Introduction The impact of screening for latent tuberculosis infection (LTBI) in immigrants for tuberculosis (TB) prevention in the UK is dependent on effective strategies for identifying at-risk groups. Here we investigate effectiveness of the primary-care (PC) registry to identify immigrants for enrolment to LTBI screening at the time of GP registration, based on their HIV status.

Methods We performed a 11-year retrospective cohort study of PC registrations, cross-referenced with foreign-born TB (FB-TB) notifications for immigrants entering the UK after 1999 that had HIV testing performed. We assumed LTBI screening to prevent prospective TB would be effective if notifications occurred ≥6 months after PC registration. The primary outcome was the proportion of FB-TB preventable with screening at PC registration and compared between subgroups stratified according to HIV status.

Results 624 of 857 FB-TB cases were HIV tested (72.8%) and 84 were HIV seropositive (13.4%). 56 (67%) HIV positive FB-TB cases occurred in immigrants from TB incidence regions 500/100,000. Overall 511 (63%) FB-TB cases were preventable if screened at PC registration. The proportion with unpreventable TB was significantly higher for the HIV positive compared to the HIV negative FB-TB subgroup; [19% vs 10%; RR (95% CI) = 1.89 (1.25 to2.84), p=0.003]. Compared with patients that were HIV negative, time to PC registration after UK entry was longer [median (IQR); 1515 (555–2202) days vs 415 (36–1558) days; p<0.005) and time to disease notification shorter [median (IQR); 587 (208–1182) days vs 1163 (669–1854) days, p<0.005] for HIV positive FB-TB patients.

Conclusion Targeted LTBI screening at the time of primary-care registration may be a less effective preventative strategy for HIV positive immigrants at highest TB risk. Our data supports the need to promote early registration and extended screening to include HIV and other blood-borne viruses as part of an integrated immigrant screening programme.