Epidemiology of hepatitis B infection in the Northern Territory of Australia

Abstract

From August 1989 to may 1993, extensive studies on the epidemiology of hepatitis B virus (HBV) infection in the Northern Territory were conducted by means of community-based surveys, morbidity and mortality data review and mathematical modelling. The incidence of clinically apparent acute HBV infection in the Top End was 12 per 100,000 per annum, with a marked ethnic difference between Aborigines (42 per 100,000) and non—Aborigines (4 per 100,000), and an odds ratio of 10.4 (95% confidence intervals 3.2-33.8). Sixty percent of Aboriginal cases of acute hepatitis B occurred in children less than 10 years of age, whereas non-Aboriginal cases occurred in adults aged 20-29, mostly with behavioural risk factors. These findings confirm the importance of immunising Aboriginal children to reduce the future incidence of hepatitis B infection and hepatoma.

Average incidence rate of primary hepatocellular carcinoma (PHC) was 5.2 per 100,000 for the Aboriginal population and 0.5 per 100,000 for the non-Aboriginal population, giving a relative risk of 10.4 (95% confidence intervals, 4.0—26.7). However, the incidence of PHC was not significantly different between Aboriginal Australians aged 40 and over (30.9 per 100,000) and persons from high risk areas such as Southern Europe, Asia or the Pacific (18.9 per 100,000) (X2l = 0.97, P > 0.05). PHC was more frequent in males (2.3 per 100,000) than in females (0.7 per 100,000), with the relative risk of 3.4 (95% confidence intervals 1.3—9.3). The incidence of PHC increased with age in Aboriginal and non—Aboriginal people; the trend was statistically significant in Aborigines (X2l =. 4.74, P < 0.05) but not in non-Aborigines (X2l = 3.43, P > 0.05). The study showed that 63.6% of Aboriginal cases and 50.0% of non-Aboriginal cases of PHC had evidence of hepatitis B surface antigen positivity. This prevalence of the virus in Aboriginal patients with carcinoma (63.6%) was much higher than that seen in Aborigines from community-based surveys (13.1%) (X2l = 21.7, P < 0.001). The results indicate the importance of hepatitis B in the aetiology of PHC in Aboriginal people, and the importance of immunisation for its prevention.

Serological markers of HBV infection were detected in 28.7% of‘school children (46.9% of 439 Aborigines, 13.7% of 556 children from the "low prevalence" groups and 32.1% of 109 "other" ethnic groups). There was evidence of HBV infection in 12.8% of school staff from "low prevalence" ethnic backgrounds and in 37.9% of teachers from other ethnic (including Aboriginal) backgrounds. in Northern Territory schools, the prevalence of HBV infection is high in children and school staff from ethnic groups previously known to be at higher risk of HBV infection. For students and staff from ethnic backgrounds expected to be at low risk, HBV prevalence is greater than in individuals from similar backgrounds in other parts of Australia. These data suggest that horizontal transmission of HBV occurs in school settings in the Northern Territory, and strengthen the rationale for HBV immunisation of all children and staff at schools. Such a policy would also be applicable to other situations where there is contact between ethnic groups with high and low rates of HBV infection.

The immunogenicity of a recombinant hepatitis B vaccine (Engerix B) was examined in Aboriginal children four years after the introduction of universal vaccination for Aboriginal infants. Among 612 vaccinees, 375 (61.3%[95% confidence intervals 57.3% — 65.1%l) seroconverted (anti—HBs titres greater than or equal to 100 mIU/ml). There was a rapid decline in anti-HBs levels in Aboriginal children with time since immunisation; 7.8% of children appeared to revert to seronegative status for each year that had elapsed since the first vaccine dose. The overall response to hepatitis B vaccination in Aborigines was less than in other reported groups; this may be due to cold-chain failure in specific communities, to impaired responses reflecting malnutrition and genetic influence, and to titres which fall more rapidly over time.

Multivariate analysis showed that geographic region, increasing birthweight and delayed timing of the first vaccine dose after birth were predictors of increased antibody response to the vaccine; but that timing of the second and third doses did not significantly affect response. These findings have practical implications; firstly, the considerable flexibility in timing of hepatitis B vaccine doses makes it possible to integrate this vaccine with other childhood vaccines and to resume vaccination schedules when vaccinees miss the second or third doses; secondly, there is a need to evaluate the requirement for a booster dose of vaccine in Aboriginal children; and to improve the cold chain in those communities where seroconversion rates were lowest.

Epidemiological models were used to estimate rates of horizontal and vertical transmission of hepatitis B virus in a mixed race Australian community.