SARS-CoV-2 PCR and antibody testing for an entire rural community: methods and feasibility of high-throughput testing procedures

and feasibility of high-throughput testing procedures High-vol ume, com mu nity-wide ascert ain ment of SARS-CoV-2 prevalence by PCR a nd an tib ody testi ng was successfully perfor med usi ng a c om muni ty-led , drive-t hrou gh m odel wit h str ong o pera tio nal suppor t, well-trai ned testin g u nits, a nd a n effective t ech nic al pla tform . Ayesha Appa , MD 1 ; Gabri el Chamie, MD, MPH 1 ; Aenor Sawyer , MD 2 ; Kimberly Balt zell, RN , PhD, MS 3 ; Kathryn Dippell, MS N, NP 3 ; Salu Ribei ro, BS, MS 4 ; Elias Duart e, BA 1 ; Joann a Vinde n, MPH 1 ; CLIAHUB Consortium 5,6 ; Jona than Krame r-Feldma n, BA 4 ; Shahrya r Rahda ri, BS 4 ; Doug MacI ntosh, RN 7 ; Kathe rine Nicholson, R N 7 ; Jona than Im, BA 8 ; Diane Havlir, MD 1 ; Bryan Gr eenho use, MD 1,5 1 University Of California, San Fr ancisco Division of HIV, Infectious Diseases, and Gl obal Medicin e; 2 University of California San Francisco De partmen t of Or thope dic Surgery; 3 Univer sity of California San Francisco School of Nursing; 4 Universi ty of California San Francisco School of Med icine; 5 Chan Zuckerberg Biohub ; 6 University of California San Francisco Depa rtmen t of Labora t ory Medicine; 7 San Francisco Sta te Universi ty School of Nurs ing; 8 University of California San Francisc o School of Pharmacy . CC-BY 4.0 International licenseIt 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 (which was not certified by peer review) The copyright holder for this preprint this version posted May 30, 2020. ; https://doi.org/10.1101/2020.05.29.20116426doi: medRxiv preprint NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice. . CC-BY 4.0 International licenseIt 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 (which was not certified by peer review) The copyright holder for this preprint this version posted May 30, 2020. ; https://doi.org/10.1101/2020.05.29.20116426doi: medRxiv preprint . CC-BY 4.0 International licenseIt 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 (which was not certified by peer review) The copyright holder for this preprint this version posted May 30, 2020. ; https://doi.org/10.1101/2020.05.29.20116426doi: medRxiv preprint 2

Since the b eginning of the COVID-19 pandemic, impaired access t o tes ting across the Uni ted Stat es has limited ou r unde rstandi ng of epidemiology and thus limit ed disease co ntrol . With cle ar evidence of asymptomatic infecti on 1 but minimal systematic active surveilla nce ac ross larger communities, addi tional efforts to condu ct large-scale t esting wer e need ed to un dersta nd the b read th of COVID-19 disease. While th ere h ave been o the r efforts t o p rovide drive-thr ough tes ting (mostly using PCR for symptomatic or e xposed individuals) 2-5 , no standar d procedures e xiste d to safely a nd efficiently condu ct “pop-up” testing using PCR and antibodi es for an enti re community. H ere, we d e scribe the p rocedu res and method ology associated wit h safe, h igh volume comprehensive t esting for SA RS-CoV-2, the first effort to p erform community-wide, universal PCR and antibody t esting to our kn owledge. By obt aining rapid, compr ehensive informa tion abo ut active and past infec tion, we offer this as one model to augment diseas e surveillanc e for rural o r suburban popula tions . Body: PRIOR TO TESTING Community Mobilization (Patient and Public Involvement) Support from key stake holde rs in the co mmunity was crucial to this project ’s success; in Bolinas, this project was initi ate d by and co-led by community members, who served as le aders th roughou t the planning and ope rati onal proc ess. O ther key stakeholde rs included th e major com munity-based heal th organiza tion, t he Depar tmen t of Public Health , and th e Fire Depa rtmen t. M ost of these grou ps, toge the r with study leade rship, par ticipa ted in a vi rtual Town Hall th e week prio r to study s tar t to int roduce the study to th e community and answer qu e stions. Addi tion ally, specific community liaisons engaged pe ople expe riencing homel essness, th e Latin x community, and home-bound elders to maximize pa rticipa tion . In summary, while each community has distinct ne eds, we found tha t an e arly need s assessment with regard t o community mobilizati on to id e ntify essential community pa rtne rs was an most importan t ea rly step. Registration & Pre-test Survey Town residents and l ocal first resp onders were invited t o regist er onlin e, using a custom interface cr ea ted on a H IPAA-compliant platform in par tne rship with study lead e rship and community liaisons. Resid ents wer e direc ted t o begi n the proc ess by providing contact infor mation for 2-factor authen tica tion (eith er phon e or email) to ensure securi ty and confirm ability to re turn resu lts. If they were not a ble to us e the online in terface , they could call a local facility, wher e first-respond er voluntee rs helped p eopl e regist er onlin e. Participant inclusio n in the st udy was confirmed by providing their zip cod e or in dicating thei r status as first r espond ers. Par ticipants co mpleted an online conse nt and survey, w hich included questions ab out t he hous ehold as well as demographics, cont act informa tion, t rav el and movement . CC-BY 4.0 International licenseIt 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 May 30, 2020. ; https://doi.org/10.1101/2020.05.29.20116426doi: medRxiv preprint 3 information, symptoms, a nd medical hist ory. Each household was schedul ed for 1 5-minute appointme nts allowing no mor e than five persons pe r car, an d they rec eived a co nfirmation of thei r appointme nt tim e by their d esired mod e of contact (SMS, email). The onlin e and t elephonic e xpe rience including website, consen t, and su rvey was available in bot h English and Spanish. On the d ay of testing, participan ts were emailed o r te xt ed with a brief summary of what to exp ect durin g their t esting expe rience . Sample Data Management Robust sample id entificati on was a key aspect to e nsuring successful data manag ement, and an importan t challenge to add ress in th e community-based, “pop-up” contex t. In accordance with Clinical Laborat ory Improvemen t Amen dments ( CLIA) regulations specified by th e clinical labora tory, ou r labels containe d two iden tifiers, nam e and dat e of birth (see Bo x 1 for exampl e below). Labels additio nally containe d a random le tt er code in hum a n readabl e and QR cod e format, to serve as a scannable identifier linking e ach specimen to a uniq ue particip ant r ecord in t he onlin e dat ab ase. Our site was no t equipp ed for on-deman d label prin ting in each lane , so all pre-r e gistered participan ts had lab els pre-pri nted the m orning prior to tes ting. Each lan e contain ed a packet of alphabe tized , pre-prin ted wa ter-r esistan t cryo labels. Each pa rticipan t had 4 iden tical labels pe r shee t: two to be use d on the two specimen con taine rs, one on a l ab requisi tion she et, a nd one spar e label . If the par ticipan t regist er ed onsit e, th e ad ministrat or ei ther : 1) used an onsite l abe l printe r available in some lanes or 2) used a set of lab els with a unique barcod e but o the rwise blank, a nd handwrot e the participan t’s name and dat e of birth on t he labels and r equisiti on form. Box 1: Sample labels A. Preregistered p arti cip ant B. Empt y la bel for parti cip an t registeri ng o nsite DURING TEST DAYS STAFFING & FLOW In orde r to comple te t esting of more tha n 1,800 individuals over four days, two t ents were s et up in a large lot with a l ane on ei the r sid e of each ten t, t o crea te four t otal la nes for tes ting (Figure 1). Participants ’ first int eracti ons were with primarily community voluntee rs outsid e the t esting ar ea, who then dir ecte d particip ants t oward medic al staff and voluntee rs in each lan e for te sting. . CC-BY 4.0 International licenseIt 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 May 30, 2020. ; https://doi.org/10.1101/2020.05.29.20116426doi: medRxiv preprint 4 Site Entry & Lane Triag e To help facilitat e ent ry into th e site and prevent in terf erenc e with regula r traffic flow in the area, traffic controll ers wer e sta tioned at the int ersect ion of the main r oad. Once cars arrived in th e designate d are a, par ticipan ts were me t by a “greeter” who passed out su rgical ma sks to all participa nts (Supplemental Tabl e 1). If Spanish-English transla tion wer e requi red, a volunte er transla tor was engage d at this poin t (site en try) to help navigat e the t esting sit e exp erie nce. Local community members voluntee red for th ese t hre e roles , as hea lthcare exp erienc e was not re quired bu t local knowledge and a welcoming presence was very helpful. Nex t, a “tri age gree te r” with a table t con firmed participa nts' p re-regis tra tion and appointme nt time and scre ened for symptoms of COVID-19, including fever, cough, shortness o f breath, fa tigue, myalgias, anosmia, and dysgeusia. If a pa rticipan t (or anyone in th e vehicle) were symptomatic, th ey were direc ted to th e specified symptoma tic lane. If asymptomatic, th e tri age gre e ter dir ect ed participan ts to the shor tes t lane . Becaus e of the symptom assessment , tri age gre ete rs were volunt eers in the hea lthcar e field who were comfor t able with ta blet us e. Using th e online pl a tform on the table t, the t riage gre et er indicat ed which lane t he particip ant moved in to, allowing each lane’s administ rat or (sitting in the t ent) to view th e queue of participan ts in th eir lan e and pre pare for their ar rival. Testing B ay & Tent Staffing/Fl ow: There was one t esting bay pe r lane, each staffed by 4 people: two dyads of 1) professional phlebot omist who perform ed tes ting (”tester”) and 2) a healthc are volun tee r (primarily nursing, medical, and pha rmacy studen ts) who served as the “tes t assistan t” (Figure 2). Inside the ten t, an addition al volunte er provid ed administr a tive support to each l ane. Fin ally, each t e nt had two t ent supervisors, each of whom was a gradua te-level t rained n urse, physician, o r tr ain ed volunte er. As a participan t app roached the testing bay i n any given lane, the test assis tant confir med the pa rticipan t’s name, dat e of birth , and wheth er t hey h ad symptoms that day. Witho ut en teri ng the te nt th emselves, the t est assist ant in th e bay relay ed this i nformation back to the adminis tra tor in t he ten t, who logged symptoms and prepar ed a t est kit for e ach participan t. To facilitate throughpu t, t est kits cont ain ing all supplies necessary to compl et e te sting in the bay (alcohol wipe, lancet , microtai ner, gauze, tongu e dep ressor , swab, viral tra nspor t media, bioha zard bag) were assembled in advance. The a dministra tor’s r ole was primarily to loc ate the a ppropria te pa rticipa nt labels and afix lab els to 1) the micro tain e r, 2) the viral tr anspor t media tube, and 3) requ isition she et. Labeling and tes t kit prep ara tion was ide ally performed in advance of th e par ticip ant re aching the testing bay, facilit ate d by the administ rat or’s ability t o view their lan e’s queu e in t he online pla tform. In the testing bay, a ca r pulled in to th e bay and turne d off the engine . If particip a nts arrived on foo t or other vehicl e, to accommodate those wit hout access to a ca r, th ey were sea ted in a chair in the middl e of the lane . The tes te r expl ained the pro cedure, a nd complet ed finger stick th en oropha ryngeal/mid- turbina te swab (see Test Proce dures for more de tail). The t est assista nt main tain ed distance from t he participan t during specime n collectio n, b ut was on hand to pass i tems to t he t est ers. Ex tra t est assistants wer e tr ained , with additi onal t est assistan ts helping as runn ers/quali ty control le ads when not working actively in the tes ting bays. Onc e a particip ant ha d complet ed tes ting, th e test assistan t verbally repor ted compl etion to th e in-ten t admi nistra tor, an d the a dministra tor n ote d whethe r tes ts were . CC-BY 4.0 International licenseIt 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 May 30, 2020. ; https://doi.org/10.1101/2020.05.29.20116426doi: medRxiv preprint 5 successfully administered and that the la bel barcod e matched the da tabas e barco de. The pa rticipan t exit ed th e lane an d tes ting site. Each tent was also st affed with two ten t supervisors, whose rol e was to t rouble-s hoot all activities in th e tes ting bay and te nt, incl uding responding to p articip ant qu estion s, and ensuring opera tions ran efficiently. Se e Suppleme ntal Table 1 for summary of staffing requ ired per d ay. Finally, on-site staff were scre ened with an emai l-based questi onnair e befor e each day to ensure th ey did not have symptoms associated with COVID-1 9 (Appendix 1). TEST PROCEDURES Our t esting str ategy employed bo th bloo d collection for an tibody tes ting and upp er respi rat ory tract sampling for PCR testing . With r ega rd to collec tion of blood, our goal was to maximize commu nity participa tion by lowering ba rrie rs to sam pling through use of a finger prick techn i que (vs. phlebotomy), while collecting enough blo od to b e sufficient to r un quan tita tive, labo rat ory-bas ed tes ts 6 . Please see Appendi x 2 for detai led proc edur es utiliz ed for sample collecti on. PPE Requirements We constr ucted p ersonal p rot ective equ i pment (PPE) requirements using the foll owing framework adapt ed from Worl d Heal th Organiza tion guiding principl es: 1) consider the type of contac t with participa nts, 2) incorpor at e tra nsmission dynamics and environment al factor s pertin ent t o the testing sit e, and 3) utiliz e stewardshi p an d appropri ate PPE re-use when possible 7 . For each rol e, we define d the d egree of contact with p articip ants and whe the r po ssible to adher e to physical distancing while p erfo rming the rol e. For exampl e, th e tes ter r ole involved physical contact with pa rticipa nts’ han ds and pro ximity to unmasked pa rticipa nts during o ropharyngea l and mid- turbina te specim en collectio n tha t may lead to sne eze o r cough. As follows, PPE requirem ents for this role wer e the mos t stringen t, including c overalls (or gown), gloves, respirat or, an d face shield. Conversely, the t est assis tant did n ot hav e physical contact with pa rticipa nts nor were th ey in close Box 2: Test Procedure Pearls Finger Prick: • Ensure par ticipants have wa rm hands pri or to finger prick (eg. use of car he ate rs, hand warmers, hair d ryers). • Use sufficient gauge lancet : we used a 17 -gauge push button lanc et bla de with 2m m depth. • Prick 3rd or 4th digit of non-dominant ha nd at lat eral asp ect of fingertip , the n immediately supinate h and well below level of hea rt t o collect specimen . Respirat ory Sam plin g: • Instruc t masked par ticipants to move the ir mask down to their chin for o ropha ryngeal collection, then r eplace b ack over mouth (but not nose) for mid-turbina te coll ecti on. . CC-BY 4.0 International licenseIt 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 May 30, 2020. ; https://doi.org/10.1101/2020.05.29.20116426doi: medRxiv preprint 6 proximity duri ng specimen collecti on but did talk with particip ants pri or to testi ng and were also handling specimens afte r collecti on. As s uch, the r ecommended PPE for this role was a surgical mask, face shield, and gloves. Pleas e see Suppl emental Tabl e 1 for det ailed PPE recommendations for al l roles . When conside ring gener alizabili ty to oth er tes ting appro aches, t he most impor ta nt consider ation should be given to the movem ents and pa rticipa nt inte ractio ns involved in each role wit h PPE recommenda tions base d on associat ed e xposure risk. Finally, with regard to PPE reuse, we mo deled our guid elines aft er our me dical ce nter an d CDC guidelines to minimize was te of mate rial s (Supplemental Table 2) 8 9 . In brief, gow ns and gloves were never reus ed, bu t face shields and masks (either surgical or respir ators) wer e safe ly removed, cleane d and stor ed for reus e thro ughout the day. Post-Testing At end of each testi ng day, blood sample s in microtainers wer e sto red uprigh t in small cardboar d specimen box es, and viral tra nsport medi a in biohaza rd bags was ideal ly stored uprigh t as well. Specimens wer e tra nspor ted t o the lab each evening for accessioning ove rni ght. Participa nts were counseled t o exp ect PCR results within 3 -7 days and antibody results withi n 4-6 weeks. Study st aff planned t o call each par ticipan t with a p ositive PCR result and dir ect th eir r esults to the Dep artm ent of Public Health. Ad dition ally, both posi tive and negative r esults wer e deliver ed via t he same online platform th rough which participa nts regi stere d. Particip ants r eceived an SMS o r e mail with a code tha t allowed them to login to view th eir resul ts. Alt ern atively, th ey had the o ption of c alling a hotline for addition al suppor t. Testing Site Throughput In tot al, 1,840 par ticipan ts were t est ed o ver 4 days using this 4-lane drive-throug h or walk-up model. Seven p articip ants r eceived hom e-based t esting on a supplem ent al 5th a nd final day of testing, to to tal 1,847 par ticipan ts overall . Fewer participan ts were sched uled on t he first day of testing to allow for study staff and voluntee r acclimation to thei r rol es. On Day 1, the fewes t num ber of particip ants were tes ted (n=338 pa rticipan ts), increas ing to the highes t number test ed on Day 4 (n=571 participan ts). Notab ly, performanc e on Days 2-4 reflects staffing described above , where as the re were fewe r personn el available to staff the testi ng bays on Day 1. Figure 3 depicts th e numbe r of participa nts tes ted pe r hour acr oss all lanes for e a ch of the four days of onsite tes ting. Wh en including th e hours during which time th e t esting sit e was fully open for appointme nts (9am - 5pm), the median number of particip ants t est ed per ho ur o nsite was 57 (interqua rtil e range 47-67). The par ticipa nts included in this an alysis were 1,801 p articipan ts with time- stamped checkou ts (compared to the t ot al of 1,847), slightly underestima ting act ual throughpu t. Identified Areas for Improvement . CC-BY 4.0 International licenseIt 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 May 30, 2020. ; https://doi.org/10.1101/2020.05.29.20116426doi: medRxiv preprint 7 While this was overal l a successful endea vor, we faced a number of challenges tha t othe rs may improve upon in th e future . Our l abeling system largely worked well when par tici pants were p re- registe red, but when par ticipan ts regist e red onsit e or ide ntified an erro r in thei r name or da te of birth , handwriting lab els was both time-consu ming and error p rone . Possible improve ments include using an onsite lab el prin ter al ong with a barcod e scanner in each lan e to a utoma te t his pr ocess as much as possible. Anoth er challe nge was verification of par ticipant id entifie rs when all par ties wer e wearing masks and maintaining physical distancin g. While t ent supe rvisors emphasiz ed on going closed-loop communication be tween t eam and pa rti cipants, al tern ative st rat egies to verb al communication may offer an improvement . For exampl e, tes t assistants could use a small white boa rd to write p articip ant identifiers a nd visually confirm these de t ails with the pa rticipan t and t he administ rato r. Finally, despit e quality con trol measu res, a small number of viral transpo rt medi a tubes leak ed materi al upon rec eipt in t he lab . As such, we recommend using tightly sealing vial s, inspecting vials prior to tes ting and uprigh t stor age of samples in individual biohazard b ags. An additio nal suggestion would b e to t es t the e ntir e process in advanc e of testing roll out, from onsite r egistra tion t o sample collec tion t o lab repo rting. Given th e speed with wh ich our efforts were planned, we wer e limited to t esting of in dividual pieces of the pr otocol with a final “dress rehe arsal” conducted just pri or to op ening. This issu e could also be mitiga ted by having an ex perienc ed te am performing the sam e oper atio ns in othe r locations. Tha t said, our resu l ts demonst rate tha t even in face of rapid planning and new op era tional sy stem developme nt, we wer e able to successfully exceed ou r goals for testing .

In summary, high-volume, community-wide ascert ainmen t of SARS-CoV-2 prevale nce by PCR and antibo dy testing was feasible a nd co uld be perform ed successfully when conducted in a community- led, drive-th rough model, with minimal s tar t up time. This op era tional mod el may be generali zable to those conduc ting any sort of high-throug hput tes ting for SARS-CoV-2, regardl ess of sampling methodol ogy. Key Messages 1. High-volume, community-wide ascertain ment of SARS-CoV-2 prevalence by PCR and antibody t esting could be conduc ted suc cessfully in a “pop-up” manner, using a multi-lane, drive-through mod el. 2. Early identificatio n of key community partners for all ph ases of the pr oject was cri tical to its success. 3. Robust da ta managem ent, including con t ingencies for onsite registr atio n, is a crucial part of early prep ara tion. 4. Proficient tes ting units, with multi ple lay ers of opera tional sup por t, can perfo rm increasingly efficient respir ato ry and fingerstick bloo d sampling over time. . CC-BY 4.0 International licenseIt 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 May 30, 2020. ; https://doi.org/10.1101/2020.05.29.20116426doi: medRxiv preprint 8 Contributors, Sources & Acknowledgements This article was crea ted t o share p rocess es, procedu res, an d lessons lea rned by o ur group in crea ting a novel method of community-based testi ng that we hop e will prove useful to ot he rs considering expandi ng access to SARS-CoV-2 testing. None of th e auth ors have po ten tial confl icts of interes t or competing int eres ts to disclose, and auth or at tribu tion has be en included in subm ission process. W e would also like to acknowledge A ndrew Kobylinkski for contributing his considera ble technical exp ertis e to this projec t. Finan cial Sup por t This work was primarily supported by th e Bolinas Community Land Trust. A dditio nal sources of suppor t included funding from the Chan Zuckerb erg Biohub Inves tigato r progr am (BG) and the Na tiona l Institutes of Heal th gran t 5T32AI007641- 17 (AA). IRB Ap prov al The aforemen tione d procedures wer e su bmitted to th e University of California Sa n Francisco institu tional r eview board for app roval, a nd the study was deem ed public heal th s urveillance no t requiring I RB oversight [IR B number 20-3 0636]. . CC-BY 4.0 International licenseIt 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 May 30, 2020. ; https://doi.org/10.1101/2020.05.29.20116426doi: medRxiv preprint 9

1. Furukawa NW , Brooks JT, S obel J . Evidence Suppor ting Transmission of Sever e Acute R espira tory Syndrome Coronavirus 2 Whil e Presympt omatic or Asymptoma tic. Emer g Infec t Dis 2020;26(7) doi: 10.3201/eid2607.201595 [published Online First : 2020/05/04] 2. Ton AN, J ethwa T, W at ers K, et al . COVID-19 drive through tes ting: An effective strat egy for conserving personal p rot ective equi pme nt. Am J Infe ct Co ntrol 2020 doi : 10.1016/j.ajic.2020.04.010 [published O nline First : 2020/04/17] 3. Kwon KT, Ko JH, Shin H, et al. Drive-Th rough Screeni ng Center for COVID-19: a Safe and Efficient Screening System agains t Massive Community Outb reak. J Korea n Med Sci 2020;3 5(11):e123. doi: 10.3346/jkms.2020.35.e123 [publish ed Online Fi rst: 2020/03/23] 4. UW COVID-19 Drive Thru Clinic AKA "T he Carvid" Clinic [Available from: https://aspr tracie .hhs.gov/techn ical-reso urces/resourc e/7856/uw-covid-19 -drive- thru-clinic- aka-the-carvid-clinic accessed May 11th 2020. 5. Verily PBb. COVID-19 Community Based Testing Guide 2020 [Availabl e from: https://www.projectbas eline .com/covid-19-guide/ accessed May 11th 2020. 6. Krammer F, Simon V. Se rology assays to manage COVID-19. Science 2020 doi: 10.1126/science.abc1227 [published Onl ine First: 2020/05/15] 7. Ration al use of person al pro tective eq uipment (PPE) for coronavirus disease (COVID-19): World Health Organiza tion , 2020:WHO refer enc e number: WHO/2019-nCoV/IPC PPE_us e/20.2. 8. Personal Prot ective Equipmen t Ext end ed Use and Re use Guid ance: UCSF He alt h Hospital Epidemiology and Infection Pr evention ; 2020 [4/7/2 020:[Available from: https://infectionco ntr ol.ucsfmedicalcen t er.o rg/sites/g/files/tkssra4681/f/Reuse_Guidelin es_PPE .pdf accessed 5/15 2020. 9. Int erim U.S . Guid ance for Risk Assess ment and Public Heal th Manag ement of Healthca re Person nel with Potenti al Exposur e in a Heal thcar e S etting to Patien ts with Coronavirus Disea se 2019 (COVID-19 ): Centers for Disease Contr ol and Prevention ; 2020 [Available from: https://www.cdc.gov/coronavirus/2019- ncov/hcp/guidance-risk-assesment-hcp.html accessed 4/13/20. . CC-BY 4.0 International licenseIt 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 May 30, 2020. ; https://doi.org/10.1101/2020.05.29.20116426doi: medRxiv preprint 20ft Pedestrian Overflow Area(cones at 20ft spacing) Non-registered cars exit KeySliding side wallSide wallRefrigeratorPainted traffic lineSpider box/power drop8ft table (x8)6ft table (x4)Tents 20’ X 20’ (6 total) Exit Vehicle & Pedestrian Registration Check GeneratorDumpsters Entrance Storage(limited) Restrooms (clean) Staff Break Area(clean zone) Storage(limited) LANE 4 LANE 1 LANE 2LANE 3 Supplies computer Refrigerator Testing assistantTester Table Table FloatSupervisor Specimen storage tableLEGEND: TENT Administrators Supplies computer Specimen storage table TESTING BAY TESTING BAY