Reconstructing the population dynamics of foreign residents in Japan to estimate the prevalence of infection with Mycobacterium tuberculosis

•Notification rate of tuberculosis among foreign residents has steadily increased in Japan.•To calculate the risk of tuberculosis by country of birth, the population dynamics was reconstructed using the McKendrick equation.•The recent increase of residents from Myanmar and Vietnam was identified.•The estimated prevalence of infection ranged from 3.5% to 21.3%, all greater than the crude estimate.•Our methodology proposes a seminal approach towards establishing prevalence of infection by age, sex and residential status. Among newly notified tuberculosis cases in Japan, both the number and the proportion of foreign-born cases have steadily increased over time. As Japan prepares to introduce pre-entry tuberculosis screening for foreign-born persons entering Japan, various epidemiological evidence is needed to evaluate its effectiveness, including the prevalence of tuberculosis among current foreign residents in Japan, by country of birth. Yet as of today, even the underlying population dynamics has yet to be quantified. The present study therefore aimed to firstly reconstruct the demographic prevalence of foreign residents by the length of stay in Japan and by country of birth, and secondly, to estimate the prevalence of infection from notification data among foreign residents in Japan. We employed the McKendrick partial differential equation model to reconstruct the dynamics among six Asian countries which account for 80% of foreign-born tuberculosis patients notified in Japan i.e. China, the Philippines, Vietnam, Nepal, Indonesia, and Myanmar. Compared with China and the Philippines, the recent remarkable increase in the number of residents who had arrived within 5 years from Myanmar and Vietnam was identified. Assuming that the risk of primary tuberculosis given infection is 5%, the estimated prevalence of infection ranged from 3.5% to 21.3%, and all the estimates were more than three times greater than the crude estimate that ignored the time since immigration. The proposed method may be used to further estimate the prevalence by age, sex and residential status, which could potentially provide critical evidence towards establishing policies to control tuberculosis among foreign-born persons in Japan, and also possibly among migrants globally.