Abstract
Introduction
Nosocomial case (NC) of COVID-19 infections is a challenge for hospitals. We report the results of a seven-month prospective cohort study investigating COVID-19 patients to assess unexpected cases (UC) (no COVID-19 precautionary measure application since admission) and NC.
Patients and methods
Investigation by an infection control team of 844 patients with COVID-19 infection hospitalized for more than 24 hours (cases).
Results
A total of 301 UC were identified (31% after contact tracing) with a total of 129 contact patients, and 27 secondary cases for 59 of them. In geriatric wards, 50% of cases were UC. NC represented 18% of cases (37% in geriatric wards), mainly identified after contact tracing of wandering cases.
Conclusion
A rapid infection control response is essential to contain nosocomial transmission, along with detailed contact tracing and screening policy. Dealing with wandering elderly patients remain challenging for HCWs.
Keywords: SARS-CoV-2, Cross-transmission, Nosocomial infection, Contact tracing
1. Introduction
The severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) pandemic continues to place a significant burden on health services worldwide. Nosocomial cases (NC) of COVID-19 infection are a challenge for hospitals where patients are at high risk of severe infection [1].
Quick identification of any situation at risk of epidemic spread within hospitals is a key feature to prevent cross-transmission. COVID-19 patients hospitalized without specific barrier measures for COVID-19 represent one of these situations because those unexpected cases (UC) can lead to contamination of other patients within contact patients or healthcare workers (HCW). Furthermore, tackling nosocomial acquisition is necessary to adjust infection control measures in order to ensure patients and HCWs safety [2], [3] . Because understanding the risk of spread due to UC or nosocomial SARS-CoV-2 acquisition is fundamental to implement infection control strategies, we conducted a prospective cohort study to describe UC, contact patients, and NC as well as circumstances for these situations to identify specificities and epidemiological situations at risk of epidemic spread within hospitals.
2. Methods
This prospective cohort study took place at Bordeaux university hospital (France), with more than 3,000 beds. We prospectively included all patients with COVID-19 infection diagnosed by RT-PCR hospitalized for more than 24 hours from August 3, 2020 to February 28, 2021 (i.e., case). Patient data were collected by an infection control team (ICT) from electronic medical records: age, gender, hospitalization unit, reason for hospitalization, onset of symptom, context of diagnosis (symptoms, before surgery or transfer, contact tracing).
The implemented infection control program included dedicated COVID-19 wards; medical masks for HCWs, patients and visitors; restriction of visitors; screening patients with RT-PCR on admission and after seven days of hospitalization. Suspected or confirmed cases were isolated in a single room for contact and droplet precautions. Furthermore, we stepped up the use of personal protective equipment (PPE) when caring for all patients, irrespective of their COVID-19 status: eye protection, respirator FFP2 masks and aprons for performing aerosol-generating procedures; aprons for contact with patients or their environment. Every COVID-19 diagnosis was notified to the ICT for investigation.
Upon COVID-19 case notification, the ICT reviewed all records to identify: (1) UC when COVID-19 precautionary measures had not been applied since admission; (2) contact patients (shared room or other exposure to UC, notably wandering elderly patients) and secondary cases during hospitalization among contact patients; (3) NC depending on the time between admission and the date of diagnosis: probably nosocomial (from 8 to 14 days) or certainly nosocomial (beyond 15 days) [4]. The origin of COVID-19 was considered as community-acquired between 0 and 2 days, and as undefined between 3 and 7 days after admission.
The cumulative incidence of cases was the number of cases divided by the number of patients hospitalized for more than 24 hours (global and per month) during the study period. The incidence of NC (i.e., probably and certainly nosocomial) was the number of cases who acquired nosocomial COVID-19 divided by the number of patients hospitalized for more than 24 hours during the study period.
3. Results
3.1. Study population
A total of 844 patients with COVID-19 were hospitalized for more than 24 hours during the study period (Table 1 ). The ICT investigated a median of 28 cases per week (range: 1–49). During the study period, a total of 431,103 patients were hospitalized for more than 24 hours; thus, the global incidence of cases was 1.9 for 1,000 admissions and varied from 1.3 to 2.2 cases for 1,000 admissions per month.
Table 1.
Description of the 844 COVID-19 cases hospitalized FOR more than 24 hours depending on their status of unexpected COVID-19 case or not - period from August 3, 2020, to February 28, 2021 - Bordeaux University Hospital – France.
| Variable | Unexpected COVID-19 |
Total | |
|---|---|---|---|
| Yes (n = 301) | No (n = 543) | n = 844 | |
| Gender, ref = male [N (%)] | 172 (34.6) | 325 (65.4) | 797 (58.89) |
| Age [mean ± SD)] | 68.4 ± 22,8 | 64.3 ± 22,3 | 65.8 ± 22.5 |
| Ward specialty [N (%)] | |||
| Surgery | 40 (57.1) | 30 (42.9) | 70 (100) |
| Geriatric | 70 (50.7) | 68 (49.3) | 138 (100) |
| Medicine | 155 (38.5) | 248 (61.5) | 403 (100) |
| Intensive Care Unit | 7 (6.5) | 100 (93.6) | 107 (100) |
| Emergency Room | 29 (23.0) | 97 (77.0) | 126 (100) |
| Reason for hospitalization [N (%)] | |||
| Impaired general condition | 54 (85.3) | 14 (14.7) | 68 (100) |
| COVID-19 | 33 (11.6) | 251 (88.4) | 284 (100) |
| Infection | 49 (64.5) | 27 (35.5) | 76 (100) |
| Surgical reason | 33 (62.3) | 20 (37.7) | 53 (100) |
| Medical reason | 69 (61.1) | 44 (38.9) | 113 (100) |
| Missing data | 63 | 187 | |
| COVID-19 classification [N (%)] | |||
| Community-acquired | 126 (20.5) | 490 (79.5) | 616 (100) |
| Undefined | 44 (57.9) | 32 (42.1) | 76 (100) |
| Probably nosocomial | 58 (85.3) | 10 (14.7) | 68 (100) |
| Certainly nosocomial | 73 (86.9) | 11 (13.1) | 84 (100) |
3.2. Unexpected cases
Unexpected cases were identified in 36% (n = 301, 95%CI [33.7; 38.3]) of our cases with a monthly variation from 3% to 50%. Patients were more frequently women, older, and UC concerned more than 50% of cases in surgery and geriatrics units (Table 1). UC were mostly diagnosed because of contact tracing after exposure (31%, 95%CI [27.6; 34.4]), for symptoms of COVID-19 during hospitalization (24%, 95%CI [21.3; 26.7]) or before surgery or transfer (23%, 95%CI [20.5; 25.5]). COVID-19 was suspected upon admission for 33 UC, but COVID-19 precautionary measures were lifted because of a first negative RT-PCR test.
3.3. Contact patients
Investigations identified 129 contact patients from 59 UC (from 1 to 19; mean 6.49; SD 6.94); primarily in surgery (60%, 95%CI [49.7; 70.3]) and medicine units (21%, 95%CI [17.5; 24.5]) rather than geriatric (11%, 95%CI [9.2; 12.8]) or emergency units (7%, 95%CI [5.9; 8.1]). The difference in rates depending on the type of medical specialty is significant (p < 0.001). A total of 27 secondary cases with a positive RT-PCR test were identified. Secondary acquisition was more frequent in geriatric units (62%, 95%CI [51.4; 72.6]) than in other units (19.5% in medical units, 7% in emergency units, and 0% in surgical or intensive care units); it was also more frequent in wandering patients (60%, 95%CI [49.7; 70.3]) than in patients from shared rooms (29.7%, 95%CI [24.7; 34.7]).
3.4. Nosocomial cases
Among our population study, 18% (n = 152, 95%CI [6.2; 29.8],) was possibly (8.1%, 95%CI [3; 13.2], n = 68) or certainly nosocomial cases (9.9%, 95%CI [3.6; 16.2], n = 84) (Table 1). NC were more frequently observed in geriatric units (37.7%, 95%CI [31.8; 43.6]) than in other units. NC was identified after contact tracing for 42% (95%CI [35.4; 48.6] or because of onset of acute symptoms for 31% (95%CI [26.2: 35.8] (Table 2 ). The global cumulative incidence of NC was 0.3 per 1,000 admissions and varied from 0.04 to 0.59 cases per 1,000 admissions per month.
Table 2.
Description of the 844 COVID-19 cases hospitalized for more than 24 hours depending on their COVID-19 status (Hospital-acquired or not) – period from August 3, 2020 to February 28, 2021 – Bordeaux University Hospital – France.
| Variable | Hospital-acquired COVID-19 |
Total | |
|---|---|---|---|
| Yes (n = 152) | No (n = 692) | n = 844 | |
| Gender, ref = male [N (%)] | 88 (57.9) | 409 (59.1) | 497 (58.89) |
| Age [mean ± SD] | 73.6 ± 18.4 | 64.1 ± 23.0 | 65.8 ± 22.5 |
| Ward specialty [N (%)] | |||
| Surgery | 10 (14.3) | 60 (85.7) | 70 (100) |
| Geriatric | 52 (37.7) | 86 (62.3) | 138 (100) |
| Medicine | 83 (20.6) | 320 (79.4) | 403 (100) |
| Intensive Care Unit | 3 (2.8) | 104 (97.2) | 107 (100) |
| Emergency Room | 4 (3.2) | 122 (96.8) | 126 (100) |
| Reason for hospitalization [N (%)] | |||
| Impaired general condition | 43 (63.2) | 25 (36.8) | 68 (100) |
| COVID-19 | 0 | 284 (92.3) | 284 (100) |
| Infection | 33 (43.4) | 43 (56.6) | 76 (100) |
| Surgical reason | 11 (20.7) | 42 (79.3) | 53 (100) |
| Medical reason | 33 (29.2) | 80 (70.8) | 113 (100) |
| Missing data | 32 | 218 | |
| Diagnostic context [N (%)] | |||
| Contact tracing | 64 (68.8) | 29 (31.2) | 93 (100) |
| Acute symptoms | 47 (40.5) | 69 (59.5) | 116 (100) |
| Before surgery or transfer | 21 (30.0) | 49 (70.0) | 70 (100) |
| Other reason | 20 (46.5) | 23 (53.5) | 43 (100) |
| Missing data | 0 | 522 | |
4. Discussion
To control transmission of SARS-CoV-2 we used a bundle of infection control measures for early detection, isolation, and systematic notification of cases for investigation. The aim of the present study was to have accurate knowledge of specificities and epidemiological situations at risk of epidemic spread within hospitals in order to have best-fit policies.
We identified frequent UC for which isolation precautions need to be adjusted and probably made it possible to limit secondary cases. Among these UC, few had secondary cases identified due to rapid identification with systematic screening and to our policy of stepping up PPE indication for all patients. To date there is currently very little evidence available on the effectiveness of interventions, excluding PPE, to prevent the spread of SARS-CoV-2 in hospital settings [5]. The proportion of NC was high among our study population and varied by ward specialty. However, comparison with the literature is complex due to various screening strategies and no universally accepted definition of NC. Incidence rates reported in seven studies with a proportion of hospital-onset COVID-19 infections among all hospitalized confirmed COVID-19 patients ranged from 0% to 15% [2]. The lower proportion of NC in surgery wards could be explained by surgery being postponed when COVID-19 screening was positive before admission. Circumstances for these risk situations more frequently concerned geriatric patients and particularly those exposed to wandering patients. Management of wandering patients is challenging for HCWs because it is both impossible to make them strictly respect barrier measures to protect other patients (wearing mask, limiting wandering, limiting physical contact with other patients), and to limit their freedom of movement through physical constraint due to ethical questions [6]. Of note, a majority of NC was screened for other reason than symptoms, highlighting the importance of effective and sustainable screening strategies to prevent cross-transmission from asymptomatic carriers [7]. We pointed the limits of screening strategies, due to RT-PCR false negative results on admission which lead HCWs to lift COVID-19 precautionary measures for symptomatic patients [8].
Our study was conducted on an exhaustive large cohort of hospitalized COVID-19 cases. However, this was an observational study without follow-up of outpatients; thus, nosocomial acquisition could be underestimated. Because data available was relatively rare in the literature we were not able to compare our results to those of other hospitals.
As rapid infection control response is essential to contain the risk of COVID-19 nosocomial transmission, a screening policy for the early detection of cases and a detailed contact tracing of patients with positive RT-PCR are necessary. Dealing with wandering elderly patients also remain challenging for HCWs in this COVID-19 context.
CRediT authorship contribution statement
Y. Carré: Formal analysis, Investigation, Writing – original draft, Writing – review & editing. M. Coppry: Writing – review & editing. C. Bataille: Writing – review & editing. L. Vivier: Writing – review & editing. A. Lasheras-Bauduin: Writing – review & editing. A.-M. Rogues: Writing – review & editing.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References
- 1.Ponsford M.J., Ward T.J.C., Stoneham S.M., Dallimore C.M., Sham D., Osman K., et al. A Systematic Review and Meta-Analysis of Inpatient Mortality Associated With Nosocomial and Community COVID-19 Exposes the Vulnerability of Immunosuppressed Adults. Front Immunol. 2021;12:17. doi: 10.3389/fimmu.2021.744696. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Abbas M., Zhu N.J., Mookerjee S., Bolt F., Otter J.A., Holmes A.H., et al. Hospital-onset COVID-19 infection surveillance systems: a systematic review. J Hosp Infect. 2021;115:44–50. doi: 10.1016/j.jhin.2021.05.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Meredith L.W., Hamilton W.L., Warne B., Houldcroft C.J., Hosmillo M., Jahun A.S., et al. Rapid implementation of SARS-CoV-2 sequencing to investigate cases of health-care associated COVID-19: a prospective genomic surveillance study. Lancet Infect Dis. 2020;20(11):1263–1271. doi: 10.1016/S1473-3099(20)30562-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Rhee C., Baker M., Vaidya V., Tucker R., Resnick A., Morris C.A., et al. Incidence of Nosocomial COVID-19 in Patients Hospitalized at a Large US Academic Medical Center. JAMA Netw Open. 2020;3(9):9. doi: 10.1001/jamanetworkopen.2020.20498. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Pham T.M., Tahir H., van de Wijgert J.H.H.M., Van der Roest B.R., Ellerbroek P., Bonten M.J.M., et al. Interventions to control nosocomial transmission of SARS-CoV-2: a modelling study. BMC Med. 2021;19(1):211. doi: 10.1186/s12916-021-02060-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Liddell K., Ruck Keene A., Holland A., Huppert J., Underwood B.R., Clark O., et al. Isolating residents including wandering residents in care and group homes: Medical ethics and English law in the context of Covid-19. Inter J Law Psychiat. 2021;74:1–16. doi: 10.1016/j.ijlp.2020.101649. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Carrara E., Ong D.S.Y., Hussein K., Keske S., Johansson A.F., Presterl E., et al. ESCMID guidelines on testing for SARS-CoV-2 in asymptomatic individuals to prevent transmission in the health care setting. Clin Microbiol Infec. 2022;28(5):672–680. doi: 10.1016/j.cmi.2022.01.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kucirka L.M., Lauer S.A., Laeyendecker O., Boon D., Lessler J. Variation in False-Negative Rate of Reverse Transcriptase Polymerase Chain Reaction-Based SARS-CoV-2 Tests by Time Since Exposure. Ann Intern Med. 2020;173(4):262–267. doi: 10.7326/M20-1495. [DOI] [PMC free article] [PubMed] [Google Scholar]