Helen McArdle doesn’t read the instructions and imagines a spike in stillbirths which isn’t there

Whenever churnos talk of spikes, like the one that wasn’t in the Glasgow cancer wards for children, I reach for my varifocals and the source material.

Here it is:

https://scotland.shinyapps.io/phs-covid-wider-impact/

The black line goes up and down between 6.5 and 2 stillbirths per 1 000 total births. It has recently climbed above 5 but was even higher in 2017 when there was no pandemic.

Nevertheless the Herald’s Health Correspondent thinks:

Aha! A spike. This will make a good headline.’

But just below the graph:

As stillbirths and infant deaths are relatively rare events in Scotland mortality rates tend to fluctuate over time just by chance.

She either didn’t read or ignored this important point and, of course, the red dotted lines indicating whether

values are unexpectedly low or high and require further investigation.

So it’s not a ‘spike’ just fluctuation within the normal range. No story here.

4 thoughts on “Helen McArdle doesn’t read the instructions and imagines a spike in stillbirths which isn’t there”

  1. The source material you linked, John, also says this about the control charts (you already know).There is also a caveat to the effect that not all stillbirths may have been included in the relevant data as there is (or was) still a little time left for registration of those stillbirth deaths. It is not expected that the number of additional deaths reported will be high. So when McArdle wrote her piece she did not mention that neither the “warning level” nor the “control level” were breached.

    As you know the trend for stillbirth deaths is downwards . In 2018/19, NRS recorded 182 stillbirths. There were 51,182 live births recorded in the same period. As in Scotland the trends for both stillbirths and live births in E&W is downwards.

    “Control charts have been used to support interpretation of these data. As numbers of deaths are relatively low, mortality rates tend to fluctuate from month to month just by chance: control charts help differentiate between expected random variation and changes which warrant further investigation.
    In this first release of information on stillbirths and infant deaths (1 July 2020), data are shown for January 2017 to May 2020, with the most recent three months (March-May 2020) being those when health and health services may have been affected by COVID-19.
    In this period the only observations which have reached a ‘warning limit’ as indicated by the relevant control chart were neonatal deaths in March 2020, where the rate was just above the upper warning limit (3.7/1,000 compared to the UWL of 3.6/1,000), but did not breach the upper control limit (the trigger for further investigation). In April and May there were fewer neonatal deaths, and the rate fell to below the upper warning limit. Rates of stillbirths and extended perinatal deaths are being closely monitored, as these approached, but did not breach, the upper warning limit in May 2020.”

    Liked by 2 people

  2. Oh for F***k
    Lets cut to the chase here and not pontificate over him
    He is a pathetic drowning man who can only clutch at straws as they float past him
    Soon his lungs shall be full of the poison
    Of the waters he chose to swim in

    Like

  3. Ah but where’s the headline in “normal fluctuations?” Or indeed the opportunity to show SNP incompetence? Correct, there isn’t one but if we can make those who don’t buy our rag think there’s one, from a carefully crafted headline, then the seed is sewn, the sub conscious alerted and another nail driven in to the body of that pesky Scottish government.

    The truth? Hell if the plebs can’t be bothered to research it properly who am I to tell them…..

    Liked by 2 people

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.