Predictors for high HIV viral load (VL) among persons with HIV in Los Angeles County using laboratory surveillance and the HIV/AIDS Reporting System (HARS)

Background: High HIV viral load (VL) in persons living with HIV may indicate absent or inadequate HIV care. We linked data from laboratory surveillance database with HIV/AIDS case registry to examine possible predictors for high VL. Methods: We selected persons reported living with HIV in LAC for whom a recent VL test was reported in 04/2006 – 10/2009. Using stratified and mixed model analyses, we examined the association between high VL (> 10,000 copies/ml) and: 1) individual level variables, such as men who have sex with men (MSM), injection drug use (IDU), care facility type, insurance, year of diagnosis, country of origin, homelessness, and frequency of VL testing; as well as 2) census tract level variables, such as median household income, education, and presence of HIV providers. We used Mantel-Haenszel Chi-square and mixed model analyses to examine statistical associations. Results: High VL was detected in 14% of the 23,852 persons who had a reported VL. Independent positively associated factors, adjusted odd ratios and 95% CI, for high VL include: Age < 30 yrs: 2.0 (1.8-2.4); >7 months between VL tests: 2.0 (1.8-2.3); US-born: 1.5 (1.3-1.7); diagnosed in pre-HAART era: 1.5 (1.3-1.6); lack of private insurance: 1.4 (1.2-1.5); Black: 1.3 (1.2-1.5); Latino: 1.2 (1.0-1.3); IDU: 1.2 (1.1-1.4); MSM: 1.1(1.0-1.3); and low high school graduation rate (p<0.001). Conclusion: High VL is associated with multiple socio-demographic and behavior risk factors. Prevention and treatment intervention programs should tailor their resources and efforts among the HIV persons and communities who have not been effectively treated.