Vu Hai Vinh1, Roselyne Vallo2, Hoang Thi Giang3, Duong Thi Huong3, Khuat Thi Hai Oanh4, Pham Minh Khue3, Nham Thi Tuyet Thanh4, Catherine Quillet2, Delphine Rapoud2, Laurent Michel5, Philippe Van de Perre2, Jonathan Feelemyer6, Jean Pierre Moles7, Amandine Cournil2, Don Des Jarlais6, Didier Laureillard8, Nicolas Nagot9. 1. Department of Infectious and Tropical Diseases, Viet Tiep Hospital, Hai Phong, Vietnam. 2. Pathogenesis and control of chronic infections, University of Montpellier, INSERM, Etablissement Français du Sang, University of Antilles, Montpellier, France. 3. Hai Phong University of Medicine and Pharmacy, Hai Phong, Vietnam. 4. Supporting Community Development Initiatives, Hanoi, Vietnam. 5. Pierre Nicole Centre, French Red Cross, CESP/Inserm 1018, Paris, France. 6. New-York University, New York. 7. Pathogenesis and control of chronic infections, University of Montpellier, INSERM, Etablissement Français du Sang, University of Antilles, Montpellier, France. Electronic address: jean-pierre.moles@inserm.fr. 8. Pathogenesis and control of chronic infections, University of Montpellier, INSERM, Etablissement Français du Sang, University of Antilles, Montpellier, France; Department of Infectious Diseases, Caremeau University Hospital, Nîmes, France. 9. Pathogenesis and control of chronic infections, University of Montpellier, INSERM, Etablissement Français du Sang, University of Antilles, Montpellier, France. Electronic address: n-nagot@chu-montpellier.fr.
Abstract
OBJECTIVE: To estimate the residual mortality rate among people who inject drugs (PWID) in a Low-Middle Income Countries context where the HIV epidemic has been controlled and methadone coverage is high. STUDY DESIGN AND SETTING: PWID from Haiphong, Vietnam, were recruited through three annual respondent-driven sampling surveys that fueled two cohorts of PWID with HIV (n = 761) and without HIV (n = 897), with bi-annual follow-up. Presumed causes of death were ascertained from medical records and/or interviews of participants family. RESULTS: Among the 1658 participants with a median follow-up of 2 years, 67 and 36 died in the HIV-positive and HIV-negative cohort, respectively, yielding crude mortality rates of 4.3 (95% Confidence interval (CI): 3.3-5.4) per 100 person-years of follow-up (PYFU) and 1.9 (CI: 1.4-2.6) per 100 PYFU. In the HIV-positive cohort, in which 81% of participants had undetectable viral load, the two main causes of death were tuberculosis and HIV-related diseases. In the HIV-negative cohort, the two main causes of death were liver-related diseases and overdose. In a time-dependent multivariable model, "unsuppressed viral load" was associated with increased risk of mortality, whereas "being on methadone" or "being employed" was associated with a lower risk. CONCLUSION: Despite a very successful HIV and methadone program, the mortality remains high among PWID in Vietnam, largely due to curable infectious diseases such as tuberculosis and viral hepatitis.
OBJECTIVE: To estimate the residual mortality rate among people who inject drugs (PWID) in a Low-Middle Income Countries context where the HIV epidemic has been controlled and methadone coverage is high. STUDY DESIGN AND SETTING: PWID from Haiphong, Vietnam, were recruited through three annual respondent-driven sampling surveys that fueled two cohorts of PWID with HIV (n = 761) and without HIV (n = 897), with bi-annual follow-up. Presumed causes of death were ascertained from medical records and/or interviews of participants family. RESULTS: Among the 1658 participants with a median follow-up of 2 years, 67 and 36 died in the HIV-positive and HIV-negative cohort, respectively, yielding crude mortality rates of 4.3 (95% Confidence interval (CI): 3.3-5.4) per 100 person-years of follow-up (PYFU) and 1.9 (CI: 1.4-2.6) per 100 PYFU. In the HIV-positive cohort, in which 81% of participants had undetectable viral load, the two main causes of death were tuberculosis and HIV-related diseases. In the HIV-negative cohort, the two main causes of death were liver-related diseases and overdose. In a time-dependent multivariable model, "unsuppressed viral load" was associated with increased risk of mortality, whereas "being on methadone" or "being employed" was associated with a lower risk. CONCLUSION: Despite a very successful HIV and methadone program, the mortality remains high among PWID in Vietnam, largely due to curable infectious diseases such as tuberculosis and viral hepatitis.