BMJ Open protocol maps fragmentation in health intelligence systems during public health emergencies
Public health emergencies expose a problem that is easy to miss when systems are functioning normally: information often arrives in pieces that do not fit cleanly together. In a new BMJ Open protocol, investigators set out to map how fragmentation is conceptualised, how it appears in practice, and how it has been addressed across health intelligence systems during public health emergencies.
The protocol is framed around a growing dependence on digital technologies, data infrastructures and analytical capacity. Modern emergency response now draws on surveillance, laboratory reporting, clinical data and other streams that can expand quickly in volume and complexity. When those streams remain siloed, the result is not just administrative inconvenience. Fragmentation can slow situational awareness, weaken coordination and complicate the translation of data into action when public health decisions must be made quickly.
One sentence captures the central problem plainly: fragmented intelligence can delay the response itself.
A scoping review built to define a messy concept
The planned review is designed as a scoping exercise rather than a narrow efficacy assessment. That choice fits the topic. Fragmentation is not a single disease endpoint or one operational failure; it is a systems-level concept that may be defined differently across disciplines, institutions and emergency settings. Some work may describe structural separation between data sources, some may focus on interoperability barriers, and some may address governance or workflow problems that prevent information from moving across organisational boundaries.
To handle that breadth, the review will follow Joanna Briggs Institute methodology for scoping reviews and will be reported in line with PRISMA-ScR. Those standards are well suited to a field where the first task is to map what has been studied, how terms have been used and where the evidence remains thin. The protocol also states that a comprehensive search will span multidisciplinary bibliographic databases and relevant grey literature, which should improve the chance of identifying operational reports, policy documents and non-traditional sources that often contain the most concrete descriptions of emergency information systems.
The focus is not simply on cataloguing papers. It is on assembling a usable picture of a fragmented domain.
Why fragmentation matters in emergency intelligence
Public health intelligence is only as useful as the speed and coherence with which it can be interpreted. During emergencies, decision-makers need integrated information on transmission patterns, resource availability, clinical severity and population risk. When those inputs are separated across systems that do not communicate well, the analytical burden shifts to humans, who must reconcile inconsistencies under pressure.
That friction has direct bedside and population-level consequences. Data delays can blunt outbreak detection. Inconsistent definitions can obscure trends. Poorly linked systems can leave decision-makers working from partial pictures just when completeness matters most. Fragmentation can also create uneven visibility across jurisdictions, especially when one part of a system has stronger digital capacity than another.
The protocol does not claim to have resolved these problems. It aims to make them visible in a structured way, which is the necessary first step before any durable improvement can be designed.
A framework intended to move from description to action
A key output will be an evidence-informed conceptual framework. That is an important feature of the protocol because fragmentation can be discussed at many levels at once: data architecture, organisational governance, technical interoperability, analytic workflows and emergency coordination. Without a framework, the field risks repeating the same vocabulary while describing different failures.
By building a framework from the mapped evidence, the review may help clarify which dimensions of fragmentation recur most often and which interventions have been proposed to reduce them. In practice, that could support future research design, measurement choices and system improvement efforts. It may also help distinguish between technical integration problems and broader coordination failures that appear similar on the surface but require different remedies.
A second sentence lands with less ceremony and more urgency: systems that cannot talk to each other will not behave like one system in a crisis.
Limits already visible in the protocol
The protocol also acknowledges constraints that will shape interpretation. Restriction to English-language publications may exclude relevant work published elsewhere, particularly in regions where emergency intelligence systems have developed under different infrastructure and governance conditions. Heterogeneity in definitions and measurement of fragmentation presents another problem. A term used to describe data silos in one paper may refer to inter-agency coordination failures in another, and those differences will complicate comparison.
These limitations do not weaken the rationale for the review. They define the edges of what the evidence can support. A scoping review cannot create uniformity where the field itself remains inconsistent, but it can expose that inconsistency and show where the conceptual gaps sit. In a field built around rapid response, that kind of map is often more useful than another isolated description of the same problem.
The value of the project lies in its structure. By combining a comprehensive search strategy with a scoping methodology, the investigators are setting out to capture not just what is known about fragmentation in health intelligence systems, but how the concept has been framed and where current approaches still fall short. The final framework may help align future work around shared definitions, clearer measurement and more coherent system design during public health emergencies. For now, the protocol makes one point unmistakable: emergency intelligence can only function at the speed of its weakest connection.