key: cord-1054912-rsjrr6ci authors: Cheung, N. N.; Boers, S. A.; Kiani deh Kiani, S.; Jansen, R. W.; Mook-Kanamori, D. O.; Janssens, L.; Feltkamp, M. C. W.; Kroes, A. C. M.; Mourik, B. C. title: On-site rapid molecular testing, mobile sampling teams and eHealth to support primary care physicians during the COVID-19 pandemic date: 2020-11-23 journal: nan DOI: 10.1101/2020.11.20.20234898 sha: 24cea222822cb6f569ada877c0b18b1b149fbfbf doc_id: 1054912 cord_uid: rsjrr6ci Objectives We evaluated the effects of on-site rapid molecular testing at a drive-through sampling facility, deployment of mobile sampling teams and implementation of an online eHealth platform as supportive measures for general practitioners (GPs) during the COVID-19 pandemic. Methods An eHealth platform was developed that allowed GPs to either refer patients to a drive-through sampling facility or to request a home visit by a sampling team. Nasopharyngeal swab samples from patients marked as urgent (n=333) were tested immediately on-site using a GeneXpert System. Non-urgent samples (n=1,460) were sent once a day to a university hospital laboratory for routine testing. Time stamps starting from referral to the moment of test report sent were recorded to calculate the turnaround time. Results The eHealth platform was rapidly adopted and used by a total of 517 GPs to test 1,793 patients in a period of 13 weeks. On-site rapid molecular testing reduced the median turnaround time to 03h:41m compared to 29h:15m for routine testing. Positive SARS-CoV-2 test results were identified amongst 84/1,477 (5.7%) and 33/316 (10.4%) patients sampled at the drive-through or at home, respectively. In the age category of >80 years, 80.4% of patients were tested by a mobile sampling team. Conclusions The combination of rapid molecular testing and eHealth reduced the time between referral and results sent back to the GP to less than four hours. In addition, mobile sampling teams helped in reaching non-mobile, elderly patient populations with a higher prevalence of COVID-19. sampling facility or to request a home visit by a sampling team. Nasopharyngeal swab samples from 27 patients marked as urgent (n=333) were tested immediately on-site using a GeneXpert System. Non-28 urgent samples (n=1,460) were sent once a day to a university hospital laboratory for routine 29 testing. Time stamps starting from referral to the moment of test report sent were recorded to 30 calculate the turnaround time. 31 32 Results 33 The eHealth platform was rapidly adopted and used by a total of 517 GPs to test 1,793 patients in a 34 period of 13 weeks. On-site rapid molecular testing reduced the median turnaround time to 35 03h:41m compared to 29h:15m for routine testing. Positive SARS-CoV-2 test results were identified 36 amongst 84/1,477 (5.7%) and 33/316 (10.4%) patients sampled at the drive-through or at home, 37 respectively. In the age category of >80 years, 80.4% of patients were tested by a mobile sampling 38 team. 39 40 The combination of rapid molecular testing and eHealth reduced the time between referral and 42 results sent back to the GP to less than four hours. In addition, mobile sampling teams helped in 43 reaching non-mobile, elderly patient populations with a higher prevalence of . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 23, 2020. Major global efforts have been undertaken since the emergence of novel coronavirus disease 2019 46 (COVID-19) to limit the spread of infection. One of the cornerstones of successful outbreak 47 management is prompt diagnosis of patients suspected of SARS-CoV-2 infection (1) . In order to 48 achieve this, mass testing campaigns have been introduced to lower testing thresholds. Drive-49 through sampling facilities are often an integral part of such campaigns, because they provide an 50 accessible, safe and efficient method for large-scale sample collection (2). An important 51 disadvantage of such facilities however are delayed turnaround times due to transportation to 52 nearby laboratories before the diagnostic test can be performed. In addition, drive-through facilities 53 are often less accessible for non-mobile and/or elderly populations. In the Netherlands, these 54 patients rely mostly on home visits by their general practitioner (GP) for sample collection. This can 55 be time-consuming and potentially hazardous for GPs due to the already increased working hours 56 during the pandemic combined with insufficient availability of personal protective equipment during 57 the early phases of the pandemic (3-5). On top of these limitations, test accessibility for GPs was 58 restricted in the Netherlanders during the early phases of the first COVID-19 wave. 59 60 To support the primary care in our region during the pandemic, we developed a system aimed at 61 reducing diagnostic turnaround times and improved test accessibility. To this aim, we implemented 62 an eHealth platform through which GPs could request a home visit by a mobile sampling team or 63 refer patients to a drive-through sampling facility. Prior to visiting, patients could fill in their 64 symptoms and risk profile online. During their visit at the facility, the patient's vital functions were 65 measured and registered in the eHealth system. After their visit, these data were sent back to the GP 66 together with the test results to aid them in their further evaluation and care of the patient. The 67 sampling facility was equipped with a mobile laboratory in which we could perform rapid molecular 68 testing to reduce the turnaround time of patient samples marked as urgent by the GP. Results were 69 . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 23, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 23, 2020. ; https://doi.org/10.1101/2020.11.20.20234898 doi: medRxiv preprint Online referral of patients and eHealth consultation 79 The Heroku (San Francisco, United States) cloud application platform and database were used to 80 develop an eHealth platform. Data was collected using questionnaires constructed with Typeform 81 (Barcelona, Spain). Site and account access were regulated with Auth0 (Bellevue, United States). Our drive-through sampling and testing facility was operational between April 21 th and June 12 th 96 2020 at a location nearby the Hague, the Netherlands. During this period, GP's could use our eHealth 97 platform to refer patients to our facility or to request a home visit by a mobile sampling team (in 98 collaboration with the regional Red Cross). Nasopharyngeal swab samples were transported directly 99 to our facility after collection. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 23, 2020. ; https://doi.org/10.1101/2020.11.20.20234898 doi: medRxiv preprint network was available on-site. When either patients or samples arrived for registration, we entered 104 their personal information in a specifically designed dashboard in R-software (Auckland, New 105 Zealand). This dashboard communicated directly with GLIMS (Ghent, Belgium), the Laboratory 106 Information System (LIS) of our Medical Microbiology department. After completion of the 107 registration form in R, sample ID stickers from GLIMS were directly printed and placed on the test-108 tube prior to sampling. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 23, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 23, 2020. The median time from referral to result sent of patients tested either immediately on-site or by 161 routine PCR at our clinical microbiology department is shown in Fig. 1 . For urgent cases, the time 162 period between registration and test initiation was reduced 43-fold and the time period between 163 test initiation and report of result was reduced 6-fold compared to non-urgent cases. For patients 164 sampled at the drive-through, this resulted in a total turnaround time of 03h:41m (95% CI: 03h:23m 165 -04h:06m) for urgent patients and 29h:15m (95% CI: 29h:12m -29h:31m) for non-urgent patients. 166 For patients sampled at home, the turnaround times were 07h:03m (95% CI: 05h:46m -07h:36m) for 167 urgent samples and 29h:10m (95% CI: 28h:28m -30h:04m) for non-urgent samples. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 23, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 23, 2020. ; https://doi.org/10.1101/2020.11.20.20234898 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 23, 2020. In addition to our primary objectives, we also evaluated the use of antigen testing. This was 207 performed on-site for n=350 samples (Fig. 4) . Comparison of these results with the CT value of the 208 routine PCR results of the exact same samples revealed that the antigen test was only able to 209 identify positive SARS-CoV-2 samples with CT value <23,4. This translated to a sensitivity of 37,1% 210 and a negative predictive value of 93,5%. 211 212 . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 23, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 23, 2020. ; https://doi.org/10.1101/2020.11.20.20234898 doi: medRxiv preprint Rapid and efficient diagnostics are key-factors in coping with the current COVID-19 pandemic and 223 should not be restricted to solely the technical qualities of a diagnostic test. Instead, efficiency 224 should apply for the entire diagnostic process starting from sample collection to delivery of the test 225 result to where it matters most, the patient and the treating physician. Our eHealth platform 226 provided GPs with an integrated healthcare system allowing for rapid patient referrals as well as 227 efficient collection and communication of patient's complaints, vitals and test results. In addition, we 228 show that on-site rapid molecular testing at a sampling facility can reduce time from referral to 229 result sent by at least twenty-four hours compared to samples from the same facility that were 230 tested by routine PCR in a nearby laboratory. 231 232 Mobile sampling teams complemented our facility by reaching the non-mobile and/or elderly 233 population. These patients resided primarily in elderly homes that are known to be associated with 234 high SARS-CoV-2 transmission-, and mortality rates (10). The importance of rapid testing in this 235 particular population was reflected in the 3-fold higher percentage of urgent requests and a 2-fold 236 higher percentage of positive tests compared to patients sampled at the facility. A potential 237 selection bias where the referring GP adheres more strictly to referral guidelines before requesting a 238 home visit cannot be excluded. Nonetheless, current mass testing campaigns in the Netherlands 239 mainly consist of drive-throughs that are nearly inaccessible to this particular population and can 240 therefore benefit from the addition of mobile sampling teams. 241 242 One of the advantages of performing a diagnostic test in a readily established laboratory is 243 assurance of high quality standards and minimal risk of exposure for laboratory workers. We 244 demonstrate that on-site diagnostics can be successfully implemented within the same quality 245 framework as a conventional microbiology laboratory. 246 247 . CC-BY-NC-ND 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 23, 2020. ; https://doi.org/10.1101/2020.11.20.20234898 doi: medRxiv preprint Next to on-site molecular testing, antigen tests are potential candidates for widespread rapid SARS-248 CoV-2 diagnostics. Our evaluation of one of such tests showed a limited sensitivity, which was in line 249 with a previous evaluation (8). However, the test was simple to perform and did produce consistent 250 results with high specificity. Also newer generations of antigen tests are entering the market with 251 improved sensitivity. Therefore, the antigen test might prove valuable in the early diagnosis of 252 patients with high viral loads, in low-resource settings where no molecular diagnostics are available 253 or in surveillance settings where frequent and fast screening is desired (11, 12) . is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted November 23, 2020. ; https://doi.org/10.1101/2020.11.20.20234898 doi: medRxiv preprint Impact of delays on effectiveness of contact tracing strategies for COVID-19: a modelling 273 study Drive-through 275 testing: a unique, efficient method of collecting large volume of specimens during the SARS-CoV-276 2 (COVID-19) pandemic Physician deaths from corona virus (COVID-19) disease On the death of 100 + Italian doctors from COVID-19 Critical supply shortages -the need for ventilators and 282 personal protective equipment during the Covid-19 pandemic Multi-center evaluation of cepheid xpert® xpress SARS-CoV-2 point-of-care test during 286 the SARS-CoV-2 pandemic Brief validation of the novel GeneXpert 289 Xpress SARS-CoV-2 PCR assay Development and potential usefulness of the COVID-19 Ag Respi-Strip diagnostic 294 assay in a pandemic Context It is made available under a perpetuity.is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprintThe copyright holder for this this version posted November 23, 2020. ; https://doi.org/10.1101/2020.11.20.20234898 doi: medRxiv preprint is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprintThe copyright holder for this this version posted November 23, 2020. ; https://doi.org/10.1101/2020.11.20.20234898 doi: medRxiv preprint