Evaluating the potential impact of the new Global Plan to Stop TB: Thailand, 2004–2005
Jay K Varmaa, Daranee Wiriyakitjarb, Sriprapa Nateniyomb, Amornrat Anuwatnonthakatec, Patama Monkongdeec, Surin Sumnapand, Somsak Akksilpe, Wanchai Sattayawuthipongf, Pricha Charunsuntonsrig, Somsak Rienthongb, Norio Yamadah, Pasakorn Akarasewic, Charles D Wellsa, Jordan W Tapperoa
Tuberculosis (TB) remains one of the world’s leading causes of severe illness and death, particularly in developing countries. Since 1993, efforts to control TB have focused on a strategy known as DOTS, which emphasizes passive case detection and standardized, directly observed treatment of sputum smear-positive TB cases.1 Widespread adoption of DOTS has greatly expanded the number of patients cured of TB, but global TB incidence and mortality have not declined.2 Major reasons for the failure to control TB include incomplete case-finding, inadequate and abundant TB care delivered by clinicians working outside national TB programmes, and the emergence of multidrug-resistant TB (MDR-TB) and HIV-associated TB.2 Recognizing these problems, in 2006 WHO launched the Global Plan to Stop TB 2006–2015, which calls upon countries to expand and enhance the DOTS strategy and also to implement collaboration between TB and HIV programmes, improve the diagnosis and treatment of MDR-TB, establish public–private partnerships, enable and promote research, strengthen existing health systems, and empower patients and communities.3
Global efforts to control TB have focused most intensely on 22 countries designated by WHO as having a high burden of TB; together these account for >80% of the world’s TB cases. In 2006, Thailand ranked 18th on the list of high-burden countries; an estimated 90 000 people develop TB annually giving an annual incidence of 135 TB cases per 100 000 people.4 Case notifications were declining in Thailand until an explosive HIV epidemic in the 1990s resulted in a sudden increase in TB cases. HIV-associated TB now accounts for an estimated 15% of all TB cases in Thailand.4 A WHO review in 2003 of Thailand’s TB programme recommended that Thailand strengthen its recording and reporting system, its laboratory services for TB patients, collaborative activities between TB and HIV programmes, and partnerships with the private sector.5
In response to the recognized limitations of the DOTS strategy and to the findings of the programme review, in October 2004 we began implementing a demonstration project of enhanced TB control in selected provinces in Thailand, incorporating strategies that are now recommended as part of the new Global Plan to Stop TB. We report here on findings from the first year of this demonstration project, which provided a unique opportunity to measure the potential impact of the new Global Plan in a high-burden country.
The Thailand TB Active Surveillance Network is a partnership between Thailand’s Ministry of Public Health, the Bangkok Metropolitan Administration, the United States Centers for Disease Control and Prevention, and the Research Institute of Tuberculosis, Japan. It involves all districts in three provinces (Chiang Rai, Phuket, Ubon-ratchathani) and two districts in Bangkok. The catchment area includes 3 557 249 people (according to a 2004 population estimate), 50 public health-care facilities and 279 private health-care facilities.
Before launching this project in October 2004, public health officials collected data from all public and private health facilities in the catchment area and negotiated agreements with these facilities to participate in the project. Data collection involved reviewing clinical and public health records to determine the number and type of TB cases, including those in HIV-infected patients and patients who were not Thai, occurring from October 2002 to September 2003 (hereafter referred to as 2003). This baseline time period was chosen, because provinces in Thailand routinely collect TB programme data using a fiscal year not calendar year, and some sites began implementing selected components of this programme between October 2003 and September 2004, making 2003 the most appropriate baseline year.
Active surveillance, monitoring and evaluation
During the project, public health staff in each province contacted participating health-care facilities in their catchment area at least once per month to obtain standardized information about each newly diagnosed case of TB, to monitor and evaluate the programme’s performance, to assist with case-management, and to follow up TB treatment outcomes for cases already being treated. The most important difference between this active surveillance and the baseline system was that public health staff in each province directly contacted (in person or by telephone) medical personnel working in both public and private facilities about case-finding and management each month, rather than relying on passive case reporting.
A case of TB was defined as anyone diagnosed with TB or treated for tuberculous disease. A new case of TB was defined as TB disease occurring in a person who reported having had no previous treatment or <1 month of treatment. Cases of TB occurring in people who are not Thai or in people diagnosed in nongovernmental facilities are not routinely reported to the national TB programme; however, in this project, all cases were included regardless of the patient’s nationality or type of health-care facility visited.
Staff recorded data using a modified version of the standard national TB register, entered data into an electronic database, and transferred data via secure Internet connection to the national TB programme.
Diagnosis and treatment
Before and during the implementation of the project, staff at public and private health-care facilities received refresher training about national guidelines for diagnosis, treatment and case-management. Provincial TB programmes were provided with additional resources to support the training of personnel and the monitoring and evaluation of district TB programmes, but routine TB diagnostic services (such as microscopy and radiography) and treatment services were paid for using existing funds, not project-specific funds. Treating physicians were not required as part of this project to follow any specific standard for TB care and treatment. TB programmes recorded on surveillance forms whether patients were directly observed ingesting their TB medicine. Those who might observe treatment included health-care workers, village health volunteers or family members.
As part of the demonstration project, we developed the capacity to perform mycobacterial culture at one laboratory in each province. For people who had already been diagnosed with TB, health-care facilities were asked to submit at least one sputum specimen for culture and susceptibility testing, ideally during the first month of TB treatment; sputum culture was not routinely used to diagnose TB. Isolates from Bangkok were also identified and tested for susceptibility to first-line drugs (streptomycin, isoniazid, rifampin, pyrazinamide, ethambutol) at the city’s central laboratory; all other isolates were sent from the province-level laboratory to the national TB programme’s reference laboratory for identification and susceptibility testing. Methods of sputum culture varied during 2005. Initially, all specimens were cultured on solid media, either Lowenstein–Jensen or Ogawa, but during the course of the project, sites shifted to performing solid culture only on Lowenstein–Jensen and began also to use liquid media culture with an automated reading instrument (BACTEC Mycobacteria Growth Indicator Tube 960, Becton Dickinson, Franklin Lakes, NJ, USA).
TB and HIV
During the project, nurses and physicians from public and private health-care facilities were trained in HIV counselling, testing, and care and treatment. During surveillance visits, these nurses and physicians were also encouraged to provide HIV counselling and testing to TB patients and HIV-related care and treatment to TB patients also infected with HIV. No financial incentives were provided to patients or health-care workers for undergoing or delivering HIV testing. Individual physicians used their own judgement about whether to measure CD4+ T-cell lymphocyte counts (CD4), provide prophylaxis for opportunistic infections or antiretroviral therapy, and manage other clinical conditions. When performed, blood for CD4 testing was usually drawn during the first month of TB treatment.
The protocol for this project was reviewed by the Thai Ministry of Pubic Health and the US Centers for Disease Control and Prevention and categorized as surveillance and implementation of a public health programme; thus it did not require oversight by a human subjects’ research institutional review board.
The total number of case notifications in the catchment area was 4978 in 2002, 4904 in 2003, and 5289 in 2004. Of the 4904 TB cases in 2003 (the baseline year), 4229 were new cases (118/100 000 people) and 2060 were new smear-positive cases (58/100 000 people). The total number of cases managed in all nongovernmental facilities was not known. In 2003, HIV status was known for 1678 (34%) cases, of whom 897 (53%) were HIV-infected and 781 (47%) were not. In 2003, the minimum estimated HIV prevalence among TB patients was 18% (897/4904). Public health-care facilities recorded 200 non-Thai nationals registered for TB treatment.
All 50 of the public and 53 private health-care facilities agreed to participate in the project. The 226 private health-care facilities that refused to participate were all small clinics, and only 5 reported in our baseline assessment that they had treated any TB cases in 2003; these 5 reported managing a total of 56 TB cases.
From October 2004 to September 2005 (hereafter referred to as 2005), 5841 TB cases were ascertained in the surveillance area, including 2684 (46%) pulmonary smear-positive cases, 1652 (28%) smear-negative, 320 (5%) with smear status not known or done, 1183 (20%) extra-pulmonary, and 2 (< 1%) unknown (Table 1). Compared with case reporting to the national programme in 2003, reporting during the project period was higher (rising from 4904 cases to 5841, a 19% increase), particularly for new pulmonary smear-negative cases (Table 2).
Two populations contributed substantially to the increase in case reporting: patients who did not have Thai nationality and those seen in the private sector. Reporting of cases among people who did not have Thai nationality increased 136%, from 200 cases in 2003 to 472 in 2005. Increases occurred in all sites: rising from 1 to 61 in Bangkok, 182 to 327 in Chiang Rai, from unknown to 4 in Ubon-ratchathani, and 17 to 80 in Phuket. In 2005, private sector providers reported diagnosing 634 cases (11% of all cases) compared with no cases reported to the national programme in 2003. Most private-sector cases were found in Bangkok (445) and Phuket (92), but all sites reported some cases (60 in Chiang Rai, 37 in Ubon-ratchathani). Of the 634 cases diagnosed in the private sector, 213 (34%) were smear-positive, 118 (19%) smear-negative, 91 (14%) extra-pulmonary, 211 (33%) were pulmonary with smear not done, and 1 had missing data.
Table 1. Characteristics of patients reported to Thailand’s Tuberculosis (TB) Active Surveillance Network, Thailand, October 2004–September 2005
Table 2. Tuberculosis (TB) case-finding before (October 2002–September 2003) and after (October 2004–September 2005) implementation of population-based surveillance, Thailand
Characteristics of cases
In 2005, most cases occurred in males and people aged 25–44 years (Table 1). Using injected drugs and having status as a migrant or refugee were uncommon. Altogether 10% had previously been treated for TB. Most (57%) cases had had a cough lasting >2 weeks and had abnormal X-rays (73%).
TB and HIV
Of 5841 TB cases identified in 2005, 985 (17%) had known their HIV status before their diagnosis of TB. Of the remaining 4856 cases who did not know their HIV status, 3956 (81%) underwent HIV counselling; 3410 (86%) of those who were counselled agreed to HIV testing. In the end, 1392 (24%) cases were known to be infected with HIV (985, or 70%, of whom knew their HIV status before TB diagnosis); 3120 (53%) were known to be HIV-uninfected; and HIV status was unknown for 1329 (23%). The proportion with unknown HIV status decreased from 66% (3226/4904) in 2003 to 23% (1329/5841) in 2005 (χ² P < 0.01).
CD4 counts were available for 865 (62%) HIV-infected cases (Table 3). The median CD4 count was 58 cells/mm³ (range = 0–2731) and mean 114.6. Only 114 (13%) cases had CD4 counts >200 cells/mm³. Of the 1392 HIV-infected cases, 183 (13%) had been prescribed antiretroviral treatment before their diagnosis of TB; 358 (26%) were prescribed antiretrovirals during TB treatment; 640 (46%) were not prescribed antiretrovirals before or during TB treatment; and data were missing for 211 (15%). Rates of co-trimoxazole use were higher: 504 (36%) were taking co-trimoxazole before their diagnosis of TB and 551 (40%) were prescribed co-trimoxazole during TB treatment.
Table 3. Characteristics of HIV-infected patients reported to Thailand’s Tuberculosis (TB) Active Surveillance Network, Thailand, October 2004–September 2005
Culture and drug-susceptibility testing
Of 4656 pulmonary TB cases reported in 2005, 4336 (93%) had at least one sputum smear performed, and 3024 (65%) had at least one sputum culture performed. Of cases with a culture performed, 1928 (64%) were culture-positive for Mycobacterium tuberculosis; 1687 (88%) of these were tested for drug susceptibility. No baseline data from 2003 were available.
MDR-TB was diagnosed in 60 cases, 34 (56%) of whom had been previously treated for TB. When using a denominator that included all pulmonary TB cases (regardless of whether sputum culture or drug-susceptibility testing had been performed), the prevalence of MDR-TB in previously treated cases was 5% (26/510) and in previously untreated cases was 1% (34/4146). When using a denominator of only those pulmonary TB cases that were culture positive and had an isolate that underwent drug-susceptibility testing, the prevalence of MDR-TB in previously treated cases was 15% (26/173) and in previously untreated cases was 2% (34/1514).
By implementing strategies recommended in the new Global Plan to Stop TB, we increased case-finding, collaboration with the private sector, the HIV-related services provided to TB patients, and the diagnosis of MDR-TB in the first year of a demonstration project in a country with a high burden of TB.
Increases in case-finding were particularly dramatic. Total case notifications fluctuated during 2002–2004, strongly suggesting that the substantial increase in 2005 was attributable to our project, specifically to the inclusion in the surveillance system of non-Thai patients and patients seen in the private sector. A large number of people from other countries live in Thailand. Rates of TB are thought to be high among these populations because many have migrated from countries (such as Cambodia and Myanmar) that have higher TB rates than Thailand.4 Nevertheless, Thailand’s national statistics do not routinely capture data on non-Thais, making it difficult to assess the burden of TB disease and the quality of TB treatment. This project shows that the case-notification rate can be improved by changing the guidelines of the national programme to include reporting on non-Thai patients. Similarly, private-sector providers are believed to treat large numbers of TB patients, particularly in more economically developed regions, such as Bangkok and Phuket.5
Various strategies have been attempted throughout the world to promote public–private partnerships in TB control.6 In this project, we focused first on the sharing of patient data, reasoning that this was a non-threatening first step to take towards engaging the private sector and that efforts to promote DOTS in these facilities would succeed only if there were standardized ways of comparing performance between the national TB programme and the private sector. This project demonstrates the feasibility and impact of including data about non-Thai patients and patients in the private sector in TB surveillance. Many countries with a high-burden of TB have failed to reach WHO’s target of 70% case detection. This project demonstrates how including migrant and private-sector patients may help countries achieve this target.
A major component of the new Global Plan is the aim of strengthening HIV services for TB patients. Through training, provision of necessary supplies, and aggressive monitoring, we were able to increase rates of HIV testing within a year and to provide patients with co-trimoxazole and antiretroviral treatment. Although more effort is needed to strengthen these linkages, our surveillance system provides an important baseline and ongoing monitoring system for collaborative activities between TB and HIV programmes. In contrast to sub-Saharan Africa, we found that most HIV-infected patients knew their HIV status before being diagnosed with TB and that most presented with advanced immune-system suppression.7–10 This suggests that efforts to control HIV-associated TB in Thailand should focus on providing routine, rigorous screening for TB disease in people already known to be HIV-infected. Given that the overwhelming majority of cases of TB occurred in people with CD4 counts <200 cells/mm³, providing antiretrovirals and possibly isoniazid preventive therapy to patients above this threshold could also prevent HIV-associated TB in Thailand.11,12
This project enhanced the capacity of laboratories to culture mycobacteria and likely increased the diagnosis of MDR-TB. No routine data about MDR-TB are collected in Thailand, but rates were consistent with a 2002 national drug-resistance survey (Thailand Ministry of Public Health, unpublished data, 2002). This component of the project was designed to determine, first, whether it is feasible to provide culture and drug-susceptibility testing for all TB patients and, second, whether it is worth the money and effort. This project has demonstrated the feasibility of increasing province-level laboratory capacity to culture mycobacteria; however, before further conclusions can be drawn, more work is needed to increase the proportion of patients whose samples are cultured and tested for drug susceptibility, to standardize laboratory methods and to ensure that laboratory results are integrated into patient management. Further analysis of data from this project may help inform international policies on the expansion of laboratory capacity to culture mycobacteria, which is a major source of debate given the paucity of data about whether such an expensive and technically demanding intervention can aid TB control in high-burden resource-limited settings.13
This report is subject to several limitations. First, data were collected as part of a surveillance and programme-monitoring project; we searched for and removed duplicate records by comparing names and various demographic and clinical factors, but rigorous attempts to verify the accuracy and completeness of data were not made. Based on audits by programme-monitoring staff, we believe it is unlikely that our results are affected greatly by recording error. Second, laboratory methods were not standardized throughout the period reported, which limits our ability to draw firm conclusions about the performance of laboratories. Third, baseline estimates were calculated from information collected through the national TB programme and its implementation partners. Other reporting systems, such as those for all communicable diseases and for AIDS, collect data about TB in Thailand but were not used because variability in case definitions and absence of critical data (for example, bacteriology, registration status) make it impossible to compare data accurately. Fourth, we reported only about services delivered to patients, not outcomes of TB treatment. This data will be reported separately because of the need to present detailed stratification by various patient characteristics.
Direct costs associated with implementing this project – including technical assistance but excluding the costs of medicines and clinical care routinely paid for by the public health system (for example, medical exams, microscopy, radiography) – were approximately US$ 80 per TB case (in 2004 US dollars). WHO has estimated the cost per patient treated through the national programme in Thailand to be about US$ 170; other studies from 1997 and 2002 have calculated that direct provider costs are substantially higher (US$ 200–350 per case).4,14,15 Therefore, this programme required an investment of 20–50% more than is currently spent per TB case. The high cost of this programme can be explained, in part, by the expense of starting up a complex large-scale externally funded project and of supporting the development of capacity for mycobacterial culture.
This demonstration project has already led to important outcomes in Thailand, including adoption of a national policy for HIV counselling and the testing of TB patients, modification of the national TB programme’s recording and reporting system to include HIV-related variables and non-Thai status, expansion of public–private partnerships in Bangkok to seven districts, and training in collaboration between TB and HIV programmes for public health officials from nine different Asian countries. This project demonstrates that implementing activities consistent with the new Global Plan to Stop TB is feasible and has had a high yield in one resource-limited country with a high burden of TB. Whether these activities are also cost effective and whether results would be similar in other countries with a high burden of TB is not known, but this project offers an important proof of principle to inform global TB control strategies. ■
- MC Raviglione, A Pio. Evolution of WHO policies for tuberculosis control, 1948–2001. Lancet 2002; 359: 775-80.
- C Dye, CJ Watt, DM Bleed, SM Hosseini, MC Raviglione. Evolution of tuberculosis control and prospects for reducing tuberculosis incidence, prevalence, and deaths globally. JAMA 2005; 293: 2767-75.
- World Health Organization. The Global Plan to Stop TB 2006–2015: actions for life towards a world free of tuberculosis. Geneva: WHO; 2006 (WHO/HTM/STB/2006.35).
- World Health Organization. Global tuberculosis control: surveillance, planning, financing. WHO report 2005. Geneva: WHO; 2005. (WHO/HTM/TB/2005.349).
- World Health Organization, Regional Office for South-East Asia. Third review of the national tuberculosis programme in Thailand, 17–28 November 2003. New Delhi: WHO; 2004 (SEA-TB-267).
- World Health Organization. Public–private Mix for DOTS: practical tools to help implementation. Geneva: WHO; 2003 (WHO/CDS/TB/2003.325).
- DJ Martin, JG Sim, GJ Sole, L Rymer, S Shalekoff, AB van Niekirk, et al. CD4+ lymphocyte count in African patients co-infected with HIV and tuberculosis. J Acquir Immune Defic Syndr Hum Retrovirol 1995; 8: 386-91.
- R Teck, O Ascurra, P Gomani, M Manzi, O Pasulani, J Kusamale, et al. WHO clinical staging of HIV infection and disease, tuberculosis and eligibility for antiretroviral treatment: relationship to CD4 lymphocyte counts. Int J Tuberc Lung Dis 2005; 9: 258-62.
- L Morris, DJ Martin, H Bredell, SN Nyoka, L Sacks, S Pendle, et al. Human immunodeficiency virus–1 RNA levels and CD4 lymphocyte counts, during treatment for active tuberculosis, in South African patients. J Infect Dis 2003; 187: 1967-71.
- L Abouya, IM Coulibaly, SZ Wiktor, D Coulibaly. The Côte d’Ivoire national HIV counseling and testing program for tuberculosis patients: implementation and analysis of epidemiologic data. AIDS 1998; 12: 505-12.
- M Badri, D Wilson, R Wood. Effect of highly active antiretroviral therapy on incidence of tuberculosis in South Africa: a cohort study. Lancet 2002; 359: 2059-64.
- S Woldehanna, J Volmink. Treatment of latent tuberculosis infection in HIV infected persons. Cochrane Database Syst Rev 2004; 1: CD000171-.
- Stop TB New Diagnostics Working Group. Strategic plan 2006-2015. Available at: http://www.stoptb.org/wg/new_diagnostics/assets/documents/SP%20Stop%20TB%20Dia%20WG%20-FINAL-Dec2005.pdf
- H Sawert, S Kongsin, V Payanandana, P Akarasewi, PP Nunn, MC Raviglione. Costs and benefits of improving tuberculosis control: the case of Thailand. Soc Sci Med 1997; 44: 1805-16.
- P Kamolratanakul, N Hiransithikul, N Singhadong. Cost analysis of different types of tuberculosis patients at tuberculosis centers in Thailand. Southeast Asian J Trop Med Public Health 2002; 33: 321-30.
- US Centers for Disease Control and Prevention, 1600 Clifton Road, Atlanta, GA 30333, USA.
- Thailand Ministry of Public Health, Nonthaburi, Thailand.
- Thailand Ministry of Public Health–US CDC Collaboration, Nonthaburi, Thailand.
- Chiang Rai Provincial Public Health Office, Chiang Rai, Thailand.
- Office of Disease Prevention and Control 7, Ubon-ratchathani, Thailand.
- Phuket Provincial Public Health Office, Phuket, Thailand.
- Bangkok Metropolitan Administration, Bangkok, Thailand.
- Research Institute of Tuberculosis, Tokyo, Japan.