Seroprevalence of Severe Acute Respiratory Syndrome Coronavirus 2 Among Skilled Nursing Facility Residents and Staff Members—Los Angeles County, August–September 2020

Abstract

Background

The prevalence of current or past coronavirus disease 2019 in skilled nursing facility (SNF) residents is unknown because of asymptomatic infection and constrained testing capacity early in the pandemic. We conducted a seroprevalence survey to determine a more comprehensive prevalence of past coronavirus disease 2019 in Los Angeles County SNF residents and staff members.

Methods

We recruited participants from 24 facilities; participants were requested to submit a nasopharyngeal swab sample for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) polymerase chain reaction (PCR) testing and a serum sample for detection of SARS-CoV-2 antibodies. All participants were cross-referenced with our surveillance database to identify persons with prior positive SARS-CoV-2 results.

Results

From 18 August to 24 September 2020, we enrolled 3305 participants (1340 residents and 1965 staff members). Among 856 residents providing serum samples, 362 (42%) had current or past SARS-CoV-2 infection. Of the 346 serology-positive residents, 199 (58%) did not have a documented prior positive SARS-CoV-2 PCR result. Among 1806 staff members providing serum, 454 (25%) had current or past SARS-CoV-2 infection. Of the 447 serology-positive staff members, 353 (79%) did not have a documented prior positive SARS-CoV-2 PCR result.

Conclusions

Past testing practices and policies missed a substantial number of SARS-CoV-2 infections in SNF residents and staff members.

COVID-19, SARS-CoV-2, Seroprevalence, Skilled nursing facilities

As of 19 January 2021, > 24.2 million cases of coronavirus disease 2019 (COVID-19) have been reported in the United States, including approximately 401 000 deaths [1]. Although >1% of the US population resides in long-term care facilities, 40% of COVID-19–associated deaths have occurred among these residents [2]. In Los Angeles County (LAC), the nation’s most populous county, there are 38 242 licensed beds in 381 nursing homes.

The first COVID-19 outbreak in an LAC nursing home was reported on 18 March 2020. From 18 March to 2 November 2020, the LAC Department of Public Health (DPH) investigated 533 COVID-19 outbreaks in 328 nursing homes and identified 11 137 residents and 7360 staff members with COVID-19. The average number of new nursing home outbreaks reported per week declined from 42 in April to 12 during October. The number of nursing home residents testing positive for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the virus that causes COVID-19, declined from a peak of 822 residents in the week of 26 April to 2 May 2020 to 59 residents in the week of 25–31 October 2020. The size of nursing home outbreaks declined from an average of 32 residents with COVID-19 in outbreaks beginning between 26 April 26 and 2 May to 17 in those beginning in 25–31 October.

The reasons for the decline in the number and size of COVID-19 outbreaks in nursing homes are likely multifactorial. Recognition of the role of asymptomatic transmission led to the implementation of universal masking policies in nursing homes [3–5]. Introduction and transmission of SARS-CoV-2 in nursing homes was further reduced through improved infection control procedures, such as isolating SARS-CoV-2–infected residents in physically separated “red-zones” and placing newly admitted residents in quarantine. Regulations requiring weekly surveillance testing for SARS-CoV-2 in nursing home staff and residents potentially resulted in earlier detection and containment of COVID-19 outbreaks [6]. It is unknown how much of the decline in COVID-19 cases and outbreaks in nursing homes is attributable to immunity among staff and residents who had recovered from infection acquired during prior outbreaks at their facilities.

Early in the pandemic, SARS-CoV-2 testing was available only through the DPH Public Health Laboratory (PHL). Before the recognition of asymptomatic transmission, SARS-CoV-2 testing was offered only for the first person in a nursing home who experienced symptoms; other residents and staff subsequently experiencing respiratory symptoms in the facility were presumed to have COVID-19 and isolated accordingly. Even after commercial SARS-CoV-2 tests became available in late March and April, many nursing homes did not have timely access to testing because community demand exceeded the available testing capacity. Therefore, it is likely that many nursing home residents and staff had undiagnosed COVID-19. We aimed to conduct a seroprevalence survey (SPS) to obtain a more comprehensive estimate of the prevalence of exposure to SARS-CoV-2 among residents and staff in LAC nursing homes.

METHODS

COVID-19 Surveillance

The LAC DPH conducts surveillance for approximately 10 million residents in 86 cities, excluding Long Beach and Pasadena, which have independent health departments. Healthcare providers in LAC are mandated to report to DPH all confirmed COVID-19 cases, and clinical laboratories are mandated to report all SARS-CoV-2 diagnostic test results (positive, negative, indeterminate, and inconclusive). Staff from DPH attempt to interview all persons reported with COVID-19 to identify risk factors, close contacts, and exposure settings with a potential for an outbreak (eg, nursing home). All outbreaks are assigned to DPH field teams, comprising physicians and nurses, for further investigation and management. All case reports and investigation data are entered into the Integrated Reporting Investigation and Surveillance System (IRIS), DPH’s central surveillance database.

Since June 2020, DPH has required skilled nursing facility (SNF) operators to conduct weekly surveillance testing of 25% of directly employed staff and 10% of residents, using a SARS-CoV-2 polymerase chain reaction (PCR) test. If a staff member or resident had COVID-19 diagnosed, the SNF is required to conduct outbreak testing of all staff and residents weekly until no new positive test results are detected in 2 consecutive rounds of testing. Facility staff members and residents with a positive SARS-CoV-2 PCR test result are exempt from regular testing requirements for 90 days after the date of specimen collection. A COVID-19 outbreak in an SNF is defined as the occurrence of a single facility-acquired laboratory-confirmed case of COVID-19 in a resident; an outbreak is considered over if no cases are detected for 14 days.

Study Population

There are 315 freestanding SNFs in LAC, excluding facilities that are a distinct part of a hospital and those in the cities of Long Beach and Pasadena. To conduct this study, DPH collaborated with an SNF operator that manages 24 of the facilities (8%) across 14 cities in LAC. We will refer to all 24 facilities as SNFs, though they provide a mix of skilled and unskilled nursing home services. The participating SNFs had a median licensed bed capacity (interquartile range [IQR]) of 97 (49–300) beds, a median resident census on the date of study enrollment of 67 (43–265), and a median staff census of 98 (71–449) (Table 1). By comparison, the median (IQR) licensed bed capacity among all 315 SNFs in LAC was 99 (62–124) beds. At the end of our study period on 24 September 2020, there were a total of 38 242 licensed bed across all 315 SNFs, and 28 315 beds (74%) were occupied (unpublished data from LAC Health Facilities Inspection Division).

Table 1.

Demographic Characteristics of Skilled Nursing Facility Staff and Residents in Seroprevalence Survey (Los Angeles County, August–September 2020)

Characteristic	Tested by PCR, No.	PCR Positive, No. (%)a	PCR Negative, No. (%)a	P Value	Tested by Serology, No.	Serology Positive, No. (%)a	Serology Negative, No. (%)a	P Value
SNF residents	1188	22	1166	…	856	346	510	…
Sexb
Male	608	12 (2)	596 (98)	.75c	440	187 (43)	253 (58)	.19c
Female	579	10 (2)	569 (98)		415	158 (38)	257 (62)
Age group
18–44 y	20	1 (5)	19 (95)	.36d	18	8 (44)	10 (56)	.89c
45–64 y	309	6 (2)	303 (98)		226	89 (39)	137 (61)
≥65 y	859	15 (2)	844 (98)		612	249 (41)	363 (59)
SNF staff	1833	22	1811	…	1806	447	1359	…
Sexe
Male	533	6 (1)	527 (99)	.95c	523	133 (25)	390 (75)	.66c
Female	1296	15 (1)	1281 (99)		1280	313 (24)	967 (76)
Age group
17–44 y	897	14 (2)	883 (98)	.31d	875	224 (26)	651 (74)	.17c
45–64 y	842	7 (1)	835 (99)		836	207 (25)	629 (75)
≥65 y	94	1 (1)	93 (99)		95	16 (17)	79 (83)

Abbreviations: PCR, polymerase chain reaction; SNF, skilled nursing facility.

^aPercentages may not total 100% because of rounding.

^bStatistical analysis excludes 1 SNF resident with unknown sex.

^cP value calculated using Pearson χ ² test.

^dP value calculated using Fisher exact test.

^eStatistical analysis excludes 4 SNF staff with unknown sex.

Data Collection

From 18 August to 24 September 2020, DPH nursing teams were deployed to obtain from residents and staff at the 24 SNFs nasopharyngeal (NP) swab specimen for SARS-CoV-2 PCR testing and serum for detection of immunoglobulin (Ig) G antibodies against SARS-CoV-2. Demographic information on all potential staff and resident participants was collected in advance for preregistration with DPH PHL and printing of laboratory requisition forms. Testing for SARS-CoV-2 by PCR was offered to all staff and residents who had not tested positive within the previous 90 days. Serologic testing was offered to all staff and residents who submitted a NP swab specimen and to persons who were excluded from PCR testing because they had previously tested positive for SARS-CoV-2. Specimen were packed in cold packs and shipped via courier to PHL for processing. Informed verbal consent was obtained from all participants; persons without the capacity to provide verbal consent were excluded from participation.

Laboratory Methods

For SARS-CoV-2 testing, PHL used the Hologic Panther Fusion SARS-CoV-2, Hologic Panther TMA SARS-COV-2, or Centers for Disease Control and Prevention 2019 novel coronavirus reverse-transcription PCR assays. Serum or plasma samples were used by PHL for serologic testing. All serum samples were tested by using 2 distinct automated chemiluminescent immunoassays as part of an orthogonal algorithm to improve specificity and positive predictive value: the Abbott SARS-CoV-2 assay (conducted on the Abbott i1000SR instrument that tests for IgG antibody against nucleoprotein) and the Diasorin SARS-CoV-2 assay (conducted on the Diasorin Liaison XL instrument that tests for IgG antibody against spike protein) [7]. Participants were categorized as seropositive if either test yielded a positive result.

Data Management

All participants’ demographic information and laboratory test results were stored in PHL’s laboratory information system, Sunquest Information Systems. To obtain the results of any prior SARS-CoV-2 PCR tests conducted for study participants, we cross-referenced study participants’ demographic information with IRIS by using a deterministic matching process based on first name, last name, a composite variable composed of the first 3 letters of first and last name, and a name flip along with date of birth. To obtain the history of prior COVID-19 outbreaks in each participating SNF, we reviewed the IRIS outbreak investigation reports for each facility to determine the dates of the first and last reported cases, cumulative counts of resident and staff cases, and bed capacity.

Statistical Analysis

Characteristics of PCR-positive and PCR-negative groups, and of seropositive and seronegative groups, were compared using Pearson χ ² tests and Fisher exact test for categorical variables. All statistical analyses were conducted using SAS 9.4 software.

Human Subjects Research Concerns

This project was reviewed and approved by the LAC Institutional Review Board.

RESULTS

From 18 August to 24 September 2020, we enrolled 3305 participants from 24 SNFs. All SNFs had a COVID-19 outbreak before the date of the SPS. The median (IQR) duration of the outbreak in SNFs, as measured by the interval between the first reported resident case and the last resident case before the date of the SPS, was 106 (72.5–139.75) days; the median interval between the date of the last case and the date of the SPS was 40 days (15–64.25) (Supplementary Table). Among the 1340 resident participants, 704 (53%) provided both NP swab and serum specimen, 484 (36%) provided NP swab specimen alone, and 152 (11%) provided serum specimen alone. Among the 1965 staff member participants, 1674 (85%) provided both NP swab and serum specimens, 159 (8%) provided NP swab specimen alone, and 132 (7%) provided serum specimen alone.

Of the 1188 residents who submitted an NP swab specimen, 608 (51%) were male, and the median (IQR) age was 72 years (63–83) years (Table 1). Of the 856 residents who submitted serum for serologic testing, 440 (51%) were male and the median (IQR) age was 72 (63–83) years. Of the 1833 staff members who submitted an NP swab specimen, 533 (29%) were male and the median (IQR) age was 45 (32–55) years. Of the 1806 staff members who submitted serum for serologic testing, 523 (29%) were male, and the median (IQR) age was 45 (33–55) years. There were no significant differences by sex or age between participants with or without a positive SARS-CoV-2 PCR result and between participants with or without positive SARS-CoV-2 antibodies.

Among the 856 residents who provided serum, 346 (40%) had detectable SARS-CoV-2 antibodies (Table 2). An additional 2 serology-negative residents (<1%) had a positive SARS-CoV-2 PCR result on the SPS, and 14 serology-negative residents (2%) had a documented prior SARS-CoV-2 PCR result in the DPH surveillance database (data not shown in tables), which yielded a total of 362 residents (42%) with evidence of current or past SARS-CoV-2 infection. Of the 346 residents who were SARS-CoV-2 serology positive, 199 (58%) did not have a documented prior positive SARS-CoV-2 PCR result >2 weeks before the SPS date (Table 3).

Table 2.

Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) Polymerase Chain Reaction and Anti–SARS-CoV-2 Serologic Results Among Skilled Nursing Facility Staff and Residents in Seroprevalence Survey (Los Angeles County, August–September 2020)

SNF	SNF Residents				SNF Staff
	Tested by PCR, No.	PCR Positive, No. (%)	Tested by Serology, No.	Serology Positive, No. (%)a	Tested by PCR, No.	PCR Positive, No. (%)	Tested by Serology, No.	Serology Positive, No. (%)a
A	39	13 (33)	19	6 (32)	59	2 (3)	58	7 (12)
B	70	1 (1)	52	41 (79)	113	2 (2)	122	37 (30)
C	51	0 (0)	39	22 (56)	93	1 (1)	110	47 (43)
D	52	0 (0)	45	3 (7)	54	0 (0)	54	11 (20)
E	36	0 (0)	24	15 (63)	41	1 (2)	33	20 (61)
F	47	0 (0)	38	21 (55)	71	0 (0)	80	27 (34)
G	62	0 (0)	53	4 (8)	118	0 (0)	116	9 (8)
H	64	1 (2)	38	8 (21)	67	1 (1)	68	13 (19)
I	29	2 (7)	26	12 (46)	71	0 (0)	70	18 (26)
J	48	0 (0)	37	9 (24)	75	0 (0)	75	6 (8)
K	40	0 (0)	23	7 (30)	25	0 (0)	28	4 (14)
L	40	0 (0)	29	2 (7)	77	0 (0)	78	8 (10)
M	11	1 (9)	9	5 (56)	40	3 (8)	44	16 (36)
N	35	0 (0)	29	15 (52)	43	1 (2)	44	17 (39)
O	55	0 (0)	37	8 (22)	110	2 (2)	58	12 (21)
P	55	0 (0)	41	28 (68)	82	1 (1)	83	41 (49)
Q	45	0 (0)	44	14 (32)	72	0 (0)	90	23 (26)
R	39	0 (0)	31	11 (35)	58	0 (0)	62	13 (21)
S	174	2 (1)	102	60 (59)	253	5 (2)	227	48 (21)
T	22	0 (0)	23	9 (39)	25	0 (0)	28	8 (29)
U	38	0 (0)	29	7 (24)	63	0 (0)	71	12 (17)
V	27	0 (0)	17	0 (0)	64	1 (2)	61	15 (25)
W	60	2 (3)	36	23 (64)	87	1 (1)	74	20 (27)
X	49	0 (0)	35	16 (46)	72	1 (1)	72	15 (21)
Total	1188	22 (2)	856	346 (40)	1833	22 (1)	1806	447 (25)

Abbreviations: PCR, polymerase chain reaction; SNF, skilled nursing facility.

^aThe Los Angeles County Public Health Laboratory performs an orthogonal algorithm whereby positive results in specimens are confirmed using a secondary test. Both the Abbott and the Diasorin severe acute respiratory syndrome coronavirus 2 assays were used for each serologic specimen. Participants were categorized as seropositive if either test yielded a positive result.

Table 3.

Correlating Prior Severe Acute Respiratory Syndrome Coronavirus 2 Polymerase Chain Reaction Results With Serologic Results in Seroprevalence Survey of Skilled Nursing Facility Residents and Staff (Los Angeles County, August–September 2020)

Serologic Results	Study Participants, No. (%)a
	PCR Positive	PCR Negative	No PCR Record	Total No.
SNF residents
Positiveb	147 (42)	150 (43)	49 (14)	346
Negative	14 (3)	426 (84)	70 (14)	510
Total	161	576	119	856
SNF staff
Positiveb	94 (21)	212 (47)	141 (32)	447
Negative	3 (0.2)	1046 (77)	310 (23)	1359
Total	97	1258	451	1806

Abbreviations: PCR, polymerase chain reaction; SNF, skilled nursing facility.

^aParticipants' prior PCR results were obtained from the Los Angeles County Department of Public Health surveillance database (Integrated Reporting Investigation and Surveillance System); PCR results obtained <2 weeks before the seroprevalence survey were excluded. Percentages may not total 100% because of rounding.

^bThe Los Angeles County Public Health Laboratory performs an orthogonal algorithm whereby positive results in specimens are confirmed using a secondary test. Both the Abbott severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) assay and the Diasorin SARS-CoV-2 assay were used for each serologic specimen. Participants were categorized as seropositive if either test yielded a positive result.

Among 1806 staff members who provided serum, 454 (25%) had evidence of current or past SARS-CoV-2 infection. Among the 454 staff members with current or past infection, 447 (98%) had detectable SARS-CoV-2 antibodies, 4 (1%) had a positive SARS-CoV-2 PCR result on the SPS, and 3 (1%) had a documented prior positive SARS-CoV-2 PCR result in the DPH surveillance database. Of the 447 staff members who were SARS-CoV-2 serology positive, 353 (79%) did not have a documented prior positive SARS-CoV-2 PCR result. The 2650 resident and staff participants who had prior SARS-CoV-2 PCR test results had been tested a median (IQR) of 4 (2–6) times.

Of the 793 participants with a positive serologic test, 241 (30%) had documentation of a prior positive SARS-CoV-2 PCR test result; the first positive SARS-CoV-2 PCR result was a median (IQR) of 104 (87–125) days before the date of the SPS (Figure 1). Of the 1869 participants with a negative serologic result, 17 (1%) had a prior positive SARS-CoV-2 PCR result; the first positive SARS-CoV-2 PCR result was a median (IQR) of 101 (84–118) days before the date of the SPS.

Figure 1.

Time interval between first prior positive polymerase chain reaction (PCR) test result and date of seroprevalence survey (SPS) among residents and staff in 24 participating skilled nursing facilities (Los Angeles County, August–September 2020; n = 258). Shown are the weeks elapsed since the first positive severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) PCR result and the date of the SPS for all participants with a history of both a positive PCR result and a serologic specimen obtained on the SPS date. For those with a positive serologic result (n = 241), the median interval (interquartile range [IQR]) between positive PCR test date and date of positive serologic result was 14 (12–17) weeks. For those with a negative serologic result (n = 17), the median (IQR) interval was 14 (12–16) weeks. Participants’ prior PCR results were obtained from the Los Angeles County Department of Public Health surveillance database (Integrated Reporting Investigation and Surveillance System); PCR results obtained <2 weeks before the SPS were excluded. The Los Angeles County Public Health Laboratory performs an orthogonal algorithm whereby positive results in specimens are confirmed using a secondary test. Both the Abbott and Diasorin SARS-CoV-2 assays were used for each serologic specimen. Participants were categorized as seropositive if either test yielded a positive result.

DISCUSSION

We conducted a SPS at 24 LAC SNFs to determine more comprehensively the prevalence of current and past SARS-CoV-2 infection. The majority of resident and staff study participants had COVID-19 that was not detected with past testing policies and practices. The results indicate that past COVID-19 outbreaks in SNFs were potentially larger than previously recognized based on counting persons with positive test results alone. It is likely that the large number of persons in SNFs with undiagnosed COVID-19 complicated efforts to control past outbreaks. Undiagnosed COVID-19 in staff members could have resulted in hidden introduction at some of these SNFs, and undetected COVID-19 in residents could have contributed to unrecognized transmission within facilities. The recent declines in the number of COVID-19 cases and outbreaks in SNFs could be partly attributable to the substantial proportion of residents and staff with immunity to SARS-CoV-2. Conversely, SNFs might experience an increase in the number of COVID-19 cases as their proportion of new susceptible residents and staff increases over time. Therefore, continued rigorous adherence to infection control procedures by SNF staff will be needed to prevent and control COVID-19 outbreaks.

A greater proportion of residents were seropositive for SARS-CoV-2 compared with staff. The exact reasons for why residents seem to be at increased risk for COVID-19 compared with staff are not known. It is likely that staff members benefited from having access to personal protective equipment to reduce their risk of infection within SNFs. Although residents are encouraged to wear a face covering at all times, adherence to this recommendation is unclear and nonmedical face coverings are not intended to protect against infection. Moreover, the majority of residents in our study resided in multiple-occupancy rooms, which is typical of most nursing facilities in the United States. Therefore, in addition to their own personal risk for acquiring COVID-19, residents could have the added risk of infection based on their roommate’s exposures (eg, visits from friends and family). Although all SNFs in LAC now conduct daily symptom screening for respiratory illness and have physically separated “red zones” to isolate residents with COVID-19, residents still remain at increased risk for infection from a roommate or staff member who might have asymptomatic or undetected COVID-19 [8].

Our results allow us to estimate the relative risk for having COVID-19 among SNF residents, compared with the general LAC population. There have been 2 assessments of the seroprevalence of SARS-CoV-2 in the LAC general population to date. A community seroprevalence study conducted in April 2020 identified that 35 (4.1%) of 865 participants were seropositive for SARS-CoV-2 [9]. Another study assessed SARS-CoV-2 seropositivity among 790 LAC university students from 29 April 29 to 8 May 2020 and demonstrated a seroprevalence of 4.0% [10]. By comparison, SNF residents and staff in our study, conducted approximately 4 months after the community studies, were 10 and 6 times more likely, respectively, to have evidence of past SARS-CoV-2 infection. Of note, the seropositivity rate among SNF staff in our study also exceeded the rate reported among healthcare personnel at a large academic medical center in LAC from 26 May to 5 June 2020 (approximately 8%) [11]. Another seroprevalence study conducted among hospital and SNF healthcare workers in Rhode Island, from 17 July to 28 August 2020, identified seropositivity rates of 3.1% (95% confidence interval, 2.7%–3.5%) among hospital personnel and 13.1% (11.5%–14.9%) among nursing home personnel; the lower seropositivity rate compared with our study could result from differences in community transmission (ie, risk for staff exposure) and in the types of healthcare workers who participated in the study (ie, intensity of direct patient contact) [12]

Our study has limitations. First, we enrolled a convenience sample of participants from 24 SNFs, which means our results are likely not representative of all 315 SNFs in LAC. Although our results cannot be generalized to all 315 freestanding SNFs within DPH jurisdiction, the characteristics of residents and the infection control practices at the participating SNFs are typical of other centers in LAC. Second, the concentration of antibodies against SARS-CoV-2 decline over time, so participants whose concentrations fell below the level of detection would have been misclassified as SARS-CoV-2 unexposed [13]. In our study, 8.7% of residents and 3.1% of staff who had a documented prior positive PCR result were seronegative for SARS-CoV-2. Furthermore, residents who were infected may have left the facility and been replaced by new residents who were not exposed to the previous outbreaks there, which would also reduce our ability to ascertain the prevalence of past infection at the facility. Finally, our deterministic matching algorithm could have missed prior PCR results if the associated name did not exactly match that provided for enrollment in the study, as well as PCR results for persons residing outside LAC (which would have been reported instead to the relevant jurisdiction).

Our study provides a more comprehensive estimate of the prevalence of exposure to SARS-CoV-2 among SNF residents and staff. These results provide context for assessing the effectiveness of past policies and programs for preventing and controlling COVID-19 in SNFs. Our results can also inform the design of studies to assess the effectiveness of COVID-19 vaccines in preventing infection and outbreaks in SNFs. We demonstrated that a substantial proportion of SNF staff members and residents were potentially unaware of their past infection status. Thus, misclassifying unvaccinated persons with unknown infection as susceptible might reduce the observed effectiveness of the vaccine. Incorporating an orthogonal serologic testing strategy with high test specificity (ie, >99.5%) to minimize the potential for false-positive results could improve classification of persons with evidence of past infection [7]. Although a positive qualitative SARS-CoV-2 antibody test result does not provide a good correlate for protection from subsequent infection, such tests could help investigators understand the potential misclassification of immune status in vaccine effectiveness studies [7].

In conclusion, we demonstrated that past testing practices and policies missed a substantial number of SARS-CoV-2 infections in SNF residents and staff members. Our results could inform vaccine rollout strategies, interpretation of trends in COVID-19 incidence in SNFs by adjusting the denominator for nonsusceptible persons, and the design of potential vaccine effectiveness studies, because unvaccinated persons could be misclassified as susceptible.

Supplementary Data

Supplementary materials are available at The Journal of Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.

Notes

Acknowledgments. We thank members of the Los Angeles County Department of Public Health coronavirus disease 2019 response team, who investigated outbreaks in skilled nursing facilities (Barbara Weiser, Christopher Lindshield, Clarence Monteclaro, Douglas Melnick, Farida Faal, Harold Burger, Hassan Mohamedali, Heather Readhead, Homer Boyd, Jacqueline Bowles, Jasmine Sharma, Karen Streeter, Kathleen Melez, Maxine Liggins, Rajnish Birla, Rosita San Diego, Sheree Poitier, Sherry Thomas, Juliana Aguayo, Yi Tsung Chien, Fatima Ebreo, Patricia Flores, Erika Goff, Vanessa Harbour Bejarano, Sue Kim, Proscovia Lubwama, Angela Madison, Pia Magante, Melany Manalo, Jean Mitchell, Mila Mulugeta, Kathy Ngo, Nguyet Nguyen, Lynn Nottingham, Phuong Ha Pham, Maria Poon, Rodel Rutaquio, Jennifer Smith, Samuel Tan, Duong Tran, Anh Trinh, Genevieve Valenzuela, and Michelle Williams), collected specimens from study participants (Amaka Chukwuma, Jane Maynard, Maria Lewis, Maribel Castillon, Melisa Jaramillo, Nina Silguero, Gabriela Centeno, Karen Sordan, Charles Micu, Olajumoke Akinsaya, Geraldine Chima, Valerie Bragg, Olga Vega, Michelle Chung, Alicia Mancillas, Jonathan Cajuli, and Bernadine Talanoa), and processed laboratory specimens (William Chen, Carey Perkins, Peera Hemarajata, Hector Rivas, Juan Lopez, Lee Borenstein, and Michael Brown), along with other Public Health Laboratory central accessioning, serology, and molecular epidemiology unit staff.

Disclaimer. The contents are those of the authors and do not necessarily represent the official views of, nor an endorsement by, the Centers for Disease Control and Prevention, US Department of Health and Human Services, or the US government.

Financial support. This project was supported by the Centers for Disease Control and Prevention, US Department of Health and Human Services..

Potential conflicts of interest. All authors: No reported conflicts All authors have completed the ICMJE Form for Disclosure of Potential Conflicts of Interest. No disclosures were reported.. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.