The data from Scotland is non-sensical.

[StatisticalSense]

According to the most recent data on the government dashboard (coronavirus.data.gov.uk/healthcare?areaType=nation&areaName=Scotland) the number of people in hospital with Corona in Scotland was 248 yesterday, however the number of admissions has been in single figures for several weeks (to the extent that you have to go back to the end of may to get 248 cumulative admissions). Clearly the average time in hospital with Covid is not several months and therefore either Scotland has a sizable problem with in hospital transmission that is not happening in other parts of the UK (unlikely) or someone is fudging the data somewhere along the lone.

Why is the time in hospital not several months?

At the beginning of June there were 700 odd confirmed cases in hospital, so they’ve got rid of 2/3 of them in that time.
www.gov.scot/binaries/content/documents/govscot/publications/statistics/2020/04/coronavirus-covid-19-trends-in-daily-data/documents/trends-in-number-of-people-in-hospital-with-confirmed-or-suspected-covid-19/trends-in-number-of-people-in-hospital-with-confirmed-or-suspected-covid-19/govscot%3Adocument/COVID-19%2BDaily%2Bdata%2B-%2BTrends%2Bin%2Bdaily%2BCOVID-19%2Bdata%2B-%2B20%2BAugust%2B2020.xlsx

Well if dh's recovery time is anything like those in hospital then I can well believe some of the 248 have beennin for months

@GreyishDays
The time in hospital is not going to be significantly different between England and Scotland unless treatment is badly failing in one country (either through excess deaths in England which the data makes clear aren't happening or through slower recovery in Scotland). Clearly it makes no sense that an admissions figure that is typically less than 10% of that in England can lead to the number of people in hospital only being around 50% lower.

I must admit I’m surprised by this too.

I agree it makes no sense. I wonder if it's a bit like the PHE, where they were counting anyone who had ever had a Covid test as a Covid death? If you were 90 and in hospital and caught Covid you're probably not going home as soon as you've recovered from Covid because there's still whatever else was wrong with you in the first place. But that doesn't mean you're in hospital because of Covid anymore.

Could be counting long stay patients . We had lots of patients stay longer than 6 months - longest was four years in hospital but there were definitely longer . Think rehab wards, psych units, long stay dementia care ... it stands to reason that they would have higher numbers of Covid19 . I’m guessing they aren’t removing from the numbers unless discharged .

I’ve been confused by the Scottish hospital figures for ages too. They are proportionately massively higher than the English figure.

Are they in hospital with something else long term and at some point tested positive ?

Same thing occurred to me today. We are about a tenth of population in scotland vs England and yet have 50%of hospitalised pits of England. Something doesn’t add up. But there is so much manipulation of the stats going on it’s so hard to follow what is actually comparable.

@positivelycatastrophic

Could be counting long stay patients . We had lots of patients stay longer than 6 months - longest was four years in hospital but there were definitely longer . Think rehab wards, psych units, long stay dementia care ... it stands to reason that they would have higher numbers of Covid19 . I’m guessing they aren’t removing from the numbers unless discharged .

Not so sure psych wards added to the numbers. I've not heard much recently but a while back I read there was a lower than average number of cases in psych patients. In fact, a psychiatric drug, an anti psychotic, was included in trials as a potential treatment.

How are England counting their covid patients?

The number we get given as staff in our trust only includes those who have a positive covid test in the last 14 days. I don’t know if that’s the same number that’s being reported nationally.

So you might be admitted, spend a few days on a ward, go to ICU and spend 4 weeks on a ventilator and then get discharged after 7-8 weeks in total but you’d only count as a covid patient for the first 2.

Is it Scotland counting too many as in hospital or England counting too little? How do we know? What do other countries look like for comparison?

People have been hospitalised for months though. The number of people in hospital has been reducing over time, there is nothing nonsensical about it.

@premiumshoes

People have been hospitalised for months though. The number of people in hospital has been reducing over time, there is nothing nonsensical about it.

But why such a high proportion compared to England? Do England stop counting after a while?

But why such a high proportion compared to England? Do England stop counting after a while?

I don't know. Why is the assumption on the thread that Scotland are doing it wrong?

“Clearly the average time in hospital with Covid is not several months”

I might be, I don’t know... but there were a few stories in local papers and stuff with patients being discharged from hospital after about 3 months, so I mean... it could be...

The patients admitted to hospital are hardly likely to be suddenly ok a few days later, they have to be pretty damn ill to be admitted.

Looking at the most recent icu report with data up to 30th July it has the median length of stay in hospital for those who have been in critical care as 20 days. That includes time in hospital before being transferred to critical care, time in critical care and time in hospital after being discharged. The interquartile range looks quite big though. The median stay in critical care is 12 days for patients they have outcome data for. But 25% spend 28 days or more there.

The "people in hospital figure" is a piece of nonsense. These are people who are in hospital who have tested positive, not people who are in hospital BECAUSE OF Covid. Massive difference. I would assume that many of these 200 odd people are very ill and/or very elderly with a whole host of underlying conditions, and who would have been in hospital anyway. It's an irrelevant measure.

You're spot on that the key figure is new admissions to hospital and those are way down.

Travelling tabby is a much better site than the official one for the fugures www.travellingtabby.com/scotland-coronavirus-tracker/ as those are undoubtedly being spun for political purposes.

@premiumshoes

But why such a high proportion compared to England? Do England stop counting after a while?

I don't know. Why is the assumption on the thread that Scotland are doing it wrong?

Is there that assumption? Certainly not by me. Another SNP fan who can’t see any wrong in anything the Scottish Government does?

It's quite possible that there are a number of long-term unwell patients in Scotland, some people are spending many months in hospital. Equally it's reasonable to know if the figures are being calculated on the same basis and they may not be; France has also had a much higher proportion of hospitalisations than England, while we know Wales counts admissions differently (anyone with any respiratory issue is counted as a covid admission, which is clearly daft heading into winter). While things are calm it would be good for the four nations at least to standardise how they track data; if in disagreement both numbers can be reported. Perhaps you should raise your question to both the Scottish government via their website and Westminster via the coronavirus every page.

From the scot gov pages explaining their counting and confirming that it's not people in hospital as a consequence of Covid:

"NHS Boards provide Scottish Government with daily management information on the numbers of patients in acute hospitals beds with confirmed and suspected COVID-19 at midnight. This includes all patients in hospital, including in intensive care, and community and long stay hospitals."

"The number of confirmed cases may include people who are in hospital for other reasons but have previously tested positive for COVID-19."

Previously tested positive - at some point, way back in March....

The average time in hospital might not be several months, but when you are at the tail of an outbreak curve, the ones who have been in months are the only ones left. Comparing them to the infection rate, or current cases, as you are doing, doesn't make sense. England is not at the tail end of a curve, so you would expect a different ratio.

Lots of people in England are still catching it, going into hospital, then out again/dying quickly. There will also be a cohort who are long-termers. As the new infections go down, this cohort will appear bigger compared to the number of cases.

Also, Covid is the name for the symptoms resulting from infection from coronovirus. If someone is still suffering the symptoms, despite not being contagious or testing positive, they are still being treated for Covid. They are still on the Covid ward. We don't know, for sure, all the details around contagion, so the figures you see are not contagious people, they are people suffering from having been infected.

You are making the assumption, statistically, that all hospital cases contribute the same variance to an analysis. Some will contribute only a day or two and some months and months. To understand why these figures look this way, you should look at the curves from survival analyses, not a normally distributed, average.