Background: Much research shows that blood calcidiol (25(OH)D3) levels correlate strongly with SARS-CoV-2 infection severity. There is open discussion regarding whether low D3 is caused by the infection or if deficiency negatively affects immune defense. The aim of this study was to collect further evidence on this topic. Methods: Systematic literature search was performed to identify retrospective cohort as well as clinical studies on COVID-19 mortality rates versus D3 blood levels. Mortality rates from clinical studies were corrected for age, sex, and diabetes. Data were analyzed using correlation and linear regression. Results: One population study and seven clinical studies were identified, which reported D3 blood levels preinfection or on the day of hospital admission. The two independent datasets showed a negative Pearson correlation of D3 levels and mortality risk (r(17) = −0.4154, p = 0.0770/r(13) = −0.4886, p = 0.0646). For the combined data, median (IQR) D3 levels were 23.2 ng/mL (17.4–26.8), and a significant Pearson correlation was observed (r(32) = −0.3989, p = 0.0194). Regression suggested a theoretical point of zero mortality at approximately 50 ng/mL D3. Conclusions: The datasets provide strong evidence that low D3 is a predictor rather than just a side effect of the infection. Despite ongoing vaccinations, we recommend raising serum 25(OH)D levels to above 50 ng/mL to prevent or mitigate new outbreaks due to escape mutations or decreasing antibody activity.

Background Much research shows that blood calcidiol (25(OH)D3) levels correlate strongly with SARS-CoV-2 infection severity. There is open discussion regarding whether low D3 is caused by the infection or if deficiency negatively affects immune defense. The aim of this study was to collect further evidence on this topic. Methods Systematic literature search was performed to identify retrospective cohort as well as clinical studies on COVID-19 mortality rates vs. D3 blood levels. Mortality rates from clinical studies were corrected for age, sex and diabetes. Data was analyzed using correlation and linear regression. Results One population study and seven clinical studies were identified, which reported D3 blood levels pre-infection or on the day of hospital admission. They independently showed a negative Pearson correlation of D3 levels and mortality risk (r(17)=-.4154, p=.0770/r(13)=-.4886, p=.0646). For the combined data, median (IQR) D3 levels were 23.2 ng/ml (17.4 – 26.8), and a significant Pearson correlation was observed (r(32)=-.3989, p=.0194). Regression suggested a theoretical point of zero mortality at approximately 50 ng/ml D3. Conclusions The two datasets provide strong evidence that low D3 is a predictor rather than a side effect of the infection. Despite ongoing vaccinations, we recommend raising serum 25(OH)D levels to above 50 ng/ml to prevent or mitigate new outbreaks due to escape mutations or decreasing antibody activity.