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


Early identification and successful treatment of persons with TB disease remains the most effective means of preventing disease transmission.[47] Therefore, inmates who are likely to have infectious TB should be identified and begin treatment before they are integrated into the general correctional facility population (i.e., at the time of admission into the correctional system). When possible, newly arrived inmates should not be housed with other inmates until they have been appropriately screened for TB disease. Screening programs in the correctional setting also allow for the detection of substantial numbers of persons with LTBI who are at high risk for progressing to TB disease and would likely benefit from a course of treatment. This secondary benefit of screening programs is often limited by inability to initiate and ensure completion of LTBI treatment, particularly in short-term correctional facilities. In addition to screening at intake, routine (i.e., at least annual) screening of long-term inmates and correctional facility staff (e.g., custody and medical) should be incorporated into the TB-control program.[48,49]

How screening activities should be implemented depends on multiple factors, including 1) the type of facility, 2) the prevalence of TB infection and disease in the facility, 3) the prevalence of TB in the inmates' communities, 4) the prevalence of other risk factors for TB (e.g., HIV) in the inmate population, and 5) the average length of stay of inmates in the facility. The type of screening recommended for a particular facility is determined by an assessment of the risk for TB transmission within that facility. The risk assessment should be performed at least annually and should be made in collaboration with the local or state health department. A facility's TB risk can be defined as being minimal or nonminimal. A facility has minimal TB risk if

  • no cases of infectious TB have occurred in the facility in the last year,

  • the facility does not house substantial numbers of inmates with risk factors for TB (e.g., HIV infection and injection-drug use),

  • the facility does not house substantial numbers of new immigrants (i.e., persons arriving in the United States within the previous 5 years) from areas of the world with high rates of TB, and

  • employees of the facility are not otherwise at risk for TB.

Any facility that does not meet these criteria should be categorized as a nonminimal TB risk facility.

Symptom Screening. Whenever possible, health-care professionals should perform the initial screening. However, correctional officers in jails (particularly those housing minimal numbers of inmates) frequently administer health intake questionnaires. If custody staff members conduct the intake screening, they should receive adequate periodic training in taking a medical history, making necessary observations, and determining the appropriate disposition of inmates with signs or symptoms of possible medical problems. Staff conducting medical intake should receive appropriate counseling and education regarding medical confidentiality.

During their initial medical screening, inmates should be asked if they have a history of TB disease or if they have been treated for LTBI or TB disease previously. Documentation of any such history should be obtained from medical records, if possible. Inmates should be observed for the presence of a cough or evidence of significant weight loss. All incoming inmates in any size jail, prison, or other detention facility (e.g., immigration enforcement) should be immediately screened for symptoms of pulmonary TB by being asked if they have had a prolonged cough (i.e., one lasting ≥3 weeks), hemoptysis (i.e., bloody sputum), or chest pain. The index of suspicion should be high when pulmonary symptoms are accompanied by general, systemic symptoms of TB (e.g., fever, chills, night sweats, easy fatigability, loss of appetite, and weight loss). Inmates should be interviewed systematically (i.e., using a standardized questionnaire) to determine whether they have experienced symptoms in recent weeks. Inmates who have symptoms suggestive of TB disease should immediately receive a thorough medical evaluation, including a TST or QFT-G, a chest radiograph, and, if indicated, sputum examinations.

Persons with symptoms suggestive of TB disease or with a history of inadequate treatment for TB disease should be immediately placed in an AII room§ until they have undergone a thorough medical evaluation. If deemed infectious, such persons should remain in isolation until treatment has rendered them noninfectious. Facilities without an on-site AII room should have a written plan for referring patients with suspected or confirmed TB to a facility that is equipped to isolate, evaluate, and treat TB patients.

Symptom screening alone is an unsatisfactory screening mechanism for TB, except in facilities with a minimal risk for TB transmission. The use of symptom screening alone often will fail to detect pulmonary TB in inmates.

Chest-Radiograph Screening. Screening with chest radiographs can be an effective means of detecting new cases of unsuspected TB disease at intake to a correctional facility. In addition, radiographic screening requires fewer subsequent visits than a TST (i.e., only those inmates with suspicious radiographs or TB symptoms require follow-up). However, such screening will not identify inmates with LTBI. One study demonstrated that screening inmates with a chest radiograph doubled the TB case-finding rate and reduced the time from intake into the correctional facility to isolation substantially compared with TST testing (2.3 days and 7.5 days, respectively), thereby reducing the risk for TB exposure for other inmates and staff.[50] Digital radiographs (miniature or full-size) provide enhanced imaging and improved storage and readability. A miniature radiograph can be performed in <1 minute and exposes the patient to approximately one tenth the radiation dose of a conventional radiograph. One cost-effectiveness analysis of miniature chest radiography for TB screening on admission to jail indicated that more cases were detected with this method than either TST or symptom screening, and the cost of radiograph screening was less per case detected.[51] The extent to which radiologic screening is used in a given institution should be dictated by multiple factors, including 1) local epidemiologic characteristics of TB disease; 2) inmate length of stay; 3) the ability of the health-care professionals within the facility to conduct careful histories, tuberculin skin or QFT-G testing, and cross-matches with state TB registries; and 4) timeliness of the radiographic study and its reading. Screening with chest radiographs might be appropriate in certain jails and detention facilities that house substantial numbers of inmates for short periods and serve populations at high risk for TB (e.g., those with high prevalence of HIV infection or history of injection-drug use and foreign-born persons from countries in which TB prevalence is high).

Inmates who are infected with HIV might be anergic and consequently might have false-negative TST results. However, routine anergy panel testing is not recommended because it has not been demonstrated to assist in diagnosing or excluding LTBI.[52] In facilities that do not perform routine radiographic screening for all inmates, a chest radiograph should be part of the initial screening of HIV-infected patients and those who are at risk for HIV infection but whose status is unknown.

In facilities with on-site radiographic screening, the chest radiograph should be performed as part of intake screening and read promptly by a physician, preferably within 24 hours. Persons who have radiographs suggestive of TB should be isolated immediately and evaluated further. Sputum-smear and culture examinations should be performed for inmates whose chest radiographs are consistent with TB disease and might be indicated for at least certain persons who are symptomatic, regardless of their TST, QFT-G, or chest radiograph results because persons with HIV and TB disease might have "negative" chest radiographs in addition to false-negative TST or QFT-G results.

Mantoux TST Screening. Tuberculin skin testing using 0.1 mL of 5 tuberculin units (TU) of purified protein derivative (PPD) is the most common method of testing for TB infection. Multiple-puncture tests (e.g., the tine test) should not be used to determine whether a person is infected. Persons who have a documented history of a positive TST result (with a millimeter [mm] reading), a documented history of TB disease, or a reported history of a severe necrotic reaction to tuberculin should be exempt from a routine TST. For persons with a history of severe necrotic reactions and without a documented positive result with a millimeter reading, a QFT-G may be substituted for the TST. Otherwise, such persons should be screened for symptoms of TB and receive a chest radiograph unless they have had one recently (i.e., within 6 months) and are not symptomatic. Pregnancy, lactation, or previous vaccination with Bacillus Calmette-Guerin (BCG) vaccine are not contraindications for tuberculin skin testing. The TST is not completely sensitive for TB disease; its sensitivity ranges from 75%--90%.[53,54] Despite this limitation, skin testing, along with use of a symptom review, frequently constitutes the most practical approach to screening for TB disease.

A trained health-care professional should place the TST and interpret the reaction 48--72 hours after the injection by measuring the area of induration (i.e., the palpable swelling) at the injection site. The diameter of the indurated area should be measured across the width of the forearm. Erythema (i.e., the redness of the skin) should not be measured. All reactions, even those classified as negative, should be recorded in millimeters of induration.

In the majority of cases, a TST reaction of ≥10 mm induration is considered a positive result in inmates and correctional facility employees. However, an induration of ≥5 mm is considered a positive result in the following persons:

  • persons infected with HIV,

  • persons who are recent contacts of patients with TB disease,

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

  • organ transplant recipients and patients with other immunocompromising conditions (e.g., persons receiving≥15 mg/day of prednisone for ≥1 month), and

  • persons suspected of having TB disease.

Persons who have a positive TST result and no symptoms suggestive of TB disease should be evaluated with a chest radiograph within 72 hours after the skin test is interpreted. Persons who have symptoms suggestive of TB disease should be evaluated immediately and placed in an AII room until TB is ruled out (see Symptom Screening).

The use of two-step testing can reduce the number of positive TSTs that would otherwise be misclassified as recent skin-test conversions during future periodic screenings. Certain persons who were infected with M. tuberculosis years earlier exhibit waning delayed-type hypersensitivity to tuberculin. When they are skin tested years after infection, they might have a false-negative TST result (even though they are truly infected). However, this first skin test years after the infection might stimulate the ability to react to subsequent tests, resulting in a "booster" reaction. When the test is repeated, the reaction might be misinterpreted as a new infection (recent conversion) rather than a boosted reaction. For two-step testing, persons whose baseline TSTs yield a negative result are retested 1--3 weeks after the initial test. If the second test result is negative, they are considered not infected. If the second test result is positive, they are classified as having had previous TB infection. Two-step testing should be considered for the baseline testing of persons who report no history of a recent TST and who will receive repeated TSTs as part of an institutional periodic skin-testing program. In the majority of cases, a two-step TST is not practical in jails because of the short average length of stay of inmates.

In the past, a panel of other common antigens was often applied with the TST to obtain information regarding the competence of the patient's cellular immune system and to identify anergy. More recently, however, anergy testing has been demonstrated to be of limited usefulness because of problems with standardization and reproducibility, the low risk for TB associated with a diagnosis of anergy, and the lack of apparent benefit of preventive therapy for groups of anergic HIV-infected persons. Therefore, the use of anergy testing in conjunction with a TST is no longer recommended routinely for screening programs for M. tuberculosis infection in the United States.[52]

Intracutaneous inoculation with BCG is currently used worldwide as a vaccine against TB. BCG is a live attenuated Mycobacterium bovis strain that stimulates the immune system to protect against TB. No reliable method has been developed to distinguish TST reactions caused by vaccination with BCG from those caused by natural mycobacterial infections, although reactions of ≥20 mm of induration are not likely caused by BCG.[55] TST is not contraindicated for persons who have been vaccinated with BCG, and the TST results of such persons are used to support or exclude the diagnosis of M. tuberculosis infection. A diagnosis of M. tuberculosis infection and treatment for LTBI should be considered for any BCG-vaccinated person who has a positive TST reaction. The same criteria for interpretation of TST results are used for both BCG-vaccinated and nonvaccinated persons.[56]

QuantiFERON®-TB Gold Test. In May 2005, the U.S. Food and Drug Administration (FDA) licensed QFT-G. This in-vitro diagnostic test measures the amount of interferon-gamma produced by cells in whole blood that have been stimulated by mycobacterial peptides. The peptides used in the test mimic proteins known as ESAT-6 and CFP-10, which are present in M. tuberculosis but absent from all BCG strains and from the majority of commonly encountered non-TB mycobacteria. The test is intended for use as a diagnostic tool for M. tuberculosis infection, including both TB disease and LTBI. As with a TST, QFT-G cannot distinguish between LTBI and TB disease and should be used in conjunction with risk assessment, radiography, and other diagnostic evaluations. The advantages of QFT-G compared with TST are that 1) results can be obtained after a single patient visit, 2) the variability associated with skin-test reading can be reduced because "reading" is performed in a qualified laboratory, and 3) QFT-G is not affected by previous BCG vaccination and eliminates the unnecessary treatment of persons with false-positive results. QFT-G does not affect the result of future QFT-G tests (i.e., no "boosting" occurs). Limitations of the test include the need for phlebotomy, the need to process blood specimens within 12 hours of collection for the most recent version of the test, the limited number of laboratories that process the test, and a lack of clinical experience in interpreting test results. The elimination of the second visit for reading the TST, however, is likely to render the QFT-G competitive in cost-benefit considerations.

Although the performance of QFT-G has not been evaluated sufficiently in select populations of interest (e.g., HIV-infected persons), available data indicate that QFT-G is as sensitive as TST for detection of TB disease and more specific than TST for detection of LTBI.[57,58] CDC guidelines for QFT-G recommend that QFT-G can be used in place of TST in all circumstances in which TST is currently used.[58] This includes initial and periodic TB screening for correctional facility inmates and employees and testing of exposed persons in contact investigations. Because data are insufficient regarding performance of QFT-G in certain clinical situations, as with a negative TST result, a negative QFT-G result alone might not be sufficient to exclude M. tuberculosis infection in these situations. Examples of such clinical scenarios include those involving patients with severe immunosuppression who have had recent exposure to a patient with TB and patients being treated or about to undergo treatment with potent tumor necrosis factor alpha (TNF-a) antagonists.

Correctional facilities and local health departments should collaborate to ensure effective TB screening in the correctional setting. Inmates might provide inaccurate information on admission for multiple reasons, ranging from forgetfulness and confusion to deliberate misrepresentation. Health departments should perform cross-matches with the local TB registry and search for matches on known aliases, birth dates, maiden names, and other personal information for inmates suspected of having TB infection. A readily accessible record of previous TB history, drug-susceptibility patterns, treatment, and compliance can be useful in determining the disposition of a given patient with suspected TB.

The following procedures should be used for the initial screening of inmates and detainees (depending on their length of stay in the facility and the type of facility) and for all correctional facility employees, regardless of the type of facility.

Inmates in Minimal TB Risk Facilities. Inmates in all minimal TB risk correctional and detention facilities should be evaluated on entry for symptoms of TB. Persons with symptoms of TB should be evaluated immediately to rule out the presence of infectious disease and kept in an AII room until they are evaluated. If the facility does not have an AII room, the inmate should be transported to a facility that has one. In addition, all newly arrived inmates should be evaluated for clinical conditions and other factors that increase the risk for infection or the risk for progressing to TB disease, including the following:

  • HIV infection,

  • recent immigration,

  • history of TB,

  • recent close contact with a person with TB disease,

  • injection-drug use,

  • diabetes mellitus,

  • immunosuppressive therapy,

  • hematologic malignancy or lymphoma,

  • chronic renal failure,

  • medical conditions associated with substantial weight loss or malnutrition, or

  • history of gastrectomy or jejunoileal bypass.

Persons with any of these conditions require further screening with a TST, a QFT-G, or a chest radiograph within 7 days of arrival. Regardless of the TST or QFT-G result, inmates known to have HIV infection or other severe immunosuppression, and those who are at risk for HIV infection but whose HIV status is unknown, should have a chest radiograph taken as part of the initial screening. Persons who have an abnormal chest radiograph should be further evaluated to rule out TB disease; if TB disease is excluded as a diagnosis, LTBI therapy should be considered if the TST or QFT-G result is positive.

Inmates in Nonminimal TB Risk Prisons. Immediately on arrival, all new inmates should be screened for symptoms, and any inmate with symptoms suggestive of TB should be placed in an AII room and evaluated promptly for TB disease. If the facility does not have an AII room, the inmate should be transported to a facility that has one. Inmates who have no symptoms require further screening with a TST, a QFT-G, or a chest radiograph within 7 days of arrival. Regardless of their TST or QFT-G status, inmates known to have HIV infection or other severe immunosuppression, and those who are at risk for HIV infection but whose HIV status is unknown, should have a chest radiograph taken as part of the initial screening. Persons who have an abnormal chest radiograph should be further evaluated to rule out TB disease; if TB disease is excluded as a diagnosis, LTBI therapy should be considered if the TST or QFT-G result is positive.

As the rate of TB disease in the United States has decreased, identification and treatment of persons with LTBI who are at high risk for TB disease have become essential components of the TB elimination strategy promoted by ACET.[59] Targeted testing using the TST or QFT-G identifies persons at high risk for TB disease who would benefit from treatment for LTBI. Prisons offer an excellent public health opportunity for identifying persons at high risk for TB who can be screened for TB infection and placed on LTBI therapy, if indicated. If the TST is used, a two-step testing procedure should be strongly considered when obtaining a baseline reading. A single step QFT-G is an adequate baseline. Inmates with a positive test should be evaluated for LTBI therapy after TB disease is excluded.

Inmates in Nonminimal TB Risk Jails and Other Short-Term Detention Facilities. As in prisons, all new detainees in nonminimal TB risk jails should be screened on entry for symptoms, and any detainee who has symptoms suggestive of TB should be placed immediately in an AII room and evaluated promptly for TB disease. If the facility does not have an AII room, the inmate should be transported promptly to a facility that does have one. Detainees without symptoms require further screening with a TST, a QFT-G, or a chest radiograph within 7 days of arrival. Regardless of the TST or QFT-G result, detainees known to have HIV infection, and those who are at risk for HIV infection but whose HIV status is unknown, should have a chest radiograph taken as part of the initial screening. Persons who have a positive result should be further evaluated to rule out TB disease.

The primary purpose of screening in correctional settings is to detect TB disease. TST or QFT-G screening in jails to initiate LTBI therapy often is not practical because of the high rate of turnover and short lengths of stay. Although not all jail detainees have short lengths of stay, determining which detainees will be in the jail for a long term is difficult. Nationwide, approximately half of persons detained in local jails are released within 48 hours of admission. Thus, even if all detainees can be tested at intake, a large proportion will be unavailable to have their TSTs read or to be evaluated when QFT-G test results are available. Of those still in custody, a substantial percentage will be released before the radiographic and medical evaluation is completed. In a 1996 study, 43% of detainees at a county jail in Illinois who had a positive TST result were released or transferred before their evaluation could be completed.[3]

A substantial proportion of detainees who are incarcerated long enough to begin LTBI therapy will be released before completion of treatment. A San Francisco study indicated that approximately 62% of detainees who were started on LTBI treatment were released before completion.[40] These data illustrate the challenges of implementing a testing and treatment program for LTBI in jails with highly dynamic detainee populations. Certain jails have adopted a targeted approach of performing TSTs only on new detainees who are at high risk for TB disease (e.g., detainees with known HIV infection). Screening for TB and treating LTBI are most effective within the jail setting if resources dedicated to discharge planning and reliable access to community-based treatment are available. Modest interventions (e.g., education and incentives [see Glossary]) in the jail setting can lead to improvements in linking released detainees to postrelease medical care and increase the likelihood that therapy will be completed.[60,61]

Persons in Holding or Booking Facilities. City, county, and other law enforcement authorities frequently have facilities that hold arrestees and detainees for short periods of time, ranging from hours to multiple days. TB symptom screening is recommended for all persons at the time of entry into these facilities. Any detainee who has symptoms suggestive of TB should be immediately isolated and transferred to a facility or hospital in which the detainee can be placed in an AII room and evaluated promptly for TB disease.

Employees in All Correctional and Detention Facilities. A medical history relating to TB should be obtained from and recorded for all new employees at the time of hiring, and a physical examination for TB disease should be required. The results of the screening and examination should be kept confidential; access should be granted to public health and infection control medical professionals only when necessary. In addition, a TST or QFT-G should be mandatory for all employees who do not have a documented history of a positive result. To improve the accuracy of the baseline result, a two-step TST or a single-step QFT-G should be used for the initial screening of employees who have not been tested during the preceding 12 months. Persons who have a positive TST or QFT-G result should have a chest radiograph taken and interpreted and should be required to have a thorough medical evaluation; if TB disease is excluded as a diagnosis, such persons should be considered for LTBI therapy. All employees should be informed that they should seek appropriate follow-up and testing for TB if they are immunosuppressed for any reason (e.g., have HIV infection). Any employee who has symptoms suggestive of TB should not return to the workplace until a clinician has excluded a diagnosis of infectious TB disease.

Other Persons Who Might Need to be Screened. Certain persons who are neither inmates nor employees but who visit high-risk facilities on a regular basis also should be considered for screening. These persons might include contractors (e.g., food handlers and service workers), volunteers, and those providing religious ministries. Screening of these persons should follow the same procedures as those outlined for employees.

Long-term inmates and all employees who have a negative TST or QFT-G result should have follow-up testing at least annually. Persons who have a history of a positive test result should be screened for symptoms of TB disease. Annual chest radiographs are unnecessary for the follow-up evaluation of infected persons. Test results should be recorded in medical records and in a retrievable aggregate database of all TST or QFT-G results. Personal identifying information should be kept confidential.

Correctional facilities can use multiple strategies to ensure annual screening of long-term inmates for newly acquired TB infection. Certain institutions schedule annual screening on the inmate's date of birth or on the anniversary of the inmate's most recent test. Other institutions and systems suspend inmate movement and screen the entire population on the same day every year. Methods of screening a subset of the inmate population (e.g., on a monthly basis) are beneficial because they provide an ongoing assessment of M. tuberculosis transmission within the facility.

Results from TST or QFT-G testing should be analyzed periodically to estimate the risk for acquiring new infection in a correctional facility; however, this analysis should be completed by using only the test results of facility employees and inmates who have remained in the facility continually during the interval between testing. The conversion rate equals the number of employees or inmates whose test results have converted from negative to positive (i.e., the numerator) during a specific interval divided by the total number of previously negative employees or inmates who were tested during the same interval (i.e., the denominator). In certain facilities, conducting an analysis of test results for specific areas or groups within the facility might be appropriate.

More frequent screening is needed when a conversion rate is substantially higher than previous rates or when other evidence of ongoing transmission is detected. A cluster (i.e., either two or more patients with TB disease that are linked by epidemiologic or genotyping data or two or more TST or QFT-G conversions occurring in the correctional facility among inmates who are epidemiologically linked) or other evidence of person-to-person transmission also warrants additional epidemiologic investigation and possibly a revision of the facility's TB prevention and control protocol.

Facilities in which the risk for infection with M. tuberculosis is minimal might not need to maintain a periodic screening program. However, requiring baseline TST or QFT-G testing of employees would enable medical staff to distinguish between a TST or QFT-G conversion and a positive TST or QFT-G result caused by a previous exposure to M. tuberculosis. A decision to discontinue periodic employee screening should be made in consultation with the local or state health department.

HIV counseling, testing, and referral (CTR) should be routinely recommended for all persons in settings in which the population is at increased behavioral or clinical risk for acquiring or transmitting HIV infection, regardless of setting prevalence.[62] Because correctional facilities are considered settings in which the population is at increased risk for acquiring or transmitting HIV, routine HIV CTR is recommended for inmates. Furthermore, HIV infection is the greatest risk factor for progression from LTBI to TB disease.[63,64] Therefore, HIV CTR should be routinely offered to all inmates and correctional facility staff with LTBI or TB disease if their HIV infection status is unknown at the time of their LTBI or TB disease diagnosis.[64,65] Correctional facilities should be particularly aware of the need for preventing transmission of M. tuberculosis in settings in which persons infected with HIV might be housed or might work.[66]

Correctional and detention facilities are strongly encouraged to collect and analyze data on the effectiveness of their TB screening policies and procedures. Working in conjunction with their state or local TB-control program, correctional and detention facilities should refine their screening policies and procedures as indicated by such data. In the absence of local data that justify revision, correctional and detention facilities should adhere to the screening recommendations detailed above.

§ Formerly called a negative pressure isolation room, an AII room is a single-occupancy patient-care room used to isolate persons with suspected or confirmed infectious TB disease. Environmental factors are controlled in AII rooms to minimize the transmission of infectious agents that are usually spread from person to person by droplet nuclei associated with coughing or aerosolization of contaminated fluids. AII rooms should provide negative pressure in the room so clean air flows under the door gap into the room, an air flow rate of 6-12 air changes per hour (ACH), and direct exhaust of air from the room to the outside of the building or recirculation of air through a high efficiency particulate air (HEPA) filter.


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.