UCSF

The illness caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) – a novel coronavirus identified in Wuhan at the end of 2019 that led to a global pandemic – primarily manifests as a lung infection with symptoms ranging from those of a mild upper respiratory tract infection to severe pneumonia, acute respiratory distress syndrome (ARDS), and death. Severe illness with coronavirus disease 2019 (COVID-19) can occur in healthy individuals and also in people with underlying medical conditions, including people with rheumatic diseases. Rheumatic diseases involve the dysregulation of the immune system, lead to systemic inflammation of the joints, muscles, bones, and organs, and are more prevalent in older ages.1,2 People with rheumatic disease have a higher prevalence of several comorbidities (such as pulmonary and kidney disease, heart disease or hypertension, obesity, and diabetes)3-8 and may be receiving immunosuppressive or immunomodulatory medications which can increase the risk of serious or opportunistic infections.9,10 In general, immunosuppression and the presence of comorbidities are associated with an increased risk of serious infection in people with rheumatic diseases.11 In addition, the impact of previous epidemics caused by coronavirus infections like severe acute respiratory syndrome 1 (SARS-1) and middle east respiratory syndrome (MERS) on patients with rheumatic diseases or other immune-mediated inflammatory diseases (IMIDs) has been scarcely reported.12 Therefore, in the midst of the pandemic, the implications of COVID-19 for people living with rheumatic diseases was a considerable concern. In March 2020, to address this knowledge gap, the COVID-19 Global Rheumatology Alliance (GRA) registry was developed by a global network of rheumatologists, scientists, and patient representatives.13 The GRA registry is a physician-reported registry of people with rheumatic diseases diagnosed with COVID-19 and includes information on patient demographics, rheumatic disease characteristics, immunomodulatory medications used for the treatment of rheumatic disease, and comorbidities, as well as COVID-19 diagnoses, treatments, outcomes, and complications. The overall goal of this dissertation is to use data from the GRA registry to study COVID-19 outcomes in people with rheumatic diseases to advance rheumatology care in the COVID-19 pandemic. The dissertation is organized into three chapters each describing a key aim of this dissertation. The first chapter pools data from three global COVID-19 registries of individuals with rheumatic diseases, inflammatory bowel disease (IBD), and psoriasis to compare the association between tumor necrosis factor inhibitor (TNFi) monotherapy and COVID-19-related hospitalization or death among individuals with IMIDs, with other commonly prescribed immunomodulatory regimens. Data from the pooled analysis found that TNFi monotherapy was associated with fewer hospitalizations or deaths compared with other immunomodulatory regimens including methotrexate (MTX), azathioprine/mercaptopurine (AZA/6MP), and janus kinase inhibitors (JAKi). In addition, TNFi combination therapy was associated with more favorable outcomes when MTX was used instead of AZA/6MP. These findings support the continued use of TNFi monotherapy during the pandemic and suggest that clinicians should weigh the risks versus benefits of de-escalating treatment or changing medications when a patient is receiving concomitant TNFi and AZA/6MP. The second chapter describes the development and evaluation of a prediction model for COVID-19 ARDS in people with rheumatic diseases and the development of a simple risk-score calculator for use in clinical settings. The prediction model was developed using a series of supervised machine learning algorithms and information that can be easily obtained at COVID-19 exposure or onset and predicted ARDS with good discrimination in the test set and in external validation sets. Age, daily glucocorticoid dose, pulmonary hypertension, interstitial lung disease, chronic kidney disease, anti-CD20 monoclonal antibody use, diabetes, hypertension, active rheumatic disease, and morbid obesity were identified as the most influential factors in predicting the onset of ARDS. A simple and interpretable regression-based risk-score calculator also predicted ARDS with good discrimination in the test set and in external validation sets. The risk-score calculator has the potential to guide risk-stratification and the treatment of COVID-19 among people with rheumatic diseases during the pandemic. The third chapter links data from the GRA registry to a robust array of country-level factors and uses a novel methodological approach to investigate potential mechanisms of the disparate impact of COVID-19 on people with rheumatic diseases, globally. Data from this analysis indicated that a range of factors related to geographical residence impacted COVID-19 outcomes independent of known patient-level demographic and clinical risk factors. Namely, lower country socioeconomic status, environmental exposures, higher demands on or lower capacity of health resources, and fewer government-imposed containment measures were independently associated with COVID-19-related death after controlling for patient demographics and clinical characteristics. These findings highlight the importance of environmental and societal factors as potential explanations of the observed global health disparities during the pandemic. The study designs and analytical methods utilized in this work seek to quantify and address important biases inherent to real-world conveniently sampled data. Together, findings from the three chapters provide important evidence to advance rheumatology care in the COVID-19 pandemic and lay foundation for a new research agenda to address regional disparities in COVID-19 outcomes in people with rheumatic diseases.