Table 2.
Crude and adjusted seropositivity rates by total antibody, IgG, and IgM ELISA
|
Crude estimates |
Adjusted estimates* |
|||||
|---|---|---|---|---|---|---|
| Seropositivity rate for total antibody, n (% [95% CI]) | Seropositivity rate for IgG, n (% [95% CI]) | Seropositivity rate for IgM, n (% [95% CI]) | Seropositivity rate for total antibody, % (95% CI) | Seropositivity rate for IgG, % (95% CI)† | Seropositivity rate for IgM, % (95% CI) | |
| RT-PCR-negative close contacts (n=880) | 40 (4·5% [3·4–6·1]) | 34 (3·9% [2·8–5·4]) | 16 (1·8% [1·1–2·9]) | 4·1% (2·9–5·7) | 1·7% (0·0–5·3) | 0·5% (0·0–1·8) |
| RT-PCR-negative close contacts in the Shenzhen cohort‡ (n=288) | 17 (5·9% [3·7–9·2]) | 16 (5·6% [3·4–8·8]) | 9 (3·1% [1·7–5·8]) | 5·5% (3·2–8·9) | 3·5% (0·0–8·7) | 2·0% (0·4–5·1) |
| Individuals residing in neighbourhood without reported cases (n=350) | 1 (0·3% [0·0–1·6]) | 0 (0·0% [0·0–1·1]) | 0 (0·0% [0·0–1·1]) | 0·0% (0·0–1·1) | 0·0% (0·0–1·0) | 0·0% (0·0–0·0) |
| Individuals residing in neighbourhood with reported cases (n=50) | 0 (0·0% [0·0–1·1]) | 0 (0·0% [0·0–1·1]) | 0 (0·0% [0·0–1·1]) | 0·0% (0·0–6·8) | 0·0% (0·0–7·0) | 0·0% (0·0–6·6) |
Sensitivity and specificity of ELISA used for calculating adjusted estimates of seropositivity rates were 98% and 99% respectively for total antibody, 96% and 98% for IgG, and 90% and 99% for IgM, on the basis of results from GeurtsvanKessel and colleagues12 and Lassaunière and colleagues.11
Sensitivity and specificity of the IgG ELISA were based on an assay validation study done by Shenzhen Centers for Disease Control and Prevention (CDC; unpublished); validation samples were collected more than 14 days after symptom onset in 23 hospitalised patients with COVID-19 and compared with 44 prepandemic controls.
The Shenzhen cohort was defined as individuals who were included in a previous study that characterised the epidemiology and transmission of COVID-19 in Shenzhen, by Bi and colleagues.8