From the fall of 1997 through the spring of 1998, we reviewed the available medical records of all patients with RSV, confirmed by enzyme-linked immunoassay (ELISA) or viral culture, at Children’s Memorial Medical Center in Chicago. At the center, acute respiratory infections in all admitted infants and children of all ages require a rapid test for RSV; these infections include the clinical diagnoses of bronchiolitis, bronchitis, cold, croup, asthma with fever, influenza, and viral pneumonia.

The patients were divided into three groups: children admitted to the ward only (ward), children admitted to the ward and the pediatric intensive-care unit (PICU) (ward + PICU), and children admitted to the PICU who also required mechanical ventilation (ward + PICU + ventilation). Children whose RSV infection was judged to play no role in their hos-pitalization were excluded from analysis (see Results), and children with a known risk factor for RSV infection were analyzed separately.

Total hospital charges, PICU expenses, and ventilator costs—excluding physician charges—were ascertained for each patient from the Business Development Office. If RSV was not both the admitting and the discharge diagnosis, we determined what role, if any, RSV played in that child’s hospitalization and we used only that part of the cost attributable to RSV. For example, if a hospitalized child’s condition worsened and subsequent testing confirmed a newly acquired RSV infection, we used only the amount of time required to return to the baseline state to calculate the cost of the patient’s RSV-related hospitalization. Data abstracted from the charts for each patient included age, sex, race, gestational age, type of insurance (private or public aid), and length of stay (or the portion of the length of stay attributable to RSV).
canadian pharmacy online

We calculated the mean charge per case, the mean cost per case (with a cost/charge ratio of 0.67 at Children’s Memorial), and the mean length of stay for each group. We then calculated the weighted mean charge, cost, and length of stay for each group. Because of the differences in the size of each subgroup, we used weighted averages. For instance, multiplying the mean charge for each group by the percentage of patients in the study from that group and adding the results yields the weighted mean charge. We used public aid as a crude surrogate for socioeconomic status.