key: cord-0258498-k0w01ppp authors: Strathdee, S. A.; Abramovitz, D.; Harvey-Vera, A.; Vera, C.; Rangel, G.; Artamonova, I.; Chaillon, A.; Ignacio, C.; Calderon, A.; Martin, N. K.; Patterson, T. L. title: Prevalence and correlates of SARS-CoV-2 seropositivity among people who inject drugs in the San Diego-Tijuana border region date: 2021-08-07 journal: nan DOI: 10.1101/2021.08.05.21261671 sha: f2d5dc78565ecc54ce98aa7628ff9c8d3b797a8a doc_id: 258498 cord_uid: k0w01ppp Background: People who inject drugs may be at elevated SARS-CoV-2 risk due to their living conditions and/or exposures when seeking or using drugs. No study to date has reported upon risk factors for SARS-CoV-2 infection among people who inject drugs or sex workers. Methods and Findings: Between October, 2020 and June, 2021, participants aged [≥]18 years from San Diego, California, USA and Tijuana, Baja California, Mexico who injected drugs within the last month underwent interviews and testing for SARS-CoV-2 RNA and antibodies. Binomial regressions identified correlates of SARS-CoV-2 seropositivity. Of 386 participants, SARS-CoV-2 seroprevalence was 36.3% (95% CI: 31.5%-41.1%); 92.1% had detectable IgM antibodies. Only 37.5% had previously been tested. Seroprevalence did not differ by country of residence. None tested RNA-positive. Most (89.5%) reported engaging in [≥]1 protective behavior [e.g., facemasks (73.5%), social distancing (46.5%), or increasing handwashing/sanitizers (22.8%)]. In a multivariate model controlling for sex, older age, and Hispanic/Latinx/Mexican ethnicity were independently associated with SARS-CoV-2 seropositivity, as was engaging in sex work (AdjRR: 1.63; 95% CI: 1.18-2.27) and having been incarcerated in the past six months (AdjRR: 1.49; 95% CI: 0.97-2.27). Presence of comorbidities and substance using behaviors were not associated with SARS-CoV-2 seropositivity. Conclusions: This is the first study to show that sex work and incarceration were independently associated with SARS-CoV-2 infection. Despite engaging in protective measures, over one-third had evidence of infection, reinforcing the need for a coordinated binational response. Risk mitigation and vaccination is especially needed among older and Hispanic people who inject drugs and those with less agency to protect themselves, such as those who are sex workers or incarcerated. The disproportionate burden of COVID-19 on under-represented minorities is well documented [1] [2] [3] . However, data are lacking on people with substance use disorders who may be vulnerable to acquiring 41 and transmitting SARS-CoV-2 due to their living conditions (e.g., homelessness, incarceration) and drug-42 related behaviors (e.g. smoking, vaping, sharing drug paraphernalia, sex work) [4, 5] . In a nation-wide 43 study, those with substance use disorders, especially African Americans and opioid users, were at 44 increased risk of COVID-19 [6] . However, this study could not determine whether these factors were 45 independently associated with COVID-19 diagnosis since risk factor data were not available, and the 46 sample was restricted to patients accessing health care. In a telephone survey of active and former drug 47 injectors in Baltimore, Maryland, compared to former substance users, those who were actively using 48 substances were less likely to report social distancing [5] . There is also evidence that the COVID-19 49 pandemic has interrupted drug supplies as well as harm reduction and addiction treatment services [7] . California, Mexico is Tijuana, a city of ~1.5 million people, located twenty minutes from San Diego and 60 located on a major drug trafficking route. In an attempt to limit SARS-CoV-2 transmission, the U.S.- Mexico border was closed to essential travel on March 21, 2020 for the eight months duration of the 62 study. . CC-BY-NC-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 7, 2021. antigens (Nucleocapsid, Spike S1-S2, Spike S1, Spike S1-RBD, Spike S2) within a multiplex format 80 based on photonic ring resonance. A machine learning algorithm was used to call results using the . CC-BY-NC-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 7, 2021. ; https://doi.org/10.1101/2021.08.05.21261671 doi: medRxiv preprint Statistical Analysis: SARS-CoV-2 prevalence was calculated with 95% confidence intervals (CIs) based 91 on the Binomial distribution. Those testing indeterminate were excluded. The Cochran-Armitage test was 92 used to assess trends in SARS-CoV-2 seroprevalence. Characteristics of participants testing SARS-CoV-2 seropositive versus seronegative were compared Role of funding source: The funders had no involvement in the study design, collection, analysis, 103 interpretation or writing of this rpreport, nor the decision to submit the paper for publication. Of 405 participants tested, none had detectable SARS-CoV-2 RNA. Considering serologic evidence of 106 . CC-BY-NC-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 7, 2021. ; https://doi.org/10.1101/2021.08.05.21261671 doi: medRxiv preprint Older age and identifying as Hispanic, Latinx, or Mexican were significantly associated with SARS-CoV-2 117 seropositivity (Table 1) . Considering potential community exposures in the past six months (Table 2) , 118 those engaging in sex work were significantly more likely to test seropositive compared to those who did 119 not (17.9% vs. 9.3%, p=0.01, [Relative Risk (RR): 1.53; 95% CI: 1.12-2.09]. Being incarcerated was 120 associated with marginally higher SARS-CoV-2 seropositivity (RR: 1.43; 95% CI: 0.99-2.09). We did not 121 observe injection drug use, smoking, vaping specific drugs or any other substance use behaviors to be 122 associated with seroprevalence, nor were any co-morbidities (i.e., HIV, HCV, Type 2 diabetes, 123 hypertension). .85 Highest year of school completed (IQR) 11.0(8.0,12.0) 11.0(7.0,12.0) 11.0(7.0,12.0) .99 Married or common law .08 . CC-BY-NC-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 7, 2021. ; https://doi.org/10.1101/2021.08.05.21261671 doi: medRxiv preprint .14 Visited shooting galleries* . CC-BY-NC-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 7, 2021. ; https://doi.org/10.1101/2021.08.05.21261671 doi: medRxiv preprint . CC-BY-NC-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 7, 2021. ; https://doi.org/10.1101/2021.08.05.21261671 doi: medRxiv preprint In a multivariate model that controlled for sex (Table 3) , each year in age was associated with a 2% 128 increase in SARS-CoV-2 seropositivity (Adjusted relative risk (AdjRR): 1.02; 95% CI: 1.01-1.03). Identifying as Hispanic/Latinx/Mexican was also independently associated with SARS-CoV-2 sharing drug paraphernalia in the prior six months ( Table 1) . None of these behaviors were significantly 141 . CC-BY-NC-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 7, 2021. ; https://doi.org/10.1101/2021.08.05.21261671 doi: medRxiv preprint associated with SARS-CoV-2 serostatus. Those testing SARS-CoV-2 seropositive were more likely to 142 report having stopped smoking in the last six months and to have had a COVID-19 test. Compared to non-sex workers, sex workers were just as less likely report protective behaviors but were 144 significantly more likely to report being exposed to someone with COVID-19 (20% vs. 3.8%, p=0.001) or 145 to have low/very low food security (94.3% vs. 77.9% p=0.02). In this community-based study of people who inject drugs in the San Diego-Tijuana border region, over Consistent with other studies, subjects who were older [6] and who identified as Hispanic, Latinx, or 165 Mexican [2] were significantly more likely to test SARS-CoV-2 seropositive. This demonstrates that even 166 . CC-BY-NC-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 7, 2021. among the lowest socioeconomic strata in the US and Mexico, ethnic disparities in SARS-CoV-2 167 seroprevalence persist. We did not observe substance use behaviors, such as smoking, vaping, or use of specific drugs such as 169 opiates or stimulants to be associated with an elevated risk of SARS-CoV-2 infection, as others have 170 hypothesized [4,6]. Instead, we observed that circumstances in the risk environment, notably sex work 171 and incarceration, were associated with significantly higher SARS-CoV-2 seroprevalence. To our knowledge, ours is the first study to show that sex work is independently associated with higher 173 SARS-CoV-2 prevalence after controlling for potential confounders such as sex, age, and ethnicity. In an 174 unpublished study in Denmark, SARS-CoV-2 seroprevalence was higher among sex workers (12.2%) 175 than people experiencing homelessness (6.8%) or the general population (2.9%) [18], but the 176 independent effects of these and other factors were not assessed. Although the majority of our sample reported that COVID-19 had greatly affected their income and 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 7, 2021. ; https://doi.org/10.1101/2021.08.05.21261671 doi: medRxiv preprint protections [22] . In an earlier study, we found that female sex workers in Tijuana who inject drugs were 194 more vulnerable to offers of unprotected sex in exchange for more money or drugs compared to those 195 that did not inject [23] . Our finding that recent incarceration was independently associated with SARS-CoV-2 seropositivity could Our study also found that over half of those testing SARS-CoV-2 seropositive had not been tested prior to 203 study enrolment. These findings underscore the need to improve community outreach to provide testing, 204 vaccines, and treatment, for example using mobile syringe exchange programs. Our ability to detect some associations was limited due to statistical power. Ours was a non-random 207 sample and the cross-sectional study design precludes drawing causal inferences. Participants 208 experiencing symptoms may have changed behaviors, such as stopped smoking or sought COVID-19 209 testing. Of note, none of the self-collected swabs tested positive for SARS-CoV-2 RNA. The sensitivity of the 211 pooling approach could have been impacted by: 1) the viral load of any particular infected individual; 2) 212 the consistency with which swabs were obtained; 3) storage, shipping, and transport conditions; or 4) of 213 diluting out (via pooling) any viral SARS-CoV-2 RNA collected below the limit of detection. In this study, 214 we limited our pools to ≤10 swabs, and previously validated our approach with up to 30 samples per pool 215 where the limit of detection was estimated at 2.4 copies/µL [25] . Therefore, it is unlikely that we 216 experienced loss of sensitivity due to pooling. Since SARS-CoV-2 antibodies may wane over time especially among patients who are asymptomatic or 218 mildly symptomatic [26] [27] , infections that occurred earlier in the epidemic may have gone undetected. . CC-BY-NC-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 7, 2021. ; https://doi.org/10.1101/2021.08.05.21261671 doi: medRxiv preprint However, the majority of participants testing SARS-CoV-2 seropositive had detectable IgM antibody titers, 220 which is suggestive of recent infection [28] . Some misclassification could have occurred among those 221 testing indeterminate who were recently infected, but these were few in number and would have tended 222 to underestimate SARS-CoV-2 prevalence, dampening any observed associations. Since COVID-19 vaccines did not become available to adults 18-65 years of age in San Diego County participants who had previously been tested for SARS-CoV-2 or had received COVID-19 vaccine was 235 very low. Efforts are needed to mitigate risks and provide COVID-19 vaccines to older and Hispanic 236 people who inject drugs, especially those with less agency to protect themselves, such as those who are 237 sex workers or incarcerated. . CC-BY-NC-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 7, 2021. ; https://doi.org/10.1101/2021.08.05.21261671 doi: medRxiv preprint the Center for AIDS Research, Dr Pamina Gorbach for assistance with survey measures, Dr. Davey Smith for helpful suggestions on the study design and Sharon Park for assistance with manuscript 243 preparation. . CC-BY-NC-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 7, 2021. ; https://doi.org/10.1101/2021.08.05.21261671 doi: medRxiv preprint COVID-19 and Racial/Ethnic Disparities Disparate Health Impact on the Hispanic/Latinx Population in the United States Racial/Ethnic 251 Disparities In COVID-19 Exposure Risk, Testing, And Cases At The Subcounty Level In California Collision of the COVID-19 and Addiction Epidemics The health and social 255 consequences during the initial period of the COVID-19 pandemic among current and former people 256 who inject drugs: A rapid phone survey in COVID-19 risk and outcomes in patients with substance use 259 disorders: analyses from electronic health records in the United States A rapid review of the impacts 262 of "Big Events " on risks, harms, and service delivery among people who use drugs: Implications for 263 responding to COVID-19 Maverick™ SARS-CoV-2 Multi-Antigen Serology Panel v2 268 01030ART-01 Target specific 270 serologic analysis of COVID-19 convalescent plasma Presentation of the results of the "Survey on Prevalence 272 of Covid-19 in Baja California Health literacy and changes in 276 pattern of drug use among participants at the Stockholm Needle Exchange Program during the 277 COVID-19 pandemic Preliminary indications 279 of the burden of COVID-19 among people who inject drugs in England and Northern Ireland and the 280 impact on access to health and harm reduction services Characteristics of 282 COVID-19 in Homeless Shelters : A Community-Based Surveillance Study COVID-19 Prevalence 285 among People Experiencing Homelessness and Homelessness Service Staff during Early Community 286 Transmission San Diego Epidemiology and Research for COVID Health (SEARCH) SARS-CoV-2 294 antibody prevalence among homeless people, sex workers and shelter workers in Denmark: a 295 nationwide cross-sectional study Associated With COVID-19 in Women at High Risk of HIV Infection in Rural Kenya During Novel Coronavirus (SARS-CoV-2) Era in the Niger Delta Region: Relationships Between 301 Knowledge, Preventive Practice, and Transmission Potential Sex workers must not be forgotten in the 303 COVID-19 response A comparison of registered 305 and unregistered female sex workers in Social and structural factors 307 associated with HIV infection among female sex workers who inject drugs in the Mexico-US border 308 region California State Prison System: An Observational Study of Decarceration, Ongoing Risks, and Risk 311 No Evidence of SARS-CoV-2 Seminal 313 Shedding Despite SARS-CoV-2 Persistence in the Upper Respiratory Tract Clinical and immunological assessment of 318 asymptomatic SARS-CoV-2 infections Epidemiological feature, viral shedding, and antibody 320 seroconversion among asymptomatic SARS-CoV-2 carriers and symptomatic/presymptomatic 321 COVID-19 patients The authors gratefully acknowledge the La Frontera study team and participants in San Diego and Tijuana, staff at Genalyte and Fluxergy for assistance interpreting laboratory results, laboratory staff at