Cybersecurity/Ireland
After the ransomware attack, surgery within 48 hours fell from 75.3 to 69.5 percent
Ireland's national hip fracture audit shows what a ransomware attack cost: surgery within 48 hours fell from 75.3 to 69.5 percent.
Using the Irish Hip Fracture Database with 99 percent coverage for 2021, researchers measured the effect of the May 2021 Health Service Executive cyberattack on the seven Irish Hip Fracture Standards. Surgery within 48 hours declined significantly from 75.3 to 69.5 percent, pressure ulcer rates worsened from 2.1 to 3.3 percent, and orthogeriatric review improved from 80.9 to 85.7 percent.
What happened
On 14 May 2021 the Health Service Executive of Ireland was hit by a cyberattack that took out its IT infrastructure entirely. Radiology, blood tests, patient management systems and much else stopped working across the national health service.
The cost of such attacks is usually reported in euros, in downtime hours, and in records breached. What they cost patients is harder to establish, because hospitals rarely have a measurement system running through the outage.
Ireland does. The Irish Hip Fracture Database, run by the National Office of Clinical Audit, tracks seven defined standards of hip fracture care across all Irish hospitals. Researchers at Tallaght University Hospital and the National Office of Clinical Audit used it to measure what the attack did.
Surgery within 48 hours fell from 75.3 percent to 69.5 percent.
The test
Data were obtained from the Irish Hip Fracture Database for all Irish hospitals, covering the seven Irish Hip Fracture Standards before the cyberattack and at incremental periods afterwards.
There was 99 percent data coverage for 2021, which is exceptional for any national audit and is what makes the comparison meaningful.
Hip fractures are a good choice for this question. They are common, they affect an elderly and vulnerable population, the care standards are defined and time-critical, and the outcome measures are unambiguous. Surgery within 48 hours is a standard because delay beyond it produces worse outcomes.
One methodological detail matters a great deal: data were entered prospectively by each hospital audit coordinator before the attack and retrospectively afterwards.
The numbers
| Standard | Before | After | p |
|---|---|---|---|
| Surgery within 48 hours (IHFS 2) | 75.3% | 69.5% | <0.001 |
| Ward admission (IHFS 1) | 27.7% | 24.9% | 0.095 |
| Pressure ulcer rate | 2.1% | 3.3% | 0.024 |
| Orthogeriatric review | 80.9% | 85.7% | 0.001 |
| Data coverage, 2021 | 99% | ||
| Data entry before attack | prospective | ||
| Data entry after attack | retrospective | ||
The result
In the matched analysis, surgery within 48 hours declined significantly from 75.3 to 69.5 percent at p below 0.001. Ward admission showed a trend toward deterioration, 27.7 to 24.9 percent at p = 0.095, which did not reach significance.
Pressure ulcer rates worsened over the full year, from 2.1 to 3.3 percent at p = 0.024.
Orthogeriatric review improved significantly in both analyses, from 80.9 to 85.7 percent at p = 0.001.
The authors’ reading is that the cyberattack disproportionately affected IT-dependent standards, with significant declines in timely surgery and a trend toward deterioration in ward admission. They suggest orthogeriatric review may have improved because geriatrician availability increased when elective activity was disrupted.
How this compares to what came before
The literature on healthcare cyberattacks is dominated by case reports and surveys. Individual hospitals describe their experience of an outage; reviews aggregate those descriptions. A systematic review in this same area catalogued 316 patient-level harms across 59 studies and could not perform quantitative synthesis because the outcomes were structurally incompatible.
That is the state of the evidence: rich in anecdote, almost empty of rates.
A national audit with 99 percent coverage and seven pre-defined standards sidesteps the problem entirely. The measurement system existed before the attack, continued through it, and was not designed for this question – which is what makes it credible.
The finding that IT-dependent standards suffered most is intuitive and had not previously been demonstrated with national data.
What it means
The most useful thing here is the method. Any country with a national clinical audit can do this after an incident, and most have not.
The specific numbers also give cyber risk a clinical unit. A 5.8 percentage point drop in timely hip fracture surgery across a national health service translates into a countable number of patients operated on later than the standard, in a population where delay matters.
The pressure ulcer finding extends the window. That deterioration ran across the full year, well after the acute IT outage, suggesting effects that persist beyond the restoration of systems.
The orthogeriatric improvement is the most interesting result and the least reliable. An explanation exists – elective activity stopped, geriatricians were available – and this design cannot test it. It is a useful reminder that a disruption reallocates resources as well as destroying capacity, and the net effect is not uniformly negative.
Business ideas from this paper
Clinical impact assessment after cyber incidents. Hospitals quantify downtime and recovery cost. Quantifying clinical standard deviation using existing audit data is a different and more consequential assessment, and the methodology is established here.
Audit continuity planning. Data entry shifted from prospective to retrospective after the attack, which degrades the evidence permanently. Designing audit systems that survive an IT outage is a specific, solvable problem.
Cyber insurance underwriting on clinical metrics. Policies are currently priced on IT exposure. Pricing them on measurable clinical standard deterioration would align incentives with what actually matters.
Downtime procedures for time-critical pathways. Surgery within 48 hours is the standard that broke. Paper-based contingency procedures specifically for time-critical surgical pathways are cheap and most hospitals’ downtime plans are generic.
What could be wrong with it
This is an observational before-and-after study of a single national event in 2021. The pandemic, staffing changes and everything else that happened in Irish hospitals that year are confounded with the cyberattack and cannot be separated.
Data were entered prospectively before the attack and retrospectively afterwards. Retrospective entry is a known source of bias, and the direction is that recorded performance tends to be worse when reconstructed from notes.
The orthogeriatric improvement illustrates the design’s limits precisely: the authors offer a plausible explanation that their study cannot test.
One country, one injury type, one incident. Hip fracture care is unusually well instrumented and other pathways may have suffered more or less.
The ward admission result did not reach significance at p = 0.095 and should be read as a trend.
How sure can you be?
We score this 9 out of 10.
It is a peer-reviewed paper in The Surgeon, using national audit data with 99 percent coverage from the National Office of Clinical Audit – an independent body – across all Irish hospitals, with pre-defined standards measured by a system that existed before the event, and named institutions at Tallaght University Hospital and the National Office of Clinical Audit.
It loses a point because the audit data are not openly deposited.
The independence of the measurement is what earns the score. The audit was not set up to study cyberattacks, which means nobody chose the standards to suit the conclusion.
Do this today
If you run a hospital, find out which of your clinical standards depend on IT and which would survive a paper-based day. Surgery within 48 hours was the one that broke in Ireland.
If you manage a clinical audit, write down what happens to data entry during an IT outage. The shift from prospective to retrospective in this study permanently weakened the evidence about the event.
If you buy or sell cyber insurance for healthcare, this paper is the first national-scale quantification of clinical harm from an attack. It belongs in the underwriting conversation.
Source: The effect of a national healthcare cyberattack on the Irish hip fracture standards, The Surgeon, May 2026. PMID 42091386 · doi.org ().
Just Out Tech explains new research in plain language. This article was drafted with AI assistance and checked by a human against the original source.
- IT-dependent standards were hit hardest: surgery within 48 hours fell 5.8 percentage points, p less than 0.001.
- Pressure ulcer rates worsened over the full year, from 2.1 to 3.3 percent.
- Orthogeriatric review improved from 80.9 to 85.7 percent - possibly because elective activity stopped and geriatricians were free.
- Data were entered prospectively before the attack and retrospectively afterwards, which is a known bias in exactly the wrong direction.
Data came from the Irish Hip Fracture Database covering all Irish hospitals via the National Office of Clinical Audit, with 99 percent coverage for 2021.
Questions people ask
What happened in May 2021?
On 14 May 2021 the Health Service Executive of Ireland suffered a cyberattack resulting in complete loss of IT infrastructure - radiology, blood tests, patient management systems and much else.
Why hip fractures?
Because Ireland runs a national hip fracture audit with near-complete coverage and seven defined standards, several of which are time-critical. It is one of the few places where the clinical cost of an IT outage can be measured precisely.
Why did orthogeriatric review improve?
The authors suggest increased geriatrician availability because elective activity was disrupted. That is a plausible explanation the study design cannot test.
Can this be attributed to the attack?
Not definitively. This is a before-and-after study of a single national event in 2021, so the pandemic and everything else happening that year are confounded with it.