key: cord-273235-jxsldz4o authors: Kennelly, Sean P; Dyer, Adam H; Noonan, Claire; Martin, Ruth; Kennelly, Siobhan M; Martin, Alan; O’Neill, Desmond; Fallon, Aoife title: Asymptomatic carriage rates and case-fatality of SARS-CoV-2 infection in residents and staff in Irish nursing homes date: 2020-09-28 journal: Age Ageing DOI: 10.1093/ageing/afaa220 sha: doc_id: 273235 cord_uid: jxsldz4o BACKGROUND: SARS-CoV-2 has disproportionately affected nursing homes (NH). In Ireland, the first NH case COVID-19 occurred on 16/03/2020. A national point-prevalence testing program of all NH residents and staff took place (18/04/2020–05/05/2020). AIMS: To examine characteristics of NHs across three Irish Community Health Organisations (CHOs), proportions with COVID-19 outbreaks, staff and resident infection rates symptom-profile, and resident case-fatality. METHODS: Forty-five NHs surveyed requesting details on occupancy, size, COVID-19 outbreak, outbreak timing, total symptomatic/asymptomatic cases, and outcomes for residents from 29/02/2020–22/05/2020. RESULTS: Surveys were returned from 62.2% (28/45) of NHs (2043 residents, 2,303 beds). Three-quarters (21/28) had COVID-19 outbreaks (1741 residents, 1972 beds). Median time from first COVID-19 case in Ireland to first case in these NHs was 27.0 days. Resident incidence was 43.9% (764/1741): 40.8% (710/1741) laboratory-confirmed, with 27.2% (193/710) asymptomatic, and 3.1% (54/1741) clinically-suspected. Resident case-fatality was 27.6% (211/764) for combined laboratory-confirmed/clinically-suspected COVID-19. Similar proportions of residents in NHs with “early-stage” (<28 days) versus “later-stage” outbreaks developed COVID-19. Lower proportions of residents in “early” outbreak NHs had recovered compared to those with “late” outbreaks (37.4% vs 61.7%; χ(2) = 56.9, P < 0.001). Of 395 NH staff across twelve sites with confirmed COVID-19, 24.7% (99/398) were asymptomatic. There was a significant correlation between the proportion of staff with symptomatic COVID-19 and resident numbers with confirmed/suspected COVID-19 (Spearman’s rho = 0.81, P < 0.001). CONCLUSION: This study demonstrates the significant impact of COVID-19 on the NH sector. Systematic point-prevalence testing is necessary to reduce risk of transmission from asymptomatic carriers and manage outbreaks in this setting.  A significant proportion of residents and staff with COVID-19 may be asymptomatic  Outbreak timing ('early' vs 'late') may impact clinical outcomes for residents  Systematic mass-testing of nursing home residents and staff allows timely identification of asymptomatic individuals  Identification and isolation of asymptomatic individuals is essential for outbreak eradication and recovery in this setting SARS-CoV-2 and the related illness COVID-19 has disproportionately affected nursing homes (NH) since emerging in late 2019 [1] . NH residents are among the frailest in society, with multiple comorbidities and high levels of care needs. NHs vary in size, staffing, governance, and integration with wider health systems locally, nationally, and internationally [2] . Policies on testing and reporting of COVID-19 in NHs differ between countries. Care home residents accounted for 53-82% of COVID-19-related deaths where confirmed and probable deaths are reported (Belgium, Canada, France, Ireland) [1] . Guidance published for long-term care facilities on infection prevention and management in the context of COVID-19 [3, 4] can be challenging to implement due to variability in facilities. Ireland has a "mixed market" of NH provision. Public NHs, owned and operated by the governmentfunded Health Service Executive (HSE), and private NHs, by individual providers/provider entities. All are registered with the Health Information and Quality Authority (HIQA) and comply with regulations for standards of care [5] . There were 31,220 beds across 581 NHs in December 2018 [6] : the majority private NHs with approximately 25,000 residents and 121 (21%) public NHs with approximately 5,000 residents. The first laboratory-confirmed COVID-19 case in Ireland occurred in the community on 29/02/2020 [7] . The first COVID-19 NH clusters were reported on 16/03/2020. The incidence of COVID-19 in long-term care residents in Ireland was 133/1000 on 06/05/2020. As of 20/05/2020, 258 NH clusters were reported accounting for 4,872 cases and 851 deaths. The national COVID-19-related casefatality figure for NH residents aged ≥70 years is 21% [7] . This may be an underestimate, failing to capture earlier attributable deaths. Given emerging evidence on the impact of COVID-19 on NHs and on asymptomatic infection, the HSE directed a National Ambulance Service (NAS)-led testing protocol in NHs followed by a national point-prevalence COVID-19-mass-testing program of all residents and staff in NHs from 18/04/2020-05/05/2020. Following this, systematic testing was completed at two-weekly intervals as new NH COVID-19 cases were identified. To examine characteristics of NHs across three Irish Community Health Organisations (CHOs) in Dublin/Eastern Ireland, the proportion with COVID-19 outbreaks, symptom profile for staff and residents, and resident mortality rates. An information sheet and survey document was distributed to lead Nursing and/or Medical Officers in forty-five NHs across three CHOs, followed by a telephone call to obtain consent and aid survey completion, ensuring correct interpretation of information. For inclusion in analysis, an outbreak was defined as ≥1 resident with laboratory-confirmed COVID-19. NHs with one to two staff members and no residents affected weren't considered outbreaks on the basis that infection may have been acquired outside of the NH setting. An "active" outbreak was defined as being within twenty-eight days of symptom onset for most recent laboratory-confirmed case in a resident/staff member. Data are reported as proportions (and percentages) and mean/median with standard deviation (SD)/interquartile ranges (IQR) as appropriate. Data on NH size (number of beds: <50, 51-100, >100) and occupancy (percentage of beds: <75%, 75-85%, 86-95%, >95%) were categorised for betweengroup analysis. Median time from first confirmed COVID-19 case in Ireland to first case in included NHs was used to distinguish between "early" and "late" outbreaks in context of the first-wave of pandemic. Comparisons of proportions between groups was carried out using a Chi-square statistic. Spearman Rank correlations were used to analyse correlations between variables. The World Medical Association -Declaration of Helsinki Ethical Principles were followed [8] , adhering to all standards in the acquisition, anonymisation, and reporting of data through consultation with the local research ethics committee. Complete surveys were returned from 62.2% ( (300/312) occurred in "outbreak" NHs with a mortality rate of 17.2% (300/1741). An outbreak was recorded in 75.0% (21/28) of facilities: four public and seventeen private [ Table 1 ]. Occupancy rates at the start of the study period were 95.1% and 87.7% in public and private NHs respectively, decreasing to 75.2% in public and 73.2% in private NHs by 22/05/2020. Eight NHs (38.1%) had ≥80% single rooms in line with regulatory standards [9] . There was no association between adherence to this standard and outbreak occurrence (χ2=1.37, p=0.24). with a "late" first case. NHs with "early" outbreaks had a higher number of deaths expressed as a proportion of total residents but similar CFR for residents with confirmed/suspected COVID-19 as NHs with "late" outbreaks [ Table 3 ]. By 29/02/2020, 55.8% (396/710) of residents with laboratory-confirmed COVID-19 had recovered. A lower proportion had recovered in NHs with "early" outbreaks compared to NHs with "late" outbreaks. A greater proportion of residents remained in isolation in the "early" vs "late" group [ Table 3 ]. Six NHs (6/21, 28.6%) had no residents with confirmed/suspected COVID-19 in isolation at the end of the study period; two (18.1%; 2/11) in the "early" and four (40.0%; 4/10) in the "late" outbreak group. In ten "COVID-19-outbreak" NHs reporting total staffing-levels the median proportion of residents with confirmed/suspected COVID-19 was 43.7% (IQR 34.6-53.4%) overall. Two had a staff/resident ratios <1, six of 1-2 and two with >2 staff members per resident. NHs with staff/resident ratios of <1, those with "late" outbreaks, though this difference was not statistically significant (χ2=2.75, p=0.10) [ Table 3 ]. This large epidemiological study of NHs, residents, and staff demonstrates the disproportionate impact of COVID-19 on this part of the health sector. Within twenty-one NH outbreak clusters in this report, 43.9% of residents had COVID-19 over the twelve-week period (eighty-three-days). Additionally, 29.0% of staff had COVID-19 in ten "outbreak" NHs reporting total staffing. Mass point-prevalence and contact-tracing testing revealed asymptomatic/pre-symptomatic infection in 27.2% of residents, and 24.7% of staff with COVID-19. Overall staffing ratios appeared to impact on case-fatality rates across sites. The CFR due to suspected/confirmed COVID-19 in residents was 26.7%, in contrast to the overall Irish national CFR (as of 06/06/2020) of 5.6%, almost half of which are deaths attributable to NH residents. This bias in the recording of national case-fatality has led some commentators to suggest NH-related mortality should be examined separate to other community deaths [10] . NH residents have increased comorbidities and frailty. Much of the increased mortality risk is associated with these factors. Usual annualised morality rate of residents is estimated at 31.8% [11] typically equating to an expected rate of 7.2% over eighty-three-days. In the eighty-three-days of this study, the mortality rate was 15.3%, overall and 17.2% in "COVID-19-outbreak" NHs, more than double the normal expected rate. With regards international comparisons on NH COVID-19 infection rates and case-fatality, results are consistent with a report on eighty-nine residents from a US facility [12] , with an infection rate of 64% and CFR of 26%, and a series from 394 residents in four UK NHs with an infection rate of 40%, and CFR of 26% [13] . In the UK series 43% of residents and 4% of staff were asymptomatic. This is the first report to investigate the effect of outbreak timing and the potential impact of concurrent mass-testing on outcomes. Earlier in the pandemic, testing was on the basis of symptoms. There were significant challenges accessing timely testing, PPE availability, and clinical knowledge regarding potential "atypical" and asymptomatic presentation of COVID-19 in NHs. When comparing outcomes for NHs with "early" vs "late" outbreaks, while there were similar rates of infection and CFR among residents, greater proportions of asymptomatic staff were identified as a result of masstesting in NHs with later outbreaks. It is possible that some asymptomatic staff-carriers in "early" outbreak NHs had undetectable viral loads by the time of testing, but had unwittingly propagated the spread of infection within NHs at early stages of the pandemic when testing was performed on the basis of symptoms. This potentially resulted in "late" outbreak NHs having greater success in controlling outbreaks in a timely fashion, as evidenced by fewer people in isolation at the end of the study period. Current evidence suggests that asymptomatic carriage is as virulent as symptomatic carriage, highlighting the importance of mass-testing [14] . In NHs with "early" outbreaks, asymptomatic infected individuals were only identified when mass-testing was performed, by which time the outbreak was well-established. Public NHs appeared to have been impacted disproportionately severely relative to private NHs, although we must be cautious interpreting this finding as five small public facilities returned outcomes (4 with outbreaks There are limitations to this study. Despite the sample size, it represents 5% of Irish NHs, and 8% with COVID-19 clusters. 38.8% (17/45) of NHs did not respond within the timeframe and were excluded. Eleven "outbreak" NHs (52.4%, 11/21) did not provide total staffing information. Information, including COVID-19-related deaths, was self-reported, increasing risk of bias, although senior NH medical/nursing staff were best placed to give accurate information. We did not have information on the nature, and severity of symptoms which would be of interest in light of reports of "atypical" symptoms in this cohort. This study demonstrates the substantial impact of COVID-19 on NH residents and staff, supporting evidence that has emerged during the pandemic to date. Proportions of asymptomatic staff/residents and increasing knowledge of "atypical" COVID-19 presentations, supports systematic mass-testing in NH settings. The ability to test, trace, and isolate asymptomatic residents/staff in a timely manner is an essential part of outbreak eradication and recovery in this setting. Mortality associated with COVID-19 outbreaks in care homes: early international evidence. Article in LTCcovid.org, International Long-Term Care Policy Network COVID-19 in nursing homes Infection Prevention and Control guidance for Long-Term Care Facilities in the Context of COVID-19. Geneva, World Health Organization The impact of COVID-19 on people who use and provide long-term care in Ireland and mitigating measures. Country report available at LTCcovid.org, International Long-Term Care Policy Network Overview of the Health System Response to date: Long-term residential healthcare settings WMA Declaration of Helsinki -Ethical Principles for Medical Research Involving Human Subjects National Standards for Residential Care Settings for Older People in Ireland The Importance of Long-term Care Populations in Models of COVID-19 Mortality in older care home residents in England and Wales Presymptomatic SARS-CoV-2 Infections and Transmission in a Skilled Nursing Facility SARS-CoV-2 infection, clinical features and outcome of COVID-19 in United Kingdom nursing homes Clinical Course and Molecular Viral Shedding Among Asymptomatic and Symptomatic Patients With SARS-CoV-2 Infection in a Community