Impact of hepatitis coinfection on hospitalization rates and causes in a multicenter cohort of persons living with HIV

Abstract

Background: Chronic viral hepatitis is a potentially important determinant of health care utilization among persons living with HIV. We describe hospitalization rates and reasons for hospitalization among persons living with HIV stratified by coinfection with hepatitis B virus (HBV) and/or hepatitis C virus (HCV).

Methods: Laboratory, demographic, and hospitalization data were obtained for all patients receiving longitudinal HIV care during 2010 at 9 geographically diverse sites. Hepatitis serostatus was assessed by hepatitis B surface antigen and/or hepatitis C antibody. ICD-9 codes were used to assign hospitalizations into diagnostic categories. Negative binomial regression was used to assess factors associated with all-cause and diagnostic category-specific hospitalizations.

Results: A total of 2793 hospitalizations were observed among 12,819 patients. Of these patients, 49.3% had HIV monoinfection, 4.1% HIV/HBV, 15.4% HIV/HCV, 2.5% HIV/HBV/HCV, and 28.7% unknown hepatitis serostatus. Compared with HIV monoinfection, the risk of all-cause hospitalization was increased with HIV/HBV [adjusted incidence rate ratio 1.55 (1.17 to 2.06)], HIV/HCV [1.45 (1.21 to 1.74)], and HIV/HBV/HCV [1.52 (1.04 to 2.22)]. Risk of hospitalization for non-AIDS-defining infection was also higher among patients with HIV/HBV [2.07 (1.38 to 3.11)], HIV/HCV [1.81 (1.36 to 2.40)], and HIV/HBV/HCV [1.96 (1.11 to 3.46)]. HIV/HBV was associated with hospitalization for gastrointestinal/liver disease [2.55 (1.30 to 5.01)]. HIV/HCV was associated with hospitalization for psychiatric illness [1.89 (1.11 to 3.26)].

Conclusions: HBV and HCV coinfection are associated with increased risk of all-cause hospitalization and hospitalization for non-AIDS-defining infections, as compared with HIV monoinfection. Policy-makers and third-party payers should be aware of the heightened risk of hospitalization associated with coinfection when allocating health care resources and considering models of health care delivery.

Figure 1. Unadjusted Hospitalization Rates by Diagnostic Category

Unadjusted rates of all-cause (panel A) and diagnostic category-specific (panel B) hospitalization. Rates are standardized as hospitalizations per 100 PY of AU7 follow-up. Unadjusted negative binomial regression was performed to construct 95% confidence intervals.

Figure 2. Adjusted Incidence Rate Ratio of Hospitalization by Diagnostic Category

Adjusted incidence rate ratios and 95% confidence intervals were calculated using negative binomial regression and are interpreted as the relative rate of admissions compared to the reference group (HIV mono-infection) after controlling for age, gender, race, HIV risk factor, CD4 count, HIV-1 RNA, ART, insurance and clinical care site.