Friday Report – Issue 28             Covid-arg.com

By Matt Fletcher, Nicola Oliver and John Roberts

COVID-19 Actuaries Response Group – Learn. Share. Educate. Influence.

COVID-19 is still one of the hottest topics for scientific papers and articles. The COVID‑19 Actuaries Response Group will provide you with a regular Friday update with a curated list of the key papers and articles that we’ve looked at recently.

Clinical and Medical News

Effect of Hydroxychloroquine

Hydroxychloroquine is an antimalarial drug that has been tested extensively for the treatment of COVID-19. So far, there has been little evidence to suggest that it is effective in either treating or preventing the disease. Results from a recently concluded multicentre, blinded, randomized clinical trial, are presented here.

Patients hospitalised for less than 48 hours with COVID-19 displaying symptoms of respiratory illness were randomised to receive hydroxychloroquine or placebo in a 1:1 ratio. the primary outcome was clinical status 14 days after randomisation; 28 secondary outcomes were also identified, which included 28-day mortality.

Analysis of the results finds no significant difference between the groups. These findings do not support the use of hydroxychloroquine for treatment of COVID-19 among hospitalized adults.

Characteristics of Hospitalized COVID-19 Patients Discharged and Experiencing Same-Hospital Readmission

As evidence continues to emerge regarding the long-term health implication of those who have had COVID-19, the question of potential hospital readmission of COVID patients is addressed in this next report by the CDC.  Using data from the Premier Healthcare Database (PHD), which contains 25% of US hospital discharges, the CDC sought to understand the frequency of, and risk factors for, readmission. A hospital readmission was defined as subsequent hospitalisation within 2 months of the initial hospital admission, (‘index hospitalisation’).

Around 10,000 patients were readmitted, (9%), which included ~1700 patients who were readmitted more than once. The median interval was 8 days between admissions (interquartile range = 3–20 days). Risk factors for readmission included:

• Presence of pre-existing conditions
• Discharge to a care home
• Age ≥65 years
• Hospitalisation prior to the index hospitalisation
• Non-BAME groups

Results from Phase 3 Vaccine Trial

The first phase 3 trial of a candidate vaccine against SARS-CoV-2 to report results is that of the mRNA-based vaccine candidate, BNT162b2. This vaccine was developed by Pfizer and BioNTech, and the press release states that the vaccine was found to be, ‘more than 90% effective in preventing COVID-19 in participants without evidence of prior SARS-CoV-2 infection in the first interim efficacy analysis’.

The manufacturers are planning to make an emergency submission for approval to the FDA in the third week of November. This is certainly cause for optimism, but without detailed analysis, one should remain cautious.

We will be publishing a Bulletin on vaccines in the coming weeks.

Ethnicity and clinical outcomes in COVID-19

The association between ethnicity and COVID is a topic that has been frequently explored in studies since the beginning of the pandemic. This week, The Lancet published a systematic review and meta-analysis which explored the relationship between ethnicity and clinical outcomes in COVID-19.

Data from 50 studies which included 18,728,893 patients were analysed. In particular, the authors sought to identify ethnic differences in the risk of becoming infected with SARS-CoV-2 as well as subsequent intensive therapy unit (ITU) admission (a surrogate marker for severe COVID-19 pneumonia) and death.

This meta-analysis reports the following:

• Individuals from Asian and Black ethnic groups are more likely to be infected by SARS-CoV-2 compared to those of White ethnicity
• Individuals of Asian ethnicity may be at increased risk of death compared to White patients
• Asian individuals had a higher risk of severe infection
• Racism and structural discrimination may also contribute to an increased risk of worse clinical outcomes within ethnic minority communities

CPAP management of COVID-19 respiratory failure

Continuous positive airway pressure, (CPAP), is a non-invasive method for delivering respiratory support to patients who have respiratory failure. One of the benefits of using this approach to respiratory support is that it does not require an intensive care setting to be utilised.

This retrospective case–control study  evaluated the impact on survival in patients with COVID-19 who were admitted to the acute inpatient unit of the Wrightington, Wigan and Leigh Teaching Hospitals NHS Foundation Trust who were treated with CPAP.

CPAP was found to be significantly (HR 0.38, 95% CI 0.36 to 0.40) associated with lower risk of death in patients with hospital stay equal to, or below 7 days. For longer hospitalisation, CPAP was associated with increased risk of death, (HR 1.72, 95% CI 1.40 to 2.12).

CPAP represents an effective, low-cost, ward-based therapy for treating respiratory failure but in the short-term only.

Modelling

Mobility network models of COVID-19 explain inequities and inform reopening (Chang, S., Pierson, E., Koh, P.W. et al, Nature (2020)) (link)

This paper uses anonymised location data from 98 million people in the largest US metropolitan areas early in the pandemic (1 March to 2 May 2020) to capture the effects on changes in mobility on the spread of COVID-19. It finds that a small minority of superspreader locations (for example, restaurants, gyms, and churches) account for a large majority of infections, and that therefore reducing maximum occupancy may be more effective than reducing mobility.

The chart shows (as an example) the predicted increase in infections from reopening different points of interest (POIs) in the Chicago metropolitan area.

The model also correctly predicts the higher infection rates seen amongst disadvantaged socioeconomic groups and specific racial groups simply by using mobility data. The authors note that this is partly because these groups have been less able to reduce mobility as sharply, and partly because they tend to visit more crowded locations.

The paper makes various policy suggestions to reduce the spread of infection overall, focussing in particular on actions that reduce racial and socioeconomic disparity in infection.

Data

REACT Study Round 6 Report (link)

The Imperial College REACT study on infectivity released its final Round 6 report yesterday. Covering the period from 16^th October to 2^nd November (so just prior to the second lockdown in England) it confirms the view from the interim report that infectivity had reached 1.3% by the end of the period, up from 0.6% in the Round 5 report. This would translate to around 1m infections, and a run rate of around 100k new infections per day. This remains around twice the ONS view – an inconsistency which appears to remain unexplained.

However, and most notably, it doesn’t suggest a value for R during the period. This is because the sharp increase seen in the first half of the period has been followed by a noticeable decline, but then infections appear to have picked up again in the last few days. This has been seen across all regions, despite the differing tierings which applied, so might suggest a national effect such as the end of half term is the cause.

ONS Infectivity Update (link)

Hot on the heels of the REACT study comes the weekly ONS update. This shows overall infectivity in England still increasing, but much more slowly, from 618k to 654k, or 1 in 85 of the population. The daily level of new infections has been in the 50,000 area now for three weeks, broadly consistent with the levelling we see reported by REACT.

Of note this week there are more encouraging signs from Northern Ireland, where infectivity levels appear to be dropping after four weeks of stringent restrictions. The situations are less clear in Wales and Scotland however, not helped by the wide confidence intervals that we see with the devolved administrations due to much lower volumes of data.

The latest SAGE estimate for R has fallen to a range of 1.0 to 1.2 for the UK, and 1.1 to 1.2 for England. Regionally the lowest mid-point estimate is for the North West (1.0), and the highest for the South East and South West (1.3). Our own estimate for England, based on hospital admissions had fallen fairly consistently for a month, but has increased back up to 1.1 in the last few days, in line with a further surge in new admissions after a level period.

Other

Community testing – Lateral Flow Tests

Public Health England (PHE) and the University of Oxford have been carrying out clinical evaluation of Lateral Flow Tests – these are rapid turnaround tests for COVID-19 that can be processed without the need for lab equipment and provide results within a short timescale (typically under half an hour). A preliminary report from their evaluation process has just been published (link).

The assessment process has four separate phases – forty devices were put forward for testing, of which six have made it to the third stage. The test being used in Liverpool’s whole-city pilot is the Innova SARS-CoV-2 Antigen Rapid Qualitative Test and is nearing completion of the testing process.

The Innova evaluation shows that the overall false positive rate was 0.32% – there was some evidence that field testing gave higher false positive rates than laboratory-based testing (0.06% vs 0.39%), with challenges in interpreting “weak” test results. This is higher than the estimated figure published for current PCR tests but still means that the proportion of false positives is likely to be low. It appears that the false negative rate may be of more concern when carried out in the community – self-trained members of the public were able to correctly evaluate only 58% of samples known to be positive, compared to 73% of trained healthcare workers and 79% among laboratory scientists. The tests do appear to detect higher proportions of individuals with high viral loads (over 95%).

Because they are inexpensive and relatively easy to use, it’s quite possible that these tests can be very important in controlling the spread of COVID-19 in future – they will be able to identify large numbers of people who may not have previously been tested and who may be asymptomatic, but who may be likely to spread the disease.

Assessing the AbC-19 Antibody Test

Also, on the subject of Lateral Flow tests, earlier this year the government announced the purchase of 1m such tests from Abingdon Health at a cost of £75m, these being to test for antibodies. The BMJ this week has published a study (link), commissioned by PHE, into the effectiveness of these tests, whilst questioning whether the timing and transparency around the publication of these results was appropriate.

The results of the AbC-19 Rapid Test show that sensitivity of 92.5% was achieved. However, when this was subdivided into those who had received a positive PCR test result and those who had not, the results were 94.2% and 84.7% respectively. The effectiveness for those who did not know they had been infected was thus much lower.

The BMJ notes that for those who received a positive test, the likelihood of this being correct was just 82%, and that in the 1m tests bought, there would be 18,900 false positive results. It notes in its review that these independent results are significantly lower than a previous report, funded by the consortium developing and producing the test which suggested sensitivity of 97.7% and specificity of 100%.

And finally….

This collection of 226 Haikus provides a partial record of the coronavirus pandemic and the Trump presidency from early January 2020 to 1 November 2020 and is an alternative way to view the timeline of events. Beautifully entitled: The Covidiot Project: Haiku for Our Times

13 November 2020