‘Most egregious use of data I have seen in my entire clinical life’ - former WHSCT Consultant
The veracity of Western Health and Social Care Trust (WHSCT) claims regarding outcomes for patients in Derry’s Altnagelvin Hospital requiring emergency general surgery (EGS) is being challenged.
There are also concerns about the level of over-inference or optimism the WHSCT attaches to outcome measures.
The data relates to patient outcomes following the centralisation of EGS provision in Altnagelvin.
Speaking to The Derry News, London-based statistician Paul Bassett said there was “a lot of uncertainty around the Trust’s figures”.
Mr Bassett specialises in the application of statistics in medical research and clinical trials.
A former WHSCT Consultant described the Trust’s claims as the “most egregious use of data I have seen in my entire clinical life”.
While an independent clinical professional said it was a “poorly thought through use of metrics”.
They added:
The outcome (mortality) itself is poor and there is a lack of methodological clarity about how the patients were counted at a basic level. Small changes between admission groups from new pathways of care could have a big effect on these data.There are other measures much more valuable to measuring the quality of emergency surgical care at individual patient level.It is easy to come up with these if you look at the most recent reports about pancreatitis, bowel obstruction and gastrointestinal (GI) bleeding, from the National Confidential Enquiry [Into Patient Outcome and Death (NCEPOD)]. Why isn’t there a bigger focus on taking processive care measures from these reports?It is disappointing that there are highly paid professionals not doing adequate due diligence on using information. Especially when the initial consultation seemed to involve significant missteps - said the clinical professional.
Processive care measures are the specific steps or activities carried out to deliver patient care. These are potentially more important than weak (statistically uncertain) indicators of mortality outcomes.
The WHSCT “temporarily” suspended EGS provision at Enniskillen’s South West Acute Hospital (SWAH) in December 2022.
Altnagelvin is currently providing EGS for local patients, in addition to patients who have had to travel from Fermanagh and Tyrone to Derry because, three and a half years later, the SWAH EGS suspension remains in place.
Jimmy Hamill from the Fermanagh-based Save Our Acute Services (SOAS) campaign said the disparities in the WHSCT’s data first emerged at a meeting of the Assembly Health Committee.
On March 13, 2025, the Committee held a session titled ‘Review of Emergency General Surgery at the South West Acute Hospital: Department of Health; Western Health and Social Care Trust’.
Appearing before the Committee were, Dr Tomas Adell (Head of Elective Care, Department of Health); Mr Mark Gillespie (Director of Surgery, Paediatrics and Women's Health, WHSCT); and Dr Brendan Lavery (Medical Director, WHSCT).
Chairman, Philip McGuigan (Sinn Féin) voiced concern about “double emergency department (ED) waits”.
He said:
It is of major concern that people are waiting in EDs twice [in SWAH and in Altnagelvin], before travel and after travel.How are you getting on with the [RQIA] recommendation on increasing the number of ambulances?
According to the Hansard record of the meeting, Dr Lavery replied that patient outcomes were “actually better”.
He added:
We use a company called CHKS [Caspe Healthcare Knowledge Systems] which analyses the admissions across every trust in England, Wales and Northern Ireland. CHKS looks at every admission - thousands of admissions - and does a statistical analysis of age, sex, comorbidities and the diagnosis that the patient was admitted with using 250 categories of diagnoses. CHKS uses all that data to generate a risk-adjusted mortality index (RAMI).At the time of the temporary suspension, the figure for the South West Acute was 110, and the figure for Altnagelvin was 85.
We got the figures in July or August last year [2024]. Effectively, the RAMI scores for Altnagelvin have continued to fall - to the extent that, if you extrapolate the data to look at mortality rates, you find that, due to the change that we have made, every 40 days, one patient survives who would not have survived.
Later in the session Dr Lavery said:
The Public Health Agency (PHA) has independently reviewed the information. There is no adverse outcome for patients who live in that area.
"SOAS was immediately dubious about the WHSCT’s RAMI claims,” said Mr Hamil.
He added:
We subsequently discovered that, writing to the Trust’s Medical Director on October 3, 2024, Joanne Mc Clean, the Director of Public Health, described the WHSCT’s data as needing ‘some more work.
The Derry News has seen a copy of the confidential email which SOAS obtained under a Freedom of Information (FOI) request to the PHA.
In it, Ms McClean wrote:
While you could say … there is no evidence of an increase in in-hospital mortality following the service change … I don’t think you could go beyond that without some more work.
She added:
The biggest issue is that the absolute number of deaths per month is very small and subject to significant month to month variation … This means that the RAMI figures calculated are likely to have wide confidence intervals.
A ‘confidence interval’ is the statistical range which tells where a true population value is likely to fall.
Mr Hamill continued:
SOAS was also aware that in December 2018 CHKS published its ‘Hospital mortality measures’ guidance which stated: ‘Indicators that count events such as deaths suffer from huge uncertainty.
That document said:
To be confident of a rate (to within 10 percentage points) approximately 1,000 deaths must be included in the dataset.
Many smaller hospital trusts have fewer than this number of deaths in a whole year, and analysis of a smaller subgroup of deaths (a specific condition, for example) would require proportionately more years of data before an acceptable degree of confidence about the underlying rate can be reached.
The former WHSCT Consultant told The Derry News, there was no way the Trust had this amount of deaths.
They added:
When we go to scientific meetings, whenever data is presented, there are always confidence intervals, and when you see confidence intervals as wide as they are in the Trust’s data, you can have no confidence. You cannot draw any conclusions from the data.
Speaking to the Assembly Health Committee, the Trust’s Medical Director just took the numerical change and said this is an improvement but with the confidence intervals, you cannot say that. It could have improved. It could have disimproved. We have no idea.
The PHA told the Trust not to use the data but it went ahead and used it. It is the wrong analysis.
The claim that for every 40 days one life is saved is outrageous because the Medical Director has derived that data from the RAMI. He looked at the reduction in mortality and he extrapolated it to the number of deaths. He said the number of deaths is now is reduced by X every 40 days.
“It is all complete nonsense. The whole methodology is flawed. It is the most egregious use of data I have seen in my entire clinical life.
Mr Hamill said SOAS realised rigorous interrogation of the Trust’s RAMI data was essential. He explained:
We therefore commissioned a ‘Statistical evaluation of data on the change in Emergency Surgery provision in Western Health and Social Care Trust’ (November 19, 2025) - an analysis of the Trust’s RAMI data by Paul Bassett (Statsconsultancy Ltd).
Mr Bassett subsequently told The Derry News:
Because of the nature of the hospital and the fact mortality is quite an unusual occurrence, there is a lot of uncertainty around the [WHSCT] figures.
It is hard to say with any certainty that things have gotten better or worse because the numbers are quite small. If there have been some changes, which they are claiming, there is no statistical justification that those differences are genuine ones and not just a chance variation over time.
There is certainly no evidence to suggest it is what we normally call a statistically significant change, in other words that the difference is unlikely to be due to chance and as a result is likely to be a genuine effect.Although there have been some slight improvements, the uncertainty around the estimates is such that it is highly likely that those are just due to chance.There is no definitive conclusion that those are genuine improvements in the performance as a result of the centralising of ESG provision in Altnagelvin Hospital.
Essentially, the WHSCT’s has been criticised for using analytic approaches that have significant uncertainty but attach strong inference without any degree of statistical skepticism or triangulation with other critical metrics of process.
This has been described as “a very poor example of medical leadership and strategic planning for change of this magnitude”.
On Monday (July 6, 2026) WHSCT issued a press release titled, ‘Western Trust highlights independent evidence of safer Emergency General Surgery pathways following inaccurate public claims’.
It has also established an Emergency General Surgery Information Hub.
The Trust said:
Independent evidence has confirmed continued improvements in Emergency General Surgery patient pathways across the Western Trust, with both the RQIA Inspection Report and the independent CHKS review highlighting better patient outcomes, improved patient flow and enhanced patient safety.The full Trust statement is available.
⏩Catherine McGinty is a journalist covering the North West.
No comments