HIV Articles  
Back 
 
 
Epidemiology of Hepatitis B Virus Infection in a US Cohort of HIV-Infected Individuals during the Past 20 Years
 
 
  Clinical Infectious Diseases Feb 1 2010;50:426-436
 
"Liver-related complications have become an increasingly important cause of morbidity and mortality in human immunodeficiency virus (HIV)-infected patients since the advent of highly active antiretroviral therapy (HAART) [1, 2]; 1 study showed that 15% of deaths of HIV-infected adults were related to liver disease [3].......This comprehensive examination of the epidemiology of HBV infection in a large cohort of HIV-infected individuals highlights the continued and significant burden of HBV infection in HIV-infected adults, with nearly 40% of patients with HIV infection also having coinfection with HBV. Despite the gradual decrease in prevalence of chronic HBV infection over the past 2 decades, overall, 11% of patients with HBV infection had chronic HBV infection..........More worrisome is the finding that HIV-infected individuals are continuing to develop incident HBV infections at a rate that has remained unchanged for the past 8 years and appears to result from continued high-risk sexual behavior. Effective prevention methods are needed and must overcome the effects of HIV-associated immune dysfunction on HBV transmission and vaccine effectiveness. The associations of HBV-active HAART use and higher CD4 cell count with decreased risk of incident HBV infection suggest additional benefits of HAART and provide further rationale for increased use of HBV-active HAART.....Active coinfection with HIV and HBV significantly influences the use of HAART in such patients, impacting the selection and timing of antiretroviral therapy (ART) initiation [9, 10] and increasing the risk of HAART-related hepatotoxicity [11, 12, 13].....The prevalence of positive HBV serological markers has been reported to be as high as 80% in some HIV-infected populations, with a prevalence of chronic HBV infection of up to 10%......Of the 1872 participants who were HBV negative at the time of diagnosis of HIV infection, 181 (9.7%) became HBV infected during follow-up, and of those patients, 37 (20.4%) developed chronic HBV infection (compared with the 7.2% of patients with chronic HBV infection at the time of diagnosis of HIV infection; P<.001)."
 
Updated Dec 1 2009: DHHS Panel's Recommendations:
· Antiretroviral therapy should be initiated in all patients with a history of an AIDS-defining illness or with a CD4 count <350 cells/mm3 (AI).
· Antiretroviral therapy should also be initiated, regardless of CD4 count, in patients with the following conditions: pregnancy (AI), HIV- associated nephropathy (AII), and hepatitis B virus (HBV) coinfection when treatment of HBV is indicated (AIII).
· Antiretroviral therapy is recommended for patients with CD4 counts between 350 and 500 cells/mm3. The Panel was divided on the strength of this recommendation: 55% voted for strong recommendation (A) and 45% voted for moderate recommendation (B) (A/B-II).
· For patients with CD4 counts >500 cells/mm3, the Panel was evenly divided: 50% favor starting antiretroviral therapy at this stage of HIV disease (B); 50% view initiating therapy at this stage as optional (C) (B/C-III).
· Patients initiating antiretroviral therapy should be willing and able to commit to lifelong treatment and should understand the benefits and risks of therapy and the importance of adherence (AIII). Patients may choose to postpone therapy, and providers, on a case-by-case basis, may elect to defer therapy based on clinical and/or psychosocial factors.

 
Rating of Recommendations: A = Strong; B = Moderate; C = Optional Rating of Evidence: I = data from randomized controlled trials; II = data from well-designed nonrandomized trials or observational cohort studies with long-term clinical outcomes; III = expert opinion
 
Helen M. Chun,1 Ann M. Fieberg,3,4 Katherine Huppler Hullsiek,3,4 Alan R. Lifson,3,4 Nancy F. Crum-Cianflone,2,4 Amy C. Weintrob,4,6 Anuradha Ganesan,4,5 Robert V. Barthel,7 William P. Bradley,4,8 Brian K. Agan,4 and Michael L. Landrum4,8 and the Infectious Disease Clinical Research Program HIV Working Groupa
 
1Naval Health Research Center and 2Naval Medical Center San Diego, San Diego, California; 3University of Minnesota, Minneapolis; 4Infectious Disease Clinical Research Program, Uniformed Services University of the Health Sciences, and 5National Naval Medical Center, Bethesda, Maryland; 6Walter Reed Army Medical Center, Washington, DC; 7Naval Medical Center Portsmouth, Portsmouth, Virginia; and 8San Antonio Military Medical Center, Fort Sam Houston, Texas
 
ABSTRACT
 
Background. The epidemiologic trends of hepatitis B virus (HBV) infection in human immunodeficiency virus (HIV)-infected patients over the past 20 years are largely unknown.
 
Methods. Prevalence and risk factors for HBV infection overall, at the time of HIV infection, and after HIV infection were examined in an ongoing observational HIV cohort study. Risk factors for HBV infection at the time of diagnosis of HIV infection were evaluated using logistic regression, and risk of incident HBV infection after diagnosis of HIV infection was evaluated using Cox proportional hazards models.
 
Results. Of the 2769 evaluable participants, 1078 (39%) had HBV infection, of whom 117 (11%) had chronic HBV infection. The yearly cross-sectional prevalence of HBV infection decreased from a peak of 49% in 1995 to 36% in 2008 (p<.001). The prevalence of HBV infection at the time of diagnosis of HIV infection decreased during 1989-2008 from 34% to 9% (p<.001). The incidence of HBV infection after diagnosis of HIV infection decreased from 4.0 cases per 100 person-years during the pre-highly active antiretroviral therapy (HAART) era to 1.1 cases per 100 person-years during the HAART era (p<.001); however, this incidence remained unchanged during 2000-2008 (p=.49), with >20% of HBV infections occurring after HIV infection being chronic. Decreased risk of HBV infection after diagnosis of HIV infection was associated with higher CD4 cell count and the use of HBV-active HAART. Receipt of >1 dose of HBV vaccine was not associated with reduced risk of HBV infection after diagnosis of HIV infection.
 
Conclusions. Although the burden of HBV infection overall is slowly decreasing among HIV-infected individuals, the persistent rate of HBV infection after diagnosis of HIV infection raises concern that more-effective prevention strategies may be needed to significantly reduce the prevalence of HBV infection in this patient population.
 
Liver-related complications have become an increasingly important cause of morbidity and mortality in human immunodeficiency virus (HIV)-infected patients since the advent of highly active antiretroviral therapy (HAART) [1, 2]; 1 study showed that 15% of deaths of HIV-infected adults were related to liver disease [3]. Coinfection with hepatitis B virus (HBV) is a well-recognized cause of liver-related complications in individuals with HIV infection and is associated with an increased risk of mortality [4, 5, 6, 7, 8]. Active coinfection with HIV and HBV significantly influences the use of HAART in such patients, impacting the selection and timing of antiretroviral therapy (ART) initiation [9, 10] and increasing the risk of HAART-related hepatotoxicity [11, 12, 13].
 
Because of the significant burden and clinical impact of HBV in HIV-infected individuals, understanding of the epidemiologic trends and risk factors associated with HBV infection in HIV-infected populations is crucial. The prevalence of positive HBV serological markers has been reported to be as high as 80% in some HIV-infected populations, with a prevalence of chronic HBV infection of up to 10% [6]. However, currently available data are somewhat limited. Previous investigations defined chronic HBV infection with use of a single positive hepatitis B surface antigen (HBsAg) result, without including other serological markers, and reported only prevalent cross-sectional data [7, 14, 15, 16, 17, 18, 19]. Data on incident HBV infection, especially in HIV-infected individuals, remain limited; however, such data could improve understanding of HBV transmission and offer opportunities for improving prevention of HBV infection in HIV-infected adults. One study reported the incidence of acute HBV infection in a large HIV cohort as 12.2 cases per 1000 person-years and the prevalence of chronic HBV infection as 7.6% [20]. That study, however, used a window of observation during the period 1998-2001, and the proportion of infections that ultimately became chronic was not reported.
 
Although the overall prevalence of HBV infection in the United States is likely to be underestimated [21, 22], data indicate a decreasing overall incidence of acute HBV infection in the general US population over the past 2 decades [23, 24]. However, similar data assessing HBV infection trends among HIV-infected individuals have not been published to our knowledge. Therefore, the aim of the present study was to describe the epidemiology of HBV infection in a prospective, observational cohort, the US Military HIV Natural History Study (Department of Defense [DoD] NHS). Our specific objectives were to (1) examine the prevalence of HBV infection by year, (2) determine the prevalence of and factors associated with HBV infection present at the time of diagnosis of HIV infection, and (3) evaluate the incidence of and factors associated with HBV infection after diagnosis of HIV infection.
 
Discussion
 
This comprehensive examination of the epidemiology of HBV infection in a large cohort of HIV-infected individuals highlights the continued and significant burden of HBV infection in HIV-infected adults, with nearly 40% of patients with HIV infection also having coinfection with HBV. Despite the gradual decrease in prevalence of chronic HBV infection over the past 2 decades, overall, 11% of patients with HBV infection had chronic HBV infection. Furthermore, HBV infections after diagnosis of HIV infection continue to occur, and although the number of such HBV infections has decreased in the HAART era, incidence rates have remained unchanged for the past 8 years, suggesting that additional efforts will be needed to further reduce the incidence rate of HBV coinfection. Lastly, susceptibility to HBV appears to be inversely correlated with immune function, further compounding HBV infection prevention efforts for HIV-infected individuals.
 
Decreased prevalence of HBV infection at the time of diagnosis of HIV infection, combined with decreased but now stable incidence of HBV infection after diagnosis of HIV infection, has resulted in a decreasing overall burden in our cohort, or yearly cross-sectional prevalence, of HBV infection. Unfortunately, despite the gradual decrease in the prevalence of HBV infection overall in our cohort, prevalence, including that of chronic disease, remains high. A previous US cohort study from the HAART era reported the prevalence of chronic HBV infection among HBV unvaccinated participants as 7.6% [20], which is similar to the 6.8% prevalence of chronic HBV infection among unvaccinated participants seen in the HAART era in our study. Other investigations have reported similar estimates for the prevalence of chronic HBV infection among HIV-infected individuals [6]. In addition, African American ethnicity was not clearly associated with increased risk of HBV infection; this result was similar to recent US trends showing decreasing racial disparities in incidence of acute HBV infection [23]. Furthermore, the association of immunologic function, as reflected by CD4 cell count, with susceptibility to HBV suggests that the high prevalence of HBV infection in this population may not be attributable to only similar modes of transmission of HIV and HBV, but also a possible increased susceptibility to HBV in HIV-infected individuals.
 
Our analysis helps improve the understanding of HBV transmission in individuals with HIV infection. For HBV infection diagnosed at the time of diagnosis of HIV infection, reflective of HBV transmission prior to diagnosis of HIV infection, data are encouraging. Prevalence has decreased significantly in the past 20 years and is continuing to decrease, which is perhaps reflective of recent US acute HBV infection epidemiologic trends [23, 24]. Because these infections occurred on an individual level before HIV-associated immune dysfunction, prior receipt of HBV vaccine was associated with reduced risk, and a minority of these infections became chronic. However, this model differs from that of HBV transmission after HIV infection. For HBV infections acquired after diagnosis of HIV infection and, therefore, after the development of HIV-associated immune dysfunction, 1 in 5 resulted in chronic HBV infection, and vaccine effectiveness may be reduced [30]. Although the interpretation of observational data must always be cautious, our results suggest that both increased use of vaccination in HIV-uninfected populations at high risk of infection and improved behavioral interventions in those with HIV infection would best reduce the burden of HBV infection in adults with HIV infection. Efforts to address only a single transmission model may be only partially effective in reducing the overall burden of HBV infection among HIV-infected individuals, and effective prevention methods in one model may not necessarily be effective in the other because of differences, including altered immunocompetence.
 
Although the incidence of HBV infection after diagnosis of HIV infection was generally lower in the HAART era, it remained unchanged during 2000-2008, about 100-fold higher than the rate of HBV infection in the US military, and 500-fold higher than the rate of acute HBV infection in the general US population [24, 31]. The reasons for a lack of continued decrease in the incidence of HBV infection in our cohort are not clear, but this observation is similar to recent trends for other STIs in the general US population [32, 33]. The association between HBV infection and STIs in our study supports the conclusion that HBV is acquired predominantly through sexual transmission in our cohort before and likely after diagnosis of HIV infection. The presumed sexual transmission of HBV infection after diagnosis of HIV infection emphasizes the need for improved methods of effective behavioral risk reduction counseling for patients with HIV infection. Various intervention methods have been shown to impact sexual behavior and reduce risk of incident STIs among HIV-infected and HIV-uninfected adults [34, 35]. Incorporating HIV infection prevention counseling in the routine medical care of HIV-infected individuals is recommended [36]. Such strategies are important for both prevention of HIV transmission and prevention of other coinfections, such as HBV infection. Without improved prevention efforts, incident HBV infection among HIV-infected individuals will likely continue to occur.
 
Intriguingly, the associations between a reduced risk of incident HBV infection and both higher CD4 cell count and use of HBV-active ART and/or HAART suggest that HBV susceptibility is impacted by both immunologic and virological factors. Our results agree with a previous investigation that found reduced risk of acute HBV infection among HIV-infected individuals associated with use of HAART and other ART [20]. Our findings suggest that there is a potential HBV preventive benefit of HBV-active HAART in HIV-infected individuals not infected with HBV and provide evidence of a possible nontraditional benefit of HAART—specifically, reduced susceptibility to HBV infection in an individual and reduced incidence of HBV infection in a population.
 
As a cohort analysis, our study has limitations. Some characteristics of the DoD NHS cohort may limit comparison with other patient groups, including the limited number of female patients and injection drug users, the inability to systematically collect data on sexual risk behavior, and the high CD4 cell counts of participants at the time of diagnosis of HIV infection because of routine military screening. However, we feel that our findings are generalizable because our participants, both active duty and civilian, are representative of a large cross-section of the US adult population, and our findings are in agreement with epidemiologic data regarding HBV infection and other STIs in the United States. Second, HBV serologic findings in HIV-infected individuals may fluctuate [37], making diagnosis and classification for investigation difficult; however, participants in our study received periodic testing of HBV serological markers, and our criteria for inclusion and definitions for infection required results from multiple serological examinations. In addition, as an observational study, ART and vaccinations were not randomly administered, precluding our ability to determine the direct effects of either ART or vaccination on HBV acquisition. Finally, some groups experienced relatively few events, limiting some comparisons and conclusions.
 
Among HIV-infected individuals in the United States, the prevalence of HBV infection, including chronic HBV infection, is gradually decreasing overall; however, it remains substantial. More worrisome is the finding that HIV-infected individuals are continuing to develop incident HBV infections at a rate that has remained unchanged for the past 8 years and appears to result from continued high-risk sexual behavior. Effective prevention methods are needed and must overcome the effects of HIV-associated immune dysfunction on HBV transmission and vaccine effectiveness. The associations of HBV-active HAART use and higher CD4 cell count with decreased risk of incident HBV infection suggest additional benefits of HAART and provide further rationale for increased use of HBV-active HAART.
 
Results
 
Overall prevalence of HBV infection.Among the 2769 participants included in the analysis, the median number of HBV screenings per participant was 4 (IQR, 2-9), and the median time between HBV screenings was 6.4 months (IQR, 5.6-10.1 months). The median time from the last HBV screening to censoring was 1.2 months (IQR, 0-12.2 months). Characteristics of eligible participants are shown in Table 1. Crude mortality was higher during the pre-HAART era (28.2%) than during the HAART era (1.8%); however, it was similar by HBV status (pre-HAART era, 29.0% among HBV-infected persons and 27.4% among HBV-uninfected persons; HAART era, 3.5% among HBV-infected persons and 1.2% among HBV-uninfected persons). Annual rates of enrollment and study discontinuation in the pre-HAART and HAART eras were similar, regardless of HBV status (data not shown).
 
Of the 2769 participants, 1078 (38.9%) were HBV infected over the entire observation period, and of those patients, 117 (10.9%) had chronic HBV infection. During each year, about 40% of patients seen were infected with HBV (Figure 2). The cross-sectional prevalence of HBV infection and chronic HBV infection peaked in 1995 (49.2% and 6.8%, respectively). Since 1996, the cross-sectional prevalence of both HBV infection and chronic HBV infection have decreased significantly (p<.001 for each comparison). The overall prevalence of chronic HBV infection among unvaccinated and vaccinated participants seen in the HAART era was 6.8% and 2.3%, respectively (p<.001). The estimated prevalence of chronic HBV infection reached its lowest level in 2008 at 3.9% (95% CI, 3.1%-4.6%).
 
Prevalence of HBV infection at the time of diagnosis of HIV infection. Of the 1885 participants with recent HIV infection (HIV seroconversion 3 years earlier) and known HBV status at the time of diagnosis of HIV infection, 456 (24.2%) were HBV infected, and of those patients, 33 (7.2%) had chronic HBV infection. The prevalence of HBV and chronic HBV infection at the time of diagnosis of HIV infection decreased during 1989-2008 (p<.001 for HBV infection; p=.002 for chronic HBV infection) (Figure 3). In multivariate analysis (Table 2), a higher risk of HBV infection at the time of diagnosis of HIV infection was associated with older age (OR, 2.03; 95% CI, 1.72-2.40 per 10-year increase), male sex (OR, 7.15; 95% CI, 2.58-19.82), and a history of gonorrhea (OR, 1.92; 95% CI, 1.44-2.56) or syphilis (OR, 1.63; 95% CI, 1.11-2.38). Receipt of HBV vaccine prior to diagnosis of HIV infection was associated with reduced risk (OR, 0.33; 95% CI, 0.24-0.46). When included in the multivariate model, hepatitis C virus status (unknown for 28% of participants) was not associated with risk of HBV infection (HR, 1.56; 95% CI, 0.68-3.58); results including hepatitis C virus status were otherwise similar (data not shown). Separate multivariate analyses for participants who had received a diagnosis of HIV infection in the pre-HAART or HAART era had similar results (data not shown).
 
Incidence of HBV infection after HIV infection. Of the 1872 participants who were HBV negative at the time of diagnosis of HIV infection, 181 (9.7%) became HBV infected during follow-up, and of those patients, 37 (20.4%) developed chronic HBV infection (compared with the 7.2% of patients with chronic HBV infection at the time of diagnosis of HIV infection; p<.001). There were 2521 and 7301 years of follow-up available in the pre-HAART and HAART eras respectively. The rate of incident HBV infection in the pre-HAART era (4.0 cases per 100 person-years of follow-up; 95% CI 3.2-4.7 cases per 100 person-years) was significantly higher than that in the HAART era (1.1 cases per 100 person-years; 95% CI, 0.9-1.4 cases per 100 person-years; p<.001) (Figure 4). Similarly, the rate of chronic HBV infection in the pre-HAART era (1.2 cases per 100 person-years; 95% CI, 0.7-1.6 cases per 100 person-years) was significantly higher than the rate in the HAART era (0.12 cases per 100 person-years; 95% CI, 0.03-0.19 cases per 100 person-years; p<.001). The rate of incident HBV infection appeared to increase during 2000-2008, but this trend was not statistically significant (p=.49).
 
In time-to-event models (Table 3), multivariate risk of incident HBV infection was significantly increased for male patients (HR, 7.79; 95% CI, 2.86-20.86). Risk was significantly reduced for patients with higher CD4 cell counts during follow-up (HR, 0.90; 95% CI, 0.84-0.96 per 100-cell increase) and patients receiving HAART containing HBV-active agents (HR, 0.31; 95% CI, 0.19-0.51), compared with those not receiving ART. Patients receiving mono/dual therapy containing HBV-active agents also had reduced risk of incident HBV infection (HR, 0.23; 95% CI, 0.07-0.72), and those receiving HAART not containing HBV-active drugs did not have a significantly reduced risk of HBV infection, although there were only 3 and 4 HBV infections in these 2 subcategories, respectively. Having received >1 dose of vaccine was not associated with reduced risk of infection (HR, 0.86; 95% CI, 0.62-1.18). When included in the multivariate model, HIV RNA level during follow-up (unknown for 12% of participants, because this test was not widely available prior to 1996) was not associated with risk of HBV infection (HR, 1.14; 95% CI, 0.92-1.40 per 1 log10 copies/mL increase); results including HIV RNA level were otherwise similar (data not shown).
 
Methods
 
Participants. The DoD NHS is an ongoing, continuous enrollment observational cohort of HIV-infected DoD beneficiaries with >4900 participants that has been in existence since 1986 and is described elsewhere [25, 26]. All active duty military members and their adult dependents who have received a diagnosis of HIV infection and who are seen at participating sites and are able to provide written consent are eligible for participation. Because of military patient referral patterns and continued study participation as civilians after completion of military service, participants in the DoD NHS represent a broad cross-section of the adult US population from all regions of the country. HIV exposure category is not routinely captured, although a previous study suggested that injection drug use in this cohort is rare [27]. Approval for this research was obtained from the institutional review board at each participating site.
 
Definitions and inclusion. HBV infection was defined as meeting at least 1 of the following criteria: (1) concurrently reactive for HBsAg and hepatitis B core antibody (HBcAb), (2) concurrently reactive for HBcAb and hepatitis B surface antibody, or (3) reactive for HBsAg or HBcAb on 2 separate occasions. Chronic HBV infection was defined as having a reactive HBsAg on 2 separate occasions at least 6 months apart. A participant was classified as HBV uninfected at the time of diagnosis of HIV infection if both HBsAg and HBcAb tests were initially nonreactive on or after the date of diagnosis of HIV infection.
 
Because HBV screening became uniform in 1989, individuals enrolled before 1989 were excluded from all analyses. Participants without a documented date of diagnosis of HIV infection and those whose HBV infection status could not be determined at any time were also excluded (Figure 1). For the evaluations of HBV infection at the time of diagnosis of HIV infection and incidence of HBV infection after diagnosis of HIV infection, 2 different, partially overlapping subgroups of participants were defined a priori.
 
For analyses, ethnicity was categorized as white; African American; Hispanic, Puerto Rican, or Mexican; or other. An individual was considered to have a sexually transmitted infection (STI) if there was a medical history or laboratory evidence of syphilis, genital herpes, or gonococcal or chlamydial infection. The presence of an AIDS-defining illness was defined using 1993 Centers for Disease Control and Prevention criteria, with the exception of an isolated CD4 cell count <200 cells/µL [28]. The definition of HAART was similar to that used in previous studies [24]. ART was categorized as none, mono/dual ART, or HAART. Mono/dual ART and HAART categories were further subdivided into HBV-inactive and HBV-active therapy, with HBV-active therapy defined as the use of lamivudine, emtricitabine, or tenofovir. Hepatitis B vaccination was defined as receipt of 1 vaccine dose.
 
Design and statistical analysis. Descriptive statistics were used to describe the eligible participants overall and the 2 subgroups. Median values were given with interquartile ranges (IQRs). Proportions were compared using χ2 tests. To address the 3 main objectives of the study, 3 analyses were conducted. To determine the prevalence of HBV infection by year in our cohort, we examined the participants meeting the aforementioned criteria for overall inclusion. For these participants, the yearly cross-sectional prevalence of HBV infection and chronic HBV infection was calculated by determining the proportion of participants seen in a particular year who had serological evidence of HBV infection during of before that year. Changes in prevalence over time were examined using the Cochran-Armitage test for trend.
 
For the second objective, the prevalence of and risk factors for HBV infection at the time of diagnosis of HIV infection were determined for the subgroup of participants with known recent HIV infection (defined as having a documented HIV seroconversion window of <3 years) and known HBV status within 6 months after diagnosis of HIV infection. Univariate and multivariate logistic regression models were used to examine risk of HBV infection in this group. For sensitivity analyses, the models were repeated separately according to the era during which HIV infection was diagnosed: the pre-HAART era (before 1996) or the HAART era (1996 or later).
 
For the third objective, the incidence of and risk factors for HBV infection after diagnosis of HIV infection were determined for all eligible HIV-infected participants (not just persons who experienced HIV seroconversion) who were known to be HBV uninfected at the time of diagnosis of HIV infection. Participants were followed up from diagnosis of HIV infection to HBV infection or the last study visit. The number of HBV and chronic HBV events, person-years at risk, and rates (cases per 100 person-years of follow-up) were calculated for the entire study period and for specific time intervals (pre-HAART and HAART eras). Poisson regression analyses were used to test differences in rates during those intervals. Univariate and multivariate Cox proportional hazards models, stratified by era of diagnosis of HIV infection (pre-HAART or HAART era), were used to explore the relationship between factors and risk of incident HBV infection. Time-updated covariates used all available measurements during the period of observation.
 
Significance was defined a priori as p<.05; all P values were 2-sided. Age, sex, and ethnicity were included in all multivariate models, in addition to variables that were statistically significant in univariate analyses (at p<.05). Odds ratios (ORs), hazard ratios (HRs), and rates are presented with 95% confidence intervals (95% CIs). All analyses were conducted using SAS software, version 9 (SAS Institute). All figures were generated using the R programming language, version 2.0.0. The "loess" and "predict.loess" functions in R were used to generate the local linear robust fit smoothing lines and pointwise 95% CIs depicting both the prevalence and incidence of HBV infection and chronic HBV infection [29].
 
 
 
 
  iconpaperstack view older Articles   Back to Top   www.natap.org