Trust

Covid Passports Reduce Trust in the Health Authorities

It’s often claimed that imposing Covid passports, as many EU countries have done, reduces trust in the health authorities. After all, trying to force someone to doing something isn’t the best way to win their trust. Persuasion usually works better.

However, despite the intuitive plausibility of the idea, good evidence has been hard to come by. A new Danish study changes that. (The study was recently published online, and it hasn’t yet been peer reviewed.)

To examine the effect of Covid passports on trust in the health authorities, Frederik Jørgensen and colleagues analysed data from a repeated daily survey of the Danish population, which has been going since May of 2020.

Crucially, the survey continued throughout November of 2021, during which the Danish government re-introduced Covid passports. Note: the Danish “corona passport” serves as evidence of vaccination or previous infection, so it does at least recognise natural immunity.

On 8th November, the Government held a press conference announcing this decision. During the press conference, Denmark’s Prime Minister explained that life would become “more burdensome” for the unvaccinated, whom he referred to as “a small group that does not play according to the rules”.

The researchers tracked two outcome measures over the weeks before and after the 8th November press conference. The first was simply whether one agreed with the statement, “I trust the political strategy behind the health authorities’ advice.”

The second was based on six statements, including the one above. The other five statements included, “I have been given clear information on the reasons for the health authorities’ advice,” and “The advice of the health authorities are sufficient to prevent the spread of infection.” This composite measure was labelled “collective action motivation”.

The authors main finding is shown in the chart below. It’s important to note that the press conference was held on the first day of week 45.

The interpretation is slightly complicated, so bear with me. Each circle (and associated confidence interval) corresponds to the difference between vaccinated and unvaccinated people relative to the benchmark week, which is week 41.

For example, the black circle for week 44 (which is almost exactly on the red zero line) indicates that the difference in trust between vaccinated and unvaccinated people in week 44 was the same as it was in week 41. Units along the y-axis are percentage points.

The key thing to notice is that the circles for weeks 45 and 46 are all below the red zero line. This indicates that the difference in trust between vaccinated and unvaccinated people became significantly larger after the 8th November press conference, with unvaccinated people becoming even less trusting than they were before.   

Jørgensen and colleagues’ finding constitutes particularly strong evidence that Covid passports are bad for trust sinces it combines a between-group comparison (vaccinated versus unvaccinated) with a before-after comparison (pre versus post press conference). The size of the effect is between 7 and 13 percentage points, depending on the measure.

Denmark is known to be one of the world’s most trusting countries. And in those characterised by less social trust, the researchers note, “adverse consequences of pressuring unvaccinated may be even more negative”.

NHS Admits 25% of Patients With Covid Were Admitted For Something Else

We’re publishing a guest post today from our in-house doctor, a senior clinician in private practice who used to work for the NHS. This one’s a bit of a shocker.

On Thursdays, the NHS releases the weekly summary data in relation to Covid patients. Normally this is a more granular version of the daily summaries – it has some hospital level detail and figures on non-Covid workload for comparison. Usually interesting but not especially informative.

Yesterday was an exception. Placed down at the bottom of the page, almost like a footnote, was a ‘Primary Diagnosis’ Supplement. Graph 1 shows the information contained in that spreadsheet. I find it astonishing. In essence it shows that since June 18th, the NHS has known its daily figures in relation to ‘Covid inpatients’ were unreliable at best and deliberately untrue at worst.

The yellow bars are what the NHS has been telling the nation were Covid inpatients. The blue bars are the numbers of inpatients actually suffering from Covid symptoms – the difference between the two are patients in hospital who tested positive for COVID-19 but were being treated for something different – where Covid was effectively an incidental finding but not clinically relevant.

For example, on July 27th, the total number of beds occupied by Covid patients was reported as 5,021. However, until today, we were not allowed to know that only 3,855 of those were actually admitted with Covid as the primary diagnosis. There has been a fairly consistent over estimate of the true number by about 25% running back to mid-June – figures before that date are “not available”.