A Population-Based Study Examining Hepatitis B Virus Infection and Immunization Rates in Northwest China

Background and Aim Current baseline data regarding the prevalence of hepatitis B virus (HBV) infections and the immune status in hyperendemic areas is necessary in evaluating the effectiveness of ongoing HBV prevention and control programs in northwest China. This study aims to determine the prevalence of chronic HBV infections, past exposure rates, and immune response profiles in Wuwei City, northwest China in 2010. Methods Cross-sectional household survey representative of the Wuwei City population. 28,579 participants were interviewed in the seroepidemiological survey ≥1 year of age. House to house screening was conducted using a standard questionnaire. All serum samples were screened by enzyme-linked immunoassays for the presence of hepatitis B surface antigen, antibodies against HBV surface antigen, and antibodies to the hepatitis B core antigen. Results Among individuals ≥1 year of age, 7.2% (95%CI: 6.3–8.1%) had chronic HBV infections, 43.9% (CI: 40.4–47.4%) had been exposed to HBV, and 23.49% (CI: 21.6–25.3%) had vaccine-induced immunity. Multi-factor weighted logistic regression analysis showed that having household contact with HBV carriers (OR = 2.6, 95%CI: 2.3–3.0) and beauty treatments in public places (OR = 1.2, 95%CI: 1.1–1.3) were the risk factors of HBV infection in whole population. Having household contact with HBV carriers (OR = 3.8, 95% CI: 2.2–6.5) and lack of hepatitis vaccination (OR = 2.0, 95% CI: 1.4–3.3) were the risk factors for HBV infection in children aged 1–14 years. Conclusions Hepatitis B infection remains a serious public health problem in northwest China. Having household contact with HBV carriers and beauty treatments in public places represented HBV infection risk factors. Hepatitis B vaccine immunization strategies need further improvement, particularly by targeting the immunization of rural migrant workers.


Introduction
Hepatitis B virus (HBV) infections represent the leading cause of illness and death in China. Every year, an estimated 300,000 persons in China die from HBV-related liver cancer or cirrhosis, accounting for 37-50% of HBV-related deaths worldwide [1,2]. This disease results in tremendous economic and healthcare burdens. The Chinese National Hepatitis Seroepidemiological Survey found that the prevalence of hepatitis B surface antigen (HBsAg) in individuals aged 1-59 years was 7.2%, and that the hepatitis B epidemic in China shifted from high to intermediate endemicity following implementation of effective nationwide vaccination programs [3]. However, the incidence of HBV in western China remains high with a prevalence of HBsAg of 8.2% in 2006 [3], which was significantly higher than rates observed in eastern China and most Western countries [3,4].
In order to prevent and control HBV transmission, a comprehensive prevention strategy was initiated in 1992 that included universal vaccination of infants; screening of all pregnant women for HBV, administration of postexposure prophylaxis to infants born to HBsAg positive women, catch-up vaccination of children and adolescents, and vaccination of adults who were at increased risk of infection [5]. However, many newborns, children, and adolescents were not vaccinated due to high vaccination costs, especially in economically underdeveloped regions in western China [6,7]. The hepatitis B immunization coverage was 33.5% among children born between 1992-2005 in western China, which was significantly lower than other regions of China [7].
The reported incidence rate of HBV in Wuwei City increased from 571/100,000 cases in 2005 to 742/100,000 cases in 2008, which was the highest in China and 8 times higher than the national average according to National Disease Supervision Information Management System (NDSIMS) which is a direct reporting network system for notifiable infectious diseases in China [8]. However, analysis of surveillance data has not always resulted in similar conclusions. For example, one study indicated that the incident rate of HBV infections was lower than the reported incident rate described by the NDSIMS and that the infection rates did not increase between [2005][2006][2007] in China [9,10]. Other studies that focused on high-risk groups showed different HBV prevalence [11][12][13][14][15]. Therefore, it is unclear whether the prevalence of HBV infections in western China has decreased in recent years because population-based studies have been rather sparse. This report describes a population-based study designed to determine the HBV infection baseline rates in Wuwei City and to estimate HBV infection and exposure rates in addition to characterizing the immune status in the population.

Study population and sampling methods
The target population consisted of local residents from Wuwei City (Gansu province). Participants were selected by the Fourth Military Medical University (FMMU) and are representative of the Wuwei City population (1.81 million). There were 4 administrative divisions: Liangzhou District, Gulang County, Minqin County, and Tianzhu Tibetan Autonomic County. Each region consisted of urban and rural populations. The Wuwei population was divided into 8 sections according to 4 administrative divisions characteristic with urban or rural populations. A stratified random cluster sampling was used according to the proportions of each section (Fig 1). 48,011 persons were selected from 4 communities and 25 villages.

Approach
House to house investigations were conducted by trained, local centers of disease control (CDC) staff members between June 2010 and September 2010. Face-to-face interviews with study subjects or their parents or guardians (if study subjects were ,15 years of age) using a standard questionnaire were used to capture demographic information including gender, age, education, occupation, ethnicity, marital status, household contacts with HBV carriers, immunization history, and history of invasive medical procedures. For children ,15-years of age, immunization status was determined by asking parents or guardians about the child's immunization records. If at least one vaccine dose was administered, respective individuals were defined as 'vaccinated'. If the vaccination certificate was unavailable, the vaccination status was recorded as 'unknown'. Because there are no personal immunization records for adults, the immunization information for adults was based on memory (vaccinated, unvaccinated, unknown). Education levels of the study participants (and their respective occupations) was determined and based on the Chinese social classification criteria. Four field pilots were conducted before the survey. Appropriate visiting times were considered to ensure high response rates. The accuracy of 5% of the questionnaires was verified via telephone interviews conducted by FMMU staff members and the information was 99.64% consistent.

Laboratory testing
Serum was collected from blood samples (5 ml for individuals .6 years of age and 2 ml for children #5 years old) and stored at 280uC. All serum samples were screened by enzyme-linked immunoassays (ELISA) for the presence of HBsAg, antibodies against HBV surface antigen (anti-HBs), and antibodies to the hepatitis B core antigen (anti-HBc) (Wantai Laboratory System, Beijing). Specimens that tested negative were considered negative and not tested further. Specimens testing positive for HBsAg were re-tested by EIA (Abbott Laboratories, Chicago, Illinois), and samples that were double positive were considered to be confirmed positive. 300 specimens were randomly selected and retested using the Abbott EIA (Abbott Laboratories) for the detection of anti-HBc and anti-HBs. The sensitivity and specificity were 100% and 99.4% for anti-HBs and 98.7% and 97.2% for anti-HBc, respectively.

Definition of HBV infection, exposure, and vaccination status
Chronic HBV infection was defined by a positive serum HBsAg result. Exposure to HBV was defined by a positive anti-HBc result, which indicates either past or present infection. Individuals positive for anti-HBs and negative for anti-HBc were considered to have vaccine-induced immunity [4].

Data analysis
All data were entered twice into an EpiData 3.1 software database and checked for consistency. After verifying accuracy, the data were analyzed at FMMU using SAS 9.2 (SAS Institute, Cary, NC). The Taylor Series Linearization method was used to account for the complex sampling design and provided the exact estimate of the standard errors. The weighted prevalence rates of HBV markers were calculated based on gender, age, education level, occupation, ethnicity, and other selected determinants. Weights were created to account for the different selection probabilities of persons across sub-districts, communities and villages, age, and gender. The weight for each person i was expressed as follows: W kji = P k 6P j|k 6P i , where P k equals the reciprocal of the sampling weights of sub-districts k, P j|k equals the reciprocal of the conditional inclusion probability of community/village j within the selected sub-districts k, and P i is an adjustment factor for person i according to the age and gender composition of the 2010 census population of Wuwei City obtained from the local statistics bureau. The prevalence rates of HBV seromarkers include the point estimates and their estimated 95% confidence intervals (95% CI). 95% CIs were compared for HBV marker prevalence of each variable; 95% CI which did not overlap were considered as statistically significant P = 0.05 [3].
Univariate and weighted multiple logistic regression analyses for HBV chronic infections, exposure to HBV, and vaccine-induced immunity were analyzed to determine the unadjusted and adjusted odds ratios (OR) and 95% confidential intervals (95%CIs) of selected determinant risk factors associated with HBV exposure/ infection among the population studied. Multivariate analysis provided useful insight into risk factors for both adults and children in addition to showing a clear effect of the selected risk factors.

Ethical considerations
The study protocol was approved by the Ethics Committee of FMMU and consent was obtained from all participants or parents/guardians. All participants or parents/guardians signed the informed consent and provided their written consent on the first page of the questionnaire in this study.

Prevalence of chronic HBV infection and exposure characteristics
The prevalence of chronic HBV infection in children and adolescents were 1.3% (aged 1-4 year age group), 2.5% (5-9 year age group), and 3.4% (10-14 year age group), respectively. Prevalence increased with age, peaking in the 40-49 year old age group (8.7%), and decreasing in the 50-59 (8.5%), 60-69 (6.2%), and the $70 (5.10%) year-old age groups (Table 1). HBV infection was more common in men (7.9%) than in women (6.9%). Cadre was the predecessor of the civil servants/public officers, they had to pass medical examination before they were hired, the HBsAg positive one was not to be hired before 2010. Cadre had the lowest prevalence rate (4.4%) followed by health care workers (4.6%), and farmers had the highest prevalence (9.1%). Prevalence rates based on ethnicity were as follows: individuals of Han decent had the highest HBsAg prevalence (7.3%), followed by individuals of Tibetan decent (5.8%), other minority ethnic groups (5.2%), and last individuals of Uigur decent (4.7%). Among persons aged 15-59 years, the lowest prevalence was in the junior college and undergraduate degree population (5.8%). Unvaccinated individuals had higher prevalence rates of HBsAg (8.5%) compared to levels observed in HBV vaccinated individuals (5.0%, OR = 0.6, CI: 0.5-0.7). Persons who had household contact with HBV carriers (OR = 2.6, CI: 2.2-3.0) had the highest HBsAg prevalence rate (15.3%) among all subgroups examined. Having household contact with HBV carriers (OR = 3.8, 95% CI:2.2-6.5) and lack of hepatitis vaccination (OR = 2.0, 95% CI: 1.4-3.3) were the risk factors for HBV infection in children aged 1-14 years.

Prevalence of HBV exposure based on age
Prevalence of exposure to HBV increased with age from 14.1% among persons 5-9 years of age to 63.1% among persons .69 years of age (Table 2). Exposure to HBV was similar between men (43.9%) and women (43.9%), but more common in individuals of Han (44.2%) and Uigur (43.1%) decent than in persons of Tibetan (24.6%) decent or of other ethnicities (31.5%) ( Table 2). The prevalence was higher in persons who were living with partners, were divorced, widowed and had household contact with HBV carriers. High HBV exposure rates were observed in rural migrant workers or other temporarily employed individuals (53.9%) having a history of acupuncture, surgery, endoscopic examination, beauty treatments in public places, and having received dental work.

Discussion
In northwest China, hepatitis B infections represent one of the most important threats to public health accounting for approximately 43% of notifiable infections yearly [16]. In this study, the prevalence of HBsAg was 7.2% in the general population and  2.3% of the 1-9 year olds had chronic HBV infections. The HBV infection rates in Wuwei City remain a serious concern even after the implementation of the routine HBV immunization program established in 2002 and the HBV catch-up vaccination program for children and adolescents established in 2009. The prevalence of chronic HBV infection in Wuwei City in 2010 was 7.4% among persons 1-59 years of age, which is slightly higher than previous national survey results obtained in 2006 (7.2%). However, the prevalence of HBsAg among children 5-9 years of age in Wuwei City was significantly higher than rates reported from a national survey in 2006 for children of a similar age, possibly due to the lower vaccine coverage rates at birth between 2001-2005 in western China than in eastern China [6,17], which resulted in increased HBV infection risks [7,18]. Having household contact with HBV carriers has been considered to be important risk factors for hepatitis B infection, which included mother-to-child transmission and other members transmission. The OR value of household contact with HBV carriers was 2.6 for the all the participants, 3.8 for 1-14 years of age and 3.0 for the 15-59 years of age, which suggested that this risk factor was more important for children than adults, the OR value was similar with the study of acute hepatitis B in Shanghai [15]. A recent study in Wuwei City indicated that the rate of mother-tochild transmission of HBV was still high (7.3%) and that the absence of HBV vaccination after birth was a possible independent risk factor for increased rates of mother-to-child HBV transmission [19]. Lack of HBV vaccination was the risk factor of HBV chronic infection and exposure to HBV for all the participants. Beauty treatments, which included barber shop shaving, tattooing and ear piercing, were common in western China. In this study, 26.3% (7,374/28,044) of the general population had the history of       beauty treatments, though these public places lacked of effective health surveillance compared with medical units [20]. Exposure to HBV (past or chronic infections) in the Wuwei population was 7 times more common than other chronic infections, consistent with the fact that most individuals exposed to HBV in adulthood are able to ''clear'' the virus without persistence of HBsAg in serum. The prevalence of HBV exposure was 41.2% among persons 1-59 years of age and was significantly higher than rates described in the national survey in 2006 (34.1%), especially for children 1-14 years of age. These observations suggested that Wuwei City was endemic for HBV with prevalence rates greater than the national average. The prevalence of anti-HBc antibodies among children 1-4 years of age was higher than children 5-9 years of age. The higher prevalence of anti-HBc antibodies in children may partly be due to the immunization status of their mothers since the prevalence of anti-HBc positive mothers (43.1%, CI: 39.3-46.8%) was significantly higher than the prevalence of anti-HBc negative mothers (30.9%, CI: 27.6-34.1%) which decreased with age [21]. The OR value of household contact with HBV carriers for exposure to HBV was 1.8 for the all the participants, 5.4 for 1-14 years of age and 1.9 for the 15-59 years of age, which showed that household contact was more important for children than adults. Invasive medical procedure (acupuncture, surgery and endoscopy examination) and blood transfusion were also the risk factors for expose to HBV, which suggested the medical factors were also played an important role for the expose to HBV and health surveillance to the medical units should be strengthened.The prevalence of anti-HBs was 49.5% among persons 1-59 years of age which was slightly lower than rates reported in the national survey of 2006 (50.1%). However, the prevalence rates among children 1-4 years of age was significantly lower than rates reported in the national survey of 2006 [3]. The prevalence of anti-HBs sharply declined from 66.5% in 1-year olds to 39.3% in 4-year olds and a significant decrease in    serum anti-HBs observed within the first 3 years after vaccination may have been due to administration of low vaccine dosages. A recent large population study in China showed that the antibody response rate following administration of 10 mg/dose hepatitis B vaccine was higher than the response elicited by the 5 mg/dose administered to newborns, suggesting that raising the HBV vaccine dose may be necessary in infants born in endemic areas [22]. The prevalence among children 15-19 years old was significantly higher than the national average reported in 2006 due to catch-up vaccinations administered to individuals 14-15 years old and individuals 15-24 years old attending junior colleges during 2009-2010.
The differences in prevalence rates among individuals presenting with vaccine-induced immunity gradually disappeared between rural and urban areas, and in different regions among persons ,20 years of age as the HBV vaccination cost diminished to no charge for newborns and other select populations. Although the prevalence of chronic infections and HBV exposure in Wuwei City was 27 and 9 times higher, respectively, than rates observed in the United States between 1999-2006, the prevalence of vaccine-induce immunity in Wuwei City was slightly higher than United States rates between 1999-2006 [4,23] suggesting that the comprehensive HBV vaccination was successful. However, the prevalence of vaccine-induce immunity in Wuwei City was significantly lower than rates reported in United States between 1999-2006 among [13][14][15][16][17] year-olds because about 15% of individuals in this age group acquired immunity as a result of infection. Among individuals .20 years of age, the majority (62.3%) presenting with anti-HBs antibodies acquired immunity following exposure to HBV.
Before the HBV vaccine was widely available in China, HBV infections were acquired rapidly either perinatally or transmitted between children 1-4 years of age [24]. After routine implementation of the HBV immunization program in 2002 the primary obstacle to overcome was vaccine administration to children of lower socioeconomic levels, which hampered the efficacy of the vaccination program in infants since some parents did not precisely follow the immunization schedules, especially at younger ages due to low socio-economic status and residence in remote areas that limited access to health and educational services [25,26]. Besides, the delayed administration of the first HBV vaccine dose to premature infants was a significant risk factor for HBV motherto-child transmission in individuals living in Wuwei City [19]. Because a high proportion of women of childbearing age (15-49 years of age) remain HBsAg positive (7.2%, CI: 6.0-8.5%) the national program must continue to improve immunization practices if higher levels of vaccination are to be reached. Specifically, timely administration of the HBV vaccine combined with hepatitis B immune globulin need to be administered to at risk individuals [27].
After implementing the universal HBV immunization policy for all newborns, in addition to the catch-up programs for children and adolescents [28], protection of susceptible adults has become more important than ever in China. Sexual contact with chronically infected persons has been recognized as one of the most efficient routes of HBV transmission [29]. Data from this study showed high prevalence rates of HBsAg among sexually active persons from adolescence to mid-adulthood, strongly suggesting that sexual transmission of HBV represents a major infection route after disposable syringe programs were implemented, in addition to the implementation of safe blood transfusion practices [30]. Therefore, the coverage of the current HBV vaccination program needs to be expanded to include sexually active young adults in China. Considering the high mobility of the population and the nature of the vaccine immune response in this group, the traditional 5 mg/dose, 0-1-6 schedule used for immunizing infants is not suitable for adults living in an endemic area. To explore cost effectiveness, vaccination schedules designed to expand the vaccination coverage and duration of protection in adults is a priority [31,32].
The present study has some limitations. First, the study design did not account for occult infections, that is, individuals seronegative for HBsAg and acute infections that would preclude detection of anti-HBc antibodies [33]. Second, some selection bias may have occurred since 31.1% of potential participants were excluded before the interview since they worked outside of the city as rural migrant workers (in most cases), and this was most true status in western China. This could have caused an underestimation in the prevalence of HBsAg since most of these individuals presented with high levels of HBsAg (primarily male farmers, 16-40 years of age). This highly mobile group represents both victims and carriers of HBV infections. This is a growing population in western China which should be the target of prevention strategies [34]. Third, the risk factors associated with chronic HBV infections were obtained from a cross-sectional survey and should be verified using new HBV infection surveys because some risk factors (including the vaccination history for adults) occurred at a long time ago and was obtained based on memory, recall bias may exist. For these reasons, the prevalence of HBV infection is likely to be underestimated.
In conclusion, this population-based study in western China provided information that will be useful in planning preventive strategies. Although vaccination against HBV infection has become a standard practice, we found that the prevalence of HBV was as high as 7.2%. HBV infections will continue to remain a serious public health problem in western China in the coming decades. Hepatitis B vaccine immunization strategies therefore need to be improved further and designed to include coverage of highly mobile adult populations.