192 



the case of a baby exposed to cocaine, the average length of stay was eleven (1 1) days 

 longer than for one without this exposure. To estimate the incremental days attributable 

 to drugs, the total number of Medicaid births involving maternal cocaine use (8% of all 

 births) was multiplied by 11 days. 



4) Additional Days for Medical Treatment Due to Substance Abuse Complications - In 

 addition to being a risk factor for getting certain Ulnesses, active substance abuse at the 

 time of hospitalization can also complicate an illness and add to the patient's length of 

 stay. For example, substance abuse can compromise the inmiune system, reducing the 

 body's ability to fight infection or some substance abuse problems (e.g., delirium 

 tremens) need to be stabilized before doctors can treat the primary medical condition. To 

 estimate the cost of substance abuse comorbidity, we computed the difference in length 

 of stay between those discharges with the same primary diagnosis with and without 

 substance abuse as a secondary diagnoses, controlling for age and sex. The total number 

 of incremental days identified in this way were counted as substance abuse-related 

 Medicaid days.* 



For each of these four categories, we estimated 1991 costs by multiplying the identified 

 substance abuse-related days by an average hospital inpatient per diem cost of $750. This per 

 diem estimate was based on 1990 Medicaid costs per day inflated by the hospital component of 



' With respect to this fourth category, our analysis understates the impact of substance abuse 

 comorbidity due to limitations of medical reporting (See Underestimation Issues). 



-17- 



