Finally got round to listening to this week's @BBCMoreOrLess. Some helpful explanation of govt stats on SARS CoV2 infection rates and false positives. Am reminded how slow I am to get my head around the statistics/maths of this. 1/5 bbc.co.uk/sounds/play/p0…
The fraction, percentage or rate of any test result is obviously dependent on which group of people you test. So random testing may yield a low rate of positive results but if you target your testing at people who think they may be positive, the rate goes up quite fast. 2/5
Sounds really obvious but it isn't. Several leading politicians and journalists are struggling with this fiddly bit of science and even the wonderful @d_spiegel got caught out trying to explain false positive rates to @JuliaHB1. 3/5
Over the years, I have peer reviewed thousands of scientific documents for journals, PhD students, grant funders etc. You learn when NOT to jump to conclusions about stats that aren't clear, and instead to ask for more explanation. 4/5
But we can't expect everyone to have a sixth sense about confusing statistics. That means careful use of verifiable data by govt with simple explanations from objective media sources. Fact check journalism exemplified by @TimHarford and team will give us all more confidence. 5/5
• • •
Missing some Tweet in this thread? You can try to
force a refresh
COVID Sitrep: Only a short update as on call this weekend with this amazing team covering covid ICU. TLDR: Things may be settling but the work is far from over. 1/8
Lots of talk this week about new covid infections hitting a plateau. This is probably driven by children going back to school. So not ideal, but better than the rebound peak we feared would happen after New Year. 2/8
New hospital admissions continue to fall but slowly. A large proportion of these are those 'incidental' covid patients admitted for other 'usual' medical problems not covid disease. 3/8
COVID sitrep: Some positive news to report this week. We are definitely feeling a fall in NHS hospital and ICU admissions with covid in London this week. The rest of the UK should see a similar pattern in the coming days. This below appeared on our ICU board this week! 1/10
As mentioned in previous weeks, being confident we are past the peak is a big deal for the NHS. Every day things get a little easier, a little simpler. We can do more ‘usual’ NHS work. We can get back to clearing the backlog of surgery and other patient care. 2/10
The colours on the national map are getting lighter every week which reassures us that the number of new infections continues to fall in most areas of the UK. 3/10
COVID sitrep: Lots of topics to choose from this week. Good news is we have probably passed the peak in new COVID-19 infections in several UK regions. This week’s map (right) is not so dark. But what will happen next is not so clear. A thread from the NHS perspective. 1/12
All eyes have been on London as the first UK region to get hit by the omicron variant. Cases seem to have peaked both in younger and older people (light blue line=<60yrs, dark blue line=60+ yrs). But we aren’t sure whether new case rates will fall quickly or plateau out. 2/12
Why so cautious when others seem confident cases are falling? Well, passing the peak defines the NHS response. When we know we are past the worst we stop planning extra covid capacity, re-focus on the patients we have, and plan more routine NHS work. A big shift in strategy. 3/12
First COVID sitrep of the New Year: The number of omicron variant COVID-19 infections remains extremely high in the UK and still far higher than anything we have seen up until last month. Here’s what’s happening in the NHS… 1/14
London was the first UK region to get hit by the omicron variant before spreading across the country. This week’s million dollar question is whether new infections have peaked in London. Many say yes, but it seems too early to be sure…. 2/14
…and being sure we are past the peak is a key moment in the NHS response to each wave. Once we *know* we are past the worst, we can stop planning extra covid capacity, we can re-focus on the patients we have, and all the other routine NHS work. This makes a big difference. 3/14
COVID sitrep for New Year’s Eve: The wave of omicron variant COVID-19 infections continues to spread around the world. Hospital admissions are rising across the UK but there are important differences to earlier waves…. 1/12
This omicron variant seems to spread much more easily. Infection rates are smashing all previous records. 189,000 ‘cases’ yesterday. Yes, we are testing a lot more people, but the proportion who test positive is also going up (blue line in graph). 2/12
The big discussion now is how harmful the omicron variant might be. Lots of reports of high rates of ‘incidental’ infections in hospital - patients testing positive for coronavirus who were admitted to hospital for a different illness. So what’s happening on the ground? 3/12
COVID sitrep for Christmas Eve: The huge wave of COVID-19 infections, driven by the new omicron variant, is spreading across the country. Hospital admissions are now rising in London but the good news is we think things won’t be as bad as last winter. 1/12 bbc.co.uk/news/uk-597587…
The wave of infections with SARS CoV-2 is simply huge, passing 100,000 ‘cases’ two days in a row and way in excess of the previous record peak of 81,000 in January 2021. Yesterday 119,000 people tested positive 2/12
Before anyone suggests it, this is NOT simply because we are doing more tests. Firstly, the proportion of tests done which are positive has increased too (blue line in graph). Secondly, people get tested for a reason - symptoms or a covid contact – it’s not random behaviour. 3/12