Table 1

Characteristics of those with follow-up IgG antibody testing with regression results for seroconversion to negative

		Survey with serology results (N=955)
		Seropositive (N=779)		Seronegative (N=176)		Bivariate‡‡ PR (95% CI), p value	Multivariable‡‡ PR (95% CI), p value
Characteristic	Category	n	%	n	%	Bivariate‡‡ PR (95% CI), p value	Multivariable‡‡ PR (95% CI), p value
Age (years)	20–29	101	73.7	36	26.3	1.00 (REFERENCE)	1.00 (REFERENCE)
	30–39	175	82.6	37	17.5	0.66 (0.43 to 0.99), p=0.045	0.69 (0.45 to 1.04), p=0.08
	40–49	159	82.4	34	17.6	0.68 (0.45 to 1.04), p=0.07	0.74 (0.49 to 1.11), p=0.14
	50–59	209	82.3	45	17.7	0.65 (0.44 to 0.97), p=0.03	0.77 (0.52 to 1.14), p=0.18
	60–69	124	85.0	22	15.1	0.57 (0.36 to 0.92), p=0.02*	0.60 (0.37 to 0.97), p=0.04*
	70+	11	84.6	2	15.4	0.57 (0.36 to 0.92), p=0.02*	0.60 (0.37 to 0.97), p=0.04*
Gender†	Female	620	82.1	135	17.9	1.00 (REFERENCE)	1.00 (REFERENCE)
	Male	154	79.4	40	20.6	1.15 (0.84 to 1.59), p=0.39	1.18 (0.85 to 1.63), p=0.32
Race/ethnicity†	White	413	81.0	97	19.0	1.00 (REFERENCE)	1.00 (REFERENCE)
	Black	91	81.3	21	18.8	0.98 (0.64 to 1.50), p=0.92	1.06 (0.69 to 1.63), p=0.78
	Hispanic	125	83.9	24	16.1	0.84 (0.56 to 1.27), p=0.40	0.86 (0.57 to 1.29), p=0.47
	Asian/Pacific Islander	123	80.4	30	19.6	0.95 (0.65 to 1.39), p=0.80*	0.97 (0.65 to 1.44), p=0.87*
	American Indian	3	100.0	0	0
	Two or more races	5	83.3	1	16.7
	Non-specified/missing	19	86.4	3	13.6
Work contract	Full-time day	521	82.8	108	17.2	1.00 (REFERENCE)*
	Full-time other	110	76.4	34	23.6	1.00 (REFERENCE)*
	Part-time day	95	76.6	29	23.4	1.06 (0.74 to 1.51), p=0.75*
	Part-time other	35	94.6	2	5.4	1.06 (0.74 to 1.51), p=0.75*
	Missing	18	85.7	3	14.3
Work location	ICU	35	71.4	14	28.6	1.00 (REFERENCE)	1.00 (REFERENCE)
	ED	72	82.76	15	17.2	0.62 (0.33 to 1.17), p=0.14	0.61 (0.33 to 1.13), p=0.12
	Hospital units (non-ICU)	209	80.4	51	19.6	0.69 (0.41 to 1.15), p=0.15	0.7 (0.43 to 1.14), p=0.15
	Other	456	82.9	94	17.1	0.60 (0.37 to 0.97), p=0.04	0.58 (0.36 to 0.93), p=0.02
	Missing	7	77.8	2	22.2
Work county‡	Nassau	176	80.0	44	20.0	1.00 (REFERENCE)
	Manhattan	84	78.5	23	21.5	1.09 (0.69 to 1.70), p=0.72
	Queens	93	82.3	20	17.7	0.88 (0.55 to 1.43), p=0.62
	Richmond	88	82.2	19	17.8	0.86 (0.52 to 1.41), p=0.54
	Suffolk	85	83.3	17	16.7	0.85 (0.50 to 1.46), p=0.56
	Westchester	26	74.3	9	25.7	1.31 (0.70 to 2.44), p=0.39
	Other/unknown	227	83.8	44	16.2	0.78 (0.52 to 1.14), p=0.21
Residential county‡	Nassau	219	79.6	56	20.4	1.00 (REFERENCE)
	Bronx	21	80.8	5	19.2	0.98 (0.43 to 2.22), p=0.95
	Kings	36	81.8	8	18.2	0.92 (0.47 to 1.81), p=0.81
	New York	26	74.3	9	25.7	1.23 (0.64 to 2.36), p=0.53
	Queens	141	82.9	29	17.1	0.87 (0.58 to 1.31), p=0.51
	Richmond	86	86.9	13	13.1	0.68 (0.39 to 1.19), p=0.18
	Suffolk	141	83.9	27	16.1	0.84 (0.55 to 1.28), p=0.42
	Westchester	50	79.4	13	20.6	1.06 (0.62 to 1.83), p=0.82
	Other/unknown	59	78.7	16	21.3	1.24 (0.73 to 2.11), p=0.42
	New York City	310	82.9	64	17.1	1.00 (REFERENCE)	1.00 (REFERENCE)
	Long Island	360	81.3	83	18.7	1.08 (0.80 to 1.46), p=0.60	1.25 (0.92 to 1.7), p=0.15
	Other	109	79.0	29	21.0	1.32 (0.88 to 1.99), p=0.18	1.5 (1 to 2.24), p=0.05
PCR	Positive	391	89.9	44	10.1	1.00 (REFERENCE)	1.00 (REFERENCE)
	Negative	61	75.3	20	24.7	2.45 (1.51 to 3.96), p=0.001	2.22 (1.38 to 3.59), p=0.001
	No PCR test	327	74.7	111	25.3	2.50 (1.80 to 3.47), p<0.0001	2.56 (1.83 to 3.58), p<0.0001
	Equivocal	0	0	1	100
Re-exposure§	Low	463	83.1	94	16.9	1.00 (REFERENCE)
	Medium	203	79.0	54	21.0	1.24 (0.91 to 1.68), p=0.17
	High	113	80.1	28	19.9	1.17 (0.80 to 1.73), p=0.42
Reinfected/ill¶	No	728	81.3	167.0	18.7	1.00 (REFERENCE)
	Yes	51	85.0	9.0	15.0	0.82 (0.44 to 1.52), p=0.52
Recent travel**	No	718	81.4	164	18.6	1.00 (REFERENCE)
	Yes	60	83.3	12	16.7	0.94 (0.55 to 1.60), p=0.81
	Missing	1	100	0	0
Public transport††	No	681	81.6	154	18.4	1.00 (REFERENCE)
	Yes	94	81.7	21	18.3	0.99 (0.65 to 1.52), p=0.98
	Missing	4	80.0	1	20.0
Days between antibody testing		Mean=140 SD=10 Minimum=105 Maximum=174		Mean=139 SD=12 Minimum=62 Maximum=158		0.99 (0.98 to 1.00), p=0.21	0.99 (0.98 to 1.01), p=0.37

*In regression models, the following categories were combined: Age=60+ years of age; race/ethnicity (Asian/Pacific Islander, American Indian, two or more races); shift=full time versus part-time.
†Administrative data collected from self-reports in prespecified fixed categories.
‡New York City=Bronx, Kings, New York, Queens, and Richmond; Long Island=Nassau and Suffolk; other=Westchester and other/unknown.
§‘Do you believe you may have been exposed again to COVID-19?’ (1–3=low; 4–6=medium; 7–9=high suspicion of re-exposure).
¶‘If so, did you feel ill?’ (Yes/No).
**‘In the past month, have you traveled outside of the NY metropolitan region and northeast?’ (Yes/No).
††‘Do you usually take public transportation to work?’ (Yes/No).
‡‡Clopper-Pearson 95% CIs. SAS GLIMMIX procedure modelling the probability of being seronegative (includes key sociodemographic and work characteristic variables), using a Newton-Raphson optimisation and type III test for fixed effects, missing values were excluded from the analyses; bolded prevalence ratios (PRs) indicate statistically significant associations at p<0.05. A PR greater than 1 is compared with the reference group; similarly, a PR less than 1.00 indicates that the group was less likely to seroconvert to negative.
ED, emergency department; ICU, intensive care unit; n, subsample count; N, sample count.