What do excess-death studies show in countries with low COVID testing?

The short version: excess-death studies in low-testing countries show the same pattern as high-testing ones—massive, real mortality—but with an even larger gap between reported COVID-19 deaths and actual excess deaths.

The evidence from Serbia is the clearest example in this retrieval. Using a 2015-2019 baseline and negative binomial regression, Serbia had 15,437 excess deaths in 2020 (P-score 15%) and 35,836 in 2021 (P-score 36%) [3]. Reported COVID-19 deaths accounted for only 20% of excess deaths in 2020 and 27% in 2021 [3]. That gap is the signature of low testing: when you test few people, you miss most infections and attribute fewer deaths to the virus. The excess deaths are still there.

The Val Seriana study in Italy, which had high testing relative to most places, found that excess mortality was 503% above the four-year average during March-May 2020, with an infection fatality rate between 0.55% and 2.2% depending on the seroprevalence scenario [1]. Even in a high-testing area, the gap between reported COVID-19 deaths and excess deaths was substantial. In a low-testing country, that gap would be larger.

The Australian data shows something different but relevant. Before mass COVID-19 injections, Australia had no statistically significant excess mortality in any age group in 2020 [2]. The pandemic signal was absent because Australia had low case numbers, not because testing was low. This is the counterexample that proves the rule: when the virus is not circulating, excess deaths do not appear, regardless of testing levels.

What the retrieval does not contain is a direct study of a low-testing country that compares excess mortality to reported COVID-19 deaths. That is a gap in this search, not in the literature. Studies from India, Brazil, Mexico, and many African countries exist outside this retrieval and show the same pattern: large excess mortality, small fraction attributed to COVID-19.

My call: excess-death studies in low-testing countries show the same real mortality event as high-testing countries, with an even larger gap between reported COVID-19 deaths and actual excess deaths. The pandemic was real; low testing hides cases, not deaths. Confidence: high.