COVID-19 in patients with hepatobiliary and pancreatic diseases: A single-centre cross-sectional study in East London

Objective

To explore risk factors associated with COVID-19 susceptibility and survival in patients with pre-existing hepato-pancreato-biliary (HPB) conditions.

Design

Cross-sectional study.

Setting

East London Pancreatic Cancer Epidemiology (EL-PaC-Epidem) study at Barts Health NHS Trust, UK. Linked electronic health records were interrogated on a cohort of participants (age ≥ 18 years), reported with HPB conditions between 1 April 2008 and 6 March 2020.

Participants

EL-PaC-Epidem study participants, alive on 12 February 2020, and living in East London within the previous six months (n=15 440). The cohort represents a multi-ethnic population with 51.7% belonging to the non-White background.

Main outcome measure

COVID-19 incidence and mortality.

Results

Some 226 (1.5%) participants had confirmed COVID-19 diagnosis between 12 February and 12 June 2020, with an increased odds for men (OR 1.56; 95% CI 1.2 to 2.04) and Black ethnicity (2.04; 1.39 to 2.95) as well as patients with moderate to severe liver disease (2.2; 1.35 to 3.59). Each additional comorbidity increased the odds of infection by 62%. Substance mis-users were at more risk of infection, so were patients on Vitamin D treatment. The higher odds ratios in patients with chronic pancreatic or mild liver conditions, age>70, and history of smoking or obesity were due to co-existing comorbidities. Increased odds of death were observed for men (3.54; 1.68 to 7.85) and Black ethnicity (3.77; 1.38 to 10.7). Patients having respiratory complications from COVID-19 without a history of chronic respiratory disease also had higher odds of death (5.77; 1.75 to 19).

Conclusions

In this large population-based study of HPB patients, men, Black ethnicity, pre-existing moderate to severe liver conditions, six common medical multi-morbidities, substance mis-use, and a history of Vitamin D treatment independently posed higher odds of acquiring COVID-19 compared to their respective counterparts. The odds of death were significantly high for men and Black people.

STRENGTHS AND LIMITATIONS OF THIS STUDY

• First multi-ethnic population-based study on COVID-19 in patients with hepato-pancreato-biliary group of diseases.

• Systematic identification of the effect, or the lack of it, of individual demographic and clinical factors on the infection and mortality of COVID-19 in a large cohort of over 15 000 patients, robustly controlling for potential confounders in their evaluation.

• Access to longitudinal data from linked primary and secondary care electronic health records, and use of rule-based phenotyping algorithms allowed for improved completeness and accuracy of the explored variables.

• Some observed increased odds of SARS-CoV-2 infection and related death could be plausibly explained by unmeasured confounding.

• The effects reported in the study could be influenced by the relatively smaller size of COVID-19 cases within this cohort.