sharing and collaboration with African Union (AU) Member States, global and regional bodies, and partners. Systems include: Is adaptable for use at the regional and country level .
Assists with the use of EMS data for action to inform policy, advocacy, response efforts and assess the impact of EBS. The inclusion of these system requirements in the EMS development along with the collaborative and iterative prototyping approach used to develop it are described in this manuscript. We listed the DHIS2 applications used, modified, or created as part of the EMS development by the following application types: data entry, data visualisation and reports, and system configuration and maintenance. We further described the early implementation of the EMS by analysing aspects of the data recorded between June 2020 to December 2022. Data elements and indicators evaluated include staff using the system; number, type, and risk level of events detected by year; the most frequently reported events by year; and data completeness and timeliness indicators. For data completeness, we looked at the following variables: event start date, detection date (by Africa CDC and country), event reported date (by Africa CDC and country), verification date, final risk assessment, lab result status, laboratory confirmation date and event intervention date (an intervention is defined as any response action taken by Africa CDC and country e.g., EOC activation, staff deployed, etc). Data, inclusive of both signals and events captured, were exported from the EMS into a comma-separated values (CSV) file and analysed in MS Excel. Signals are defined as the initial detection (by IBS or EBS) of a potential public health event, prior to verification. Events are “a manifestation of disease or an occurrence that creates a potential for disease” [13]; events can be infectious, zoonotic, food safety, chemical, radiological or nuclear in origin, and are transmitted by persons, vectors, animals, goods/food or through the environment. In the context of EBS and EWAR, an event refers to a signal that has been verified. For the country-level adaptations, we described the system attributes and indicators that were adapted, and the timeline associated with country level development and implementation. No personally identifiable information was captured as part of this analysis. The United States Centers for Disease Control and Prevention (US CDC) deemed this project non-research and a routine public health practice.
Discussion Prior to the EMS, Africa CDC used ad-hoc tools that were work intensive and required repetitious formatting and report production, reducing the time available for other critical EBS-related work. The EMS automated these tasks and enabled Africa CDC staff to collaboratively focus on detection, analysis, risk assessment, and response to events. The system allows for a multisectoral, One Health approach to EBS as it captures events affecting multiple populations and environments; this approach improves awareness of potential public health events like zoonotic outbreaks before they impact human health. EMS report generation has increased and diversified since inception showing the utility of this application and the potential for additional products [25]. Data entry completeness and timeliness metrics have improved over time. Completeness of key dates that rely on input only from the EMS users (i.e., the first 7 variables in Table 4)—as compared to those where the data is provided by sources outside the EMS system (e.g., lab)—are particularly striking from 2020 to 2022, with the proportion of missing key dates decreasing from an average of 31% to < 1%. Also, the percentage of events meeting the proposed timeliness indicators has increased from an average of 11% in 2020 to 44% in 2022. These improvements from 2020 to 2022 are likely due to multiple factors like the adding of system prompts for data inaccuracies and increasing the skill level and experience of the EMS users over time through supportive supervision and multiple training sessions. Ultimately Africa CDC has a goal to have at least 80% of events meeting the described timeliness indicators outlined in the EBS framework [12–13], and there is room to improve. Of particular concern is the low number of events, especially high and very high-risk events, where Africa CDC intervention dates are not recorded. The lack of data could indicate a true lack of action, or potentially highlight a need to link the EMS with the systems being used by the Africa CDC Emergency Preparedness and Response Division for monitoring response related activities. Improvement could be made by further incorporating automated alerts when critical fields are left blank or by creating mandatory data entry fields for critical dates. Supportive supervision and mentoring may also need to be strengthened to improve consistency across the team for routine EBS tasks—for instance, there was variability seen in the initial risk assessment level determined for the same disease (Table 3). An automated M&E dashboard and report have been slated for the next phase of development. This will ensure real-time timeliness tracking, which could also help improve timeliness as an alert component could be set up to bring more attention to events not being addressed in a timely fashion. There is also a need to expand the current linkages of the EMS to other data sources. Specifically, there is an interest from both Africa CDC, and AU Member States in adapting the EMS, to link this system to other existing systems housing IBS data being collected as part of IDSR or other routine surveillance systems. Linking these data can help identify existing and predict future endemic disease outbreaks with the use of reporting thresholds overlaid on historic data. Other areas include linking to additional social media applications, laboratory and genomics data sets, and big data to expand the detection aspects of the EMS. To address the long-term maintenance and sustainability of the EMS, following the development, HISP-SA conducted a knowledge transfer to the Africa CDC system managers. At the country-level, countries have set up country teams with expertise in managing DHIS2 who now support country customization and maintenance. Since the EMS was developed, other systems have been created or expanded their use case to incorporate event management. These include systems like EMS2 developed by WHO for use by WHO headquarters and regional office staff. This system is marked for the eventual expansion to the country-level in the coming years. SORMAS is also another open-source system initially developed to support IBS related data entry but has since been evaluated and adapted to support EBS data and event management [26–27]. Given that Africa CDC has set a goal to establish EBS, possibly using a variant of the EMS, in most AU Member States by 2030, it will be important to be aware of and in alignment with other EMSs established in Africa. Continued interoperability and standardisation of key EBS data elements will be critical for successful integration and regional collaboration around event detection and response in Africa.
Conclusion Africa CDC developed and implemented an EMS that would help streamline the existing EBS workflow and improve continental and AU Member State EWAR. While there is still much more work that needs to be completed to ultimately envision this goal, most initial objectives have been addressed with the advent of the Africa CDC EMS. The EMS provides an EBS data management solution not previously available that can be adapted to support other global regions and countries. It is in alignment with and supports Africa CDC’s effort to strengthen data standards and data sharing across Africa’s disease surveillance community, and the system is expected to be steadily improved over the coming years. Areas considered for EMS expansion include strengthening linkages to IBS and other data sources to improve overall epidemic intelligence and early warning capacity. Additionally, the inclusion of more response related functions within the EMS could improve the utility and use for response related staff and thus the communication between and timeliness associated with surveillance and response efforts. With the apparent timeliness limitations posed by conducting EWAR activities at the continental and global level that are so dependent upon country-level data, building adaptable open-source systems like this will only improve EWAR overall as they are further established at a national level. Thus, advocacy and support for the continual establishment and improvement of systems like this are desperately needed.
Acknowledgments The authors would like to thank the US CDC GDDOC (Ray Arthur, Christine Manthey, James Fuller)), WHO (Johannes Christo Schnitzler (EIOS)), UKHSA (Ashley Sharp, James Elston (SITAware)), Nigeria Centers for Disease Control (Womi-Eteng Oboma Eteng (SITAware)), European Commission Joint Research Centre (Luigi Spagnolo (EIOS)) and staff who were part of the initial consultation for the development of the Africa CDC EMS. In addition, the authors appreciate the entire epidemic intelligence team of the Africa CDC for their support in piloting the system and providing feedback for improvements.
[END]
---
[1] Url:
https://journals.plos.org/digitalhealth/article?id=10.1371/journal.pdig.0000546
Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.
via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/
