Data is not a mirror. It does not simply reflect reality back at us — it selects, frames, and in doing so, inevitably excludes. The most powerful use of data is not confirmatory but exploratory: the patient, unglamorous work of tracking real changes in real communities, driven by genuine curiosity about what’s happening rather than what we hope or assume to be true. When data is used to justify decisions already made, it stops being a tool for understanding, becoming something closer to a weapon.
It can beget a kind of institutional confidence that in turn can become its own form of danger, creating certainty in those who believe they are doing the right thing, armed with data that appears to support them, to make decisions that will profoundly affect folks they may never meet. The debate over emergency surgery at South West Acute Hospital (SWAH) in Enniskillen is, on the surface, a local Northern Irish healthcare dispute.
But a closer look reveals a set of genuinely difficult tensions that cut to the heart of how we make life-and-death decisions in modern healthcare — tensions that familiar to anyone who has followed similar battles in Shropshire, Lincolnshire, the Scottish Highlands, or rural Wales.
Who Gets to Decide What Counts as Evidence?
For this post, I have leaned heavily on a statistical report compiled by independent statistical consultant, Paul Bassett which was commissioned by Save Our Acute Services (SOAS). It throws and important light on a issue that has been bubbling away in Fermanagh and wider the border area but which has struggled to get a fair hearing further afield.
The Western Health and Social Care Trust has clinical experts. They have consultants, medical directors, and years of surgical experience. When they look at their Risk Adjusted Mortality Index (RAMI) data and conclude that outcomes have improved, they are not acting in bad faith. They genuinely believe it.
But belief, even expert belief, is not the same as statistical proof. An independent statistician commissioned to examine the same data reached an entirely different conclusion — not because the numbers are different, but because the analytical framework applied to them is more rigorous. This tension between clinical authority and independent statistical scrutiny is not unique to Northern Ireland.
It surfaced prominently in the Mid Staffordshire National Health Service (NHS) scandal, where reassuring mortality statistics masked serious care failings for years. It appeared again in debates over the reconfiguration of stroke services in London and Manchester, where clinicians and statisticians disputed what the outcome data actually demonstrated. The question of who is qualified to interpret evidence — and whose interpretation carries institutional weight — remains one of the most consequential unresolved problems in healthcare governance.
When the Data Is Too Thin to Follow
Modern evidence-based medicine was built on the principle that we should follow the data. But what happens when the data is too thin to follow anywhere with confidence? The RAMI data provider itself recommends approximately 1,000 deaths for reliable comparisons. The Western Trust has around 100 per year. This is not a minor methodological quibble. It means that the entire delineation of “improved outcomes” rests on figures whose confidence intervals are so wide that almost any conclusion could be drawn from them.
This problem is not confined to SWAH. Research published in the British Medical Journal has repeatedly highlighted how small hospital trusts lack the patient volumes needed to generate statistically meaningful quality indicators, yet are routinely ranked and compared using exactly those measures.
The Dr Foster Hospital Guide, which for years published hospital mortality rankings in national newspapers, was criticised by statisticians on precisely these grounds — that apparent differences between institutions frequently reflected statistical noise rather than genuine variation in care quality. The SWAH situation is, in this sense, a local manifestation of a systemic flaw in how healthcare performance is measured and communicated across the entire NHS.
This creates a genuine tension for policymakers everywhere. You cannot wait indefinitely for statistically perfect data before making service decisions — hospitals must be run, budgets must be set, configurations must be decided. But neither can you responsibly present statistically fragile findings as settled evidence of improvement. There is no clean question to where exactly that line sits — between necessary pragmatism and misleading certainty.
What the Data Simply Cannot See
Perhaps the most troubling aspect of this case is what the data simply cannot see. Mark McGuigan was 61 years old (you can hear his story here), from Roslea in Fermanagh. He developed gallstone problems, was sent directly to Altnagelvin under the new pathway, waited three days in an Emergency Department (ED) chair, developed sepsis, then pancreatitis, then necrotising fasciitis, and died on 17th November 2025 — never having reached surgery. His death will not appear in the Trust’s RAMI statistics. RAMI counts inpatient surgical deaths. He died in intensive care in Belfast.
This is not an edge case anomaly. It is a structural blindspot that researchers have long recognised. The phenomenon known as the “streetlight effect” — measuring what is easy to measure rather than what most needs measuring — distorts policy in ways that are rarely acknowledged.
The Care Quality Commission (CQC) in England has similarly acknowledged that existing mortality metrics miss significant categories of patient harm, particularly those arising from delays and care fragmentation. When we choose our outcome measures, we are simultaneously choosing which harms become visible and which remain invisible.
Rural Lives and an Unspoken Bargain
There is an equity dimension to this case that deserves direct naming. The principle that time is critical in emergency medicine is well-established and universally applied — except, it seems, when the patients in question live in rural areas far from centralised services. The Royal College of Emergency Medicine (RCEM) has published evidence of one excess death for every 72 patients waiting 8–12 hours in Emergency Departments.
The National Institute for Health and Care Excellence (NICE) guidelines on emergency surgical care consistently emphasise timely access as a core determinant of outcome. NHS England’s own Getting It Right First Time (GIRFT) programme has acknowledged that transfer times and journey distances represent genuine clinical risks in emergency presentations.
Yet in case after case — from the reconfiguration of services in Cumbria and North Yorkshire to the ongoing debates about district general hospital viability across Wales and Scotland — rural communities are effectively being asked to accept higher personal risk so that centralised services can demonstrate better aggregate statistics.
The Nuffield Trust and the Health Foundation have both published work highlighting how rurality functions as a persistent and largely unaddressed health inequality in United Kingdom healthcare planning. That bargain — your inconvenience and risk in exchange for our improved institutional metrics — is rarely made explicit, and almost never consented to.
What Happens When Institutions Know and Carry On Anyway
Finally, there is the question of what happens when an institution knows its evidence is contested and continues using it anyway. The Public Health Agency (PHA) privately cautioned the Western Trust that its conclusions went beyond what the data could support. An independent statistical review confirmed no significant improvement. Yet the Trust continued — and apparently continues — to make its “lives saved” claims publicly.
This pattern will be recognisable to those who followed the Morecambe Bay maternity scandal (also here and here), where internal concerns were repeatedly downplayed in public communications, or the later stages of the Mid Staffordshire crisis (see Francis report here), where board-level confidence persisted long after warning signs had accumulated. The UK’s Healthcare Safety Investigation Branch (HSIB) was established in April 2017 partly in recognition that NHS organisations have structural incentives to present their performance in the most favourable available light.
When a public body presents statistically questionable findings to justify permanent service changes, and no mechanism exists to effectively challenge or correct this in real time, the democratic legitimacy of the entire decision-making process is undermined. The HSSIB, which succeeded HSIB in 2023, has broader powers — but its remit remains focused on individual incidents rather than the systemic misuse of outcome data.
The SWAH case will eventually be resolved one way or another. But the tensions it surfaces — about expertise, evidence, measurement, equity, and accountability — will not resolve themselves. They will simply reappear, wearing different faces, in the next community asked to accept the loss of services they depend on.
Until the NHS develops genuinely robust mechanisms for independent statistical scrutiny of service change decisions, and until rural health equity is treated as a serious policy priority rather than an afterthought, the people of Fermanagh and West Tyrone will not be the last to find themselves on the wrong side of numbers that don’t tell the whole story.
Data used well is an act of care as much as analysis. It asks not only what can be measured, but what matters — and who is being missed. Until that standard is applied consistently, the people most affected by major decisions will continue to find themselves on the wrong side of statistics that were never really designed to find them in the first place.
“Intelligence is quickness to apprehend as distinct from ability,
which is capacity to act wisely on the thing apprehended.”
— Alfred North Whitehead
Mick is founding editor of Slugger. He has written papers on the impacts of the Internet on politics and the wider media and is a regular guest and speaking events across Ireland, the UK and Europe. Twitter: @MickFealty
Discover more from Slugger O’Toole
Subscribe to get the latest posts sent to your email.