Background:  HIV⁺ individuals smoke at double to triple the rate of the general population and experience greater morbidity and mortality from smoking. Small studies have examined characteristics of HIV⁺ smokers, but few have examined smoking patterns of large patient (pt) numbers in routine care from geographically distinct US sites. Understanding patient characteristics of HIV⁺ smokers will inform evidence-based smoking cessation interventions.

Methods:  A convenience sample of HIV⁺ patients at the University of Washington (UW) and University of Alabama at Birmingham (UAB) were assessed on psychological characteristics and smoking status using a computerized Pt Based Metric (PBM) from 2006 to 2009. Variables included demographics, depression (PHQ-9), substance use (ASSIST), alcohol use (AUDIT-C), and HIV measures (VL, CD4, ART receipt). Univariate analyses and multinomial logistic regression (MLR) was used to determine associations between these factors and smoking status.

Results:  Among 2064 patients (UW = 922, UAB = 1142, mean age = 43.6, 80% male, 56% white, 49% public insurance, 15% uninsured, 51% lifetime substance abuse, and 15% “at risk” alcohol use), 44% were current smokers, 22% were former smokers and 34% never smoked. UW had a higher prevalence of active smoking compared to UAB (46% vs 42%, P <0.001). Current smokers were significantly more likely to ever have used drugs or alcohol and reported higher levels of depressive and panic symptoms compared to persons who had never smoked, while former smokers reported intermediate levels of these measures. Among patients on ART, current smokers had lower rates of virological suppression (<50 copies/mL) compared to former or never smokers (62% vs 70% or 66%, P <0.05). In MLR analyses, relative to never smokers, current smokers were more likely to be white males (OR = 1.9, 95%CI 1.4 to 2.8), white females (OR = 2.9, 1.6 to 5.1), uninsured (OR = 1.9, 1.4 to 2.7) or publicly insured (OR = 2.2, 1.7 to 2.9), to have current (OR = 6.7, 4.5 to 9.9) or prior substance abuse (OR = 4.9, 3.8 to 6.5), and to be an “at risk” drinker (OR = 2.1, 1.4 to 3.0).

Conclusions:  HIV⁺ smokers have multiple concurrent risk factors for medical and psychiatric conditions, including higher rates of depression, panic, alcohol, other substance use, and lower rates of viral load suppression. Former smokers fall on a continuum on nearly all psychological and medical factors between current and never smokers. Effective smoking cessation interventions in HIV⁺ patients need to encompass the complexity of their psychological, substance, and medical risks.