To the Editor: The report by Khatri and Frieden (Oct. 31 issue)1on tuberculosis control in India echoes the official line ofthe Indian government, health policy bureaucrats, and the WorldHealth Organization. Sadly, data collection in India cannotbe taken at face value, and the accuracy of the impressive curerates has been questioned.2 The 200,000 new health workers alludedto are but a small fraction of those required to take on theadditional burden imposed by direct observation. This shortageconstrains the Revised National Tuberculosis Control Program(RNTCP) to recommend direct observation of only 6 of the 18continuation-phase doses, and this incomplete supervision ata time when the illness is improving and the patient is leastcompliant has been dismissed as only partially observed therapy.
Marginalized persons (homeless persons, alcoholics, migrants,and drug abusers) are not enrolled lest they spoil neat quarterlycalculations. Thus, the most "successful" centers providingdirectly observed treatment, short course (DOTS) are also thosewith the highest rates of exclusion of potential patients.3
Finally, Khatri and Frieden underestimate the extent and theeffect of multidrug-resistant tuberculosis and the human immunodeficiencyvirus (HIV) on tuberculosis control. Multidrug-resistant tuberculosis,the main saboteur, occurs far more frequently than in 1 to 3percent of cases. At our referral mycobacterial laboratory inMumbai, 60 percent of all strains are multidrug-resistant.4The RNTCP has abjured itself of the responsibility of treatingthese patients who are the true "untouchables" of the Indianhealth care system. HIV-control programs and tuberculosis-controlprograms continue to be run in isolation, despite the fact thatIndia has the world's largest dually infected population.5
Zarir F. Udwadia, M.D. Hinduja Hospital and Research Center Mumbai 400 016, India zfu{at}vsnl.com
References
Khatri GR, Frieden TR. Controlling tuberculosis in India. N Engl J Med 2002;347:1420-1425. [Free Full Text]
Balasubramanian VN, Oommen K, Samuel R. DOT or not? Direct observation of anti-tuberculosis treatment and patient outcomes, Kerala State, India. Int J Tuberc Lung Dis 2000;4:409-413. [Web of Science][Medline]
Singh V, Jaiswal A, Porter JD, et al. TB control, poverty and vulnerability in Delhi, India. Trop Med Int Health 2002;7:693-700. [Medline]
Udwadia ZF, Hakimiyan A, Rodrigues C, Jillisgar T, Mehta A. A profile of drug-resistant tuberculosis in Bombay. Chest 1996;110:Suppl:228s-228s. abstract.
To the Editor: Khatri and Frieden omit mention of an importantpopulation group: children. As many as half of children livingin contact with adults who have tuberculosis may be infectedwith tuberculosis through their caregivers' coughs and breaths.Bacille Calmette–Guérin vaccination limits thedissemination of tuberculosis but does not prevent primary infection.Childhood tuberculosis often remains unrecognized. Diagnosison the basis of positive sputum smears is rarely possible, becausechildren produce little sputum. Children with tuberculosis havenonspecific symptoms, including fevers, malaise, and stunting.Only those with overt pulmonary or disseminated disease areeasily identified as having tuberculosis. Tracing of contactshas not been routine in much of the world, even though evaluationof and chemoprophylaxis in child contacts until they are provedto be uninfected represent the gold standard. We thus allowtubercle bacilli to survive in a large pool of infected children,which is especially disturbing in the light of the perpetuationof drug-resistant organisms. Furthermore, some 10 to 15 percentof children with unrecognized infection will have full-blownbut preventable tuberculosis in adulthood.
Solutions to these problems include incorporation of childhoodtuberculosis into national tuberculosis-control programs, developmentof contact-tracing programs, and exploration of the feasibilityof chemoprophylaxis for exposed children. Better diagnostictools for childhood tuberculosis are also needed.
Jane G. Schaller, M.D. Tufts University School of Medicine Boston, MA 02111 jschaller{at}lifespan.org
Jeffrey Starke, M.D. Baylor College of Medicine Houston, TX 77030
The authors reply: The Indian tuberculosis-control program hasnow treated more than 1.5 million patients and saved more than250,000 lives, but it is true that, in some areas, not all patientswith diagnosed tuberculosis are included in the program. Theprogram recommends that all patients, and certainly no lessthan 90 percent of them, be treated according to the DOTS strategy.The proportion of patients newly diagnosed who are treated inthe program has increased steadily to 94 percent. Contrary toUdwadia's assertion, areas that fail to enroll a high proportionof patients in the program tend to have lower cure rates —a reflection of weaker implementation of the program. For example,the area with the lowest proportion of patients with diagnosedtuberculosis who were treated in the program in the most recentquarter was also the area with the lowest cure rate.1 Directobservation is essential in the intensive phase of treatment,when the burden of organisms and the risks of treatment failureand development of drug resistance are highest; in the continuationphase, the program uses direct observation for at least thefirst of three doses per week.
The proportion of patients at a referral hospital who have multidrug-resistanttuberculosis has no relevance to the actual proportion of peoplein the community who have multidrug-resistant disease.2 Allvalid studies in India have found rates of multidrug resistanceof 1 to 3 percent among previously untreated patients.3,4,5,6Multidrug-resistant tuberculosis is a symptom of poor performanceof programs; the highest priority is to prevent multidrug-resistanttuberculosis by effective treatment. As we state in our article,more than 1 million patients each year with newly diagnosedtuberculosis do not yet have access to basic treatment, andthe top priority must be to ensure that they have such access.Both HIV-positive and HIV-negative patients are treated in theprogram; models to improve coordination are being evaluated.
With regard to the issues raised by Schaller and Starke, thenew program includes treatment of children with active diseaseas well as investigation of contacts and preventive treatmentof children who are contacts of those with infectious cases.These efforts have met with varying degrees of success in differentparts of the country.
There are several errors in our article. Clinical features areprovided in Table 2, not Table 1. On page 1422, the top lineof the right column should read "More than 200,000" rather than"Nearly 200,000." The sentence beginning on line 12 of thatcolumn should read "By September 2001, about 3.4 million symptomaticpatients had been assessed for tuberculosis, and in the caseof nearly 800,000, treatment had been started — in morethan half of them, within the previous 12 months."
G.R. Khatri, M.D., D.P.H. Ministry of Health and Family Welfare New Delhi 110 058, India
Thomas R. Frieden, M.D., M.P.H. New York City Department of Health and Mental Hygiene New York, NY 10013 tfrieden{at}health.nyc.gov
References
Tuberculosis control — India. New Delhi, India: Directorate General of Health Services, 2003. (Accessed January 16, 2003, at http://www.tbcindia.org.)
Guidelines for surveillance of drug resistance in tuberculosis. In: Index of /emc/amrpdfs. Geneva: World Health Organization and International Union Against Tuberculosis and Lung Disease, 1997. (WHO/TB/96.216.) (Accessed January 30, 2003, at http://www.who.int/emc/amrpdfs/.)
Paramasivan CS, Bhaskaran K, Venkataraman P, et al. Surveillance of drug resistance in tuberculosis in the state of Tamil Nadu. Indian J Tuberc 2000;47:27-33.
Espinal MA, Simonsen L, Laszlo A, et al. Anti-tuberculosis drug resistance in the world. Report no. 2: prevalence and trends. Geneva: World Health Organization, 2000:181. (WHO/CDS/TB/2000.278.)
Paramasivan CN, Venkataraman P, Chandrasekaran V, Bhat S, Narayanan PR. Surveillance of drug resistance in tuberculosis in two districts of South India. Int J Tuberc Lung Dis 2002;6:479-484. [Medline]
Mathur ML, Khatri PK, Base CS. Drug resistance in tuberculosis patients in Jodhpur district. Indian J Med Sci 2000;54:55-58. [Medline]
Correction
to
Khatri and Frieden, 
Previous
