The effectiveness of the neonatal diagnosis-related group scheme

Abstract

The goal of this study is to investigate the effectiveness of the neonatal diagnosis-related group scheme in patients affected by respiratory distress syndrome. The variable costs of individual patients in the same group are examined. This study uses the data of infants (N = 243) hospitalized in the Neonatal Intensive Care Unit of the Gaslini Children's Hospital in Italy in 2016. The care unit's operating and management costs are employed to estimate the average cost per patient. Operating costs include those related to personnel, drugs, medical supplies, treatment tools, examinations, radiology, and laboratory services. Management costs relate to administration, maintenance, and depreciation cost of medical equipment. Cluster analysis and Tobit regression are employed, allowing for the assessment of the total cost per patient per day taking into account the main cost determinants: birth weight, gestational age, and discharge status. The findings highlight great variability in the costs for patients in the same diagnosis-related group, ranging from a minimum of €267 to a maximum of €265,669. This suggests the inefficiency of the diagnosis-related group system. Patients with very low birth weight incurred costs approximately twice the reimbursement set by the policy; a loss of €36,420 is estimated for every surviving baby with a birth weight lower than 1,170 grams. On the contrary, at term, newborns cost about €20,000 less than the diagnosis-related group reimbursement. The actual system benefits hospitals that mainly treat term infants with respiratory distress syndrome and penalizes hospitals taking care of very low birth weight patients. As a result, strategic behavior and "up-coding" might occur. We conduct a cluster analysis that suggests a birth weight adjustment to determine new fees that would be fairer than the current costs.

Fig 1. Sample variability.

Fig 2. Cluster analysis.

Fig 3. Within group variability.