Abu Dhabi now has one patient-safety system connecting every hospital in the emirate.
The headline numbers are large: 100% hospital connectivity, more than 60,000 users and over 70,000 patient-safety-related cases collected in the system’s first year.
DoH says the platform now connects every Abu Dhabi hospital, serves more than 60,000 users and has collected over 70,000 safety-related cases, creating a sector-wide dataset for earlier risk detection.
Abu Dhabi patients, healthcare providers, clinical quality teams and organisations building AI-supported patient-safety systems.
A unified taxonomy and sector-wide dataset can surface recurring safety patterns and near misses that may be difficult to detect when facilities report in isolation.
The first-year figures are reported by DoH and do not by themselves prove that AI caused a reduction in patient harm. The 70,000 cases include near misses and early-warning opportunities, not only harmful incidents.
DoH says the platform will expand to additional healthcare facilities while implementation and outcomes continue to be measured.
Patients do not need to enrol in the system. For care decisions, continue to use normal provider channels and ask clinicians how safety alerts or AI-supported tools affect an individual case.
Healthcare professionals gain a shared safety-reporting layer; patients may benefit if cross-facility patterns lead to earlier preventive action.
The system is used by Abu Dhabi healthcare facilities and professionals rather than being a public sign-up service.
Watch for published outcome data showing which risks were detected earlier, what interventions followed, and whether preventable-harm rates fall over time.
But one of those numbers needs translating immediately.
Seventy thousand cases does not mean 70,000 patients were harmed.
What the system actually connects
The Department of Health – Abu Dhabi began building the unified Patient Safety System around a common safety-data standard, with a 2025 partnership with RLDatix establishing a central portal and shared taxonomy for patient-safety reporting across the emirate.
One year into implementation, DoH says the platform now connects 100% of Abu Dhabi hospitals, as well as long-term care facilities, dialysis centres and fertility centres. More than 60,000 healthcare users can work through the system.
The point is to stop treating each facility as an isolated reporting island. When the same kinds of near misses or safety signals appear across multiple providers, a sector-wide dataset can make the pattern visible earlier.
The 70,000 figure is not a harm count
DoH says the system collected more than 70,000 patient-safety-related cases during its first year.
Around 70% represented opportunities to identify early warning signals and learn from potential risks before they resulted in patient harm.
That distinction matters. Patient-safety systems are supposed to capture near misses, hazards and weak signals as well as incidents that actually injure someone. A system that only records harm after it happens would miss much of the information needed for prevention.
Where AI enters the picture
The platform uses healthcare-sector-wide data to identify patterns and early safety signals. DoH describes the goal as a shift from reacting to incidents toward predicting and preventing them.
That does not mean an AI system is independently making clinical decisions for every patient in Abu Dhabi. The public description is about analysing safety information, surfacing patterns and helping professionals intervene earlier.
That sits alongside a wider UAE healthcare-AI shift Robius has already tracked. In July, we looked at how AI is already being used in UAE hospitals for diagnostic support. This system operates at a different layer: not diagnosing one scan, but learning from safety data across an entire healthcare network.
There are operational gains too
DoH says the unified workflow has reduced the time required to report risks, saving more than 2,000 hours of staff time.
It also reports an improvement in measured patient-safety culture among healthcare professionals from 68% to 73%. DoH says the latest result is 21 percentage points above the international benchmark it cites from the US Agency for Healthcare Research and Quality.
Those figures are useful indicators, but they are not the same thing as proving that the AI itself caused a specific reduction in medical harm. The stronger claim supported by the public data is that Abu Dhabi now has a much broader, standardised safety-reporting layer and can analyse risks across facilities rather than one organisation at a time.
Why the architecture matters
The less glamorous part of the project may be the most important.
Before the AI can find patterns, hospitals need to describe safety events in a consistent way. DoH’s 2025 announcement focused heavily on a unified taxonomy so healthcare organisations classify and interpret safety events using the same structure.
That is what makes cross-hospital analysis possible. If every provider records the same event differently, a central dashboard becomes a pile of incompatible reports.
Abu Dhabi standardised the reporting layer first, then put automation and analysis on top of it.
What to watch next
The system is still expanding beyond hospitals. The next useful evidence will be outcome data: which recurring risks were detected earlier because of the shared dataset, what interventions followed, and whether specific categories of preventable harm fall over time.
For now, the milestone is substantial without exaggerating it.
Every Abu Dhabi hospital is connected to the same patient-safety reporting system. More than 70,000 cases have fed the learning layer. And most of those cases should not be read as injuries — they are exactly the kind of warnings a prevention system is designed to catch before an injury happens.
Sources
Gulf News — first-year Patient Safety System results reported from DoH, 19 September 2026
Checked 19 September 2026. The 70,000 figure covers patient-safety-related cases, including near misses and early-warning opportunities; it should not be read as 70,000 incidents of patient harm.
Robius.news — Dubai, UAE — 2026 | Built to be first. Built to be trusted.
