HIV-associated comorbidities as mediators of the association between people living with HIV and hospital-acquired infections

Abstract

Background: Hospital-acquired infections (HAIs) lead to poor health outcomes in hospitalized patients and may be disproportionately affecting the aging population of people living with HIV (PLWH). This study determined the association between HIV and HAIs, and analyzed the potential mediating effects of comorbidities.

Methods: The Louisiana Hospital Inpatient Discharge Database for the years 2011-2015 was used. All patients with at least 1 HAI diagnosis within this source population were included as cases in the case-control study, and a 1:1 ratio of controls was randomly selected from the same hospitals.

Results: Of the 1,852,769 eligible hospital discharges that occurred from 2011 through 2015, there were 7,422 patients with at least 1 HAI. Marginal logistic regressions of the case-control sample showed a strong association between HIV and central line-associated bloodstream infections (CLABSIs), but an inverse association between HIV and any HAI. However, the mediation analyses revealed that having at least 1 comorbidity mediates the association between HIV and CLABSIs.

Discussion: The unexpected inverse association between HIV and HAI could be attributed to the sample size of the exposed group of patients, or it could be explained by the mechanisms of treatment for HIV patients.

Conclusions: This study found that people living with HIV are at an increased risk of developing a CLABSI, which is consistent with the published literature. The mediation analyses indicated that having at least 1 comorbidity mediated the association between HIV and CLABSI diagnosis.

Keywords: Chronic diseases; Comorbid conditions; Infectious disease; Louisiana; Mediation analysis.