The following sections summarize Australia’s premigration health screening program and the dataset. Further detail is available elsewhere (22).

Given the primary aim of the new screening program, we found a low yield in detecting TB. TB is rare in children migrating to Australia; 15.6 cases per 100,000 child applicants were diagnosed over the study period, and annual Australia TB notifications in overseas-born children ranged from 4.5 to 8.8 cases per 100,000 children during 2015–2018 (7). Further, the effect of including IGRA and TST in screening on case detection is uncertain. Of the 17 TB-positive children in whom IGRA or TST was performed, 7 had a negative test for infection (all IGRA), and in the other 10 children, it cannot be assumed that the positive IGRA or TST result led to the diagnosis of TB. Although a test for infection is often included in the approach to TB diagnosis in children, the results do not confirm or exclude active disease and so cannot be interpreted in isolation (25,26).

IGRA and TST detect LTBI in children, who might benefit from TB preventive treatment (4). We found that 3.3% of children screened had LTBI, equating to 30 applicants screened per LTBI case detected. Factors associated with LTBI matched well-established risk factors, namely increasing age, history of TB contact, and coming from a region with high TB prevalence (acknowledging variability within regions). These findings support previous evidence showing greater yield from migrant TB screening targeting higher-risk groups (14). Increased positivity with TST versus IGRA also reflects known test characteristics; IGRA has greater specificity in BCG-vaccinated children (26).

Although the LTBI prevalence of 3.3% found in this population was low, that prevalence was higher than the recently modeled estimate of 2.1% for overseas-born children (0–14 years of age) in Australia (12), likely reflecting targeted screening criteria. LTBI prevalence might have been higher if children 11–14 years of age had also been screened, given LTBI prevalence increases with age in TB-endemic settings. LTBI prevalence was much lower than in some other international migrant screening programs. An analysis of premigration LTBI screening in 67,334 children 2–14 years of age bound for the United States (from countries with TB incidence rate >20 cases per 100,000 children) found a higher LTBI prevalence of 12% (27).

Characteristics of those screened largely reflected screening criteria. Although overall screening completeness could not be determined, screening was nearly universal among humanitarian applicants 5–9 years of age, who all met criteria. Although complete assessment of the risk factors prompting testing in temporary applicants was not possible, a surprisingly high proportion (9.8%) of temporary applicants underwent IGRA or TST; the proportion testing positive (3.6%) was similar to the proportion in permanent applicants (3.8%). In contrast, humanitarian applicants had significantly lower odds of LTBI than permanent applicants despite the presumption of high LTBI risk and mandated testing. This difference is likely because permanent applicants were only screened if they were from a high-burden setting, and the humanitarian intake at the time was predominantly from countries with relatively low TB prevalence (largely refugee applicants from Syria and Iraq) (28). Nevertheless, the proportion of positivity (2.1%) in humanitarian applicants was lower than previous LTBI prevalence estimates in refugee-background children tested in Australia, including from the Middle East (11,29).

Almost all children with positive screening results had further investigation for TB, including chest radiography, as per guidelines. Most were allocated a TBHU requiring linkage to follow-up care in Australia. The updated screening was associated with a marked increase in children with TBHUs. These additional TBHUs might have enabled TB preventive treatment that might not otherwise have occurred, for potential substantial benefit for these children. However, TBHUs require use of services and resources both for families and the health system in Australia. Time from visa application to follow up after migration is often months to years. This delay from a positive test to commencement of TB preventive treatment (in addition to the unknown time between infection and positive test) risks missing the period when disease is more likely to develop in children with LTBI (5). Which might reduce both the individual and public health benefit of preventive treatment.

After migration, Australia guidelines recommend testing migrants <35 years of age from high-burden TB settings for LTBI (30). In practice, this testing occurs infrequently unless migrants are assigned a TBHU, linked with specific services (e.g., refugee health clinics), or there is another specific requirement (e.g., occupational screening) (31,32). Systematic premigration LTBI testing linked to appropriate follow-up care theoretically offers several advantages, including broader testing within mandated IMEs, targeting migrants from high-risk settings, supporting source country TB infrastructure, and detecting infection promptly, given LTBI reactivation risk is highest early after migration (21,33–35). Some migrant groups have also reported a preference for premigration LTBI testing over postmigration testing (36). Researchers analyzing premigration TB screening (with chest radiography) suggest considering expansion to include LTBI, citing reactivation of overseas-acquired LTBI as the primary driver of TB in low-burden countries (13,33,37,38). Some reviews and modeling estimates have supported the potential effectiveness and cost-effectiveness of premigration LTBI screening, particularly when targeted to migrants from high-burden settings (16,33). Others have found limited cost-effectiveness, particularly if screening costs are included in cost-effectiveness calculations (rather than being borne by migrants), and have highlighted the limited likely contribution of migrant screening to TB elimination in low-burden settings (39). However, observational data from premigration health screening programs detecting LTBI, particularly in children, remain limited, underscoring the importance of analyses such as this one (33,40).

Ideally, assessment of the overall effectiveness of the new screening program would consider the full cascade of care, from premigration LTBI detection to postmigration TB preventive treatment uptake and completion, as well as TB incidence (33). Losses over the cascade of care are common: of 8,231 children in whom LTBI was detected before migration to the United States, 70% were followed after migration but rates of diagnosis revision and low treatment completion were substantial (27). However, given that TB will not develop in most children with LTBI even without TB preventive treatment, the low LTBI prevalence in our analysis implies a large number of children would need to be screened to prevent 1 case of TB.

The first limitation of our analysis is that the study population represents visa applications rather than persons, meaning repeat applicants are not accounted for (22). No data were available regarding BCG immunization. Of note, BCG vaccine is included in the immunization schedule of many applicant source countries (e.g., India, China, and the Philippines), and a large proportion of screened applicants likely received the BCG vaccine in infancy. However, in TB-endemic countries, a positive TST indicates infection with M. tuberculosis regardless of BCG status, and false positivity decreases with increasing age (41,42). Further, in children, a positive TST requires clinical follow up regardless of BCG status.

We were unable to assess screening completeness directly against criteria because of privacy limitations on data format. Our denominator for the proportion of children screened was inflated by including children <2 years and 11–14 years of age, whose age placed them outside of screening criteria. Because of these factors, the cohort represents a mix of children screened based on age, visa stream, and clinical risk criteria. Nevertheless, the demographics of the cohort broadly align with screening criteria, and sensitivity analyses restricted to children 5–9 years of age did not alter our conclusions, suggesting our group largely reflects children meeting screening criteria. Further, our results reflect real-world implementation of the updated screening program. Finally, no data were available regarding care linkage in Australia or clinical outcomes.

Our analysis provides the results of Australia’s systematic premigration TB screening in children, one of few such programs globally. Notwithstanding inherent delays between infection, screening, and TB preventive treatment, the individual benefit for children with LTBI detected by premigration health screening could be considerable, with potential future reductions in TB incidence in overseas-born children in Australia. However, the low proportion of LTBI found in our analysis, despite the program targeting higher-risk children, suggests a modest yield from the current program toward substantial reductions in TB after migration.

TB elimination in Australia is unlikely to be achieved through migrant screening alone, and strategies such as investment in TB programs in high-burden settings, alongside strong local public health measures, might have greater impact (10,39,40). Key questions remain unanswered that will influence whether, and how, migrant screening can best contribute to TB and LTBI detection. These questions include the effects of changing migration patterns on screening yield, trade-offs between screening older migrants (with higher infection prevalence) versus children (with greater individual benefit for those with LTBI), acceptability and cost for migrants, health resourcing implications, and ultimately, effectiveness for reducing TB burden, toward the overarching goal of TB elimination.

In conclusion, in low-burden settings, additional interventions to prevent TB are inherently likely to have lower yield and be less cost-effective because of already low TB incidence. Shifting screening costs onto migrants might improve apparent cost-effectiveness but has implications for equitable access to screening. Given TB epidemiology in Australia, targeted migrant screening might still make a valuable contribution toward TB elimination even if absolute yield is low.