Change in country-level COVID-19 lethality is associated with improved testing: no apparent role of medical care or disease-specific knowledge.

AIM: Reductions in the case fatality rate of COVID-19 in the unvaccinated have been credited to improvements in medical care. Here I test whether either of these factors predicts reductions in the case fatality rate, and whether observed reductions are better explicable by improved ascertainment of mild cases. METHODS: Using weighted log-log regression, I compute the association between changes in the case fatality rate and test density between 3 July 2020 and 5 January 2021 in 162 countries; and check whether case fatality rate change is associated with either per capita medical spending (proxy for critical care access) or timing of the pandemic (proxy for COVID-specific knowledge). RESULTS: The median test density increased from 175 tests per thousand population to 1200, while the median case fatality rate dropped from 4.1% to 2.0%. While the case fatality rate was higher at both timepoints in Europe/North America than Africa / Asia, its association with test density was similar across countries. For each doubling in test density, the mean case fatality rate decreased by 18% (P<0.0001) with a median (interquartile rate) country-level decline of 20% (5-30) per doubling of test density. The rate of change of the case fatality rate was not associated with either medical care access or COVID-specific knowledge (all P>0.10). CONCLUSIONS: Declines in the case fatality rate were adequately explained by improved testing, with no effect of either medical knowledge or improvements in care. The true lethality of COVID-19 may not have changed much at the population level. Prevention should remain a priority.

Estimating total morbidity burden of COVID-19: relative importance of death and disability.

OBJECTIVE: Calculations of disease burden of COVID-19, used to allocate scarce resources, have historically considered only mortality. However, survivors often develop postinfectious 'long-COVID' similar to chronic fatigue syndrome; physical sequelae such as heart damage, or both. This paper quantifies relative contributions of acute case fatality, delayed case fatality, and disability to total morbidity per COVID-19 case. STUDY DESIGN AND SETTING: Healthy life years lost per COVID-19 case were computed as the sum of (incidence*disability weight*duration) for death and long-COVID by sex and 10-year age category in three plausible scenarios. RESULTS: In all models, acute mortality was only a small share of total morbidity. For lifelong moderate symptoms, healthy years lost per COVID-19 case ranged from 0.92 (male in his 30s) to 5.71 (girl under 10) and were 3.5 and 3.6 for the oldest females and males. At higher symptom severities, young people and females bore larger shares of morbidity; if survivors' later mortality increased, morbidity increased most in young people of both sexes. CONCLUSIONS: Under most conditions most COVID-19 morbidity was in survivors. Future research should investigate incidence, risk factors, and clinical course of long-COVID to elucidate total disease burden, and decisionmakers should allocate scarce resources to minimize total morbidity.

Role of Ascertainment Bias in Determining Case Fatality Rate of COVID-19.

Case fatality rate (CFR) is used to calculate mortality burden of COVID-19 under different scenarios, thus informing risk-benefit balance of interventions both pharmaceutical and nonpharmaceutical. However, observed CFR is driven by testing: as more low-risk cases are identified, observed CFR will decline. This report quantifies test bias by modeling observed CFR as log-log-linear function of test density (tests per population) in 163 countries. CFR declined almost 20% (e.g. from 5% to 4%) for each doubling of test density (p < 0.0001); this association did not vary by continent (interaction p > 0.10) although at any given test density CFR was higher in Europe or North America than in Asia or Africa. This effect of test density on observed CFR is adequate to hide all but the largest true differences in case survivorship. Published estimates of CFR should specify test density, and comparisons should correct for it such as by applying the provided model.