Private Eye

Tour Dates




Staying Alive




Press Info

Interview Feature

Press Quotes

Tour Reviews



Log in

November 14, 2012

Medicine Balls 1326
Filed under: Private Eye — Dr. Phil @ 1:05 pm

Show Me the Data


On October 17th, Tim Kelsey, the National Director for Patients and Information at the NHS Commissioning Board and founder of Dr Foster, said he ‘should be sacked’ if the NHS doesn’t undergo ‘a data revolution’ under his leadership. Both Kelsey and David Cameron are fond of citing the publication of outcome data for adult heart surgeons in England as proof of a more transparent, accountable NHS. Alas, as the Telegraph spotted, the scheme has stalled due to a lack of funding.


The publication of comparative clinical outcomes was one of the key recommendations of the Bristol heart inquiry and in 2004, heart surgeon Sir Bruce Keogh – now clinical director of the NHS – managed to persuade his 240 colleagues to publish the results of adult heart surgery. Dramatic improvements in survival rates followed. As Kelsey puts it: ‘In some procedures, more than a third of patients are living when they might previously have died and adult heart surgery in England is measurably, demonstrably and statistically better than anywhere else in Europe.’ Or at least it was until they stopped publishing the data.


Mortality ratios don’t give the full picture of how a surgeon, unit or hospital is performing, but if they’re high they warrant proper investigation and questions from patients and relatives. This methodology not only helped spot the Bristol heart scandal but guided the unit’s eventual turnaround. When an external investigation finally took place at Bristol, changes were implemented that saw the mortality ratio drop from 29% to 3.5% within three years. Mid Staffordshire hospital had a significantly high mortality ratio from 1998. The Francis report (due in January 2013) might finally tell us, 15 years later, why mortality ratios and other statistical alerts were ignored for so long. Julie Bailey, founder of Cure the NHS, has published a book – ‘From Ward to Whitehall’ – that shows how patients and relatives were ignored too. The truth of the statistics is often revealed by a visit the ward.

Professor Brian Jarman’s analysis of child heart surgery mortality (Eye last) is not perfect, but it still identified Oxford as a significant outlier. The official Central Cardiac Audit Database (CCAD) did not, and it was left to a whistleblower to get surgery suspended there after four deaths in 2010 (Eyes passim). The Dr Foster Unit (DFU) and CCAD both want to find a fair way of comparing child heart surgery units in England, so it would make sense for them to share data and expertise. But when the DFU applied to have access to data and coding from CCAD, the request was declined because it didn’t ‘demonstrate value to patients.’


CCAD argues that risk adjusted mortality for child heart surgery isn’t yet perfect enough to publish comparative outcomes, but they’ve done it in New York since 1997. In the UK and Ireland, paediatric intensive care has used risk adjusted mortality predictions for over ten years, constantly modified to keep it up to date, and valued and trusted by all the units (


Child heart surgery teams are under huge pressure in their understaffed, over-stretched units awaiting the outcome of Jeremy Hunt’s review of the Safe and Sustainable reorganization. Bristol Children’s Hospital has now been put on an official warning by the Care Quality Commission after a series of deaths were linked to understaffing on the cardiac ward. The sooner resources and expertise are pooled in fewer centres, the better. MD’s guess is that when the reorganization finally happens, official mortality ratios will start to be published.

The NHS needs not just to measure clinical outcomes in real time, but act swiftly on them if they give cause for concern. Breast surgeon Ian Paterson is alleged to have performed over a thousand inappropriate cancer procedures since 1994. Who was monitoring his outcomes? The National Joint Registry has data on the comparative results of individual orthopaedic surgeons, and the success rates vary widely. Why can’t patients see them? And who has a clue how their GP compares? Good data costs money, but not nearly as much as the avoidable harm of secretive, substandard care. Kelsey knows this, but he’ll have a job persuading much of the NHS.