A Research Of The Prevalence Of HIV And Syphilis Among Drug Users In Xichang County Of Sichuan Province

BACKGROUNDThe main factor associated with HIV and other blood-borne transmitted diseases is the multi-person re-use or sharing of syringes and needles. In addition, the indirect sharing of equipment such as water, cotton, cookers and other drug preparation equipment has been attributed to assisting HIV transmission. Particularly where IDUs sell sex to help fund their drug use, injecting drug users (IDUs) are also vulnerable to HIV infection through sexual transmission. They are also at high risk from other blood-borne and sexual-transmitted infections such as hepatitis B and C, and syphilis, and from premature death from drug overdose or sepsis infection. Because of a large population base, a small increase of HIV prevalence in the general population in China would mean a large increase in the number of domestic infections and a significant contribution to the total number of infections around the world. The traditional approach to identify risk factors for HIV infection among IDUs is to conduct cross-sectional survey in a defined area. However, a more definitive approach to identify risk factors for HIV infection is to assemble a cohort of initially uninfected individuals who are then followed prospectively at prescribed time intervals for seroconversion. This design, the seroincidence cohort, overcomes the major limitation of the cross-sectional seroprevalence study by establishing a temporal causal relationship between expose and disease occurrence measured by seroconversion. Several cohort studies have been conducted among IDUs in China, most in the provinces of Yunnan and Guangxi, which have a high HIV prevalence. These studies showed varying but generally high seroconversion rates. China has a large geographic area and a variety of cultures and customs; more cohort studies among IDUs should provide valuable data for understanding the whole picture of the HIV epidemic. In addition, comprehensive evaluation of the factors associated with HIV, HCV, HBsAg, syphilis seroconversion in the IDU cohort would assist in developing intervention programs among this population. The aim of this 4-year cohort study was to estimate the trends of HIV, HCV, HBsAg, syphilis incidences and HIV/HCV coinfection rates, mortality rates and death causes, as well as to identify the independent factors for the incidences to HIV, HCV, HBsAg, syphilis, as well as mortality. At the almost same time, we start three cross-sectional surveys to know the trends of prevalence of HIV, HCV, HBsAg, syphilis and high risk behaviors among IDUs in Xichang County of Sichuan Province, China.OBJECTIVES1. To investigate HIV, HCV, HBsAg and syphilis sero-incidence and contributed factors among IDUs in a drug trafficking city of southwest Sichuan Province and to knowledge trends of HIV risk behaviors in a prospective cohort of IDUs; to compare with trends of the prevalence of HIV, HCV, HBsAg and syphilis and risk behaviors in three cross-sectional surveys among IDUs.2. To investigate cohort retention and factors associated with Socio-demographic characteristics, drug use and sexual behaviors at the baseline among IDUs in Xichang, Sichuan Province.3. To study the mortality in the 4-year follow-up study of IDUs in Xichang of Sichuan Province.METHODS1. In November 2002, a community-based baseline survey was conducted to recruit 333 HIV-seronegative IDUs for a prospective cohort study in Xichang County of Sichuan Province, China. Follow-up visits were conducted every 6 months to investigate drug use and sexual behaviors, and blood specimens were also collected to test for HIV antibody, HCV antibody, HBsAg, and syphilis antibody. Univariate and multivariate Poisson regression model were conducted to analyze factors contributed to HIV, HCV, HBsAg, syphilis seroconversion. Factors associated with participants completing the 4-year follow-up on time were analyzed using chi-square test in univariate analysis and Logistic regression model in multivariate analysis. Descriptive statistical analyses were used to knowledge the trends of drug use and sexual behaviors in a prospective cohort of IDUs.2. Through November 2002, 376 injection drug users were recruited and then followed up for four year as part of a prospective cohort study conducted in Xichang site. Deaths were recorded with a designable questionnaire during the follow-up period. Socio-demographics and risk behaviors were recorded at the time of screening. Trends in mortality were analyzed using Kaplan-Meier survival analysis and Cox regression analysis.3. In May 2004, Oct 2005 and June 2007, three cross-sectional surveys were conducted in drug detoxification in Xichang County of Sichuan Province, China. A questionnaire was formed to investigate drug use and sexual behaviors, and blood specimens were also collected to test for HIV antibody, HCV antibody, HBsAg, and syphilis antibody. The trends of drug use and sexual behaviors in three cross-section studies were analyzed by Logistic regression.RESULTS1. HIV prevalence at baseline was 11.3% (43/379), and total HIV seroincidence was 2.19 per 100 person-years (95% confidence interval (CI): 1.12-3.27), and all subtypes of 16 HIV-1 seroconversion were CRF₀7BC. 13 HIV-1 seroconversion people are firstly infected with HCV. HIV incidence per 100 person-years declined first and then rose at the fourth follow-up year, from 3.17 in 2002-2003 (95% CI: 0.97-5.37) to 1.81 in 2003-2004 (95% CI: 0.04-3.58), and 1.37 in 2004-2005 (95% CI: 0-3.28) and 1.82 in 2005-2006 (95% CI: 0-4.33). Multivariate Poisson regression model showed that risk factors, which were significantly associated with HIV seroincidence, included minority ethnics (Relative ratio (RR), 4.31; 95% CI, 1.56-11.91; P=0.0049) and having shared needles and syringes in the past 3 months≥1 times/week (RR, 32.51; 95% CI, 10.43-101.35; P<0.0001).2. HCV prevalence at baseline was 71.0% (269/379), and total HCV seroconversion was 31.63 per 100 person-years (95% CI: 22.95-40.32). HCV incidence per 100 person-years declined first then rose at the fourth follow-up year, from 42.07 in 2002-2003 (95% CI: 27.02-57.12) to 31.65 in 2003-2004 (95% CI: 13.74-49.56) to 15.59 in 2004-2005 (95% CI: 1.92-29.25) then to 20.08 in 2005-2006 (95% CI: 0.40-39.76). Drug injection in the past 3 months>7 times per week (RR, 8.30; 95% CI, 4.70-14.68; PO.0001) was significantly associated with HCV seroincidence.3. HBsAg prevalence at baseline was 16.9% (64/379), and HBsAg incidence was 11.24 per 100 person-years (95% CI: 8.30-14.19), decreasing 14.16 in 2002-2003 (95% CI: 8.91-19.40) to 9.15 in 2003-2004 (95% CI: 4.36-13.94) to 8.97 in 2004-2005 (95% CI: 3.11-14.82), then rising to 10.68 during 2005-2006 (95% CI: 1.32-20.04). In multivariate Poisson regression model, drug injection in the past 3 months≥7 times/week (RR, 6.62: 95% CI: 3.85-11.38; P<0.0001) and syphilis infection (RR, 2.16; 95% CI: 1.22-3.82; P=0.0086) were significantly associated with HBsAg seroconversion.4. Syphilis prevalence was 15.3% (58/379) at the baseline and total seroconversion in this cohort is 4.15 per 100 person-years (95% CI, 2.49-5.82), decreasing from 6.04 in 2002-2003 (95% CI: 2.76-9.32) to 3.18 in 2003-2004 (95% CI: 0.64-5.73) to 0.87 in 2004-2005 (95% CI: 0-2.58), then rising to 6.75 in 2005-2006 (95% CI: 0.13-13.36). In multivariate Poisson regression, female (RR, 4.09; 95% CI, 1.79-9.34; P=0.0008), and married or cohabited (RR, 2.65; 95% CI, 1.19-5.92, P=0.0173) were significantly associated with syphilis seroconversion.5. During the 48-month follow-up period, cohort retention rate was 59.2% (197 of 333). Multivariate Logistic regression model showed that factors significantly contributed to cohort retention included appearing at the 6-month follow-up visit (OR, 4.01; 95% CI, 2.35-6.84; P<0.0001), and junior high school or lower (OR, 2.11; 95% CI, 1.31-3.40; P=0.0020). Reduction trend in sharing needles or syringes in the past 3 months was obvious. However, unprotected sexual behaviors were less amenable to change.6. A total of 45 deaths were found during the 4-year follow-up study. The Mortality rate of this cohort was 32.96 per 1000 person-years (95% CI: 23.33-42.59), decreasing from 77.32/1000 person-years at the first year (95%CI: 48.68-105.95), to 32.06/1000 person-years at the second year(95%CI: 13.12-51.01), to 15.27/1000 person-years at the third year (95% CI: 1.88-28.65), to 3.01/1000 person-years at the forth year (95% CI: 0-8.90). Among 45 deaths, 68.9% (N=31) died of heroin overdoses. Total overdose mortality is 22.89/1000 person-years (95% CI: 14.83-30.95), declining by year from 42.07 (95% CI: 27.02-57.12), to 31.65 (95% CI: 13.74-49.56), to 15.59 (95% CI: 1.92-29.25), to 3.01 (95% CI:0-8.90), respectively. The mean age of overdose death is 30.5±6.2 years. In the final Cox regression model drug injection in the past 3 months>7 times per week (hazard ratio (HR), 3.69; 95% CI: 1.96, 6.94; P<0.0001) were found to be associated with all-causes mortality, and≥10 year history of heroin use (HR, 2.53; 95% CI: 1.19, 5.39; P=0.0161), and drug injection in the past 3 months>7 times per week (HR, 3.60; 95% CI: 1.70, 7.66; P=0.0009) were found to be associated with overdose-specific mortality in the final Cox regression model.7. In three cross-sectional studies among IDUs, the prevalence of drug injection past 3 months≥7 times per week (2007 vs. 2004, adjusted odds ratio (AOR), 0.35; 95% CI, 0.24-0.51, P<0.01) and having shared needles and syringes past 3 months (2007 vs. 2004, AOR, 0.48; 95% CI, 0.29-0.78, P<0.01) showed decreasing trends, but the trends of risk sexual behaviors have no significant changes.CONCLUSIONS1. Our study showed the rapid transmission of HIV, HCV, HBsAg and syphilis among IDUs in Sichuan Province, China. By reducing risky behaviors related to drug injection and sexual practices, HIV, HCV, HBsAg and syphilis seroincidence are decreasing in the first three years. However, effective risk reduction is not risk elimination. HIV, HCV, HBsAg and syphilis seroincidence are rising at the fourth year. Substantial 'residual' injection and sexual risk behavior is likely to remain even after the great majority of IDUs are aware of AIDS, hepatitis C, hepatitis B, and syphilis and have taken steps to reduce risk behavior.2. The main reasons for loss of follow-up in study cohort included a high rate of incarceration and early death; most of these deaths resulted from overdose of illicit drugs. Cohort retention among IDUs with junior high school or lower and not appearing at the 6-month follow-up visit was low in this cohort. Follow-up of the study cohort is still ongoing. The retention strategies are to be reexamined and revised to increase the retention rate, and thereby minimize potential selection bias.3. Heroin overdose could rapidly be becoming a significant contributor to premature mortality among drug users in Xichang County, Sichuan Province. Future studies should be conducted to focus on fatal and no-fatal overdose and contributed factors among IDUs to provide targeted interventions in reducing unnatural mortality, especially increasing attraction and retention of drug users to methadone maintenance treatment (MMT).4. The results of our three cross-sectional studies among IDUs at the same area showed that drug injection in the past 3 months and having shared needles and syringes have been declining, but the unprotected sexual behaviors have no obvious changes.