The coronavirus disease 2019 (COVID-19) pandemic has strained hospitals and healthcare systems worldwide, with bed capacity and throughput posing considerable challenges during surges. Reference Emanuel, Persad and Upshur1 In the United States, >70% of hospitals have <200 beds and most have a combination of single- and multiple-occupancy rooms, which complicates the placement of COVID-19 patients in cohorts. Reference Stenehjem, Hyun and Septimus2 With an estimated 30% of cases asymptomatic, rapid and reliable testing is important to safely placing inpatients in cohort. Reference Oran and Topol3 Small and critical-access hospitals often lack the volume or capacity for on-site molecular testing. Many safety-net hospitals are financially vulnerable and lack capital for new large-throughput molecular testing equipment. Reference Reiter, Jiang and Wang4 Point-of-care (POC) tests can be easily deployed in these settings. The IDNOW SARS-CoV-2 test (Abbott Laboratories, Abbott, IL) is a rapid POC test that provides results within 15 minutes. Our urban safety-net community hospital deployed admission testing using IDNOW in patients with and without COVID-19 symptoms. Paired parallel swabs were sent to a reference laboratory for confirmatory testing. We evaluated IDNOW accuracy and its impact on infection prevention workflow.
Methods
In June 2020, severe acute respiratory coronavirus virus 2 (SARS-CoV-2) testing for all admissions was implemented regardless of patient symptoms at our 151-bed hospital. Two computerized provider orders were created, one for testing patients with symptoms suggestive of COVID-19 and one for screening patients without suspicion of COVID-19. Each order triggered paired parallel nasal sample and nasopharyngeal sample collection. Testing was performed on site using IDNOW testing with nasal samples and at a reference laboratory using the Roche COBAS SARS-CoV-2 assay (Roche Diagnostics, Mannheim, Germany) with nasopharyngeal samples. The nasopharyngeal samples for COBAS were collected and placed in viral transport media (VTM) and were then refrigerated at 4–8°C. Samples were transported thrice daily to the reference laboratory 10.7 km (6.7 miles) away. The nasal samples were collected and transported immediately at room temperature in a sterile tube to the laboratory on site, where IDNOW testing was performed in accordance with the manufacturer’s instructions. Results were posted in the electronic medical record (EMR). Positive results were considered critical and were communicated to clinicians verbally.
We included patients who underwent paired parallel testing in the emergency department for admission from June 1, 2020, to October 31, 2020, when testing protocols changed. We assessed demographic characteristics and the diagnostic accuracy (ie, sensitivity, specificity, and positive and negative predictive values) of IDNOW testing comapred to COBAS was calculated using McNemar testing with exact binomial 95% confidence intervals (CIs). To evaluate the impact of test accuracy on infection prevention, median times between IDNOW and COBAS results were calculated for all admissions and for false-negative IDNOW results. We searched for nosocomial cases, defined as those who tested positive after initial dual negative tests on admission. The University of Illinois at Chicago Institutional Review Board approved this study and granted a waiver of the requirement for informed consent. The Abbott and Roche companies were not involved in the study design or analysis.
Results
Overall, this study included 2,674 admissions with 2,163 (81%) paired IDNOW and COBAS results. Demographically, 1,122 (52%) patients were male, median age was 43 years (IQR, 28–57), 789 (36%) were Black, and 1,153 (53%) were Hispanic or Latino. The diagnostic accuracy of IDNOW compared to COBAS is summarized in Table 1. During the study period, the SARS-CoV-2 test positivity rate in Chicago ranged from 4% to 11%. The median times from sample collection to IDNOW and COBAS results were 0.8 hours (IQR, 0.6–1.2) and 33.8 hours (IQR, 24.2–47.2), respectively. Among 27 false-negative IDNOW results, the median time difference was 15.8 hours (IQR, 1.0–39.2) between IDNOW and COBAS results. No nosocomial cases were identified during our study.
Diagnostic Accuracy of IDNOW for SARS-CoV-2
Note. CI, confidence interval; PPV, positive predictive value; NPV, negative predictive value.
Discussion
Universal SARS-CoV-2 testing on admission can be an important strategy to safely place patients in cohorts in hospitals with double-occupancy rooms. IDNOW was easily deployed and identified 74% of patients with COBAS-confirmed SARS-CoV-2. Sensitivity was higher (85%) for patients admitted with clinical suspicion of COVID-19 and lower (56.4%) for those without. These results align with prior studies demonstrating lower accuracy of IDNOW compared with other platforms. Reference Harrington, Cox and Snowdon5,Reference Basu, Zinger and Inglima6 Many reasons for this have been postulated, including sample collection modality, symptom onset, user variability, and lag between sample collection and testing. Reference Kucirka, Lauer, Laeyendecker, Boon and Lessler7 Despite these limitations, IDNOW played an important role in identifying cases early and improving throughput in combination with robust infection prevention protocols and engineering controls. The attack rate of COVID-19 in double-occupancy rooms is high. One report showed that 39% of exposed roommates converted within 5 days after exposure. Reference Karan, Klompas, Tucker, Baker, Vaidya and Rhee8 Nosocomial transmission risk has increased with new highly contagious variants. Reference Klompas, Pandolfi, Nisar, Baker and Rhee9 At our facility, patients with negative IDNOW results and suspected COVID-19 were placed in single-occupancy rooms pending COBAS results. Patients without symptoms of COVID-19 and negative IDNOW results were placed in cohorts using a zoned double-occupancy strategy that involved floor-to-ceiling plexiglass barriers, face masks for source control, and the use of a commode to defer sharing a bathroom pending COBAS results. Reference Cadnum, Jencson and Donskey10 With a median time difference of 33.8 hours between IDNOW and COBAS, significant exposure and nosocomial transmission pose a risk in dual occupancy rooms. Despite this risk, no nosocomial cases were identified in our study period, highlighting the importance of engineering controls and infection prevention protocols.
This study had several limitations. It was conducted before the emergence of highly infectious variants such as SARS-CoV-1 ο (omicron), and it was conducted at a single site, potentially limiting generalizability. Nevertheless, these results contribute a valuable assessment of the diagnostic accuracy of IDNOW in symptomatic and asymptomatic individuals. We have also provided real-world data on pragmatic implementation of rapid testing and infection prevention strategies. Our hospital is representative of many safety-net hospitals with double-occupancy rooms, and our approach may provide a valuable model for testing, infection prevention protocols, and engineering controls.
Acknowledgments
The authors acknowledge the colossal efforts of healthcare workers and essential workers during the COVID-19 pandemic.
Financial support
No financial support was provided relevant to this article.
Conflicts of interest
All authors report no conflicts of interest relevant to this article.
Open access
