Racial/Ethnic Disparities in Continuous Viral Suppression Among People Living with HIV in Virginia

Introduction: Viral suppression (VS), a key step along the HIV care continuum and an essential component of HIV transmission prevention, is frequently evaluated as a cross-sectional outcome and much attention has been given to the achievement of an undetectable viral load during a single calendar or patient-year. However, successful disease management and treatment as prevention depend on an individual's ability to achieve VS over time. Recent work suggests disparities in HIV care continuum outcomes not present in cross-sectional analyses may emerge over time. This study examined the relationship between race/ethnicity and the achievement of VS for two consecutive years (termed “continuous VS”) in a sample of persons living with HIV (PLWH) in Virginia, as well as in the subset who were engaged in Ryan White-funded care.

 

Methods: Using surveillance and HIV care data from Virginia’s Care Markers Database (CMDB), we identified PLWH in Virginia as of December 31, 2015, who were diagnosed with HIV on or before December 31, 2013. Client records were matched with Ryan White service utilization data to identify recipients of Ryan White-funded care. We used multivariate logistic regression to estimate the association between race/ethnicity and continuous VS (most recent viral load < 200 copies/mL in both 2014 and 2015), and interaction terms between race/ethnicity and gender to investigate how the association may vary by gender.

Results: In our sample of PLWH in Virginia (n=23,047), 6,793 (29%) achieved continuous VS and 7,535 (33%) had evidence of Ryan White-funded care in 2014. In multivariate analysis, among PLWH in Virginia, Black persons had significantly lower odds of continuous VS compared to White persons (adjusted OR=0.92, 95% CI: 0.84–0.99). Ryan White recipients were significantly more likely to achieve continuous VS than non-Ryan White recipients (adjusted OR=8.89, 95% CI: 8.29–9.54). When the sample was limited to PLWH with evidence of Ryan White-funded care (n=7,535), the association between Black race and continuous VS was no longer statistically significant (p= 0.5213); however, Hispanic persons had higher odds of continuous VS (adjusted OR=1.27, 95% CI: 1.03, 1.56) as compared to White persons. The interaction between gender and Hispanic ethnicity had a p-value of 0.1491; Hispanic females were more likely to have continuous VS than Hispanic males, though this did not reach a significance level of p<.05.

Conclusions: Significant racial/ethnic disparities in continuous VS exist among PLWH in Virginia, but Ryan White-funded care may attenuate, or alter, these disparities. Additionally, these findings suggest that continued exploration of racial/ethnic disparities by gender is needed.