key: cord-0953431-sx3rz2hd authors: Grys, T. E.; McAulay, K.; Ingram, D.; Duffy, C.; Williams, A.; Arumugam, S.; Ma, J.; Macar, U.; Han, G.; Chang, S.-F.; Sia, S. K.; Mayer, K. title: SafeSwab, a sample collection and dispensing device for near-patient testing date: 2021-08-22 journal: nan DOI: 10.1101/2021.08.20.21262386 sha: 088003eef397ca9bdf9d3527034d003ff33e6a2f doc_id: 953431 cord_uid: sx3rz2hd The COVID-19 pandemic has accelerated the pace of innovation around virtual care visits and testing technology. Here we present the SafeSwab (Safe Health Systems, Los Angeles, CA), an integrated, universal sample collection and dispensing device that is designed to minimize user error and enable rapid testing in a point of care or self-testing format. The SafeSwab was used with the Safe Health Systems HealthCheck digital health application to enable self-testing by patients using lateral flow tests for SARS-CoV-2 antigen or for antibodies against SARS-CoV-2. Patients (n=74) using the SafeSwab produced a valid rapid test result in 96% of attempts, and 96% of patients felt confident that they had collected a good sample. The Safe HealthCheck app has an integrated image analysis algorithm, AutoAdapt LFA, that interprets a picture of a rapid test result, and the algorithm interpreted the result correctly 100% of the time. The SafeSwab was found to be versatile and easy to use for both self-collected nasal sampling as well as fingerstick blood sampling. The use of Safe Health Systems HealthCheck app allows an integrated solution for patient instruction and test interpretation The SARS-CoV-2 pandemic has spurred an intense period of research and innovation like none ever seen before. The U.S. Food and Drug Administration (FDA) has recently authorized antigen tests for use in non-laboratory settings, including the home (1). . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted August 22, 2021. ; https://doi.org/10.1101/2021.08.20.21262386 doi: medRxiv preprint Although these tests are relatively simple to perform for those who have laboratory experience or medical training, they are dependent on adequate specimen collection and adequate performance of several time and volume-dependent steps. The motivation to approve home-based testing was the urgency of the pandemic. It is likely, however, that tests for other infectious diseases and chronic diseases will be cleared through the 510(k) and premarket approval (PMA) process for use in non-laboratory settings. The most successful among them will be those that optimize the testing process in a way that increases the robustness of the test and minimizes the chance of user error. We introduce here the SafeSwab, an integrated sample collection and dispensing device that is broadly applicable to near-patient testing, whether point-of-care or home use. The device is designed to minimize errors of collection, handling, and measuring. The SafeSwab was used in combination with a digital health solution called Safe HealthCheck (Safe Health Systems), to allow patients to test themselves. Safe HealthCheck is a smart phone application that integrates symptom checking, virtual consults, animated instructional videos, and an automated machine learning algorithm (AutoAdapt LFA) for automated test interpretation (Arumugam et al., under review). Here we outline a usability study in the context of COVID-19, where our primary outcome was participant satisfaction with the devise and test validity. Specific assay performance is not addressed here. . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 22, 2021. ; https://doi.org/10.1101/2021.08.20.21262386 doi: medRxiv preprint SafeSwab. The SafeSwab ( Figure 1 ) is a collection device that allows for integrated sample collection and dispensing. After using a standard lancet, the absorbent tip of the SafeSwab can be used to collect a fingerstick blood sample. Or, the swab tip can be extended to reveal approximately 1 cm of swab surface to collect a nasal sample. Distal to the tip is a reservoir, which can be filled with any buffer to suit the test being performed. Twisting the reservoir releases the buffer to flow down the barrel, carrying the sample out of the absorbent tip. When held over a rapid testing device, the sample can be placed directly into the sample inlet. Approach. To evaluate the usability of the SafeSwab, we recruited individuals undergoing standard of care (SOC) SARS-CoV-2 testing [nasopharyngeal (NP) swab tested using one of two Emergency Use Authorized (EUA) PCR methods, RealTime SARS-CoV-2 assay on the m2000 (Abbott Laboratories) system or Alinity m SARS- Antigen testing population. Participants in the antigen testing arm were recruited via the drive-through COVID-19 testing site at an academic hospital campus. After providing informed consent (IRB protocol 20-010688), participants were asked to remain in their vehicle (to simulate a home environment) and were provided with a small tray containing a Flowflex™ SARS-CoV-2 Antigen Rapid Test cartridge (ACON Laboratories), a SafeSwab pre-filled with Flowflex™ SARS-CoV-2 Antigen Rapid Test . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Self-testing. Participants in both studies used the Safe HealthCheck phone application to complete the testing process. First the app prompts the user to scan a QR code on the test cartridge pouch, initiating the cognate animated instructional video (motion504, Minneapolis, MN). The video shows the user the entire testing process. After the video, still images from the video instructions with accompanying text take the user through the testing process step by step on subsequent screens ( Figure 2) . The app includes a built-in timer and subsequent prompt for the user to take a picture of the cartridge (also via the app). The test image is sent to an Amazon Web Server where it is processed by the AutoAdapt LFA image analysis algorithm, which uses machine learning to detect the test strip and any bands present, then uses a lookup table containing information on positions of test vs. control lines to interpret the result (2). . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted August 22, 2021. ; https://doi.org/10.1101/2021.08.20.21262386 doi: medRxiv preprint During the study, the result was not returned to the user, but was stored in a deidentified database for documentation purposes. Survey. Each participant was asked to complete a usability study at the end of their experience. During the survey, participants were asked to provide their age and maximum educational level and to answer 5-7 usability questions anonymously. Analysis. The survey results were collated to assess SafeSwab Usability. To assess SafeSwab performance, as opposed to specific assay performance, the primary . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted August 22, 2021. ; https://doi.org/10.1101/2021.08.20.21262386 doi: medRxiv preprint All 32 participants in the antibody testing arm produced a valid result (control line present) and of these, 32 were positive for IgG while 3 were negative. The AutoAdapt LFA algorithm correctly interpreted all 32 results compared to a human observer, including the three IgG-negative samples. Patients participated 6-10 weeks after their onset of symptoms. For this reason, despite the inclusion of separate IgM and IgG zones on the ACON SARS-CoV-2 IgG/IgM Rapid Test cartridge, we only evaluated IgG results, as IgM is known to be sporadically detected more than 4 weeks following infection (8) . Venous blood was not collected to compare the results to an EUA approved antibody assay to minimize participant discomfort and as the primary focus was usability rather than specific assay performance. The usability survey was completed by 21 (53%) antigen testing participants and 25 (78%) antibody testing participants (Table 2) . Although some (24-28%) of respondants felt that they would like help collecting the sample, 100% felt that the instructions for transfering the sample from the SafeSwab into the cartridge were clear, and ultimately, 96% of respondants felt that they had collected a good sample, including all of the antibody testing. Surprisingly, those who thought that they would like help collecting the sample were mostly (5 of 7 for antibody and 3 of 5 for antigen) from the group who self-identified as having medical or laboratory training. This may reflect a self-awareness about the importance of sample collection. Here we introduce SafeSwab, a new testing device that enables sampling of an anterior nasal sample or a fingerstick blood sample, and with an integrated buffer chamber, enables easy delivery of the sample and buffer into a testing device. The SafeSwab is ideal for near-patient testing such as with a point-of-care device. We . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted August 22, 2021. ; https://doi.org/10.1101/2021.08.20.21262386 doi: medRxiv preprint implemented the SafeSwab in conjunction with the Safe Health Systems HealthCheck digital health application to determine if patients could perform an antigen test after taking their own anterior nasal sample, or a serology test after collecting their own fingerstick blood. In a typical antigen test, the swab is placed into a buffer tube, swirled for a period of time, then removed, and a dropper cap placed on top. Then drops are placed into the test cartridge. With the SafeSwab, these multiple steps are integrated into one device. In an antibody test, the blood is typically transferred to the sample well of the cartridge with a miniature plastic pipette or capillary tube, both of which require skill and dexterity. Then, buffer must be measured into the sample well. We see the primary advantages of the SafeSwab to be reduced chance of user error, such that it is easy enough for a lay user. Despite all patients having documented PCR results, three antibody tests were negative. A previous study showed this antibody test to be highly accurate in a set of samples drawn in the first 1-4 weeks after diagnosis (9) . It is possible that antibody levels in these three patients had decreased, or that they had a low response; variability in anti-SARS-CoV-2 IgG response is well documented (10) . Combining valid results of both antigen and antibody tests, 63 of 69 (91%) patients generated the correct result compared to the EUA approved PCR reference standard. From an operational standpoint, 69 of 72 (96%) patients generated a valid result. In all cases (n=72), the AutoAdapt LFA algorithm interpreted the test correctly when compared to the human interpretation. Design and performance characteristics of this algorithm are outlined in a recent preprint (2) . In this study, the Safe Health System HealthCheck application was used to guide the participant through the testing steps and transfer the images to the AudoAdapt LFA . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 22, 2021. ; https://doi.org/10.1101/2021.08.20.21262386 doi: medRxiv preprint algorithm. This digital health platform has the capacity to integrate features such as provision of reliable medical information, symptom checking algorithms, connection to healthcare providers through virtual consults, self-testing instructions, automated test interpretation (using the AutoAdapt LFA program), integration of test results into a patient record (if the patient chooses to link accounts), and public health reporting for reportable diseases. These features, when paired with the SafeSwab, could enable and accelerate new models of patient care, including increasing access to testing and decreasing cost of healthcare. . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted August 22, 2021. ; https://doi.org/10.1101/2021.08.20.21262386 doi: medRxiv preprint Food and Drug Administration Adaptable Automated Interpretation of Rapid Diagnostic Tests Using Few-Shot Learning Verification and Validation of SARS-CoV-2 Assay Performance on the Abbott m2000 and Alinity m Systems Evaluation of Abbott BinaxNOW Rapid Antigen Test for SARS-CoV-2 Infection at Two Community-Based Testing Sites Diagnostic accuracy and feasibility of patient self-testing with a SARS-CoV-2 antigen-detecting rapid test Head-to-head performance comparison of selfcollected nasal versus professional-collected nasopharyngeal swab for a WHO-listed SARS-CoV-2 antigen-detecting rapid diagnostic test Diagnostic performance and characteristics of anterior nasal collection for the SARS-CoV-2 antigen test: a prospective study Longitudinal Profiling of Antibody Response in Patients With COVID-19 in a Tertiary Care Hospital in Beijing Retrospective clinical evaluation of 4 lateral flow assays for the detection of SARS-CoV-2 IgG Antibody Responses in COVID-19: A We would like to thank the hard work of the research coordinators and study nurses who supported the study. We also acknowledge ACON Laboratories (San Diego, CA) for supplying kits for evaluation. It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted August 22, 2021. ; https://doi.org/10.1101/2021.08.20.21262386 doi: medRxiv preprint . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Twisting the buffer reservoir (C), distal to the swab tip, breaks the seal, releasing buffer down the shaft and through the swab tip, carrying the sample into the testing kit.. CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)The copyright holder for this preprint this version posted August 22, 2021. ; https://doi.org/10.1101/2021.08.20.21262386 doi: medRxiv preprint