Abstract: Objective: To analyse which variables associated with ICU admission for COVID-19 were linked to higher hospital costs according to the APR-DRG classification. Design: Retrospective, observational, and analytical study. Setting: COVID-19 ICU in a tertiary hospital. Patients: Adults (>18 years) with a confirmed diagnosis of SARS-CoV-2 infection. Interventions: Predictive models using multiple logistic regression. Main variables of interest: Hospital cost, APR-DRG, mechanical ventilation. Results: A total of 799 patients were analyzed and categorized into tertiles based on hospital stay costs, resulting in three groups: 266 patients with lower costs (median ;6160 [p25: 3962 p75: 6160]), 314 with intermediate costs (median ;16,446 [p25: 10,653 p75: 18,274]), and 219 with higher costs (median ;26,085 [p25: 26,085-p75: 51,523]). The best predictive model, with an AIC of 490.09 and an R2 of 0.32, identified the following factors as significantly associated with higher hospital costs: ICU length of stay (OR: 1.05; 95% CI: 1.03-1.07; p<0.01), development of VAT/VAP (OR: 4.72; 95% CI: 2.83-7.85; p<0.01), OXA-48 infection (OR: 2.65; 95% CI: 1.25-5.61; p=0.01), pulmonary embolism (OR: 6.42; 95% CI: 2.17-19.26; p<0.01), smoking history (OR: 2.22; 95% CI: 1.49-3.74; p<0.01), and vasopressor requirement (OR: 1.79; 95% CI: 1.22-2.86; p=0.01). The area under the curve (AUC) was 0.866 (p<0.01). Conclusions: Prolonged ICU stay, infectious and thromboembolic complications, smoking history, and vasopressor requirement were significantly associated with higher hospital costs.

Authorship

Download reference