Epidemiology of asthma and bronchial hyperresponsiveness in Auckland children

Abstract

In New Zealand (NZ) asthma mortality rates are higher than in any other developed country studied, and hospital admission rates for asthma are among the highest. Within NZ, both mortality and hospital admission rates for asthma are higher among Polynesians than among Europeans. To investigate the possibility that these differences were due to differences in asthma prevalence, a study was initiated to measure the prevalence and severity of asthma among both Europeans and Polynesians in Auckland using methods identical to those used in an overseas study. From all primary schools in the Auckland Urban Area, classes were sampled by two-stage, random, stratified sampling. In the winter of 1985, the children in the sampled classes were studied employing a questionnaire and a rapid, portable method of bronchial challenge testing with histamine which had been used to study schoolchildren aged 8 to 10 years in two areas in New South Wales (NSW), Australia. Participation was 83%, the participants comprising 1084 European, 509 Maori, and 460 Pacific Island children, who were closely matched in gender and age to the NSW children. Prevalence of various items was compared including asthma symptoms, diagnosed asthma, and bronchial hyperresponsiveness (BHR), defined as a 20% fall in the one-second forced expiratory volume (FEV1), on or before the administration of a cumulative dose of 7.8 µmol histamine. The prevalence rates of respiratory symptoms, BHR, severe BHR and BHR combined with symptoms were similar among the 1084 Auckland European children and 769 inland NSW children, but lower among 718 coastal NSW children than the other two sites. In Auckland, the prevalence of respiratory symptoms was higher in Maoris than among Europeans or Pacific Islanders, whereas prevalence of diagnosed asthma was similar in all three groups. Europeans had the highest rate of BHR and severe BHR followed by Maoris then Pacific Islanders. Maori and Pacific Island children reporting symptoms were less likely to receive prophylactic medication than European children. The study design also enabled assessment of the epidemiological correlates of BHR, which had been little studied in random populations before 1985. Among the Auckland children, bronchial responsiveness was found to be unimodally distributed, with no clear separation between children with a diagnosis or symptoms of asthma and those without. The presence or absence of BHR did not accurately identify children with a diagnosis or symptoms of asthma. A l4-variable multiple logistic regression model was developed which explained 27% of the variability in BHR rates. The main variables independently increasing BHR risk were: low baseline FEV1, a history of wheeze, diagnosed asthma, atopic conditions, or bronchodilator treatment, and European (as opposed to Polynesian) ethnicity. Climatic conditions at the time of the test were less strongly associated. In summary, neither the international nor the inter-ethnic differences in asthma mortality and asthma admission rates were adequately explained by differences in the prevalence of asthma symptoms, diagnosed asthma or BHR; however, differences in asthma severity and in the management of asthma may be contributory. In this large population-based study of children, BHR was associated independently with a number of different variables but did not precisely identify children reporting asthma symptoms or a diagnosis of asthma. Thus a single bronchial challenge test is not an ideal tool on its own for diagnosing asthma or measuring asthma prevalence.