Cost-effectiveness of targeted screening for active pulmonary tuberculosis among asylum-seekers: A modelling study...

Screening asylum-seekers for active pulmonary tuberculosis is common practice among many European countries with low incidence of tuberculosis. The reported yields vary substantially, partly due to the heterogeneous and dynamic nature of asylum-seeking populations. Rather than screening all new arrivals (indiscriminate screening), a few countries apply targeted screening based on incidence of tuberculosis in asylum-seekers' country of origin. However, evaluations of its cost-effectiveness have been scarce. The aim of this modelling study was to assess whether the introduction of a screening threshold based on the tuberculosis incidence in the country of origin is sensible from an economic perspective. To this end, we compare the current, indiscriminate screening policy for pulmonary tuberculosis in Germany with a hypothetical targeted screening programme using several potential screening thresholds based on WHO-reported incidence of tuberculosis in countries of origin. Screening data is taken from a large German federal state over 14 years (2002-2015). Incremental cost-effectiveness is measured as cost per case found and cost per case prevented. Our analysis shows that incremental cost-effectiveness ratios (ICERs) of screening asylum-seekers from countries with an incidence of 50 to 250/100,000 range between 15,000€ and 17,000€ per additional case found when compared to lower thresholds. The ICER for screening asylum-seekers from countries with an incidence <50/100,000 is 112,000€ per additional case found. Costs per case prevented show a similar increase in costs. The high cost per case found and per case prevented at the <50/100,000 threshold scenario suggests this threshold to be a sensible cut-off for targeted screening. Acknowledging that no screening measure can find all cases of tuberculosis, and that reactivation of latent infections makes up a large proportion of foreign-born cases, targeting asylum-seekers from countries with an incidence above 50/100,000 is likely to be a more reasonable screening measure for the prevention and control of tuberculosis than indiscriminate screening measures.