Serum albumin levels have been associated with prognosis in several conditions among older adults. The aim of this study is to assess the prognostic value in mortality of serum albumin in older adults with SARS-CoV-2 infection. Methods. Cohort observational study with consecutive older-adults (≥65 years old), with confirmed SARS-CoV-2 infection admitted to a university hospital between March–May 2020. A logistic regression model was fitted to assess the impact of albumin levels on in-hospital mortality adjusted by potential confounders. Results. Among a total of 840 patients admitted to the hospital, 405 (48%) were older adults with a total of 92 deaths (23%) among them. Those who died were older, had more comorbidities, higher inflammation status and lower levels of serum albumin at admission [3.10 g/dL (0.51) vs. 3.45 g/dL (0.45); p < 0.01. Serum albumin levels at admission were negatively correlated with inflammatory markers such as C-Reactive protein (Pearson Coeff −0.4634; p < 0.001) or IL-6 (Pearson’s Coeff −0.244; p = 0.006) at admission but also to other clinical outcomes such time to clinical stability (Pearson’s Coeff −0.259; p < 0.001). Severe hypoalbuminemia associated with increased risk of mortality was defined as ≤3 g/dL at admission according to the AUC/ROC analysis (0.72 95% CI 0.63–0.81) In a multivariate logistic regression model adjusting by age, inflammation, comorbidities and severity at admission severe hypoalbuminemia was a strong predictor of in-hospital mortality (OR 2.18 95% CI 1.03–4.62; p = 0.039). Conclusion. Severe hypoalbuminemia with ≤3 g/dL is an independent risk factor for mortality among older adults with SARS-CoV-2 infection. There is a consistent correlation between albumin levels and inflammatory biomarkers. Further studies are needed to determine whether the supplementation of albumin as coadjuvant treatment will have a positive impact on the prognosis of this infection.