Information was provided on costs estimated in the study. However, hospital charge data rather than costs appear to have been used to account for the costs of CMV treatment. It is not clear whether these charge data were converted to costs. It would also have been helpful to have set a base price year rather than using actual cost and charge data which were collected over a period of nearly three years. Future analysis may also wish to consider costs to others in society, particularly given the significant difference in the mean length of therapy for patients between the two groups, and the lower number of inpatient days required for the intervention.