Chronic Disease Prevalence in Migrant Groups

Five waves of the HABITAT survey data were analyzed to determine the prevalence of self-reported chronic diseases and behavioral risk factors over 9 years by group country of birth: Australian-born, immigrants from high-income countries, and immigrants from low–middle-income countries. Poisson regression was used to estimate the prevalence ratios (PRs) because of its ability to estimate PR consistently and effectively in prospective studies. Poisson regression provides better analysis than logistic regression because of providing unbiased, more interpretable, and easier to communicate ratios [20 (link)]. To avoid overestimating the error of the estimated risk, a robust error variance procedure (sandwich estimation) was used (Zou, 2004). Preliminary bivariate analyses indicated that employment status, income, and education qualifications were not significantly associated with any of the diseases or risk factors under study. All regression models were adjusted for age, sex, education, and gross household income. All socio-demographic covariates analysed are listed in Table 1. We reported estimated PRs with 95% confidence intervals (CIs) at a significance level of p < 0.05. All analyses were conducted using Stata 14.0 SE.