Screening for tuberculosis in migrants and visitors from high-incidence settings: present and future perspectives

Literature search strategy

To inform the content of this review we conducted a literature search of five databases (Ovid MEDLINE Epub Ahead of Print, Ovid MEDLINE In-Process & Other Non-Indexed Citations, Ovid MEDLINE, Ovid Embase and Ovid Cochrane Database of Systematic Reviews) from database inception to December 2017. The following key words (or close variations) were used: “tuberculosis”, “latent tuberculosis” “screening”, “public health surveillance”, “follow-up”, “health undertaking”, “migrant”, “refugee”, “asylum seeker”, “overseas”, “foreign”, “arrival”, “entrant”, “visa applicant”, “transient”. Key words were mapped to MeSH (Medical Subject Headings), allowing the search for matching content. We also searched the reference lists of relevant papers. Review authors provided content expertise to identify important articles in the topic area.

Pre-migration screening for TB disease

Regular migrants from TB-endemic settings may not have an increased risk of TB disease compared with the general population in their country of origin [18], but their likelihood of developing TB often exceeds that of the population that they will be entering. For this reason, a number of low-incidence countries mandate pre-migration screening for all migrants from high-incidence settings [19].

Pre-migration TB screening, implemented routinely in the UK and in selected other European countries, the USA, Canada, Australia, and New Zealand, involves the systematic screening of prospective regular migrants and refugees in their country of origin. This is often undertaken as part of a more comprehensive health screening (conducted during the visa application process for regular migrants). The goal of pre-migration screening is to identify individuals with TB disease [20] and, notably in Canada and Australia, to identify migrants at high risk of subsequently developing TB after arrival [21, 22].

Figure 1 presents generic screening algorithms for TB disease and LTBI in migrants or visitors from settings with a high incidence of TB. Screening algorithms vary considerably between countries [21], but the major principles and algorithms are broadly similar.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1

Example of generic screening algorithms used in migrants or visitors from high-incidence settings. TB: tuberculosis; M. tuberculosis: Mycobacterium tuberculosis; TST: tuberculin skin test; IGRA: interferon-γ release assay; LTBI: latent TB infection. ^#: if no signs or symptoms suggestive of TB disease (and chest radiograph negative in case of recent TB contact); ^¶: history of previous TB and HIV status also assessed; ⁺: some countries perform IGRA in TST-positive children (to increase test specificity in bacille Calmette–Guérin-vaccinated individuals); ^§: e.g. Xpert MTB/RIF; ^ƒ: sometimes done irrespective of TST/IGRA result, e.g. young children in recent close contact with an infectious TB case (USA) to avoid a chest radiograph if TST/IGRA negative.

The first stage of screening for TB disease typically involves a combination of symptom-based, physical and chest radiography assessment. Confirmatory bacteriological testing may involve sputum testing using smear microscopy, PCR (e.g. Xpert MTB/RIF) and/or culture. Additional tests for extrapulmonary TB, such as fine needle aspiration biopsies of enlarged superficial lymph nodes, may be performed if indicated. If TB disease is identified, verification of a bacteriological response to therapy is usually required before migration is permitted (figure 1).

Canada [23], the USA [24], Australia [25] and the UK [26] require applicants above a certain age threshold (usually 10 years; 15 years in the USA) to have a plain chest radiograph if they originate from a high TB incidence country (generally defined as an estimated TB incidence ≥40 per 100 000 population). However, most other low TB incidence countries do not have this requirement [19]. The USA, Australia, and Norway [21, 24, 25] also require all child migrants aged 2–10 years (2–14 years in the USA) to undergo an interferon-γ release assay (IGRA) or a tuberculin skin test (TST) for LTBI. A chest radiograph is performed only if these initial tests for infection are positive. Such testing also provides an opportunity to consider LTBI treatment in children without evidence of TB disease.

Pre-migration screening policies are underpinned by evidence from observational studies that have demonstrated high case-finding yields among screened populations and significant reductions in the risk of TB following the implementation of pre-migration screening [27–29]. Since the yield of pre-migration screening correlates with the incidence of TB in the source country [18], pre-migration screening policies focus on regular migrants from high-incidence countries. The effectiveness of pre-migration screening to reduce TB importation is influenced by the algorithm employed and the quality of the screening process implementation (e.g. accuracy of chest radiograph reading). Drawing upon the available observational data [29], the World Health Organization (WHO) has issued a conditional recommendation that people migrating from high- to low-incidence countries should be screened for TB disease before arrival [30].

A diagnosis of TB may be associated with stigma and usually results in delayed visa processing with substantial delays required to complete TB treatment, especially treatment for drug-resistant TB. The detection of LTBI or minor radiographic abnormalities may also contribute to stigma and exacerbate underlying anxiety. These negative consequences of pre-migration screening, together with concerns that a TB diagnosis may adversely affect the application outcome, may fuel practices that undermine the validity of the screening process.

Screening for LTBI

The WHO recommends that systematic LTBI screening and treatment should be considered for high-risk populations in upper-middle- and high-income countries, using either an IGRA or a TST [31]. One of the reasons for considering LTBI screening among migrants from high to low TB incidence countries is that migrants are at their highest risk of developing TB disease within the first 5 years after migration. TB could be prevented if LTBI is detected and adequate preventive therapy provided [32]. Migrants may also have comorbidities such as diabetes mellitus, chronic kidney disease or other immune compromising conditions that could increase their risk of future TB disease progression [33].

National policies for LTBI screening among migrants have different thresholds to define a high TB incidence country, prioritise different groups of migrants (i.e. asylum seekers versus regular migrants), and recommend different diagnostic tests and preventive therapy regimens [34, 35]. UK guidelines state that healthcare professionals “should maintain a coordinated programme to detect” LTBI and offer preventive treatment to new entrants aged ≤35 years [36]. Guidelines from the USA Preventive Services Task Force recommend that asymptomatic adults with likely previous TB exposure be screened for LTBI, including “persons who were born in, or are former residents of, countries with higher TB prevalence” [37]. A review of policies from 29 OECD countries found that 25 countries (86.2%) routinely screened all migrants for TB disease and 16 (55.2%) offered LTBI screening, usually post-migration, of which 15 offered LTBI screening to all migrants, including asylum seekers and refugees (in eight countries this was compulsory) [19].

A study from the UK conducted in recent migrants from high TB incidence areas demonstrated that among migrants with LTBI, those who had LTBI treatment based on a positive IGRA had a significantly reduced risk of TB during the observation period compared with untreated LTBI-positive patients (incidence rate ratio 0.17, 95% CI 0.05–0.60) [38].

Considerations for LTBI testing include the accuracy of the test, the risk–benefit ratio of LTBI treatment and the impact of the test result upon clinical decision making [39, 40]. Concerns about poor effectiveness of therapy are often related to reports of low rates of LTBI treatment completion, ranging from 7% to 83% [41]. Uptake of LTBI testing and LTBI treatment completion rates can be improved with adequate patient support, and some studies have reported that patient-centred interventions (such as pharmacist-led LTBI clinics, screening offered in community colleges and free monthly drug dispensation directly through the TB clinic) improve uptake of LTBI testing (up to 75%) and LTBI treatment completion rates, among those who accepted treatment (up to 94%) [42–44].

The cost-effectiveness of LTBI screening among migrants is an important consideration for policy makers. Cost-effectiveness studies have reported mixed results and are highly context specific [45]. One modelling study found that cost-effectiveness was optimal when a single IGRA test was offered to all migrants aged ≤35 years from countries with an estimated TB incidence ≥150 per 100 000 population, with the estimated ability to identify 92% of infected migrants [46]. Another modelling study found that pre-migration screening of Chinese and Indian students going to the USA would be cost-effective (and prevent 157 cases of TB annually) from the destination country perspective if the students paid for the screening themselves [47]. In a scoping review on the cost-effectiveness of LTBI screening among migrants from high-incidence countries, of the nine studies that met the inclusion criteria, four studies stated that the IGRA was the most cost-effective test for LTBI screening generally and two others noted that LTBI screening was cost-effective only if carried out among migrants with recent close TB contact [48]. Most cost-effectiveness models do not consider full programmatic scale-up of LTBI screening and the use of short-course LTBI treatment regimens has not been considered. Therefore, more accurate estimates of the burden of migrant screening to destination countries and migrant populations are needed.

Screening and diagnostic tools for TB and LTBI

Chest radiography and microbiological tests for the detection of Mycobacterium tuberculosis are widely used as screening and diagnostic tools to assess migrants for TB disease, whereas the TST and IGRA are used for LTBI screening or as the initial step in TB disease assessment (e.g. in children).

Chest radiography is a key TB screening test. Due to its high sensitivity and moderate specificity for TB, the primary role of chest radiography in TB screening is to identify individuals who should undergo microbiological testing to rule out pulmonary TB [49]. The use of digital radiography has a number of advantages, including less radiation exposure, better image quality, reduced costs, less image variability, and the convenience and efficiency of electronic transmission of images, which facilitates off-site reading and quality control by specialist radiologists [49]. It is thus no surprise that digital chest radiography is preferred by many immigration authorities [49–51]. The accuracy of chest radiography is also critically dependent on the skills of the individual interpreting the image, which can contribute to poor inter-reader reliability. To address this, some countries have stringent requirements for the type of physicians allowed to read immigration screening radiographs, as well as for the method of reporting and categorising abnormalities [21, 51, 52]. Recent advances in machine learning and computing power have led to the development of computer-aided programs that can analyse digital chest radiographs for the presence of TB-compatible abnormalities [53, 54]. This could eliminate the problem of inter-reader variability. However, the accuracy and cost-effectiveness of this nascent technology remains unclear due to limited evidence [55, 56]. Computer-aided programs may be useful in quality control and in aiding radiologists with their reading.

Microbiological tests for detecting M. tuberculosis are typically only required if TB disease is suspected, either due to the presence of symptoms, clinical findings or radiographic abnormalities. Smear microscopy to identify acid-fast bacilli has been used to diagnose TB disease for well over a century, and remains a valuable test as it is rapid, cheap and provides information on the degree of infectiousness. However, it has poor sensitivity that limits its value as a stand-alone test to rule out active disease. In one study of 1179 Vietnamese individuals applying for an immigration visa to the USA, smear microscopy missed 76% of culture-confirmed cases [57]. The addition of routine sputum culture in visa applicants with presumptive TB has been associated with a substantial decline in post-migration case detection in the USA since 2007 [58, 59], although temporal factors may have contributed to this decline. Between 2007 and 2012, approximately 1.5 million migrants and refugees to the USA were screened using a culture-based algorithm. About the same number of migrants and refugees were screened by a smear-based algorithm [59]. Among the 4032 TB cases diagnosed by culture, 2195 (54.4%) were sputum smear negative. The annual number of reported TB cases among migrants screened with the culture-based algorithm, within 1 year after arrival, decreased from 1511 to 940 from 2007 to 2012 [59]. A UK study showed that migrants screened at sites where sputum culture testing was performed on all samples had increased odds of being diagnosed with bacteriologically confirmed TB at pre-migration screening (OR 2.4, 95% CI 1.9–3.0; p<0.0001) compared with those being screened at sites with smear-based testing alone [60].

While culture is widely accepted as the reference standard for diagnosing active TB [52, 61, 62], its limitations include high cost and delayed diagnosis due to slow bacterial growth [63]. PCR-based tests, such as the fully automated Xpert MTB/RIF assay, are increasingly used, since they offer higher sensitivity than smear microscopy and provide results much faster than culture. However, Xpert is substantially less sensitive than culture in patients with paucibacillary disease, missing almost one-third of sputum smear-negative TB cases [64]. Recent guidelines from the WHO [65] and other major professional organisations [61] recommend Xpert as the first microbiological test to perform in someone with presumptive TB; a growing number of countries have introduced this within their local screening algorithms.

TSTs and IGRAs detect T-cell responses to M. tuberculosis antigens. They are primarily used in the assessment of LTBI. IGRAs are more specific than TSTs in individuals with previous exposure to nontuberculous mycobacteria or previous vaccination with bacille Calmette–Guérin (BCG) [66]. Among commercially available IGRAs, the T-SPOT.TB appears to be the more sensitive test, but the QuantiFERON Gold assay is broadly equivalent, and is simpler to perform and standardise [37].

Post-migration follow-up

While pre-migration screening primarily focuses on the detection of TB disease, it also identifies migrants who have no microbiological evidence of TB disease at the time of migration, but who are at an increased risk of developing TB disease in the future. Indicators of future TB risk include chest radiograph abnormalities consistent with previous TB infection or disease, or recent contact with an infectious TB patient [67, 68]. The post-migration incidence of TB in high-risk migrants identified in this way is estimated to be more than 100 times higher (relative risk (RR) 102–416) than in the general population in the destination country [69]. This risk is substantially higher than that of TB in other high-risk groups, such as close contacts of patients with TB (RR 47) [70], patients using tumour necrosis factor-α inhibitors (RR 2–29) [71], patients undergoing solid organ transplantation (RR 27) [72], patients with chronic renal failure requiring renal replacement therapy (RR 8) [73], or children with haematological malignancies or solid cancers (RR 17) [74]. Offering these high-risk migrants LTBI treatment should therefore be a priority for those receiving countries with a low incidence of TB.

Some countries have established post-migration follow-up programmes for migrants at increased risk of developing TB. The follow-up programmes may consist of one medical follow-up visit in the destination country (the usual approach adopted in the USA [59]) or can encompass multiple follow-up visits spanning ≥2 years (the usual approach in Australia [75, 76]). Follow-up programmes that conduct serial chest radiographs aim to identify cases of TB disease early, reducing secondary TB transmission within the community and improving the outcome of affected individuals [76]. To date, the effectiveness (in particular the cost-effectiveness) of such programmes remains unclear, since many high-risk migrants do not attend routine follow-up visits [17]. It seems reasonable to consider LTBI treatment as an alternative or as an addition to chest radiograph follow-up, as the benefit of LTBI treatment for high-risk groups is well documented [77, 78].

A substantial proportion of TB cases diagnosed soon after migration are likely to have been present at the time of pre-migration screening or occur among short-term visitors (for whom screening is not performed routinely). As elaborated earlier, the number of missed cases can be reduced by adding a requirement for sputum culture for patients with radiographic abnormalities or symptoms during pre-migration screening.

The considerable challenges of providing TB care to migrants from different linguistic and cultural backgrounds need to be acknowledged and addressed. Healthcare systems in receiving countries need to provide access to culturally appropriate and high-quality healthcare for migrants in order to detect and treat TB in this group. Training of clinicians is required, along with removal of any restrictions in access to healthcare such as those that only allow treatment for emergency care. Discussions about the benefits of LTBI treatment in someone who is completely asymptomatic can be particularly challenging, especially when health beliefs are not aligned with evidence-based Western medicine [79]. Those involved in the care of migrants need to be aware of and address important cultural differences in health beliefs [80].

Asylum seekers and refugees

Asylum seekers and refugees, defined as persons who have fled from their country of origin for fear of persecution and have applied for protection [81], are a particularly high-risk group for developing TB [82–84]. Many refugees originate from countries with a high incidence of TB [85] and often continue to live in conditions of deprivation in the destination country. The same is true for asylum seekers, defined as persons who seek safety from persecution or serious harm in a country other than their own, and who await a decision on the application for refugee status under relevant international and national instruments [11]. For asylum seekers and refugees, limited access to healthcare, combined with economic vulnerability, may contribute to significant delays in TB diagnosis. For refugees and asylum seekers with TB, flight from persecution may interrupt their treatment, leading to poor outcomes and acquired drug resistance, as well as potential ongoing TB transmission [86, 87]. Time spent in crowded refugee camps increases the risk of TB exposure and infection. Refugees may also have other risk factors for TB disease, such as malnutrition, chronic kidney disease, diabetes mellitus and HIV co-infection.

The WHO has published detailed guidelines for establishing effective pre-migration TB care and prevention among asylum seekers and refugees [85]. These recommendations focus on strengthening the health system in countries of origin and implementing the core elements of an effective TB programme. In settings torn by war or civil unrest, strengthening poor local healthcare infrastructure may not be possible and screening is only feasible at or after entry to a low-incidence country.

If accepted by a receiving country, refugees usually require routine symptom-based and radiographic screening to exclude TB disease with attention to physical and psychological comorbidities. Such screening may be implemented prior to migration, at the point of entry or after arrival, depending on the receiving country [88]. Some receiving countries screen refugees for LTBI pre-migration, given their increased risk of infection and subsequent progression to disease [39]. LTBI treatment or serial screening for disease can then be initiated for those testing positive with a TST or IGRA. High treatment completion rates for LTBI therapy can be achieved with appropriate social and healthcare support [89].

The yield of TB cases from screening low-risk refugees within Europe has been found to be low [39]. A German study suggested that TB cases identified at entry screening were concentrated in asylum seekers from only a few countries and that screening was most efficient in migrants from those countries (e.g. Cameroon, Eritrea, Gambia, Georgia, Pakistan, Russia and Somalia) [90]. Hence, TB control policies must be tailored to the risk profile of refugee populations, and be implemented in the broader context of appropriate social, economic and healthcare support throughout the migration process. Europe has experienced an unprecedented influx of refugees and asylum seekers in recent years. It is essential that access to early TB diagnosis and care among refugees is ensured, and that the coordination of TB screening efforts in Europe is improved to benefit refugee populations, to protect the public at large and to achieve the targets of the End TB Strategy [91, 92].

Short-term visitors and those with long-duration multiple-entry visas

The risk of TB in visitors, i.e. those granted temporary admission to stay in a country for <3–6 months at a time, in particular among those who travel regularly from high- to low-incidence settings, is not well known. Unlike permanent or long-term migrants, short-term visitors are rarely screened [7, 21]. In an era where movement of people across borders is considered essential for business, trade and economic development, screening policies may be in tension with policies to decrease perceived barriers to travel. However, given the dramatic increase in people movement between high and low TB incidence settings, consideration should be given to addressing the challenge of TB in short-term visitors, if the WHO goal of TB elimination is to be realised [93].

In recent decades, the pattern of international migration has changed in complexity, with a greater emphasis on short-term visits, frequent returns and circular migration where cumulative risk factors and exposure should be considered [94]. In OECD countries, inbound tourism contributes more than 4% of gross domestic product, 5% of employment and 20% of service exports, and continues to increase year by year [95].

The limited information about the health status of visitors, and their need for and use of medical services, has hindered an objective evaluation of this area of public health policy [96]. In some countries, visitors are intentionally excluded from data collection. For example, in the USA, TB cases are not included in official case counts when a person is in the USA for <90 days [9]. Despite this, we know that TB is a concern in these groups and studies have shown up to 20% of notifications in the USA occur among visitors or temporary residents [97]. In Australia, the National Notifiable Disease Surveillance data from the period 2012–2016 identified that short-term overseas visitors (where this data was captured) made up 7% of all TB notifications [98].

The case study in box 1 shows data from a simple modelling exercise to assess the TB risk among visitors coming from China and India to Australia if long-term (10-year) multiple-entry (allowing up to 3 months stay per year) visas for these nationalities were to be introduced.

Another case study by Weinberg et al. [9] reported cases of TB among foreign-born, temporary workers in the USA, working predominantly in the tourism industry, and highlighted delayed presentation and diagnostic delays of up to 3 months due to a lack of awareness among clinicians. This was further complicated by problems in TB contact tracing due to frequent changes of location and use of bunk-style accommodation among temporary workers [9]. A study from the United Arab Emirates among migrants undergoing medical visa screening found that the TB prevalence among new applicants (first screening) was significantly higher (49.3 per 100 000) than among migrants who underwent repeat screening for visa renewal (25.2 per 100 000) [99]. Travel back to the country of origin was, however, not captured in the study and it is therefore not possible to determine whether the risk of TB decreased significantly despite return visits to the home country.

The economic and other benefits derived from short-term travel provide an imperative to strike a balance between promoting ease of travel and a safe environment that limits the spread of communicable diseases, in particular TB. Greater policy coherence, the development of long-term strategic approaches, and engagement with a range of public and private stakeholders (e.g. employers of short-term migrants) will be essential. Given the high global rates of TB disease and infection, the identification of all short-term visitors who pose a risk of TB importation and domestic transmission is not feasible [8], but better risk stratification may guide practical steps to minimise risk. As a minimum, policy makers must ensure that any legal, financial and social barriers to access healthcare are removed for anyone with a potentially serious infectious disease, enabling early diagnosis and treatment [100].

Paediatric perspectives

Historically, TB control programmes have focused almost exclusively on the identification and treatment of infectious adult cases, with little recognition that children suffer high disease burdens in TB-endemic areas with uncontrolled transmission of M. tuberculosis [101]. It is estimated that around 1 million children developed TB disease in 2016 [102] with most cases occurring in TB-endemic settings where TB is likely to be a major, albeit unrecognised, contributor to child mortality [103, 104]. The rising tide of drug-resistant TB also affects children, with high rates of multidrug-resistant TB reported in children from settings with evidence of multidrug-resistant TB transmission [105, 106].

Migrant children can be exposed to TB in their country of origin, as well as in neighbouring countries or refugee camps during periods of displacement. After arrival in a receiving country with a low TB incidence, children may be exposed to visitors from TB-endemic countries within migrant communities or when visiting their country of origin. The risk of TB disease progression is highly dependent on the child's age at the time of infection (children aged <2–5 years are particularly vulnerable), the time since infection (>90% of disease progression occurs within 12 months of primary infection) and the child's immune status (severely malnourished and HIV-infected children are most vulnerable) [107].

Given the vulnerability of young children to rapid disease progression, screening for TB infection and the provision of LTBI treatment is important. LTBI detection is complicated by the poor specificity of the TST in children routinely vaccinated with BCG at birth [108]. IGRAs are not affected by BCG vaccination. Although indeterminate results have been a concern and blood collection is challenging in young children [109], reliable IGRA results are achieved in the vast majority of refugee children [110]. There is evidence that IGRAs are accurate even in children aged <2 years and that indeterminate results, if adequate blood volumes have been secured, usually have other (non-age-related) explanations such as an acute infection [111]. Results from the UK suggest that children with a positive TST and negative IGRA are not at risk of developing TB [112].

Children with documented LTBI, especially those who are young and vulnerable, benefit from appropriate LTBI treatment. Unfortunately, with LTBI no test can determine whether the infecting organism might be drug resistant, emphasising the importance of taking a careful TB exposure history [101]. Despite the availability of effective LTBI treatment, repeat exposure and future re-infection may occur after successful treatment, particularly if a migrant returns to their country of origin [113]. Therefore, in order to protect the most vulnerable children, it is important to vaccinate young children (aged <5 years) of migrant families born in low-incidence settings with BCG, if a prolonged return visit to the country of origin or another TB-endemic setting is planned. Table 2 provides an overview of issues to consider regarding paediatric TB in migrant populations, with a brief summary of challenges and practice recommendations.

The diagnosis of TB disease is often delayed in migrant children if this is not considered a likely diagnosis by the local clinician. It is important to ensure clinicians who manage these children consider the diagnosis of TB when they present with symptoms of the disease, which are often nonspecific in nature.

Child-friendly TB drug formulations are often unavailable in low TB incidence countries, as pharmaceutical companies lack the financial incentives to support local drug registration. This limits the availability of low-volume drugs that are unlikely to generate a profit, although special access schemes may facilitate use of these “orphan drugs”. Given the availability of quality-assured child-friendly, water-dispersible fixed-drug combination tablets via the Global Drug Facility [114], it is hoped that children in low TB incidence settings will also have access to these drugs.

Conclusions and outlook

A substantial proportion of TB cases in many countries with a low incidence of TB occur in migrants originating from high-incidence settings. Therefore, strategies to prevent and manage TB in international migrants are important to meet TB elimination targets in low-incidence countries. Existing national TB screening programmes that aim to reduce the risk of TB after arrival, while minimising the burden of screening to migrants, vary considerably. Table 3 outlines current evidence gaps and/or future research needs to inform optimal screening practices for TB in migrant populations. It is hoped that improvements and refinements in approaches will lead to expanded TB prevention efforts among migrants at highest risk of developing disease.

Supplementary material