Economic BurdeE Associated With Hospital Postadmission Dehydration

摘要

Background: Development of dehydration after hospital admission can be a measure of quality care, but evidence describing the incidence, economic burden, and outcomes of dehydration in hospitalized patients is lacking. Objective: The objective of this study was to compare costs and resource utilization of U.S. patients experiencing postadmission dehydration (PAD) with those who do not in a hospital setting. Methods: All adult inpatient discharges, excluding those with suspected dehydration present on admission (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM] codes for dehydration: 276.0,276.1,276.5), were identified from the Premier database using ICD-9-CM codes. PAD and no-PAD (NPAD) groups were matched on propensity score adjusting for demographics (age, sex, race, medical, elective patients), patient severity (All Patient Refined Diagnosis-Related Groups severity scores), and hospital characteristics (geographic location, bed size, teaching and urban hospital). Costs, length of stay (LOS), and incidence of mortality and catheter-associated urinary tract infection (CAUTI) were compared between groups using the ttest for continuous variables and the x2 test for categorical variables. Results: In total, 86,398 (2.1%) of all the selected patients experienced PAD. Postmatching mean total costs were significantly higher for the PAD group compared with the NPAD group ($33,945 vs $22,380; P < .0001). Departmental costs were also significantly higher for the PAD group (all P < .0001). Compared with the NPAD group, the PAD group had a higher mean LOS (12.9 vs 8.2 days), a higher incidence of CAUTI (0.6% vs 0.5%), and higher in-hospital mortality (8.6% vs 7.8%) (all P < .05). The results for subgroup analysis also showed significantly higher total cost and longer LOS days for patients with PAD (all P < .05). Conclusions: The economic burden associated with hospital PAD in medical and surgical patients was substantial.