icon-folder.gif   Conference Reports for NATAP  
2015, October 7-11
San Diego
Back grey_arrow_rt.gif
HIV, Hypertension, and Diabetes Control Vary Between and Within HIV Clinics
  IDSA/IDWeek 2015, October 7-11, San Diego
Mark Mascolini
Tight control of HIV replication, blood pressure, and diabetes varied substantially across HIV clinics in the Veterans Administration (VA) system [1]. And within individual clinics, good control of one condition did not correlate with good control of the others.
With collaborators in the VA and other institutions, University of Iowa researchers who conducted this study observed that good control of common HIV-related comorbidities has become critical to care of people with HIV infection. But little is known about differences in quality of comorbidity control across HIV specialty clinics. They planned this cross-sectional analysis (1) to gauge variation in HIV control, hypertension control, and diabetes control between HIV clinics in the VA, and (2) to determine clinic-level correlations between HIV control and control of the two comorbidities.
Seeing about 25,000 veterans with HIV, the VA is the largest HIV care provider in the United States. This analysis involved HIV-positive veterans who received care at a single VA clinic in 2013. The investigators defined viral control as a last viral load at or below 200 copies in veterans taking antiretroviral therapy. Hypertension control meant a last blood pressure below 140/90 mm Hg in HIV-positive veterans. And diabetes control meant a last hemoglobin A1c below 9% in veterans with HIV.
The researchers limited the analysis to clinics with more than 25 veterans receiving care for HIV plus either hypertension or diabetes. Statistical analysis of clinic disease-control rates adjusted for demographics, other comorbidities, nadir CD4 count, residential ZIP-code poverty rates from census data, and time receiving HIV care in the VA. The investigators defined high-performing clinics as those in the top one-fifth for each measure (HIV, blood pressure, and A1c control).
The study group included 21,334 veterans, 97% of them men, 50% black, and 44% white. Age averaged 54, and years in care averaged 8.8. Almost all veterans (94%) were taking antiretroviral therapy, 90% had a viral load below 200 copies; 39% had a hypertension diagnosis, and 69% of them had hypertension control; 17% had a diabetes diagnosis, and 87% of them had diabetes control.
Among 109 clinics analyzed for HIV load, control rates ranged from 73% to 100% and averaged 90%. Among 75 clinics assessed for hypertension, control rates ranged from 51% to 84% and averaged 68%. Among 42 clinics in the diabetes analysis, control rates ranged from 71% to 98% and averaged 88%.
Across 75 clinics, HIV control did not correlate with blood pressure control (r = 0.18, P = 0.26). Across 42 clinics, HIV control did not correlate with A1c control (r = 0.13, P = 0.40). And across 42 clinics, A1c control correlated only modestly with blood pressure control (r = 0.27, P = 0.08). Fewer than 5% of clinics ranked as a high-performing clinic for both HIV control and hypertension control.
The researchers concluded that "control measures are not highly correlated at [the] clinic level" and "there does not appear to be a single construct of globally 'high performing' clinics." They suggested that quality-improvement programs aimed at one condition "may have no or unintended effects on other domains of care."
1. Ohl M, Richardson K, Parker V, et al. HIV viral control and comorbidity control are not highly correlated at the level of the HIV clinic. IDWeek 2015, October 7-11, San Diego. Abstract 729.