Revised 20 Sept 2021. Data series updated 20 Oct 2021
Previous posts at this blog have pointed out the importance of hospitalisation outcomes in understanding what is happening and what to do about the serious coronavirus outbreak in New South Wales.
One of the messages GMO Pundit is advocating is that hospitalisation statistics may provide an early signal of continuing improvement in managing the distressing outbreak usually discussed in terms of increasing numbers of diagnosed infections. These high case numbers are genuinely worrying, not least because they accompanied by some seriously adverse health effects for many people and their families and pose huge challenges for medical staff in hospitals.
Extremely informative weekly surveillance reports are available from the New South Wales public health agencies. Another just issuing (Week 39 ending 2 October 2021) provides an opportunity to update this post as the epidemic and available statistics continues to evolve.
From these reports is possible to obtain weekly statistics for the chances of staying away from hospital and the intensive care unit
The numbers for weekly new diagnosed cases and weekly admissions to hospital are large and relatively precise as measures of success in managing the outbreak and mitigating harm.
What hasn't been much commented about to the Pundit's knowledge, is how these hospitalisation stats are a predictable trend that can be projected forward to suggest future improvements are likely.
This is shown in the graph below produced by GMO Pundit from New South Wales government reports
The graphic provides a simple summary of likely improvement in health protection outcomes from the heroic public health effort in New South Wales. (These include vaccinating 3,829,992 people (47 percent of those eligible) by 21 August.)
With this extension of the series of weekly statistics data the trend seen in the earlier reports is still seen but is weakened and is more uncertain. There is an overall trend to fewer severe outcomes of infection over time. The trend in weekly ICU admissions is more clear cut. (The initial weekly reports weeks 23-27 give fewer details of hospitalisation so the ratio used is bases on totals for week 23 through to 27.)
The Pundit suggests that this trend is driven by the rapid and targeted rollout of vaccination protection in New South Wales and that, hopefully, we can expect further improvement over the next month.
Lag adjusted hospital disease severity ratios
The NSW government weekly COVID-19 reports state that in this outbreak ,hospitalisation lags onset of illness by 5 days. This means the ratios given in the early figure are likely underestimate the chances of severe illness outcomes with infected .
The correct this bias, hospitalisation ratios were also corrected for lags using the prior week's new case numbers to calculate ratios. This leads to the following graph when week 39 (ending 2 October) results became available:
This graph suggests that chances of being hospitalised if infected by the Delta virus variant were worryingly high at 20 to 30 percent at the start of the outbreak.
Those chances severe illness were more moderate over the later weeks of the outbreak at about 12 percent, and chances of ICU entry were also lower than at the start of the outbreak.
Overall there is less frequent progression of infection to severe illness as the epidemic progresses. This indicates that rising vaccine protection is starting to soften disease severity.
Fortunately by 20 September is had become clear that we have passed a peak in daily diagnosis of infection numbers for NSW. Vaccination also halted transmission of disease.
Two hoorays for vaccines. Less disease transmission and less harm to infected people.
In a 30 September post the Pundit explains the answers he found in Portugal to the questions posed by the current post:
10 Oct 2021
Then there is this on Twitter
Although hospitalisation levels are a complex metric depending on hospital stay, the tweet graphic suggests that Victoria may have lower rates of hospitalisation than New South Wales did.
Afternote added 30 Sep 2021.
Although not using the same measures that were plotted above, the following graphs confirm hospitalisation rates have improved in 2021 compared to 2020:
Afternote 9 Sept 2021.
Improvements in disease outcome are seen in other countries in the late stages of the epidemic. Eg here are calculation for England based on government records.
And there is other encouraging news:
And this (10 Sep 2021)
No comments:
Post a Comment