Case Fatality Rate of COVID-19: Meta-Analysis Approach

<jats:p>
          </jats:p>

indicator/SH.STA.DIAB.ZS?view=chart) was extracted from World Bank data. Analyses were done using Stata software version 13 (Stata Corp, College Station, TX).
In addition to subgroup analysis, meta-regression was conducted to explain the heterogeneity. The meta-regression results showed that there is an inverse association between the number of COVID-19 tests per million and CFR (P = 0.001); also, being in the WPRO region is related with a lower CFR (P = 0.004). In addition to the mentioned results, Spearman correlation showed a significant negative correlation between the number of COVID-19 tests per million and CFR (Spearman's rho = -0.43, P = 0.001).
The overall CFR was estimated to be 3.3 in the present study. Compared to the initial reports from China, the estimated CFR in our analysis is nearly 1.5 times the value reported by Wu Z and McGoogan JM 3 which was 2.3% (1023 deaths among 44 672 cases) and also nearly 2.5 times the value reported by Wu et al. 9 who reported the overall symptomatic CFR as 1.4% (0.9-2.1%) in their study. However, in recent studies from China, the overall CFR was estimated at 3.06% (95% CI 2.02-4.59) 10 which is almost similar to the results of our study.
Despite the controversy surrounding COVID-19 CFR, it is clear that CFR of COVID-19 is noticeably lower than that CFRs of SARS (9.5%) and MERS (34.4%) but higher than seasonal influenza (0.1%). 11 The results suggested that the higher the number of COVID-19 tests per million, the lower the estimated CFR. The number of COVID-19 tests per million performed in each country can be probably considered as an appropriate proxy for the level of health care services such as the numbers of critical care beds, leading to a lower CFR. However, this can be explained by the fact that countries with more tests tend to find more mild patients with less fatality.
Comparison of different WHO regions shows that the highest CFR pertains to African countries, possibly due to the fact that preventive services, followed by health care and facilities in these countries, are weaker, leading to higher disease fatality. The surprising thing is that after African countries, European countries have the highest CFR. These countries, many of which are tourist destinations (such as Italy, Spain, Germany, etc), have more passengers than epicenter countries, and this is likely to lead to the rapid growth of the disease and the overcrowding of patients in hospitals, finally leading to higher CFR.
Previous studies have shown that CFR is higher among the elderly than young people, 3,10 and it is higher among people with a past medical history. 7 However, the results of our analysis show that although the mortality rate is higher in older countries (3.8%), the CFR is not higher (2.1%) in countries with a higher prevalence of diabetes (10% or more).
In summary, the CFR is 3.3, being the highest in AFRO and the lowest in WPRO. Also, it is smaller in countries with more COVID-19 tests, and greater in older countries. It is important to note that given that a large proportion of patients are mild/asymptomatic and unidentified, it is more likely that the estimated CFR is an overestimation.