Making treatment decisions with HIV infection: a pilot study of patient preferences

The importance of understanding patient preferences in making treatment decisions is widely recognized. This pilot study utilized a forced-choice paired-comparison method in which 28 ambulatory HIV-infected patients were given a computer-generated presentation of all possible pairs of eight different treatment options for their disease (FDA-approved medications, experimental and alternative treatments, no medication). Preferences were analyzed using binary multidimensional scaling analyses to determine the utility of paired-comparison models for the study of treatment-decision making and to identify factors influencing patient decision making. Results indicated that a three-dimensional model provided the best fit for the data. One dimension correlated with medications that raise CD4+ lymphocyte counts (r = 0.92, p < 0.001) and a second dimension correlated with frequency of dosing (r = 0.97, p < 0.0001). Patients' internal consistency of decision making was inversely correlated with severity of AIDS dementia symptoms as measured by performance on a neuropsychological test battery (r = -0.55, p < 0.0025). This finding indicates that AIDS dementia may significantly hinder patients' ability to use a rational (internally consistent) decision-making strategy in making treatment choices. Results also suggested that AIDS patients base treatment decisions primarily on the likelihood of raising CD4+ cell counts and restrictiveness of dosing regimens, but are not influenced by FDA approval status, volume of empirical support for the medications, or even possible harmful side effects. The implications of these findings for the treatment of patients with AIDS are discussed.