COVID-19 in patients with autoimmune diseases: characteristics and outcomes in a multinational network of cohorts across three countries

Eng Hooi Tan, Anthony G. Sena, Albert Prats‐Uribe, Seng Chan You, Waheed‐Ul‐Rahman Ahmed, Kristin Kostka, Christian Reich, Scott L. DuVall, Kristine E. Lynch, Michael E. Matheny, Talita Duarte‐Salles, Sergio Fernández‐Bertolín, George Hripcsak, Karthik Natarajan, Thomas Falconer, Henry M. Spotnitz, Anna Ostropolets, Clair Blacketer, Thamir M. Alshammari, Heba Alghoul, Osaid Alser, Jennifer C. E. Lane, Dalia Dawoud, Karishma Shah, Yue Yang, Lin Zhang, Carlos Areia, Asieh Golozar, Martina Recalde, Paula Casajust, Jitendra Jonnagaddala, Vignesh Subbian, David Vizcaya, Lana Yin Hui Lai, Fredrik Nyberg, Daniel R. Morales, Jose Posada, Nigam H. Shah, Mengchun Gong, Arani Vivekanantham, Aaron Abend, Evan Minty, Marc A. Suchard, Peter R. Rijnbeek, Patrick Ryan, Daniel Prieto‐AlhambraLara 2021 · DOI

TL;DR

This multinational study analyzed electronic health records from over 133,000 patients with autoimmune diseases diagnosed with COVID-19, revealing that COVID-19 is a significantly more severe disease than influenza in this population, leading to increased complications and mortality. The findings underscore the vulnerability of individuals with autoimmune conditions to severe COVID-19 outcomes.

What Problem Does This Paper Address?

The COVID-19 pandemic presented a significant challenge for individuals with autoimmune diseases. Public health advisories recommended shielding for these patients due to concerns about increased risk of infection and severe complications, potentially exacerbated by their underlying conditions and immunosuppressive therapies. However, there was a critical lack of robust data characterizing the specific risks, clinical presentations, and outcomes for this vulnerable group. Existing studies were often limited in sample size and geographic scope. This paper aims to fill that knowledge gap by providing a comprehensive, multinational analysis of COVID-19 characteristics and outcomes in patients with prevalent autoimmune diseases, comparing them to similar patients hospitalized with influenza.

Methods

The study employed a multinational network retrospective cohort design, leveraging electronic health records from multiple data sources across the USA, Spain, and South Korea. A key methodological innovation was the use of the Observational Medical Outcomes Partnership (OMOP) Common Data Model, which standardized data from disparate sources, enabling a federated network analysis without direct patient-level data sharing. This approach allowed for the aggregation of data from six distinct databases, including hospital electronic health records and claims data. Patients were included if they had a prevalent autoimmune disease and a clinical or laboratory-confirmed diagnosis of COVID-19 (January-June 2020) or influenza (September 2017-April 2018). The primary outcomes assessed were death and complications within 30 days of hospitalization.

Datasets

The study utilized data from six electronic health record databases:

  • Columbia University Irving Medical Center (CUIMC) - USA
  • Health Insurance Review and Assessment (HIRA) - South Korea
  • IQVIA Open Claims - USA
  • Optum EHR - USA
  • Information System for Research in Primary Care-Hospitalization Linked Data (SIDIAP-H) - Spain
  • Department of Veterans Affairs (VA-OMOP) - USA

These databases provided a broad geographic and demographic representation, allowing for a multinational comparison.

Key Insights and Ideas

  • Vulnerability of Autoimmune Patients: The study confirms that patients with autoimmune diseases are a vulnerable population for COVID-19, experiencing more severe outcomes compared to influenza.
  • Comorbidity Burden: Patients with autoimmune diseases, particularly those hospitalized with COVID-19, frequently presented with significant comorbidities like hypertension, chronic kidney disease, and heart disease, which are known risk factors for severe COVID-19.
  • Respiratory and Cardiac Complications: COVID-19 in this cohort led to a high incidence of severe respiratory complications (ARDS, pneumonia) and cardiac issues, mirroring findings in broader COVID-19 populations but with potentially higher frequency.
  • Thromboembolism Risk: The observed higher incidence of venous thromboembolic events (VTEs) in COVID-19 patients with autoimmune diseases, compared to influenza patients, highlights a potential synergistic effect of the disease and the underlying autoimmune condition on coagulation pathways.
  • Data Standardization is Key: The use of the OMOP Common Data Model was crucial for harmonizing data from diverse sources, enabling a large-scale, multinational analysis that would be challenging otherwise.

Key Results

  • The study analyzed data from 133,589 patients diagnosed and 48,418 hospitalized with COVID-19 who had prevalent autoimmune diseases.
  • Compared to patients hospitalized with influenza, those hospitalized with COVID-19 experienced more respiratory complications (pneumonia, ARDS) and had higher 30-day mortality (ranging from 6.3-24.6% for COVID-19 vs. 2.2-4.3% for influenza across different databases).
  • Prevalent comorbidities in hospitalized COVID-19 patients included hypertension (45.5-93.2%), heart disease (29.0-83.8%), and Type 2 diabetes (30.8-74.3%).
  • Severe outcomes within 30 days post-hospitalization were frequent, with ARDS (2.1-42.8%) and pneumonia (12.6-53.2%) being the most common.
  • Mortality rates were generally higher in US and Spanish databases (16.3-24.6%) compared to South Korea (6.3%).

Limitations and Future Work

The authors acknowledge several limitations:

  • Data Completeness and Accuracy: While efforts were made to standardize data, potential misclassifications, underestimation of conditions, and variations in coding practices across databases exist. Claims data may not always reflect actual diagnoses.
  • Hospitalization Rate vs. Severity: The higher hospitalization rate for COVID-19 compared to influenza might not solely reflect disease severity but could also be influenced by quarantine measures or increased monitoring for COVID-19 patients.
  • Underestimation of Mortality: Mortality data capture can vary by database; inpatient deaths are recorded, but deaths after discharge may not be.
  • Testing Limitations: In the early stages of the pandemic, limited testing may have led to underdiagnosis, particularly for less severe cases.

Future work suggested includes investigating specific predictors of poor outcomes in COVID-19 patients with autoimmune diseases to better inform clinical management and risk stratification.

Summary updated

Abstract

Compared with influenza, COVID-19 is a more severe disease, leading to more complications and higher mortality.
BibTeX
@inproceedings{tan2021covid,
  author = {Eng Hooi Tan and Anthony G. Sena and Albert Prats‐Uribe and Seng Chan You and Waheed‐Ul‐Rahman Ahmed and Kristin Kostka and Christian Reich and Scott L. DuVall and Kristine E. Lynch and Michael E. Matheny and Talita Duarte‐Salles and Sergio Fernández‐Bertolín and George Hripcsak and Karthik Natarajan and Thomas Falconer and Henry M. Spotnitz and Anna Ostropolets and Clair Blacketer and Thamir M. Alshammari and Heba Alghoul and Osaid Alser and Jennifer C. E. Lane and Dalia Dawoud and Karishma Shah and Yue Yang and Lin Zhang and Carlos Areia and Asieh Golozar and Martina Recalde and Paula Casajust and Jitendra Jonnagaddala and Vignesh Subbian and David Vizcaya and Lana Yin Hui Lai and Fredrik Nyberg and Daniel R. Morales and Jose Posada and Nigam H. Shah and Mengchun Gong and Arani Vivekanantham and Aaron Abend and Evan Minty and Marc A. Suchard and Peter R. Rijnbeek and Patrick Ryan and Daniel Prieto‐Alhambra},
  title = {{COVID-19 in patients with autoimmune diseases: characteristics and outcomes in a multinational network of cohorts across three countries}},
  booktitle = {Lara},
  year = {2021},
  doi = {10.1093/rheumatology/keab250},
}
Figure