The first case of coronavirus disease 2019 (COVID-19) in Connecticut was reported in a nursing home (NH) on March 15, 2020. Within the next 2 months, 80.0% of Connecticut’s 215 NHs reported at least 1 case of COVID-19, accounting for 61.6% of COVID-19 deaths in the state.1 Residents were initially tested for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) only if symptomatic, as per recommendations from the Centers for Disease Control and Prevention. In early May, NHs were prioritized and selected for point prevalence surveys to provide a baseline for residents not previously identified as infected. We describe the results of these surveys in a targeted subset of Connecticut NHs between May 2 and 19, 2020.
We prioritized NHs in which establishment of baseline resident SARS-CoV-2 status would improve control measures, such as cohorting of individuals, based on a low proportion of previously infected residents (but at least 1 case), a high number of residents with unknown SARS-CoV-2 status, and evidence of at least 1 newly identified case in the previous 7 days. NHs with data verified by authors and completing surveys by May 19 were included. Several lower-priority NHs that had executed testing independently or expressed interest in surveys were also included.
Nasopharyngeal swabs were tested via polymerase chain reaction–based methods for detection of SARS-CoV-2 using 6 platforms in 8 laboratories (eMethods in the Supplement). Verbal consent and specimen collection were obtained by NH staff for residents without a prior confirmed SARS-CoV-2–positive test result. Symptoms were assessed by NH staff on the day of the survey, including atypical presentations in elderly individuals, and for 14 days after testing, following guidelines from the Centers for Disease Control and Prevention.
NH quality rating and number of licensed beds were obtained from the Nursing Home Compare database and case rates and location from the state.
The surveys were conducted as part of the state’s public health response for outbreak control and, therefore, were exempt from the need for institutional review board approval.
Point prevalence surveys were conducted in 33 NHs across Connecticut, representing 15.3% of NHs statewide (n = 215). The geographic distribution of included and remaining NHs is shown in the Figure. Included NHs had a quality rating of 3.58 stars (vs 3.93 stars in the remaining NHs; P = .24) and 135 beds (vs 127 beds; P = .23), and the case rate in the towns in which they were located was 617 cases/100 000 individuals (vs 1263/100 000; P < .001).
Overall, 2117 residents were tested (median per NH, 51; range, 14-242) and 601 (28.3%) were positive. Of the 601 positive residents, 530 (88.2%) were asymptomatic when sampled; 11.7% (62/530) developed symptoms within 14 days (presymptomatic). All SARS-CoV-2–positive residents were asymptomatic or presymptomatic at the time of testing in 45.5% of NHs (Table). The median time from the first case to the survey was 37 days (range, 6-54). Nineteen facilities had at least 50% of residents testing positive (range, 50%-94%), with testing occurring a median of 37 days (range, 7-54) from detection of the first facility case.
In a sample of NHs in Connecticut with at least 1 COVID-19 case in the week preceding point prevalence surveys, 28% of residents tested positive, of which 78% remained asymptomatic and 10% were presymptomatic. The proportion of asymptomatic residents is higher than in previous smaller survey studies of long-term care facilities, which found percentages of 50% to 55%.2-4
The study limitations include sampling of selected NHs in 1 state and no staff testing. The high proportion of asymptomatic patients may be overestimated due to challenges in ascertaining symptoms in elderly individuals with atypical or mild presentations, exclusion of symptomatic patients who previously tested positive, or the possibility of symptom resolution before testing. In addition, COVID-19 rates in surrounding communities were not factored into NH prioritization and repeat testing was not performed.
NHs house particularly vulnerable populations because of their age, rates of comorbidities, and clustering.5 Point prevalence surveys may be necessary to limit spread in NHs, with a prioritized rollout in situations with limited control and testing capacity. Repeated testing in NHs may also be useful.4,6
Corresponding Author: Sunil Parikh, MD, MPH, Department of Epidemiology of Microbial Diseases, Yale School of Public Health, 60 College St, New Haven, CT 06520 (sunil.parikh@yale.edu).
Accepted for Publication: July 24, 2020.
Published Online: August 10, 2020. doi:10.1001/jama.2020.14984
Author Contributions: Drs Parikh and Leung had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Drs Parikh and O’Laughlin contributed equally to this work.
Concept and design: Parikh, O’Laughlin, Ehrlich, Leung.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Parikh, O’Laughlin, Ehrlich, Harizaj.
Critical revision of the manuscript for important intellectual content: Parikh, O’Laughlin, Ehrlich, Campbell, Durante, Leung.
Statistical analysis: Parikh, Ehrlich.
Administrative, technical, or material support: O’Laughlin, Ehrlich, Campbell, Harizaj, Durante, Leung.
Supervision: Parikh, Leung.
Conflict of Interest Disclosures: None reported.
Disclaimer: The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.
Additional Contributions: We thank the Connecticut Department of Public Health team including Barbara Cass, RN; Anu Paranandi, DO, MPH; Erin Grogan, RN, MS; Naissa Piverger, MPH; Meghan Maloney, MPH; Ellen Neuhaus, MD; Surjit Sethuraman; Kim Hriceniak, RNC, BSN; Kristin Soto, MPH; and Terry Rabatsky-Ehr, MS, MPH, for building and maintaining a nursing home surveillance system for coronavirus disease 2019. We thank Ben Gagne and members of the Connecticut National Guard for assisting in the deployment of point prevalence survey test kits. We thank Linda Niccolai, PhD, and team from the Yale School of Public Health for assistance with developing and executing the nursing home surveillance system. No individuals listed received compensation for their contributions to this work.
3.Feaster
M, Goh
YY. High proportion of asymptomatic SARS-CoV-2 infections in 9 long-term care facilities, Pasadena, California, USA, April 2020.
Emerg Infect Dis. 2020;26(10). doi:
10.3201/eid2610.202694
PubMedGoogle Scholar 4.Sanchez
GV, Biedron
C, Fink
LR,
et al. Initial and repeated point prevalence surveys to inform SARS-CoV-2 infection prevention in 26 skilled nursing facilities: Detroit, Michigan, March-May 2020.
MMWR Morb Mortal Wkly Rep. 2020;69(27):882-886. doi:
10.15585/mmwr.mm6927e1PubMedGoogle ScholarCrossref 5.Richardson
S, Hirsch
JS, Narasimhan
M,
et al; Northwell COVID-19 Research Consortium. Presenting characteristics, comorbidities, and outcomes among 5700 patients hospitalized with COVID-19 in the New York City area.
JAMA. 2020;323(20):2052-2059. doi:
10.1001/jama.2020.6775PubMedGoogle ScholarCrossref