Introduction Although 80% of all TB cases in the UK occur in foreign born persons, TB risk in the immigrant population is largely unknown due to uncertain estimates of migration. The evaluation of screening models to prevent immigrant TB depends on informed estimations of this risk.
Objective To evaluate TB risk in a cohort of immigrants with new immigrant GP registration status (Flag-4) in Leicestershire; and to estimate efficacy of a screening model that uses Flag-4 registration and testing with interferon gamma release assays (IGRAs) for identifying latent infection with M tuberculosis (LTBI).
Methods All Flag-4 registered immigrants between January 2000 and December 2010 were included and collated with TB notification data for the same period. TB cases arising in registered immigrants were included for estimation of case rate using Kaplan–Meier curves. Cumulative TB rates were expressed as time after UK entry and time after GP registration and compared between immigrant subgroups stratified by WHO incidence in country of origin (150–499/100 000 or 500+/100 000) and age group at time of registration (<16, 16–35 or =36 years). The number needed to screen was calculated using an overall prevalence estimate of 25% IGRA positivity, with all cases occurring in this subgroup.
Results 564 cases were recorded in 34 764 immigrants. The median (IQR) observation was 2198 (982–3329) days after UK entry and 956 (358–1888) days after GP registration. There was no difference in risk with time after UK entry or GP registration and the TB rate rose linearly over 10 years. In our cohort, the 5-year cumulative TB rate was significantly higher for immigrants from regions with incidence of 150–499 than those from 500+. The TB rate was also significantly higher in adults than children, and highest in adults aged 16–35 years (Abstract P55 table 1). For this age band, the estimated number needed to screen (95% CI) with IGRAs to identify one immigrant developing TB in 5 years was 78.8 (73.2–85.2) persons.
Conclusions A new immigrant screening model using the Flag 4 GP registry and IGRA testing may be effective for identifying at-risk immigrants.