Why has Warrington got more infections per head than London?

[Confused52]

If you look here https://www.centreforcities.org/coronavirus/#wherein

you will see the confirmed cases per 100 000 head of population for towns and cities in England and Wales. On 22 April the figures for Warrington was 237. On the same day the figure for London was 224.

Other places in England worse than Warrington were:

Birmingham, 228, Luton 256, Liverpool 260, Birkenhead 262, Newcastle 266, Sheffield 270, Sunderland 270

and a day in earlier in Wales - Cardiff 407, Newport 327 and Swansea 368.

Milton Keyes our New Town partner is at just 150.

What is Warrington doing wrong?

[asperity]

Statistics and damned lies as they say. It depends whether you're looking at cases in total or per size of population. Warrington has less than 500 cases in total whereas London has over 27000. Tower Hamlets, with only slightly larger population, has 578 cases.

https://coronavirus.data.gov.uk/#local-authorities

[Confused52]

1 hour ago, asperity said:

Statistics and damned lies as they say. It depends whether you're looking at cases in total or per size of population. Warrington has less than 500 cases in total whereas London has over 27000. Tower Hamlets, with only slightly larger population, has 578 cases.

https://coronavirus.data.gov.uk/#local-authorities

Well what I listed was "confirmed cases per 100 000 head of population". So the numbers are comparable.

After lockdown +6 days London and the rest of the country should have been on a similar downward curve because that is what social distancing does, it turns big towns into what appear as small towns as regards social mixing. London had a two week lead on growing cases so it is very odd that we should have more cases pero head of population that they do.

This is the latest positive test data for Warrington it seems to have strange spikes what are they caused by?

[asperity]

And the numbers for Warrington and Tower Hamlets (in London, chosen at random from all the London boroughs) are comparable, so what's the problem?

Actually I'm getting a bit peeved with all the negativity around. The tiny bit of news I see on the BBC while I'm waiting for the weather forecast (which is also presented somewhat negatively by the BBC for some reason) all seems to be people complaining that their life isn't perfect. Hey! Newsflash! Life isn't perfect, just get on with it!

[Sparky71]

It's probably just down to luck or bad luck.

[Observer II]

Who's doings all these "tests" in Warrington ?    Sounds like we should have herd immunity by now !    😷

[Confused52]

3 hours ago, Observer II said:

Who's doings all these "tests" in Warrington ?    Sounds like we should have herd immunity by now !    😷

I give in, the numbers are from the official numbers of Public Health England, the dates are the date of the swab test and the numbers are the quantity of positives where the subjects address lies in the Borough of Warrington. The source of number in the Centre for Cities graphs is the same source, PHE, but a day earlier for England and two days earlier for Wales. 

If you look at the bar chart you will see that it looks like there are two trajectories for the fall in numbers as well as looking as if the measurement starting from 8th April look strangely high (or low) for a naturally decaying infection, the graph is just too jerky. All I want is to be able to know how to interpret data to asses what risks I am prepared to take. I don't see the point in criticising I just want to understand the facts that I am given.

This leads to a doubt about the comparison in the Cities report. Sheffield Council say their rate seems higher than expected because they are doing more testing but there is no hint about what is happening here. Depending on which set of points one believes the infection rate here in Warrington could be either much higher than it was at lockdown or about the same. When considering what to do about life in general and considering risk it is a significant point. 

Herd immunity sarcasm doesn't help me understand anything.  With the benefit of the back of a metaphorical fag packet and some wild assed guessing: if 50% are asymptomatic and 20% get it seriously but only half of those get admitted to hospital and are tested we get the total infections in Warrington are approx. 500x2x5x2= 5000 out of a population of 210000. So about 2.5% of the population have had an infection, so nowhere near herd immunity with its minimum of 60%. So 97.5% of us are still susceptible. Even more are susceptible in Greece although I don't see why it matters.

 

[Observer II]

Lot of "ifs" in that - how many people were actually tested in total out of a population of around 200,000 ?   So presumably a sample from which they make projections and conclusions.    😷

[Davy51]

Apparently, treatment in Warrington Hospital is turning a very positive corner with the use of modified  CPAP sleep apnea machines to keep airways open on patients & is helping patients to fight the virus.

[Confused52]

49 minutes ago, Observer II said:

Lot of "ifs" in that - how many people were actually tested in total out of a population of around 200,000 ?   So presumably a sample from which they make projections and conclusions.    😷

You appear not to be reading properly today, I hope you are OK. It was not a random sample test, it was the number treated in hospital and tested positive. Well that is what the 497 which I approximated to 500 was meant to be. The extrapolation to a number infected was a guess by me using that ratios that are generally bandied about and work out to 10 infections per tested infected individual. The random testing of a sample would have to include non-infected people that is not what these numbers are about. My problem is that I don't believe they are actually a series of numbers from a single method because they are too different day to day.

[asperity]

What this fella said:

https://youtu.be/REWeBzGuzCc

[Evil Sid]

So there are things that we know we know.

There are things that we don't know we know but we know we  know we don't know them.

There are things that we don't know we don't know but we don't know we don't know.

And then there is conspiracists which say that we don't know we don't know but really we do know and the government hushed it up but my mate in the pub told he got it from fred the delivery guy who got it from a reliable source of his cousin who works at a pub in washington and overheard it from two guys discussing the cia in the bar .

[Confused52]

 

 

This is the seven day moving average of new cases in the WBC area up until yesterday. So the question is where are the extra cases being found because they seem to pre-date any easing?

[Gary]

And just to put another fly in the ointment - we had mass testing in Warrington with a drive-thru facility - but none of these tests are included in the Warrington figures and will be just added to the national figures. Also most people who have had it, got it or going to get it, will not even be tested for some time yet. So the information we are getting is a very rough guide.

[Confused52]

Gary,

Looking at the notes for the COBR presentation the data that government gives out includes Pillar 1 and 2 cases so the cause of the flattening could be Care Home staff and families, in which case it should flatten out in a week or so and resume a downward trajectory. To be clear the numbers in the graph are those from Public Health England and will be very different from the hospital figures. To get a sense of what is happening we really need a set of data where the thing being measured is the same every day so the constant changing and in some cases not publishing data regularly is unhelpful for the analysis of risk. 

[Confused52]

New graph same data source and layout but nine days later than the last one on this thread.

 

 

[Confused52]

and the graph for data up to Thursday sample dates using data supplied by PHE on Sat 6/6/20, the numbers are the 7 day rolling average of new positive tests.

Note this is exactly the same data as shown in the deckzero page but including later data available since their capture.

 

 

The significant point is that this data, the data used by deckzero, is incompatible with their conclusion that the Rt is 04. If the Rt was less than 0.5 the number of new cases would be falling by half in the average period to symptoms and a test which is likely to be seven days. The fact is that the curve is trending up which is incompatible with an overall R value of less than 0.5.

There is no way of actually calculating an R value from this data but it being close to 1 is way more likely than it being much lower at 0.4. The message given out by WBC is dangerous, we need to continue being vigilant.

The often cited Cambridge data only gives regional values for a reason, you cannot calculate R for an area as small as Warrington because the numbers and data are inadequate. The Cambridge data is from a model of mobility and infection rates from PHE. It is good that Gary was more cautious than the WG in his wording but WBC need talking to.

[Confused52]

The peak of cases in Warrington was therefore higher than London and the count per day was above that of London from the 6th of April to today. No-one has given an explanation.

[Confused52]

….. and the new data... 

The counts shown are historical seven day rolling averages of new-cases (positive-tests) per day, this is more useful than actual daily counts and is what the government use too.

There are three sets of geographical area data, (in Red) Warrington, (in Green) North-West region and (in Blue) England. The counts for the North West and England are adjusted to be the number of cases that would have been found on average in a population equal to the size of Warrington. That makes the height of the curves at any given date comparable one with an other.

The solid colour lines are the new set of numbers issued at the end of last week by Public Health England and include cases in Pillar 1 (Patients and staff in Hospitals and Care Homes tested at PHE labs) and Pillar 2 all other tests done for the NHS by none PHE-run labs (the new super-labs).

The dotted lines of the same colour are the old figures released before the end of last week with only Pillar 1 cases included. Until the arguments about Leicester the gap between the actual cases and the published figures were hidden from the view of the public.

In assessing risk it is also worth remembering that the latest ONS infection survey of showed that four out of five people who tested positive were not showing symptoms on the day they were tested.

The source of all of this data is Public Health England and the ONS, all of it is public data.

 

[Gary]

Thanks for all the time and effort you have put into this and I wish I could get straight answers and info off Public Health England. There are so many inconsistencies in the daily reporting figures I am actually starting to wonder why I bother. On at least seven occasions in the past month figures for confirmed new cases have decreased from the previous day with no given explanation.

The authorities keep telling me Warrington's infection rate is lower than other areas, yet your graphs suggest otherwise.

I am also soon to be provided information showing Warrington has the lowest COVID-19 death rate in the North West. 

[asperity]

While that sounds like good news, let us not forget that this isn't a competition. I'm sure that if everyone keeps sensible we'll all be back to normal soon. Went for a drink with my mate who I haven't seen since the pubs shut, wonderful! The pub was organised perfectly although whether they can prosper with low levels of custom for long remains to be seen. Fingers crossed.

[Confused52]

On 7/7/2020 at 8:02 AM, Gary said:

I am also soon to be provided information showing Warrington has the lowest COVID-19 death rate in the North West. 

Gary,

I am sorry to say that it may be a while until you get that evidence. Using the Death Registration data from ONS up to 26th June Warrington had 108.1 deaths per 100k population with Covid19 on the death certificate in all locations (Hospital, Care Home, Home and the ubiquitous Other). The highest rate in the North West was 142.7 in South Lakeland. The best rate was 42.7 in Ribble Valley. I always check these claims about how good Warrington is. it is worrying that they suggested they could provide it. Places with lower rates include some of our close neighbours so if they still make the claim ask for the basis on which they make it!

I have not provided the data because as Asp says, it is not a competition. If you want the numbers in order to seek truth I will happily provide them direct.

[Confused52]

Some new numbers released by PHE gives the number of cases above 2 for each week in Middle Layer Statistical Areas.

I have added those areas of Warrington's counts up and found how many cases more than that were in Warrington over the same period, I also counted up the equivalent cases for the whole of England in the same period. I assigned the cases (2 or 1 a week) pro-rata to the over 3 cases (an assumption). Using that data I calculated the rate of total cases in the time period for each area per 100k population, the I worked out the ratio of that case rate to the equivalent rate for England. So over the period that there have been cases in Warrington Fairfield and Howley MSOA (not the same as the council ward) had 264% of the English average number of cases.

The 25 Warrington MSOA names are listed here by the cumulative rate, relative to the English average. It is difficult to make any guesses at causation but if the Council trot out the usual Index of Deprivation guff it would  currently look like a patent lie. I could change my mind about that with more data which they might have but will probably not make public.

Fairfield & Howley 264% Kingswood & Whittle Hall 234% Winwick & Burtonwood 222% Penketh & Cuerdley 189% Callands & Old Hall 188% Hood Manor & Sankey Bridges 181% Bewsey & Dallam 181% Central Warrington 174% Culcheth West & Croft 171% Great Sankey 166% Orford 163% Latchford 160% Woolston 128% Lymm 127% Hulme 124% Blackbrook 120% Stockton Heath & Hillcliffe 93% Grappenhall 89% Locking Stumps 82% Culcheth East & Hollins Green 73% Padgate & Paddington 59% Houghton Green & Fearnhead 58% Appleton Thorn, Stretton & Walton 44% Appleton 32% Oakwood & Gorse Covert 15%
If anyone wanted I could put in on a map with roads as background to make the actual areas clear, however usually this thread is a simple monologue.

[Observer II]

Judging by the news up-dates,  rather than "poverty",  their seems to be a link between rates of infection and the IQ of a population, as displayed by their compliance or otherwise with scientific advice.     😷

[Evil Sid]

The significant part of that statement is "scientific advice".

The problem being that that advice changes depending on whose scientist you listen to.

The two metre thing for instance. saw some bod on tv talking about it saying that we should stick to the two metre distance where possible. where not possible then one metre would be acceptable.

Peoples interpretation "One metre is fine"

wear a mask in shops and other enclosed public places. "unless you have breathing or other difficulties" peoples interpretation "we don't need to wear a mask".

I have problems wearing a mask. nothing to do with breathing just seeing. I wear glasses, but only to see with, wearing a mask means that when i breath out the glasses steam up and i am looking through a thick fog. a bit of a problem when trying to see the sell by date on a product or trying to navigate a supermarket aisle without walking into the shelves.

people will always come up with some excuse of why something does not apply to them.

Reading a story the other night that mentioned how to determine the IQ of an angry mob. Take the IQ of the least intelligent mob member  an divide by the number of people in the mob and you have the mob IQ. Probably applies to certain population areas as well....🤫