Western Europe has suffered a recent upsurge in coronavirus infections after having suppressed the initial spike. The US, far less successful in quelling the first wave, continues to record hundreds of covid deaths daily. Still, extrapolating from international seroprevalence surveys, these countries’ estimated cumulative infection rates remain under ten percent — far below the seventy percent widely regarded as the threshold for achieving herd immunity. Are there countries whose infection rates place them closer to the magic number?
Granted, data on deaths and their causes are unreliable in many parts of the world. Still, death rates are more reliable than diagnostic testpositive case counts, which vary widely based on availability and selection criteria. As a gross approximation, here’s an algorithm for estimating a country’s cumulative covid infection rate based on death rate:
 Retrieve the nation’s cumulative covid death count per million population.
 Retrieve the nation’s median age.
 Estimate that nation’s covid fatality rate. For the US I’ve been using 0.85%, but that percentage must be ageadjusted because older people are more likely to die of covid infection. The median age of the US population is 38 years. Here’s the quickanddirty age adjuster I’ve been using that fairly closely approximates international seroprevalence results:

 For countries with a median age above 38, multiply 0.85% by 1.1 to the power of years in excess of 38. E.g., if a country’s median age is 42, the estimated fatality rate is 0.85 x 1.1^{4}, or 1.24%.
 If a country’s median age is below 38, divide 0.85% by 1.1 to the power of years under 38. E.g., if a country’s median age is 30, the estimated fatality rate is 0.85/1.1^{8}, or 0.40%.

 Multiply death count per million by estimated fatality rate, then multiply that product by 100,000 to arrive at the percentage of the country’s population infected by covid. (Deaths lag infections by about 3 weeks, so the algo estimates cumulative infections as of early September.)
As of 26 September, the countries with estimated covid infection rates of at least ten percent are:

 Bolivia = 30 percent
 Peru = 30 percent
 Ecuador = 19 percent
 Mexico = 18 percent
 Panama = 15 percent
 Iraq = 14 percent
 Brazil = 13 percent
 Colombia = 13 percent
 Chile = 11 percent
 Honduras = 11 percent
 Guatemala = 10 percent
Latin America clearly dominates the list. Not one of these countries is even halfway to herd immunity.
A seroprevalence survey from Brazil was published a week ago in Lancet. At least 200 immunity tests were conducted June 47 for each of 83 cities; the calculated seroprevalence was 3.1 percent. How does that compare with my algorithm? Brazil’s covid death count on June 27 — 21 days after the midpoint of data collection — was 57,103. Brazil’s population is 209.5 million, so the death count per million was 272. Based on the algo, Brazil’s prevalence on June 6 would have been 5 percent. That’s within shouting distance of the survey’s 3 percent. Any number of explanations could be proposed to account for the difference, including refusal to participate by more than half of the sampled subjects. “These difficulties were compounded by the rapid spread of disinformation through social media characterising the interviewers as swindlers, or of even being part of a plot to spread the virus. Interviewers were arrested in 27 cities and in eight cities tests were destroyed by local police forces.” It’s also possible that the algorithm is overestimating prevalence for populations with median age lower than that of the US.
LikeLike
As of 14 October, continuity prevails. Argentina has now edged out Guatemala for the tenth spot on the list; the others remain the same. Each of the countries on the list added around one percent to its infection rate over the past 18 days.
LikeLike