Version 2.093

I have been very lazy about reporting on my analysis of Covid-19 deaths in Portugal, but it grows old being dismissed by people as if I were crazy. Maybe I will write a more in-depth post over the weekend. By my accounts there are at least five thousand deaths in excess, since March 9, which is when I deemed the data started to deviate from average. 

However, this analysis is very fraught with uncertainty because the death certificate data can be revised for up to two years. For example, when I published my first analysis back in April, on that exact day, 2020 deaths were revised up by about 600 all in one swoop. It made sense for it to be so, since the variance of the data had gotten rather strange at some points, so I had my doubts about it making sense. 

There is another lemon to juice, though: a lot more people have been dying since 2016. More people also died during the Troika years, but the math and explanations made sense. For example, in 2013, there was a heat wave that supposedly was responsible for killing 1.700 people in a few days in July and, during the winter of 2015, about 5.500 deaths above expected were attributed to a severe flu season. If you take out those deaths, the totals seem reasonable historically.

My point that things don't make sense anymore is illustrated by the following: in 2015, with that severe flu season that killed 5.500 people above expected, less than 109 thousand people in total died; but, in 2018, with a mild flu season that killed about 3.700 people above expected, the total annual deaths were 113.598. 

Let's subtract the effect of the flu for each year to get a back of the envelope expectation of death annually. Does this mean that our expected annual deaths went from 103,5 thousand to almost 110 thousand in the span of three years? It does not make sense to me. Even if you were to do the analysis during the September-August timeframe, rather than the regular calendar year, you still come up with that enormous disparity.

Ever since the Socialist Party took office, the number of death certificates issued annually has been at least 110 thousand every year, which had never happened before. My guess -- and it's a just a guess, because I am clearly not versed in the twisted logic that is common in Portugal -- is that when you cut funding from healthcare in a country that is already under stress, people die.