Identification of acute vaccine-preventable hepatitis in individuals with chronic hepatitis in British Columbia between 1991 and 2007

1BC Centre for Disease Control, Vancouver, British Columbia; 2Department of Public Health Sciences, University of Toronto, Toronto, Ontario; 3School of Population and Public Health, University of British Columbia, Vancouver, British Columbia Correspondence: Dr Jane A Buxton, BC Centre for Disease Control, 655 West 12th Avenue, Vancouver, British Columbia V5Z 4R4. Telephone 604-707-2573, fax 604-707-2516, e-mail Jane.buxton@bccdc.ca Viral hepatitis causes a considerable burden of morbidity and mortality worldwide. Despite sharing a common name, hepatitis viruses vary in their modes of transmission and outcomes of infection. Hepatitis A virus (HAV) is transmitted fecal-orally, directly (1) or indirectly through ingestion of contaminated food or water (2,3) or, rarely, parenterally (4-6). Although transmission routes for hepatitis B virus (HBV) and hepatitis C virus (HCV) may be similar (7) in Canada, HBV is mainly transmitted sexually (8) whereas HCV is primarily transmitted parenterally, with the majority of incident and prevalent HCV infections occurring among injection drug users (IDUs) (3,9-11). HAV infection typically results in mild illness without sequelae; case fatality is usually low (0.1% to 0.3%), but may be higher in older persons and those with comorbidities (3). HAV infection has no chronic carrier state and confers lifelong immunity (2). The development of chronic HBV following acute HBV infection is inversely associated with age, with a 90% risk of chronic infection among HBV-infected infants and 1% to 5% among adults (3). People with chronic viral hepatitis who become acutely infected (suprainfected) with HAV or HBV are at an increased risk of liver failure (12); chronic HBV and HCV coinfection is associated with poorer outcomes and an accelerated progression of liver disease (7,13-15). Between 1991 and October 2007, 33,250 chronic HBV and 59,080 HCV cases were reported in British Columbia (BC). Both acute HAV and HBV are vaccine preventable, and these vaccines are provincially funded for certain susceptible populations including orIgInal artICle


Case identification and data management
Nominal data and personal health numbers were used to link cases of acute HBV and HAV reported between 1991 and October 2007 to previously reported cases of chronic HBV and HCV.If individuals had more than one HAV-, HBV-or HCV-positive result, the date of the first positive result was considered to be the date of identification.Acute HAV is diagnosed by positive immunoglobulin M antibody to HAV with a compatible clinical history or exposure.HBV is considered to be acute if a case is positive for hepatitis B surface antigen and immunoglobulin M antibody to hepatitis B core antigen, or if the individual tested positive for hepatitis B surface antigen and cleared it within six months in the context of compatible clinical history or probable exposure.If reported HBV infection was not classified as acute, it was considered to be chronic.Individuals with positive reports of acute hepatitis within 14 days of chronic hepatitis identification were excluded from the analyses to account for delays in laboratory testing and reporting.
On completion of data linkages using a computer algorithm, personal identifiers (first and last name and personal health numbers) were deleted, and the anonymized data were subsequently analyzed.Data were cleaned of cases with probable data entry errors.Because BC guidelines recommend that high-risk infants undergo testing one to six months after completion of HBV vaccine series (administered at zero, one and six months after birth), any HBV infections identified within seven months of birth were excluded.Similarly, because infants at risk for vertical transmission of HCV should be tested at or after six weeks of age, cases reported before the age of six weeks were excluded.
The study received approval from the Behavioural Research Ethics Board of the University of British Columbia (Vancouver, BC).

Statistical analyses
Data linkages and analyses were performed using SAS version 9.1 (SAS Institute Inc, USA).To determine the identification rate of hepatitis, P.E.O.P.L.E 32 Population Estimates and Projections were obtained from BC Stats.The mean population estimate during the observation period was used to calculate the cumulative identification rate.Because data were only available until October 2007 and no seasonal variations were found, the annual identification rate for 2007 was prorated.Differences in age at hepatitis identification and time between chronic and acute hepatitis detection were examined using t tests.ANOVA was used to test for differences in age at hepatitis diagnosis, and the time between chronic and acute hepatitis virus identification according to type of chronic hepatitis infection.To compare differences across sex and HAs, c 2 and Fisher's exact tests were used.A two-sided P<0.05 was considered to be statistically significant.

rESuLtS
Data cleaning removed a total of 90 cases; the majority were excluded to ensure temporality (ie, acute HAV or HBV diagnosis occurred after chronic hepatitis diagnosis [not before or within 14 days]).Between 1991 and October 2007, 30 BC residents with chronic HBV and 104 with HCV were subsequently reported to be infected with HAV (Table 1); of these, 13 had HAV after both chronic HBV and HCV.During this period, the overall cumulative population rate of HAV suprainfection identification was 3.1 per 100,000.In comparison, the cumulative identification rate for individuals diagnosed with HAV alone was 82.3 per 100,000, and the mean age at diagnosis was significantly younger than the suprainfected cases -31.6 years (95% CI 31.1 to 32.2) versus 35.9 years (95% CI 33.9 to 37.8).
In the same period, 162 individuals with chronic HCV were suprainfected with acute HBV (Table 2).The overall cumulative population rate of acute HBV/chronic HCV suprainfection was 4.1 per 100,000.There were significant differences across HAs for age at acute HBV diagnosis (P=0.02).In particular, residents of Interior HA were significantly older than Fraser HA residents diagnosed with acute HBV (mean age 43.6 years [95% CI 37.2 to 50.1] versus 34.5 years [95% CI 32.4 to 36.7]) (data not shown).Among BC residents diagnosed with acute HBV only, the cumulative identification rate was 39.2 per 100,000.
Most of the suprainfection burden was among individuals with chronic HCV (Figure 1).Among those who became infected with HAV, more BC residents were chronically infected with HCV than with HBV (P<0.0001).Furthermore, significantly more men than women developed hepatitis suprainfection (P<0.0001), and this relationship persisted for each hepatitis suprainfection.Among suprainfected BC residents, some sex differences persisted (Table 3).Women were significantly younger than men when they were diagnosed with HAV (P=0.02) and acute HBV (P=0.0002) after identification of chronic hepatitis.However, there were no significant sex differences for the mean time to acute hepatitis identification after diagnosis of chronic hepatitis coinfection.

DISCuSSIon
Between 1991 and October 2007, 121 HAV cases among individuals chronically infected with HBV or HCV, and 162 acute HBV cases among individuals chronically infected with HCV were identified in BC.Among chronic HCV cases, 0.2% were suprainfected with HAV and 0.3% were identified with acute HBV (data not shown).However, a substantial number of acute HAV and HBV cases likely go undetected and the number of acute HBV cases is underestimated.The Provincial Laboratory of Northern Alberta found that 53.1% of HCV-positive individuals had been previously infected with or immunized against HAV, and 44.3% had been infected with HBV.One-quarter had been exposed to both HAV and HBV (16).Among nonvaccinated adults 20 to 39 years of age in BC, 15.7% had been previously infected with HAV (17).Vancouver residents accounted for 23.1% of HAV suprainfection cases.In late 1997, an HAV outbreak among men who have sex with men (MSM) began in Vancouver, when the number of HAV cases reported at the end of the year increased by 100% compared with the previous few months (18).To address the outbreak, the Vancouver Coastal Health Authority established a mass HAV vaccination campaign for MSM (18).In 1999, an HAV outbreak occurred in Vancouver's Downtown Eastside, where residents included IDUs who have a high prevalence of HCV (5,19) and are at risk for both HBV and HCV through common routes of transmission.Vaccination uptake during outbreaks tends to be high (5), and HAV immunization policies have developed in response to trends in HAV infection (20); HAV vaccination has been free of cost for IDUs and MSM in BC since 1998 and 2001, respectively.Our data suggested that HAV vaccination of HCV-reactive persons has successfully reduced the number of HAV cases since its implementation.Although reported cases of HAV were declining before publicly funded immunization programs for high-risk groups, from 900 cases in 1992 to 478 cases in 2006; since 2001, fewer than 80 cases of HAV are reported annually and, in 2007, only 41 cases were reported in BC.In a previous study of risk factors for HAV in four of five regional HAs in BC, cases that were identified as IDUs, and hence likely had a high prevalence of HCV (21,22), declined from 34 in 1998, and 23 in 1999, to zero cases in 2001 to 2004 (20).Recently, HAV infection was found to be more common among travellers to developing countries (17).Therefore, HAV infections could be decreased by targeting immunization at travellers to HAV-endemic countries.
Although the HBV vaccine was made available to IDUs in 1992 and to HCV-reactive persons in 1997, its influence on temporal trends is less apparent.In 1998, the number of acute HBV cases among persons previously diagnosed with HCV increased (which may be a consequence of increased testing), before declining after 1998.In the United States, 79% of newly acquired HBV cases are associated with high-risk sexual activity and injection drug use (23).To increase vaccination coverage among high-risk adults, HBV vaccine could be offered in settings in which a high proportion of adults have HBV risk factors, including sexually transmitted infections/ HIV testing, and treatment centres and primary care facilities serving MSM and IDUs (21).In BC, HBV vaccine has been funded for routine vaccination among grade 6 students since 1992, and infants born on or after February 1, 2001.This translates into lower acute HBV incidence, and declines have already been noted in adolescents and young adults within 10 years after its inception, with fewer than 50 cases annually in BC since 2006 (24).
Women were significantly younger when they were diagnosed with acute hepatitis.Females are more likely to get tested (25) and, in BC, women are routinely tested for HBV during pregnancy.However, more men develop hepatitis suprainfection because a greater number of acute HBV and HCV cases occur among males.Males are less likely to interact with the health care system (26) and, thus, may not be immunized against vaccine-preventable hepatitis following chronic hepatitis diagnosis, which may contribute to more suprainfections among males.Timeliness of vaccine administration following hepatitis diagnosis is imperative.Among individuals chronically infected with HCV, early HBV immunization is important because they may mount a better immune response to the vaccine before the onset of cirrhosis (27).Some cases of acute HBV infection in persons with HCV may have previously received HBV vaccine, but did not mount an adequate response; again this supports the need for early follow-up and immunization of identified individuals with HCV.

Characteristic
Acute HBV/chronic HCV Acute HBV only Our findings have some limitations: HAV, HBV and HCV infections are underdiagnosed and under reported.Because HAV may result in mild illness, individuals may not suspect infection and, thus, will not seek testing.Among individuals expressing symptoms of acute hepatitis, physicians will generally screen for all hepatitis viruses.
Most HCV tests in BC are performed at the BC Centre for Disease Control (Vancouver, BC); therefore, quality can be assured.HCVpositive sera are reflex tested for HAV and HBV immune status to determine vaccine eligibility; these results are included in the HCV report and sent to both the requesting physician and local public health.However, HBV testing occurs at hospital and private laboratories throughout the province, and determination of HAV immune status is not routinely performed in persons with HBV-positive results.In 1996 and 1997, the number of people seeking testing for hepatitis increased because they received notification to be tested due to past receipt of a blood product.HAV outbreaks were also occurring, which could explain the increase in the number of positive acute hepatitis cases identified among the chronically infected in 1998, but not the subsequent decline.The numbers presented may be a conservative estimate of the true burden of hepatitis suprainfection as a result of misclassification.
Despite these limitations, the data highlight the value of timely public health follow-up.In BC, public health follow-up is performed at the local level and may be dependent on local policies, resources and priorities.Acute HAV and HBV can be prevented through immunization.HAV suprainfection has declined considerably since 1998/1999, when publicly funded HAV vaccine was introduced for persons with HCV and HBV, indicating that BC's immunization policy is effective.However, acute HBV infection after HCV diagnosis continues, despite the introduction of publicly funded HBV vaccine for IDUs (at high risk for HCV) in 1992 and in 1997 for persons with HCV.Vaccination coverage for HBV may be improved among IDU and HCV-reactive persons by comprehensive case follow-up when HCV infection is identified.However, because persons with HCV may not use primary care regularly, health care facilities frequented by this vulnerable population can be targeted and immunization status assessed at all health care encounters.