Prevention and Control of Tuberculosis in Correctional and Detention Facilities: Recommendations from CDC

Endorsed by the Advisory Council for the Elimination of Tuberculosis, the National Commission on Correctional Health Care, and the American Correctional Association


Morbidity and Mortality Weekly Report. 2006;55(27):1-44. 

In This Article

Diagnosis and Treatment of Latent Tuberculosis Infection and Tuberculosis Disease

The principles of diagnosis and treatment of LTBI and TB disease discussed in this section are guidelines and not meant to substitute for clinical experience and judgment. Medical providers not familiar with the management of LTBI and TB disease should consult a person with expertise. All facilities' local operations procedures should include plans for consultation with and referral to persons with expertise in TB and should include criteria delineating when consultation and referral are indicated.

Although the index of suspicion for TB disease varies by individual risk factors and prevalence of TB in the population served by the correctional facility, correctional facilities typically are considered higher-risk settings (see Screening). A diagnosis of TB disease should be considered for any patient who has a persistent cough (i.e., one lasting ≥3 weeks) or other signs or symptoms compatible with TB disease (e.g., hemoptysis, night sweats, weight loss, anorexia, and fever). Diagnostic tests for TB include the TST, QFT-G, chest radiography, and laboratory examination of sputum samples or other body tissues and fluids.

Persons exposed to inmates with TB disease might become latently infected with M. tuberculosis depending on host immunity and the degree and duration of exposure. Therefore, the treatment of persons with TB disease plays a key role in TB control by stopping transmission and preventing potentially infectious cases from occurring.[92] LTBI is an asymptomatic condition that can be diagnosed by the TST or QFT-G.

A baseline screening TST result of ≥10 mm induration is considered positive for the majority of correctional facility staff and inmates, and these persons should be referred for medical and diagnostic evaluation. However, for correctional facility staff and inmates who have had a known exposure in a correctional facility (i.e., close contact with an inmate or staff member with infectious TB disease) after having a previous (baseline) TST value of 0 mm, TST results of ≥5 mm should be considered positive and interpreted as a new infection. Correctional facility staff and inmates with a screening baseline TST result of ≥1 mm, but <10 mm, who are subsequently exposed to TB disease, should be considered newly infected if they have TST values increase by ≥10 mm on retest ( Table 3 ). For example, a baseline TST result with 8 mm induration and a repeat TST result 1 year later with 18 mm induration would indicate a new infection. However, a repeat TST result with 12 mm induration would not indicate a new infection.

When decisions are made for the diagnosis and treatment of LTBI and choosing the cut-off value for a positive reaction, certain risk factors (e.g., immunocompromising conditions and known contact with a TB patient) should be assessed. Correctional facility staff and inmates who have TST indurations of 5-9 mm should be advised that their results might be an indication for treatment under certain conditions.

Interpretation of the TST might be complicated by previous vaccination with BCG, anergy, and the "boosting" effect. Detailed recommendations describing how the TST should be interpreted in relation to these possible confounders have been published.[64,93]

A correctional facility staff member or inmate who refuses testing for M. tuberculosis infection should first be educated regarding the importance of routine screening of correctional facility staff and inmates. If the person continues to refuse to have a TST, the option may be offered for the person to be tested using the QFT-G test (and vice versa). The decision to offer an alternative test depends on the reason for refusal and should be consistent with the patient's underlying wishes (e.g., offering QFT-G in place of TST is acceptable if the patient objects to having injection of a substance but agrees to having blood drawn).

®-TB Gold Test Data

Interpretation of QFT-G data is initially performed electronically; an approved interpretation method is automatically performed by the software supplied by the manufacturer ( Table 4 ).[58] A complete description of the test's interpretation is included in the product insert.

Persons who have a positive QFT-G result should be referred for a medical and diagnostic evaluation. On serial testing, a person with QFT-G results changing from negative to positive should be referred for medical and diagnostic evaluation and considered to be a QFT-G converter. Risk factors (e.g., the facility's prevalence of TB disease and personal risk factors) should be assessed when making decisions about the diagnosis and treatment of LTBI.

Persons with Suspected Pulmonary TB. Multiple types of abnormalities demonstrated on chest radiographs are strongly suggestive of pulmonary TB disease, including upper-lobe infiltration, cavitation, and pleural effusion. Infiltrates can be patchy or nodular and observed in the apical or subapical posterior upper lobes or superior segment of the lower lobes. If radiographic or clinical findings are consistent with TB disease, further studies (e.g., medical evaluation, mycobacteriologic examinations of sputa or tissue, and comparison of current and prior chest radiographs) should be performed.[65] Persons with TB pleural effusions might have concurrent unsuspected pulmonary or laryngeal TB disease.[94] These patients should be considered infectious until pulmonary and laryngeal TB disease is excluded. Patients with suspected extrapulmonary TB disease also should be suspected of having pulmonary TB until concomitant pulmonary disease is excluded.

The radiographic presentation of pulmonary TB in HIV-infected persons might be atypical. Apical cavitary disease is less common among such patients than HIV-negative patients. More common findings among HIV-infected persons are infiltrates in any lung zone, mediastinal or hilar adenopathy, or, in rare cases, a normal chest radiograph.[65,95,96,97]

Persons with LTBI. To exclude pulmonary TB disease, a chest radiograph is indicated for all persons in whom LTBI is diagnosed. If chest radiographs do not indicate pulmonary TB, and no symptoms consistent with TB disease are present, persons with positive test results for TB infection should be considered for treatment for LTBI. Persons with LTBI typically have normal chest radiographs, although they might have abnormalities suggestive of previous TB disease or other pulmonary conditions. In certain patients with TB symptoms, pulmonary infiltrates might be apparent on chest computed tomography scan or magnetic resonance imaging study but not on chest radiograph. Previous, healed TB disease typically produces radiographic findings that differ from those associated with current TB disease. These findings include nodules, fibrotic scars, calcified granulomas, and apical pleural thickening. Nevertheless, a chest radiograph by itself cannot be used to distinguish between current and healed TB. Nodules and fibrotic scars might contain slowly multiplying tubercle bacilli and pose substantial risk for progression to TB disease. Calcified nodular lesions (i.e., calcified granulomas) and apical pleural thickening indicate lower risk for progression to TB disease.[65]

Pregnant Women. Because TB disease is dangerous to both the mother and the fetus, a pregnant woman who has a positive TST or QFT-G result or who is suspected of having TB disease should receive a chest radiograph (with shielding consistent with safety guidelines) as soon as feasible. If symptoms or other high-risk conditions (e.g., HIV infection) are identified, a chest radiograph might have to be performed during the first trimester of pregnancy.[64,65,98]

Sputum examination is a key diagnostic procedure for pulmonary TB disease[93] and is indicated for the following inmates and correctional facility staff:

  • persons suspected of having pulmonary TB disease because of a chest radiograph consistent with TB disease, particularly those with any respiratory symptoms suggestive of TB disease;

  • persons with chest radiographic findings suggestive of previous, healed TB disease;

  • HIV-infected persons with any pulmonary symptoms (regardless of chest radiograph findings); or

  • persons suspected of having pulmonary TB disease for which bronchoscopy is planned (all sputum specimens should be collected and final results of staining for AFB should have been reviewed before proceeding with bronchoscopy[67]).

Persons requiring smear- and culture-sputum examination should submit at least three sputum specimens (collected 8-24 hours apart, with at least one specimen collected in the early morning).[71,99] Specimens should be collected in a sputum induction booth or in an AII room. In resource-limited settings without environmental containment, collection is safer when performed outdoors. Patients should be instructed how to produce an adequate sputum specimen, and a health-care professional should supervise and observe the collection of sputum, if possible.[93] For patients who are unable to produce an adequate sputum specimen, expectoration might be induced by inhalation of an aerosol of warm, hypertonic saline.[71]

Detection of AFB in stained smears by microscopy can provide the first mycobacteriologic indication of TB disease. A positive result for AFB in a sputum smear is predictive of increased infectiousness; however, negative AFB sputum-smear results do not exclude a diagnosis of TB disease if clinical suspicion is high. In 2002, only 63% of U.S. patients with reported positive sputum cultures had positive AFB sputum smears.[100]

Although smears allow for the detection of mycobacteria, definitive identification, strain typing, and drug-susceptibility testing of M. tuberculosis can be performed only via culture.[93] A culture of sputum or other clinical specimen that contains M. tuberculosis provides a definitive diagnosis of TB disease. In the majority of cases, identification of M. tuberculosis and drug-susceptibility results are available within 28 days using recommended rapid methods (e.g., liquid culture and DNA probes). A negative culture result is obtained in approximately 14% of patients with confirmed pulmonary TB disease.[100] Testing sputum with certain techniques (e.g., nucleic acid amplification [NAA]) facilitates the rapid detection and identification of M. tuberculosis, but should not replace culture and drug-susceptibility testing in patients with suspected TB disease.[88,101,102] Recommendations for use and interpretation of NAA tests in the diagnosis of TB disease have been published previously.[101,102]

Laboratories should report positive smear results within 24 hours of collection and positive cultures within 24 hours of the notation of the positive culture. Drug-susceptibility tests should be performed on initial isolates from all patients to assist in the identification of an effective anti-TB regimen. Drug-susceptibility tests should be repeated if 1) sputum specimens continue to be culture-positive 3 months after initiation of treatment or if 2) persons whose cultures had converted to negative subsequently revert to positive.[65,93]

Treatment for LTBI is essential to controlling and eliminating TB disease in the United States because it substantially reduces the risk that TB infection will progress to TB disease.[23] Certain persons are at high risk for developing TB disease once infected, and every effort should be made to begin these persons on a standard LTBI treatment regimen and to ensure that they complete the entire course of treatment for LTBI. Before treatment for LTBI is started, TB disease should be ruled out by history, medical examination, chest radiography, and when indicated, mycobacteriologic studies.

Correctional facility staff and inmates in the following high-risk groups should be given treatment for LTBI if their reaction to the TST is ≥5 mm, regardless of age[64,65]:

  • HIV-infected persons,

  • recent contacts of a TB patient,

  • persons with fibrotic changes on chest radiograph consistent with previous TB disease, and

  • patients with organ transplants and other immunocompromising conditions who receive the equivalent of ≥15 mg/day of prednisone for ≥1 month.

All other correctional facility staff and inmates should be considered for treatment of LTBI if their TST results are ≥10 mm induration. If QFT-G is used, any correctional facility staff member or inmate with a positive QFT-G result should be considered for LTBI treatment. Decisions regarding initiation of LTBI treatment should include consideration of the likelihood of the patient continuing and completing LTBI treatment under supervision if released from the facility before the treatment regimen is completed.

Persons with previously positive TST results who have previously completed treatment for LTBI (i.e., ≥6 months of isoniazid, 4 months of rifampin, or another regimen) do not need to be treated again unless concern exists that reinfection has occurred. Other persons who might be poor candidates for treatment of LTBI include those with a previous history of liver injury or a history of excessive alcohol consumption; active hepatitis and end-stage liver disease are relative contraindications to the use of isoniazid or pyrazinamide for treatment of LTBI.[64,103] If the decision is made to treat such patients, baseline and follow-up monitoring of serum aminotransaminases are recommended.

Standard regimens have been developed for the treatment of LTBI ( Table 5 ). The preferred treatment for LTBI is 9 months of daily isoniazid or biweekly dosing administered by DOT. Although regimens are broadly applicable, modifications should be considered for certain populations (e.g., patients with HIV infection) and when drug resistance is suspected.

Reports of severe liver injury and death associated with the combination of rifampin and pyrazinamide for treatment of LTBI prompted ATS and CDC to revise previous recommendations. These recommendations now state that this regimen typically should not be offered for the treatment of LTBI.[64,103,104,105,106,107] If the potential benefits substantially outweigh the demonstrated risk for severe liver injury and death associated with this regimen and the patient has no contraindications this regimen may be considered; a physician with experience treating LTBI and TB disease should be consulted before use of this regimen.[103] Clinicians should continue the appropriate use of rifampin and pyrazinamide in standard multidrug anti-TB regimens for the treatment of TB disease.[65]

For all LTBI treatment regimens, nonadherence to intermittent dosing results in a larger proportion of total doses missed than daily dosing; therefore, all patients on intermittent treatment should receive DOT. In addition, DOT should be used with daily dosing of LTBI treatment whenever feasible. Patients with the highest priority for DOT are those at the highest risk for progression from LTBI to TB disease, including persons with HIV infection and persons who are recent contacts of infectious patients with pulmonary TB.

Contacts of patients with drug-susceptible TB disease who once tested negative but subsequently have a positive TST result (i.e., ≥5 mm) should be evaluated for treatment of LTBI. The majority of persons who are infected will have a positive TST result within 6 weeks of exposure; therefore, contacts of patients with drug-susceptible TB disease who have initial negative TSTs should be retested 8-10 weeks after the end of exposure to a patient with suspected or confirmed TB disease.[108] Persons with TB infection should be advised that they can be re-infected with M. tuberculosis if re-exposed.[109,110,111] If they have not been treated previously, HIV-infected persons (regardless of TST result or previous LTBI treatment history), persons receiving immunosuppressive therapy (regardless of TST result or previous LTBI treatment history), and persons with a known previous (to current exposure) positive TST also should be considered for LTBI treatment.

Treatment of LTBI should not be started until a diagnosis of TB disease has been excluded. If the presence of TB disease is uncertain because of an equivocal chest radiograph, a standard multidrug anti-TB therapy might be started and adjusted as necessary, depending on the results of sputum cultures, drug-susceptibility tests, and clinical response.[65] If cultures are obtained without initiating therapy for TB disease, treatment for LTBI should not be initiated until all cultures are reported as negative, which might take 6-8 weeks.

Treatment for LTBI caused by drug-resistant M. tuberculosis organisms is complex and should be conducted in consultation with the local health department's TB control program and persons with expertise in the medical management of drug-resistant TB. Often this will require waiting for results of susceptibility testing of the isolate from the presumed source patient. Treatment should be guided by in vitro susceptibility test results from the isolate to which the patient was exposed.[65,112,113]

Routine laboratory monitoring during treatment of LTBI is indicated only for patients with abnormal baseline tests and for persons at risk for hepatic disease. Baseline laboratory testing is indicated only for persons infected with HIV, pregnant women, women in the immediate postpartum period (typically within 3 months of delivery), persons with a history of liver disease, persons who use alcohol regularly, and persons who have or who are at risk for chronic liver disease.[64]

All patients should undergo clinical monitoring at least monthly. This monitoring should include 1) a brief clinical assessment regarding the signs of hepatitis (i.e., nausea, vomiting, abdominal pain, jaundice, and yellow or brown urine) and 2) education about the adverse effects of the drug(s) and the need for prompt cessation of treatment and clinical evaluation should adverse effects occur. All aspects of the clinical encounter should be conducted in private and in the patient's primary language.

Severe adverse events associated with the administration of tuberculin antigen or treatment of LTBI or TB disease (e.g., those resulting in hospitalization or death) should be reported to MedWatch, FDA's Safety Information and Adverse Event Reporting Program at telephone 800-FDA-1088, by facsimile at 800-FDA-0178, or via the Internet by sending Report Form 3500 (available at Instructions regarding the types of adverse events that should be reported are included on MedWatch report forms. In addition, severe adverse effects associated with LTBI treatment should be reported to CDC's Division of Tuberculosis Elimination at telephone 404-639-8118.

A decision to initiate treatment (i.e., combination anti-TB chemotherapy) should be made on the basis of epidemiologic information; clinical, pathological, and radiographic findings; and the results of microscopic examination of AFB-stained sputum smears and cultures for mycobacteria. A positive AFB-smear result provides strong inferential evidence for the diagnosis of TB, and combination chemotherapy should be initiated promptly unless other strong evidence against the diagnosis of TB disease is present (e.g., a negative NAA test). If the diagnosis is confirmed by isolation of M. tuberculosis or a positive NAA test, treatment should be continued until a standard course of therapy is completed. Because as few as 50% of patients with positive sputum culture results for M. tuberculosis will have negative sputum AFB-smear results,[93] when initial AFB-smear results are negative, empiric therapy for TB is indicated if the clinical suspicion for TB disease is high. Regardless of the decision to begin anti-TB treatment, diagnoses other than TB should be considered and appropriate evaluations undertaken in patients with negative AFB-smear results. A diagnosis of culture-negative pulmonary TB can be made if sputum cultures are negative, the TST result is positive (in this circumstance, a reaction of ≥5 mm induration is considered positive), a clinical or radiographic response is observed 2 months after the initiation of therapy, and no other diagnosis has been established. An adequate regimen for culture-negative pulmonary TB includes an additional 2 months of isoniazid and rifampin to complete 4 months of treatment.[65] If no clinical or radiographic response is observed by 2 months, treatment can be stopped, and other diagnoses (including inactive TB) should be considered. If AFB-smear results are negative, and suspicion for TB disease is low, treatment can be deferred until the results of mycobacterial cultures are known and a comparison chest radiograph is available (typically at 2 months). Among persons who have not begun treatment and in whom suspicion of TB is low, treatment of LTBI should be considered if 1) cultures are negative, 2) the TST result is positive (≥5 mm induration), and 3) the chest radiograph is unchanged after 2 months. A person with TB expertise should be consulted for unusual or complex situations.

Individualized case management should be provided for all patients with TB disease.[114,115,116] In addition, patient management should be coordinated with officials of the local or state health department; suspected or confirmed TB cases should be reported to the local or state health department in accordance with laws and regulations. Regimens for treating TB disease should contain multiple drugs to which the organisms are susceptible. For persons with TB disease, treatment with a single drug can lead to the development of mycobacterial resistance to that drug. Similarly, adding a single drug to a failing anti-TB regimen is not recommended because it can lead to resistance to the added drug.[65]

For the majority of patients, the preferred regimen for treating TB disease consists of an initial 2-month phase of isoniazid, rifampin, pyrazinamide, and ethambutol, followed by a continuation phase of isoniazid and rifampin lasting ≥4 months, for a minimum total treatment period of 6 months ( Table 6 and Table 7 ). The decision to stop therapy should be made on the basis of the number of doses taken within a maximum period (not simply a 6-month period).[65] Persons with cavitary pulmonary TB disease and positive cultures of sputum specimens at the completion of 2 months of therapy should receive a longer, 7-month continuation phase of therapy (total duration: 9 months) because of the substantially higher rate of relapse among persons with this type of TB disease.[65]

If interruptions in TB therapy occur, the decision should be made whether to restart a complete course of treatment or continue the regimen as originally intended. In the majority of instances, the earlier the break in therapy and the longer its duration, the more serious the effect and the greater the need to restart the treatment from the beginning. Continuous treatment is more important in the initial phase of therapy, when the bacillary burden is highest and the chance of developing drug resistance is greatest. Although no evidence on which to base detailed recommendations exists, examples of practical algorithms for managing interruptions in therapy have been described previously.[65]

For HIV-infected persons who are receiving antiretroviral therapy, TB treatment regimens might need to be altered. Whenever possible, the care of persons with concomitant TB and HIV should be provided by or in consultation with persons with expertise in the management of both TB and HIV-related disease.[65] To prevent the emergence of rifampin resistance, persons with TB, HIV, and CD4+ T-lymphocyte cell counts <100 cells/mm3 should not be treated with highly intermittent (i.e., once- or twice-weekly) regimens. These patients should instead receive daily therapy during the intensive phase (i.e., first 2 months) and receive daily dosing or 3 doses per week by DOT during the continuation phase.[117] Antiretroviral therapy should not be withheld because the patient is being treated for TB if it is otherwise indicated. Nevertheless, beginning both antiretroviral therapy and combination chemotherapy for TB at nearly the same time is not advisable. Although data on which to base recommendations are limited, experience in the fields of HIV and TB suggests that treatment for TB should be initiated first. Delaying the initiation of antiretroviral therapy until 4-8 weeks after starting anti-TB therapy is advantageous because it 1) better enables providers to ascribe a specific cause to a drug side effect, 2) decreases the severity of paradoxical reactions, and 3) decreases adherence challenges for the patient. Until controlled studies have been conducted that evaluate the optimal time for starting antiretroviral therapy in patients with HIV infection and TB, this decision should be individualized on the basis of 1) the patient's initial response to treatment for TB, 2) the occurrence of side effects, and 3) the availability of multidrug antiretroviral therapy. Because drug-drug interactions might be less frequent with use of rifabutin, substitution of rifabutin for rifampin might be indicated with certain antiretroviral medications. Detailed information on TB treatment in HIV-infected persons has been published.[65,107] Updates are posted on the Internet as new findings become available (at,, and

Drug-susceptibility testing should be performed on all initial isolates from patients with TB disease. When results from drug-susceptibility tests become available, the treatment regimen should be adjusted accordingly[65,113,114,118,119] ( Table 6 and Table 7 ). Medical providers treating patients with drug-resistant TB disease should seek expert consultation and collaborate with the local health department for treatment decisions.[65]

The primary determinant of treatment outcome is patient adherence to the drug regimen. Thus, careful attention should be paid to measures designed to enable and foster adherence.[65,119,120] DOT is the preferred treatment strategy for all persons with TB disease and high-risk (e.g., HIV infected) persons with LTBI. DOT should be used throughout the entire course of therapy whenever feasible. Practitioners providing treatment to inmates should coordinate DOT with the local health department on an inmate's release. The local health department also may be involved in monitoring therapy for correctional facility staff.[65]

Achieving completion of treatment for LTBI or TB disease often is difficult, particularly in correctional facilities. Movement of inmates both within and outside of correctional systems interferes with continuity of care and might lead to treatment default.[121] Comprehensive case management that includes discharge planning and coordination with other correctional facilities and health departments is needed to ensure completion of therapy for patients with TB disease and LTBI.[42]

Multiple studies have demonstrated that inmates have relatively low LTBI treatment completion rates, particularly those in jails who are likely to be released before their therapy has been completed.[14,28,40,122] For a substantial proportion of inmates, referrals for follow-up after release are not made; of inmates whose appointments are scheduled, 40%-60% will not attend their first clinic visit.[36,40] Multiple interventions have been attempted to improve LTBI treatment completion in this population, including patient education while in jail, use of incentives, and use of DOT.[61,122,123] None of these strategies has had substantial success, although patient education and use of DOT have increased completion rates modestly in certain situations.[61,122] Active case management, as recommended for TB disease, should be considered as a next step in improving the completion rates for LTBI treatment.[14,42]


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.