ObjectiveThis retrospective cohort identified the association of human serum albumin (HSA) with adverse outcomes (septic shock, in-hospital and out-of-hospital mortality) in elderly hospitalized patients who have community-acquired pneumonia (CAP) and specific body mass index (BMI).Materials and methodsThis research included hospitalized CAP individuals (≥ 60 years) and was conducted at a teaching hospital in western China. All the patients were categorized into three populations based on two BMI cutoff values (18.5 kg/m2 and 24 kg/m2). The data was acquired from medical records, local government mortality databases, and telephone interviews. Binomial logistic regression analysis was used to explore the associations between low HSA and septic shock and in-hospital mortality, and Cox regression analysis was used to explore the association between low HSA and out-of-hospital mortality.ResultsA total of 627 patients were included in the analysis of in-hospital death and septic shock, and 431 patients were included in the analysis of out-of-hospital death. The study showed that 120 elderly patients with CAP (19.14%) died in the hospital, while 141 patients (32.71%) died out of the hospital, and 93 patients (14.83%) developed septic shock. No differences in in-hospital and out-of-hospital mortality were observed for BMI values < 18.5 kg/m2 or BMI ≥ 24 kg/m2, regardless of whether HSA was ≥ 40 g/l or < 40 g/l. When 18.5 kg/m2 ≤ BMI < 24 kg/m2, patients with HSA < 40 g/l had both higher in-hospital and out-of-hospital mortality compared with those with HSA ≥ 40 g/l (in-hospital death: 26.13% vs. 11.46%, p < 0.001; out-of-hospital death: 46.15% vs. 19.17%, p < 0.001). No significant differences were observed in the incidence of septic shock between patients with HSA < 40 g/l and those with HSA ≥ 40 g/l either in the overall population or when the BMI values were divided according to the cutoff values of 18.5 kg/m2 and 24 kg/m2. After further logistic regression analysis and adjustment for potential confounders, the results showed that when 18.5 kg/m2 ≤ BMI < 24 kg/m2, elderly CAP patients with HSA < 40 g/l had a higher risk of in-hospital and out-of-hospital mortality compared with those with HSA ≥ 40 g/l (in-hospital death: HR = 1.964, 95%CI = 1.08–3.573; out-of-hospital death: HR = 2.841, 95%CI = 1.745–4.627).ConclusionsHSA levels can predict the risk of in-hospital and out-of-hospital mortality in elderly patients with CAP and normal BMI values. However, HSA cannot predict the risk of septic shock in elderly patients hospitalized with CAP, irrespective of their BMI classification.