Efficient management of diarrhea in the acquired immunodeficiency syndrome (AIDS). A medical decision analysis

To compare the efficacy and cost effectiveness of alternative strategies for the evaluation and medical management of diarrhea in patients with the acquired immunodeficiency syndrome (AIDS).

Medical decision analysis using a decision tree.

Three management strategies were compared: full evaluation, which included stool culture, ova and parasite examination, stain for protozoa, blood cultures, esophagogastroduodenoscopy with biopsy, and colonoscopy with biopsy; limited evaluation, which included stool culture, ova and parasite examination, stool stain for protozoa, and blood cultures; and minimal evaluation, which included only a stool culture. Treatment was based on the outcome of the diagnostic workup. Patients without a specific diagnosis were treated symptomatically with diphenoxylate hydrochloride. Nonrespondents to initial symptomatic treatment and patients with recurrent diarrhea after specific therapy were given the full evaluation. Average probabilities of diagnostic and therapeutic success were calculated from previous reports. Costs were compiled based on diagnostic and therapeutic expenditures.

Under baseline assumptions, the remission rates for diarrhea were 75.2%, 74.8%, and 74.8% in patients undergoing the full, limited, and minimal evaluations, respectively. The respective costs of the three strategies were $5419, $1997, and $1700 per patient in remission. A sensitivity analysis showed that the differences in efficacy and cost among the full, limited, and minimal strategies remained unchanged over a broad range of possible rates of diagnostic or therapeutic success.

The minimal evaluation in all patients, with the full evaluation reserved only for nonrespondents to symptomatic treatment, is efficacious and is the most cost-effective strategy for managing AIDS-related diarrhea.