Skip to main content

Impact of DOTS expansion on tuberculosis related outcomes and costs in Haiti



Implementation of the World Health Organization's DOTS strategy (Directly Observed Treatment Short-course therapy) can result in significant reduction in tuberculosis incidence. We estimated potential costs and benefits of DOTS expansion in Haiti from the government, and societal perspectives.


Using decision analysis incorporating multiple Markov processes (Markov modelling), we compared expected tuberculosis morbidity, mortality and costs in Haiti with DOTS expansion to reach all of the country, and achieve WHO benchmarks, or if the current situation did not change. Probabilities of tuberculosis related outcomes were derived from the published literature. Government health expenditures, patient and family costs were measured in direct surveys in Haiti and expressed in 2003 US$.


Starting in 2003, DOTS expansion in Haiti is anticipated to cost $4.2 million and result in 63,080 fewer tuberculosis cases, 53,120 fewer tuberculosis deaths, and net societal savings of $131 million, over 20 years. Current government spending for tuberculosis is high, relative to the per capita income, and would be only slightly lower with DOTS. Societal savings would begin within 4 years, and would be substantial in all scenarios considered, including higher HIV seroprevalence or drug resistance, unchanged incidence following DOTS expansion, or doubling of initial and ongoing costs for DOTS expansion.


A modest investment for DOTS expansion in Haiti would provide considerable humanitarian benefit by reducing tuberculosis-related morbidity, mortality and costs for patients and their families. These benefits, together with projected minimal Haitian government savings, argue strongly for donor support for DOTS expansion.

Peer Review reports


Between 1997 and 2002 the incidence of active TB increased in most low and middle income countries [1]. This occurred despite the availability of adequate tools for diagnosis and treatment, and an effective TB control strategy – which has been labelled DOTS. This strategy, originally developed in sub-Saharan Africa, is being promoted by the World Health Organisation (WHO) [2] because it is feasible in high-burden settings [3], cost-effective even in low income countries [4], and can result in substantial reduction in TB incidence [5].

However, between $290 – $500 million (US) is required annually for the additional training, equipment and infrastructure needed to implement and maintain DOTS in all low and middle income countries [6, 7]. As a result DOTS expansion has lagged considerably behind WHO targets [8].

In the America's, Haiti is the poorest country [9], with the highest incidence of smear positive pulmonary TB (138/100,000 in 2002) [8] and HIV seroprevalence (4.5% in 2002) [10]. In 2002 the WHO estimated that only 37% of the population had access to DOTS programmes [11], only 49% of all smear positive cases were diagnosed, and only 71% of those diagnosed were successfully treated [8]. We have compared projected TB related outcomes and costs if the current control programme is maintained (status quo), or if DOTS is expanded to reach WHO benchmarks, for 100% of the population in Haiti.


General description of model and the two strategies compared

We developed a decision analysis model, incorporating multiple Markov processes, (Markov modelling) using Tree-Age Pro Release 6.0, (Tree-age Inc., Williamstown, MA). This model calculated the probability of TB-related events expected to result over 20 years, starting in 2003, from two alternate strategies for TB control. These probabilities were calculated for a hypothetical fixed cohort with the age structure, socio-demographic and health characteristics and size of the population of Haiti in 2002. The information on the population of Haiti, summarized in Table 1, was taken from publicly accessible information provided by international agencies including the World Bank, WHO, and the United Nations Program on AIDS (UNAIDS) [810].

Table 1 Summary of key input data for Haiti in 2002(All costs in US dollars)

The strategies compared were to continue the present national TB control programme (status quo), or DOTS expansion. With the status quo strategy there would be no change over the 20 year time frame of the analysis, from January 2003 levels of: DOTS coverage [11], case finding [8], treatment outcomes [8], incidence of smear positive TB [8], prevalence of initial TB drug resistance [12, 13], and HIV seroprevalence [10]. Since incidence of disease did not change with the status quo, risk of TB infection [14], and LTBI prevalence also remained unchanged throughout the period of analysis. Current levels of DOTS coverage have been achieved with assistance from foreign donors – hence we implicitly accounted for current foreign assistance. However, we did not explicitly add in these expenditures, since they would have been added to both strategies, making both more expensive, but without changing the differences between them.

With DOTS expansion, we assumed that DOTS would be expanded from the level of coverage in January 2003, (Year 1) [11], to reach 100% of government health facilities by the end of Year 3. Case detection would increase from current WHO estimated levels [8] to 70%, and treatment success (cure and treatment completion) to 85% – the WHO targets [3], although treatment failure and TB mortality would not change [8]. Rates of initial drug resistance [12, 13] and HIV prevalence [10] would not change with DOTS over the full 20 years. DOTS implementation would halve pre-diagnostic health system visits [15, 16], and health system delays [1719], although patient delays would not change. Hospitalisation duration would decrease by two-thirds [15, 16]. The number of patients investigated for each new case of TB diagnosed would increase from 4 to 16 [20]. Retreatment failures (multi-drug resistant TB) would be treated with second line drugs obtained from the Green Light Committee [21] with 48% cure, and 12% mortality [22]. Most importantly DOTS expansion was assumed to result in a 6% annual decline in incidence, as described in Peru following national DOTS implementation [5]. This would produce corresponding reductions in the risk, and prevalence of TB infection [14].

Health states and transitional probabilities (see Figure 1)

Figure 1

Sample decision analysis tree for Haitian adult, initially without TB infection, nor HIV infection who acquires new TB infection.

At the start of Year 1 (assumed to be 2003), cohort members were considered to be in one of five TB-related states, and one of three HIV-related states. The TB-related states were: 1) no tuberculosis infection; 2) recent latent tuberculosis infection (LTBI) – acquired within 2 years; and, 3) long-standing LTBI – acquired more than 2 years ago; 4) active tuberculosis; and, 5) treated, or spontaneously resolved active TB. States 2–5 were further sub-classified into 3 groups: drug-sensitive, single-drug resistant, or multi-drug resistant TB – with likelihood based on surveys of drug resistance in Haitian populations [12, 13]. The proportion with recent or long-standing LTBI was calculated based on the age structure of the population [23], and incidence of smear-positive active disease [8], using the Styblo formula [14].

Key pathogenetic model assumptions regarding reactivation and cure rates for HIV-negative and HIV-positive individuals, summarized in Table 2, were based on published cohort studies and randomized trials. In HIV negative persons, active TB would develop in the first two years after new TB infection in 5% of the population [24]. In other studies the risk of disease in the first two years after new TB infection ranged from 4% [25] to 10% [26]. Given the 80% protective effect afforded by previous latent TB infection in HIV uninfected persons [27], re-infection would be followed by a 1% risk of active TB in the first two years. In latently infected persons risk would be 0.1% annually after the first two years – based on prospective follow-up of two tuberculin positive cohorts – of young military recruits [28], and Vietnamese refugees [29].

Table 2 Probabilities of outcomes with different TB and HIV health states

The probability of transfer out and default varied by strategy. We assumed that the treatment outcome of "transferred-out" was equivalent to default [30]. For individuals who defaulted from therapy we assumed an overall cure rate of 62%, calculated from the proportion of defaulters after different lengths of therapy [31], and cure rates in randomised trials of regimens of 3 or 4 months duration [3234].

The HIV-related states were: 1) no HIV infection; 2) early HIV, defined as having no clinical manifestations; or, 3) late HIV infection – defined as clinical AIDS. HIV-related survival and annual rates of transition from early to late HIV states were based on a Ugandan cohort [35]. Probability of acquiring HIV infection was the same in both strategies. Annual risk of HIV infection was estimated to be 0.49% – calculated from the general population prevalence [10] divided by the average years of survival in a low-income setting [35] (and thus assumed to be uniform for the entire population). We assumed no consequence of HIV re-infection.

The same model was used for HIV infected or uninfected. The proportion of HIV infected persons entering the model was 0.045 – corresponding to the estimated seroprevalence in Haiti in 2002 (10). Every year, 0.5% of the uninfected population could acquire new HIV infection, thus changing from an HIV uninfected to an HIV infected state. The major difference in the model for HIV infected and uninfected was in regard to the probability of development of active TB following TB infection and mortality – without TB, or during and after treatment for active TB. The risk of active TB in persons with LTBI, and HIV infection has been studied in a number of settings. We could find only one study that categorized cohort members into the states of new or old TB infection and early or late HIV [36]. This was used for the base case estimate of 3.4% annual risk of active TB disease in persons with long-standing LTBI and HIV infection.

We could not find published estimates of risk of disease following new TB infection in HIV infected individuals. Therefore we extrapolated from HIV negative persons, in whom the risk of development of active TB following new LTBI infection is 20–50 times higher than with long-standing LTBI [2426]. Given the 3.4% annual risk of active TB in early HIV infection and long-standing LTBI [36], we assumed the risk of TB disease would be 10 times higher, or 34% per year during the first two years following new TB infection.

Response to treatment of TB disease was not affected by HIV infection. HIV-infected individuals with smear positive active TB were assumed to have no chance of spontaneous cure – hence 100% mortality without treatment. If treated, TB mortality would be 2.25 times higher during treatment [3739], and 2.2 times higher after treatment [3, 40, 41]. Mortality from active TB disease with MDR strains would be 100% [42, 43].

Model calculations

Beginning in 2003 (Year 1), the model determined the proportion of the cohort developing smear positive active TB, dying from TB, or dying from other causes in each year for 20 successive years. The risk of death from other causes was derived from the age distribution and HIV sero-prevalence of the cohort, and country-specific life tables published by WHO [44]. The probability of TB related outcomes depended on cohort members' TB and HIV related states (infected or not, and new or old), as detailed above, and summarized in Table 2. Clinical outcomes also varied according to whether each TB related state was diagnosed and treated.

Cohort members who survived to the end of each year in the model, entered the following year of the simulation. Health states at the beginning of each year depended on the events during the preceding year. As a (simplified) example, some members of the cohort entered Year 1 in the health state of no TB infection. If they survived Year 1 without acquiring TB or HIV infection they entered Year 2 in the same state. However if they acquired new TB infection during Year 1, they entered Year 2 in the "new TB infection" state. During Year 2, they could develop active TB (with probability as shown in Table 2), die from other causes, or remain with new TB infection (entering Year 3 with this health state). If they developed active TB they could be diagnosed and treated, or remain undiagnosed. Likelihood of diagnosis, and treatment outcomes varied according to the strategy being analyzed (see Table 1), and underlying drug resistance. The probability of having pan- sensitive, single drug resistant, or multi-drug resistant underlying strains, was determined from two surveys of initial drug resistance in Haitian TB patients [12, 13]. If undiagnosed they could die of TB (probability in Table 2) or other causes (from Life Tables), or survive and enter Year 3 with undiagnosed TB disease.

The model generated a single value for the cumulative proportion of the cohort that developed active TB, died of TB, or died of other causes throughout the 20 years for each strategy. This was then multiplied by the size of the population to generate expected total number of persons developing each outcome over the period of analysis.


Costs were estimated from societal, and governmental perspectives [45]. Costs modelled from the government perspective included TB-related health care costs for governments and health care providers, plus costs for DOTS implementation, maintenance, and drug costs. TB related costs from the societal perspective included government costs, plus patients' and families' out-of-pocket expenditures, lost wages (including time caring for ill family members), plus productivity losses resulting from disability and death [45]. All costs were expressed in 2003 US dollars, and all future expenditures and outcomes were discounted 3% annually [46].

A questionnaire regarding the household impact of fatal adult illness [47] was adapted, pre-tested in Montreal, and translated into French and Creole – to measure out-of-pocket costs and lost income for patients and families for pre-diagnostic, hospitalisation, treatment and follow-up visits. This was administered by trained interviewers to 84 consenting adults in their second and third months of therapy for new smear positive pulmonary TB, at the same facilities where health system costs were ascertained. To estimate government TB expenditures, we surveyed administrators of 8 rural, and 8 urban health facilities, including 3 TB sanatoria, 5 general hospitals, 5 general clinics, and 3 TB dispensaries. This study was approved by the Institutional Review Board of McGill University, and the National TB control programme of Haiti.

Costs for DOTS implementation and maintenance were based on a DOTS expansion project in Ecuador [48], pro-rated to Haiti, based on their respective per capita gross national incomes [49]. Drug costs were calculated from the model estimates of number of new and retreatment cases, and unit costs from the Global Drug Facility [50]. Patients with active TB were assumed to be 50% disabled (productivity loss) from symptom onset until diagnosis, unable to work while hospitalised, and 50% disabled for the remainder of the first two months of treatment [1719]. Patients who were not diagnosed or who failed treatment were assumed to have 50% disability throughout their illness. Productivity loss from death in HIV uninfected was estimated as per capita annual income [9] times the number of years remaining in the model, with appropriate discounting [46]. For HIV infected the productivity loss was estimated based on the number of years they would be anticipated to survive after development of active TB based on total survival of 9.8 years (35) and the number of years they had survived with HIV infection before developing active TB.

Sensitivity analyses

We conducted extensive one-way sensitivity and threshold analyses, to assess the robustness of our findings to variations in key assumptions. Each parameter was varied individually, except for pathogenetic parameters that were based on well characterized, carefully studied cohorts. We examined the effect of varying rates of decline in tuberculosis incidence following DOTS expansion, increasing HIV seroprevalence and drug resistance, case detection rate, and higher costs of DOTS implementation, maintenance and drugs. Where possible, sensitivity analyses reflected ranges from the published literature. If these were not available, other variations were used such as halving or doubling costs. We also modelled composite "best case" and "worst case" scenarios, based on combinations of favourable and unfavourable assumptions for influential model assumptions – reflecting findings of the initial one-way sensitivity analyses.


As shown in Table 3, current government expenditures averaged $432 per TB patient, of which hospital services accounted for 64%, and TB drugs less than 5%. For the 84 TB patients and families surveyed, total TB related out-of-pocket expenses, and lost income averaged $334 – equivalent to 76% of average per capita income of Haitians in 2002. Of 34 patients who could compare earnings before onset of TB and at the time of the survey, 9 (26%) had a significant drop in income (average 65%), while only one (3%) reported increased income.

Table 3 Summary of health system and patient costs in Haiti

With the status quo strategy, 226,590 TB cases, and 107,070 TB-related deaths are projected to occur in Haiti over 20 years (Table 4). This morbidity and mortality will result in total societal costs of $378 million, including government costs of $59 million. The DOTS expansion strategy is projected to avert 63,080 TB cases, prevent 53,120 TB deaths, and result in societal savings of $131 million, over 20 years. Of the societal savings, 73% will result from deaths averted, and 20%, or approximately $26 million would reflect savings for patients and their families.

Table 4 Projected cumulative TB incidence, related mortality, and costs with two strategies for TB control in Haiti: base case analysis

We project a reduction in government expenditures for hospital services from $38 million to $22 million. However these savings would be offset by the need for initial investment for DOTS expansion, plus increased recurrent government expenditures for smear microscopy, directly observed treatment, multi-drug resistant (MDR) TB treatment, supervision, training, and quality control. As a result, net government savings are projected to be only $4 million over the full 20 years.

The substantial societal savings with DOTS expansion was robust in all sensitivity analyses, including doubling the prevalence of drug resistance, or doubling the costs for initial DOTS expansion, for TB drugs, or for ongoing supervision and training (Table 5). Of note – even if the incidence did not decline at all under DOTS expansion, DOTS is still predicted to result in substantial society savings through reduced mortality and reduced costs for patients and their families. However, government savings were much more susceptible to changes in assumptions- reflecting the very modest government savings in the base case analysis. Haitian government savings would be significant only if foreign donors supported the costs for DOTS expansion.

Table 5 Sensitivity analysis for projected savings over 20 years with DOTS expansion in Haiti

As seen in Figure 2, the DOTS expansion strategy would begin to result in societal savings within 4 years, although government savings would only begin after 15 years. A small increase in HIV sero-prevalence would result in a substantial increase in the number of TB cases, deaths, and TB-related societal costs (Figure 3), although savings with DOTS would be greater. As seen in Figure 4, societal savings would increase by more than $4 million for every 5% increase in case detection. This reflects the lower mortality that would result when more smear positive cases are detected and treated.

Figure 2

Year by year costs or savings with DOTS expansion in Haiti from societal and government perspectives. Dashed line – government perspective, solid line – societal perspective.

Figure 3

Effect of differences in HIV sero-prevalence on the projected cumulative total TB related costs over 20 years in Haiti with the two strategies. Dashed line – status quo, solid line – DOTS expansion.

Figure 4

Effect of changes in the case detection rate with DOTS expansion on total societal savings over 20 years in Haiti.


We project that DOTS expansion in Haiti, to reach WHO targets, would cost an initial $4.2 million and result in 28% reduction in TB cases, 49% reduction in mortality and net societal savings of $131 million over 20 years. However, we project that the Haitian government would have to make significant initial investment and only begin to achieve savings after 15 years. Taken together, these findings should provide a powerful argument for foreign donors to support initial DOTS expansion in Haiti.

Several findings deserve comment. We have projected savings totalling $26 million for patients and their families from DOTS expansion. Among the patients surveyed, direct out-of-pocket expenditures and lost income represented 49% and 27% respectively of the average yearly per capita income in Haiti. And patients surveyed also reported a substantial drop in income at the time of the survey compared to prior to onset of their illness. Savings are anticipated with DOTS expansion, because the decentralisation of diagnostic and treatment services should result in faster diagnosis [1719], reduced hospitalisation [15, 16], and reduced out-of-pocket expenses for treatment and follow-up. Such projected benefits should make DOTS expansion a top priority for donors supporting the Millennium Development goal of poverty reduction [51].

The estimated current government expenditures of $432 per TB patient were surprisingly high – very close to the average per capita income of $440 [9]. We project potential savings from reduced hospitalization, but these will only be realized if hospital expenditures are actually reduced (i.e., by closing beds and reducing staffing). As well any such savings will be offset by increased expenditures for directly observed treatment, and staff supervision, quality control and training. These costly activities are essential for maintenance of a proper DOTS programme [52], but are not necessarily components of TB control programmes prior to DOTS expansion. Therefore new positions must be created, and new workers found with different knowledge and skills – a formidable challenge in many low-income countries. And, immediate government expenditures are required, while savings will only begin after 15 years. As Harold Wilson pointed out "a week is a long time in politics"; most governments will consider 15 years far too long to wait for a payback. Hence, there is little immediate incentive for the Haitian government to implement DOTS.

Another important component of projected government expenditures following DOTS expansion will be lab costs, totalling $9.6 million, or 17% of all TB related government expenditures over 20 years. This reflects the high labour costs for smear microscopy [53], which can take up to 20 minutes of technician time per specimen [52]. In Peru only 2% of TB suspects investigated are smear positive [54], meaning that up to 150 negative smears are examined for each new case found. If treatment success exceeds 85%, enhanced case detection is essential for the long-term epidemiologic impact of DOTS, given the high mortality, and contagiousness of undiagnosed cases. However, given our projections of substantial expenditures for smear microscopy, a high priority should be given to replacing this labour intensive cornerstone of the DOTS strategy.

This study is novel in that we conducted direct surveys to ascertain costs of health facilities, the national TB programme, and TB patients themselves. In this impoverished nation, patient and government expenditures for TB are remarkably high. We combined this direct data gathering with estimates of disability, costs for DOTS expansion, and TB drugs from published experience elsewhere. Additional strengths include the complex decision analysis model which incorporated five TB related health states (with one of three underlying TB drug resistance states) and three HIV related health states within the same model. As well published estimates of risk of TB and HIV infection, HIV progression, development and outcomes of active TB, and mortality were used.

However, there were a number of important potential limitations of this decision analysis. These included our assumptions regarding impact of DOTS on incidence, prevalence of HIV and drug resistance, stability of the population, methods of calculation of annual risk of infection, and costs for DOTS expansion. The assumption that incidence would decline 6% annually following DOTS expansion was based on observations in Peru after nationwide DOTS implementation [5]. This rate of decline is midway between two recent estimates – of 4.3% annually observed with DOTS in China [55], and 7.5% annually predicted for countries achieving WHO targets [56]. And, in sensitivity analyses, societal savings were still substantial even with the extreme assumption of no change in incidence at all. We calculated annual risk of new TB infection from estimated incidence using the Styblo formula [14] – which has been criticized [57]. However, the same formula was applied to both strategies in all years, so inaccuracies in the estimate of infection rates would have been similar for both strategies. We may have over-estimated societal savings because we did not account for other illnesses causing health care costs or disability among those who survived with active TB. However these potential costs should be relatively low since the average age of the patients with active TB in Haiti was 34 so that even those with the most gains in survival would only be 54 by the end of 20 years. As well we did account for mortality from HIV infection (from published studies), and all other causes (from WHO life tables for the general population of Haiti). Thus the most important future societal costs among the added survivors with DOTS, were accounted for in the analysis.

Costs for DOTS expansion are difficult to estimate as there is little published experience to date. For this analysis, costs for expansion were based on published experience in Ecuador [48]. Given the social, economic, and epidemiologic differences between Ecuador and Haiti, these estimates may not be considered valid for Haiti. However these estimated costs were higher than actual costs incurred for a DOTS expansion project in India [58] where the economic situation is very similar to Haiti [9].

A very important limitation of our analysis is the assumption that basic government health services would continue to operate. Therefore we only accounted for the additional costs of DOTS expansion and maintenance. Continued basic government health services in Haiti, depends upon political stability, which is among the worst in the Americas. This problem has contributed to the current problems of the national TB control programme. But, the DOTS strategy has been successfully implemented and maintained amidst major civil conflicts in Mozambique and Nicaragua. Therefore we believe our findings should not be dismissed as overly optimistic by foreign donors considering investment in TB control in Haiti.

We assumed that HIV seroprevalence would remain constant over the next 20 years even though HIV seroprevalence has risen in most developing countries over the past two decades. However if HIV seroprevalence did increase, with a corresponding increased incidence of TB, societal saving would be greater with DOTS (Figure 3), because of improved case detection with corresponding reduction in mortality of undiagnosed active TB in HIV infected. We did not model the potential effect of large scale provision of antiretroviral therapy (ART). This is just being introduced in Haiti and other low income countries, so the costs, efficacy, and population impact of ART are currently unknown. Our assumption of unchanged drug resistance may be incorrect. However, in sensitivity analysis, even when the prevalence of drug resistance was doubled, results were very similar. We also assumed no population growth for Haiti, despite current annual growth of 1.4% [23]. However, a larger population would simply mean more new infections, and new active cases – with either strategy. And, as with higher HIV sero-prevalence, DOTS would be even more cost-saving relative to the status quo strategy.

We have projected that DOTS expansion in Haiti could prevent a large number of TB cases, and TB deaths, with substantial resultant societal savings. But this would require significant Haitian government investment – which may be difficult to ensure, given current political instability and the prospect of little payback after many years. Given this, and the substantial potential humanitarian, economic, and public health benefits, we conclude that foreign donors should strongly consider investing in DOTS expansion in Haiti.


  1. 1.

    Corbett EL, Watt CJ, Walker N, Maher D, Williams BG, Raviglione MC, et al: The growing burden of tuberculosis: global trends and interactions with the HIV epidemic. Arch Intern Med. 2003, 163 (9): 1009-1021. 10.1001/archinte.163.9.1009.

    Article  PubMed  Google Scholar 

  2. 2.

    Kochi A: The global tuberculosis situation and the new control strategy of the World Health Organization. Tuberc. 1991, 72: 1-6. 10.1016/0041-3879(91)90017-M.

    CAS  Article  Google Scholar 

  3. 3.

    World Health Organization: Treatment of Tuberculosis: Guidelines for National Programmes. Geneva. 1997, Second

    Google Scholar 

  4. 4.

    Murray CJL, Styblo K, Rouillon A: Tuberculosis in developing countries: burden, intervention and cost. Bull Int Union Against Tuberculosis. 1990, 65 (1): 2-20.

    Google Scholar 

  5. 5.

    Suarez PG, Watt CJ, Alarcon E, Portocarrero J, Zavala D, Canales R, et al: The dynamics of tuberculosis in response to 10 years of intensive control effort in Peru. J Infect Dis. 2001, 184: 473-478. 10.1086/322777.

    CAS  Article  PubMed  Google Scholar 

  6. 6.

    Floyd K, Blanc L, Raviglione M, Lee JW: Resources required for global tuberculosis control. Science. 2002, 295 (5562): 2040-2041. 10.1126/science.1069771.

    CAS  Article  PubMed  Google Scholar 

  7. 7.

    Sachs J: Macroeconomics and Health: Investigating in Health for Economic Development. World Health Organization. 2001

    Google Scholar 

  8. 8.

    World Health Organization: Global Tuberculosis Control: Surveillance, Planning, Financing. WHO Report. 2003, Geneva, Switzerland

    Google Scholar 

  9. 9.

    The World Bank Group. Haiti Data Profile. 12-8-2003. []

  10. 10.

    UNAIDS: Joint United Nations Programme on HIV/AIDS. 2002, []

    Google Scholar 

  11. 11.

    Lee JW, Espinal M, Jaramillo E: Report of site visit to evaluate DOTS expansion in Latin America. World Health Organization, editor. 2002

    Google Scholar 

  12. 12.

    Pitchenik AE, Russell BW, Cleary T, Pejovic I, Cole C, Snider D: The prevalence of tuberculosis and drug resistance among Haitians. N Engl J Med. 1982, 307 (3): 162-165.

    CAS  Article  PubMed  Google Scholar 

  13. 13.

    Scalcini M, Carré G, Jean-Baptiste M, Hershfield E, Parker S, Wolfe J, et al: Antituberculosis drug resistance in Central Haiti. Am Rev Respir Dis. 1990, 142: 508-511.

    CAS  Article  PubMed  Google Scholar 

  14. 14.

    Styblo K: The relationship between the risk of tuberculosis infection and the risk of developing infectious tuberculosis. Bull Int Union Tuberc. 1985, 60 (3–4): 117-119.

    Google Scholar 

  15. 15.

    Floyd K, Skeva J, Nyirenda T, Gausi F, Salaniponi F: Cost and cost-effectiveness of increased community and primary care facility involvement in tuberculosis care in Lilongwe District, Malawi. Int J Tuberc Lung Dis. 2003, 7 (9 Suppl 1): S29-S37.

    CAS  PubMed  Google Scholar 

  16. 16.

    Floyd K, Wilkinson D, Gilks C: Comparison of cost effectiveness of directly observed treatment (DOT) and conventionally delivered treatment for tuberculosis: experience from rural South Africa. BMJ. 1997, 315 (7120): 1407-1411.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  17. 17.

    Sherman LF, Fujiwara PI, Cook SV, Bazerman LB, Frieden TR: Patient and health care system delays in the diagnosis and treatment of tuberculosis. Int J Tuberc Lung Dis. 1999, 3 (12): 1088-1095.

    CAS  PubMed  Google Scholar 

  18. 18.

    Yamasaki-Nakagawa M, Ozasa K, Yamada N, Osuga K, Shimouchi A, Ishikawa N, et al: Gender diference in delays to diagnosis and health care seeking behaviour in a rural area of Nepal. Int J Tuberc Lung Dis. 2001, 5 (1): 24-31.

    CAS  PubMed  Google Scholar 

  19. 19.

    Wandwalo ER, Mørkve O: Delay in tuberculosis case-finding and treatment in Mwanza, Tanzania. Int J Tuberc Lung Dis. 2000, 4 (2): 133-138.

    CAS  PubMed  Google Scholar 

  20. 20.

    Caminero Luna JA: Guia de la Tuberculosis para Medicos Especialistas. Union Internacional contra la Tuberculosis y Enfermedades Respiratorias (UICTER), editor. 6432. 2003, Imprime en France, Imprimerie Chirat, Depot Legal 2003 No 6432

    Google Scholar 

  21. 21.

    Gupta R, Cegielski JP, Espinal MA, Henkens M, Kim JY, Lambregts-van Weezenbeek CSB, et al: Increasing transparency in partnerships for health – introducing the Green Light Committee. Tropical Medicine and International Health. 2002, 7 (2): 970-976. 10.1046/j.1365-3156.2002.00960.x.

    Article  PubMed  Google Scholar 

  22. 22.

    Suarez PG, Floyd K, Portocarrero J, Alarcon E, Rapiti E, Ramos G, et al: Feasibility and cost-effectiveness of standardised second-line drug treatment for chronic tuberculosis patients: a national cohort study in Peru. Lancet. 2002, 359 (9322): 1980-1989. 10.1016/S0140-6736(02)08830-X.

    Article  PubMed  Google Scholar 

  23. 23.

    CIA – The World Factbook. CIA – The World Factbook. 12-9-2003. Washington, DC, Central Intelligence Agency – USA, []

  24. 24.

    Sutherland I: The evolution of clinical tuberculosis in adolescents. Tuberc. 1966, 47: 308-

    Google Scholar 

  25. 25.

    Van Zwanenberg D: The Influence of the number of bacilli on the development of tuberculous disease in children. Am Rev Respir Dis. 1960, 82: 31-44.

    CAS  PubMed  Google Scholar 

  26. 26.

    Ferebee SH: Controlled chemoprophylaxis trials in tuberculosis. Adv Tuberc Res. 1969, 17: 28-106.

    Google Scholar 

  27. 27.

    Menzies D: Issues in the management of contacts of patients with active pulmonary tuberculosis. Can J Public Health. 1997, 88 (3): 197-201.

    CAS  PubMed  Google Scholar 

  28. 28.

    Comstock GW, Edwards LB, Livesay VT: Tuberculosis morbidity in the US Navy: its distribution and decline. Am Rev Respir Dis. 1974, 110: 572-580.

    CAS  PubMed  Google Scholar 

  29. 29.

    Nolan CM, Elarth AM: Tuberculosis in a cohort of Southeast Asian refugees: A five-year surveillance study. Am Rev Resp Dis. 1988, 137: 805-809.

    CAS  Article  PubMed  Google Scholar 

  30. 30.

    Cummings KC, Mohle-Boetani J, Royce SE, Chin DP: Movement of tuberculosis patients and the failure to complete antituberculosis treatment. Am J Respir Crit Care Med. 1998, 157 (4 Pt 1): 1249-1252.

    CAS  Article  PubMed  Google Scholar 

  31. 31.

    Chee CBE, Boudville IC, Chan SP, Zee YK, Wang YT: Patient and disease characteristics, and outcome of treatment defaulters from the Singapore TB control unit – a one-year retrospective survey. Int J Tuberc Lung Dis. 2000, 4 (6): 496-503.

    CAS  PubMed  Google Scholar 

  32. 32.

    Parthasarathy R, Prabhakar R, Somasundaram PR: A controlled clinical trial of 3- and 5-month regimens in the treatment of sputum-positive pulmonary tuberculosis in South India. Am Rev Respir Dis. 1986, 134: 27-33.

    Google Scholar 

  33. 33.

    East African/British Medical Research Councils Study: Controlled clinical trial of five short-course (4-month) chemotherapy regimens in pulmonary tuberculosis: Second report of the 4th study. Am Rev Respir Dis. 1981, 123: 165-170.

    Google Scholar 

  34. 34.

    Singapore Tuberculosis Service/British Medical Research Council: Long-term Follow-up of a clinical trial of six-month and four-month regimens of chemotherapy in the treatment of pulmonary tuberculosis. Am Rev Respir Dis. 1986, 133: 779-783.

    Google Scholar 

  35. 35.

    Morgan D, Mahe C, Mayanja B, Okongo JM, Lubega R, Whitworth JA: HIV-1 infection in rural Africa: is there a difference in median time to AIDS and survival compared with that in industrialized countries?. AIDS. 2002, 16 (4): 597-603. 10.1097/00002030-200203080-00011.

    Article  PubMed  Google Scholar 

  36. 36.

    Wood R, Maartens G, Lombard CJ: Risk factors for developing tuberculosis in HIV-1 – Infected adults from communities with low or very high incidence of tuberculosis. J Acquir Immune Defic Syndr. 2000, 23: 75-80.

    CAS  Article  PubMed  Google Scholar 

  37. 37.

    Murray J, Sonnenberg P, Shearer SC, Godgrey-Faussett P: Human immunodeficiency virus and outcome of treatment for new and recurrent pulmonary tuberculosis in African patients. Am J Respir Crit Care Med. 1999, 159: 733-740.

    CAS  Article  PubMed  Google Scholar 

  38. 38.

    Chaisson RE, Clermont HC, Hole EA, Cantave M, Johnson MP, Atkinson J, et al: Six-month supervised intermittent tuberculosis therapy in Haitian patients with and without HIV infection. Am J Respir Crit Care Med. 1996, 154: 1034-1038.

    CAS  Article  PubMed  Google Scholar 

  39. 39.

    Desvarieux M, Hyppolite PR, Johnson WD, Pape JW: A novel approach to directly observed therapy for tuberculosis in an HIV-endemic area. Am J Public Health. 2001, 91 (1): 138-141.

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  40. 40.

    Whalen C, Horsburgh CR, Hom D, Lahart C, Simberkoff M, Ellner J: Accelerated course of human immunodeficiency virus infection after tuberculosis. Am J Respir Crit Care Med. 1995, 151: 129-135.

    CAS  Article  PubMed  Google Scholar 

  41. 41.

    Connolly C, Reid , Davies G, Sturm W, McAdam K, Wilkinson D: Relapse and mortality among HIV-infected and uninfected patients with tuberculosis successfully treated with twice weekly directly observed therapy in rural South Africa. AIDS. 1999, 13: 1543-1547. 10.1097/00002030-199908200-00015.

    CAS  Article  PubMed  Google Scholar 

  42. 42.

    Beck-Sague C, Dooley SW, Hutton MD, Otten J, Breedan A, Crawford JT, et al: Hospital outbreak of multi-drug resistant Mycobacterium tuberculosis infections: Factors in transmission to staff and HIV-infected patients. JAMA. 1992, 268: 1280-1286. 10.1001/jama.268.10.1280.

    CAS  Article  PubMed  Google Scholar 

  43. 43.

    Edlin BR, Tokars JI, Grieco MH, Crawford JT, Williams J, Sordillo EM, et al: An outbreak of multi-drug resistant tuberculosis among hospitalized patients with the acquired immunodeficiency syndrome. New Engl J Med. 1992, 326 (23): 1514-1521.

    CAS  Article  PubMed  Google Scholar 

  44. 44.

    Lopez AD, Salomon J, Ahmad O, Murray CJL, Mafat D: Life Tables for 191 Countries: Data, Methods and Results. 2000, Geneva, World Health Organization (GPE Discussion Paper Series: No.9)

    Google Scholar 

  45. 45.

    Drummond MF, O'Brien B, Stoddart GL, Torrance GW: Methods for the economic evaluation of health care programmes. 1997, New York: Oxford University Press, 2

    Google Scholar 

  46. 46.

    Weinstein MC, Siegel JE, Gold MR, Kamlet MS, Russell LB: Recommendations of the Panel on Cost-Effectiveness in Health and Medicine. JAMA. 1996, 276 (15): 1253-1258. 10.1001/jama.276.15.1253.

    CAS  Article  PubMed  Google Scholar 

  47. 47.

    Ainsworth M, Koda G, Lwihula G, Mujinja P, Over M, Semali I: Measuring the Impact of Fatal Adult Illness in Sub-Saharan Africa – An Annotated Household Questionnaire. LSMS Living Standards Measurements Study Working Paper No.90. 1992, Washington, DC, The World Bank

    Google Scholar 

  48. 48.

    Vaca J, Peralta H, Gresely L, Cordova R, Kuffo D, Romero E, et al: DOTS implementation in a middle income country – Development and evaluation of a novel approach. Int J Tuberc Lung Dis. 2005, 9 (5): 521-527.

    CAS  PubMed  Google Scholar 

  49. 49.

    World Bank Website. 2004, Access date: April 15, 2004, []

  50. 50.

    Global Drug Facility: First-Line tuberculosis drugs& formulations currently supplied/to be supplied by the global TB drug facility. World Health Organization, editor. 2003, Access date: November 15, 2003, []

    Google Scholar 

  51. 51.

    The World Bank Group: Achieving the MDG's and related outcomes: A framework for monitoring policies and actions. 2004, []

    Google Scholar 

  52. 52.

    Management of Tuberculosis: A guide for low income countries. 2000, International Union Against Tuberculosis and Lung Disease, Fifth

    Google Scholar 

  53. 53.

    Roos BR, van Cleeff MRA, Githui WA, Kivihya-Ndugga L, Odhiambo JA, Kibuga DK, et al: Cost-effectiveness of the polymerase chain reaction versus smear examination for the diagnosis of tuberculosis in Kenya: a theoretical model. Int J Tuber Lung Dis. 1997, 2 (3): 235-241.

    Google Scholar 

  54. 54.

    Suarez P: Government of Peru MoH. Tuberculosis en el Perú Informe 2000. 2001, Lima, Peru

    Google Scholar 

  55. 55.

    The effect of tuberculosis control in China. Lancet. 2004, 364 (9432): 417-422. 10.1016/S0140-6736(04)16764-0.

  56. 56.

    Elzinga G, Raviglione MC, Maher D: Scale up: meeting targets in global tuberculosis control. Lancet. 2004, 363 (9411): 814-819. 10.1016/S0140-6736(04)15698-5.

    Article  PubMed  Google Scholar 

  57. 57.

    Borgdorff MW: Annual risk of tuberculous infection: time for an update?. Bull World Health Organ. 2002, 80 (6): 501-502.

    PubMed Central  Google Scholar 

  58. 58.

    Khatri GA, Frieden TR: Controlling Tuberculosis in India. N Engl J Med. 2002, 347 (18): 1420-1425. 10.1056/NEJMsa020098.

    CAS  Article  PubMed  Google Scholar 

  59. 59.

    Burgess AL, Fitzgerald DW, Severe P, Joseph P, Noel E, Rastogi N, et al: Integration of tuberculosis screening at an HIV voluntary counselling and testing centre in Haiti. AIDS. 2001, 15: 1875-1879. 10.1097/00002030-200109280-00018.

    CAS  Article  PubMed  Google Scholar 

  60. 60.

    Dasgupta K, Schwartzman K, Marchand R, Tannenbaum 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 (6): 2079-2086.

    CAS  Article  PubMed  Google Scholar 

  61. 61.

    Comstock GM: How much isoniazid is needed for prevention of tuberculosis in immunocompetent adults. Int J Tuberc Lung Dis. 1999, 3 (10): 847-850.

    CAS  PubMed  Google Scholar 

  62. 62.

    Nolan CM, Aitken ML, Elarth AM, Anderson KM, Miller WT: Active tuberculosis after isoniazid chemoprophylaxis of Southeast Asian refugees. Am Rev Respir Dis. 1986, 133: 431-436.

    CAS  PubMed  Google Scholar 

  63. 63.

    Dye C, Espinal MA, Watt CJ, Mbiaga C, Williams BG: Worldwide incidence of multidrug-resistant tuberculosis. J Infect Dis. 2002, 185 (8): 1197-1202. 10.1086/339818.

    Article  PubMed  Google Scholar 

  64. 64.

    Whalen CC, Johnson JL, Okwera A, Hom DL, Huebner R, Mugyenyi P, et al: A trial of three regimens to prevent tuberculosis in Ugandan adults infected with the human immunodeficiency virus. N Engl J Med. 1997, Uganda-Case Western Reserve University Research Collaboration, 337 (12): 801-808. 10.1056/NEJM199709183371201.

  65. 65.

    Guelar A, Gatell JM, Verdejo J, Podzamczer D, Lozano L, Aznar E, et al: A prospective study of the risk of tuberculosis among HIV-infected patients. AIDS. 1993, 7: 1345-1349.

    CAS  Article  PubMed  Google Scholar 

  66. 66.

    Grzybowski S, Barnett GD, Styblo K: Contacts of cases of active pulmonary tuberculosis. Bull IUAT. 1975, 50: 90-106.

    CAS  Google Scholar 

  67. 67.

    Fischl MA, Uttamchandani RB, Daikos L, Poblete RB, Moreno JN, Reyes RR, et al: An outbreak of tuberculosis caused by multiple-drug resistant tubercle bacilli among patients with HIV infection. Ann Intern Med. 1992, 117: 177-183.

    CAS  Article  PubMed  Google Scholar 

  68. 68.

    Small P, Shafer R, Hopewell P: Exogenous reinfection with multidrug-resistant Mycobacterium tuberculosis in patients with advanced HIV infection. N Engl J Med. 1993, 328: 1137-1144. 10.1056/NEJM199304223281601.

    CAS  Article  PubMed  Google Scholar 

  69. 69.

    Daley CL, Small PM, Schecter GF, Schoolnik GK, McAdam RA, Jacobs WR, et al: An outbreak of tuberculosis with accelerated progression among persons infected with the human immunodeficiency virus. New Engl J Med. 1992, 326 (4): 231-235.

    CAS  Article  PubMed  Google Scholar 

  70. 70.

    Stead WW: Management of health care workers after inadvertent exposure to tuberculosis: A guide for use of preventive therapy. Ann Intern Med. 1995, 122: 906-912.

    CAS  Article  PubMed  Google Scholar 

  71. 71.

    Rieder HL: Epidemiologic basis of tuberculosis control. 1999, Paris, France, International Union Against Tuberculosis and Lung Disease, 1-162. First

    Google Scholar 

  72. 72.

    Grzybowski S, Enarson DA: The fate of cases of pulmonary tuberculosis under various treatment programmes. Bull Int Union Tuberc. 1978, 53 (2): 70-74.

    CAS  PubMed  Google Scholar 

  73. 73.

    Horwitz O: Public health aspects of relapsing tuberculosis. Am Rev Respir Dis. 1969, 99: 183-193.

    CAS  PubMed  Google Scholar 

  74. 74.

    Cohn DL, Catlin BJ, Peterson KL, Judson FN, Sbarbaro JA: A 62-dose, 6-month therapy for pulmonary and extrapulmonary tuberculosis. A twice-weekly, directly observed, and cost-effective regimen. Ann Intern Med. 1990, 112 (6): 407-415.

    CAS  Article  PubMed  Google Scholar 

  75. 75.

    Results at 5 years of a controlled comparison of a 6-month and a standard 18-month regimen of chemotherapy for pulmonary tuberculosis. Am Rev Respir Dis. 1977, 116 (1): 3-8.

  76. 76.

    Somner AR: Short-course chemotherapy in pulmonary tuberculosis. A controlled trial by the British Thoracic Association (third report). Lancet. 1980, 1 (8179): 1182-1183.

    CAS  PubMed  Google Scholar 

  77. 77.

    Controlled clinical trial comparing a 6-month and a 12-month regimen in the treatment of pulmonary tuberculosis in the Algerian Sahara. Algerian working group/British Medical Research Council cooperative study. Am Rev Respir Dis. 1984, 129 (6): 921-928.

  78. 78.

    Benator D, Bhattacharya M, Bozeman L, Burman W, Cantazaro A, Chaisson R, et al: Rifapentine and isoniazid once a week versus rifampicin and isoniazid twice a week for treatment of drug-susceptible pulmonary tuberculosis in HIV-negative patients: a randomised clinical trial. Lancet. 2002, 360 (9332): 528-534. 10.1016/S0140-6736(02)09742-8.

    Article  PubMed  Google Scholar 

  79. 79.

    Espinal MA, Kim SJ, Suarez PG, Kam KM, Khomenko AG, Migliori GB, et al: Standard short-course chemotherapy for drug-resistant tuberculosis: treatment outcomes in 6 countries. JAMA. 2000, 283 (19): 2537-2545. 10.1001/jama.283.19.2537.

    CAS  Article  PubMed  Google Scholar 

  80. 80.

    Mitnick C, Bayona J, Palacios E, Shin S, Furin J, Alcantara F, et al: Community-based therapy for multidrug-resistant tuberculosis in Lima, Peru. N Engl J Med. 2003, 348 (2): 119-128. 10.1056/NEJMoa022928.

    Article  PubMed  Google Scholar 

  81. 81.

    Deschamps MM, Fitzgerald DW, Pape JW, Johnson WD: HIV Infection in Haiti: Natural History and Disease Progression. AIDS. 2000, 14: 2515-2521. 10.1097/00002030-200011100-00014.

    CAS  Article  PubMed  Google Scholar 

  82. 82.

    Malkin JE, Prazuck T, Simonnet F, Yameogo M, Rochereau A, Ayerour J, et al: Tuberculosis and human immunodeficiency virus infection in West Burkina Faso: clinical presentation and clinical evolution. Int J Tuberc Lung Dis. 1997, 1 (1): 68-74.

    CAS  PubMed  Google Scholar 

  83. 83.

    Johnson JL, Okwera A, Vjecha MJ, Byekwaso F, Nakibali J, Nyole S, et al: Risk factors for relapse in human immunodeficiency virus type 1 infected adults with pulmonary tuberculosis. Int J Tuberc Lung Dis. 1997, 1 (5): 446-453.

    CAS  PubMed  Google Scholar 

  84. 84.

    Sonnenberg P, Murray J, Glynn JR, Shearer S, Kambashi B, Godfrey-Faussett P: HIV-1 and recurrence, relapse, and reinfection of tuberculosis after cure: a cohort study in South African mineworkers. Lancet. 2001, 358 (9294): 1687-1693. 10.1016/S0140-6736(01)06712-5.

    CAS  Article  PubMed  Google Scholar 

  85. 85.

    Pulido F, Peña JM, Rubio R, Moreno S, González J, Guijarro C, et al: Relapse of tuberculosis after treatment in human immunodeficiency virus-infected patients. Arch Intern Med. 1997, 157: 227-231. 10.1001/archinte.157.2.227.

    CAS  Article  PubMed  Google Scholar 

  86. 86.

    Deas J: Haiti's Response to Tuberculosis and Malaria. Application to the Global Fund to Fight AIDS TaM, editor. 2002

    Google Scholar 

Pre-publication history

  1. The pre-publication history for this paper can be accessed here:

Download references


We are grateful to many staff of the National TB Programme in Haiti, Sarah Hoibak for assistance in the health facility and patient costs questionnaire, Jason McKnight for assistance in data analysis, and Catherine Michaud for seemingly endless revisions of this manuscript.

Supported by a grant from the Rockefeller Foundation Drs. Schwartzman and Menzies are recipients of research career awards from the Fonds de Recherche en Santé du Québec. Dr. Barr is the recipient of a Robert Wood Johnson Generalist Physician Faculty Scholar Award.

Author information



Corresponding author

Correspondence to Dick Menzies.

Additional information

Competing interests

The author(s) declare that they have no competing interests.

Authors' contributions

V. J. – Conception of project, data gathering, drafting manuscript

W. M. – Data gathering and critical revisions of manuscript

K. S. – Conception of project, data analysis, critical revisions of manuscript

O. O. – Conception of project, data gathering, data analysis, critical revisions of manuscript

G. B. – Conception of project, data analysis, critical revisions of manuscript

F. G. – Conception of project, and critical revision of manuscript

D. M. – Conception of project, data gathering, data analysis, drafting manuscript

Vary Jacquet, Willy Morose, Kevin Schwartzman, Olivia Oxlade, Graham Barr, Franque Grimard and Dick Menzies contributed equally to this work.

Authors’ original submitted files for images

Rights and permissions

Open Access This article is published under license to BioMed Central Ltd. This is an Open Access article is distributed under the terms of the Creative Commons Attribution License ( ), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Reprints and Permissions

About this article

Cite this article

Jacquet, V., Morose, W., Schwartzman, K. et al. Impact of DOTS expansion on tuberculosis related outcomes and costs in Haiti. BMC Public Health 6, 209 (2006).

Download citation


  • Foreign Donor
  • Health System Delay
  • Government Saving
  • Societal Saving
  • Haitian Government