CovRelex-SE: Adding Semantic Information for Relation Search via Sequence Embedding

In recent years, COVID-19 has impacted all aspects of human life. As a result, numerous publications relating to this disease have been issued. Due to the massive volume of publications, some retrieval systems have been developed to provide researchers with useful information. In these systems, lexical searching methods are widely used, which raises many issues related to acronyms, synonyms, and rare keywrds. In this paper, we present a hybrid relation retrieval system, CovRelex-SE, based on embeddings to provide high-quality search results. Our system can be accessed through the following URL: https://www.jaist.ac.jp/is/labs/nguyen-lab/systems/covrelex-se/. © 2023 Association for Computational Linguistics.

Trends in Postpartum Depression before and during the COVID-19 Pandemic and Correlates of Treatment Choices: A Population-Based Retrospective Cohort Study

Objectives: COVID-19-related stressors - including social distancing, material hardship, increased intimate partner violence, and loss of childcare, among others - may result in a higher prevalence of depression among postpartum individuals. This study examines trends in postpartum depression in the US from 2018 to 2022, as well as correlates of treatment choices among women with postpartum depression. Method(s): 1,108,874 women aged 14-64 in the Komodo Healthcare Map with 1+ live birth between April 2018 and December 2021 and had continuous enrollment 2+ years before and 4+ months after the delivery date were included. Prevalence of depression during postpartum (within 3 months after delivery) was calculated before (April 2018-March 2020) and during (April 2020-March 2022) COVID-19. Multinomial logistic regression was used to investigate correlates of treatment choices (no treatment, medication-only, psychotherapy-only, or both). Result(s): The prevalence of postpartum depression increased from 9.7% pre-pandemic to 12.0% during the pandemic (p < 0.001). Among 119,788 women with postpartum depression in 2018-2022, 47.0% received no treatment, 35.0% received medication-only, 10.0% received psychotherapy-only, and 7.4% received both within one month following their first depression diagnosis. Factors associated with an increase in the odds of receiving medication-psychotherapy treatment (vs. no treatment) included older ages;commercial insurance coverage;lower social vulnerability index;history of anxiety or mood disorder during and before pregnancy;and being diagnosed by a nurse practitioner, physician assistant, or behavioral care practitioner (vs. physician). Similar patterns were observed for medication-only and psychotherapy-only treatments. Conclusion(s): In this large, nationally representative sample of US insured population, the prevalence of postpartum depression increased significantly by 2.3 percentage-points during the pandemic (or a relative increase of 23.7%). Nonetheless, almost half of women with postpartum depression received no treatment, and only 7.5% received both medication and psychotherapy. The study highlighted potential socioeconomic and provider variation in postpartum depression treatment.Copyright © 2023

Impact of a Real-Time Electronic Alert on Antibiotic Treatment Duration for Pneumonia in Hospitalized Patients

Background. Limiting antibiotic prescribing to the shortest effective duration reduces antibiotic-associated adverse events and resistance. Up to two-thirds of patients receive excessive durations of therapy for pneumonia. This study evaluated the effect of a stewardship intervention to reduce excess antibiotic duration for inpatients with pneumonia. Methods. A dashboard was developed to generate real-time alerts when inpatients at an academic medical center received antibiotics with an indication of community- or hospital-acquired pneumonia for more than 5 or 7 days, respectively. From November 2019 through April 2021, alerts were regularly reviewed by the antibiotic stewardship (AS) team and intervened upon when patients exceeded the guideline recommended duration of therapy for pneumonia without additional indications for continuing antibiotics. We compared inappropriate duration of therapy pre- and post-implementation of the dashboard by calculating the mean number of excess days of antibiotics beyond the recommended duration. Patients with SARS-CoV-2 infection and patients on hospital services that care for patients with cysticfibrosis, bronchiectasis, or immunocompromising conditions were excluded. Four other hospitals within the same health system that did not utilize the dashboard generated alerts served as a comparison group. Results. During the intervention period, the AS team reviewed 834 patients with dashboard alerts and documented 115 interventions. For alerts reviewed without intervention, reasons for lack of intervention included active Infectious Diseases consult, additional infection diagnosis requiring a longer duration, and delayed clinical improvement. In the post-implementation period there was a mean of 1.28 excess days of antibiotics for pneumonia compared to the pre-implementation mean of 1.36 excess days. In comparison, aggregating data from the hospitals not utilizing the dashboard, there was a mean of 0.67 excess days post-intervention, compared to a mean 0.62 days pre-intervention. Conclusion. The pneumonia dashboard is a potentially valuable stewardship tool which may reduce excess days of antibiotics for pneumonia. The dashboard's impact may be improved by daily review and excluding patients with additional infection diagnoses.

THE SUBSEQUENT IMPACT OF COVID-19 FOR BOTH IMMUNE DYSREGULATION AND OPPORTUNISTIC INFECTIONS

SESSION TITLE: Post-COVID-19 Infection Complications SESSION TYPE: Case Report Posters PRESENTED ON: 10/17/2022 12:15 pm - 01:15 pm INTRODUCTION: The COVID-19 pandemic remains a burden to healthcare worldwide. Current literature suggests a possible link between COVID-19 survivors and opportunistic infections. We present a case of an immunocompetent male who presented with pneumocystis jirovecii pneumonia (PJP) in the setting of a recent COVID-19 infection. CASE PRESENTATION: A 65-year-old man with 65 pack-year smoking history, COPD, and recent COVID-19 pneumonia requiring hospitalization 1 month prior, presented with 2 days of dyspnea. His physical exam was notable for hypoxia requiring supplemental oxygen and bibasilar crackles. WBC was elevated at 15,500. ABG was significant for hypoxemia. A CT chest demonstrated bilateral peripheral mixed ground glass and consolidative opacities (Figure 1). Upon admission, the patient received ceftriaxone and azithromycin for presumed community acquired pneumonia. However, the patient continued to clinically decompensate with increasing oxygen requirements. As such, a repeat CT was ordered which demonstrated bilateral ground glass opacities, interstitial scarring, and subpleural honeycombing (Figure 2). A bronchoscopy was also performed;bronchoalveolar lavage was positive for PJP by PCR but with negative DFA. The patient was started on trimethoprim/sulfamethoxazole (TMP/SMX) and prednisone. After 3 weeks, the patient clinically improved and was discharged to a skilled nursing facility for rehabilitation. Subsequent CT scan 1 month after initial presentation demonstrated fibrotic changes and bronchial wall thickening (Figure 3). DISCUSSION: In our case, the patient was an immunocompetent male with underlying COPD and recent COVID-19 pneumonia, found to have PJP by PCR. The PCR test for PJP has a higher sensitivity compared to DFA (1), so our patient's incongruent positive PCR and negative DFA test results may represent true PJP or an organizing pneumonia with colonization. A lung biopsy with histology ultimately could have confirmed the diagnosis but was not performed in this case given the patient's clinical improvement with steroids and TMP/SMX (2). Previous studies have demonstrated that COVID-19 can cause immune dysregulation via decreased T cell count and thus can increase the risk for opportunistic infections (3). Furthermore, multiple case reports have shown concurrent COVID-19 and PJP in immunocompetent patients (1). Based on his findings, we believe that our patient was at increased risk for and subsequently developed PJP as a direct consequence of his recent COVID-19 infection. CONCLUSIONS: COVID-19 has been identified as a predisposing factor for subsequent chronic conditions. Studies have demonstrated the capability of COVID-19 infection to weaken the immune system for opportunistic infections as well as remodel the pulmonary architecture. Both conditions can confer high morbidity and mortality for COVID-19 survivors. As such, a close surveillance of this population is warranted. Reference #1: Chong WH, Saha BK, Chopra A. Narrative review of the relationship between COVID-19 and PJP: does it represent coinfection or colonization? Infection (2021) 49:1079–1090. doi:10.1007/s15010-021-01630-9 Reference #2: Culebras M, Loor K, Sansano I, Persiva Ó, Clofent D, Polverino E, Felipe A, Osorio J, Muñoz X, Álvarez A, et al. Histological Findings in Transbronchial Cryobiopsies Obtained From Patients After COVID-19. Chest (2022) 161:647–650. doi:10.1016/j.chest.2021.09.016 Reference #3: Qin C, Zhou L, Hu Z, Zhang S, Yang S, Tao Y, Xie C, Ma K, Shang K, Wang W, et al. Dysregulation of immune response in patients with coronavirus 2019 (COVID-19) in Wuhan, China. Clin Infect Dis (2020) 71:762–768. doi:10.1093/cid/ciaa248 DISCLOSURES: No relevant relationships by Duc Do No relevant relationships by Clara Suh

LIGHTSITE III (Interim Analysis): Evaluation of Multiwavelength Photobiomodulation in Dry Age-Related Macular Degeneration Using the LumiThera Valeda Light Delivery System

Purpose : Dry age-related macular degeneration (AMD) is a leading contributor to visual impairment across the globe. No current treatment exists to improve visual function or reduce disease progression outside of vitamin supplementation and lifestyle changes. LIGHTSITE III is evaluating multiwavelength photobiomodulation (PBM) therapy using the LumiThera Valeda® Light Delivery System in dry AMD Methods : LIGHTSITE III (NCT04065490) is a prospective, double-masked, randomized, sham-controlled, parallel group, multi-center study to assess the safety and efficacy of PBM in dry AMD. Target enrollment was approximately 96 subjects (144 eyes). Subjects are treated with six series of PBM/Sham treatments (3x per week for 3 weeks) delivered over a 24-month period with a 13-month efficacy analysis of data. PBM therapy consists of low-level light exposure to selected tissues resulting in positive effects on mitochondrial output and improvement in cellular activity. Valeda is used to deliver multiwavelength PBM treatment using 590, 660 and 850 nm of light. Subjects are assessed for clinical and safety outcomes (i.e., best-corrected visual acuity (BCVA), low- luminance BCVA, contrast sensitivity, reading speed, color vision, VFQ-25 and perimetry). Independent OCT, FAF and color fundus imaging outcomes at selected timepoints are analyzed by a masked imaging reading center Results : A total of 148 eyes from 100 subjects with dry AMD have been enrolled and randomized in a 2:1 design (PBM:Sham). The majority of subjects are female (68%) and Caucasian (99%). The average age at enrollment was 75 years and mean time since dry AMD diagnosis is 4.9 years. COVID-19 interference has been minimal and not significantly impacted subject enrollment or retention. Clinical and anatomical outcome data from the interim analysis conducted at Month 13 is presented. Results from the 21-month time point are expected at end of 2022 Conclusions : LIGHTSITE III provides the largest, randomized controlled trial evaluating the effects of PBM in dry AMD subjects. PBM therapy may offer a new treatment strategy with a unique mechanism and modality for patients with dry AMD.

A Single Center, Retrospective Analysis of Inpatient Neurologic Consultations of Patients Infected with SARS-CoV-2

Objective: To determine if inpatient neurological consultations differ between COVID-19 and non-COVID-19 respiratory infections. Background: The COVID-19 pandemic has posed challenges to healthcare systems across the world, and neurological complications of COVID-19 have garnered increased concern in medicine and the public. Neurological consultation for patients with viral-mediated disease is common;it is unknown whether the neurologist's approach to inpatient consultation of patients with COVID-19 should be altered. Design/Methods: We performed a retrospective chart analysis of inpatient neurologic consultations at three major hospitals comprising the University of Pennsylvania Health System. We compared the reason for neurologic consultation and final diagnosis of 62 patients with COVID-19 between March 2020 and April 2021 to 56 patients with non-COVID-19 respiratory virus (defined as Influenza A, Influenza B, Respiratory Syncytial Virus, Rhinovirus, or Adenovirus) between January 2019 and January 2020. Secondary metrics included mortality and level of care. A frequency and means analysis were completed to evaluate the relative difference between groups on all primary and secondary metrics. Results: Stroke was the only diagnosis more common in the COVID population as compared to the non-COVID virus population (14% vs. 9%). Neurology was consulted more frequently for altered mental status in the COVID-19 population (27% vs. 18%);however, the ultimate diagnosis was toxic-metabolic encephalopathy due to infection, not a consequence of COVID-19 itself. Neurology was consulted significantly later in the hospital course of COVID-19 (3.1 vs. 0.96 days), despite a higher mortality in the other population (30% vs. 19%). Conclusions: Patients requiring inpatient neurologic consultation with a diagnosis of COVID-19 or another respiratory virus were found to be remarkably similar in terms of their ultimate neurologic diagnosis, with the exception of stroke, which was more common in the COVID-19 population. These results suggest the neurological approach to patients with COVID-19 should be similar to that in patients with other respiratory infections.

The Effect of Foreign Direct Investment Inflow on Exports: Evidence from Vietnam

Foreign direct investment (FDI) and export are now often regarded as two of the most important drivers of economic growth on a worldwide scale. The impact of foreign direct investment on Vietnam's exports is investigated in this study. The data for the time period 1985-2020 was obtained from the World Bank and the Vietnam General Statistics Office. The years 1985 to 2020 were chosen to evaluate the evolution of macroeconomic parameters since 1986. The impact of the Covid-19 epidemic on renovation reform. The Johansen co-integration test proved that FDI and domestic investment (DI) had a long-term positive impact on Vietnam's export growth. The Granger causality test revealed that there is a one-way relationship between FDI and export in the near term, but no such relationship exists between DI and export. The result of the variance decomposition study demonstrates that the FDI sector has a bigger impact on Vietnam's export growth than the DI sector. Furthermore, export activities are vulnerable to FDI sector shocks. As a result, in recent years, FDI has been regarded as the most important factor of export growth in Vietnam.

Prior COVID-19 infection: an underappreciated factor in vaccine hesitancy in the USA.

Despite tremendous efforts to quickly identify the 'vaccine hesitant' in the USA, what has emerged instead is a complex picture of a highly heterogeneous unvaccinated population. Although numerous factors have been implicated in influencing US COVID-19 vaccine decision-making, the role that prior coronavirus disease 2019 (COVID-19) infection may play in vaccine receipt has been largely uninvestigated. Using data from two separate US national surveys, the US COVID-19 Trends and Impact Survey and the Household Pulse Survey, we find that roughly one-quarter of unvaccinated survey respondents has had a prior COVID-19 infection. Prior COVID-19 infection halves the odds of receiving the vaccine. This information is consequential for ongoing vaccine outreach efforts.

Community-led tuberculosis and COVID-19 risk communication in the Vietnamese migrant population in Japan

Background and Aims: Vietnamese migrants have become the largest group of newly notified tuberculosis (TB) cases among overseas-born in Japan in 2019. Migrant population access to information regarding TB and coronavirus disease 2019 (COVID-19) is limited in Japan mainly due to language barriers. A previous survey of the Vietnamese in Japan reported that Facebook (FB) is the most popular information channel. We aimed to develop digital health programmes, and to assess their effectiveness for the promotion of risk communication for TB and COVID-19. Methods: Digital health programmes were developed through multi-sectoral collaboration with Vietnamese and Japanese health professionals, and Vietnamese social media. Programmes included a real-time online seminar with FBbased live-streaming, a Vietnamese video of the seminar, and Vietnamese digital news articles. We reviewed the implementation of the health programmes and their outcomes. Results: Eighty-five people participated in the real-time online seminar. TB and COVID-19-related topics including prevention, access to information for testing, care, support and COVID-19 vaccines were discussed. Participants had opportunities to interact with health professionals during question-and-answer sessions, and listen to experiences of TB patients. The seminar video had over 5,100 views via FBbased Vietnamese social media and sites by the following day. Some Vietnamese posted health inquiries, and others sought help with access to medical care in Japan. Conclusions: A community-led digital health approach helped disseminate TB and COVID-19-related information, and promote health seeking behaviour among Vietnamese migrants in Japan. Multi-sectoral collaboration and utilisation of online channels were the key to success of this programme.

U.S. frontline workers and COVID-19 inequities.

We overcome a lack of frontline worker status information in most COVID-19 data repositories to document the extent to which occupation has contributed to COVID-19 disparities in the United States. Using national data from over a million U.S. respondents to a Facebook-Carnegie Mellon University survey administered from September 2020 to March 2021, we estimated the likelihoods of frontline workers, compared to non-frontline workers, 1) to ever test positive for SARs-Cov-2 and 2) to test positive for SARs-Cov-2 within the past two weeks. Net of other covariates including education level, county-level political environment, and rural residence, both healthcare and non-healthcare frontline workers had higher odds of having ever tested positive for SARs-Cov-2 across the study time period. Similarly, non-healthcare frontline workers were more likely to test positive in the previous 14 days. Conversely, healthcare frontline workers were less likely to have recently tested positive. Our findings suggest that occupational exposure has played an independent role in the uneven spread of the virus. In particular, non-healthcare frontline workers have experienced sustained higher risk of testing positive for SARs-Cov-2 compared to non-frontline workers. Alongside more worker protections, future COVID-19 and other highly infectious disease response strategies must be augmented by a more robust recognition of the role that structural factors, such as the highly stratified U.S. occupational landscape, have played in the uneven toll of the COVID-19 pandemic.

The impact of the COVID-19 pandemic on intravitreal bevacizumab use

Purpose : The COVID-19 pandemic has led to stay-at-home orders which may lead to fewer medical and ophthalmology visits. Efficacy of treatments for retinal vascular diseases such as wet age-related macular degeneration (wet AMD) depend on strict adherence to regimens of anti-vascular endothelial growth factor (VEGF) injections. The purpose of this study is to describe and compare the number of bevacizumab injections administered in 2019 and 2020 to determine if the COVID-19 pandemic may have affected usage. Methods : We conducted a single-center, retrospective observational study which compared the number of bevacizumab injections in January through November 2020 with the number of bevacizumab injections in 2019. The total number of injections for each month and year were compared.Results : The total number of bevacizumab injections in 2020 (Jan-Nov) was 683 compared to 801 in 2019 (Jan-Nov), a 14.7% reduction (p = 0.02). Significantly fewer injections were administered in March, April, and May 2020 (p < 0.01), and from August to November (p < 0.01). The fewest number of injections for 2020 were administered in April (n = 45) and May (n = 45). Conclusions : In this patient population, a significant reduction in bevacizumab injections occurred during the COVID pandemic, and the drop in injections was first seen in March 2020 which coincided with stay-at-home orders within the catchment area. These data suggest that the Covid-19 pandemic likely affected the number of patients coming to the retina clinic. Additional data is needed to determine if other factors may have contributed to this decline, and if these missed visits has adversely affected visual acuity outcomes.

The Impact of the COVID-19 pandemic on aflibercept and ranibizumab anti-VEGF Injections

Purpose : The COVID-19 Pandemic resulted in a substantial decrease in outpatient ophthalmology clinic visits due to stay-at-home orders. Retinal vascular diseases and wet age-related macular degeneration require adherence to treatment regimens with antivascular endothelial growth factor (anti-VEGF) injections. The purpose of this study was to assess how the COVID-19 pandemic affected the number of injections compared to prepandemic values. Methods : This retrospective, observational clinical study assessed all anti-VEGF injections of aflibercept and ranibizumab for both 2019 (Jan-Dec) and 2020 (Jan-Nov) at an academic center. The total number of injections for each year was compared, as well as the number of injections for each month in 2020 compared to the average number of monthly injections in 2019. Both 1-sample and 2-sample t-tests were conducted, and a p-value less than 0.05 was considered significant. Results : The total number of aflibercept and ranibizumab injections in 2019 (Jan-Nov) were 4989 and 522 respectively, compared to 4855 and 515 in 2020 (Jan-Nov). No significant difference in injections was noted between 2019 and 2020 for either aflibercept (p=0.31) or ranibizumab (p=0.80). When analyzing each month in 2020 compared to the average number of aflibercept injections per month in 2019, there were significantly fewer aflibercept injections administered in February (p<0.01), August (p=0.03), and November (p<0.01), but significantly higher injections administered in July (p<0.01) and October (p<0.01). For ranibizumab, significantly fewer injections were administered in March (p=0.02), April (p=0.02), and May (p<0.01), but a higher number of injections were administered in September (p<0.01) and November (p<0.01). Conclusions : Despite stay-at-home orders, there was no significant decrease in the overall number of aflibercept or ranibizumab injections during the COVID pandemic compared to the prior year. Although further information is needed to determine if these injections represent both new and return patients, these data suggest that in this specific population, patients still returned to their retina specialist for care despite the ongoing pandemic.

Using race- and age-specific COVID-19 case data to investigate the determinants of the excess COVID-19 mortality burden among Hispanic Americans

BACKGROUND Age-adjusted COVID-19 mortality estimates have exposed a previously hidden excess mortality burden for the US Hispanic population. Multiple explanations have been put forth, including unequal quality/access to health care, higher proportion of pre-existing health conditions, multigenerational household composition, and disproportionate representation in telecommute-unfriendly occupations. However, these hypotheses have been rarely tested. OBJECTIVE We examine age-stratified patterns of Hispanic COVID-19 mortality vis-à-vis patterns of exposure to evaluate the multiple posited hypotheses. METHODS We use a combination of public and restricted data from the Centers of Disease Control and Prevention and leverage national and subnational race- and age-stratified COVID-19 mortality and case burdens/advantages to evaluate the workplace vulnerability hypothesis. We also use individual-level information on prior health conditions and mortality from the case data to assess whether observed patterns are consistent with the other hypotheses. RESULTS Our results indicate that the disproportionate burdens for both COVID-19 case and mortality for the Hispanic population are largest among the working-age groups, supporting the hypothesis that workplace exposure plays a critical role in heightening vulnerability to COVID-19 mortality. We find little evidence to support the hypotheses regarding multigenerational household composition, pre-existing health conditions, or unequal quality/access to health care. CONCLUSION Our findings point to the key roles played by age structure and differential exposure in contributing to the disproportionately severe impact of COVID-19 on the Hispanic population. CONTRIBUTION We contribute evidence to explain the driving factors in the observed excess COVID-19 mortality burden among Hispanics. Our findings underscore the importance of focusing on more robust workplace protections, particularly for working-age minority populations. © 2021. D. Phuong Do & Reanne Frank.

Healthcare worker attendance during the early stages of the COVID-19 pandemic: A longitudinal analysis of fingerprint-verified data from all public-sector secondary and tertiary care facilities in Bangladesh

Background The COVID-19 pandemic has overwhelmed hospitals in several areas in high-income countries. An effective response to this pandemic requires health care workers (HCWs) to be present at work, particularly in low- and middle-income countries (LMICs) where they are already in critically low supply. To inform whether and to what degree policymakers in Bangladesh, and LMICs more broadly, should expect a drop in HCW attendance as COVID-19 continues to spread, this study aims to determine how HCW attendance has changed during the early stages of the COVID-19 pandemic in Bangladesh. Methods This study analyzed daily fingerprint-verified attendance data from all 527 public-sector secondary and tertiary care facilities in Bangladesh to describe HCW attendance from January 26, 2019 to March 22, 2020, by cadre, hospital type, and geographic division. We then regressed HCW attendance onto fixed effects for day-of-week, month, and hospital, as well as indicators for each of three pandemic periods: a China-focused period (January 11, 2020 (first confirmed COVID-19 death in China) until January 29, 2020), international-spread period (January 30, 2020 (World Health Organization's declaration of a global emergency) until March 6, 2020), and local-spread period (March 7, 2020 (first confirmed COVID-19 case in Bangladesh) until the end of the study period). Findings On average between January 26, 2019 and March 22, 2020, 34.1% of doctors, 64.6% of nurses, and 70.6% of other health care staff were present for their scheduled shift. HCWs' attendance rate increased with time in 2019 among all cadres. Nurses' attendance level dropped by 2.5% points (95% confidence interval (CI) =-3.2% to -1.8%) and 3.5% points (95% CI = -4.5% to -2.5%) during the international-spread and the local-spread periods of the COVID-19 pandemic, relative to the China-focused period. Similarly, the attendance level of other health care staff declined by 0.3% points (95% CI=-0.8% to 0.2%) and 2.3% points (95% CI = -3.0% to -1.6%) during the international-spread and local-spread periods, respectively. Among doctors, however, the international-spread and local-spread periods were associated with a statistically significant increase in attendance by 3.7% points (95% CI =2.5% to 4.8%) and 4.9% points (95% CI=3.5% to 6.4%), respectively. The reduction in attendance levels across all HCWs during the local-spread period was much greater at large hospitals, where the majority of COVID-19 testing and treatment took place, than that at small hospitals. Conclusions After a year of significant improvements, HCWs' attendance levels among nurses and other health care staff (who form the majority of Bangladesh's health care workforce) have declined during the early stages of the COVID-19 pandemic. This finding may portend an even greater decrease in attendance if COVID-19 continues to spread in Bangladesh. Policymakers in Bangladesh and similar LMICs should undertake major efforts to achieve high attendance levels among HCWs, particularly nurses, such as by providing sufficient personal protective equipment as well as monetary and non-monetary incentives.

Unequal burdens: assessing the determinants of elevated COVID-19 case and death rates in New York City's racial/ethnic minority neighbourhoods.

BACKGROUND: The disproportionate burden of the COVID-19 pandemic on racial/ethnic minority communities has revealed glaring inequities. However, multivariate empirical studies investigating its determinants are still limited. We document variation in COVID-19 case and death rates across different racial/ethnic neighbourhoods in New York City (NYC), the initial epicentre of the U.S. coronavirus outbreak, and conduct a multivariate ecological analysis investigating how various neighbourhood characteristics might explain any observed disparities. METHODS: Using ZIP-code-level COVID-19 case and death data from the NYC Department of Health, demographic and socioeconomic data from the American Community Survey and health data from the Centers for Disease Control's 500 Cities Project, we estimated a series of negative binomial regression models to assess the relationship between neighbourhood racial/ethnic composition (majority non-Hispanic White, majority Black, majority Hispanic and Other-type), neighbourhood poverty, affluence, proportion of essential workers, proportion with pre-existing health conditions and neighbourhood COVID-19 case and death rates. RESULTS: COVID-19 case and death rates for majority Black, Hispanic and Other-type minority communities are between 24% and 110% higher than those in majority White communities. Elevated case rates are completely accounted for by the larger presence of essential workers in minority communities but excess deaths in Black neighbourhoods remain unexplained in the final model. CONCLUSIONS: The unequal COVID-19 case burden borne by NYC's minority communities is closely tied to their representation among the ranks of essential workers. Higher levels of pre-existing health conditions are not a sufficient explanation for the elevated mortality burden observed in Black communities.