Factors associated with initiation of prolonged analgesic use among patients in the hiv outpatient study (HOPS)

Background: Analgesic use is common but remains poorly described among human immunodeficiency virus (HIV)-infected persons in the highly active antiretroviral therapy era. Methods: We studied HIV Outpatient Study participants during 1996 to 2008. We used Cox proportional hazards regression to assess variables associated with initiation of prolonged analgesia (90 consecutive days of analgesics); logistic regression to explore variables associated with initiation of prolonged opioid analgesia among those taking any prolonged analgesia; and linear regression to determine temporal trends in prolonged analgesia. Results: Among 4180 patients, 931 (22%) initiated prolonged analgesia. Factors independently associated (P<0.05) with prolonged analgesia included: age above 40 years (hazard ratio=1.20), female sex (1.43), injection drug use as an HIV risk factor (1.33), public healthcare payer (1.88), nadir CD4+ less than 200 cells/mm (1.29), tobacco use (1.43), prior opportunistic infection(s) (1.25), antidepressant use (1.76), and anxiolytic use (1.51). Independent correlates of prolonged opioid analgesia were white race (odds ratio=1.64), baseline CD4+ less than 350 cells/mm (1.88), and anxiolytic use (1.87). Prolonged analgesia ranged from 11% to 15% each year. Conclusions: In the highly active antiretroviral therapy era, up to 15% of HIV Outpatient Study patients used prolonged analgesic therapy each year. Variables associated with the initiation of prolonged analgesia included HIV and non-HIV-related factors.