Digital surveillance tools have transformed what is technically possible in disease detection and response. What was once a process that took days compiling facility reports, aggregating case counts, producing an epidemiological summary can now happen in minutes with properly configured digital infrastructure.
The question is not whether digital tools improve surveillance speed. They do. The question is: what does "real-time" actually mean in resource-limited health systems, and what does it require beyond the technology itself?
What Real-Time Surveillance Actually Means
"Real-time surveillance" is a term that has been adopted so widely and applied so loosely that it risks losing meaningful content. In a global health context, real-time surveillance does not mean continuous data streaming with millisecond latency. It means that surveillance data is available to decision-makers in a timeframe that allows them to act on it before the decision window closes.
For epidemic-prone disease surveillance, the relevant decision windows are:
- Immediate notification (within 24 hours): When a case of a Category 1 epidemic-prone disease viral haemorrhagic fever, meningococcal disease, yellow fever is detected, the LGA DSNO and State Epidemiologist need to know within 24 hours, not at the end of the weekly reporting cycle.
- Alert threshold crossing (within 48 hours): When the number of cases in a facility or LGA crosses the outbreak alert threshold, the investigation team needs to be deployed within 48 hours of the alert not when the monthly aggregate data is reviewed.
- Response coordination (within 72 hours): When investigation confirms a probable outbreak, the coordination mechanism laboratory support, rapid response team deployment, risk communication should be activated within 72 hours of confirmation.
These decision windows define what "real-time" means in practice. A surveillance system that delivers data within these windows is a real-time system, regardless of what technology it uses. A system with sophisticated digital infrastructure that delivers data outside these windows is not.
The Digital Surveillance Stack
A modern digital surveillance architecture in West Africa typically combines several tools, each serving a distinct function:
DHIS2: Aggregate Reporting and Programme Monitoring
DHIS2 handles routine disease surveillance reporting weekly facility reports, monthly aggregate case counts, programme performance monitoring. Its strength is aggregation and trend analysis across large organisational unit hierarchies. A national programme manager in Abuja can see facility-level reporting completeness in Imo State without requesting a report from the state level. For detailed configuration guidance, see What is DHIS2? and DHIS2 Tracker Configuration for Outbreak Surveillance.
SORMAS: Outbreak Case Management and Contact Tracing
SORMAS (Surveillance Outbreak Response Management and Analysis System) is purpose-built for active outbreak response. It manages individual case records, contact lists, follow-up schedules, and laboratory specimens through a structured workflow designed for the rapid pace of outbreak investigation. Where DHIS2 manages programme data, SORMAS manages event data the individual-level, time-critical information needed during active outbreak response.
Community Surveillance Applications
Community surveillance detection of health events at community level before they present to formal health facilities is the earliest point in the surveillance chain where digital tools can accelerate detection. Mobile-based community reporter applications allow community health workers and community informants to report suspected cases or unusual health events in near real-time, triggering immediate investigation before facility-level reporting would have identified the cluster.
In the polio eradication programme, community informant networks with over 2,960 informants trained and equipped were a key source of acute flaccid paralysis case notification that preceded formal facility-level reporting in a significant proportion of detected cases.
Geographic Information Systems (GIS)
GIS mapping of case distribution transforms surveillance data from a table of numbers into a spatial picture of transmission. For outbreak response, geographic visualization enables immediate identification of transmission clustering, proximity to vulnerable populations, and travel corridors that may be amplifying spread. DHIS2's built-in mapping module handles standard thematic maps; for more complex spatial analysis, integration with dedicated GIS platforms adds capability.
The Non-Digital Prerequisites
The seduction of digital surveillance tools is that they appear to solve problems that are actually about people, processes, and power not technology. A SORMAS instance with no trained investigation officers does not accelerate outbreak response. A DHIS2 dashboard that no one opens does not inform decisions. A community surveillance application used by untrained community reporters produces noise, not signal.
Real-time digital surveillance requires four non-digital foundations:
1. Trained human capacity at every level
The data entry officer at the facility level needs to be able to operate the reporting tool and understand which patients meet case definitions. The LGA DSNO needs to interpret the dashboard and initiate the correct response protocol when an alert fires. The state epidemiologist needs to coordinate specimen transport, laboratory communication, and national notification simultaneously. Technology accelerates the work of trained people; it does not replace the training.
2. Connectivity that matches the surveillance design
A surveillance system that requires real-time data synchronisation to function cannot be deployed in areas with intermittent connectivity without designing an offline capability and a synchronisation protocol. The design must match the infrastructure. This requires an honest assessment of connectivity conditions in the deployment area not an assumption that mobile network coverage maps reflect the connectivity that health workers actually experience at facility level.
DHIS2's Progressive Web App (PWA) mode and SORMAS's offline functionality provide offline data capture with synchronisation when connectivity is restored. These features must be explicitly configured, tested in offline conditions, and included in user training not assumed to work automatically.
3. Response protocols that connect detection to action
The fastest detection system in the world has zero value if the detection does not trigger a defined, authorised response. Every alert threshold crossing must have a corresponding protocol: who is notified, who initiates investigation, what is the investigation checklist, what is the escalation pathway, what resources can be deployed and by whose authority.
Designing the response protocol is as important as designing the detection system. The 60% reduction in outbreak confirmation time achieved in the WHO programmes I support was not entirely a function of faster data it was also a function of faster decision-making once the data arrived. Response protocols that eliminated ambiguity about who does what, when, were as important as the digital tools that delivered the signal.
4. Power and device infrastructure
Digital tools require power. In primary health care facilities in Nigeria, grid power is frequently unreliable. Facilities without reliable power cannot run digital systems without a separate power solution solar panels, battery backup, or generator access. Device procurement without power infrastructure is a recurring implementation failure that is obvious in hindsight and preventable with an infrastructure assessment upfront.
Digital Surveillance and Data Quality
A digital surveillance system is not a data quality system. It is a data management system. It stores, transmits, and displays whatever is entered into it accurately or inaccurately, completely or incompletely.
The transition from paper-based to digital surveillance frequently produces an initial deterioration in apparent data quality not because quality actually worsens, but because digital systems make incompleteness and inconsistency visible in ways that paper-based aggregate reporting conceals. A state that previously received 78% of its paper reports and noted it as a minor compliance issue now sees a completeness rate of 78% displayed in red on a real-time dashboard, triggering escalation.
This visibility is a feature, not a bug. The digital system has not created a data quality problem it has made an existing problem visible. But it requires preparedness: programme leadership needs to be ready for the initial visibility of problems that digital implementation reveals, and to respond analytically rather than defensively.
For a comprehensive treatment of data quality in digital surveillance systems, see DHIS2 Data Quality: How to Build Systems That Produce Reliable Data.
What the Future of Digital Surveillance Looks Like
The trajectory of digital surveillance in West Africa is moving in three directions simultaneously:
Artificial intelligence and predictive analytics
Machine learning models trained on historical surveillance data can identify outbreak signals earlier than threshold-based alert systems detecting the subtle changes in case distribution, age profile, or geographic clustering that precede a measurable increase in case counts. Early pilots of AI-assisted signal detection are underway in several African health systems. The limiting factor is data quality: predictive models are only as good as the historical data they are trained on, and surveillance data from resource-limited settings carries systematic biases and gaps that models must be designed to accommodate.
Interoperability between systems
The vision of a fully integrated surveillance architecture where laboratory results automatically update case records in SORMAS, which feeds aggregate counts into DHIS2, which triggers notifications in a response coordination platform is technically achievable but requires significant investment in system integration standards and API development. FHIR (Fast Healthcare Interoperability Resources) is the emerging standard, and investment in FHIR-compliant system design now is an investment in interoperability at scale.
Community surveillance at scale
The expansion of smartphone access and mobile network coverage in West Africa is steadily extending the reach of community-based surveillance. As community informant networks become increasingly digital with smartphone-based reporting replacing paper-based community registers the detection frontier moves closer to transmission events in real time.