Table 2.
Correlation of vitamin D concentrations with COVID-19 infections and outcomes in recent studies.
| Reference | Country | n | Population type | Study design | Vit D doses | Outcomes |
|---|---|---|---|---|---|---|
| Tan et al., 2020 [62] | Singapore (a tertiary academic hospital) | 43 | Adults, age ≥50 yrs | Cohort observational (15 Jan–15 April 2020) |
Vitamin D 1000 IU, Mg 150 mg, and vitamin B12 500 μg (oral) | (i) A fewer patients who received vitamin D, Mg and vitamin B12 required subsequent oxygen therapy compared to controls (3/17 vs. 16/26, p = 0.006) (ii) In multivariate analysis, patients treatment with vitamin D, Mg and vitamin B12 showed a significant protective effects against clinical deterioration (p = 0.041) after adjusting for age, gender and comorbidities |
| Present study | 20 European countries | Cases and death/1 M population | Adults | Retrospective (as of 20 May 2020) | NA | A significant negative correlation was observed for levels of mean vitamin D with COVID-19 cases (p = 0.033) but not with death (p = 0.123) per million of population |
| IIie et al., 2020 [19] | 20 European countries | Cases and death/1 M population | Adults | Retrospective (as of 8 April 2020) | NA | A negative correlation was observed between levels of mean vitamin D and COVID-19 cases (p = 0.050) and death (p = 0.053) per million of population |
| Alipio 2020 [61] | Southern Asian countries (three hospitals) |
222 | NA | Retrospective multicentre study | NA | (i) The differences in the mean levels of vitamin D were significant within the mild, ordinary, severe and critical cases of COVID-19 (p < 0.001) (ii) Vitamin D status showed a significant association with clinical outcomes (p < 0.001) |
| Lau et al., 2020 [20] | USA (a single tertiary academic medical center) | 20 | Adults, mean age 65.2 yrs | Retrospective observational study (27 March–21 April 2020) |
NA | A high vitamin D insufficiency was observed in ICU patients (84.6%) than in the floor patients (57.1%) (p = 0.29) |
| Glicio et al., 2020 [17] | South Asia (two tertiary medical centers) | 176 | Adults, age ≥60 yrs | Retrospective (as of 5 May 2020) |
NA | (i) Severe patients had a low level of vitamin D than mild patients (ii) Subjects with pre-existing medical conditions had a low level of vitamin D |
| Hastie et al., 2020 [18] | UK (UK Biobank data 2006–2010 for vitamin D and ethnicity) | 449 | Adults, age 37–73 yrs | Cross-sectional (16 March–14 April 2020) | NA | (i) Vitamin D levels showed a significant association with COVID-19 infection in univariate analysis (p = 0.013) but not after adjustment for confounders (p = 0.208) (ii) Ethnicity showed a significant association with COVID-19 infection univariably |
| Darling et al., 2020 [15] | UK (UK Biobank data 2006–2010 for BMI, vitamin D and ethnicity) | 580 cases and 723 control | Adults, mean age 57.7 yrs | Retrospective | NA | (i) No significant difference was observed for vitamin D levels between COVID-19 cases and control group (ii) Vitamin D status was significantly lower in those of Asian, Black and Mixed ethnicity (p < 0.0010) compared with those of White ethnicity (iii) Vitamin D levels were significantly lower in those with obesity (p < 0.001). Overweight or obese person; living in London; being male and being of Asian, Black or Mixed ethnicity was associated with a higher odd of positive cases (iv) In regression model, the interaction between BMI and vitamin D status did not predict test result in the available data set |
| Li et al., 2020 [21] | Mainland of USA (48 states and Columbia district) |
– | 1,609,488 cases and 91,094 deaths | Retrospective (22 Jan–23 May 2020) | NA | (i) Latitudes were marginally associated with cases (p = 0.0792) and deaths (p = 0.0599) (ii) Sunlight and vitamin D, with latitude as an indicator, possibly associated with reduced risks for both COVID-19 cases and mortality |
| De Smet et al., 2020 [16] | Belgium (Central network hospital) | 186 cases, 2717 controls | Adults, median age 71 yrs (cases), 68 yrs (control) | Retrospective observational (1 March–7 April 2020) | NA | (i) Patients with COVID-19 had significantly a low median value of vitamin D and higher vitamin D deficiency compared to control subjects (p = 0.0016, p = 0.0005, respectively) (ii) This difference were more pronounced in male COVID-19 subjects than male control subjects that increased with advancing radiological stage and were not confounded vit D-impacted comorbidities |
| Daneshkhah et al., 2020 [14] | Hospitals and clinics from different parts of the world | 5000 cases | Age up to 80 yrs | As of March 21, 2020 | NA | About 15% reduction in the number of severe COVID-19 cases given a normal vitamin D status within a population |
| Raharusuna et al., 2020 [22] | Indonesia (Government hospital) | 780 cases | Adults, mean age 54.5 yrs | Retrospective cohort study (2 March 2–24 April 2020) | NA | (i) In univariate analysis, older and male cases with pre-existing medical condition and below normal vitamin D levels were associated with higher odds of death (ii) After adjustment of confounders (age, sex and comorbidity), vitamin D levels showed a strong relationship with COVID-19 mortality |