A screenshot of a cell phone

Description automatically generated
Bias by graphing? The left-hand scale needs to go up to at least 400 and maybe 1000 and reveal the spike to be really a meaningless blip or even less.

A ‘spike’ in hospital acquired infections or in already infected admissions? A ‘spike’ or a tiny ‘blip?’ Graphs can be misleading depending on how you scale them. The graph shows the number of blood samples containing bacteria in each three-month period. In a three-month period, thousands will be treated. I can’t find the figures. There are 244 paediatric beds in the unit and around 400 children in care on any one day. If it were to be accurate rather than melodramatic, the graph would need to use daily, rather than quarterly figures, and still need to be extended upward SIXTEEN times higher making the spike seem, as it is, really a tiny blip. If they insist on using quarterly figures in a vain attempt to generate any visible data and pattern of change, then it would need to be extended upward as much as 50 times or perhaps more depending on how many children were treated in a period of three months and you’d need a magnifying glass to see the so-called spike.

The HPS report contains some important caveats and a call for caution. Neither word, nor the warnings they make, appear in the BBC Scotland report:

It is not possible to determine whether changes in episodes are confounded by changes in the patient population and their underlying medical conditions.

The spike/blip may be as much the result of more children coming in already infected as of more becoming infected after being admitted.

In the monthly analysis of environmental bacteria positive blood cultures, the numbers are small and should be treated with caution. (p21)

The increases are statistically insignificant and becoming in 2019 almost nil.

Here’s how they misreported:

A screenshot of a social media post

Description automatically generated
Sickening exploitative reporting

From BBC Scotland, TV broadcast, radio and online, today:

Higher than expected rates of some infections were found in cancer patients at Glasgow’s children’s hospital in 2017, a report has said. Health Protection Scotland said there was an upturn in “gram-negative” bacteria over a 10-month period. This came a year before cancer wards were closed because of a link to contaminated water. The mother of Milly Main said she was “100%” certain contaminated water caused her daughter’s fatal infection. The 10-year-old died in August 2017 after treatment for leukaemia at the Royal Hospital for Children (RHC).

Way down the page BBC Scotland admits:

The Health Protection Scotland analysis said that over the four-year period blood data for the Royal Hospital for Children was “broadly consistent” with other large children’s hospitals in Scotland.

So, there was no story in the first place, really?


Click to access 2_review-of-nhsggc-paediatric-haemato-oncology-data.pdf

What the Health Board pointed out:

  • One occasion when the number of infections linked to environmental organisms was greater than expected for this group of patients (Table 5). The period in question was June 2018 which was already being investigated by the infection control team and was identified as being potentially linked to the water supply.
  • At no other time between 2013 and 2019 did the rate of infections linked to environmental organisms exceed the upper range of expected levels. This includes 2016 and 2017.
  • An increase in Gram negative infections (including both environmental and enteric, i.e. intestinal infections) was noted in 2017 however this remained within expected levels for the unit. During this time there was an investigation into the possibility that two of these cases may have been linked which was later confirmed not to be the case. This investigation was reported to HPS as per mandatory guidance.
  • Since the move to Ward 6A and 4B in September 2018, infection rates have been similar to other Scottish paediatric units.
  • For a particular group of infections, known as gram positive infections, the rates have fallen and are now lower than elsewhere in Scotland.
  • No single source of ‘exposure’ of infections has been identified across the six year period.