All health organisations collect data, but failing to analyse it can put lives at risk

Nurse Lucy Letby was reportedly arrested last week after data pertaining to infant fatalities in Cheshire was studied, writes Justin Thomas
CHESTER, ENGLAND - JULY 04:  (Editors Note: Number plate of unrelated vehicle has been pixelated) A police officer stands outside a house in Chester after a healthcare professional working at the Countess of Chester Hospital was arrested on suspicion of murdering eight babies and attempting to kill six others, on July 4, 2018 in Chester, United Kingdom. A female health care worker at the Countess of Chester Hospital has been arrested on suspicion of murdering eight babies. Cheshire Police having been have been investigating the deaths of 17 newborns at the neonatal unit between March 2015 and July 2016.  (Photo by Christopher Furlong/Getty Images)

Last week in the English county of Cheshire, nurse Lucy Letby was reportedly arrested on suspicion of murder. The alleged victims in the case were eight newborn babies. This arrest is part of a broader probe into the deaths of 17 infants at the neonatal unit at the Countess of Chester Hospital between March 2015 and July 2016.

The case came to light when concerned parties began exploring the data concerning unexplained and unexpected infant fatalities.

One of the organisations involved, Mothers and Babies: Reducing Risk through Audits and Confidential Enquiries (MBRRACE-UK), crunches numbers, transforming data into intelligence to help healthcare organisations improve by highlighting how they compare nationally with similarly sized and focused organisations.

MBRRACE-UK reports showed that between 2014 and 2015 the rate of neonatal deaths at the Countess of Chester unit more than doubled, increasing from 1.32 deaths for every 1,000 births, to 2.96 babies dying for every 1,000 born.

Comparing this data to other similarly sized units, the Countess of Chester neonatal unit was an outlier. It was clear that something was wrong: neonatal mortality had gone from unremarkable to among the highest in the land.

This is undeniably a tragic case. However, if there is one positive to take away from this unfolding nightmare, it is that the spike in deaths was quickly detected – and investigative and remedial actions were rapidly initiated.

This has not always been the case. We only have to look back at the horrific case of Harold Shipman who, in January of 2000, was found guilty of murdering fifteen patients under his care. Shipman is actually suspected of killing perhaps as many as 250 patients between 1975 and 1998.

According to Dr John Chisholm, former chairman of the British Medical Association's general practitioners committee, one of the reasons Shipman went undetected for as long as he did, was related to systemic failures in the monitoring and analysing of death and cremation data.


Read more from Justin Thomas:


Although we live in the information age, lots of data sources routinely go unanalysed or are merely analysed in isolation. When we do start to look at this data, examining it in a joined-up way, we can get clarity about what is going right and what is going wrong within our organisations.

For example, I once worked in the UK health service for a large group of mental health hospitals (under the umbrella of Lancashire Care NHS Foundation Trust). From an analysis of our incident reporting data, we noticed that there was a significant annual spike in serious incidents – mostly suicides – every April, going back as far as the records existed.

This alone was useful to know. However, we couldn’t explain it until we also looked at the database recording staff leave and absence.

It turned out that April was also the time of year when lots of staff took holidays. This mass exodus in April was due to an existing HR policy stipulating that leave not taken before April 30 could not be carried over into the next financial year.

Changing this policy ensured wards were no longer dangerously low on staff during that month, or any other time of the year, contributing to improved patient safety and the reduction of serious incidents.

How many years might this situation have persisted had we not analysed the data we collected? How many lives were saved?

It is not only in healthcare; many other sectors are guilty of collecting data that they fail to analyse. Collecting bits of information can give us a false sense of security. But if we don’t interpret the data we collect, we can never hope to benefit from it.

Some organisations still have a rather talismanic approach to data and technology. "If we collect lots of bits of information using the latest gadgets," the line goes, then "it's all good".

Data can help us, but only if we have the willingness and ability to translate it into knowledge, to transform it into robust and actionable intelligence. Unfortunately, many of our organisations are still driven by the data imperative – unthinkingly collecting data just because it feels like the right thing to do.

Such organisations amass data mountains that are never meaningfully mined for the nuggets of information that can transform organisations and protect us from systemic malevolence and malpractice. And, at least in the health sector, doing so could save lives.

Dr Justin Thomas is professor of psychology at Zayed University

Follow The National's Opinion section on Twitter

Justin Thomas

Justin Thomas

Justin Thomas is a professor of psychology at Zayed University and a columnist for The National