Tuberculin Skin Test Screening Practices Among US Colleges and Universities

Transcription

1 Tuberculin Skin Test Screening Practices Among US Colleges and Universities Karen A. Hennessey, PhD, MSPH; Joann M. Schulte, DO; Linda Cook, MA; Marjeanne Collins, MD; Ida M. Onorato, MD; Sarah E. Valway, DMD, MPH Context. Concern about transmission of Mycobacterium tuberculosis on college campuses has prompted some schools to institute tuberculin skin test screening of students, but this screening has never been evaluated. Objective. To describe tuberculin skin test screening practices and results of screening in colleges and universities in the United States. Design and Setting. Self-administered mail and telephone questionnaire in November and December 1995 to a stratified random sample of US 2-year and 4- year colleges and universities. Main Outcome Measures. Type of tuberculin screening required; types of schools requiring screening; number and rate of students with positive skin test results and/or diagnosed as having tuberculosis. Results. Of the 3148 US colleges and universities, 624 (78%) of 796 schools surveyed responded. Overall, 378 schools (61%) required tuberculin screening; it was required for all new students (US residents and international students) in 161 (26%) of 624 schools, all new international students but not new US residents in 53 (8%), and students in specific academic programs in 294 (47%). Required screening was more likely in 4-year vs 2-year schools, schools that belonged to the American College Health Association vs nonmember schools, schools with immunization requirements vs schools without, and schools with a student health clinic vs those without (P for all). Public and private schools were equally likely to require screening (64% vs 62%; P =.21). In the 378 schools with screening requirements, tine or multiple puncture tests were accepted in 95 (25%); test results were recorded in millimeters of induration in 95 (25%); and 100 (27%) reported collecting results in a centralized registry or database. Of the 168 (27%) of 624 schools accepting only Mantoux skin tests and reporting results for school years through , 3.1% of the students screened had positive skin test results (median percentage positive, 0.8%). International students had a significantly higher case rate for active tuberculosis than US residents (35.2 vs 1.1 per students screened). Conclusions. Widespread tuberculin screening of students yielded a low prevalence of skin test reactors and few tuberculosis cases. To optimize the use of limited public health resources, tuberculin screening should target students at high risk for infection. JAMA. 1998;280: From the Division of Tuberculosis Elimination, National Center for HIV, STD, and TB Prevention, Centers for Disease Control and Prevention, Atlanta, Ga (Drs Hennessey, Schulte, Onorato, and Valway); and the University of Pennsylvania, Philadelphia (Dr Collins and Ms Cook). Dr Hennessey is currently an Epidemic Intelligence Service Officer with the Epidemiology Program Office assigned to the Polio Eradication Activity in the National Immunization Program at the Centers for Disease Control and Prevention. Preliminary results of this study were presented at the 1997 International Conference of the American Thoracic Society/American Lung Association, San Francisco, Calif, May 18-20, Reprints: Division of Tuberculosis Elimination, 1600 Clifton Rd (Mailstop E10), Centers for Disease Control and Prevention, Atlanta, GA TUBERCULOSIS on college and university campuses has been raised as a concern because of the large number of students who come from areas of the world where tuberculosis is common. In 1997, 39% of all tuberculosis cases in the United States occurred among foreignborn persons and, from 1986 through 1993, the tuberculosis case rate among foreign-born persons was 31 per , more than 5.5 times the rate among USborn persons. 1,2 National tuberculosis surveillance data do not include information on whether a person with tuberculosis is a college or university student. However, from 1993 through 1997, a total of 7476 incident cases of tuberculosis (6% of all tuberculosis cases reported from 1993 through 1997) were reported in the United States among persons aged 18 through 24 years, the age group most likely to be in college (Centers for Disease Control and Prevention [CDC], unpublished data). Of these cases, 62% were foreign born; most came from 4 countries: Mexico(28%), Vietnam(14%), the Philippines (9%), and India (5%). Thirty-six percent of all foreign-born tuberculosis cases aged 18 through 24 years were diagnosed as having tuberculosis less than 1 year after entering the United States. Some of these persons may have been screened after arrival in the United States as a part of routine refugee or immigrant screening. Foreign-born college students may be either US residents or international students, ie, students who are residents of foreign countries but who come to the United States to study. In 1995, approximately international students came to the United States to study and constituted 3% of the US college student population. 3 Of the international students, 58% came from Asia, an area of the world where many countries have 2008 JAMA, December 16, 1998 Vol 280, No. 23 Tuberculosis Screening Practices in US Colleges Hennessey et al

2 tuberculosis rates estimated to be more than 20 times that of the United States. 4 Screening for active tuberculosis disease is required of refugees and persons legally immigrating to the United States, but no medical screening is required for international students before entering the United States. 5 Studies in the 1980s at 3 different colleges found that overall, approximately 1.3% of students had positive skin test results; among foreign-born or international students, however, the rate of positive skin test results was around 50%. 6-8 A recent outbreak of tuberculosis at a university demonstrated how Mycobacterium tuberculosis can easily spread on college campuses given the prolonged close contact students have with each other. 9 To address concerns of potential M tuberculosis transmission on campus, some schools have instituted required tuberculin skin test screening of students. However, little is known about tuberculin screening practices in colleges and universities in the United States since a comprehensive study has never been conducted. This article reports the results of a CDC/American College Health Association (ACHA) survey of colleges and universities across the United States, describing the type and extent of required tuberculin skin testing of students and the number of students identified with positive skin test results or active tuberculosis disease as a result of required screening. METHODS A self-weighted, stratified, random sample of 796 schools was selected from the 3148 colleges and universities referenced in Peterson s guides to 2- and 4- year colleges Because a previous CDC/ACHA survey of prematriculation immunization practices among college and universities found requirements to be more likely among 4-year schools, public schools, and schools that were ACHA members, we believed the same was likely to be true for any tuberculin skin test screening. 13 Therefore, schools were randomly sampled within these strata, ie, whether a school was 2- or 4- year, ACHA member or nonmember, and public or private. The 796 schools were mailed survey questionnaires in November For schools that were ACHA members, the ACHA contact person at the school received the questionnaire. For nonmember schools, the questionnaire was mailed to the contact personlistedinthepeterson sguides. 11,12 A cover letter was included with all survey forms asking that the questionnaire beforwardedtothepersonateachschool who had knowledge of and access to information about each school s health screening practices. Schools that did not respond by mail were surveyed by telephone. After all survey forms were received, they were reviewed by the authors. Schools with responses that appeared invalid or inconsistent were called, the questionnaire was reviewed with the person who completed the form, and corrections were made to responses as needed. Associations between school characteristics and having a tuberculin screening requirement were measured using the normal approximation for the comparison of 2 proportions. 14 If schools reported any tuberculin skin test screening requirements, we collected data on the number of students screened and the number with positive skin test results from school years through The definition of a positive or negative skin test result was determined by each student s health care provider. The percentage of students with positive skin test results was calculated using those schools that reported results of required screening and accepted the Mantoux skin test method only. All schools were asked to report if they had any tuberculosis cases among their students from school years through , regardless of whether any tuberculin screening was required. Additional information asked about the cases included how the student was diagnosed as having tuberculosis, whether the student was US born or foreign born, and whether the student was an international student or US resident. To measure the yield of required tuberculin skin test screening, we estimated the rate of tuberculosis cases identified through required screening for schools that required screening of all new students or new international students. As denominators for these rates, we used published figures of total enrollment for US residents and international students. For 4-year schools, we assumed that the number of international students screened per year was one quarter of the school s total international student enrollment; the number of US residents screened per year was one quarter of the school s total enrollment minus the international students. For 2-year schools, the same method was used, but we assumed that the number of students screened was half the total enrollment. 11,12 School records and individual student medical records were not reviewed to verify reported tuberculin skin test results or diagnoses of active tuberculosis. RESULTS Of the 796 schools included in the survey, 624 (78%) responded: 402 (64%) by mail and 222 (36%) by telephone. There Table 1. Characteristics of the 624 Schools Surveyed Characteristics No. (%) 2-Year school 263 (42) 4-Year school 361 (58) American College Health 196 (31) Association member American College Health 428 (69) Association nonmember Public school 308 (49) Private school 316 (51) Student health clinic on campus 361 (58) No student health clinic 259 (41) Missing data on student health clinics 4 (1) Immunizations required for school entry 337 (54) No immunization requirement 270 (43) Missing data on immunization 17 (3) requirement Required any tuberculosis screening 378 (61) was no difference between schools that responded to the survey and those that did not with regard to any of the sampling strata, ie, whether they were a 2- vs 4-year school, public vs private school, or ACHA member. There were also no differences across the sampling strata by type of response (mail or telephone). Additionally, the response rates were uniform across strata; therefore, the proportions from this study represent a weighted proportion of schools nationwide. The characteristics of the 624 schools responding to the survey are shown in Table 1. In 1995, the median total student enrollment at these schools was 1963 (range, ) and median international student enrollment was 33 (range, ). Of the 624 schools that responded to the survey, 337 surveys (54%) were completed by staff from student health services, 284 (46%) by school administrators, and for 3 schools (0.5%) this information was unavailable. Of the 284 schools whose forms were completed by administrative staff, 245 (86%) did not have student health clinics and thus no clinic staff could have completed the forms. Overall, 378 (61%) of the 624 schools required tuberculin skin test screening for some subset of their students. All new students (US residents and international students) were required to have tuberculin skin tests in 161 (26%) of the 624 schoolsandallnewinternationalstudents (but not new students from the United States) in 53 (8%) of the schools. Overall, 294 (47%) of the schools required tuberculin screening in specific academic programs; 231 (62%) of 373 schools with health care programs; 142 (32%) of 446 with education programs; and 42 (19%) of 221 with social work programs. Required tuberculin screening was more likely in 4- vs 2-year schools, in ACHA member vs nonmember schools, schools with a student health clinic vs those without, and schools with immunization requirements JAMA, December 16, 1998 Vol 280, No. 23 Tuberculosis Screening Practices in US Colleges Hennessey et al 2009

3 Table 2. Factors Associated With Required Tuberculosis Screening No. (%) of Schools That Characteristics Require Screening 2-Year school 131/254 (52) 4-Year school 247/350 (71) American College Health Association member 160/195 (82) American College Health Association nonmember 218/409 (53) Public school 191/300 (64) Private school 187/304 (62) Student health clinic on campus 283/359 (79) No student health clinic 95/242 (39) Immunizations required for school entry 255/333 (77) No immunizations required 118/261 (45) *P values computed using the normal approximation for the comparison of 2 proportions. vs those with no immunization requirements (P). Public and private schools were equally likely to have required tuberculin screening (64% vs 62%; P =.21) (Table 2). When schools were divided into quartiles by size of school, 41% of the schools in the lowest quartile (ie, 832 students) had skin test screening requirementswhile68% to72% ofschools ineachoftheother3quartileshadscreening requirements. Reported tuberculin screening practices of the 378 schools with required screening are shown in Table 3. Exemptions from skin test screening were allowed in 85% of the schools for various reasons such as reports of prior positive skin test results, BCG vaccination, or religious or philosophical reasons. Tine or multiple puncture tests were accepted in 25% of the schools. Skin test results were required to be recorded in millimeters of induration in only 25% of the schools and thus it was not possible to analyze results by millimeters of induration. Of the 378 schools with required screening, only 100 (27%) reported that they collected all skin test results in a centralized registry or database. For new students, tuberculin skin testing was part of a prematriculation requirement and these skin tests were done by each student s health care provider and results submitted to the college or university prior to the student enrolling. For colleges and universities that required tuberculin skin testing for students in specific academic programs (eg, nursing, elementary education), this requirement would occur after the student had enrolled, usually when the student began working in a health care setting or elementary school. The skin tests for these students could be done on campus, if there was a health clinic. Overall, 42% of the 378 colleges and universities requiring tuberculin skin tests reported that no skin testing was done at the school, 44% could do skin testing on campus and also accepted results from elsewhere, 12% reported that required skin P Value*.21 Table 3. Reported Practices Among the 378 Schools With Required Tuberculosis Screening Screening Practices No. (%) Exempted from screening Prior positive skin test results 227 (60) BCG vaccine 94 (25) Religious/philosophical reasons 123 (33) No exemptions 58 (15) Type of skin test accepted Mantoux only 256 (68) Tine (multiple puncture) or Mantoux 95 (25) Unknown 27 (7) Skin test results recorded as Millimeters of induration only 95 (25) Millimeters of induration or 224 (59) positive/negative Results not recorded 29 (8) Unknown 30 (8) Central database of skin test results includes All students required to be screened 100 (27) Only students screened at school 25 (7) Only students with positive 20 (5) skin test results Results not collected 209 (55) Unknown 24 (6) testing was done on campus only, and 2% had missing information. Analyses of screening results were limited to the 168 schools that required screening, accepted Mantoux skin tests only, and reported results of their required screening (Table 4). These 168 schools represent 27% of all schools surveyed and 66% of schools with required screening. For all 168 schools combined, a total of students were screened and 3.1% of the students had positive tuberculin skin test results; the median percent positive for the 168 schools was 0.8%. There was no difference in percentage of students with positive skin test results between schools collecting results in a central database and schools using paper records; both found 3% of all students to have positive skin test results. When analyzed by type of screening program, overall percentages of positive skin test results ranged from 2.1% for the 16 schools that screened only new students to 22.9% for the 4 schools that screened only international students. All 168 schools reported that students with positive skin test results were referred to either the health department or their private physician. Fifty-nine (35%) of these 168 schools also referred students to the student health clinic. No information was available about any follow-up for students with positive skin test results (eg, chest radiographs, isoniazid preventive therapy) because it was not feasible to collect these data from each student s private health care provider. However, 139 (82%) of the 168 schools had student health clinics on campus and of these, 52 (37%) reported they could offer isoniazid at the clinic. During the 4-year period studied, 114 casesofactivetuberculosisdiseaseinstudents were reported from 68 (11%) of the 624 schools surveyed; no tuberculosis cases were reported in 556 schools (89%). Of the 114 cases, only 32 cases (26 [81%] of whom were foreign born) were identified as a result of required tuberculin skin test screening: 13 were from schools that required screening of all new students, 11 from schools screening international students, and 8 from screening students in specific academic programs. Of the 214 schools that required screening of all new US resident or new international students (n = 161) or new international students but not new US resident students (n = 53), enrollment figures for international students were not available for 20 schools. (No tuberculosis cases were identified through required screening at these 20 schools.) Thus, analysis of rates of tuberculosis cases identified through required tuberculin screening was limited to the 194 schools with enrollment figures available. Among these schools, 24 cases of tuberculosis were reported to have been identified through required screening, for an estimated overall rate of 4.7 per students screened. When analyzed by whether students were US residents or international students, international students had markedly higher estimated rates of tuberculosis identified through required screening: 35.2 per vs 1.1 per among US residents (Table 5). Tuberculosis cases rates could not be estimated for the 8 cases (5 of whom were foreign born) identified through required screening in specific academic programs because no information was available to estimate the numbers of students screened in these programs. Of the remaining 82 cases of tuberculosis among students that were not identified as a result of required screening,67were identified from symptoms or through contact investigations, and the reason the case was identified was unknown for 15. Thirty-nine of the 82 cases were US born,39wereforeignborn,andthebirthplace was unknown for JAMA, December 16, 1998 Vol 280, No. 23 Tuberculosis Screening Practices in US Colleges Hennessey et al

4 COMMENT In the United States, tuberculin screening of the general population is not considered an effective method for identifying infection with M tuberculosis or active cases of tuberculosis and is not included as part of the strategy to eliminate tuberculosis from the United States. 15,16 Studies have shown that widespread tuberculin screening of lowrisk populations finds few persons with positive skin test results or active tuberculosis and that targeted screening of high-risk persons is a much more costeffective approach The CDC recommends screening only those persons at high risk for infection with M tuberculosis. 15,20 This includes contacts of infectious tuberculosis cases, human immunodeficiency virus (HIV) infected persons, injection drug users, health care workers, other immunocompromised persons, and foreign-born persons from countries with high tuberculosis rates. This strategy was adopted because tuberculin screening of low-risk persons diverts limited public health resources from higher-priority activities such as treating and following up tuberculosis cases and their contacts. Additionally, it is likely that many positive skin test results from widespread screening in lowrisk persons do not represent infection with M tuberculosis. 16,21,22 Weighted analysis from our study found that approximately 60% of all US colleges and universities have some form of required tuberculin screening and that 3% of the general student population screened (including international students) had positive skin test results (median percentage of positive test results for these schools, 0.8%). In targeted screening programs, 23% of the international students had positive skin test results. Required tuberculin screening identified few tuberculosis cases. Overall, we estimated that 4.7 tuberculosis cases were identified for every students screened. However, the rate of active tuberculosis identified through screening among international students was 32 times higher than among US residents. It is possible that these rates are underestimated if schools were not informed of all tuberculosis cases identified through tuberculin screening. Of the 5 tuberculosis cases identified through required screening who were US residents, 2 were foreign born. Data were not available to determine how many students who were US residents were US born. Because most of the tuberculosis cases among students were not identified as a result of required screening, additional studies examining risk factors for these students might indicate Table 4. Percentage of Students With Positive Skin Test Results Among Schools With Required Screening Programs, Fall 1992-Fall 1995* Screening Criteria Schools No. Screened New students only International students only Students in specific programs only Combinations of the above Total the missed opportunities for preventing the development of tuberculosis. Our results were limited by the fact that survey questionnaires were not completed by all schools, were completed by a variety of personnel (eg, administrators, student health staff), and medical records were not reviewed by the authors to verify diagnoses. The estimated percentage of students with positive skin test results may be high if schools that instituted required screening did so because of a higher risk of tuberculosis on their campus. Conversely, because many schools exempted from screening those students who had received BCG or reported prior positive skin test results, screening results could underestimate the true prevalence of infection in the student body. Despite these limitations, estimates of the percentage of positive skin test results among US residents are comparable with results of other studies of US populations of similar age. 6-8,23,24 Problematic screening practices were identified among some schools with required screening and should be highlighted for any institution with, or considering, a screening program. Despite published guidelines recommending only the use of the Mantoux method for skin testing, 16 many schools accepted skin test results from tine tests. Multiple puncture devices are less specific than the Mantoux test and should not be used. 25 Fifty-nine percent accepted skin test results as positive or negative. It is essential that millimeters of induration be recorded because interpretation of skin test results depends on a combination of millimeters of induration and risk Skin Test Positive Results *Schools that accepted Mantoux skin tests only and reported results of required screening. New students only include US residents and international students. Skin Test Positive Results, % Table 5. Estimated Tuberculosis Rates for Cases Identified Through Required Screening of All New Students Who Are US Residents or International Students, Fall 1992-Fall 1995 Type of Student Schools* Cases No. Screened Rate per Students Screened Relative Rate US Resident Reference International Student *A total of 194 schools are included in this table: 141 schools required screening of all new US resident and new international students, and 53 required screening of new international but not new US resident students. For the 141 schools requiring screening of both types of students, information for US resident and international students is shown for each type of student. Estimated from enrollment figures. Permanent residents of the United States; 3 cases were US born and 2 cases were foreign born. Residents of other countries coming to the United States to study. factors for tuberculosis. For example, contacts of a tuberculosis case or persons infected with HIV are considered skin test positive at 5 mm induration; foreign-born persons at 10 mm; and persons with no risk factors for tuberculosis infection are considered positive at 15 mm induration. 16 Because tuberculin reactivity as a result of receiving BCG vaccine wanes over time and rarely persists more than 10 years after vaccination, 26 students with a history of BCG vaccination should be included in any tuberculin skin test screening requirements, unless they provide documentation of a prior positive skin test result. Any student who had received BCG vaccine and has a tuberculin skin test reaction of 10 mm or more of induration should be considered to have a positive reaction and should be referred to a physician to rule out active tuberculosis and evaluate for preventive therapy. 16,26 Only 25% of schools that required students to be screened collected all results in a database or central registry. Schools must collect and analyze data to determine if screening is necessary on their campus. Decisions regarding tuberculosis screening in communities are usually based on local epidemiologic data. However, colleges and universities should have their own criteria based on the demographics of their students. If schools require screening for tuberculosis, then (1) the goals of any screening should be clearly defined and mechanisms must be in place to ensure that students receive anyfollow-up(eg, chestradiographs, medical evaluation, preventive therapy) that is needed; (2) widespread screening of the general student population should not be JAMA, December 16, 1998 Vol 280, No. 23 Tuberculosis Screening Practices in US Colleges Hennessey et al 2011

5 done, instead students at high risk for infection, eg, international or foreign-born students from countries with high rates of tuberculosis, students working in health care settings, and immunocompromised students, should be targeted for screening; (3) only the Mantoux method for skin testing should be used; (4) students vaccinated with BCG should not be excluded from screening; and (5) results of screening (all skin test results, demographic, clinical, and follow-up information) must be collected and periodically analyzed to evaluate the usefulness of screening and ensure that the goals of screening are met. The modification and/ References 1. Centers for Disease Control and Prevention. Tuberculosis morbidity US, MMWR Morb Mortal Wkly Rep. 1998;47: McKenna MT, McCray E, Onorato I. The epidemiology of tuberculosisamongforeign-bornpersons in the United States, N Engl J Med. 1995; 332: Davis TM, ed. Open Doors 1994/95: Report on International Educational Exchange. New York, NY: Institute of International Education; Raviglione MC, Snider DE, Kochi A. Global epidemiology of tuberculosis: morbidity and mortality of a worldwide epidemic. JAMA. 1995;273: Immigration Act of 1990, Pub L No , 104 Stat Nelson ME, Fingar AR. Tuberculosis screening and prevention for foreign-born students: eight years experience at Ohio University. Am J Prev Med. 1995;11(suppl 1): Susmano S. Testing international students for tuberculosis. J Am Coll Health. 1990;39: Quillan S, Malotte K, Shlian D. Evaluation of a tuberculosis screening and prophylaxis program for international students. J Am Coll Health. 1990;38: Braden CR. Infectiousness of a university student with laryngeal and cavitary tuberculosis: investigative team. Clin Infect Dis. 1995;21: Levy PS, Lemeshow S. Sampling of Populations Methods and Applications. New York, NY: John Wiley & Sons Inc; Stern JD, ed. Peterson s Guide to Two-Year Colleges th ed. Princeton, NJ: Peterson s Guides; Pendleton EH, ed. Peterson s Guide to Four- Year Colleges th ed. Princeton, NJ: Peterson s Guides; Baughman AL, Williams WW, Atkinson WL, Cook LG, Collins M. The impact of college prematriculation immunization requirements and risk for measles outbreaks. JAMA. 1994;272: Snedecor GW, Cochran WG. Statistical Methods. 7th ed. Ames: Iowa State University Press; American Thoracic Society/CDC. Control of tuberculosis in the United States: American Thoracic Society. Am Rev Respir Dis. 1992;146: Centers for Disease Control and Prevention. A strategic plan for the elimination of tuberculosis in the United States. MMWR Morb Mortal Wkly Rep. 1989;38(S-3): Driver CR, Valway SE, Cantwell MF, Onorato IM. Tuberculin skin test screening in schoolchildren in the United States. Pediatrics. 1996;98: Henry PM, Mills WA, Holtan NR, et al. Screening for tuberculosis infection among secondary school students in Minneapolis-St. Paul: policy implications. Minn Med. 1996;79: or continuation of any screening program should be based on these analyses. This study was funded by grant H75/CCU from the Centers for Disease Control and Prevention to the American College Health Association. Dr Collins was the principal investigator. The authors would like to acknowledge Robert Pratt for his assistance with data management. 19. Mohle-Boetani JC, Miller B, Halpern M, et al. School-based screening for tuberculosis infection: a cost-benefit analysis. JAMA. 1995;274: Centers for Disease Control and Prevention. Essential components of a tuberculosis prevention and control program: screening for tuberculosis and tuberculosis infection in high-risk populations. MMWR Morb Mortal Wkly Rep. 1995;44(RR-11): Huebner RE, Schein MF, Bass JB. The tuberculin skin test. Clin Infect Dis. 1993;17: StarkJR. Thetuberculinskintest.PediatrAnn. 1993;22: Trump DH, Hyams KC, Cross ER, Struewing JP. Tuberculosis infection among young adults entering the US Navy in Arch Intern Med. 1993; 153: Lifson AR, Halcon L, Johnston AM, et al. Tuberculin skin testing among economically disadvantaged youth: studies of US Job Corps students. Am J Epidemiol. In press. 25. Starke JR, Jacobs RF, Jereb J. Resurgence of tuberculosis in children. J Pediatr. 1992;120: Centers for Disease Control and Prevention. The role of BCG vaccine in the prevention and control of tuberculosis in the United States. MMWR Morb Mortal Wkly Rep. 1996;45(RR-4): JAMA, December 16, 1998 Vol 280, No. 23 Tuberculosis Screening Practices in US Colleges Hennessey et al