Published online July 2, 2007
PEDIATRICS Vol. 120 No. 1 July 2007, pp. 90-99 (doi:10.1542/peds.2006-2168)
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow P3Rs: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when P3Rs are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Flaherman, V. J.
Right arrow Articles by Royce, S. E.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Flaherman, V. J.
Right arrow Articles by Royce, S. E.
Related Collections
Right arrow Infectious Disease & Immunity

ARTICLE

Cost-effectiveness of Alternative Strategies for Tuberculosis Screening Before Kindergarten Entry

Valerie J. Flaherman, MD, MPHa, Travis C. Porco, PhD, MPHb, Elliot Marseille, PhD, MPPc and Sarah E. Royce, MD, MPHb

a Department of Pediatrics, University of California, San Francisco, California
b California Department of Health Services, Richmond, California
c Health Strategies International, Orinda, California


   ABSTRACT
TOP
 ABSTRACT
METHODS
 RESULTS
 DISCUSSION
 APPENDIX: CHOICE OF BASELINE...
 REFERENCES
 
OBJECTIVE. We undertook a decision analysis to evaluate the economic and health effects and incremental cost-effectiveness of using targeted tuberculin skin testing, compared with universal screening or no screening, before kindergarten.

METHODS. We constructed a decision tree to determine the costs and clinical outcomes of using targeted testing compared with universal screening or no screening. Baseline estimates for input parameters were taken from the medical literature and from California health jurisdiction data. Sensitivity analyses were performed to determine plausible ranges of associated outcomes and costs. We surveyed California health jurisdictions to determine the prevalence of mandatory universal tuberculin skin testing.

RESULTS. In our base-case scenario, the cost to prevent an additional case of tuberculosis by using targeted testing, compared with no screening, was $524897. The cost to prevent an additional case by using universal screening, compared with targeted testing, was $671398. The incremental cost of preventing a case through screening remained above $100000 unless the prevalence of tuberculin skin testing positivity increased to >10%. More than 51% of children entering kindergarten in California live where tuberculin skin testing is mandatory.

CONCLUSIONS. The cost to prevent a case of tuberculosis by using either universal screening or targeted testing of kindergarteners is high. If targeted testing replaced universal tuberculin skin testing in California, then $1.27 million savings per year would be generated for more cost-effective strategies to prevent tuberculosis. Improving the positive predictive value of the risk factor tool or applying it to groups with higher prevalence of latent tuberculosis would make its use more cost-effective. Universal tuberculin skin testing should be discontinued, and targeted testing should be considered only when the prevalence of risk factor positivity and the prevalence of tuberculin skin testing positivity among risk factor–positive individuals are high enough to meet acceptable thresholds for cost-effectiveness.

Key Words: tuberculosis • cost-effectiveness • school health services • mass screening • tuberculin test

Abbreviations: TST—tuberculin skin testing • TT—targeted tuberculin skin testing • PTCG—Pediatric Tuberculosis Collaborative Group • RQ—risk factor questionnaire • QALY—quality-adjusted life year • BESS—Medicare Part B Extract Summary System

Routine tuberculin skin testing (TST) of children entering kindergarten is often practiced in the United States, despite recommendations of major policymaking organizations.1–3 One reason for its persistence may be that these organizations have not collaborated on specific alternative recommendations for specific ages. For example, in 2004, the Pediatric Tuberculosis Collaborative Group (PTCG) released guidelines recommending that universal TST should be replaced by risk factor screening followed by targeted TST for individuals with ≥1 positive response on the risk factor questionnaire (RQ).4 The recommended RQ is based on evidence from a variety of studies in pediatric populations5–11 and includes questions on country of birth, travel history, tuberculosis exposure, and close contact with someone with positive TST results. However, the PTCG did not specify optimal age groups for testing, and its guidelines suggest that testing in older age groups may be preferable.

In the absence of specific guidelines, public health officials have made individual decisions based on the available evidence and local preferences. In California, local health officers mandate universal prekindergarten TST in many counties and school districts. Universal TST is also required by the state's Medicaid program, through the screening component of the Early and Periodic Screening Diagnosis and Treatment benefit. A decision analysis showed moderate cost-effectiveness for universal screening in Santa Clara county in 1995,12 and individual health officers and program officers have used these findings to make decisions regarding screening. Children in districts and jurisdictions that do not require universal TST may receive TST, risk factor screening followed by targeted TST (TT), or no screening at all at the well-child check before kindergarten entry. Similar inconsistencies have been noted in other states.13,14

TT has been shown to be more cost-effective than universal screening for some populations of US children.12,15 However, the cost-effectiveness of using the newly developed RQ for prekindergarten screening has not been evaluated. To determine the economic and health consequences of alternative strategies for screening children in this age group, we undertook a decision analysis comparing mandatory universal TST, mandatory TT using the RQ, and no mandatory screening. We used a range of estimates for risk factor prevalence, risk of progression to disease, and proportion of TST-positive individuals in the population. We also examined the possible costs and health outcomes expected to be associated with these 3 strategies in California.


   METHODS
TOP
 ABSTRACT
 METHODS
RESULTS
 DISCUSSION
 APPENDIX: CHOICE OF BASELINE...
 REFERENCES
 
Model
We developed a computer-based, deterministic, decision-analysis model that calculated costs and benefits from the perspective of the health care system. Cost per case of tuberculosis averted was the key outcome. Incremental costs and health outcomes were calculated by comparing results for universal screening versus TT and for TT versus no screening. Clinical and economic outcomes were discounted at a rate of 3% annually over 20 years. All costs were converted to 2004 US dollars by using the medical care component of the Consumer Price Index.16 The model was implemented in CLISP (Fig 1). 17 Screening alternatives evaluated, model characteristics, and data sources used for input specification are discussed below.


Figure 1
View larger version (12K):
[in this window]
[in a new window]

  		FIGURE 1 Decision tree for cost-effective analysis. Branches that represent screening options begin from square decision nodes, branches that represent chance events begin from circular chance nodes, and triangular terminal nodes represent possible outcomes. A, Alternative screening approaches. B, Decision tree for individuals who received TST. LTBI indicates latent tuberculosis infection.

 
Screening Strategies Evaluated
The 3 screening alternatives were as follows. The first was universal screening, in which all children would receive TST before kindergarten entry. The second alternative was risk factor screening. Before kindergarten entry, all children would receive risk factor screening for tuberculosis with the RQ recommended by the PTCG. Children with positive results for ≥1 risk factor would receive TST. The third alternative was no screening. Before kindergarten entry, TST would not be administered unless a provider suspected tuberculosis (eg, the child had symptoms of tuberculosis disease or the child was identified as a contact of a patient with an active case of tuberculosis).

We made the following clinical assumptions. (1) All children entering kindergarten are between 4.5 and 6 years of age. (2) Screening would not identify any prekindergarten children with active tuberculosis disease beyond those who would have been identified clinically. (3) Positive TST results for children 4.5 to 6 years of age would not result in a source case investigation (consistent with current national guidelines).18 (4) Tuberculosis disease from 5 to 10 years of age would not initiate a contact investigation. (5) Isoniazid therapy would reduce the incidence of tuberculosis by 70%.3,19–22

Model Input Estimation and Sensitivity Analyses
Estimates for model inputs were obtained from the medical literature and from data provided from selected California health jurisdictions. The medical literature was searched by using Medline, manual searches of references in available literature articles, and personal questioning of tuberculosis control experts. All primary sources were obtained and examined, with questioning of authors if available.

We established a base-case scenario by using the best estimates available for each parameter. For each input, we then determined a plausible range for sensitivity analyses, on the basis of high and low estimates found in the relevant literature. We conducted 1-way sensitivity analyses for each of 4 key model inputs. We also conducted 2-way sensitivity analyses for selected pairs of inputs. The probabilities used at baseline, the minimal and maximal values for each parameter from the medical literature, and the plausible ranges used for sensitivity analyses are presented in Table 1. The rationale for the choice of baseline values is described in the Appendix.


View this table:
[in this window]
[in a new window]

  		TABLE 1 Tuberculosis-Associated Probabilities From Available Literature Reports and Range for Sensitivity Analysis

 
We used thresholds for cost-effectiveness of $50000 and $100000. These thresholds were chosen because $50000 per quality-adjusted life year (QALY) saved is used frequently as an appropriate cost-effectiveness threshold for clinical and preventive interventions,23,24 and we estimated that, on the basis of expected mortality rates for our cohort, individuals who developed tuberculosis disease would lose between 1 and 2 QALYs. We also used $100000 per QALY because there is substantial evidence that the traditional $50000 threshold may be too low.25

Published reports of the specificity of TST range from 90% to 99%.26–29 However, this range was inconsistent with our base-case parameters. Our base-case parameter for the prevalence of TST positivity in the risk factor–negative population was 0.37%, which implies a specificity of not less than 99.6% in that group. The true specificity of TST in our population is not known. However, with higher assumed specificity, any screening intervention seems more cost-effective. Therefore, to allow comparison with current practice in the most conservative manner, we assumed as follows. For our base-case scenario, we postulated that TST had different test characteristics in these 2 populations because of different exposure to BCG vaccine and atypical mycobacteria, both of which may tend to increase false-positive rates and are more common in areas from which individuals tend to immigrate to California.30–32 We then used the highest published specificity, 99%, for the risk factor–positive population and assumed a higher specificity of 99.9% for the risk factor–negative population. We then performed a sensitivity analysis by using a high specificity of 99.9% for both populations.

Costs
Cost estimates are presented in 2004 US dollars. Major costs are summarized in Table 2. A detailed description of cost derivation is available from the corresponding author.


View this table:
[in this window]
[in a new window]

  		TABLE 2 Cost Inputs for Decision Analysis

 
Costs of TST
The cost of a tuberculin skin test was assumed to be $10.21, based on 2003 Medicare Part B Extract Summary System (BESS) data.33 We made a conservative estimate that 5% must be repeated, leading to an expected cost per patient tested of $10.73. We assumed that an average of 10 minutes of licensed practical or licensed vocational nurse time would be needed to read, to interpret, and to chart the TST results, and we obtained data on estimated wage and fringe benefits from the Bureau of Labor Statistics.34 Total costs were $14.29 for each patient who had a TST both placed and read.

Costs of Risk Factor Screening
The RQ was assumed to require 2 minutes of licensed practical nurse time and 0.5 minutes of physician time for review of results, leading to a total cost of $1.41.

Costs of Evaluation and Treatment of TST-Positive Cases Identified Through Screening
For each positive TST result in the prekindergarten screening cohort, we assumed 1 medical visit and 1 chest radiograph, at costs based on BESS data.33 We assumed that all those who completed therapy would use 9 months of isoniazid, at an estimated medication cost based on 2004 Red Book wholesale prices,35 and that those who did not complete therapy would receive 2 months of isoniazid. We assumed 10 minutes of time for nurse management of refills and symptoms. Isoniazid-associated hepatitis is a very rare event in this age group. A health department-based study found no cases of hepatotoxicity among 1468 children 0 to 14 years of age,36 and no case report of significant toxicity from this cause in children <7 years of age could be found in the literature.37–40 Therefore, our model did not include any costs for treatment of hepatotoxicity.

Costs of Active Tuberculosis Disease
We estimated the average cost for a future case of tuberculosis, for age groups 5 to 14 years and 15 to 24 years. The components of this cost were based on expert opinion regarding the standard of care, although we note that there might be variation among practitioners. Included were costs for medical doctor visits estimated by using BESS data,33 costs for antimycobacterial drugs estimated by using the 2004 Red Book,35 costs for sputum assessments estimated by using Current Procedural Terminology codes, and costs for radiology and other laboratory studies estimated by using BESS data.33 Nurse case management time was estimated by using Bureau of Labor Statistics data. Directly observed therapy costs were calculated by averaging estimates from 3 articles.41–43 Hospitalization costs were calculated from the Centers for Disease Control and Prevention Cost of Hospitalization Study, which includes all costs of hospitalization for tuberculosis according to age group.44 Costs of contact investigation for those developing active tuberculosis in adulthood were derived from a California cohort of investigations in 1999 to 2000.45 Costs for contacts were derived by using the aforementioned estimates and data from the US Public Health Service study46 and the published literature regarding isoniazid toxicity in adults.36,47,48

Survey
To determine the scope of possible changes in screening policy, we surveyed all 61 California health jurisdictions regarding whether TST was required for school entry at kindergarten or at any other time. We asked jurisdictions that had access to TST results from prekindergarten screening to provide summary information, including data on place of birth and incidence of TST positivity. Population estimates for California children 5 to 9 years of age were obtained from the California Department of Finance for 2004.49 The survey instrument is available from the corresponding author.


   RESULTS
TOP
 ABSTRACT
 METHODS
 RESULTS
DISCUSSION
 APPENDIX: CHOICE OF BASELINE...
 REFERENCES
 
Base-Case Results
Table 3 shows costs and projected disease outcomes for universal testing, TT using the RQ, and no mandatory screening, with the assumptions of our base-case scenario. Net total health care system costs per 100000 children were $61393 for no screening, $1029532 for TT, and $1531294 for universal screening. For 100000 individuals presenting at prekindergarten age, 6.5, 4.7, and 3.9 cases are expected to arise with no screening, TT, and universal screening, respectively.


View this table:
[in this window]
[in a new window]

  		TABLE 3 Base-Case Results for 100 000 Children Screened

 
In this base-case scenario, TT would avert 1.8 cases of tuberculosis per 100000 children screened, compared with no screening, and universal screening would avert an additional 0.74 cases of tuberculosis per 100000 children screened, compared with TT. The cost to prevent an additional case of tuberculosis with TT, as opposed to no screening, was $524897. The cost to prevent an additional case of tuberculosis with universal screening, instead of TT, was $671398.

Of 100000 children screened, no children would begin isoniazid prophylaxis with the no-screening alternative, 795 children would begin isoniazid prophylaxis with TT, and 1170 children would begin isoniazid prophylaxis with universal screening. Because 795 children would begin isoniazid prophylaxis with TT who would not have received isoniazid with no screening, and because TT would avert 1.8 cases of tuberculosis, compared with no screening, 440 children would be treated with isoniazid to prevent 1 case of tuberculosis if TT were used instead of no screening. Because 375 children would begin isoniazid prophylaxis with universal screening who would not have received isoniazid with TT, and because universal screening would avert 0.74 cases of tuberculosis, compared with TT, 507 children would begin isoniazid prophylaxis to prevent 1 additional case of tuberculosis if universal screening were used instead of TT.

Sensitivity Analyses
The incremental cost of using universal testing instead of TT was very sensitive to variations in the probability of TST positivity in the risk factor–negative population. Varying this parameter over a plausible range resulted in estimates of $316054 to $11804901 per case of tuberculosis prevented by using universal testing instead of TT (Fig 2A). The incremental cost-effectiveness of using TT instead of no screening was very sensitive to variations in the probability of TST positivity in the risk factor–positive population. Varying the probability of TST positivity over a plausible range resulted in estimates of $36984 to $653547 per case of tuberculosis prevented by using risk factor screening instead of no screening (Fig 2B).


Figure 2
View larger version (25K):
[in this window]
[in a new window]

  		FIGURE 2 Univariate sensitivity analyses. A, Sensitivity analysis of cost-effectiveness of universal screening, compared with TT, over a plausible range of prevalence of TST positivity in a risk factor–negative population. B, Sensitivity analysis of cost-effectiveness of TT, compared with no screening, over a plausible range of prevalence of TST positivity in a risk factor–positive population. C, Sensitivity analysis of cost-effectiveness of universal screening, compared with TT, and of TT, compared with no screening, over a plausible range of risk factor prevalence. D, Sensitivity analysis of cost-effectiveness of screening over a plausible range of risk of progression to disease. Each contour line represents cost per case averted x 104.

 
Both the incremental cost-effectiveness of universal screening in comparison with TT and the incremental cost-effectiveness of TT in comparison with no screening were somewhat sensitive to changes in estimates of the prevalence of risk factor positivity (Fig 2C). Both estimates of incremental cost-effectiveness were very sensitive to changes in estimates of the risk of progression to disease (Fig 2D).

In a sensitivity analysis that assumed that the specificity of the test was 99.9% for both the risk factor–negative and risk factor–positive populations, TT was shown to avert 4.6 cases of tuberculosis per 100000 children screened, compared with no screening, at a total cost of $941965. Therefore, TT had an incremental cost-effectiveness of $204175 per case of tuberculosis averted, with these assumptions about specificity. The incremental cost-effectiveness of universal screening, compared with TT, was unchanged in this sensitivity analysis.

Threshold Analysis
The incremental cost of using universal testing instead of TT to prevent an additional case of tuberculosis did not fall below $100000 when any plausible range for risk factor prevalence was combined with any plausible range for the prevalence of TST positivity among the risk factor–negative children. The incremental cost-effectiveness of using TT instead of no screening fell below $50000 when the prevalence of risk factor positivity was ≥18% and the prevalence of TST positivity among the risk factor–positive children was ≥13%. The incremental cost of using TT instead of no screening fell below $100000 when the prevalence of risk factor positivity was ≥10% and the prevalence of TST positivity among the risk factor–positive children was ≥7% (Fig 3).


Figure 3
View larger version (13K):
[in this window]
[in a new window]

  		FIGURE 3 Contour map for threshold analysis of the effect of simultaneous variation in the prevalence of risk factor positivity and the prevalence of TST positivity, given risk factor positivity, on incremental costs per case averted, comparing no screening and TT.

 
Survey Results
Forty-two (69%) of 61 California health jurisdictions responded to the survey. Of those that responded, 16 (38%) required TST before kindergarten entry. The counties that required TST before school entry included the largest health jurisdictions. Population estimates for California showed that, in 2004, ≥51% of California children overall lived in health jurisdictions that required TST before kindergarten entry and ~252405 children entered kindergarten in areas of California that required universal TST.


   DISCUSSION
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
APPENDIX: CHOICE OF BASELINE...
 REFERENCES
 
Our results show that a majority of California children live in jurisdictions that mandate universal screening for tuberculosis before kindergarten entry and that the incremental cost of universal screening, compared with TT, for prevention of tuberculosis disease is very high. Under our base-case scenario, the incremental cost of using universal screening in place of TT is $671398 per tuberculosis case prevented, both for the screened child and through secondary transmission. On the basis of our survey results indicating that ≥252405 children 5 years of age live in jurisdictions that require universal testing, each year of using TT instead of currently mandated universal testing would save California $1.27 million. Each year of using TT in place of currently mandated universal testing would result in only 1.89 additional cases of tuberculosis over the subsequent 20 years. The $1.27 million in annual savings would be better invested in more cost-effective strategies to prevent tuberculosis.

Our results support strongly the recommendations of the PTCG for discontinuing universal TST of children. The incremental cost of universal screening is very high under our base-case scenario and remains quite high over a wide range of plausible circumstances. However, our results also suggest that the RQ recommended by the PTCG is not optimal for our cohort. The positive predictive value of the recommended RQ for predicting positive TST results in our base-case population is only 1.59%, and our study shows that TT remains costly, compared with no screening, when the prevalence of TST positivity in the population screened is relatively low. Foreign birth has been shown to be the strongest predictor of positive TST results for children without clinical risk factors for tuberculosis.5,7 If foreign birth were the only question used to predict risk factor status, then TT might prove to be more cost-effective than it is using the currently recommended RQ. In Santa Clara and Los Angeles, for example, ~15% to 18% of foreign-born kindergarteners have positive TST results and 7% of the kindergarteners are foreign born.12,50 Threshold analyses show that the incremental cost-effectiveness of TT, compared with no screening, falls below $100000 when these estimates are used for the prevalence of foreign birth and the prevalence of TST positivity among foreign-born children.

Therefore, improving the risk factor screening tool may make TT more cost-effective. The choice of age group for screening may also have a strong impact on cost-effectiveness. There are 3 reasons why screening of kindergarteners may be much less cost-effective than screening of older age groups. First, the prevalence of latent tuberculosis infection is relatively low among kindergarteners, compared with older school-aged children. Our sensitivity analyses indicate that both universal screening and TT are much more cost-effective when the prevalence of TST positivity is higher. Second, the risk of transmission from subjects who develop active tuberculosis during childhood is much lower than the risk of transmission from subjects who develop tuberculosis during adolescence or later in adulthood, leading to an overall low rate of secondary transmission for kindergarteners.51 Our estimates of secondary transmission would be higher if our cohort were older. Third, 5 years of age marks the beginning of the "favored age," in which children are less likely to develop tuberculosis disease.51 Our sensitivity analysis of the risk of progression to disease shows that incremental costs per case of tuberculosis averted are lower in populations with a higher risk of progression to disease. Our results suggest that tuberculosis screening would be more cost-effective in older age groups, in which the prevalence of TST positivity, the risk of developing active disease, and the risk of transmitting disease are higher. However, additional research is needed to determine whether increased nonadherence and hepatotoxicity in older children outweigh the benefits of later screening.10,37,40

Our results differ from those obtained in Santa Clara with a screening cohort from 1992, in which TT was found to be cost-effective.12 In contrast to our work, the Santa Clara study used a single question regarding place of birth to determine risk status and the prevalence of TST positivity in their foreign-born kindergartners was 18%, which would improve cost-effectiveness substantially. In addition, our study used a 20-year time horizon, whereas the Santa Clara analysis used a lifetime horizon, although the impact is lessened by the 3% annual rate of discount used in both studies.

There are several important limitations to our study. First, the risk of progression to disease and the specificity of TST are both parameters that are challenging to estimate for US children. However, examination of a wide range of plausible estimates for these parameters showed no incremental cost-effectiveness less than $390000 per tuberculosis case averted. Second, our model assumes that universal TST or TT detects no cases of active tuberculosis disease in prekindergarteners and that children with active tuberculosis have symptoms that prompt their clinicians to perform TST at the time of their prekindergarten well-child checks. One universal TST program identified no active cases,12 but another detected an average of 0.2 cases per 100000 kindergarteners screened per year.50 If these cases would not have been identified clinically in routine well-child checks, then the incremental cost-effectiveness of screening would be somewhat lower than we report. Third, this model did not incorporate QALYs, because of the paucity of data on QALYs associated with either tuberculosis disease in childhood or isoniazid treatment for kindergarteners. Additional research in this area is needed.

With little economic or clinical argument to support universal prekindergarten TST, this practice has outlived its historical utility. Most, if not all, school districts in California fail to meet the threshold for cost-effective tuberculosis screening with either universal screening or TT using the RQ recommended by the PTCG. We recommend that universal TST before kindergarten entry be discontinued in all California school districts and that TT be considered only where and when the prevalence of risk factor positivity and the prevalence of TST positivity among risk factor–positive individuals are high enough to meet acceptable thresholds for cost-effectiveness. Savings should be redirected to more cost-effective methods to prevent tuberculosis. Screening with TT may be cost-effective in middle school and high school, and districts may wish to consider screening for those age groups while awaiting additional research regarding the optimal age for screening. Our evidence suggests that screening of immigrants may be very cost-effective, and additional research is needed to explore the cost-effectiveness of replacing the RQ recommended by the PTCG with a single question regarding county of birth. Screening at an older age with a tool with higher positive predictive value may yield most of the benefits of prekindergarten TT at a much lower cost.


   APPENDIX: CHOICE OF BASELINE VALUES FOR MODEL INPUT ESTIMATION
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 APPENDIX: CHOICE OF BASELINE...
REFERENCES
 
The baseline estimate for the proportion of risk factor–positive children comes from Northern California Kaiser Permanente.7 The baseline estimate for the proportion of risk factor–positive children with positive TST results comes from the same cohort.7 The minimal value for this parameter comes from North Carolina,52 and the maximal value comes from Santa Clara12 and from Los Angeles.53 In the latter 2 cohorts, only foreign-born subjects were identified as risk factor positive. Because the positive predictive value of foreign birth for the outcome of positive TST results is higher than that of other questions on the RQ, this maximal value may be an overestimate of this parameter for the purposes of our analysis. The baseline estimate for the proportion of risk factor–negative children who have positive TST results comes from the Kaiser Permanente cohort.7 The minimal value comes from the South Bronx,8 and the maximal value comes from Los Angeles County data examining the proportion of United States-born children entering kindergarten who were found to have positive TST results.53 Because some United States-born children would have other risk factors and would screen risk factor positive on the RQ recommended by the PTCG, this maximal value may be an overestimate. The baseline estimate for the probability of treatment completion comes from San Diego.54 The maximal estimate for this parameter comes from Maryland.55

The risk of progression to disease is very difficult to estimate for current US children. Available estimates identified from reviews22,56 and from examination of the literature come from historical populations,57,58 from largely adult populations,59–61 or from cohorts weighted heavily with very young children.62 Because routine isoniazid therapy was initiated for US populations in 1952, there is no recent US estimate for the risk of progression to disease in this age group.63 Our baseline estimate is the most widely cited estimate for this parameter, that is, 90.2 cases per 100000 annually without discounting, from a Puerto Rican cohort enrolled in 1949 to 1951 and monitored for 18 to 20 years.57 Our minimal estimate for this parameter comes from a Hong Kong cohort.64 An alternative figure of 55.7 cases per 100000 annually without discounting was found in 1924 to 1934 in Massachusetts.58


   ACKNOWLEDGMENTS
 
This publication was made possible by grant KL2 RR024130 from the National Center for Research Resources, a component of the National Institutes of Health and the National Institutes of Health Roadmap for Medical Research.


   FOOTNOTES
 
Accepted Feb 28, 2007.

Address correspondence to Valerie J. Flaherman, MD, MPH, University of California San Francisco, 3333 California St, San Francisco, CA 94118. E-mail: flahermanv{at}peds.ucsf.edu

The contents of this article are solely the responsibility of the authors and do not necessarily represent the official view of the National Center for Research Resources or the National Institutes of Health.

The authors have indicated they have no financial relationships relevant to this article to disclose.


   REFERENCES
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 APPENDIX: CHOICE OF BASELINE...
 REFERENCES
 

  1. American Academy of Pediatrics, Committee on Infectious Diseases. Red Book: Report of the Committee on Infectious Diseases. Elk Grove Village, IL: American Academy of Pediatrics; 2003
  2. US Preventive Services Task Force. Screening for tuberculosis infection. Available at: www.ahrq.gov/clinic/uspstf/uspstubr.htm. Accessed July 1, 2006
  3. American Thoracic Society. Targeted tuberculin testing and treatment of latent tuberculosis infection. Am J Respir Crit Care Med. 2000;161 :S221 –S247[Free Full Text]
  4. Pediatric Tuberculosis Collaborative Group. Targeted tuberculin skin testing and treatment of latent tuberculosis infection in children and adolescents. Pediatrics. 2004;114 :1175 –1201[Abstract/Free Full Text]
  5. Saiman L, San Gabriel P, Schulte J, Vargas MP, Kenyon T, Onorato I. Risk factors for latent tuberculosis infection among children in New York City. Pediatrics. 2001;107 :999 –1003[Abstract/Free Full Text]
  6. Besser RE, Pakiz B, Schulte JM, et al. Risk factors for positive Mantoux tuberculin skin tests in children in San Diego, California: evidence for boosting and possible foodborne transmission. Pediatrics. 2001;108 :305 –310[Abstract/Free Full Text]
  7. Froehlich H, Ackerson LM, Morozumi PA. Targeted testing of children for tuberculosis: validation of a risk assessment questionnaire. Pediatrics. 2001;107(4) . Available at: www.pediatrics.org/cgi/content/full/107/4/e54
  8. Ozuah PO, Ozuah TP, Stein RE, Burton W, Mulvihill M. Evaluation of a risk assessment questionnaire used to target tuberculin skin testing in children. JAMA. 2001;285 :451 –453[Abstract/Free Full Text]
  9. Hsu K, Christiansen D, O'Connor D, Bernardo J, Hacker K. Self-assessment of tuberculosis infection risk by urban adolescents. Arch Pediatr Adolesc Med. 2003;157 :1227 –1231[Abstract/Free Full Text]
  10. Pong AL, Anders BJ, Moser KS, Starkey M, Gassmann A, Besser RE. Tuberculosis screening at 2 San Diego high schools with high-risk populations. Arch Pediatr Adolesc Med. 1998;152 :646 –650[Abstract/Free Full Text]
  11. Lobato MN, Hopewell PC. Mycobacterium tuberculosis infection after travel to or contact with visitors from countries with a high prevalence of tuberculosis. Am J Respir Crit Care Med. 1998;158 :1871 –1875[Abstract/Free Full Text]
  12. Mohle-Boetani JC, Miller B, Halpern M, et al. School-based screening for tuberculous infection: a cost-benefit analysis. JAMA. 1995;274 :613 –619[Abstract]
  13. Gounder CR, Driver CR, Scholten JN, Shen H, Munsiff SS. Tuberculin testing and risk of tuberculosis infection among New York City schoolchildren. Pediatrics. 2003;111(4) . Available at: www.pediatrics.org/cgi/content/full/111/4/e309
  14. Driver CR, Valway SE, Cantwell MF, Onorato IM. Tuberculin skin test screening in schoolchildren in the United States. Pediatrics. 1996;98 :97 –102[Abstract/Free Full Text]
  15. Miller TL, Hilsenrath P, Lykens K, McNabb SJ, Moonan PK, Weis SE. Using cost and health impacts to prioritize the targeted testing of tuberculosis in the United States. Ann Epidemiol. 2006;16 :305 –312[CrossRef][ISI][Medline]
  16. US Department of Labor, Bureau of Labor Statistics. Consumer Price Index. Available at: www.bls.gov/cpi/home.htm. Accessed May 5, 2006
  17. Steele GL. Common Lisp: The Language. 2nd ed. Bedford, MA: Digital Press; 1990
  18. National Tuberculosis Controllers Association. Guidelines for the investigation of contacts of persons with infectious tuberculosis. MMWR Recomm Rep. 2005;54 (RR-15):1–47
  19. Comstock GW. How much isoniazid is needed for prevention of tuberculosis among immunocompetent adults? Int J Tuberc Lung Dis. 1999;3 :847 –850[ISI][Medline]
  20. International Union Against Tuberculosis Committee on Prophylaxis. Efficacy of various durations of isoniazid preventive therapy for tuberculosis: five years of follow-up in the IUAT trial. Bull World Health Organ. 1982;60 :555 –564[ISI][Medline]
  21. Dasgupta K, Schwartzman K, Marchand R, Tennenbaum TN, Brassard P, Menzies D. Comparison of cost-effectiveness of tuberculosis screening of close contacts and foreign-born populations. Am J Respir Crit Care Med. 2000;162 :2079 –2086[Abstract/Free Full Text]
  22. Ferebee SH. Controlled chemoprophylaxis trials in tuberculosis: a general review. Bibl Tuberc. 1970;26 :28 –106[Medline]
  23. Hirth RA, Chernew ME, Miller E, Fendrick AM, Weissert WG. Willingness to pay for a quality-adjusted life year: in search of a standard. Med Decis Making. 2000;20 :332 –342[Abstract/Free Full Text]
  24. Yazdanpanah Y. Costs associated with combination antiretroviral therapy in HIV-infected patients. J Antimicrob Chemother. 2004;53 :558 –561[Abstract/Free Full Text]
  25. Ubel PA, Hirth RA, Chernew ME, Fendrick AM. What is the price of life? And why doesn't it increase with inflation? Arch Intern Med. 2003;163 :1637 –1641
  26. Villarino ME, Burman W, Wang YC, et al. Comparable specificity of 2 commercial tuberculin reagents in persons at low risk for tuberculous infection. JAMA. 1999;281 :169 –171[Abstract/Free Full Text]
  27. Rose DN, Schechter CB, Adler JJ. Interpretation of the tuberculin skin test. J Gen Intern Med. 1995;10 :635 –642[ISI][Medline]
  28. Berkel GM, Cobelens FG, de Vries G, Draayer-Jansen IW, Borgdorff MW. Tuberculin skin test: estimation of positive and negative predictive values from routine data. Int J Tuberc Lung Dis. 2005;9 :310 –316[ISI][Medline]
  29. Huebner RE, Schein MF, Cauthen GM, et al. Evaluation of the clinical usefulness of mycobacterial skin test antigens in adults with pulmonary mycobacterioses. Am Rev Respir Dis. 1992;145 :1160 –1166[ISI][Medline]
  30. Sepulveda RL, Burr C, Ferrer X, Sorensen RU. Booster effect of tuberculin testing in healthy 6-year-old school children vaccinated with bacillus Calmette-Guerin at birth in Santiago, Chile. Pediatr Infect Dis J. 1988;7 :578 –581[ISI][Medline]
  31. Tissot F, Zanetti G, Francioli P, Zellweger JP, Zysset F. Influence of bacille Calmette-Guerin vaccination on size of tuberculin skin test reaction: to what size? Clin Infect Dis. 2005;40 :211 –217[CrossRef][ISI][Medline]
  32. Lifschitz M. The value of the tuberculin skin test as a screening test for tuberculosis among BCG-vaccinated children. Pediatrics. 1965;36 :624 –627[Abstract/Free Full Text]
  33. US Department of Health and Human Services. Part B Extract Summary System (BESS) data file. Available at: www.cms.hhs.gov/NonIdentifiableDataFiles/03_PartBExtractSummarySystem.asp. Accessed June 25, 2006
  34. US Department of Labor. November 2004 employment and wage estimates. Available at: www.bls.gov/oes/oes_2004_n.htm. Accessed May 6, 2006
  35. Thomson Corp. Red Book: Pharmacy's Fundamental Reference. Montvale, NJ: Thomson PDR; 2004;111
  36. Nolan CM, Goldberg SV, Buskin SE. Hepatotoxicity associated with isoniazid preventive therapy: a 7-year survey from a public health tuberculosis clinic. JAMA. 1999;281 :1014 –1018[Abstract/Free Full Text]
  37. Kopanoff DE, Snider DE Jr, Caras GJ. Isoniazid-related hepatitis: a US Public Health Service cooperative surveillance study. Am Rev Respir Dis. 1978;117 :991 –1001[ISI][Medline]
  38. Millard PS, Wilcosky TC, Reade-Christopher SJ, Weber DJ. Isoniazid-related fatal hepatitis. West J Med. 1996;164 :486 –491[ISI][Medline]
  39. O'Brien RJ, Long MW, Cross FS, Lyle MA, Snider DE Jr. Hepatotoxicity from isoniazid and rifampin among children treated for tuberculosis. Pediatrics. 1983;72 :491 –499[Abstract/Free Full Text]
  40. Stein MT, Liang D. Clinical hepatotoxicity of isoniazid in children. Pediatrics. 1979;64 :499 –505[Abstract/Free Full Text]
  41. Moore RD, Chaulk CP, Griffiths R, Cavalcante S, Chaisson RE. Cost-effectiveness of directly observed versus self-administered therapy for tuberculosis. Am J Respir Crit Care Med. 1996;154 :1013 –1019[Abstract]
  42. Snyder DC, Chin DP. Cost-effectiveness analysis of directly observed therapy for patients with tuberculosis at low risk for treatment default. Am J Respir Crit Care Med. 1999;160 :582 –586[Abstract/Free Full Text]
  43. Burman WJ, Dalton CB, Cohn DL, Butler JR, Reves RR. A cost-effectiveness analysis of directly observed therapy vs self-administered therapy for treatment of tuberculosis. Chest. 1997;112 :63 –70[ISI][Medline]
  44. Taylor Z, Marks SM, Rios Burrows NM, Weis SE, Stricof RL, Miller B. Causes and costs of hospitalization of tuberculosis patients in the United States. Int J Tuberc Lung Dis. 2000;4 :931 –939[ISI][Medline]
  45. Sprinson JE, Flood J, Fan CS, et al. Evaluation of tuberculosis contact investigations in California. Int J Tuberc Lung Dis. 2003;7(suppl 3) :S363 –S368
  46. Daley CL. Tuberculosis latency in humans. In: Rom WN, Garay SM, eds. Tuberculosis. 2nd ed. Philadelphia, PA: Lippincott Williams & Williams; 2004:85–99
  47. McElroy PD, Ijaz K, Lambert LA, et al. National survey to measure rates of liver injury, hospitalization, and death associated with rifampin and pyrazinamide for latent tuberculosis infection. Clin Infect Dis. 2005;41 :1125 –1133[CrossRef][ISI][Medline]
  48. Lichtenberg F. The Effect of Using Newer Drugs on Admissions of Elderly Americans to Hospitals and Nursing Homes: State-Level Evidence From 1997–2003. New York, NY: Columbia University and National Bureau of Economic Research; 2005
  49. State of California, Department of Finance. Race/Ethnic Population With Age and Sex Detail, 2000–2050. Sacramento, CA: State of California, Department of Finance; 2004
  50. County of Los Angeles. TB Cases Identified Through School Mandate TB Screening Activities in LAC Public Health Centers for Mandate Years 93–94 through 98–99. Los Angeles, CA: County of Los Angeles; 2001
  51. Starke JR. Tuberculosis in children. Semin Respir Crit Care Med. 2004;25 :353 –364[CrossRef][ISI][Medline]
  52. Stout JE, Nageswaran S, Tilson E. Effectiveness of a tuberculosis screening questionnaire in identifying children with latent tuberculosis infection. Presented at: annual meeting of the American Thoracic Society; May 22, 2005; San Diego, CA
  53. County of Los Angeles. Analysis of Results From the Mandated School TB Skin Test, 2004–2005, All Schools. Los Angeles, CA: County of Los Angeles; 2005
  54. LoBue PA, Moser KS. Use of isoniazid for latent tuberculosis infection in a public health clinic. Am J Respir Crit Care Med. 2003;168 :443 –447[Abstract/Free Full Text]
  55. Chang S, Wheeler LS, Farrell KP. Public health impact of targeted tuberculosis screening in public schools. Am J Public Health. 2002;92 :1942 –1945[Free Full Text]
  56. Watkins RE, Brennan R, Plant AJ. Tuberculin reactivity and the risk of tuberculosis: a review. Int J Tuberc Lung Dis. 2000;4 :895 –903[ISI][Medline]
  57. Comstock GW, Livesay VT, Woolpert SF. The prognosis of a positive tuberculin reaction in childhood and adolescence. Am J Epidemiol. 1974;99 :131 –138[Abstract/Free Full Text]
  58. Pope AS, Sartwell PE, Zacks D. Development of tuberculosis in infected children. Tuberculosis. 1939;29 :1318 –1326
  59. Nolan CM, Elarth AM. Tuberculosis in a cohort of Southeast Asian refugees: a five-year surveillance study. Am Rev Respir Dis. 1988;137 :805 –809[ISI][Medline]
  60. Fine PE, Sterne JA, Ponnighaus JM, Rees RJ. Delayed-type hypersensitivity, mycobacterial vaccines and protective immunity. Lancet. 1994;344 :1245 –1249[CrossRef][ISI][Medline]
  61. Comstock GW, Ferebee SH, Hammes LM. A controlled trial of community-wide isoniazid prophylaxis in Alaska. Am Rev Respir Dis. 1967;95 :935 –943[ISI][Medline]
  62. Ferebee S, Mount FW, Anastasiades A. Prophylactic effects of isoniazid on primary tuberculosis in children: a preliminary report. Am Rev Tuberc. 1957;76 :942 –963[ISI][Medline]
  63. Hsu KH. Thirty years after isoniazid: its impact on tuberculosis in children and adolescents. JAMA. 1984;251 :1283 –1285[Abstract]
  64. Leung CC, Yew WW, Chang KC, et al. Risk of active tuberculosis among schoolchildren in Hong Kong. Arch Pediatr Adolesc Med. 2006;160 :247 –251[Abstract/Free Full Text]
PEDIATRICS (ISSN 1098-4275). ©2007 by the American Academy of Pediatrics



This article has been cited by other articles:


Home page
 PediatricsHome page
M. N. Lobato, J. A. Jereb, and J. R. Starke
Unintended Consequences: Mandatory Tuberculin Skin Testing and Severe Isoniazid Hepatotoxicity
Pediatrics, June 1, 2008; 121(6): e1732 - e1733.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow P3Rs: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when P3Rs are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Flaherman, V. J.
Right arrow Articles by Royce, S. E.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Flaherman, V. J.
Right arrow Articles by Royce, S. E.
Related Collections
Right arrow Infectious Disease & Immunity