Prevalence of SARS-Cov-2 antibodies and living conditions: the French national random population-based EPICOV cohort

Table 1 Prevalence of antibodies against SARS-CoV-2¹ in people living in France² at the end of the first lockdown according to cumulative hospitalisation and death rates cumulated until the first week of May at départment level: the national EpiCov cohort, round 1—May 2020

	Total	ELISA-S+ ELISA-S ≥ 1.1			SN+ Neutralisation assay ≥ 40			ELISA-S+ or SN+			ELISA-S+/i ELISA-S ≥ 0.7
	N	N	%³	95% CI³	N	%³	95% CI³	N	%³	95% CI³	N	%³	95% CI³
Mainland France	12,114	785	4.5	[3.9–5.0]	656	4.1	[3.6–4.7]	892	5.5	[4.8–6.1]	1132	7.1	[6.4–7.8]
Quintile of hospitalisation rate
1st quintile (lowest rate)	1017	30	2.7	[1.5–3.9]	24	1.9	[1.1–2.7]	38	3.3	[2.0–4.6]	61	5.7	[4.0–7.5]
2nd quintile	1228	43	2.9	[1.9–3.8]	37	2.4	[1.5–3.2]	60	3.9	[2.8–5.0]	77	5.1	[3.8–6.3]
3rd quintile	1170	52	3.6	[2.5–4.7]	50	3.6	[2.5–4.8]	62	4.4	[3.1–5.6]	72	5.2	[3.9–6.6]
4st quintile	3378	148	4.1	[2.9–5.3]	115	4.7	[3.2–6.2]	170	5.6	[4.1–7.2]	245	7.2	[5.5–8.9]
5st quintile (highest rate)	5321	512	9.2	[7.4–10.9]	430	8.0	[6.3–9.7]	562	10.0	[8.2–11.7]	677	12.4	[10.5–14.3]
Sextile of death rate
1st sextile (lowest rate)	734	19	2.3	[1.1–3.4]	16	1.6	[0.8–2.5]	27	3.1	[1.8–4.4]	40	4.8	[3.1–6.5]
2nd sextile	1156	26	2.7	[1.6–3.8]	31	2.3	[1.4–3.1]	47	3.6	[2.4–4.8]	67	5.4	[3.8–6.9]
3rd sextile	892	38	3.6	[2.3–4.9]	35	3.2	[2.0–4.4]	49	4.4	[3.0–5.8]	62	5.9	[4.2–7.5]
4st sextile	2393	99	3.4	[2.5–4.4]	71	3.8	[2.5–5.1]	113	4.7	[3.3–6.1]	165	5.8	[4.3–7.3]
5st sextile	1780	91	5.3	[3.5–7.1]	84	5.7	[3.7–7.6]	106	6.4	[4.4–8.5]	134	7.7	[5.5–9.9]
6st sextile (highest rate)	5159	502	9.5	[7.6–11.3]	419	8.1	[6.3–9.9]	550	10.3	[8.4–12.2]	664	12.9	[10.9–15.0]

Bold is used to underline % and OR
¹Home sampling for finger prick/Euroimmun ELISA-S and seroneutralization tests
²People aged 15 or over, residing in mainland France, but not in care homes for the elderly or prisons
³The sampling design is taken into account for the estimation of prevalence, confidence intervals, with the SAS procsurvey procedure. The percentages are weighted by sampling weight (the inverse of inclusion probability), corrected for non-response probability and calibrated on the margin of the census. The prevalences are not equal to n/N

ISSN: 1471-2334