tions (Chesney, 2003; Bartlett et al., 2001). Nonadherence due to regimen complexity may be relieved somewhat as more antiretroviral medications are approved for once-daily dosing (Piliero and Colagreco, 2003). Other issues in adherence stem from the patient and the patient/health care provider relationship. Studies have shown that depression, lack of belief in the efficacy of the medicine, and lack of confidence in ability to adhere to the regimen predict nonadherence to HAART (Catz et al., 2000; Singh et al., 1999). There is a great deal of evidence that active alcohol or drug use contributes to nonadherence, although good adherence can be achieved among this population (Chesney, 2003; Escobar et al., 2003; Tucker et al., 2003; Lucas et al., 2002; Chesney et al., 2000). Lack of HIV-related knowledge and low literacy levels in general are also associated with poor adherence (Kalichman et al., 1999). One study of individuals with excellent adherence to HAART found that generally those with high adherence rates believed that the medication was and would continue to work and had trust and confidence in their primary care provider. They also were motivated by a strong desire to stay healthy that made taking their medications a priority, even when they were actively using drugs and alcohol (Malcolm et al., 2003).
Although these factors do predict poor adherence in overall study populations, it has also been shown that it is difficult for clinicians to predict adherence levels in individuals. In one study, physicians and clinic nurses were able to predict an individual’s adherence less than half the time (Paterson et al., 2000). This is a significant issue when it concerns members of groups that are already highly stigmatized, such as those with a mental illness, because it could lead to the denial of therapy based on a presumption of the inability to adhere. Bogart and colleagues (2000) found in a 1998 survey that physicians relied on a variety of nonmedical factors in determining whether or not to prescribe HAART, including demographic and psychiatric factors such as homelessness, age, and history of psychiatric hospitalizations.
It is also important to note that many of the factors shown to inhibit an individual’s ability to adhere are not immutable, but can be influenced with appropriate interventions. As noted earlier, there is evidence that even in populations where it is generally thought adherence will be low, such as individuals who are homeless, a significant proportion can attain high enough levels of adherence to realize some (though less than optimal) therapeutic benefit from the medication. From this evidence, it appears that certain interventions, such as treatment for depression, may increase adherence levels, allowing individuals to gain greater treatment benefit (Bangsberg et al., 2000). Appropriate adherence support provided as a routine part of HIV care offers the opportunity to get the most out of therapy and helps to reduce the likelihood that drug resistance will develop.