Long hospitalisation and high hospital charges of premature infants are important concerns for parents and health providers. In this study we determined the length of stay and hospital cost of preterm babies and assessed the predictive value of perinatal factors on the duration and the cost of hospitalisation. We reviewed hospital charges and length of stay for a retrospective cohort of 72 premature infants discharged from the neonatal intensive care unit of Marmara University Hospital between January 1999 and January 2000. The perinatal factors were analysed using forward stepwise multiple linear regression analysis. The median gestational age and birth weight of the study group was 33 (28-37) weeks and 1560 (814-2340) grams respectively. Median length of hospital stay was 19 (4-93) days whereas hospital charges were 4345 (750-23217). The hospital cost per day of hospitalisation was 250 (57-350). We developed a mathematical model to predict length of stay and hospital charges from perinatal information. The equation predicts length of stay within five days in 91% of the cases and has an R2 of 0.83. Birth weight and presence of respiratory distress requiring mechanical ventilation and bronchopulmonary dysplasia are the most powerful predictive factors in both the duration of hospital stay and hospital costs. In addition presence of intraventricular haemorrhage (IVH) is also an important factor for determining hospital costs. The equation for hospital charges predicts the costs within 1475 in 92% of the cases and has an R2 of 0.85. This information is helpful in counselling parents to provide an estimate for the duration and cost of hospitalisation as well as introducing quantitative information for health planning and allocation of resources in the country.

Long hospitalisation and high hospital charges of premature infants are important concerns for parents and health providers. In this study we determined the length of stay and hospital cost of preterm babies and assessed the predictive value of perinatal factors on the duration and the cost of hospitalisation. We reviewed hospital charges and length of stay for a retrospective cohort of 72 premature infants discharged from the neonatal intensive care unit of Marmara University Hospital between January 1999 and January 2000. The perinatal factors were analysed using forward stepwise multiple linear regression analysis. The median gestational age and birth weight of the study group was 33 (28-37) weeks and 1560 (814-2340) grams respectively. Median length of hospital stay was 19 (4-93) days whereas hospital charges were 4345 (750-23217). The hospital cost per day of hospitalisation was 250 (57-350). We developed a mathematical model to predict length of stay and hospital charges from perinatal information. The equation predicts length of stay within five days in 91% of the cases and has an R2 of 0.83. Birth weight and presence of respiratory distress requiring mechanical ventilation and bronchopulmonary dysplasia are the most powerful predictive factors in both the duration of hospital stay and hospital costs. In addition presence of intraventricular haemorrhage (IVH) is also an important factor for determining hospital costs. The equation for hospital charges predicts the costs within 1475 in 92% of the cases and has an R2 of 0.85. This information is helpful in counselling parents to provide an estimate for the duration and cost of hospitalisation as well as introducing quantitative information for health planning and allocation of resources in the country.