BACKGROUND: A total lymphocyte count (TLC) of 1200 cells/mL has been used as a surrogate for a CD4 count of 200 cells/microL in resource-limited settings with varying results. We developed a more effective method based on a decision tree algorithm to classify subjects. METHODS: A decision tree was used to develop models with the variables TLC, hemoglobin, platelet count, gender, body mass index, and antiretroviral treatment status of subjects from the University of Alabama at Birmingham (UAB) observational database. Models were validated on data from the Birmingham Veterans Affairs Medical Center (BVAMC) and Zambia, with primary decision trees also generated from these data. RESULTS: A total of 1189 patients from the UAB observational database were included. The UAB decision tree classified a CD4 count < or =200 cells/microL as better than a TLC cut-point of 1200 cells/mL, based on the area under the curve of the receiver-operator characteristic curve (P < 0.0001). When applied to data from the BVAMC and Zambia, the UAB-based decision tree performed better than the TLC cut-point of 1200 cells/mL (BVAMC: P < 0.0001; Zambia: P = 0.0009) but worse than a decision tree based on local data (BVAMC: P < or = 0.0001; Zambia: P < or = 0.0001). CONCLUSION: A decision tree algorithm based on local data identifies low CD4 cell counts better than one developed from a different population or a TLC cut-point of 1200 cells/mL.

BACKGROUND: Health care expenditures for persons infected with human immunodeficiency virus (HIV) in the United State determined on the basis of actual health care use have not been reported in the era of highly active antiretroviral therapy. METHODS: Patients receiving primary care at the University of Alabama at Birmingham HIV clinic were included in the study. All encounters (except emergency room visits) that occurred within the University of Alabama at Birmingham Hospital System from 1 March 2000 to 1 March 2001 were analyzed. Medication expenditures were determined on the basis of 2001 average wholesale price. Hospitalization expenditures were determined on the basis of 2001 Medicare diagnostic related group reimbursement rates. Clinic expenditures were determined on the basis of 2001 Medicare current procedural terminology reimbursement rates. RESULTS: Among the 635 patients, total annual expenditures for patients with CD4+ cell counts <50 cells/microL (36,533 dollars per patient) were 2.6-times greater than total annual expenditures for patients with CD4+ cell counts > or =350 cells/microL (13,885 dollars per patient), primarily because of increased expenditures for nonantiretroviral medication and hospitalization. Expenditures for highly active antiretroviral therapy were relatively constant at approximately 10,500 dollars per patient per year across CD4+ cell count strata. Outpatient expenditures were 1558 dollars per patient per year; however, the clinic and physician component of these expenditures represented only 359 dollars per patient per year, or 2% of annual expenses. Health care expenditures for patients with HIV infection increased substantially for those with more-advanced disease and were driven predominantly by medication costs (which accounted for 71%-84% of annual expenses). CONCLUSIONS: Physician reimbursements, even with 100% billing and collections, are inadequate to support the activities of most clinics providing HIV care. These findings have important implications for the continued support of HIV treatment programs in the United States.

Treatment of HIV-infected patients with HAART can result in long-term suppression of viral loads to undetectable levels. Rapid virologic rebound typically follows treatment interruption (TI), with a potential for significant loss of CD4+ cells. Patients who maintain virologic suppression despite interrupting treatment have not been well described. All patients with a pretreatment viral load (VL) > or = 5000 copies/ml, who had been on therapy for > or = 2 weeks, and who underwent a TI lasting > or = 180 days were analyzed. Patients whose maximum VL did not exceed 5000 copies/ml > or = 6 months after starting TI ("nonrebounders") were compared with those whose VL exceeded 5000 copies/ml (rebounders). Seventy-one patients were included in the analysis. Nineteen (27%) were nonrebounders. Ninety-four percent of patients in each group interrupted treatment for reasons unrelated to virologic response. Median change in CD4 count during TI was not significantly different between the nonrebounder and rebounder groups (-20.5/microl vs. -64.0/microl; p < 0.086). In a multivariate logistic regression analysis, the following factors predicted nonrebounder status: peak VL before TI (log10 copies/ml) (OR = 0.14, 95% CI = 0.04-0.48, p = 0.0016); having received HAART (vs. mono/dual therapy) as initial regimen (OR: 11.0, 95% CI: 2.04-59.8, p = 0.0054); and female gender (OR = 4.8, 95% CI = 1.09-21.5, p = 0.0384). The large majority of chronically infected HIV patients with a TI > or = 180 days interrupted treatment for reasons unrelated to virologic response. Almost 30% did not have a significant virologic rebound. Those patients were more likely to be female, had a lower peak VL prior to treatment, and their initial regimen was more likely to be HAART. Examining the immune responses of nonrebounders may contribute to the understanding of protective immunity to HIV.

In a comparison of rates of acquired immunodeficiency syndrome (AIDS)-defining malignancies (ADMs) for 1989-1996 versus 1997-2002, we found a decrease in ADMs (rate ratio, 0.31; P<.0001) and a significant increase in non-AIDS-defining malignancies (non-ADMs; rate ratio, 10.87; P<.0002). The mean CD4 cell count was lower among patients with ADMs than among those with non-ADMs. A longer duration of survival during highly active antiretroviral therapy might explain the increasing incidence of non-ADMs.