icon-    folder.gif   Conference Reports for NATAP  
  Conference on Retroviruses
and Opportunistic Infections (CROI)
February 13-16, 2017, Seattle WA
Back grey_arrow_rt.gif
Age, Weight, Being a Woman Tied to Falls in Veterans With/Without HIV
  Conference on Retroviruses and Opportunistic Infections (CROI), February 13-16, 2017, Seattle
Mark Mascolini
Veterans with HIV had a higher incidence of medically significant falls than veterans without HIV in a 130,000-person study [1]. Age over 50 independently predicted falls in veterans both with and without HIV. Other independent fall predictors were female sex, white race, and being underweight.
Veterans Aging Cohort Study (VACS) investigators set out to explore links between HIV status, age, race, and other variables and medically significant falls. This longitudinal cohort analysis included 130,107 HIV-positive and negative veterans in care between 1996 and 2009. The VACS team identified medically significant falls by injury codes and by a machine learning algorithm that identifies falls in radiology reports. To pinpoint fall predictors they used generalized estimating equations with autoregression correlation.
The analysis included 43,615 veterans with HIV and 86,492 HIV-negative veterans (34% and 66%). Both groups had a baseline age of 47 years. The HIV group had a lower proportion of whites (38% vs 40%) and a higher proportion of blacks (49% versus 47%) (P < 0.0001), a lower average body mass index (25 versus 30 kg/m2, P < 0.0001), and higher proportions with HCV infection (27% vs 11%, P < 0.0001), anemia (29% versus 14%, P < 0.0001), and dementia (0.7% versus 0.2%, P < 0.0001). Women made up 2% of each group.
Both groups took an average of 4 medications with higher proportions of the HIV group taking opioids (13% vs 11%) and lower proportions taking muscle relaxants (3% vs 5%) and anticonvulsants (2% versus 3%) (P < 0.0001 for all medication differences).
Age at first fall averaged 48 overall, and fall incidence was significantly higher in the HIV group (30 vs 27 per 1000 person-years, P < 0.0001). Fall incidence rose faster with age in HIV-positive veterans (about 20 to 45 per 1000 person-years from age 35 to 65) than in HIV-negative veterans (about 20 to 30 per 1000 from age 35 to 65).
Multivariable analysis adjusted for demographics, comorbidities, and medications determined that, compared with HIV-negative veterans under age 50, older HIV-positive veterans and older HIV-negative veterans both had higher odds of a medically significant fall (adjusted odds ratio [aOR] 1.18, 95% confidence interval [CI] 1.12 to 1.25, P < 0.0001, with HIV; aOR 1.13, 95% CI 1.09 to 1.16, P < 0.0001, without HIV). HIV-positive veterans younger than 50 had lower odds of a significant fall than HIV-negative veterans younger than 50 (aOR 0.89, 95% CI 0.85 to 0.93, P < 0.0001).
Compared with men, women had 40% higher odds of a medically significant fall (aOR 1.40, 95% CI 1.30 to 1.51, P < 0.0001). Compared with whites, blacks had almost a 20% lower chance of a fall (aOR 0.82, 95% CI 0.80 to 0.85, P < 0.0001). Compared with normal-weight veterans, underweight veterans had 35% higher odds of a fall (aOR 1.35, 95% CI 1.26 to 1.44, P < 0.0001), while overweight veterans and obese veterans both had lower odds of a fall (aOR 0.85, 95% CI 0.83 to 0.88, P < 0.0001, for overweight; aOR 0.81, 95% CI 0.78 to 0.84, P < 0.0001, for obese).
Given the significantly greater fall incidence in veterans with HIV, the researchers recommend that HIV-positive people over age 50 should be assessed for fall risk.
1. Womack JA, Murphy TE, Bathulapalli H, et al. longitudinal association between HIV status and medically significant falls. Conference on Retroviruses and Opportunistic Infections (CROI), February 13-16, 2017, Seattle. Abstract 930.