icon-folder.gif   Conference Reports for NATAP  
  Glasgow HIV
28 - 31 October 2018
Glasgow, UK
Back grey_arrow_rt.gif
Non-HIV Comorbidity Predicts Mortality: Impact Greater in Women
  HIV Drug Therapy, Glasgow 2018, October 28-31, 2018, Glasgow
Mark Mascolini
Multimorbidity with non-HIV conditions strongly predicted mortality in a 25,000-person analysis of the Netherlands ATHENA cohort [1]. Women with 3 or more comorbidities ran a higher risk of death than men with 3 or more comorbidities.
Non-HIV conditions account for growing proportions of deaths among people with HIV. And multimorbidity--the accumulation of 2 or more non-HIV conditions--has become more common in many HIV populations. ATHENA cohort investigators conducted this study to explore the relationship between multimorbidity and all-cause mortality in people with HIV and to see whether the mortality impact of multimorbidity differs by gender.
The researchers included adults with HIV-1 in active follow-up during 2000-2017 while taking combination antiretroviral therapy (ART). The study focused on 7 comorbidities: cardiovascular disease, stroke, non-AIDS malignancies, chronic kidney disease, diabetes, hypertension, and obesity. To identify independent predictors of mortality, the ATHENA team used Poisson regression adjusted for age, gender, region of origin, HIV transmission risk, smoking, current CD4 count, viral load, prior AIDS, and chronic HBV or HCV infection.
The analysis included 4813 women (19%) and 20,473 men. Median age was lower in women than men (34.3 versus 40.5), and a lower proportion of women was born in the Netherlands (27.3% versus 62.7%). Median initial CD4 count was similar in women and men (310 and 330), as was median viral load (3.5 and 4.1 log10).
Numbers of comorbidities per person rose with age. Among women 18-29, 30-39, 40-49, 50-59, 60-69, and 70+, proportions with 2 comorbidities were 5.2%, 4.6%, 8.0%, 14.0%, 19.8%, and 27.1%. Only a handful of men younger than 40 had 2 comorbidities. Among men 40-49, 50-59, 60-69, and 70+, proportions with 2 comorbidities were lower than in women in those age brackets: 4.1%, 8.5%, 14.9%, and 21.6%.
Crude mortality rates rose from 5.9 per 1000 person-years with no comorbidities to 21.0 per 1000 with 1 comorbidity, 35.2 per 1000 with 2 comorbidities, 81 per 1000 with 3 comorbidities, and 173 per 1000 with 4 or more. Multivariable Poisson regression determined that, compared with no comorbidities, 1 comorbidity raised the death risk 3.6-fold (95% confidence interval [CI] 3.2 to 4.0), 2 comorbidities raised the risk 6.1-fold (95% CI 5.3 to 7.0), 3 comorbidities raised the risk 13.9-fold (95% CI 11.6 to 16.7), and 4 comorbidities raised the risk 23.8-fold (95% CI 17.7 to 32.0).
Further analysis identified a significant interaction between gender, number or comorbidities, and mortality with the relative risk (RR) for women compared with men rising from 0.55 (0.43 to 0.70) to 0.94 (0.76 to 1.18) to 1.00 (0.73 to 1.36) to 1.69 (1.09 to 2.61) and to 2.21 (1.02 to 4.77) for people with 0, 1, 2, 3, or 4+ comorbidities (P < 0.0001). In other words, women with fewer than 2 comorbidities had lower mortality than men, but women with 3 or more comorbidities had higher mortality than men. Early use of dual-nucleoside therapy by women partly explained their excess mortality risk with more multimorbidity.
The ATHENA investigators concluded that multimorbidity strongly and independently predicts mortality in people with HIV. Although all-cause mortality was lower in women than men overall, multimorbidity affected mortality more in women than men.
1. Wit F, van der Valk M, Gisolf J, et al. Multimorbidity and risk of death differs by gender in people living with HIV in the Netherlands: the ATHENA cohort study. HIV Drug Therapy, Glasgow 2018, October 28-31, 2018, Glasgow. Abstract O115.