Advancing Toward Elimination
Once the control of tuberculosis is ensured through the treatment of active cases of tuberculosis, the next step in tuberculosis control is to focus on the prevention of new cases. The relatively long interval between infection with Mycobacterium tuberculosis and the onset of disease creates an opportunity to intervene and prevent disease by treating those with latent infection. Also, as the risk of infection declines, more cases will be the result of reactivation of infection acquired many years earlier. Increasing efforts at prevention can accelerate the decline in the number of cases of tuberculosis, maintain momentum and interest in tuberculosis control programs, and act as preparation for a final push to elimination. This chapter provides an overview of the principles of prevention through treatment of latent infection and proposes strategies for use of this approach with defined populations including foreign-born individuals, especially recent arrivals, from countries with high rates of tuberculosis, inmates of correctional facilities, infected contacts of individuals with infectious tuberculosis, and other high-risk groups that are defined locally. The committee recognizes that many of the strategies described in this chapter will require significant changes and increases in resources at all levels of tuberculosis control but is convinced of the importance of accelerating efforts for the prevention of tuberculosis.
Recommendation 4.1 To limit the spread of tuberculosis from infec
Page 87tious patients to their contacts, the committee recommends that more effective methodologies for the identification of persons with recently acquired tuberculosis infection, especially persons exposed to patients with new cases of tuberculosis, be developed and efforts be increased to evaluate appropriately and treat latent infection in all persons who meet the criteria for treatment for such infections.
Recommendation 4.2 To prevent the development of tuberculosis among individuals with latent tuberculosis infection, the committee recommends that
Tuberculin skin testing be required as part of the medical evaluation for immigrant visa applicants from countries with high rates of tuberculosis, a Class B4 immigration waiver designation be created for persons with normal chest radiographs and positive tuberculin skin tests, and all tuberculin-positive Class B immigrants be required to undergo an evaluation for tuberculosis and, when indicated, complete an approved course of treatment for latent infection before receiving a permanent residency card (“green card”). Implementation should be in a stepwise fashion and pilot programs should evaluate and assess costs.
Tuberculin testing be required of all inmates of correctional facilities and completion of an approved course of treatment, when indicated, be required, with referral to the appropriate public health agency for all inmates released before completion of treatment.
Programs of targeted tuberculin skin testing and treatment of latent infection be increased for high-incidence groups, such as HIV-infected individuals, undocumented immigrants, homeless individuals, and intravenous drug abusers, as determined by local epidemiological circumstances.
BACKGROUND AND INTRODUCTION
The recent success of tuberculosis control efforts indicates that, given adequate resources, the traditional control program that focuses on the treatment of active tuberculosis is once again functioning effectively and that a steady decline in cases should be expected. However, to make significant progress toward the elimination of tuberculosis in the United States, efforts to prevent cases from occurring must be amplified. Without an effective vaccine for the prevention of pulmonary tuberculosis, the most effective means of preventing new cases is to take advantage of the relatively long period between infection and the development of active
disease by treating individuals with latent infection to reduce the risk of disease. The Advisory Council for the Elimination of Tuberculosis (ACET) included expanded targeted testing and treatment of latent infection among the highest-priority activities for tuberculosis control programs (CDC, 1999). Recently published guidelines from the Centers for Disease Control and Prevention (CDC) and the American Thoracic Society will provide a framework for increased attention to targeted screening and will recommend new regimens for the treatment of tuberculosis infection (American Thoracic Society and CDC, 2000). As recommended by the guidelines and by ACET, local epidemiological conditions will guide the testing effort, but increased national attention will be needed in three areas. First, as noted in Chapter 2 , tuberculosis among foreign-born individuals accounts for an increasing proportion of U.S. tuberculosis cases, and slightly more than half of these cases occur within the first 5 years of the arrival of foreign-born individuals in the United States. Targeted tuberculin skin testing of newly arrived immigrants before arrival in the United States with adequate treatment for those with latent infection once they arrive could prevent a significant number of cases. Second, an outbreak of tuberculosis associated with a prison led to the recognition of the resurgence of tuberculosis in the United States, and screening for infection and treatment of latent infection should be mandatory in these settings. Finally, by challenging traditional concepts about contact investigations and expanding the definition of close contacts to include social contacts and others previously thought to be at lower risk of infection, many more recently infected individuals who are at an increased risk of developing tuberculosis will be identified. This chapter provides a brief background on the rationale for the treatment of latent infection and addresses these priority areas for the prevention of tuberculosis.
TREATMENT OF LATENT TUBERCULOSIS INFECTION
Shortly after the discovery of isoniazid in 1952, the potential of this highly bactericidal drug to prevent tuberculosis was explored. A series of trials summarized in a review by Ferebee (1970) showed that the effectiveness of isoniazid for the prevention of tuberculosis ranged from 92 percent in an outbreak among Dutch sailors, in which adherence to the regimen was strictly enforced (Veening, 1968), to 26 percent in a Tunisian community-based study in which the rate of adherence to the isoniazid regimen was estimated to be only 25 percent or less. A study of 3-, 6-, and 12-month regimens showed that a 12-month regimen was the superior regimen (International Union Against Tuberculosis, 1982) and Comstock (1999) recently presented an analysis that showed that a regimen of at least 9 months likely gives maximum protection.
Isoniazid can cause hypersensitivity reactions, paresthesia, and other adverse effects, but concerns about isoniazid hepatotoxicity have limited the use of isoniazid in the past for the treatment of latent infection. However, this controversy concerns the provision of isoniazid to individuals at relatively low risk of tuberculosis. In contrast, the use of isoniazid has never been challenged for individuals at high risk of tuberculosis, including recent tuberculin skin test converters, contacts of infectious patients, immigrants from countries with high rates of tuberculosis, and individuals with certain medical conditions, especially human immunodeficiency virus (HIV) infection. A recent 7-year study of patients receiving isoniazid for treatment of latent infection showed a rate of clinical hepatitis of only 0.1 percent (11 cases among 11,141 patients), with no deaths and only one hospitalization in a program with patient monitoring, the administration of only limited amounts of medication at each visit, and good education on the symptoms of adverse effects (Nolan et al., 1999).
In addition to isoniazid, rifampin-based regimens have recently been recommended as treatment for latent infection with M. tuberculosis (American Thoracic Society, 2000). The current recommended regimens for treatment of latent infection include isoniazid for 9 months (with an option for 6 months), rifampin and pyrazinamide for 2 months, and rifampin alone for 4 months (ATS/CDC statement). Although the rifampin-pyrazinamide regimen has the advantage of being much shorter, the experience with this regimen is relatively limited, and one pilot study that evaluated this regimen showed that it has a very high rate of toxicity. In particular, this regimen has not been evaluated with immigrant populations, and programs that use the regimen are encouraged to maintain surveillance for adverse reactions to this regimen.
CURRENT SCREENING FOR IMMIGRANTS TO THE UNITED STATES
As noted in Chapter 1 , increasingly large proportion of tuberculosis cases in the United States occur among foreign-born individuals: 7,591 (41.3 percent) cases in 1998. If the number of U.S.-born individuals with tuberculosis continues to decrease at a rate of 9 percent, as it did between 1992 and 1998, and the number of foreign-born individuals with tuberculosis remains the same, the majority of tuberculosis cases will occur among foreign-born individuals from 2002 onward. The CDC Working Group on Tuberculosis Among the Foreign-Born thus concluded that the elimination of tuberculosis in the United States will increasingly depend on the elimination of tuberculosis among foreign-born individuals.
Foreign-born individuals comprise those who arrive in the United States through a number of mechanisms. The majority who come to the
United States legally are in the United States temporarily. These arrivals include tourists, students and their families, and workers and their families. Immigrants include those who intend to reside permanently in the United States, some who have newly arrived from their country of origin, and others who adjust their status to permanent residency while they are already in the United States.
Tuberculosis is one condition for which immigrants are screened before they enter the United States. This screening currently consists of a chest radiograph for all immigrants 15 years of age and older and the examination of three sputum samples for acid-fast bacilli (AFB) for all individuals with chest radiograph findings consistent with tuberculosis. Immigrants younger than 15 years of age who are suspected of having tuberculosis or of having contact with an individual with a known case of tuberculosis are given a tuberculin skin test, and those with any indication of tuberculosis are given a chest radiograph.
Individuals with any sputum tests positive for AFB are classified as infectious and as having a Class A medical condition. This means that either they must complete an approved course of tuberculosis treatment before entering the United States or that they must start therapy and have three consecutive negative sputum tests before coming to the United States. Individuals who do not complete therapy must have a relative arrange for treatment at the intended destination in the United States, with the concurrence of the local or state tuberculosis control program. They are to be given 30 days of medication when they depart for the United States.
Individuals who have abnormal chest radiograph and sputum examinations but who are negative for AFB are classified as having a Class B medical condition: Class B1 if the radiograph indicates active disease but the individual is not infectious due to negative sputum smears or if the individual has extrapulmonary tuberculosis (such individuals are to be started on therapy and given 30 days of medication before coming to the United States); Class B2 if the radiograph indicates tuberculosis that is not clinically active; and Class B3 if the radiograph indicates old or healed tuberculosis. Individuals with Class B1 or B2 tuberculosis are told to report to the public health department at their final destination.
Any Class A or B medical classification is recorded on immigration documents (DQ [Division of Quarantine] form 75.17). If the immigrant enters one of the eight ports (Honolulu, Seattle, San Francisco, Los Angeles, Chicago, Miami, Atlanta, and New York) where U.S. Public Health Service Division of Quarantine staff are assigned, the quarantine officer gives the immigrant a form that reminds the individual of the requirement to report to the local public health department upon arrival at the
final destination. The quarantine officer sends DQ form 75.17 to the CDC Division of Quarantine, which then forwards the forms to state and local health departments. If the immigrant arrives in a port without a quarantine officer, the immigration officer is to collect the form (DQ form 75.17), notify the immigrant of the requirement to report to the local public health department upon arrival at the final destination, and then send DQ form 75.17 to the nearest port with a quarantine officer. There has been no evaluation of the rate of referral of immigration forms from ports without a quarantine officer, but it is believed that many of the forms are not collected and that when they are collected they are often held for long periods of time and sent to a port with a quarantine office in batches.
When the immigration forms are collected and sent to health departments, the follow-up rate of screening of immigrants for tuberculosis can be quite high. It has been estimated to exceed 82 percent in most areas of the United States (Binkin et al., 1996), and one tuberculosis control program has documented a follow-up program that achieved a 97.5 percent follow-up rate (Catlos et al., 1998). However, it costs the program $13–$14 immigrant to follow-up each new arrival and to encourage the individual to report for examination, and the program accounts only for those immigrants about whom the health department was notified by the Division of Quarantine.
THE NEED FOR EXPANDED SCREENING
More than a quarter (29.6 percent) of all cases of tuberculosis among foreign-born individuals occur during the first year after their arrival in the United States, and more than half (55.5 percent) occur during their first 5 years in the United States. The risk for developing tuberculosis declines the longer an immigrant from a country with a high risk of infection is in the United States, because they are no longer at a high risk of infection or reinfection. While the risk of developing tuberculosis for a recently arrived immigrant from a country with a high risk of infection is lower than that of a newly infected person, the risk for the typical immigrant from a high-risk country is many times that of the typical U.S. resident. The current screening procedure for immigrants does not and cannot detect or prevent many of the cases of tuberculosis that occur among foreign-born individuals in the United States. Recent immigrants from countries with high rates of tuberculosis have been identified as prime candidates for screening and treatment of latent infection (American Thoracic Society, 2000), but public health efforts to confront the problem have been inadequate (Binkin et al., 1996). The Working Group on Tuberculosis Among the Foreign-Born concluded,
Efforts to provide screening and preventive therapy for the foreign-born are limited. Averting future cases of TB [tuberculosis] requires linking screening programs to prevention services. However, few resources are available to health departments for prevention efforts to foreign-born persons. Also, persons who do not consider themselves ill and who are from countries where TB is reported as a stigma might be reluctant to begin or complete therapy (Centers for Disease Control and Prevention, 1998:7)
In the current system, the burden is on the health departments to encourage individuals to be screened for tuberculosis infection after they arrive in the United States.
Given the significance of tuberculosis among foreign-born individuals and given the contribution that addressing the incidence of tuberculosis among foreign-born individuals would make to the goal of tuberculosis elimination, the committee believes that the federal government should fund a coordinated national effort to screen foreign-born individuals for tuberculosis. Reliance on the efforts and resources of local health departments unfairly burdens those communities within which a disproportionate number of foreign-born individuals reside, fails to reflect the national interest in radically reducing the rate of tuberculosis among immigrants and refugees, and virtually ensures that the level of commitment will be inadequate for the task.
The strategies that must be developed are linked to the impediments to identifying those with latent tuberculosis infection and ensuring their completion of treatment for latent infection, the sheer size of the populations to be screened, the difficulties in gaining access to those who should be screened, and the cultural and linguistic barriers that make voluntary approaches to screening and treatment of limited utility. A significant departure from current methods of screening and treatment of latent tuberculosis infection among foreign-born individuals is dictated by the clear public health benefits that could be attained.
Strategies to Ensure Case Identification and Treatment Completion
A first step would be to expand the current screening of those applying to immigrate and the screening of refugees. Tuberculin skin testing for latent infection should be added to the current panel of tests used to screen for active tuberculosis (chest radiographs and sputum analysis), and a Class B4 (or other designation) should be created for individuals with a normal chest radiograph but a positive tuberculin skin test. Such testing, like other medical examinations, would occur before the immigrant or refugee came to the United States. Tuberculin skin testing already occurs in the United States for those seeking to adjust their status
after coming to the United States on temporary visas. Given the small number of tuberculosis infections detected by tuberculin skin test screening among low-risk populations and the recent recommendations of the American Thoracic Society and the CDC that screening for latent infection be targeted to those at risk, there is little justification for screening all applicants for immigration. Targeting of screening to those who come from countries with the greatest tuberculosis burdens (prevalence of infection estimated to be equal to or greater than the global median of 36 percent) would detect the greatest number of infections and would require the testing of approximately 200,000 immigrants per year. Individuals born in countries with the greatest burdens of tuberculosis account for 55 percent of the cases of tuberculosis among the foreign-born individuals each year. In particular, even though Mexico is estimated to have only a 17 percent prevalence of tuberculosis infection, Mexican-born individuals account for nearly one-quarter of the foreign-born individuals with tuberculosis each year and just under 10 percent of the total number of individuals with tuberculosis in the United States. Requiring immigrants from Mexico to be skin tested before immigration would add about 53,000 immigrants to the testing program each year, increasing the total number of immigrants to be skin tested each year to about 250,000. Among each annual cohort of immigrants from countries with the greatest tuberculosis burdens, approximately 2,100 cases of tuberculosis can be expected to occur over the course of a cohort's first 5 years in the United States. Assuming a steady influx of immigrants, this rate of tuberculosis among each cohort translates into 2,100 cases per year among the population of immigrants who arrived in the past 5 years. As noted in Appendix E , data on the rate of tuberculosis among immigrants have a number of limitations, and estimates obtained by using more recent population and immigration estimates from the 2000 census would be valuable. Identifying and Treating Latent Infection
The methodology for estimating the number of cases that could be expected to occur among foreign-born individuals, the number of cases that could be prevented by treatment of latent infection, and the costs of a program for the treatment of individuals with latent infection are contained in Appendix E . Assuming an overall rate of effectiveness of treatment for latent infection of 75 percent (allowing for nonadherence), at least 1,300 cases of tuberculosis could be prevented per year during the first 5 years that the immigrants are in the United States. (The number of cases prevented in later years, when the immigrants are at a lower risk of developing tuberculosis, was not estimated.) Overall program costs would be about $23 million per year, and the cost per case of tuberculosis pre
vented would be about $14,559, which is less than the cost of treating a patient with tuberculosis and monitoring the contacts of that patient, which is $16,391. If the results of the tuberculin skin testing before immigration were accepted and the test was not repeated, the program costs would drop to about $12,305 per case of tuberculosis prevented.
Additional savings will be realized if all individuals with a Class B tuberculosis designation were required to report for tuberculosis screening in the United States. As noted above, tuberculosis control programs are very successful in getting immigrants with a Class B tuberculosis designation to report for examination. However, although such screening is cost-effective, it is cosily, and screening of foreign-born individuals with Class B tuberculosis would account for about 12 percent of the tuberculin skin testing costs described above. Shifting the burden to report for screening to the arriving immigrant would reduce the need for intensive follow-up, and the cost per case of tuberculosis prevented could be reduced to $12,920 if all immigrants reported without follow-up.
One method of requiring examination for tuberculosis and tuberculosis infection would be to withhold the “green card” for all immigrants with a Class A or Class B tuberculosis designation and to issue this permanent residency document only when these individuals have completed screening by the health department and treatment (if indicated). This approach may shift costs from the public health programs to the Immigration and Naturalization Service (INS), but the immigrants would have a strong incentive to comply. However, the service may not be as costly if modern information technology is applied. At a minimum, the failure of an immigrant with a Class B tuberculosis designation to report to the health department for screening could be made grounds for deportation, as is already the case for immigrants with a Class A designation and a waiver to enter the United States. Issues and Concerns
The committee carefully considered the issues surrounding tuberculin skin testing and treatment of latent infection for newly arrived immigrants. An extra public session was added to the committee's schedule to allow a roundtable discussion of the ethical, legal, and practical issues of implementing these programs. After these discussions the committee concluded that the difficulties in implementing the program were largely structural but that these could be overcome with a sufficient commitment of resources. Although these costs will be significant, many of the benefits will accrue to the general programs of targeted tuberculin skin testing and the treatment of individuals with latent infection. Significant additional funding will be required for (a) CDC to provide training and moni
tor the quality of the tuberculin testing programs, (b) CDC and INS to ensure that information on the immigrants is collected accurately and is promptly communicated to the relevant health departments in the United States, and (c) the state and local health departments that will have to provide the final medical examination and that will be responsible for completion of therapy, when treatment is indicated, including start-up costs for training and infrastructure. However, improvements are needed in the current system of testing for and treatment of latent infections, and this new program could pave the way for a broad array of needed changes. Because of the magnitude of the changes required by this recommendation, the committee suggests that this be implemented in a stepwise approach and that pilot programs be developed to evaluate strategies and assess their costs.
The adoption of a mandatory tuberculin skin test screening program may trouble those concerned with the human rights of immigrants and refugees, foreign workers, and students. In a national climate characterized by occasional hostility to immigrants and assumptions about the burdens posed by foreign-born individuals, it is not difficult to understand such concerns. Nevertheless, the committee can see no fundamental reason why the logic of current tuberculosis screening should not be extended to those with latent infection. Although the threat to the public health from latent infection is not as immediate as the threat that stems from active disease, the overall epidemiological burden is significant. It is that potential burden that justifies such screening. Treatment of Latent Infections
No program of mandatory screening for latent infection could be justified unless it were linked to a program of treatment of latent infection. Recent analyses have made clear that with isoniazid monotherapy, the recommended course of therapy should be isoniazid therapy daily or twice weekly for 9 months (American Thoracic Society, 2000). A two-drug regimen of rifampin and pyrazinamide daily for as little as 2 months has also been shown to be effective at reducing the risk of disease by the same amount as the 9-month isoniazid regimen, but as noted earlier, toxicity may be an issue for the rifampin-pyrazinamide regimen. For both the long-term monotherapy and the short-course, 2-month regimens, strict adherence to the treatment regimen is crucial.
A positive tuberculin skin test should not impede the entrance of immigrants, refugees, workers and students to the United States. Should the treatment of latent infection, however, be voluntary or mandatory? Reluctance to mandate treatment of latent infection is rooted in concerns about the risks associated with such treatment as well as in constitutional
and ethical principles. As noted above, early experience with isoniazid treatment raised questions about its hepatotoxicity. However, it has always been clear that the benefits of isoniazid treatment outweigh the risks of toxicity for those at high risk of tuberculosis. Recent experience has made clear that careful clinical monitoring and the exclusion of those at very high risk of hepatotoxicity can radically reduce the risks associated with treatment of latent infection. Nevertheless, the risks are not absent. These lingering risks, however, along with concerns for individual rights, no longer justify the basic claim for relying solely on voluntary treatment programs. Payment for Treatment and Community Support
Although the committee acknowledges the importance of protecting and respecting the rights of immigrants and refugees, it believes that the vast public health benefit that would follow from universal treatment of those with latent infection, as recommended by the American Thoracic Society, justifies the imposition of therapy. The federal government, in recognition of the fact that such therapy is primarily designed to achieve a public health goal of national importance, should pay for the treatment of immigrants and refugees. Although the payment for treatment should be from public funds, the provision of treatment should be undertaken in a range of settings. Given the cultural and linguistic barriers that confront individuals who have newly arrived in the United States, much might well be attained by engaging the cooperation of community-based organizations that provide services to distinct ethnic immigrant communities. Certainly such organizations will have much to offer local tuberculosis control programs that may be called upon to provide treatment of newly arrived immigrants for latent infections. Nonpermanent Residents
Screening of immigrants for latent tuberculosis infection and treatment (if necessary) will not affect students, workers, and their families who come to the United States for long-term stays but do not seek permanent residence. In 1996, 280,000 new arrivals from countries with a high incidence of tuberculosis were students, workers, and their families, and an additional 2,000 cases of tuberculosis per year may occur among these individuals during their first 5 years in the United States. The estimated number of these cases is smaller than that for permanent visa applicants because of differences in the incidence of tuberculosis in their countries of origin. This also is likely to be an overestimate of the number of cases as many of these people would not be in the United States for the full 5
years, and the age distribution of those with immigrant visas (which included the very young and the very old) and those with student and work visas is very different. Health screening is not required for students and workers coming to the United States, and starting a screening program would require a more detailed analysis of the tuberculosis risk for this population. However, school- and employment-based screening and treatment programs should be considered for these new arrivals.
Focusing on those who will be coming to the United States in the future will not address the problem of the estimated 7 million foreign-born individuals with tuberculosis infection already in the United States legally. For them, many of whom are already citizens and the vast proportion of whom may not know of their latent infections, and for all others not screened on entry, programs of culturally sensitive and linguistically appropriate aggressive outreach will be necessary. Programs of mandatory screening would be difficult to launch, as would efforts at imposing mandatory treatment. A close interaction with the private sector, neighborhood health centers, and community-based organizations are going to be necessary if these programs are going to be practical and effective. Skin testing and treatment for latent infection are a part of the Guide to Clinical Preventive Services (DHHS, 1996) and individuals from high-risk populations, covered by Medicare will commonly be treated in the private sector. Immigrants from high-risk populations, especially immigrants, are likely to obtain care from neighborhood health centers and community-based organizations and an example of a successful collaboration with the health department in Seattle/King County is noted above. It is the process of immigration that provides a unique opportunity for screening and treatment. It is a process that provides a singular opportunity to advance the goal of tuberculosis elimination.
MANDATORY SCREENING AND TREATMENT OF LATENT INFECTION IN PRISONS AND OTHER CONGREGATE SETTINGS
If much could be achieved in terms of the goal of tuberculosis elimination through a program of mandatory screening and treatment of individuals newly arrived in the United States, considerable benefit would also follow from such screening efforts in congregate settings, where the rates of tuberculosis are historically high: homeless shelters and prisons (see the box Tuberculosis in a Custodial Setting). Prison authorities have an affirmative constitutional duty to protect inmates from infectious conditions (Greifinger et al., 1993). That duty extends not only to the provision of care but also to screening. In 1981, a federal appeals court held that the failure to screen new inmates for communicable conditions violated
the Eighth Amendment's prohibition against cruel and unusual punishment since it would expose prisoners to a preventable threat to their health (Lareau v. Manson, 651 F. 2d 96, 109 [2nd Cir 1981]). In its decision the court wrote that the “resulting threat to the well being of inmates is so serious . . . that this practice constitutes ‘punishment' in violation of the Due Process Clause.” A decade later, in the midst of growing alarm about the resurgence of tuberculosis and well-publicized concerns about multidrug-resistant tuberculosis outbreaks in institutional settings, a federal district court in Pennsylvania mandated that state authorities initiate a tuberculosis control program that included mandatory tuberculin skin test screening (Austin v. Pennsylvania Department of Corrections, WL 277511 [E.D. Pa. 1992]).
Immigrant Outreach and Treatment of Latent Infection
Please refer to the page image for an unflawed representation of this content.
Foreign-born individuals are at the highest risk of latent tuberculosis in their first 5 years in the United States. This provides an opportunity for treatment of latent tuberculosis infection prevention of disease. In King County, Washington, for example, approximately two-thirds of cases of tuberculosis occur among foreign-born individuals. With the assistance of the Annie E. Casey Foundation and the Firland Foundation, the Office of Refugee and Immigrant Assistance of the State of Washington, the Seattle/King County Health Department, and the Community Housecalls and International Medicine Clinic at Harborview Medical Center have engaged in a pilot program of immigrant outreach to increase the effectiveness of tuberculosis screening and prevention services among new refugees and immigrants.
A key component of this program has been the employment of bilingual-bicultural community members who are in good standing in their own community and who are trained to address tuberculosis-related public health and infection control questions. The outreach workers are able to address compliance concerns by providing information about tuberculosis and its transmission and treatment and by assisting with the logistics of medicine pick-up and clinic visits. They provide newly arrived immigrants a cultural as well as a linguistic link to the health care system. The outreach workers also provide newly arrived families information about immigration issues, social welfare regulations, the school system, and access to general health care.
Another key feature of the program has been the use of individual interviews and focus groups with community members on a community-by-community basis. During these interviews and group discussions the symptoms of tuberculosis, the social significance of the disease, the implications of therapy, the concerns about chronic illness in that community, and information about taking medication in the absence of illness are discussed in culture-specific terms that could affect tuberculosis control. After characterizing areas of potential misunderstanding and conflict, tuberculosis health education materials are developed to help the outreach workers (and other health care providers) negotiate these problem areas. (Materials are available via the Internet at http://healthlinks.washington.edu/clinical/ethnomed .)
The program has been extremely successful. At the time of the latest evaluation 89 immigrants from Albania, Bosnia, Ethiopia, Russia, Somalia, Ukraine, Vietnam, and other countries had been started on treatment for latent infection. Five individuals moved, 2 stopped medication on their own initiative, 26 are still on medication, and 53 have successfully completed a 6-month course of therapy. That is an 88 percent (53/60) completion rate for those who could have finished therapy and a 96 percent (53/55) completion rate for those who were eligible to complete therapy and who remained in the area. This success far exceeds the 60– 70 percent completion rates obtained by routine programs and has been obtained with a population in whom completion of treatment for latent infection is considered to be particularly difficult.
A formal evaluation of the program will be conducted, including an assessment of its costs, but sensitivity to cultural and linguistic issues must be central to its success. Another important aspect of this pilot project has been the demonstration of a successful collaboration between public-sector, academic, and charitable, nongovernmental organizations in its development and implementation. This model of the delivery of treatment for latent infection and of success in the eradication of tuberculosis that is achievable with foreign-born individuals is worthy of attention.
Tuberculosis in a Custodial Setting
Please refer to the page image for an unflawed representation of this content.
Custodial settings such as prisons can provide two important functions to society toward tuberculosis elimination. Careful monitoring of tuberculosis in these settings provides an early warning of what is occurring in the general public. This was so with the occurrence of multidrug-resistant (MDR) tuberculosis, which was first identified in New York State prisons. In addition, it is much easier to treat tuberculosis while those identified as having infection or disease are institutionalized than when they are again free, as most of them will be.
There are several reasons why the risk of tuberculosis is higher in these settings than in the general public. Those who are incarcerated or institutionalized tend to have more risk factors associated with tuberculosis, factors such as lower socioeconomic status, minority status, abuse of alcohol or other substances, and previously inadequate health care. In addition, institutions place these people who are at higher risk in close contact, often in older, poorly ventilated buildings, facilitating transmission of M. tuberculosis when cases of disease occur.
The New York State Department of Correctional Services took the situation in 1991 as a challenge and has shown significant progress. Approximately 25 percent of incoming inmates were infected with M. tuberculosis; that rate did not change during the subsequent decade. What did change was the development and spread of active disease. Tuberculosis disease incidence was 225 cases per 100,000 population in 1991. In 1998 it was 40. (This rate includes all individuals with tuberculosis in the system, including those transferred in [15/100,000] and those diagnosed while they were in the system [25/100,000]. The latter rate is similar to the 21/100,000 rate for New York City, from which three-quarters of inmates in the New York State prison system come). Although there were 39 MDR cases of tuberculosis in 1991, there was only 1 in 1998. The rate of new infections within the prison system has been decreased from 1.7 percent in 1993 to 0.25 percent in 1998 for staff and from 2.4 percent to 1.2 percent for inmates. Departmental commitment included staff dedicated to dealing with infectious diseases, mandatory education about tuberculosis and other infectious diseases for all staff and inmates, mandatory skin testing of every one in the system (inmates and staff) mandatory evaluation of all individuals with suspect cases, liberal use of respiratory isolation including all individuals awaiting sputum testing results, directly observed therapy with all antituberculosis medications, and mandatory treatment of latent infection. During the decade more than 750,000 skin tests were given.
Since the New York State corrections system releases nearly 30,000 people back into the general population every year, surveillance, treatment of latent infection, and treatment while in the system provide significant benefits to society. Continuous monitoring will continue to detect significant shifts in the general population.
It is thus clear that mandatory screening for latent tuberculosis infection would not only be permissible in prisons but could be required as a matter of law. Such a screening program, however, could be justified only if follow-up therapy for latent tuberculosis infection was available. Could such treatment be mandatory, or might the ethical principle of autonomy and the constitutional right to privacy preclude such an approach, permitting only the offer of treatment? The answer has not been clear.
There is no question that prison authorities would have the authority to impose treatment for communicable conditions. However, can they impose prophylactic interventions? In at least one federal court decision, it was held that prison authorities could require inmates to undergo immunization against diphtheria-tetanus (Zaire v. Dalsheim, 698 F. Supp, 57 [S.D.N.Y., 1988]). Using an exacting standard of review, the court held that an inmate's privacy interest did not preclude compulsory vaccination because there was “a compelling state interest in preventing the spread of deadly disease among a closely quartered population.” Nevertheless, writing in 1993, the director of the prison health system in New York State rejected mandatory treatment of latent tuberculosis infection (Greifinger et al., 1993). It would be “virtually impossible” to compel the taking of oral medication. It would be hard to justify quarantining those who required such treatment because latent infection posed no immediate threat to the health of others. Additionally, there were the clinical risks associated with isoniazid therapy. “Since the risk of developing tuberculosis is low among immunocompetent young adults, the risk of coercive treatment of latent infection appears to outweigh the benefit.” Just 3 years later, a successor New York state prison system health director adopted a policy of compulsory treatment for latent tuberculosis infection. His analysis came to a distinctly different conclusion about the risks and benefits involved.
He determined that screening for latent infection without mandated treatment of latent infection for those who are infected inadequately protected the health of the community: the prison community and the wider community. This was consistent with the opinion of the federal court (Jolly v. Coughlin, 76 F3d 468 [2nd Cir 1996]). To overcome the logistical problems of mandatory skin testing and treatment for latent infection, while at the same time protecting the health of the community and the
constitutional rights of the inmates, a new status, “tuberculin hold,” was established. Those who decline or who are unable to meet the mandatory requirements are restricted to their cells with careful monitoring for clinical or X-ray signs of disease for 1 year. The length of this status was determined on the basis of the greater risk of development of disease in the first year after infection, with the year based on the worst-case scenario that infection occurred at the time of refusal of the mandated testing and treatment of latent infection. (To meet constitutional requirements, inmates on “tuberculin hold” are provided with 1 hour of outdoor recreation time per day.) Confining inmates to their cells limits the potential spread of tuberculosis if disease should develop, but it is less restrictive than respiratory isolation, which is required in the case of inmates with disease.
The current tuberculosis control policy of the New York State Department of Correctional Services has been reviewed by a federal court, where it has been found not to violate inmates' constitutional rights (Word v. Wright, 98-CV-220 [U.S. District Court for the Western District of New York]).
Given the state of prison health care systems and the potential for abuse of prisoners, any proposal for mandatory treatment must be subject to careful review from constitutional and ethical perspectives. Given the long-term strategy of tuberculosis elimination, the relative safety of carefully monitored treatment of latent infection, the risks to prisoners—both HIV-infected prisoner and other prisoners—of exposure to tuberculosis in overcrowded settings, and policy initiatives that extend the treatment of tuberculosis infection to nonincarcerated populations, the committee concludes that a program of mandatory prophylactic treatment in prisons will serve the interests of both inmate populations and the public health more broadly.
EXPANDING CONTACT INVESTIGATIONS AND IMPROVING OUTBREAK MANAGEMENT
CDC estimates that an average investigation of each case of tuberculosis in the United States results in the identification of approximately nine close contacts. On average, 30 percent of contacts are infected and another 1 percent will have already progressed to active disease. Approximately 5–10 percent of those with latent infection (depending on the length of time since infection) will progress from latent infection to active disease, and half of these cases will occur during the first 2 years after exposure. The prevalence of tuberculosis among close contacts is approximately 700 per 100,000 population (nearly 100-fold higher than that in the general population) (Binkin et al., 1999; Moodie and Riley, 1974). There
fore, the examination of contacts or persons exposed to an individual with tuberculosis is one of the most important methods of case finding for either tuberculosis disease or latent infection.
The transmission of tuberculosis to contacts has been documented in many diverse locations such as institutions (CDC, 1987), doctor's offices (Askew et al., 1997), airplanes (Kenyon et al., 1996), crack houses (CDC, 1991), HIV respite facilities (Daley et al., 1992), drug rehabilitation centers (CDC, 1980), navy ships (DiStasio and Trump, 1990), and renal transplant units (Jereb et al., 1993). The utility and importance of contact investigations in these settings and for high-risk groups such as foreign-born individuals (Wells et al., 1997) and children under 15 years of age (Casanova et al., 1991; Fernandez et al., 1994; Goldman et al., 1994; Mehta and Bentley, 1992; Rubilar et al., 1995; Topley et al., 1996) and follow-up of patients with multidrug-resistant tuberculosis cases (CDC, 1987; Snider et al., 1985) have also been demonstrated. In addition, it has been shown that the rate of adherence to treatment for latent tuberculosis infection may be highest among contacts (Menzies et al., 1993).
In the United States, health departments currently perform 90 percent of the contact tracing investigations (Binkin et al., 1999). These investigations, as noted above, are the second highest priority (after the early identification and treatment of individuals with active tuberculosis) for tuberculosis control programs. As cases of tuberculosis have retreated into defined pockets of the population (e.g., geographic and risk behavior groups), it has become necessary to modify traditional contact tracing epidemiology to address the specific needs of the individuals in these groups. The contact tracing investigation may now include other types of domiciles such as homeless shelters, correctional facilities, nursing homes, and HIV hospices.
As all cases of tuberculosis began by contact with an individual with tuberculosis, use of a strategy of the early identification and evaluation of contacts and completion of treatment for latent tuberculosis infection would clearly move the country toward tuberculosis elimination. Unfortunately, even in a country as rich with resources as the United States, contact investigations are often less than successful. The literature offers numerous examples of both highly productive and less than adequate contact investigations (Allos et al., 1996; Barnes et al., 1991; Hedemark, 1996; Holcombe, 1996; Hussain et al., 1992; Peerbooms et al., 1995; Sasaki et al., 1995). In addition, several recent outbreaks of tuberculosis have highlighted the need for enhanced efforts to prevent transmission (Allos et al., 1996; Bock et al., 1998; Kenyon et al., 1997; Mangura et al., 1998; Nivin et al., 1998; Valway et al., 1998; Washko et al., 1998).
The goals of a contact investigation are to identify exposed individu
als who may have infection or disease and ensure that they are screened and monitored as appropriate, identify the source of tuberculosis disease transmission (this is particularly relevant for children with active tuberculosis when recent transmission is likely), and identify a tuberculosis outbreak when more newly infected persons or more tuberculosis cases are discovered during the investigation than were anticipated on the basis of previous epidemiological data. In this situation, contact tracing may then lead to expanded outbreak investigation activities.
The problems associated with contact investigations that must be addressed to achieve improvements in the effectiveness of these investigations are grouped into eight specific areas, which are identified and described in the following sections.
Public Health Infrastructure
As stated in an article in Pennsylvania Medicine,
Contact investigation is very likely the least appreciated of the activities of any tuberculosis program or clinic. It requires highly educated tuberculosis staff who have experience to deal with the different situations that occur as a result of a case of pulmonary tuberculosis coming to detection. . . . In the year described, thousands of hours of staff time were required to perform the contact work. It is unrealistic to conceive that such work could be performed by any other group than a dedicated and knowledgeable tuberculosis staff, without losing quality of performance and without incurring much greater cost and illness. (Rubin and Lynch, 1996)
A declining public health infrastructure is one problem that may affect either the ability of a tuberculosis control program to respond in a timely manner or the capacity to achieve a response (Barnes et al., 1991; Marks et al., 1999). Anticipated decreases in funding and the attendant downsizing, staff turnover, and inexperience will affect a tuberculosis control program's ability to respond effectively and to prioritize its activities. In addition, as the numbers of cases of tuberculosis decline, it is a reasonable expectation that the responsibilities of public health staff may be shifted elsewhere. When tuberculosis control resources are scarce (as in low-incidence areas), decisions about who will respond (a public health nurse, epidemiologist, or outreach staff) and who will supervise the effort must be made. Inadequate resources may lead to the assignment of untrained staff to many contact investigation tasks. As is demonstrated in the case study in the box Tale of Two Counties, these same limitations can profoundly affect an area's ability to stop the transmission of tuberculosis in an outbreak situation.
Tale of Two Counties
Please refer to the page image for an unflawed representation of this content.
In 1994, an outbreak of tuberculosis occurred in two rural counties in two adjacent states that had each previously reported less than one case of tuberculosis per year. The extent of contact investigation follow-up varied considerably between both counties, as did the results (Onorato, 1999).
Please refer to the page image for an unflawed representation of this content.
The differences between the two counties included the fact that County B used creative approaches, had culturally sensitive staff, used on-site directly observed therapy or directly observed preventive therapy, maintained better confidentiality, and provided education for local health care providers. In addition, the links between the health department and the community were stronger in County B.
This experience demonstrated the need for a long-term approach and long-term support in low-incidence areas. Contact investigation and outbreak management can be extremely complex, and increasing numbers of public health jurisdictions will be unprepared as tuberculosis becomes rarer.
Attitudinal beliefs and social or cultural differences between the health department staff and the patient or the patient's contacts can affect the staff's ability to conduct an effective investigation. The overriding influences of socioeconomic factors (e.g., homelessness, intravenous drug use, and cultural and linguistic barriers) require new strategies. Culturally sensitive and diverse staff are increasingly needed at all levels of investigation and monitoring. Similarly, staff who are skilled in identifying risk factors for HIV infection during the interview process are needed so that a contact's risk of developing tuberculosis can be properly assessed.
It is also important to clearly determine the responsibilities of the field investigator. Jurisdictions vary in the person assigned to carry out field investigation tasks. Some require assessment by a nurse, whereas others allow outreach staff to administer skin tests and collect sputum specimens from symptomatic patients. Nursing practice issues have been raised, as have liability concerns. On the other hand, resource allocation may be scarce, and the ability to conduct a skin test onsite has distinct advantages.
The failure to recognize an outbreak and to expand an investigation when one is needed can create serious problems (Barnes et al., 1991). Inadequate contact investigations can result in continued transmission of tuberculosis. Missed epidemiological links can have profound consequences as evidenced in a recent outbreak among participants in a floating card game in the rural south (Bock, 1998). (See the box Tuberculosis Outbreak.)
Difficulties can also arise that relate to the size of the setting in which the contact investigation must be conducted (e.g., in homeless shelters, multiple work sites, or airplanes and trains). Public health officials face many challenges in contact investigations that require massive screenings. These include providing sufficient follow-up to personalize the po
tential risk and the need for followthrough with public health recommendations and ensuring that skin tests are consistently performed by trained and experienced personnel, particularly when skin test administration and reading of results must be delegated to numerous local health departments and private health care providers. In some settings such as schools, institutions, or work sites, the desire on the part of the employees or administration to overtest individuals to ameliorate the hysteria often associated with knowledge of a case of tuberculosis may influence the direction of the investigation. In addition, pressure exerted by various organizations or groups to protect their members through case isolation or universal testing may be exerted.
Tuberculosis Outbreak in a Floating Card Game in the Rural South: Is the Deck Marked Against Tuberculosis Control?
Please refer to the page image for an unflawed representation of this content.
When nine tuberculosis cases occurred in a rural southern county in an 18-month period (more than had occurred in the previous 10 years combined, with the result being an annual tuberculosis case rate of 86 per 100,000 people), investigators knew they were stumped. An investigation was conducted 19 months after the index patient's case was diagnosed and his acknowledged contacts were evaluated and treated for tuberculosis infection. The investigation revealed that he should also have identified at his initial interview about contacts his two card-playing associates, an extramarital liaison, and another social contact. All four of these individuals presented with tuberculosis disease 7 to 18.5 months later; one of the individuals had HIV coinfection and died. Secondary transmission with additional cases of tuberculosis also occurred. DNA fingerprinting of the available isolates confirmed that the index patient had transmitted the disease to the additional four contacts. Treatment of latent infection could have prevented the progression of their infections to active disease.
State tuberculosis control consultants assisted regional and county staff with evaluating and controlling the outbreak. Three reasons for the inadequate evaluation of case contacts appeared likely. First, local public health workers with limited previous tuberculosis control experience failed to recognize the extent of the exposure and to extend the investigation when a high proportion of contacts were found to be infected. Second, the index patients failed to disclose illegal and illicit social contacts. Third, ethnic minorities were reluctant to participate in a contact investigation conducted by individuals from outside their community.
The outbreak demonstrates the current challenges for tuberculosis control programs in the United States. As the rates of tuberculosis decline, programs will be challenged to maintain expertise among staff who manage few cases. Skilled interviewers are needed to overcome the challenges of ensuring confidentiality when illegal and illicit social connections, such as gambling and drug-using partners, are involved. Another challenge will be to overcome barriers to minority ethnic communities' participation in the health care system.
Skin testing problems because of false-positive results due to Mycobacterium bovis BCG continue to pose obstacles to the contact investigation process. Many foreign-born children were vaccinated with BCG at birth, increasing the possibility of a false-positive test. Also, contacts with a history of BCG vaccination often fail to believe that they are at risk and therefore may not comply with recommendations for testing and treatment. In addition, false-negative tests are possible due to malnourishment, treatment with immunosuppressive drugs, and concomitant disease, such as a childhood infection or HIV infection. The need for the contact to be seen 48–72 hours after testing to have the skin test read can also be problematic for both the contact and the investigator. The need for repeat testing for those contacts who were initially skin test negative 10– 12 weeks postexposure (the “window period”) adds additional barriers to the completion of the medical evaluation. Location of contacts 3 months later can be difficult, and often, contacts do not see the need for retesting. For health departments that are already stretched for resources, retesting may assume less of a priority.
Finally, the failure to ensure completion of treatment for latent infection will result in a failed contact investigation process, no matter how successful the previous steps were. Data on the completion of treatment for latent infection vary, but data from a recent CDC study suggest that in the states and big cities that it surveyed, the completion-of-therapy rate was 57 percent, but only 44 percent of those who were eligible for therapy completed it. For one-third of those who started therapy, completion-of-therapy status was noted as either “refused/uncooperative” or “unknown.” The impact of improving completion of treatment for latent infection can be seen by considering that there were roughly 14,000 people with pulmonary tuberculosis, and at least half of these people would have been smear positive and therefore highly infectious. If each of these cases had 3 infected contacts, there would be 21,000 infected contacts; without any treatment, a conservative estimate would be that 7 percent (the standard estimate of a 10 percent risk minus 3 percent of individuals
who may have already developed disease by the time of the investigation) or 1,470 of these people would develop tuberculosis disease. Assuming a 90 percent efficacy of therapy among those who complete treatment, 582 cases of tuberculosis would be prevented if 44 percent of these people completed therapy, while 1,190 cases would be prevented if 90 percent of them completed therapy. Public Health Resources
Jurisdictions with a high or medium prevalence of tuberculosis would likely find it most efficient to place the responsibility for contact investigations within their health departments. The suggested activities to be maintained by these programs are summarized in Table 4-1 .
Jurisdictions with a low prevalence of tuberculosis may not be able to justify use of the public health resources required to maintain all of the suggested activities of a contact investigation listed in Table 4-1 , and a secondary support system should be developed. One option would be the establishment of regional response teams. It may be difficult for state tuberculosis program staff to provide on-site expertise to other states, but models of regional and state collaborations for the provision of tuberculosis services currently exist, including long-term hospitalization, education, consultation, and laboratory support. Another approach to regional or state collaboration would be the establishment of a federal “swat” team that would provide an immediate response to an outbreak. Such a federal swat team might overcome some of the problems of the provision of services by staff from one state in another state. Disease investigators from other programs, other appropriately trained staff with alternative duties, or staff with multiple duties could perform the investigations if proper guidance were available. Supervision and evaluation could be provided by properly trained and experienced public health nurses or other designated clinical staff. Regional centers of expertise could be used for consultation and technical assistance.
In states where primary public health responsibility is in counties or similar jurisdictions, vertical responsibility systems could be established. State tuberculosis control programs would be responsible for planning for incremental increases in skills, with local public health staff being responsible for the initiation of case management. Working within a network of more experienced supervisors and consultants, local public health workers could be directed by regional supervisors, who would report to state-level specialists in medicine, nursing, epidemiology, and behavioral and social aspects of tuberculosis case management.
Whether it is a jurisdiction with a high prevalence or a low prevalence of tuberculosis, these responsibilities will require adequate funding (state
and federal). The commitment to support a comprehensive program to ensure that contacts are identified, provided access to adequate and appropriate care, and monitored until the completion of therapy is a key in the advance toward the elimination of tuberculosis.
Please refer to the page image for an unflawed representation of this content.
Several problems related to management of contact follow-up by private providers that evaluate and treat patients with tuberculosis and their contacts arise. One such problem is the delay in reporting of the case to the health department once it is diagnosed (from a variety of private-
sector settings, e.g., hospitals, correctional facilities, and health centers), which may allow continued transmission. Often, private providers fail to recommend treatment for latent tuberculosis infection for contacts at high risk. In the study by Marks et al. (1999), only 50 percent of those close contacts who were at high risk for disease were placed on therapy. Problems related to ensuring public health follow-up activities for patients and contacts monitored in managed care settings are also encountered with private providers.
Although private providers are expected to report cases or suspected cases, the health departments have responsibility for ensuring that they are reported in a timely fashion. Strategies such as the use of 24-hour reporting lines, reporting fact sheets, the use of easy-to-remember or identify reporting telephone numbers, and educational efforts about reporting responsibilities and the role of the health department targeted to physicians in all settings conducted on a routine basis should promote reporting of cases or suspected cases of tuberculosis. Evaluation activities related to reporting, such as validation studies, accuracy of information, review of cases by reporting status, and time to notification could be a routine health department activity for the identification of physicians and institutions that do not report cases of tuberculosis or tuberculosis infection in a timely manner and that should accordingly be contacted.
Health departments can work collaboratively with providers in the private sector to ensure that contacts are placed on treatment and complete their treatment for latent tuberculosis infection. If a contact is identified through the health department tracking system as one who is being monitored by a private-sector provider, educational messages need to be delivered to the providers about the risk for the patient (on the basis of the transmission assessment) and the importance of the patient beginning and completing treatment for latent tuberculosis infection. Health departments should also be tracking contacts seen in the private sector for completion of treatment status and should provide assistance as needed, including directly observed therapy for treatment of latent infection.
As tuberculosis retreats into defined populations in limited geographical areas, physicians will see a tuberculosis case less often in their daily practices, and will be less likely to consider the diagnosis in their initial workup. The CDC revised edition of the Tuberculosis Core Curriculum has a small section on contact investigations and will be a useful reference when it is published. However, additional resources are needed in a format that is accessible to all physicians. As described in the discussion of education and training in Chapter 3 , medical school curricula should contain basic information on the diagnosis of tuberculosis, but in some areas, medical expertise in tuberculosis will need to be maintained
at a different level through state health departments, hot lines, centers of excellence, and so on.
Lack of Consistent National Policies for Contact Investigations
The only document that suggests a national strategy for contact investigation is the American Thoracic Society-American Lung Association Control Statement written in 1992 (ATS/ALA, 1992). This is outdated in that it does not address the situations, described earlier in the report, that challenge conventional wisdom about tuberculosis transmission. These include transmission in a social situation (the floating card game) and the higher than expected proportion of cases in which transmission occurred from smear-negative individuals, as discussed in Chapter 2 . As a result, various tuberculosis control programs use different methods for different aspects of contact investigations.
For example, once a case or a suspected case of tuberculosis is reported, decisions must be made about those cases that require contact tracing. This is often problematic, as in some areas of the country contact investigations are restricted to contacts of smear-positive individuals with pulmonary or laryngeal tuberculosis, whereas in other jurisdictions a broader definition may apply. In addition, as children are rarely infectious, in some areas a contact investigation may be replaced by a source case investigation to identify the person who transmitted disease to the child. Even with new national guidelines about the reporting of contacts (which emphasize smear status), there is still controversy about using smear status as a marker of potential transmission (Barnes, 1998; Behr et al., 1999; Corless et al., 1999; Iseman, 1997; Liippo et al., 1993; Menzies, 1997a,b; Rodriguez et al., 1996). Routine screening of all contacts of patients smear positive for AFB as part of the contact investigation process may be all inclusive but may also waste considerable resources on patients who do not have tuberculosis as their final diagnosis (environmental mycobacteria are also AFB) (Corless et al., 1999). To complicate matters further, transmission from smear-negative patients has also been noted (Behr et al., 1999).
It is also often difficult to define the period of infectiousness. When the contact is unable to remember reliably when his or her symptoms began, some jurisdictions elect to define the period of infectiousness as beginning at least 3 months before treatment started. For other jurisdictions it may be from the time of diagnosis. As there are no guidelines in this regard, there is no consistency in programmatic approach. Other problems include the lack of consistent, standardized definitions for all aspects of the contact investigation process, including definitions of “contact,” “close contact,” “other-than-close contact,” and so on. Documenta
tion problems are common throughout the investigation process (Barnes et al., 1991), and as no standardized methodology for data upkeep, analysis, or evaluation exists or is required, programs approach these tasks differently. Additionally, no national, consensus guidelines for tuberculosis outbreak investigations exist such as response plans or procedures. Tuberculosis programs were required to have an “outbreak response plan” as part of their Year 2000 Cooperative Agreement grant applications for funding from the CDC, but no guidance was given on a standardized approach.
The newly established CDC Contact Working Group could serve as a vehicle to facilitate the development of much needed standardized recommendations and implementation strategies, such as definitions, common elements, and formats for data collection (including factors that place an individual at high risk for tuberculosis and at high risk of a lack of adherence to treatment), methods for assessment of all contacts (tuberculosis skin test positive or negative) for factors that place them at high risk for tuberculosis and prioritization for follow-up on the basis of these factors, methods for ensuring counseling for the prevention of HIV infection and testing for all individuals at risk for HIV infection, methods for determination of the date of last exposure, and evaluation standards. Collaboration with the National Tuberculosis Controllers Association could assist with this process. In addition, current state or big city contact investigation guidelines (where they exist) and other related materials (algorithms, fact sheets, etc.) can be reviewed as potential models or examples.
As part of its global efforts, CDC could work with the International Union Against Tuberculosis and Lung Diseases Nursing Section's International Working Group on Contact Investigations. This group has developed a list of international referrals to be used to monitor contacts who leave the state or country.
Lack of Knowledge About Need for and Importance of Contact Investigations
Contact investigation education has three components: provider education (including public health), patient and contact education, and the patient-provider relationship. Provider Education
A major difficulty for providers is priority setting. Those who lack training and experience often do not understand the need for a systematic approach to the investigation. Providers may also lack knowledge as to who is at risk of progression to disease, and resources may be spent on
delivering services to individuals who are not at demonstrated risk. The lack of a standardized approach to tuberculin skin testing and reading of the results may also result in misinterpretations that may confound the process. Finally, there are failures to recognize an outbreak and to expand an investigation when needed (Barnes et al., 1991). Patient and Contact Education
A delay in diagnosis may result when a patient with tuberculosis fails to seek treatment. Asch and colleagues (1998) have reported that 30 percent of symptomatic patients with tuberculosis did not obtain medical attention for more than 30 days after the onset of symptoms. This delay adds to the potential for transmission to contacts. The reasons for this delay may include patient or contact mistrust of the government or health care system or related immigration issues. Cultural beliefs that cause fear of stigmatization because of tuberculosis, tuberculosis and HIV coinfection, or other factors may also interfere with the contact investigation process. Patients may be reluctant to divulge contact information if they are worried about the perceptions of others. The lifestyles of many tuberculosis patients result in competing priorities. Drug use or the need for food and shelter for homeless people may take priority over medical care. Although some patients readily provide the names of their contacts, misunderstandings about how tuberculosis is transmitted may lead to misunderstandings of what a true contact is and the patient may inadvertently miss some critical information (Shrestha-Kuwahara et al., 1999). Patients are also often reluctant to identify risk factors for HIV infection. Finally, among foreign-born individuals, there are misconceptions about the receipt and presumed protectiveness of the BCG vaccine. Many of these factors result in the failure of contacts to respond to the screening efforts (Brassard and Lamarre, 1999; Hussain et al., 1992). For example, in a church choir outbreak (Mangura et al., 1998), few persons (less than one-third) who were exposed participated in the initial skin testing and one-third of those who did participate did not return for readings of the skin test results. Similar findings have been found in other settings, such as a pediatric outpatient clinic, where, despite extensive resource dedication, only 43 percent of exposed children completed screening (Moore et al., 1998). Patient-Provider Relationship
The patient-provider relationship is often complicated by different attitudinal beliefs and social or cultural differences. Either the interviewer, the interviewee, or both may be uncomfortable with the HIV risk assess
ment questions. Also, as mentioned previously, the patient's perceptions of contact may be different from the provider's epidemiological perceptions of contact with tuberculosis (Booysen et al., 1999).
To overcome all of these obstacles, providers need skills in patient assessment, interviewing, counseling, communication, skin testing, and reading and evaluation. Models exist for some of these areas, such as training courses in contact investigation and interviewing skills offered through CDC-funded Model Tuberculosis Centers. The New Jersey Model Tuberculosis Center, for example, offers improved contact investigation interviewing techniques through innovative skills-based training. If offered on a regional basis (with adequate supportive funding), this type of training would allow more participation and interaction by providers with patients. These courses should also promote the development of skills in basic environmental assessment techniques for use in field investigations. CDC, in conjunction with the Model Tuberculosis Centers, recently offered via the Internet a self-study module on contact investigation that was reportedly quite successful. The CDC also plans to release this material on videotape and as an interactive computer-based course.
Patient and provider educational materials are needed to facilitate the contact investigation process. Educational materials for patients need to be culturally and linguistically appropriate, as well as at the correct level of literacy for the targeted population, and they need to be evaluated to assess their effectiveness. Committee members were not aware of efforts by either CDC or the Model Tuberculosis Centers to develop these materials and encourage their development, along with the other patient educational materials discussed in Chapter 3 . Before the development of the educational materials, appropriate theories and models of behavioral change need to be evaluated and used as appropriate to deliver effective, targeted messages, materials, and programs for health care providers and patients.
The patient-provider relationship would be enhanced through cultural awareness training for providers. Some tuberculosis programs have already established such training programs, and these programs or models from other disease control programs should be available to all providers. The patient-provider relationship will also be enhanced by training in skills in counseling for the prevention of HIV infection and testing for HIV infection, which should be widely available to all providers.
The lack of knowledge as to what types of data are needed can significantly affect the quality and performance of the contact investigation. This problem was noted in a study conducted in several state and big city
tuberculosis control programs (Marks et al., 1999). Their results indicate that there were great variations as to who was collecting data from the sites surveyed, as well as what information was being obtained. The study noted that less information was recorded for individuals who were tuberculin skin test negative or who did not begin treatment for latent infection, no contacts were noted for many patients with tuberculosis, no notation that identified persons at high risk for disease (especially HIV) was found, and risk factors for nonadherence to treatment were rarely recorded. In a presentation to the committee it was noted that 2 of 11 programs surveyed did not record the date of last exposure to establish time frames for investigation, and no record of follow-up skin tests was noted for 43 percent those who were initially negative (Onorato, 1999).
The lack of general prevalence data on which to base a decision to expand (or not expand) an investigation is also problematic. Although the current expansion of the National Health and Nutritional Examination Survey (NHANES) to include tuberculosis will collect some general prevalence data, given the type of sampling used for that survey, it is questionable whether the results from that survey can be extrapolated for use in contact investigations.
Regarding data accessibility, studies have shown that the inability to obtain information can profoundly affect the results of the investigation (Hussain et al., 1992; Nivin et al., 1998). For example, in a tuberculosis outbreak in a hospital nursery (Nivin et al., 1998) contact investigation difficulties included the fact that inadequate data on tuberculin skin testing of hospital employees were available and many records could not be located at all.
The CDC Working Group on Contact Investigations or a similar group should develop guidelines for data collection: what types of data are needed, where data should be maintained, how many data should be collected, and so on. Guidelines should outline and specify the types of data required for all aspects of the contact investigation process, followup, analysis, and evaluation.
There is also a need for community-specific tuberculosis prevalence data, if that is to be the measure used for comparison in the expansion of investigations. If this is not feasible, then alternative means of decision making need to be developed and disseminated. Health departments should work with health care providers to ensure that contact testing data are accessible when they are needed. Laws that allow public health agency access to data for investigations of infectious diseases should be enforced, and provider collaboration with public health agencies should be enhanced.
The current contact investigation process does not appear to adequately incorporate host susceptibility into the priority-setting process. For example, contacts who are infected with HIV may belong in the first group of contacts evaluated, regardless of the extent of contact. Another factor associated with the determination of host susceptibility relates to the adequacy of HIV risk assessment techniques used during the contact investigation process (particularly for those who are skin test negative) (Barnes et al., 1991; Barnes et al., 1996). If these techniques are not adequate, persons at the highest risk of disease may be missed.
Recent data also suggest that the concentric circle method (i.e., examining groups at lower risk of exposure only if the higher risk group had a high rate of infection) may not completely address contact investigation needs (Bock et al., 1998; Cegielski et al., 1997; Fitzpatrick et al., 1999; Mangura et al., 1998; Rothenberg, 1996). Some of these investigators have suggested that this traditional approach be modified or changed to reflect the influence of social networking. Two recent reports highlight this need. In an investigation of a church choir, all cases of tuberculosis were reported separately by each township in quite diverse communities (Mangura et al., 1998). Traditional contact investigation methods had identified no workplace or family contacts. In the investigation of participants in a floating card game described earlier in this chapter, the failure to identify the existence of numerous social networks propagated the outbreak (Bock et al., 1998). These examples demonstrate the importance of congregate activities outside of work and the importance of assumed socially defined high-risk groups.
The role of DNA fingerprinting in contact investigations needs to be further defined. Three recent reports suggest the importance of molecular epidemiology in understanding the application of the current contact investigation framework. In the first of these, the conventional means of contact investigation failed to identify epidemiological links, in which typing by restriction fragment length polymorphism (RFLP) analysis led to the detection of a community outbreak among HIV-infected persons (Tabet et al., 1994). The common place of transmission was a local bar. A second example was an investigation in Los Angeles in which RFLP analysis showed that the locations at which the homeless population congregate were important sites of tuberculosis transmission for both homeless and nonhomeless people (Barnes et al., 1997). Finally, a study in Baltimore by Bishai and colleagues (1998) suggests that measures to reduce tuberculosis transmission should be location specific instead of based on the concentric circle method. RFLP analysis in the context of contact investigations also raises the issue of reopening such investigations on the
basis of new results by RFLP analysis. The costs and benefits must still be assessed to evaluate this strategy.
If research demonstrates that new frameworks for contact investigations such as social networking or location-based screening are effective, health department staff and other providers will require training in their applicability and use. Education on tuberculosis and contact tracing strategies will need to be adapted to the characteristics of distinct population groups. Expansion of screening programs might be applicable for groups at increased risk of transmission, particularly if it seems that traditional contact investigations are not feasible.
Safety and Confidentiality
Safety and confidentiality issues reflect the changing world of tuberculosis and the associated program implications for investigators. Safety for the investigator is an issue, given the high-crime-rate areas where contact investigations are frequently conducted. Decisions need to be made about security, and plans to ensure safety should be developed as needed (Bock et al., 1998). In addition, confidentiality can be difficult to maintain in some settings, particularly during workplace- and institution-based outbreaks.
The lack of a scientific basis for some aspects of the contact investigation process affects the potential effectiveness and efficiency of the investigation and restricts the investigator's ability to stop further transmission of disease. This is particularly true in the determination of transmission factors. As noted earlier, it is still not possible to define precisely the limits of the contact investigation. Although mathematical models for analysis of potential transmission of tuberculosis are available, the factors that affect both the transmission and the acquisition of disease are variable and difficult to calculate (Nardell et al., 1991). For example, there are continued difficulties in defining the extent of “infectiousness.”
Although the transmission risk assessment provides some direction, it lacks scientific validation, and some of the variables are open to question (e.g., the relationship between transmission and smear status [as noted earlier], patients with “normal” chest X rays, and the clinical presentation in children) (Barnes, 1998; Behr et al., 1999; Corless et al., 1999; Iseman, 1997; Kenyon et al., 1997; Menzies, 1997a; Pena et al., 1999). Differences in the virulence of the infecting organism also affect transmission. The ability to define such virulence would help the investigator to better prioritize contact efforts by being able to focus on those who are
defined as potential “superspreaders” (Valway et al., 1998; van Soolingen et al., 1996; Zhang et al., 1992). The reverse is also true for children, who are not usually considered infectious. A better understanding of transmission factors as they relate to children would assist the investigator in decision making in such difficult settings as schools or day-care centers. A better definition of the protection afforded by the BCG vaccine and its relevance to exposure would also help define those at risk of transmission (Menzies, 1996). There are also continued difficulties in defining environmental factors such as direction of airflow, volume of ventilation, the presence of ultraviolet light, crowding, and volume of air space. Finally, the recent recommendations of ACET suggest that operational research is needed to (a) ensure prompt and complete identification of contacts, (b) increase the numbers of appropriate contacts who are identified, (c) increase the proportion of infected contacts who are placed on treatment for latent infection, and (d) and increase the proportion who complete therapy (CDC, 1999).
If tuberculosis elimination efforts are to be successful, prevention activities must be targeted to the groups at highest risk for progression from tuberculosis infection to disease. Contacts of patients with infectious cases of tuberculosis are such a high-risk group, and the tracing of contacts should be priorities for tuberculosis control programs. Although the focus of contact tracing is prevention, other potential benefits of the investigation may include the identification of additional cases of tuberculosis and the opportunity for education about tuberculosis disease, the risk of transmission, the tuberculosis-HIV connection, and so on. As every case of tuberculosis began as a contact, the ability to rapidly identify tuberculosis cases and to effectively conduct the subsequent contact tracing is one of the cornerstones of tuberculosis control efforts by public health agencies. Without this capacity, transmission of tuberculosis will persist, the decline in the numbers of cases of tuberculosis will stop, and tuberculosis elimination will be impossible to achieve.
Allos BM , Gensheimer KF , Bloch AB , Parrotte D , Horan JM , Lewis V , and Schaffner W. 1996 . Management of an outbreak of tuberculosis in a small community . Ann Intern Med 125 : 114–117 .
American Thoracic Society/American Lung Association . 1992 . Control of tuberculosis in the United States . Am Rev Respir Dis 146 : 1623–1633 .
American Thoracic Society and Centers for Disease Control and Prevention . 2000 . Targeted tuberculosis testing and treatment of latent infection . Am J Respir Crit Care Med 161 : 5221–5247 .
Asch S , Leake B , Anderson R , and Gelberg L. 1998 Why do symptomatic patients delay obtaining care for tuberculosis? Am J Respir Crit Care Med 157 : 1244–1248
Askew GL , Finelli L , Hutton M , et al. 1997 . Mycobacterium tuberculosis transmission from a pediatrician to patients . Pediatrics 100 : 19–23 .
Barnes PF. 1998 . Reducing the ongoing transmission of tuberculosis (Editorial) . JAMA 280(19) : 1702–1703 .
Barnes PF , Bloch AB , Davidson PT , and Snider DE. 1991 . Tuberculosis in patients with human immunodeficiency virus infection . N Engl J Med 324 : 1644–1650 .
Barnes PF , Silva C , and Otaya M. 1996 . Testing for human immunodeficiency virus infection in patients with tuberculosis . Am J Respir Crit Care Med 153 : 1488–1450 .
Barnes PF , Yang Z , Preston-Martin S , et al. 1997 . Patterns of tuberculosis transmission in central Los Angeles . JAMA 278(14) : 1159–1163 .
Behr MA , Warren SA , Salamon H , Hopewell PC , Ponce de Leon A , Daley CL , and Small PM. 1999 . Transmission of Mycobacterium tuberculosis from patients smear-negative for acid-fast bacilli . Lancet 353 : 444–449 .
Binkin NJ , Vernon AA , Simone PM , et al. 1999 . Tuberculosis prevention and control activities in the United States: An overview of the organization of tuberculosis services . Int J Tuberc Lung Dis 3(8) : 663–674 .
Binkin NJ , Zuber PLF , Wells CD , Tipple MA , and Castro KG. 1996 . Overseas screening for tuberculosis in immigrants and refugees to the United States: Current status . Clin Infect Dis 23 : 1226–1232 .
Bishai WR , Graham NM , Harrington S , et al. 1998 . Molecular and geographic patterns of tuberculosis transmission after 15 years of directly observed therapy . JAMA 280 : 1679–1684 .
Bock NN , Mallory JP , Mobley N , DeVoe B , Brooks Taylor B. 1998 . Outbreak of tuberculosis associated with a floating card game in the rural south: Lessons for tuberculosis contact investigations . Clin Infect Dis 27 : 1221–1226 .
Booysen C , Van Rie A , Warren R , et al. 1999 . The Importance of Contact in the Transmission of Tuberculosis, abstr . 600-PD, p. S37 . In: Proceedings of the IUATLD Meeting in Madrid . Paris : International Union Against Tuberculosis and Lung Disease .
Brassard P , and Lamarre V. 1999 . Evaluation of Tuberculosis Transmission in an Outpatient Pediatric Setting, abstr . 247-PD, p. S188–189 . In: Proceedings of the IUATLD Meeting in Madrid . Paris : International Union Against Tuberculosis and Lung Disease .
Brook N , Brooks M , Redden D , et al. 1999 . A Computer-Based System for TB Contact Investigation . In: Abstracts of the 1999 ALA/ATS International Conference in San Diego , California .
Casanova MC , Gonzalez MC , Perez MM , Piqueras AR , Estelles DC , and Morera LM. 1991 . The investigation of contacts of the tuberculous pediatric patient . Med Clin 97(13) : 486–490 .
Catlos EK , Cantwell MF , Bhatia G , Gedin S , Lewis J , and Mohle-Boetani JC 1998 . Public health interventions to encourage TB class A/B1/B2 immigrants to present for TB screening . Am J Respir Crit Care Med 158 : 1037–1041 .
California Department of Health Services/California Tuberculosis Controllers Association (CDHS/CTCA) . 1998 . Joint Guidelines: Contact investigation 1–45 .
Cegielski JP , Robison VS , Robinson C , McGaha P , Hassell W , Clark SC. 1997 . Community-based screening and prevention of tuberculosis in high risk neighborhoods identified with a geographic information system . In: Program and abstracts of the 37th Interscience Conference on Antimicrobial Agents and Chemotherapy; September 28– October 1, 1997; Toronto, Ontario . Abstract K184 : 361 .
Centers for Disease Control . 1991 . Crack cocaine use among persons with tuberculosis— Contra Costa County, California, 1987–1990 . Morbid Mortal Weekly Rep 40(29) : 485–489 .
Centers for Disease Control . 1987 . Multi-drug resistant tuberculosis—North Carolina . Morbid Mortal Weekly Rep 35(51–52) : 785–787 .
Centers for Disease Control . 1980 . Tuberculosis in a drug rehabilitation center—Colorado . Morbid Mortal Weekly Rep 29(45) : 543–544 .
Centers for Disease Control and Prevention . 1999 . Tuberculosis elimination revisited: Obstacles, opportunities and renewed commitment. Report of the Advisory Council for the Elimination of Tuberculosis . Morbid Mortal Weekly Rep 48(RR-9) : 1–13 .
Centers for Disease Control and Prevention . 1999 . Tuberculosis elimination revisited . Morbid Mortal Weekly Rep 48(RR-15) : 1–20 .
Centers for Disease Control and Prevention . 1998 . Recommendations for prevention and control of TB among foreign-born persons: Report of a Working Group on Tuberculosis among the Foreign Born . Morbid Mortal Weekly Rep 47(RR-16) : 1–30 .
Community-based screening and prevention of tuberculosis in high-risk neighborhoods identified with a geographic information system, abstr . K184, p. 361. In: Program and Abstracts of the 37th Interscience Conference on Antimicrobial Agents and Chemotherapy, September 28–October 1, 1997 . Washington, DC : American Society for Microbiology .
Comstock GW. 1999 . How much isoniazid is needed for prevention of tuberculosis among immunocompetant adults? Int J Tuberc Lung Dis 10 : 847–850 .
Corless JA , Stockton PA , and Davies PDO . 1999 . Is routine screening of all sputum “smear-positive” patients necessary?, abstr . 119-PD, p. S138. In: Proceedings of the IUATLD meeting in Madrid . Paris : International Union Against Tuberculosis and Lung Disease .
Daley CL , Small PM , Schecter GF , et al. 1992 . An outbreak of tuberculosis with accelerated progression among persons infected with the human immunodeficiency virus . N Engl J Med 326 : 231–235 .
Department of Health and Human Services . 1996 . Guide to Clinical Preventive Services , 2nd Ed. Report of the U.S. Preventive Services Task Force . Washington, DC : DHHS .
DiStasio AJ , and Trump DH. 1990 . The investigation of a tuberculosis outbreak in the closed environment of a U.S. Navy ship, 1987 . Military Med 155(8) : 347–351 .
Ferebee SH. 1970 . Controlled chemoprophylaxis trials in tuberculosis: A general review . Adv Tuberc Res 17 : 28–106 .
Fernandez RA , Arazo GP , Aguirre EJM , and Arribas LJL. 1994 . The study of contacts of tuberculosis patients . Ann Intern Med 11(2) : 62–66 .
Fitzpatrick LK , Agerton T , Heirendt W , Valway S , and Onorato I. 1999 Tuberculosis in a small community: A preventable outbreak, abstr . 228-PS, p. S13. In: Proceedings of the IUATLD Meeting in Madrid . Paris : International Union Against Tuberculosis and Lung Disease .
Goldman JM , Teale C , Cundall DB , and Pearson SB. 1994 . Childhood tuberculosis in Leeds, 1982–90: Social and ethnic factors and the role of the contact clinic in diagnosis . Thorax 49(2) : 184–185 .
Greifinger RB , Heywood NJ , and Glaser JB. 1993 . Tuberculosis in prison: Balancing justice and public health . J Law Med Ethics 21(3–4) : 332–341 .
Hedemark LL. 1996 . Contact investigation of a neighborhood bar patron . Investigation of contacts to tuberculosis cases: New York City, June 7–8, 1996 . In: Symposium Summary . New York : New York City Department of Health .
Holcombe JM. 1996 . Contact investigation in a rural setting: A state perspective. Investigation of contacts of tuberculosis cases: New York City, June 7–8, 1996 . In: Symposium Summary . New York : New York City Department of Health .
Hussain SF , Watura R , Cashman B , Campbell IA , and Evans MR. 1992 . Audit of a tuberculosis contact tracing clinic . Br Med J 304(6836) : 1213–1215 .
International Union Against Tuberculosis . 1982 . Efficacy of various durations of isoniazid preventive therapy for tuberculosis: Five years of follow-up in the IUAT Trial . Bull WHO 60(4) : 555–564 .
Iseman MD , 1997 . Editorial response: an unholy trinity—Three negative sputum smears and release from tuberculosis isolation . Clin Infect Dis 25 : 671–672 .
Jereb JA , Burwen DR , Dooley SW , et al. 1993 . Nosocomial outbreak of tuberculosis in a renal transplant unit: Application of a new technique for restriction-fragment-length polymorphism analysis of Mycobacterium tuberculosis isolates . J Infect Dis 168(5) : 1219–1224 .
Kawamura LM , and Green Rush AM. 1999 . TB control tech transfer toolbox: Translating TB control models into step-by-step program guides, abstr . 351-PS, p. S90. In: Proceedings of the IUATLD Meeting in Madrid . Paris : International Union Against Tuberculosis and Lung Disease .
Kellogg B , Dye C , Cox K , and Rosenow G. 1987 . Public health nursing model for contact follow up of patients with pulmonary tuberculosis . Public Health Nursing 4(2) : 99–104 .
Kenyon TA , Ridzon R , Luskin-Hawk R , et al. 1997 . A nosocomial outbreak of multidrug-resistant tuberculosis . Ann Intern Med 127(1) : 32–36 .
Kenyon TA , Valway SE , Ihle WW , Onorato IM , and Castro KG. 1996 . Transmission of multidrug-resistant Mycobacterium tuberculosis during a long airplane flight . N Engl J Med 334 : 933–938 .
Liippo KK , Kulmala K , and Eero OJT . 1993 . Focusing tuberculosis contact tracing by smear grading of index cases . Am Rev Respir Dis 148 : 235–236
Lohuis AMF , van der Werff GFM , Drost AP , Moree C , Vegter B , and Sebek MMGG . 1999 . The complete nursing assessment . The development of a standard nursing questionnaire for every patient with TB in The Netherlands, abstr . 136-PS, p. S94. In: Proceedings of the IUATLD Meeting in Madrid . Paris : International Union Against Tuberculosis and Lung Disease .
Mangura BT , Napolitano EC , Passannante MR , McDonald RJ , and Reichman LB. 1998 . Mycobacterium tuberculosis miniepidemic in a church gospel choir . Chest 113 : 234–237 .
Marks S , Taylor Z , Nguyen C , Qualls N , Shrestha-Kuwahara R , and Wilce M. 1999 . Outcomes of a CDC TB Contact Investigation Study . Preliminary results presented to the Institute of Medicine Committee on the Elimination of Tuberculosis in the United States .
Mehta JB , and Bentley S. 1992 . Prevention of tuberculosis in children: Missed opportunities . Am J Prev Med 8(5) : 283–286 .
Menzies D. 1997a . Effect of treatment on contagiousness of patients with active tuberculosis . Infect Control Hosp Epidemiol 18 : 582–586 .
Menzies D. 1997b . Issues in the management of contacts of patients with active tuberculosis . Can J Public Health 88(3) : 197–201 .
Menzies R. 1996 . Interpreting contact investigation results: The effect of BCG vaccination . Investigation of contacts to tuberculosis cases: New York City, June 7–8, 1996. In: Symposium Summary . New York : New York City Department of Health .
Menzies R , Rocher I , and Vissandjee B. 1993 . Factors associated with compliance in treatment of tuberculosis . Tubercle Lung Dis 74(1) : 32–37 .
Moodie AS , and Riley RL. 1974 . Infectivity of patients with pulmonary tuberculosis in inner city homes . Am Rev Respir Dis 110 : 810–812 .
Moore M , Schulte J , Valway SE , et al. 1998 . Evaluation of transmission of Mycobacterium tuberculosis in a pediatric setting . J Pediatr 133 : 108–112 .
Nardell E , Keegan J , Cheney S , and Etkind S. 1991 . Airborne infection: Theoretical limits of protection achievable by building ventilation . Am Rev Respir Dis 144 : 302–306 .
Nivin B , Nicholas P , Gayer M , Frieden TR , and Fujiwara PI. 1998 . A continuing outbreak of multidrug-resistant tuberculosis, with transmission in a hospital nursery . Clin Infect Dis 26 : 303–307 .
Nolan CM , Goldberg SV , and Buskin SE. 1999 . Hepatoxicity associated with isoniazid preventive therapy: A seven-year survey from a public health tuberculosis clinic . JAMA 281 : 1014–1018 .
Onorato I. 1999 . Managing TB Outbreaks . Centers for Disease Control and Prevention presentation to the Institute of Medicine Committee on the Elimination of Tuberculosis in the United States .
Peerbooms PGH , van Doornum GJJ , van Deutekom H , Coutinho RA , and van Soolingen D. 1995 . Laboratory-acquired tuberculosis . Lancet 345 : 1311–1312 .
Pena A , Anibarro L , Sanjurjp A , Nunes MJ , Garcia JC , and Romay A. 1999 . Childhood transmission of tuberculosis, abstr . 561-PD, p. S192. In: Proceedings of the IUATLD Meeting in Madrid . Paris : International Union Against Tuberculosis and Lung Disease .
Rodriguez EM , Steinbart S , Shaulis G , Bur S , and Dwyer DM. 1996 . Pulmonary tuberculosis in a high school student and a broad contact investigation: Lessons relearned . Md Med J 45(12) : 1019–1022 .
Rothenburg RB. 1996 . Social network approach in contact tracing . Investigation of Contacts to tuberculosis cases: New York City, June 7–8, 1996. In: Symposium Summary . New York : New York City Department of Health .
Rubilar M , Brochwicz-Lewinski MJ , Anderson M , and Leitch AG. 1995 . The outcome of contact procedures for tuberculosis in Edinburgh, Scotland, 1982–1991 . Respir Med 89(2) : 113–120 .
Rubin FL , and Lynch DC. 1996 . Tuberculosis control through contact investigation . Penn Med 22–23 .
Sasaki Y , Yamagishi F , and Suzuki K. 1995 . The present condition of patient's, doctor's and total delays in tuberculosis case finding and countermeasures in the future . Kekkaku 70(1) : 49–55 .
Shrestha-Kuwahara R , Wilce M , Deluca N , Rosenbaum J , and Taylor Z. 1999 . Patient provider communication during TB contact investigation, abstr . 219-PD, p. S139. In: Proceedings of the IUATLD Meeting in Madrid . Paris : International Union Against Tuberculosis and Lung Disease .
Snider DE , Kelly GD , Cauthen GM , Thompson NJ , and Kilburn JO. 1985 . Infection and disease among contacts of tuberculosis cases with drug-resistant and drug-susceptible bacilli . Am Rev Respir Dis 132 : 125–132 .
Tabet SR , Goldbaum GM , Hooton TM , Eisenach KD , Cave MD , and Nolan CM. 1994 . Restriction-fragment-length polymorphism analysis detecting a community-based outbreak among persons infected with human immunodeficiency virus . J Infect Dis 169 : 189–192
Topley JM , Maher D , and Mbewe LN. 1996 . Transmission of tuberculosis to contacts of sputum-positive adults in Malawi . Arch Dis Child 74(2) : 140–143 .
Valway SE , Sanchez MPC , Shinnick TF , et al. 1998 . An outbreak involving extensive transmission of a virulent strain of Mycobacterium tuberculosis . N Engl J Med 338(10) : 633–639 .
van Soolingen D , Lambregts-van Weezenbeek CSB , de Haas PEW , Veen J , and van Embden JDA . 1996 . Transmission of sensitive and resistant strains of Mycobacterium tuberculosis in The Netherlands, 1993–1995 . Nederlands Tijdschr Geneesk 140 : 2286–2289 .
Veening GJ. 1968 . Long-term isoniazid prophylaxis: Controlled trial on INH prophylaxis after recent tuberculin conversion in young adults . Bull Int Union Tuberc 41 : 169–171 .
Washko R , Robinson E , Fehrs LJ , and Frieden TR. 1998 . Tuberculosis transmission in a high school choir . J School Health 68(6) : 256–259 .
Wells CD , Zuber PLF , Nolan CM Binkin NJ , and Goldberg SV. 1997 . Tuberculosis prevention among foreign-born persons in Seattle-King County, Washington . Am J Respir Crit Care Med 156 : 573–577 .
Zhang Y , Heym B , Allen B , Young D , and Cole S. 1992 . The catalase-peroxidase gene and isoniazid resistance of Mycobacterium tuberculosis . Nature 358 : 591–593 .