Context. Calcifediol has been proposed as a potential treatment for COVID-19 patients. Objective: To compare the administration or not of oral calcifediol on mortality risk of patients hospitalized because of COVID-19. Design: Retrospective, multicenter, open, non-randomized cohort study. Settings: Hospitalized care. Patients: Patients with laboratory-confirmed COVID-19 between 5 February and 5 May 2020 in five hospitals in the South of Spain. Intervention: Patients received calcifediol (25-hydroxyvitamin D-3) treatment (0.266 mg/capsule, 2 capsules on entry and then one capsule on day 3, 7, 14, 21, and 28) or not. Main Outcome Measure: In-hospital mortality during the first 30 days after admission. Results: A total of 537 patients were hospitalized with COVID-19 (317 males (59%), median age, 70 years), and 79 (14.7%) received calcifediol treatment. Overall, in-hospital mortality during the first 30 days was 17.5%. The OR of death for patients receiving calcifediol (mortality rate of 5%) was 0.22 (95% CI, 0.08 to 0.61) compared to patients not receiving such treatment (mortality rate of 20%; p < 0.01). Patients who received calcifediol after admission were more likely than those not receiving treatment to have comorbidity and a lower rate of CURB-65 score for pneumonia severity >= 3 (one point for each of confusion, urea > 7 mmol/L, respiratory rate >= 30/min, systolic blood pressure < 90 mm Hg or diastolic blood pressure <= 60 mm Hg, and age >= 65 years), acute respiratory distress syndrome (moderate or severe), c-reactive protein, chronic kidney disease, and blood urea nitrogen. In a multivariable logistic regression model, adjusting for confounders, there were significant differences in mortality for patients receiving calcifediol compared with patients not receiving it (OR = 0.16 (95% CI 0.03 to 0.80). Conclusion: Among patients hospitalized with COVID-19, treatment with calcifediol, compared with those not receiving calcifediol, was significantly associated with lower in-hospital mortality during the first 30 days. The observational design and sample size may limit the interpretation of these findings.