Introduction: Monitoring antiretroviral (ARV) drug resistance is of growing importance in the management of persons infected with HIV, but few reports document how genotypic and phenotypic resistance testing (GPT) has been used among patients receiving routine outpatient care.
Methods: We studied data from participants in the HIV Outpatient Study seen at 10 HIV clinics in the United States during 1999 to 2006. We restricted analyses to patients whom we considered eligible for GPT (i.e., had a documented HIV viral load >1000 copies/mL). We used multivariable general modeling to evaluate temporal trends in use of GPT among eligible patients and to identify factors associated with being tested during 1999 to 2002 and 2003 to 2006.
Results: Of 5594 active patients, 3995 (71%) were considered eligible for GPT in at least one year during 1999 to 2006 (declining from 50.2% in 1999 to 31.2% in 2006). The fraction of eligible patients receiving GPT increased from 11.2% in 1999 to 31.0% in 2003 (P < 0.001 for trend) and then stabilized at approximately 30% through 2006. Among persons tested, the annual percentage receiving only genotype testing declined over time (90% to 56%), whereas the percentage receiving genotype and phenotype testing increased (5.4% to 39.1%). The annual use of GPT for ARV-naïve patients increased over time and after 2003 exceeded the corresponding rates for ARV-experienced patients. In multivariable analyses, low CD4 count and high HIV viral load were consistently associated with GPT. Compared with other ARV-experienced patients, those who were triple ARV-class experienced were consistently more likely to be tested, whereas ARV-naïve were less likely to be tested during 1999 to 2002 and more likely during 2003 to 2006. In addition, women and heterosexual men (vs. men who have sex with men) and black patients (vs. white) were less likely to be tested during 1999 to 2002, whereas older patients were less likely to be tested during 2003 to 2006.
Discussion: The annual frequency of GPT use has increased almost threefold since 1999. GPT use among ARV-naïve patients has increased coincident with dissemination of recommendations. Although earlier sex and racial/ethnic disparities in testing have waned, older patients were significantly less likely to be tested in recent years.