key: cord-0304147-0zjvzzrr
authors: Stokes, W.; Venner, A.; Buss, E.; Tipples, G.; Berenger, B. M.
title: Evaluation of the ID NOW Among Symptomatic Individuals During the Omicron Wave
date: 2022-05-21
journal: nan
DOI: 10.1101/2022.05.19.22275316
sha: 10f6e03ac9e809028b5add8bc676d35d16347df5
doc_id: 304147
cord_uid: 0zjvzzrr

BACKGROUND Point of Care SARS CoV 2 devices, such as the Abbott ID NOW have great potential, to help combat the COVID19 pandemic. Starting in December, 2020, the ID NOW was implemented throughout the province of Alberta, Canada (population 4.4 million) in various settings. We aimed to assess the ID NOW performance during the BA.1 Omicron wave and compare it to previous waves. METHODS The ID NOW was assessed in two distinct locations among symptomatic individuals: acute care (emergency room, urgent care, and hospitalized patients) and community assessment centres (AC) during the period January 5 to 18, 2022. Starting January 5, Omicron represented >95% of variants detected in our population. For every individual tested, two swabs were collected: one for ID NOW testing and the other for either reverse-transcriptase polymerase chain reaction (RT-PCR) confirmation of negative ID NOW results or for variant testing of positive ID NOW results. RESULTS A total of 3,041 paired samples were analyzed (1,139 RTPCR positive). 1,873 samples were from 42 COVID-19 AC and 1,168 from 69 rural hospitals. ID NOW sensitivity for symptomatic individuals presenting to community AC and patients in hospital was 96.0% [95% confidence interval (CI) 94.5 to 97.3%, n=830 RTPCR positive], and 91.6% (95% CI 87.9 to 94.4%, n=309 RTPCR positive), respectively. SARS CoV 2 positivity rate was very high for both populations (44.3% at AC, 26.5% in hospital). CONCLUSIONS Sensitivity of ID NOW SARS CoV 2, compared to RTPCR, is very high during the BA.1 Omicron wave, and is significantly higher when compared to previous SARS CoV 2 variant waves.

The ID NOW (Abbott, Illinois, United States) is approved by the United States Food and Drug 68 Administration (FDA) Emergency Use Authorization for the point of care, rapid detection of 69 severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) in individuals who are within 70 the first 7 days of symptom onset. 1 (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 

The collection and testing of COVID-19 samples followed the same procedures as described in 114 our previous paper. 2 We chose to analyze data from January 5 -18, 2022, when Omicron 115 represented >95% of SARS-CoV-2 variants detected in our population. Not all individuals with 116 ID NOW positive samples had samples collected for testing of variants of concern, due to 117 changes in our testing protocols designed to conserved lab capacity. 118 119 Individuals were given the option to have POCT by ID NOW and routine testing, or routine 120 testing alone. All individuals tested with the ID NOW had two parallel swabs collected. The first 121 swab collected was either a NP swab or OP swab, which was placed in UTM (Yocon Biology, 122

Beijing, China or GDL Korea Co. Ltd, Seoul, Korea) for RT-PCR, and transported to an 123 accredited laboratory at room temperature and stored at 4 o C until processing. The second swab 124 was an OP swab for ID NOW testing (using the OP swab provided in the ID NOW kits). The OP 125 swab for ID NOW testing was always collected second to ensure all individuals had a sample 126 available for RT-PCR (i.e. in case the individual refused the second NP or OP swab). If the ID 127 NOW test was negative, the second swab was sent for confirmatory RT-PCR testing. If the ID 128 NOW test was positive, the second swab was either sent for storage or for variant of concern 129 screening (VoC) testing. All RT-PCR tests sent for variant testing were done within 130 approximately 72 hours from time of collection. During the Omicron wave, only a subset of 131 positive samples were sent for variant testing due to resourcing pressures faced by our 132 laboratory. Positive ID NOW results without subsequent variant testing were considered true 133 positives in this study. 134 135 All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All personnel performing the ID NOW swabbing/testing were trained healthcare workers 148 (HCW), who were previously trained in NP and oropharyngeal swab collection. At time of 149 collection, they asked and recorded whether the patient had symptoms or was asymptomatic. All 150 sites and HCW were trained on the ID NOW collection, transport, and testing processes, at least 151 according to the manufacturer's instructions, prior to ID NOW implementation. AHS staff were 152 trained according to the APL POCT program, which meets CPSA accreditation standards. Each 153 ID NOW device underwent a verification process, which included testing 3 positive ID NOW 154 control swabs and 5 negative ID NOW control swabs on the ID NOW instrument before use. 155

One positive control and one negative control swab were tested on the ID NOW instrument after 156 each new box of ID NOW kits was opened, after each new HCW was trained on the instrument, 157 and, after each instrument was transported to a different site. 158 All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Table 2 and Figure 1 . In both assessment 179 centres and hospitals, ID NOW positivity rate was extremely high at 44.3% and 26.5%, 180 respectively, and is consistent with SARS-CoV-2 positivity rates observed among other COVID-181 All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. In the current Omicron study and our previous ID NOW study , sensitivity was slightly lower in 209 hospital settings compared to community assessment centres. However, its performance was still 210 higher for Omicron than what we previously observed, but not statistically significant (likely due 211 to the lower sample size in our Omicron study). 2 There may be various factors to account for the 212 decreased ID NOW sensitivity observed among hospitals compared to assessment centres. 213

Firstly, ID NOW testing is performed immediately after sample collection at assessment centres, 214

whereas hospitals require transportation to the on-site lab prior to ID NOW testing. While ID 215 NOW testing was mandated to be done within 1 hour from collection, short periods of time from (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Omicron has altered tissue tropism compared to prior variants which may make sampling from 228 the oropharynx more reliable. 12 Higher COVID-19 prevalence during the Omicron wave could 229 have also contributed. Positivity rate for all COVID-19 tests done during the study period in 230 Alberta was 36.74% compared to 4.27% and 9.09% during the wild type peaks (April 2020 and 231

December 2020), 13.56 for alpha (May 2021), and 13.47% for Delta (September 2021) [AHS 232

surveillance dashboard]. 233 234 Interestingly, a high proportion (87.0%) of the false positives detected in our study, among 235 individuals at assessment centres, were from samples collected using a nasopharyngeal swab. 236

Due to the many (45.5%) ID NOW positive samples that were subsequently not tested for 237 variants in our study, the specificity calculated in our study is likely even lower. The reasons 238 behind reduced specificity is unclear, but potentially related to altered tissue tropism favouring 239 samples from the oropharynx over the nasopharynx when tested earlier in the COVID-19 240 infection course.13 Lower specificity is unlikely to be explained by other factors, such as SARS-241

CoV-2 contamination, given that our testing was conducted across over 100 sites and there has 242 not been prior reports of contamination or high rates of false positive results with the ID NOW in 243 the literature. 244

Strengths of this study include the large sample size studied in various real-world locations, 246 including community COVID-19 assessment centres and hospitals. Due to the same testing sites, 247 procedures, and methodology, the results during Omicron can be accurately compared to our 248 results from before Omicron. 2 Due to the routine surveillance testing of many positive ID NOW 249 or RT-PCR samples for variants of concern, which included E gene RT-PCR from a consistent 250 RT-PCR testing platform (APL LDT), we were able to roughly assess the specificity of the ID 251 NOW and examine the relationship of E gene Ct values between true positive and false negative 252 ID NOW results in the face of Omicron. 253 254 Our study had several limitations. Due to the heterogeneity of our populations tested, it is 255 difficult to exclude confounders that may have contributed to ID NOW performance. However, 256

we previously observed no differences in sensitivity based on common patient, collecting and 257 testing characteristics, including age, gender, and swab type. 2 Our current study also did not 258 demonstrate any difference in performance with these characteristics, though the sample size was 259 not large enough to make any concrete conclusions. Another limitation is the inability to 260 exclusively study Omicron by itself. While the Delta variant only represented 0.7% of variants 261 detected from the assessment centres during our study period, there was still a moderately high 262 proportion (14.5%) of delta variant still circulating within our hospitals. This study did not assess 263 the Omicron BA.2 sublineage, as it was not circulating in our population during the study period. 264 265 Other limitations include missing parallel RT-PCR results that could affect the sensitivity 266 observed in our study. As previously mentioned, many (45.5%) ID NOW positive samples did 267 not undergo variant testing which would affect our calculated specificity. ID NOW sensitivity 268 may be slightly lower than we calculated due to the exclusion of 398 ID NOW negative samples 269 that subsequently did not have a second sample tested with RT-PCR. Reasons behind missing 270 parallel RT-PCR are multifactorial and include sample lost or discarded prior to testing, testing 271 sites going against guidelines and not obtaining a second swab for RT-PCR confirmation, and 272 All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission.

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 

United States Food and Drug Administration. ID NOW COVID-19 -instructions for use

(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.