Effects of vitamin D on COVID-19 risk and hospitalisation in the UK biobank

Abstract

Background: Vitamin D (VitD) plays an important role in immune modulation. VitD deficiency is associated with increased susceptibility to acute respiratory syndrome as observed in COVID-19. We evaluated potential associations between serum VitD levels and risk of COVID-19 infection and hospitalisation, within the overall and cancer populations.

Methods: We performed a nested case-control study within the UK biobank cohort, among all individuals with at least one serum VitD level measurement at baseline (2006-2010) and a COVID-19 polymerase chain reaction (PCR) results recorded, and individuals with previous cancer diagnosis. Binary multivariable logistic regression was performed to assess associations between VitD levels and risk of COVID-19 infection (positive PCR), and hospitalisation (COVID-19-positive PCR in hospital), and stratified by ethnicity.

Results: Of 151,543 participants, 21,396 tested positive for COVID-19. Of 24,400 individuals with cancer, 2,608 tested positive. In the total cohort, VitD insufficiency (Adjusted Odds Ratio (aOR) 0.97, 95% Confidence Interval (CI) 0.94-1.00) and deficiency (aOR 0.95, 95%CI 0.90-0.99) were associated with slightly lower odds of COVID-19 infection. In contrast, both VitD insufficiency (aOR 1.19, 95%CI 1.08-1.31) and deficiency (aOR 1.36, 95%CI 1.19-1.56) were associated with higher odds of COVID-19 hospitalisation. Among Asian (aOR 1.50; 95%CI 1.08-2.07) and Black (aOR 1.57; 95%CI 1.14-2.16) participants, VitD deficiency was associated with higher odds of COVID-19 infection. Among White participants, VitD insufficiency was associated with slightly lower odds of COVID-19 infection (aOR 0.97; 95%CI 0.86-0.95), while both VitD insufficiency (aOR 1.19; 95%CI 1.08-1.32) and deficiency (aOR 1.44; 95%CI 1.25-1.66) were associated with increased odds of hospitalisation. In the cancer population, vitamin D deficiency was associated with higher odds of infection only among Black participants (aOR 3.50; 95%CI 1.22-10.01); no other associations were observed.

Conclusions: Low VitD levels were associated with an increased risk of COVID-19 hospitalisation but showed only a weak association with infection risk. Black and Asian populations had higher infection risk associated with VitD deficiency, but this did not translate to increased hospitalisation. In contrast, White populations with low VitD levels exhibited a higher risk of hospitalisation. There was no evidence of an interaction between VitD levels and ethnicity affecting infection or hospitalisation risk. In the cancer cohort, no significant associations were observed for COVID-19 infection or hospitalisation.