Tuberculosis among Foreign-born Persons, Singapore, 2000–2009

We determined the proportion of foreign-born persons with tuberculosis (TB) in Singapore. This proportion increased from 25.5% in 2004 to 37.6% in 2009. Unskilled workers from countries with high incidences of TB accounted for the highest number of and greatest increase in foreign-born TB case-patients.

S ingapore, an island city-state (area 710 km 2 ) in Southeast Asia, liberalized its immigration policy and underwent rapid economic growth during 2005-2010. This policy resulted in a marked increase in its population from 4.17 million in 2004 to 5.08 million in 2010, which was largely caused by an increase in foreign-born persons comprising long-term pass holders (LTPHs) (permission to stay in Singapore >6 months), permanent residents (PRs), and naturalized citizens (1). In recent decades, mass immigration and infl ux of nonimmigrants from countries with high incidences of tuberculosis (TB) to industrialized countries have contributed to the epidemiology and incidence of TB in host countries (2)(3)(4)(5)(6)(7)(8)(9). We report the epidemiology of TB in foreign-born case-patients in Singapore during 2000-2009.

The Study
Notifi cation of TB cases to the Singapore TB Elimination Programme (STEP) Registry is mandated by law. Since 1998, the notifi cation report has included disease characteristics, sociodemographic information, country of origin, immigration status, and year of arrival in Singapore of TB patients (10). All notifi cations during 2000-2009 were obtained from the STEP Registry database. When data on country of birth were not available for persons not born in Singapore (0.5% of case-patients), the nationality of these persons was assumed to represent their country of birth. Missing information about country of birth for persons with a Singaporean nationality (17%) and arrival date (6.5%) were matched against a database provided by the Ministry of Home Affairs.
Persons not born in Singapore were considered foreign born. Foreign-born case-patients comprised naturalized citizens, PRs, and LTPHs. Singapore issues long-term passes to skilled and unskilled workers and others (students and foreign-born family members of citizens or PRs) to enable a stay >6 months in Singapore. Persons applying for permanent residency or long-term stay and renewal of longterm passes are given medical examinations that include a general physical examination, chest radiograph, and testing for HIV (11). The Department of Singapore Statistics  Figure 2, panel B). Among native-born persons, TB occurred predominantly in men >40 years of age (Table). TB in foreign-born persons was distributed equally in both sexes and occurred mostly in persons 20-39 years of age (Table). Drug resistance and extrapulmonary involvement were higher among foreign-born persons than among native-born persons.

Conclusions
The decreasing trend of foreign-born persons with TB in the fi rst half of the study period could be attributed to fewer transient migrant workers entering Singapore during the economic crisis in Asia and the severe acute respiratory syndrome outbreak in 2003. After economic recovery and liberalization of the immigration policy in Singapore in 2005, there was an infl ux of migrant workers and immigrants from countries with high incidences of TB and a corresponding increase in TB notifi cations among this population. Of these persons with TB, >75% came from 5 of the 7 countries (India, China, Indonesia, Bangladesh, and the Philippines) with highest incidences of TB (12,13).
The large increase in the number of TB case-patients from Indonesia, the Philippines, and Myanmar refl ects the infl ux of migrant workers from these countries. This fi nding is consistent with the predominance of cases reported in foreign-born persons 20-39 years of age. The lower incidence of diabetes among these younger persons than in native-born persons is not surprising. The preponderance of lymph node involvement among the foreign-born persons merits further study.
The estimated TB rate in persons born in Singapore decreased from 49.7 cases/100,000 population to 41.8/ 100,000 during 2000-2007 but increased to 46.5/100,000 in 2008 and 45.2/100,000 in 2009. This increase followed the increase in cases among foreign-born persons in 2005. Transmission from foreign-born to native-born persons may have occurred, especially in increasingly crowded urban settings of Singapore. However, this proposal cannot be verifi ed without DNA fi ngerprinting studies.
Currently, persons with untreated inactive lesions or scarring on their screening chest radiographs are given permission for a long-term stay in Singapore without further examination or follow-up. To improve screening, mandating sputum TB cultures may be prudent for these persons. If active disease is excluded, these persons can then receive prophylactic treatment.
We do not favor a policy of screening for latent TB infection because of the high probability of falsepositive tuberculin skin test results in persons vaccinated with Mycobacterium bovis BCG and the unfavorable risk-benefi t ratio of preventive therapy for persons from countries with high TB incidences, who may be more likely to have acquired latent TB in the past than recently (14). Although interferon-γ release assays will overcome the problem of false-positive tuberculin skin test results in persons vaccinated with M. bovis BCG, the cost of these assays is prohibitively high, and their positive predictive value for progression to active TB remains to be determined (15).
A strength of this study was its nationally representative data of all TB cases reported to the STEP Registry. Electronic linkage of the Registry to the 2 mycobacterial culture laboratories in Singapore also enabled inclusion of all bacteriologically positive cases. A limitation of this study was the unavailability of data for native-born and foreign-born persons by sex and age to compare ageand sex-specifi c TB case rates between these 2 groups. In addition, because of the lack of population data for country of birth, we were unable to determine the TB rate by country of origin. Also, information bias is inherent in retrospective studies, and most country of birth data were based on information submitted by the notifying physician. We were also unable to analyze the association of HIV and TB because access to HIV status in our study population   was limited. Our study highlights the need for measures to address TB among foreign-born persons in Singapore.