The Clinical Implications of High Rates of Intimate Partner Violence Against HIV-Positive Women

Introduction: Intimate partner violence (IPV) is associated with increased risk of HIV infection among women, however, whether IPV affects outcomes after HIV infection is uncertain. We assess the impact of IPV on HIV-positive women. Methods: All HIV-positive women who received outpatient HIV care in southern Alberta between March 2009 and January 2012 were screened for IPV. The associations with IPV of sociodemographic factors, health-related quality of life, clinical status, and hospitalizations were obtained from a regional database and evaluated with multivariable regression analysis. Results: Of 339 women screened, 137 (40.4%) reported experiencing IPV. Those disclosing IPV had higher rates of smoking [ adjusted prevalence ratio (APR) = 5.07; 95% confidence interval (CI): 2.72 to 9.43]; illicit drug use (APR = 7.58; CI: 2.45 to 23.26); a history of incarceration (APR = 4.84, CI: 1.85 to 12.68); depression (APR = 2.50, CI: 1.15 to 5.46); and anxiety disorders (APR = 5.75, CI: 2.10 to 15.63). Health-related quality of life was diminished with IPV (APR = 2.94, CI: 1.40 to 6.16) for poor/fair versus very good/excellent. IPV-exposed women were hospitalized 256 times per 1000 patient-years compared to 166/1000 patient-years among IPV-unexposed (P < 0.001) women. The relative risk was increased for HIV-unrelated hospitalizations (APR = 1.42, CI: 1.16 to 1.73) and for HIV-related hospitalizations after outpatient HIV care was initiated (APR = 2.19, CI: 1.01 to 4.85). Modifiable contributors to the poor outcomes included decreased use of antiretroviral therapy (APR = 0.55, CI: 0.34 to 0.91) and additional interruptions in care longer than 1 year (APR = 1.90, CI: 1.07 to 3.39). Conclusions: IPV is associated with deleterious HIV-related and HIV-unrelated health outcomes, of which, suboptimal engagement in care is a contributor. To improve outcomes, practitioners should aim to increase engagement in care of these women in particular.