Skip to main content Skip to left navigation Skip to footer
Skip Left Navigation

Browse by Topic

Latent Tuberculosis Infection, Screening, 2016

* Indicates an old grade definition

Recommendations: Screening for Latent Tuberculosis Infection


Burden of Disease

Tuberculosis causes a substantial health burden globally. Approximately one-third of the world’s population is infected with tuberculosis; in 2014, 9.6 million persons were estimated to have contracted tuberculosis, and an estimated 1.5 million deaths related to tuberculosis infection occurred worldwide.9 In the United States, 9563 new active cases of tuberculosis were reported in 2015, which corresponds to an incidence rate of 3.0 cases per 100,000 persons.7 In 2013, 555 deaths from tuberculosis were reported in the United States.28 In 2015, half of all tuberculosis cases occurred in 4 states: California, Texas, New York, and Florida. Asians represented the largest percentage of total cases (33%), followed by Hispanics (28%), African Americans (21%), and whites (13%); American Indian or Alaska Natives and Native Hawaiian or other Pacific Islanders each represented approximately 1% of cases.7 Incidence rates of active tuberculosis may be higher in populations at increased risk, owing to greater likelihood of exposure (eg, persons who have lived in countries with a high tuberculosis burden) or greater likelihood of progression from LTBI to active disease (eg, persons who are immunosuppressed). Although LTBI is asymptomatic, signs and symptoms of active tuberculosis disease may include cough, hemoptysis (coughing up blood), unexplained weight loss, night sweats, fever, chills, and fatigue.

Scope of Review

The USPSTF commissioned a systematic review of the evidence on screening for LTBI.2, 3 Evidence dating from the inception of searched databases until August 3, 2015, was included. The review focused on evidence on screening for LTBI in asymptomatic adults seen in primary care settings. It did not include evidence on screening in persons for whom LTBI screening would be considered management of a specific condition (eg, persons living with HIV), public health surveillance (ie, tracing contacts of persons with active tuberculosis disease), surveillance of employees working in high-risk settings, or screening indicated prior to the use of specific immunosuppressive medications.

Accuracy of Screening Tests

There is no direct test for the diagnosis of latent infection with M tuberculosis. Following screening, diagnosis of LTBI is based on medical history, physical examination, and exclusion of active tuberculosis disease. In the absence of a reference standard for detection of LTBI, screening test performance is based on detection of disease in persons with known active tuberculosis and nondetection of disease in populations at low risk for the disease and presumed not to have LTBI or active tuberculosis.The USPSTF identified 67 good- or fair-quality studies that provided information on the accuracy and reliability of screening tests for LTBI.2 For studies reporting on sensitivity, 8 were conducted in countries with a high burden of tuberculosis, 29 were conducted in countries with an intermediate burden, 10 were conducted in countries with a low burden, and 3 were conducted in countries with a mix of low to intermediate burden. For studies reporting on specificity, 3 were conducted in countries with an intermediate burden; 14 were conducted in countries with a low burden; and 1 was conducted in 2 countries: 1 with an intermediate burden and 1 with a low burden.When using a positive threshold of 10 mm of induration, the TST has moderate sensitivity and high specificity for detection of LTBI. Based on pooled analyses of studies reviewed by the USPSTF, when using a positive threshold of 10 mm, the TST has sensitivity of 79% (11 studies; n=988) and specificity of 97% (9 studies; n=9651).2Pooled analyses of the T-SPOT.TB test (a type of IGRA) indicate sensitivity of 90% (16 studies; n=984) and specificity of 95% (5 studies; n=1810). Pooled analyses of the QuantiFERON-TB Gold In-Tube test (another type of IGRA) indicate sensitivity of 80% (24 studies; n=2321) and specificity of 97% (4 studies; n=2053). The USPSTF identified no studies that evaluated the accuracy and reliability of sequential screening strategies.

Effectiveness of Early Detection and Treatment

The USPSTF identified no randomized clinical trials that compared screening with no screening to provide direct evidence of the benefit of screening for LTBI on health outcomes, such as rates of active tuberculosis disease, disease-specific or all-cause mortality, or tuberculosis transmission. Three good- or fair-quality trials (n=35,563) conducted in Canada, Brazil, Saudi Arabia, Spain, Czechoslovakia, Finland, Germany, Hungary, Poland, Romania, and Yugoslavia provided evidence on the benefits of treatment of LTBI.2 Trials evaluated treatment with isoniazid,29 rifampin,30 and rifapentine plus isoniazid.31The best evidence on the effectiveness of treatment was from the International Union Against Tuberculosis (IUAT) trial. This good-quality randomized clinical trial was conducted in 7 European countries (Czechoslovakia, Finland, Germany, Hungary, Poland, Romania, and Yugoslavia) among participants with fibrotic pulmonary lesions but not active tuberculosis. The trial, published in 1982, included 27,830 participants and evaluated treatment with daily isoniazid. It found that at 5 years, the relative risk (RR) of progression to active tuberculosis was 0.35 (95% CI, 0.24 to 0.52) for treatment with isoniazid (300 mg daily for 24 weeks) compared with placebo. The trial reported fewer deaths attributable to tuberculosis among participants receiving treatment with isoniazid (0 vs 3 deaths in the placebo group; RR, 0.14 [95% CI, 0.01 to 2.78]), although this difference was not statistically significant.The other 2 treatment trials compared either rifampin with isoniazid and found zero deaths in either group or rifapentine plus isoniazid with isoniazid alone and found that the combination therapy was noninferior in preventing progression to active tuberculosis. None of the treatment studies reported on transmission rates of tuberculosis.

Potential Harms of Screening and Treatment

The USPSTF identified no studies that directly reported on the harms of screening. Potential harms include stigma associated with screening and diagnostic workup and treatment of false-positive results. Five good- or fair-quality studies (n=36,043) conducted in the United States, Canada, Saudi Arabia, Brazil, Spain, Czechoslovakia, Finland, Germany, Hungary, Poland, Romania, and Yugoslavia reported on the harms of treatment.2, 29-33 Interventions evaluated included isoniazid, rifampin, and rifapentine plus isoniazid. The most consistently reported harm was hepatotoxicity. The only study that assessed harms of treatment vs placebo was the IUAT trial,30 which found an RR of 4.59 (95% CI, 2.03 to 10.39) for hepatotoxicity at 5 years among participants being treated with isoniazid (300 mg for 24 weeks) vs placebo. The IUAT trial also reported more deaths from hepatotoxicity among participants being treated with isoniazid than with placebo, although this finding was not statistically significant (0.14 vs 0 deaths per 1000 persons; RR calculated from published data, 2.35 [95% CI, 0.12 to 45.46]).The other trials compared either rifampin30, 32, 33 or rifapentine plus isoniazid31 with isoniazid. Meta-analysis of 3 trials of rifampin compared with isoniazid found a higher RR for hepatoxicity among participants being treated with isoniazid (RR, 3.29 [95% CI, 1.72 to 6.28]).2 None of these trials, which were more recent than the IUAT trial, reported any deaths from hepatotoxicity. The 1 study that reported on hepatotoxicity of rifapentine plus isoniazid vs isoniazid alone found a nonsignificant reduced RR of 0.90 (95% CI, 0.75 to 1.08) for grade 3 or 4 hepatotoxicity among participants being treated with rifapentine plus isoniazid. There also was a nonsignificant reduced RR of death from hepatotoxicity among participants being treated with rifapentine plus isoniazid versus isoniazid alone (RR, 0.83 [95% CI, 0.51 to 1.35]).A few studies also reported on gastrointestinal adverse events. Compared with placebo, participants treated with isoniazid had a higher risk of medication discontinuation because of gastrointestinal adverse events (RR, 1.33 [95% CI, 1.01 to 1.75]).29 Compared with rifampin, treatment with isoniazid had a nonsignificant increased RR of gastrointestinal adverse events (RR, 1.60 [95% CI, 0.76 to 3.40]) in 2 studies.2 All 5 studies also reported on discontinuation of treatment because of adverse events. Compared with placebo, treatment with isoniazid had an RR of medication discontinuation of 1.50 (95% CI, 1.18 to 1.89).29 Pooling results from the 3 studies of isoniazid vs rifampin found a nonsignificant increased risk of discontinuation with isoniazid (RR, 1.61 [95% CI, 0.57 to 4.57]).2 The study of rifapentine plus isoniazid vs isoniazid alone found an increased risk of discontinuation with rifapentine plus isoniazid (RR, 1.28 [95% CI, 1.03 to 1.59]).31

Estimate of Magnitude of Net Benefit

Overall, the USPSTF found adequate evidence that accurate screening tests for LTBI are available, treatment of LTBI provides a moderate health benefit in preventing progression to active disease, and the harms of screening and treatment are small. The USPSTF has moderate certainty that screening for LTBI in persons at increased risk for infection provides a moderate net benefit. The USPSTF estimated that if a hypothetical cohort of 100,000 asymptomatic adults at increased risk for tuberculosis (eg, persons born in, or former residents of, high-prevalence countries) were screened, 52 to 146 active tuberculosis cases would be prevented, 7 to 67 cases of hepatotoxicity would occur (depending on type of treatment), and 111 persons would discontinue treatment because of adverse events. The number needed to treat to prevent 1 case of LTBI from progressing to active tuberculosis would range from 111 to 314 (depending on the patient’s risk for progression), and the number needed to harm to cause 1 case of hepatotoxicity from treatment would range from 279 to 2531 (depending on type of treatment). These estimates are based on prevalence data from the 2011–2012 National Health and Nutrition Examination Survey1 and numerous assumptions about screening sensitivity and specificity (eg, using the TST with a 10-mm threshold for a positive diagnosis) and potential benefits of treatment (eg, estimated efficacy of treatment for 24 weeks of isoniazid, based on IUAT trial findings). Further information on the assumptions used is available in the corresponding evidence review.2

How Does Evidence Fit With Biological Understanding?

Tuberculosis disease is caused by M tuberculosis, which is spread through airborne transmission when a person with active pulmonary tuberculosis coughs or sneezes. When the tuberculosis bacillus is inhaled, a person can either clear M tuberculosis; develop active disease (primary tuberculosis disease), which may be infectious; or develop latent infection (LTBI), which is asymptomatic and not infectious. Latent infection can later reactivate and progress to active tuberculosis disease. Approximately 30% of persons exposed to active M tuberculosis will develop LTBI.2 Approximately 5% to 10% of persons with a positive TST result will experience reactivation of LTBI and progress to active tuberculosis disease.2-6

Response to Public Comment

A draft version of this recommendation statement was posted for public comment on the USPSTF website from March 8 to April 4, 2016. Many comments sought clarification around risk assessment of populations who should receive screening. The USPSTF clarified that given regional variations in the local populations considered at risk for tuberculosis, clinicians may consult their local or state public health agency for additional details on specific populations at risk in their community. Furthermore, the USPSTF clarified that although persons with diabetes and pregnant women are not addressed separately in this recommendation statement, they are also not excluded from the recommendation. A few public comments sought clarification on the recommended frequency of screening. Although the USPSTF sought evidence on screening frequency, there was not enough evidence available to determine an optimal screening interval. Several comments requested that the recommendation include treatment of LTBI. While the USPSTF acknowledges that treatment of LTBI contributes to the success of LTBI screening, it is beyond the scope of the USPSTF to make any specific recommendations on treatment. The CDC provides treatment guidelines for LTBI.17

Other Sections:

Other Considerations


Screening with the TST requires that patients return 48 to 72 hours after administration of the skin test for interpretation of results. When placing a TST, clinicians should plan with patients accordingly to ensure they can return in time and that the facility is able to interpret the test results within the proper time frame.26 Screening with an IGRA requires obtaining a single venous blood sample, and patients do not need to return for interpretation of results. However, clinicians should be aware of processing requirements for blood samples and ensure that venous blood samples are drawn and can reach the laboratory for processing within the appropriate time frame (8 to 30 hours, depending on the test).27

Research Needs and Gaps

Further research is needed that evaluates risk assessment tools to determine efficient ways of identifying candidates for LTBI testing and treatment. Additional research on how often LTBI screening should be performed in different subpopulations is also needed. The USPSTF identified no studies on LTBI screening or treatment in pregnant women and the potential effects on the fetus; this represents an important gap in the literature that needs further research. In addition, more studies are needed to clarify whether certain screening methods are preferable for certain risk groups.

Update of USPSTF Recommendations

The USPSTF last issued a recommendation on screening for tuberculosis in 1996. At that time, the USPSTF recommended screening for tuberculosis infection with the TST in asymptomatic, high-risk persons (A recommendation) and consideration of bacille Calmette–Guérin vaccination for selected high-risk individuals only (B recommendation). Given the changes in the epidemiology of the disease, the development of newer screening technologies, and newer methods for developing evidence-based recommendations, the USPSTF decided to update the topic and issue a recommendation using its current methodology and considering all of the available evidence, including studies published prior to 1996.

Recommendations of Others

The American Academy of Family Physicians recommends screening for LTBI in populations at increased risk.34 In 2005, the CDC, the American Thoracic Society, and the Infectious Diseases Society of America issued joint guidelines recommending that clinicians screen for LTBI only among high-risk populations and when treatment is feasible.35 In its 2013 "Guide for Primary Health Care Providers," the CDC recommended targeted testing for tuberculosis among high-risk populations only.8 The CDC identifies persons at risk for developing tuberculosis as those who have an increased likelihood of exposure to persons with tuberculosis disease (known close contacts of a person with infectious tuberculosis disease, persons who have immigrated from tuberculosis-endemic regions of the world, and persons who work or reside in facilities or institutions with those at high risk for tuberculosis) or persons with clinical conditions or other factors associated with an increased risk of progression from LTBI to tuberculosis disease (HIV infection, injection drug use, radiographic evidence of prior healed tuberculosis, low body weight, or other medical conditions). Further information on targeted testing is available from the CDC.36The WHO also recently issued guidelines on the management of LTBI. For high-income countries with an estimated tuberculosis incidence of less than 100 cases per 100,000 persons (such as the United States), the WHO recommends systematic testing for and treatment of LTBI among persons living with HIV, adult and child contacts of persons with pulmonary tuberculosis, patients initiating anti-tumor necrosis factor treatment, patients receiving dialysis, patients preparing for an organ or hematologic transplant, and patients with silicosis. Either an IGRA or the TST should be used. The WHO also recommends considering systematic testing and treatment among prisoners, health care workers, immigrants from high-burden countries, homeless persons, and illicit drug users. Either an IGRA or the TST should be used. It does not recommend systematic testing for LTBI among persons who have diabetes, engage in harmful alcohol use, smoke tobacco, or are underweight, unless they are already included in the above recommendations.37 Further information is available from the WHO.38.

Prevention TaskForce

You are now leaving Prevention TaskForce website and going to
This website contains links to other federal and state agencies and private organizations. The United States Prevention Services Taskforce (USPSTF) cannot attest to the accuracy of information provided by these website links. Linking to other sites does not constitute an endorsement by USPSTF of the sponsors or the information and products presented on the sites. You will be subject to the destination site's privacy policy when you leave the Prevention TaskForce website.

Continue Cancel