Alternative and Complementary Approaches to the Diagnosis and Treatment of Active Disease
The limitations of the DOTS strategy have stimulated numerous initiatives to improve program performance (including treatment protocols for patients carrying drug-resistant bacilli or who are infected with HIV), active case finding, collaborations within and between public and private sector health services, schemes for outpatient and community-based treatment, and integration of the management of TB and other illnesses.
Management of Drug-resistant Disease
The higher the proportion of patients carrying drug-resistant bacilli is, the greater the need for accurate resistance testing and for the provision of alternative regimens that include at least three drugs to which bacilli are fully susceptible. Of greatest importance is resistance to the two principal first-line drugs, isoniazid and rifampicin (that is, MDR-TB). The introduction of resistance testing, second-line drugs, longer treatment regimens (12 to 18 months), and rigorous bacteriological and clinical monitoring all increase program costs without necessarily ensuring high cure rates (equal to or greater than 85 percent). Indeed, achieving the same cure rates for MDR-TB patients as for patients carrying fully susceptible strains may not be possible. The cost-effectiveness of this component of a TB control program is therefore lower by an amount that depends on the nature of the resistance, the methods of testing and monitoring, and the choice of regimen. The higher costs and lower cure rates associated with treating drug-resistant TB are part of the argument for preventing the spread of resistance in the first place, as can be investigated with models of selection and transmission (Dye and Espinal 2001; Dye and others 2002; Dye and Williams 2000). Suarez and others (2002) have investigated the cost-effectiveness of managing drug-resistant TB in Peru, but because studies in other settings have yet to be published, an empirical overview is not yet possible. Further data will be available from studies in Estonia, the Philippines, and Russia in 2005.
Treatment of HIV Coinfection
Antiretroviral therapy for HIV-positive individuals is unlikely to prevent a large fraction of TB cases unless treatment can be given shortly after HIV infection is acquired (Sonnenberg and others 2005; Williams and Dye 2003). In general, antiretroviral therapy is likely to be most effective, not in reducing TB incidence, but in extending the life expectancy of HIV-positive patients successfully treated for TB (Friedland and others 2004). Antiretroviral therapy and DOTS are formally synergistic, because without undergoing both together, HIV-infected TB patients have a short life expectancy, typically less than five years.
Where the prevalence of HIV infection has been rising quickly, as in eastern and southern Africa, even the most energetic programs of TB chemotherapy may not be able to reverse the rise in TB incidence. However, mathematical modeling indicates that, even in the midst of a major HIV epidemic, early detection and cure are the most cost-effective ways of minimizing TB cases and deaths (Currie and others, 2005). One reason is that DOTS programs treat all TB cases, not just those linked with HIV. The alternatives—the prevention of HIV infection, TLTI, and antiretroviral therapy—are less promising strategies to control TB, at least for the coming decade, although they could be used in combination with DOTS.
Active Case Finding
The DOTS strategy is based on passive case detection for three reasons: (a) the majority of incipient TB cases develop active smear-positive, infectious disease more quickly than any reasonable interval between successive rounds of mass screening for TB symptoms or x-ray abnormalities; (b) the majority of patients severely ill with a life-threatening disease are likely to seek help quickly (Toman 1979); and (c) countries that have not yet implemented effective systems for passive case detection are not in a position to pursue cases more actively. The drawback of passive case finding is that the delays to diagnosis among symptomatic patients are often long, and health services never see some patients. To shorten delays and increase the proportion of cases detected, studies of risk can identify subpopulations in which TB tends to be relatively common. Systematic surveys of these subpopulations for active TB may be logistically feasible and affordable. The target populations include individuals infected with HIV, refugees (Marks and others 2001), contacts of active cases (Claessens and others 2002; Noertjojo and others 2002), health workers (Cuhadaroglu and others 2002), and drug users and prisoners (Nyangulu and others 1997). Despite the practical possibilities and the potential effect on transmission (Murray and Salomon 1998), active case finding is rarely done in high-burden countries, where the emphasis is still on implementing the basic DOTS strategy.
Case Finding and Treatment in the Private Sector
It is well known that many TB patients first seek treatment from private practitioners and that diagnosis and treatment in the private sector often do not meet internationally accepted standards (Uplekar, Pathania, and Raviglione 2001). A new scheme to deliver DOTS through the private sector (Public-Private Mix DOTS) operates through the provision of free drugs, by information exchange and patient referral, and with some financial support from participating governments. Two pilot projects in Hyderabad and Delhi, India, improved case-detection rates by 26 percent and 47 percent, respectively, and maintained treatment success close to the target of 85 percent (WHO 2004b). Other such projects are under way elsewhere in India as well as in Bangladesh, Indonesia, Nepal, the Philippines, and Vietnam (WHO 2004d).
Outpatient and Community-based Treatment
Early studies of the cost-effectiveness of TB control found that full ambulatory treatment, eliminating hospitalization during the first two months (intensive phase), was cheaper and did not compromise cure rates (de Jonghe and others 1994; Murray and others 1991). Partly as a result, ambulatory treatment has become the standard of care in many high-burden countries. The natural extension, to home- and community-based treatment, has proved to be just as effective in several African settings, and even lower in cost (Adatu and others 2003; Dudley and others 2003; Floyd and others 2003; Floyd, Wilkinson, and Gilks 1997; Moalosi and others 2003; Okello and others 2003; Sinanovic and others 2003; Vassall and others 2002; Wilkinson, Floyd, and Gilks 1997). Various schemes have been used to provide TB care in the community, in which nongovernmental organizations, volunteers (Okello and others 2003), or appointed "guardians" (Floyd and others 2003) supervise treatment, sometimes with financial incentives (Sinanovic and others 2003). Consequently, community-based care is being adopted in some countries (for example, Uganda) as standard procedure.
Integrated Management of Tuberculosis and Other Respiratory Illnesses
Surveys in nine countries found that up to one-third of patients over five years of age attending primary health centers had respiratory symptoms, of whom 5 to 10 percent were TB suspects, but only 1 to 2 percent had TB (WHO 2004e). Because TB is rare among respiratory diseases, comanaging TB with other conditions has clear advantages. The purpose of the WHO's Practical Approach to Lung Health (PAL) project is to encourage a syndromic approach to management of patients, to standardize health service delivery through the development and implementation of clinical guidelines, and to promote the necessary coordination within national health services. Preliminary investigations in the Kyrgyz Republic and Morocco suggest that PAL projects can improve the accuracy of diagnosis, encourage better practice in prescribing drugs, and strengthen primary care. However, a full analysis of costs and effects in the nine-country study remains to be done.
