Abstract
Objectives. We described risk factors associated with infectious tuberculosis (TB) and missed TB-prevention opportunities in foreign-born US residents, who account for almost two thirds of the nation’s TB patients.
Methods. In a cross-sectional study at 20 US sites of foreign-born persons diagnosed with TB in 2005 through 2006, we collected results of sputum smear microscopy for acid-fast bacilli (a marker for infectiousness) and data on visa status, sociodemographics, TB-related care seeking, and latent TB infection (LTBI) diagnosis opportunities.
Results. Among 980 persons with pulmonary TB who reported their visa status, 601 (61%) were legal permanent residents, 131 (13.4%) had temporary visas, and 248 (25.3%) were undocumented. Undocumented persons were more likely than permanent residents to have acid-fast bacilli–positive smears at diagnosis (risk ratio = 1.3; 95% confidence interval = 1.2, 1.4). Of those diagnosed 1 year or more after arrival, 57.3% reported LTBI screening opportunities; fewer than 25% actually were. Undocumented persons reported fewer LTBI screening opportunities and were less likely to be tested.
Conclusions. Progress toward TB elimination in the United States depends upon expanding opportunities for regular medical care and promotion of LTBI screening and treatment among foreign-born persons.
Tuberculosis (TB) in low-burden countries such as the United States reflects the global nature of the TB epidemic: the majority of cases occur among persons who have migrated from countries with higher TB rates. In 2012, 63% of all TB cases diagnosed in the United States occurred among people born in other countries—more than half of them from Mexico, the Philippines, India, Vietnam, and China.1
Tuberculosis is diagnosed by sputum-smear microscopy for acid-fast bacilli (AFB) and sputum culture to isolate Mycobacterium tuberculosis. The detection of AFB by sputum smear microscopy (smear-positive) is a measure both of severity of disease and infectiousness; the risk of transmission from smear-positive TB patients can be 5-fold greater than from patients whose smear microscopy results are negative for AFB (smear-negative).2–5 Persons who are smear-negative at diagnosis have less-advanced disease and are less likely to transmit infection.2 From 2000 to 2012, 46% of foreign-born persons diagnosed with TB in the United States were smear-positive at diagnosis.6
Of the 47 million persons who enter the United States every year, only the 500 000 who come as legal permanent residents (refugees and immigrants) are screened for TB before arrival based on protocols from the Centers for Disease Control and Prevention’s (CDC’s) Division of Global Migration and Quarantine.7,8 Those with chest radiographs suggestive of TB but who are smear-negative—and, since 2007, culture-negative for M. tuberculosis—are evaluated at local health departments after arrival in the United States, preferably within 90 days.9 Those not screened before arrival include the 46.5 million tourists, students, and workers with temporary visas and the approximately 500 000 undocumented persons who enter illegally; previous studies estimated that undocumented persons and those with temporary visas account for more than half of TB cases among foreign-born persons.10–12
The purpose of this study was to assess risk factors for smear-positive TB in foreign-born persons after arrival in the United States. We hypothesized that persons with fewer opportunities for TB screening would be more likely to have smear-positive sputum at TB diagnosis. The study also evaluated missed opportunities to diagnose and treat latent TB infection (LTBI) among foreign-born persons subsequently diagnosed with TB. An estimated 84% of TB cases in foreign-born residents of the United States are attributable to progression of LTBI acquired in the patient’s home country and, therefore, these cases are theoretically preventable with LTBI treatment.13 The findings should suggest approaches to reducing TB incidence among foreign-born persons in the United States.
METHODS
The Division of TB Elimination at the CDC funded the Tuberculosis Epidemiologic Studies Consortium to conduct epidemiological, economic, operational, and behavioral studies related to TB in North America; each of the 20 US sites consisted of an academic or nonprofit organization linked to a local or state health department.14 This study was a cross-sectional survey of foreign-born persons aged 15 years and older diagnosed with TB at all US sites in 2005 through 2006. Details of the study have been published.15 In brief, after obtaining consent, trained interviewers used a structured questionnaire to ask participants about demographics, date of first US arrival, visa status at US entry and at study interview (subsequently classified as permanent or US citizen, temporary, and undocumented), fears of deportation, socioeconomic status, and health insurance status at time of TB symptom onset or (for those who denied symptoms) at diagnosis, care seeking for TB, and LTBI screening and treatment opportunities in the United States. We assumed visa status at study interview to be the same as the status at diagnosis.
Clinical information, abstracted from state and city tuberculosis registries, included AFB smear status at diagnosis (recorded as positive or negative, with unknown results treated as negative), results of culture for M. tuberculosis, and anatomic site of disease (subsequently classified as pulmonary, extrapulmonary, or both). Results were weighted by gender, age, and country of origin to account for refusals. This analysis includes only participants with pulmonary TB, with or without extrapulmonary involvement. We estimated predictors of AFB-positive smear status (smear-positive or smear-negative) at diagnosis with weighted relative risks, crude and adjusted for visa status, by using log–binomial regression. We stratified models on time from first arrival in the United States to TB diagnosis: recent arrivals (defined as persons diagnosed within 6 months after arrival) and others (persons diagnosed greater than 6 months after arrival). We selected a 6-month cutoff to identify TB cases that could be affected by preimmigration screening protocols as in Lowenthal et al.16 Visa status at arrival was used for patients diagnosed within 6 months, and visa at diagnosis for those diagnosed post 6 months.
We entered predictors of AFB smear status identified in bivariate models into multivariable models, and developed final models by using backward elimination. We exponentiated model coefficients to estimate risk ratios (RRs). We estimated weighted percentages, risk ratios, and confidence intervals (CIs) with SAS version 9.3 (SAS Institute Inc, Cary, NC).
Sensitivity analysis assessed the effect on the findings of excluding persons with unknown visa status. We estimated 3 dichotomous logistic regression models, using in turn only 2 of the 3 visa classes. We selected an imputed visa class as the most likely prediction among results from each of the 3 models. Model variables included age at study interview, age at entry into the United States, region of origin, region of US residence, a combined measure of socioeconomic status that included income and education, gender, time since arrival, and whether the person had ever lived in a refugee camp. We reanalyzed selected study findings by using participants with imputed visa status.
We based identification of missed opportunities for diagnosis and treatment of LTBI on responses of participants who had lived in the United States for at least a year before their TB diagnosis.
RESULTS
Of 1496 persons enrolled in the study, 1166 (79.8%) had pulmonary involvement and more than half of those (54.3%) were smear-positive at diagnosis (Table 1). For the 1141 persons who had a known arrival date, the median time from arrival in the United States to TB diagnosis was 69 months (interquartile range = 22, 187).
TABLE 1—
Anatomical Site of Disease, Smear, and Culture Status for 1166 Foreign-Born Persons With Pulmonary Tuberculosis Who Were Residing in the United States at Diagnosis: 20 US Sites, 2005–2006
| Site of Disease | No. (%) | Sputum Smear Positive, No. (%) | Sputum Culture Positive, No. (%) | Sputum not Cultured or Results Unknown, No. (%) |
| Pulmonary onlya | 1020 (87.5) | 583 (56.3) | 855 (83.9) | 31 (2.7) |
| Pulmonary and extrapulmonaryb | 146 (12.5) | 54 (40.7) | 105 (72.7) | 16 (9.2) |
| Total with pulmonary involvement | 1166 (100.0) | 637 (54.3) | 960 (82.3) | 47 (7.0) |
Note. All participants with positive sputum smear were culture-positive.
One participant with positive culture without smear results counted as smear-negative.
One participant without smear results counted as smear-negative.
Diagnosed Within 6 Months of Arrival in the United States
Of the 166 persons (14.2%) with TB diagnosed within 6 months of arrival, 151 (91%) reported their visa status; the majority of those (58.6%) were permanent residents who would have undergone TB screening before entry (Table 2). The remaining 67 persons included 33 undocumented and 34 with temporary visas.
TABLE 2—
Association of Patient Characteristics With Sputum Smears Positive for Acid-Fast Bacilli Among 980 Tuberculosis Patients Who Reported Their Visa Status, Stratified by Time From US Arrival to Diagnosis: 20 US Sites, 2005–2006
| Characteristic | No. (% of Total)a or Median (IQR) | Smear Positive, No. (%) or Median (IQR) | Weighteda RR (95% CI) | Weighted RR Adjusted for Visa Status (95% CI) |
| Persons with tuberculosis diagnosed within 6 months after arrival (n = 151) | ||||
| Visa statusb | ||||
| Undocumented | 33 (19.2) | 20 (63.5) | 2.8 (2.3, 3.5) | |
| Temporary | 34 (22.2) | 14 (36.4) | 1.6 (1.2, 2.2) | |
| Permanent | 84 (58.6) | 16 (22.9) | 1.0 (Ref) | |
| Gender | ||||
| Male | 84 (55.8) | 30 (35.6) | 1.2 (0.9, 1.4) | 0.9 (0.7, 1.1) |
| Female | 67 (44.2) | 20 (31.3) | 1.0 (Ref) | 1.0 (Ref) |
| Age, y | 32.9 (23.1–55.6) | 30.8 (22.7–47.7) | 1.0 (0.99–1.001) | 0.999 (0.996–1.00) |
| Region of birth | ||||
| East Asia/Pacific | 73 (47.8) | 10 (13.0) | 1.0 (Ref) | 1.0 (Ref) |
| Latin America | 14 (12.3) | 10 (70.0) | 5.4 (3.8, 7.6) | 4.3 (2.8, 6.8) |
| Mexico | 31 (13.9) | 18 (62.1) | 4.8 (3.3, 6.8) | 4.1 (2.5, 6.6) |
| South Asia | 7 (6.21) | 3 (37.5) | 2.9 (1.7, 4.9) | 3.3 (1.9, 5.9) |
| Sub-Saharan Africa | 23 (15.8) | 7 (30.1) | 2.3 (1.5, 3.6) | 2.5 (1.6, 3.9) |
| Otherc | 3 (4.0) | 2 (80.0) | 6.2 (4.2, 9.0) | 7.2 (4.7, 10.8) |
| Insurance | ||||
| Yes | 31 (19.7) | 6 (23.3) | 1.0 (Ref) | 1.0 (Ref) |
| No | 120 (80.3) | 44 (36.3) | 1.6 (1.1, 2.2) | 1.2 (0.8, 1.7) |
| Fear of deportationd | ||||
| Yes | 46 (30.6) | 16 (34.8) | 1.0 (0.8, 1.3) | 1.0 (0.8, 1.3) |
| No | 100 (69.3) | 34 (32.4) | 1.0 (Ref) | 1.0 (Ref) |
| Symptoms before diagnosis | ||||
| With symptoms | 83 (52.3) | 41 (49.6) | 3.0 (2.2, 4.1) | 2.6 (1.9, 3.6) |
| No symptoms | 68 (47.7) | 9 (16.4) | 1.0 (Ref) | 1.0 (Ref) |
| Total | 151 | 50 (33.7) | ||
| Persons with tuberculosis diagnosed > 6 months after arrival (n = 829) | ||||
| Visa statuse | ||||
| Undocumented | 215 (21.4) | 140 (68.6) | 1.3 (1.2, 1.4) | |
| Temporary | 97 (12.5) | 55 (54.4) | 1.0 (0.9, 1.1) | |
| Permanent | 517 (66.1) | 281 (52.9) | 1.0 (Ref) | |
| Gender | ||||
| Male | 514 (61.2) | 317 (59.8) | 1.1 (1.1, 1.3) | 1.2 (1.1, 1.3) |
| Female | 315 (38.8) | 159 (51.2) | 1.0 (Ref) | 1.0 (Ref) |
| Age, y | 40.9 (28.3–56.3) | 40.0 (27.4–54.8) | 1.004 (1.002–1.005) | 1.001 (0.99–1.004) |
| Region | ||||
| East Asia/Pacific | 252 (34.3) | 127 (51.3) | 1.0 (Ref) | 1.0 (Ref) |
| Latin America | 155 (22.5) | 97 (63.4) | 1.2 (1.1, 1.3) | 1.2 (1.0, 1.3) |
| Mexico | 249 (18.8) | 158 (63.5) | 1.2 (1.1, 1.3) | 1.1 (1.0, 1.2) |
| South Asia | 66 (9.2) | 36 (52.5) | 1.0 (0.9, 1.2) | 1.0 (0.9, 1.2) |
| Sub-Saharan Africa | 80 (10.5) | 43 (51.5) | 1.0 (0.9, 1.1) | 1.0 (0.9, 1.1) |
| Otherf | 27 (4.7) | 15 (51.6) | 1.0 (0.8, 1.2) | 1.0 (0.8, 1.2) |
| Time since arrival | ||||
| 6–11.9 mos | 47 (6.0) | 34 (71.1) | 1.3 (1.2, 1.4) | 1.3 (1.2, 1.4) |
| 1–2 y | 68 (8.0) | 39 (57.6) | 1.0 (0.9, 1.2) | 1.0 (0.9, 1.2) |
| 3–5 y | 183 (22.0) | 100 (57.0) | 1.0 (1.0, 1.1) | 1.0 (0.9, 1.1) |
| > 5 y | 531 (64.0) | 303 (55.8) | 1.0 (Ref) | 1.0 (Ref) |
| Insurance | ||||
| Yes | 368 (48.8) | 195 (51.2) | 1.0 (Ref) | 1.0 (Ref) |
| No | 461 (51.6) | 281 (61.4) | 1.2 (1.1, 1.3) | 1.1 (1.0, 1.2) |
| Fear of deportationd | ||||
| Yes | 127 (15.4) | 72 (59.0) | 1.1 (1.0, 1.2) | 1.0 (0.9, 1.1) |
| No | 667 (84.6) | 381 (55.6) | 1.0 (Ref) | 1.0 (Ref) |
| Symptoms before diagnosis | ||||
| < 50 d before | 214 (25.2) | 140 (67.5) | 1.7 (1.6, 2.0) | 1.7 (1.5, 1.9) |
| 50–175 d before | 234 (26.2) | 134 (54.8) | 1.5 (1.3, 1.7) | 1.4 (1.3, 1.6) |
| > 175 d before | 213 (24.4) | 143 (66.3) | 1.8 (1.6, 2.0) | 1.8 (1.6, 2.0) |
| No symptoms | 168 (24.1) | 59 (36.8) | 1.0 (Ref) | 1.0 (Ref) |
| Total | 829 | 476 (56.5) | ||
Note. CI = confidence interval; IQR = interquartile range; RR = relative risk.
Proportions and relative risks were weighted for refusals, and differ slightly from raw data.
Did not include persons diagnosed within 6 mo of arrival who did not report visa status; 8 of those (58.7%) were smear-positive.
Other countries: Azerbaijan (1), Georgia (1), and United Arab Emirates (1).
Of 5 persons diagnosed within 6 mo of arrival, 3 had no opinion and 2 said that the question was not applicable (all 5 were permanent or temporary residents at entry and at diagnosis). Of 35 persons diagnosed > 6 mo after arrival, 16 had no opinion and 19 said that the question was not applicable (28 of 35 were either permanent or temporary residents at both entry and diagnosis).
Did not include 146 individuals diagnosed more than 6 mo after arrival who did not report visa status; of those, 93 (62.1%) were smear-positive.
Other countries: Austria (1), Azerbaijan (1), Bosnia and Herzegovina (3), Bulgaria (1), Denmark (1), Georgia (1), Hungary (1), Ireland (1), Kazakhstan (2), Montenegro (1), Poland (3), Romania (2), Russian Federation (2), Serbia (1), Spain (1), Turkey (1), United Kingdom (2), and Uzbekistan (2).
Approximately one third (50; 33.7%) of persons diagnosed within 6 months of arrival had sputum smears positive for AFB at diagnosis; this proportion varied significantly by the patient’s visa status. Compared with permanent residents, undocumented persons were nearly 3 times as likely and temporary visa holders 1.6 times as likely to have positive sputa at diagnosis (Table 2).
More than one quarter (28.7%) of permanent residents diagnosed within 6 months of arrival reported that their symptoms began before their arrival in the United States, one indication that they could have had TB disease at entry. Indeed, 45.9% of the 50 persons with positive smears in this period indicated that their symptoms began before arrival. A similar proportion of persons who were undocumented or had temporary visas at entry reported symptoms before arrival.
The mode of diagnosis also varied significantly by visa status. Almost three quarters (72.1%) of permanent residents were diagnosed as a result of postarrival health department screenings, whereas care seeking for symptoms was the primary reason for the TB diagnosis among both temporary residents (51.5%) and undocumented persons (93.8%; P < .001 for the comparison of temporary residents and undocumented persons with permanent residents; data not shown).
Diagnosed More Than 6 Months After Arrival in the United States
Of the 975 persons with TB who were diagnosed later than 6 months after arrival in the United States, 829 (85%) reported their visa status; almost two thirds (66.1%) of these were permanent residents, about one fifth (21.4%) were undocumented, and the remainder (12.5%) had temporary visas (Table 2). Approximately 80% of the 975 were diagnosed as a result of care seeking for symptoms, independent of visa status. Other modes of diagnosis were rare; however, permanent residents were more frequently diagnosed as part of non–TB-related medical care than persons with other visa statuses (10.1% vs 7.1% for all others; P = .02), and the undocumented were more frequently diagnosed within the framework of a contact investigation (5.4% vs 2.8% for all others; P < .001).
Almost two thirds (57.4%) of the 829 persons who reported their visa status were smear-positive at diagnosis whereas only one third (33.7%) of the 50 diagnosed within 6 months after arrival had positive smears at diagnosis (RR = 1.7; 95% CI = 1.4, 2.0).
As with recent arrivals, the proportions of patients who were smear-positive at diagnosis varied significantly by visa status. Compared with permanent residents, undocumented persons were 1.3 times as likely to be smear-positive at diagnosis (Table 2). Undocumented persons were also more likely than temporary residents to be smear-positive (RR = 1.2; 95% CI = 1.1, 1.4; data not shown). Other factors associated with being smear-positive at diagnosis were male gender, birth in Mexico or Latin America, a shorter time from arrival to diagnosis (< 1 year), and lacking health insurance. Although smear status did not differ by time from symptom onset to diagnosis, those who had symptoms were more likely to be smear-positive than those without symptoms. Notably, fear of deportation was not associated with smear status, either overall or even among the undocumented (not shown).
Table 2 shows that after we controlled for visa status, male participants were 20% more likely than female participants to be smear-positive, and those diagnosed in the period beginning 6 months to 1 year after arrival were 30% more likely to be smear-positive than those diagnosed later. After backward elimination to develop a multivariable model, only male gender and undocumented visa status were associated with being smear-positive at diagnosis. In that model, male participants were 20% more likely to be smear-positive than female participants (RR = 1.2; 95% CI = 1.1, 1.3) and the undocumented were 30% more likely than permanent or temporary residents to be smear-positive (RR = 1.3; 95% CI = 1.2, 1.4 for both). An interaction term between these 2 variables was also significantly associated with smear status (P = .02). Stratifying on gender, undocumented male participants were approximately 30% more likely than male permanent residents, and 50% more likely than temporary visa holders, to be smear-positive (RR = 1.3; 95% CI = 1.2, 1.4, and RR = 1.5; 95% CI = 1.3, 1.7, respectively). Among female participants, temporary visa holders were 20% more likely than permanent residents to be smear-positive (RR = 1.2; 95% CI = 1.1, 1.4); we observed no other differences.
Mode of diagnosis was significantly associated with sputum-smear status among foreign-born persons diagnosed greater than 6 months after arrival in the United States. More than 80% of these persons were diagnosed as a result of care seeking for symptoms, but those who were diagnosed in the course of treatment of non-TB medical conditions; during screening for employment, school, or living situations; or because they were close contacts of a TB patient were all less likely to be smear-positive (Table 3). Approximately 82% of all visa classes were diagnosed as a result of care seeking for symptoms.
TABLE 3—
Association Between Mode of Diagnosis and Presence of Acid-Fast Bacilli in Sputum Smears Among 975 Foreign-Born Persons With Tuberculosis Diagnosed > 6 Months After Arrival in the United States: 20 US Sites, 2005–2006
| Mode of Diagnosis | No. (%) | Smear Positive, No. (%) | Weighted RR (95% CI) |
| Symptoms | 786 (81.6) | 509 (64.1) | 3.7 (2.5, 5.4) |
| Non-TB medical condition | 92 (9.6) | 33 (32.6) | 1.8 (1.3, 2.8) |
| TB contact investigation | 41 (3.3) | 13 (26.4) | 1.5 (0.9, 2.5) |
| Screening for school, employment, living situation | 33 (3.4) | 6 (17.2) | 1.0 (Ref) |
| Postarrival immigration examination | 19 (1.8) | 4 (21.9) | a |
| Other | 4 (0.4) | 4 (100) | a |
| Total | 975 (100) | 569 (57.6) |
Note. CI = confidence interval; RR = relative risk; TB = tuberculosis.
Numbers too small for relative risk estimation.
Sensitivity Analysis
Of the 146 participants diagnosed with TB greater than 6 months after arrival whose visa status at diagnosis was unknown, 119 (85.9%) lived in New York City, where an executive order prohibited health department employees from inquiring about patients’ visa status.17 For the 829 participants with complete data who were included in the sensitivity analysis, the κ statistic capturing agreement between the observed and predicted visa status was 0.79.
In the reanalyses using the imputed visa class values, gender and visa status remained significant predictors of smear status. The undocumented were again more likely than both permanent residents and temporary visa holders to be smear-positive at diagnosis: RR = 1.3 for both (95% CI = 1.2, 1.4 for the comparison with permanent residents and 95% CI = 1.1, 1.4 for the comparison with temporary residents). Undocumented male participants were again more likely to be smear-positive compared with male permanent residents (RR = 1.4; 95% CI = 1.3, 1.5) and male temporary residents (RR = 1.3; 95% CI = 1.1, 1.5); for female participants, there was no statistically significant finding, which differs from the original finding with only the complete data (noted previously).
Missed Opportunities for Latent Tuberculosis Infection Screening
We divided participants diagnosed greater than 6 months after arrival into 2 groups: those diagnosed within 1 year of arrival, and those diagnosed later. Our decision to create these 2 groups was based upon the noted patterns of smear positivity between 6 months and 1 year after arrival depicted in Table 2.
Reported opportunities for LTBI testing among the 922 participants who had been in the United States at least 1 year before their TB diagnosis included enrolling in a school or college (n = 279), working in a health care facility or another job that required a health screening (n = 188), applying for a permanent visa (n = 113), becoming pregnant (n = 137), being identified as a contact to a TB patient in the United States (n = 101), and serving time or working in a prison (n = 76). Although 64.0% of persons diagnosed with TB a year or more after US entry reported that they had had 1 or more of these opportunities to be screened, only 29.1% of them reported that they had actually been tested with a tuberculin skin test (TST) at least 1 year earlier. Table 4 shows that permanent residents and temporary visa holders were both approximately 2 times more likely than the undocumented to have had a TST. When we excluded applying for a permanent visa from these testing opportunities, permanent and temporary residents were still more likely to have received a TST at least 1 year before diagnosis (RR = 1.7; 95% CI = 1.4, 2.1, and RR = 2.2; 95% CI = 1.8, 2.7, respectively). Only 14 participants diagnosed greater than 6 months but less than 1 year since arrival had an opportunity to be tested before diagnosis.
TABLE 4—
Reported Opportunities for Latent Tuberculosis Infection Testing by Visa Status for 782 Foreign-Born Persons With Known Visa Status With Tuberculosis Diagnosed a Year or More After Arrival in the United States: 20 US Sites, 2005–2006
| Visa Status at Diagnosis | No. (%) | No. (%) Who Had Screening Opportunity | Weighted RRa (95% CI) | No. (%) Who Had TB Skin Test ≥ 1 y Before TB Diagnosis | Weighted RRb (95% CI) |
| Permanent resident | 496 (67.1) | 357 (70.9) | 1.5 (1.3, 1.6) | 182 (35.1) | 1.8 (1.5, 2.2) |
| Temporary | 89 (12.3) | 67 (75.1) | 1.5 (1.4, 1.7) | 32 (38.0) | 2.2 (1.8, 2.7) |
| Undocumented | 197 (20.6) | 104 (48.4) | 1.0 (Ref) | 34 (18.1) | 1.0 (Ref) |
| Total | 782c | 528 (66.8) | 248 (32.0) |
Note. CI = confidence interval; RR = relative risk; TB = tuberculosis.
RR excludes 18 participants with unknown opportunity for latent tuberculosis infection testing.
RR excludes 315 participants with unknown tuberculin skin test and 6 participants who were already tuberculin skin test–positive.
Forty-seven participants diagnosed between months 6 and 12 were not included; 137 participants with unknown visa status at diagnosis were also excluded; of those, 56.8% had a screening opportunity and 27.2% were known to be tuberculin skin tested.
Although the American College Health Association has long recommended TB screening of foreign-born students,18 only 59 (22.6%) of those who enrolled in a school or college after arrival in the United States reported being tested by their school or college.
For foreign-born women from countries with high TB prevalence, the American College of Obstetrics and Gynecology recommends LTBI testing during prenatal care.19 Among 334 women diagnosed with TB a year or more after arrival in the United States, 137 (38.2%) reported becoming pregnant after arrival and only 35.0% of them received the TST at the time of their pregnancy. Among the 137 who reported pregnancy since arrival, 33.2% identified pregnancy as their only screening opportunity.
Participants were also asked about treatment of LTBI. Overall, of 222 persons who recalled receiving a TST at least a year before their TB diagnosis, 86 (39.9%) reported a positive test. Of those 86, only 18 (17.1%) reported receiving LTBI treatment of any kind. Thus, only 4.3% of the 922 cases diagnosed within 1 year of arrival were treated for LTBI.
DISCUSSION
As the proportion of the nation’s TB cases among the foreign-born continues to grow, reaching an expected 70% by 2020, TB elimination in the United States will increasingly depend on a comprehensive understanding of the epidemiology of TB in foreign-born populations.20 This large, multisite study adds to that understanding with the first detailed data on visa status and its association with infectious pulmonary TB. In the United States, visa status affects TB control because it determines access to health care, including screening services overseas (indirectly), and, in the United States, medical care, and public and private health insurance.21,22
The study’s main finding is an association between infectious (AFB-smear–positive) pulmonary TB and visa status. Compared with undocumented persons, legal permanent residents and US citizens are more likely to be diagnosed when they are less infectious (smear-negative). This association is strongest in the first 6 months after arrival in the United States, but it persists among persons who have resided in the United States longer.
This could be attributable in part to permanent residents’ greater access to health insurance, health care, and screening opportunities. Among study participants who were diagnosed because of care seeking for symptoms, almost two thirds (64.1%) were smear-positive at diagnosis, compared with less than a third (31.7%) diagnosed through contact investigations, screenings, and during treatment of other medical conditions. Most undocumented persons are uninsured and do not have regular medical providers who might check for TB in the course of routine examinations or through scheduled screenings. State and local TB programs receive notification of newly arrived permanent residents who require follow up, but the programs have no means to find undocumented persons or those with temporary visas who might have TB.
Some changes already implemented will reduce importation of TB disease into the United States. Approximately one third of study participants diagnosed within the first 6 months of arrival in the United States reported that they already had TB symptoms at or before arrival, and they could have had TB disease at entry. The most recent Division of Global Migration and Quarantine instructions to overseas physicians, which require sputum cultures of persons with abnormal chest radiographs (published October 2009), have shown a reduction in cases diagnosed within 6 months of arrival.16
But these changes address neither the problem of TB disease among recently arrived persons who are undocumented or who have temporary visas nor the larger problem of the 84% of TB cases that occur more than 6 months after arrival. Reducing TB morbidity among the foreign-born will require better stateside screening of foreign-born persons for TB during regular medical encounters that may identify LTBI or even active TB before the development of smear positivity. However, the estimated 11 million undocumented persons in the United States are largely without insurance and hence are unlikely to have regular medical encounters; they are thus unlikely to obtain care before symptoms develop. Implementation of the Affordable Care Act will have limited effect on insurance coverage among the undocumented, because the Affordable Care Act excludes undocumented persons.23 On the other hand, Federally Qualified Health Centers will continue to play a critical role in treating the undocumented, among other uninsured persons who constitite approximately one third of their patient base.
Statistical models of TB trends in the United States have shown that eliminating transmission of TB disease in the United States will not result in TB elimination (defined as 1 case per million persons) in the foreign-born population, because (1) an estimated 7.5 million foreign-born persons living in the United States already have LTBI acquired in their countries of birth and 5% to 10% of these, if untreated, will develop TB disease; and (2) millions of foreign-born persons arrive annually who already have LTBI acquired in their countries of origin.24,25 Elimination of TB also requires (1) improving TB-control programs in countries such as Mexico that are major origins of immigrants to the United States, and (2) improving LTBI screening and treatment of foreign-born persons in the United States.26
Although 60% of foreign-born study participants reported at least 1 screening opportunity in the 2 years before their TB diagnosis, fewer than 1 in 4 reported receiving a TST; this proportion was lower among undocumented persons. Moreover, among those who reported being tested, the vast majority (82.9%) of those who reported a positive test did not recall receiving any treatment. Some organizations already have guidelines that recommend testing for LTBI in specific foreign-born populations, such as students, pregnant women, and recent arrivals.18,19,27 However, more recent studies have called for expansion of testing to all foreign-born persons except those from nations with low TB incidence such as Canada, Australia, and Western Europe.28,29
Persons screened overseas and found to have LTBI should receive preventive TB treatment upon arrival in the United States. Particularly for the undocumented—an estimated 30% of all foreign-born persons in the United States—this will require expanded partnerships among private clinicians, TB-control programs, and funding agencies to remove cultural, organizational, and financial barriers to care. Even expanded testing and treatment of LTBI will not prevent all cases; indeed 17.1% of the 86 participants who tested positive for LTBI recalled receiving treatment of LTBI, but still developed TB disease. It is possible that some of these participants did not complete treatment, but those data were not available to us. However, it is known that treatment completion remains a significant challenge in the United States, with only 30% to 64% of patients completing treatment.30 New regimens that greatly reduce treatment duration may improve treatment completion rates for everyone, foreign-born or not.30
Limitations
This study had limitations. First, the selection of 6 months after arrival as a cutoff to distinguish between persons with likely imported TB (active TB that began before arrival) from incident TB (beginning after arrival) is somewhat arbitrary. We also analyzed a 1-year cutoff (results not shown), and results for were not different in any substantive way. For example, in the early period, the proportion of permanent residents still dominated among reported TB patients, and the proportion of smear-positives was still greatest among the undocumented and least among the permanent residents. In the later period, the undocumented, Mexicans or Latin Americans, male participants, and the uninsured were still more likely to be smear-positive at diagnosis.
A second limitation of this study was a certain amount of missing data. Persons with unknown smear status were treated as smear-negative. However, this makes the risk estimates conservative, as estimates of risk would have been higher had some of the unknowns actually had a positive smear and were analyzed as such. Visa status at diagnosis was missing for 20.8% of participants diagnosed greater than 6 months after arrival, and visa status at entry was missing for 10.8% of participants diagnosed within the first 6 months of arrival. However, a sensitivity analysis that imputed visa status did not change the study’s major conclusions. Although the study was not a national sample, the 20 enrollment sites accounted for 39% of all TB cases among foreign-born persons during the study period. In addition, study participants were of similar age, gender, country of origin, and time in the United States as all foreign-born persons diagnosed with TB during the same time period (data not shown).31
Self-reported visa status may differ from actual visa status. However, the internal consistency of the data suggests that most responded without purposeful misrepresentation as (1) most of the undocumented were from Mexico and Latin America, which is consistent with what is known about this population and (2) although they were told that they could refuse to provide any information, a large proportion of participants (25%) reported that they were undocumented. Furthermore, it is unlikely that those who misreported their visa status did so on the basis of knowledge of their sputum-smear status. It is probable that refusals to participate may have been higher among undocumented persons. However, this would not affect the association between smear status and visa status as it is unlikely that those who refused to participate were more or less likely to be smear-positive than those who agreed. Participant recall bias was likely more of an issue for reports of previous LTBI testing and treatment. Finally, our study design does not permit us to estimate the percentage of all foreign-born persons treated for LTBI who did not develop TB; only a cohort study could estimate the number of TB cases averted because of treatment of LTBI in the foreign-born.
Conclusions
Responding to the challenge of TB elimination will require efforts by both the public and private sectors to expand TB and LTBI testing and treatment among foreign-born persons in the United States. Progress toward elimination of TB in the United States will depend on expansion of opportunities for regular medical care for all persons regardless of visa status, promotion of screening and treatment of LTBI among foreign-born persons, and collaboration with international partners to improve TB control programs in less-developed nations.
Acknowledgments
This study was funded by the Centers for Disease Control and Prevention through the Tuberculosis Epidemiologic Studies Consortium.
Human Participant Protection
Institutional review board approval was received by all recruitment sites and by the Centers for Disease Control and Prevention.
References
- 1.Atlanta, GA: US Department of Health and Human Services, Centers for Disease Control and Prevention; 2013. Reported tuberculosis in the United States, 2012. [Google Scholar]
- 2.Shah NS, Cavanaugh JS, Pratt R et al. Epidemiology of smear-negative pulmonary tuberculosis in the United States, 1993–2008. Int J Tuberc Lung Dis. 2012;16(9):1234–1240. doi: 10.5588/ijtld.11.0794. [DOI] [PubMed] [Google Scholar]
- 3.Tostmann A, Kik SV, Kalisvaart NA et al. Tuberculosis transmission by patients with smear-negative pulmonary tuberculosis in a large cohort in the Netherlands. Clin Infect Dis. 2008;47(9):1135–1142. doi: 10.1086/591974. [DOI] [PubMed] [Google Scholar]
- 4.Behr MA, Warren SA, Salamon H et al. Transmission of Mycobacterium tuberculosis from patients smear-negative for acid-fast bacilli. Lancet. 1999;353(9151):444–449. doi: 10.1016/s0140-6736(98)03406-0. [DOI] [PubMed] [Google Scholar]
- 5.Moonan PK, Ghosh S, Oeltmann JE, Kammerer JS, Cowan LS, Navin TR. Using genotyping and geospatial scanning to estimate recent Mycobacterium tuberculosis transmission, United States. Emerg Infect Dis. 2012;18(3):458–465. doi: 10.3201/eid1803.111107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Centers for Disease Control and Prevention. Online tuberculosis information system, National Tuberculosis Surveillance System, United States. Available at: http://wonder.cdc.gov/tb.html. Accessed September 30, 2013.
- 7.Centers for Disease Control and Prevention. Tuberculosis screening and treatment technical instructions (TB TIs) using cultures and directly observed therapy (DOT) for panel physicians. Available at: http://www.cdc.gov/immigrantrefugeehealth/exams/ti/panel/tuberculosis-panel-technical-instructions.html. Accessed April 9, 2015.
- 8.US Citizen and Immigration Services. Policy Manual. Available at http://uscis.gov/policymanual/HTML/PolicyManual.html. Accessed January 30, 2014.
- 9. National TB Program objectives and performance targets for 2015. Available at: http://www.cdc.gov/tb/programs/Evaluation/Indicators/default.htm. Accessed April 9, 2015.
- 10.Lee A, Batalova J. Temporary admissions of nonimmigrants to the United States. Migration Information Source. Available at: http://www.migrationinformation.org/usfocus/display.cfm?ID=881. Accessed January 30, 2014.
- 11.Warren R, Warren JR. Unauthorized immigration to the United States: annual estimates and components of change, by state, 1990 to 2010. Int Migr Rev. 2013;47(2):296–329. doi: 10.1111/imre.12022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Liu Y, Painter JA, Posey DL et al. Estimating the impact of newly arrived foreign-born persons on tuberculosis in the United States. PLoS ONE. 2012;7(2):e32158. doi: 10.1371/journal.pone.0032158. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Essential components of a tuberculosis prevention and control program. Recommendations of the Advisory Council for the Elimination of Tuberculosis. MMWR Recomm Rep. 1995;44(RR-11):1–16. [PubMed] [Google Scholar]
- 14.Katz D, Albalak R, Wing JS, Combs V. Setting the agenda: a new model for collaborative tuberculosis epidemiologic research. Tuberculosis (Edinb) 2007;87(1):1–6. doi: 10.1016/j.tube.2005.12.003. [DOI] [PubMed] [Google Scholar]
- 15.Davidow AL, Katz D, Reves R, Bethel J, Ngong L the Tuberculosis Epidemiologic Studies Consortium. The challenge of multisite epidemiologic studies in diverse populations: design and implementation of a 22-site study of tuberculosis in foreign-born people. Public Health Rep. 2009;124(3):391–399. doi: 10.1177/003335490912400308. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Lowenthal P, Westenhouse J, Moore M, Posey DL, Flood J. Reduced importation of tuberculosis after the implementation of an enhanced preimmigration screening protocol. Int J Tuberc Lung Dis. 2011;15(6):761–766. doi: 10.5588/ijtld.10.0370. [DOI] [PubMed] [Google Scholar]
- 17.City of New York Office of the Mayor. Executive Order No. 34. 2003. Available at: http://www.nyc.gov/html/imm/html/rights/rights.shtml. Accessed January 30, 2014.
- 18.American College Health Association. November 2000, tuberculosis screening of college and university students.ACHA Guidelines.2012.Available at: http://www.acha.org/Publications/docs/ACHA_Tuberculosis_Screening_Apr2012.pdf. Accessed January 30, 2014.
- 19.Pulmonary disease in pregnancy. ACOG Technical Bulletin Number 224. Washington, DC: American College of Obstetricians and Gynecologists; 1996. [PubMed] [Google Scholar]
- 20.Woodruff RS, Winston CA, Miramontes R. Predicting U.S. tuberculosis case counts through 2020. PLoS ONE. 2013;8(6):e65276. doi: 10.1371/journal.pone.0065276. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Sommers BD. Stuck between health and immigration reform—care for undocumented immigrants. N Engl J Med. 2013;369(7):593–595. doi: 10.1056/NEJMp1306636. [DOI] [PubMed] [Google Scholar]
- 22.Breen JO. Lost in translation — ¿Cómo se dice, “Patient Protection and Affordable Care Act”? N Engl J Med. 2012;366(22):2045–2047. doi: 10.1056/NEJMp1202039. [DOI] [PubMed] [Google Scholar]
- 23.Insurance coverage provisions of the Affordable Care Act. CBO’s February 2014 baseline. Available at: http://www.cbo.gov/sites/default/files/cbofiles/attachments/43900-2014-02-ACAtables.pdf. Accessed May 14, 2014.
- 24.Shea KM, Kammerer JS, Winston CA, Navin TR, Horsburgh CR., II Estimated rate of reactivation of latent tuberculosis infection in the United States, overall and by population subgroup. Am J Epidemiol. 2014;179(2):216–225. doi: 10.1093/aje/kwt246. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Hill AN, Becerra J, Castro KG. Modelling TB trends in the U.S. Epidemiol Infect. 2012;14(10):1862–1872. doi: 10.1017/S095026881100286X. [DOI] [PubMed] [Google Scholar]
- 26.Schwartzman K, Oxlade O, Barr RG et al. Domestic returns from investment in the control of tuberculosis in other countries. N Engl J Med. 2005;353(10):1008–1020. doi: 10.1056/NEJMsa043194. [DOI] [PubMed] [Google Scholar]
- 27.American Thoracic Society. Targeted tuberculin testing and treatment of latent tuberculosis infection. Am J Respir Crit Care Med. 2000;161(4 pt 2):S221–S247. doi: 10.1164/ajrccm.161.supplement_3.ats600. [DOI] [PubMed] [Google Scholar]
- 28.Linas BP, Wong AY, Freedberg KA, Horsburgh CR., II Priorities for screening and treatment of latent tuberculosis infection in the United States. Am J Respir Crit Care Med. 2011;184(5):590–601. doi: 10.1164/rccm.201101-0181OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Walter ND, Painter J, Parker M et al. Persistent latent tuberculosis reactivation risk in US immigrants. Am J Respir Crit Care Med. 2014;189(1):88–95. doi: 10.1164/rccm.201308-1480OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Sterling TR, Villarino E, Borisov AS et al. for the TB Trials Consortium PREVENT TB Study Team. Three months of rifapentine and isoniazid for latent tuberculosis infection. N Engl J Med. 2011;365:2155–2166. doi: 10.1056/NEJMoa1104875. [DOI] [PubMed] [Google Scholar]
- 31.Katz D, Davidow A, Reves R Tuberculosis Epidemiologic Studies Consortium. The epidemiology of tuberculosis in foreign-born persons in the United States. Poster presented at: International Union Against Tuberculosis and Lung Disease Conference: North American Region; February 22–24, 2007; Vancouver, British Columbia.