Fig 1.
Schematic diagram illustrating method to compute the COVID-19 Testing Index (CovTI).
The CovTI was computed from a weighted sum of the four sub-indices (orange), each of which is derived from one key indicator (green). The input data for each indicator (white) and intermediate steps (grey) are shown.
Fig 2.
Percentile rankings and sub-index functions for four key indicators.
Percentile rankings, in which 100 represented to most desirable value of the indicator and 0 the least desirable, were computed for each of the four key indicators (a) DR, (b) TPR, (c) CFR, and (d) AC. Functions were fitted to the percentile ranks, which were used to compute the sub-indices (a) DRsi, (b) TPRsi, (c) CFRsi, (d) ACsi (Eqs 9–12).
Table 1.
Definitions of grouping variables for multiple linear regression of COVID-19 Testing Index (n = 147).
Table 2.
Comparison of period prevalence and detection rate estimates with other similar estimates in literature [44, 45].
Fig 3.
Comparison of the medians of Test Positivity Rate (TPR), Case Fatality Rate (CFR), proportion of Active Cases (AC), and Detection Rate (DR) among the quartiles of COVID-19 Testing Index (CovTI).
Bottom quartile included the lowest 25% of CovTI values (n = 42), 2nd quartile included CovTI values between 25th and 50th percentiles (n = 41), 3rd quartile included CovTI values between 50th and 75th percentiles (n = 41), and the top quartile included highest CovTI values above 75th percentile (n = 41). Values in brackets indicate the range of CovTI values in each quartile. Error bars represent interquartile range. Data per 00:00 GMT June 3, 2020.
Table 3.
COVID-19 Testing Index (CovTI) and sub-indices a) among top 15 countries and territories assessed (n = 165).
Table 4.
Comparison of differences in COVID-19 Testing Index between countries and states with different testing and tracing policies, geographical settings, forms of government and economic development status (n = 147).
Table 5.
Multiple linear regression analysis F(7, 139) = 7.07 (p<0.0001), adjusted R2 = 0.23 of factors associated with COVID-19 Testing Index from final model (n = 147) (data from 00:00 GMT June 3, 2020).