Talbot EA, Moore M, McCray E, Binkin NJ. Tuberculosis Among Foreign-Born Persons in the United States, 1993-1998. JAMA. 2000;284(22):2894–2900. doi:10.1001/jama.284.22.2894
Author Affiliations: Division of Tuberculosis Elimination, National Center for HIV, STD and TB Prevention (Drs Talbot, Moore, McCray, and Binkin), and the Division of Applied Public Health Training, Epidemiology Program Office (Dr Talbot), Centers for Disease Control and Prevention, Atlanta, Ga.
Context Immigration is a major force sustaining the incidence of tuberculosis
(TB) in the United States.
Objective To describe trends and characteristics of foreign-born persons with
TB and the implications for TB program planning and policy development.
Design, Setting, and Subjects Descriptive analysis of US TB surveillance data from case reports submitted
from 1993 to 1998.
Main Outcome Measure Demographic and clinical characteristics of foreign-born persons with
Results The number of TB cases among foreign-born persons increased 2.6%, from
7402 in 1993 to 7591 in 1998, and the proportion of US cases that were foreign-born
increased from 29.8% to 41.6%. During 1993-1998, the TB case rate was 32.9
per 100,000 population in foreign-born persons compared with 5.8 per 100,000
in US-born persons. Six states reported 73.4% of foreign-born cases (California,
New York, Texas, Florida, New Jersey, and Illinois). Approximately two thirds
of these cases were originally from Mexico, the Philippines, Vietnam, India,
China, Haiti, and South Korea. Among those for whom date of US entry was known,
51.5% arrived 5 years or less prior to the diagnosis of TB. Most were male
and aged 25 to 44 years. During 1993-1996, the proportion receiving some portion
of treatment under directly observed therapy increased from 27.3% to 59.1%
and approximately 70% completed therapy in 12 months. The rate of primary
resistance to isoniazid was 11.6% and to both isoniazid and rifampin was 1.7%.
Conclusions As the United States moves toward the goal of TB elimination, success
will depend increasingly on reducing the impact of TB in foreign-born persons.
Continued efforts to tailor local TB control strategies to the foreign-born
community and commitment to the global TB battle are essential.
The World Health Organization estimates that one third of the world's
population is infected with the causative organism of tuberculosis (TB), Mycobacterium tuberculosis; that there are 8 million new
cases of active TB annually; and that nearly 2 million persons die of TB each
year.1 Most infections, cases, and deaths occur
in developing countries. In a number of developed countries with substantial
levels of immigration, however, foreign-born persons increasingly contribute
to the incidence of TB and sustain TB rates.2- 4
Immigration has contributed substantially to changes in TB epidemiology
in the United States during the last decade and is considered an important
factor in the resurgence of TB during the late 1980s and early 1990s. Although
the number of reported cases of TB has decreased steadily since the peak of
the resurgence in 1992, the decline has been limited to persons born in the
United States.5,6 The success
of TB control efforts depends on successfully defining the at-risk populations,
which will assist in activities such as case finding, program planning to
meet unique service needs, and targeting prevention efforts. To highlight
national trends in characteristics of foreign-born TB patients and the potential
implications for TB program planning and policy development, we analyzed data
from the national TB surveillance system, which receives case reports on all
TB patients included in annual state morbidity totals.
All 50 states and the District of Columbia report TB cases to the national
TB surveillance system using a standardized case report form.7- 9
We analyzed data from case reports submitted from 1993 through 1998. The case
report collects information on demographic and clinical characteristics, including
country of birth and selected TB risk factors. Information on immigration
status is not collected.
Consistent with the standard definition used in national TB reporting,
a US-born person was defined as a person who was born in the United States
or its associated jurisdictions or was born in a foreign country but had at
least 1 US parent. A person who did not meet these criteria was classified
as foreign-born. If a case report did not include information regarding country
of birth, it was excluded from analysis. The case report also includes information
on the month and year of immigration to the United States. For reports that
included only the year of immigration, July was assigned as the month.
California does not submit human immunodeficiency virus (HIV) test results
to the surveillance system, but does submit the results of TB and acquired
immunodeficiency syndrome (AIDS) registry crossmatches. California TB cases
with an AIDS match were classified as HIV-positive; all others were classified
as having an unknown HIV status.
Annual population estimates by nativity, including estimates for specific
age and sex groups, were obtained from the National Population Estimates,
which contains post-1990 census estimates from the US Census Bureau.10 The population of foreign-born persons from each
birth country was obtained by applying the percentage reported in the 1990
census to the annual National Population Estimates for foreign-born persons.
Proportions were compared using the χ2 test. Linear trends
were tested by χ2 for trend.
From 1993 through 1998, 131,377 new TB cases were reported from the
50 states and the District of Columbia. Of these, 84,095 cases (64.0%) occurred
among persons born in the United States and 46,123 cases (35.1%) occurred
in foreign-born persons; reports for 1159 cases (0.9%) were missing information
on birth country and were excluded from analysis. As a result of the substantial
decrease in the number of cases in US-born but not foreign-born persons during
this period, the proportion of TB cases that occurred in foreign-born persons
increased from 29.8% to 41.6% (Table 1).
The rates for both US-born and foreign-born persons decreased during this
period; however, the foreign-born rate remained at least 5 times higher than
the US-born rate (Table 1). The
6 states that reported the most cases overall (California, New York, Texas,
Florida, New Jersey, and Illinois) reported 77,238 (58.8%) of the US total
TB cases and 33,834 (73.4%) of the TB cases in foreign-born persons. The number
of states reporting more than 50% of TB cases among foreign-born persons increased
from 6 in 1993 to 13 in 1998 (Figure 1).
The numbers of TB cases and case rates among foreign-born persons by
birth country are shown in Table 2.
In each of the 6 years of observation, approximately two thirds of foreign-born
persons with TB had come to the United States from 1 of 7 countries: Mexico,
the Philippines, Vietnam, China, India, Haiti, and South Korea. Among the
6 highest case-reporting states, Mexico was the most commonly reported birth
country for foreign-born persons with TB in Texas, California, and Illinois
and accounted for 59.5%, 33.0%, and 24.6% of these states' cases in foreign-born
persons, respectively. In Florida, Haiti was the most common birth country
(34.3%) and in New Jersey, India was most common (18.8%). In New York, 3 countries
were each reported as the birth country for approximately 10% of foreign-born
TB cases (China, the Dominican Republic, and Haiti). Table 2 shows overall and birth country–specific proportions
of persons diagnosed with TB less than 1 year, 1 to 5 years, and more than
5 years after arrival in the United States.
Numbers and percentages of TB cases and TB case rates in US-born and
foreign-born persons during 1993-1998 are shown by age group and sex in Table 3. The majority of cases occurred
in males in both the US-born and foreign-born populations. There was minimal
variation by year. The largest numbers of cases among both US-born and foreign-born
persons occurred in the 25- to 44-year-old age group, but the highest rate
among US-born persons occurred in those aged 65 years or older, and among
foreign-born persons in those younger than 5 years.
Table 4 presents data on
substance use, occupation, and residence history of US-born and foreign-born
persons with TB. Drug use, excess alcohol use, unemployment, homelessness,
and residence in a correctional or long-term care facility were reported more
frequently for US-born persons. Among foreign-born persons reporting homelessness
in the year prior to diagnosis, the most common countries of origin were Mexico
(47.6%), Cuba (5.6%), El Salvador (4.1%), Guatemala (4.1%), and Haiti (3.7%).
During 1994-1998 (the years with more than 70% of results known for each characteristic),
the proportion of persons with TB who were substance users, homeless, or residents
of a correctional facility was 29.3% for US-born persons compared with 9.3%
for foreign-born persons.
Selected clinical characteristics are presented in Table 5. Among the 6 highest case-reporting states, the percentage
of case reports with HIV test results for foreign-born persons ranged from
18.3% in Illinois to more than 50% in New York and Florida and for US-born
persons from 33.0% in Illinois to 59.0% in New York. A higher proportion of
US-born persons were infected with HIV. In both US-born and foreign born persons,
nearly three fourths (72%) of TB patients with HIV infection were aged 25
to 44 years. More than half of the reported foreign-born persons with TB and
HIV infection were from California (27.1%) or New York (26.4%). The majority
were from 1 of 2 birth countries, Mexico (26.4%) and Haiti (23.8%).
During 1993 through 1998, a higher proportion of foreign-born persons
were initially treated with at least the recommended 4-drug regimen. The proportion
increased during the 6 years of observation from 38.8% to 71.0% among US-born
persons and from 53.7% to 83.4% among foreign-born persons (χ2
for trend, P<.001). Based on follow-up data available
through 1996, a higher proportion of US-born persons received at least some
part of their antituberculosis therapy as directly observed therapy. This
proportion increased from 40.1% to 68.1% in US-born persons and from 27.3%
to 59.1% in foreign-born persons (χ2 for trend, P<.001). Among both US-born and foreign-born persons without known
rifampin resistance, approximately 70% completed therapy within 12 months.
The percentage completing therapy in 12 months increased from 64% in 1993
to 75% in 1996 (the year with the most recent data available) for both US-born
and foreign-born persons (χ2 for trend, P<.001).
Table 6 presents the frequency
of drug resistance in M tuberculosis isolates from
US-born persons, all foreign-born persons, and foreign-born persons from the
7 most common countries of birth in 1993-1998. Susceptibility test results
were available for initial isolates from more than 90% of both US-born and
foreign-born persons with culture-positive TB. Among those without prior TB,
US-born persons were less likely to have an isolate with any drug resistance,
although rates of resistance to at least isoniazid and rifampin (multidrug-resistant
TB [MDRTB]) were similar. The proportion with MDRTB decreased from 2.6% in
1993 to 1.0% in 1998 among US-born persons and from 2.3% to 1.5% among foreign-born
persons (χ2 for trend, P<.01).
Among those with prior TB, US-born persons were also less likely to have an
isolate with any drug resistance, and rates of MDRTB were significantly lower
among US-born persons.
Our findings provide a current profile of foreign-born TB patients in
the United States. The features in most striking contrast to US-born patients
include the stable numbers of patients and substantially higher TB case rates
during the study period. Tuberculosis case rates, however, should be interpreted
with the understanding that populations of foreign-born persons may be underestimated.
Also important to control and prevention efforts is the geographic variation
in proportions of foreign-born TB patients and in countries of origin.
Based on extrapolation and assuming that changes in the number of US-born
and foreign-born persons with TB continue to occur at the rate observed in
1993-1998, more than half of US cases may occur in foreign-born persons by
the year 2002. The marked decrease in cases among US-born persons was an expected
outcome of TB control efforts that prioritized prompt identification of persons
with active TB and initiation and completion of appropriate therapy.11 This approach primarily reduces ongoing transmission
and the number of cases caused by recent infection. For complex reasons, this
same programmatic approach has not been as effective in controlling rates
among foreign-born persons, but it is likely related to a higher prevalence
of latent infection in the foreign-born population.5,12
Tuberculosis case reports do not include information to determine whether
cases are a result of recent transmission or reactivation of latent infection;
however, studies using DNA fingerprinting methods to evaluate evidence for
recent transmission also support this inference.13- 15
Thus, interruption of transmission through treatment and contact investigations
alone is insufficient to reduce TB cases among foreign-born persons, and efforts
to prevent the transition from latent infection to active disease are needed
to complement these core TB control activities.
Because of the considerable geographic variations in TB in foreign-born
persons, approaches to controlling and preventing TB should be tailored locally
to at-risk foreign-born populations.16 Our
findings concerning duration of residence in the United States prior to TB
diagnosis have several implications for tailoring these approaches. In areas
where the majority of TB cases are among recent arrivals, the emphasis should
be on providing adequate follow-up for persons indicated as having TB by immigrant
and refugee screening17- 20
and on screening new arrivals who have not undergone such screening. In contrast,
in areas where the majority of cases have been in the United States several
years prior to diagnosis, the emphasis may be better placed on screening for
latent infection in foreign-born populations.
Tuberculosis case rates were high in all age groups of foreign-born
persons, including foreign-born children. Tuberculosis in children implies
recent transmission in their communities. Since the source case for children
is often an adult, foreign-born adults who infect their foreign-born children
may contribute to this high rate.21 Foreign-born
adults may also infect their US-born offspring or other children in their
care.21,22 A child may also be
infected during travel to the birth country of a parent.22
Children should be considered high priority for evaluation and treatment during
contact investigations.23,24 Investigations
to identify the source case for children are also recommended to prevent further
Foreign-born persons with TB were less likely to have risk factors for
TB, such as a history of homelessness, residence in a correctional facility,
or excess alcohol or injection drug use. Moreover, less than 10% of cases
in the foreign-born population occurred in persons with HIV coinfection, based
on the minimum estimates provided by our data. Thus, the most important risk
factor for establishing risk for TB for foreign-born persons appears to be
previous residence in a country with a high rate of TB. However, the level
of completeness of the data for these risk factors requires caution in interpreting
differences because potential ascertainment bias may exist.
Our study extends previous reports of higher levels of isoniazid resistance
in foreign-born persons.26 A higher proportion
of foreign-born patients started initial antituberculosis drug regimens of
at least 4 first-line drugs, although the proportions in both foreign-born
and US-born patients increased during the study period. These findings may
reflect increased awareness among health care professionals that high levels
of isoniazid resistance exist among foreign-born persons and that initial
drug regimens of 4 first-line drugs are recommended to prevent development
of MDRTB when there are individual patient risk factors for resistance or
when population levels of isoniazid resistance exceed 4%.27
Levels of drug resistance, especially isoniazid resistance, also have
important implications for efforts to treat latent TB infection in some foreign-born
populations. Isoniazid has been the standard for treating latent M tuberculosis infection in the absence of known contact to drug-resistant
TB. However, high levels of isoniazid resistance in some foreign-born populations,
including most of those contributing to the majority of TB cases in the United
States, raise concern about the efficacy of isoniazid preventive therapy in
these populations. Updated recommendations on the treatment of latent TB infection
provide several alternative regimens.23,28
Further study of the acceptability and cost-effectiveness of using alternative
regimens in settings where rates of isoniazid-resistant TB are high will be
helpful for planning future prevention efforts in foreign-born populations.23
As the United States moves toward the ultimate goal of TB elimination
(<1 case per 1 million persons), success will depend increasingly on reducing
the impact of TB in foreign-born persons.29
A close relationship exists between the global TB crisis and the impact of
the disease in the United States. For this reason, the Centers for Disease
Control and Prevention supports international TB control efforts, as well
as efforts to improve the prevention and control of TB among foreign-born
persons in the United States.30- 33
Continued efforts to tailor local TB control strategies to the foreign-born
community and commitment to the global TB battle are essential.