Sharing essential knowledge with health researchers and policy makers

Communities of practice

Tuberculosis And Hiv: Operational Challenges Facing Collaboration And Integration

21 Jan 2008

Source: WHO/TDR

Simon J Tsiouris ¹ , Neel Gandhi ² , Wafaa El-Sadr ³ , Gerald Friedland ⁴

¹International Center for AIDS Care and Treatment Programs (ICAP), Mailman School of Public Health, Columbia University, USA
²Division of Infectious Diseases, School of Medicine, Emory University, USA
³ICAP, Mailman School of Public Health, Columbia University, USA
⁴Dept of Medicine, Epidemiology and Public Health, School of Medicine, Yale University, USA

Abstract

Summary

The global HIV epidemic is fuelling an extraordinary increase in tuberculosis (TB) cases, and in the number of patients co-infected with both diseases. The management of TB/HIV co-infected patients is often fragmented, with little coordination of care between TB and HIV treatment programmes in many areas and at many levels. Care is further hampered by limitations of current TB diagnostic methods, limited access to and underutilization of HIV counselling and testing services, and a paucity of data regarding optimal regimens and timing of initiation of antiretroviral therapy. Collaboration between TB and HIV programmes and integration of services has been advocated, but has been hindered by a history of independent structures and functions of established national TB programmes and newly established HIV programmes, and by the inadequacies of primary care and general health care services in many countries. The substantial morbidity and mortality among TB/HIV co-infected patients necessitates a renewed focus on operational research to identify practical diagnostic tools and treatment strategies, in addition to new models of collaboration and integration between TB and HIV programmes and services.

Working paper contributed to the Scientific Working Group on Tuberculosis Research, Special Programme for Research and Training in Tropical Diseases (3–6 October 2005)

Source: http://www.who.int/tdr/publications/publications/swg_tub.htm

Introduction

Tuberculosis and human immunodeficiency virus (HIV) disease have been closely entwined since the early years of the HIV/AIDS epidemic. The two conditions interact synergistically in their epidemiologic characteristics and clinical manifestations and are clothed in stigma. Both conditions carry the broader risk of creating social, economic and political instability. The overwhelming share of the human toll of disease due to both TB and HIV is borne by resource-limited countries [1]. In sub-Saharan Africa, HIV is fuelling the TB epidemic, with the incidence rate of TB increasing from 146 per 100 000 in 1990 to 345 per 100 000 in 2003 [2]. TB is now one of the leading causes of morbidity and mortality among persons with HIV disease and, in some countries, the percentage of patients with active TB who are co-infected with HIV has reached 60% [3]. Even with appropriate management of TB, patients with HIV co-infection can have higher mortality, which may be due to concomitant conditions rather than lack of effectiveness of TB therapy [4].

The increased number of TB/HIV co-infected patients in Africa, where the co-epidemic has the greatest impact, as well as in other resource-limited areas has intensified the need to find solutions to diagnostic, therapeutic and management issues at the interface of both diseases. Knowledge and experience in the diagnosis and management of TB and HIV as separate diseases is extensive. In contrast, knowledge and experience in the diagnosis and management of co-existing HIV and TB is quite limited. While creating additional challenges, this nonetheless offers the prospect of enhancing the management of both diseases through collaborative and integrative efforts [5].

From a public health perspective, national TB control programmes (NTPs) and national AIDS control programmes (NACPs) need to develop mechanisms of collaboration which will promote interventions for disease control at a population level. The TB/HIV Working Group of the Stop TB Partnership has developed interim recommendations and policies [5]. Several demonstration projects have been carried out to assess the feasibility of such collaborative and integrative efforts in urban and rural areas [6,7], and some programmes have begun to adopt the public health oriented strategies of TB care in newly developed HIV treatment programmes [8].

At the same time, health care delivery sites, especially primary health care centres, need to be able to provide appropriate care for the TB/HIV co-infected patient through integration of their often separated TB and HIV services. In some settings this has already occurred, such as when the same primary health care provider is responsible for TB and HIV services. In many others, the delivery of HIV diagnosis, care and treatment has remained separate from services for TB. However, successful integration of TB and HIV services will not reliably enable the primary health care system to provide satisfactory care. Strengthening of the primary health care system beyond the TB and HIV services is also needed to allow for prompt recognition of HIV and TB related signs and symptoms with proper referral as appropriate for further diagnosis and management when and where necessary. Without general strengthening, efforts to control this co-epidemic are not likely to succeed.

Arguably, some of the greatest challenges in achieving effective programmatic collaboration and service integration are in the identification of practical approaches to diagnostic and therapeutic issues in the TB/HIV co-infected patient, with the goal of reducing morbidity, mortality and incidence of both active TB and HIV disease. To meet these challenges, interventions must be designed and then tested to determine their clinical and/or public health impact. Traditional clinical research strategies, such as randomized clinical trials, are a necessary part of the process of identifying answers to many questions. However, such strategies can be protracted and their results can have limited generalizability. The judicious and timely use of operational research (defined herein as the application of analytic techniques to available and modelled data to quantify or predict the impact of interventions on individual clinical outcomes, the benefits of interventions at a population level, and the ratio of benefit to cost for various interventions) is also needed to reach practical solutions to unanswered questions. An example of this is the mathematical modelling of different interventions and their impact on TB incidence which has been described [9], but more needs to be done with currently available data and data from new studies.

In this communication, we illustrate several of the treatment, diagnosis and management issues relevant to TB/HIV co-infection. Additionally, we describe the questions which need to be answered in order to achieve optimal patient outcomes through better collaboration between TB and HIV programmes and greater integration of TB and HIV services at the service delivery site with the net effect of promoting a decrease in the burden of disease created by the co-epidemic. Throughout, we recognize the need for flexibility since solutions may vary by setting depending upon prevalence of HIV and TB disease, local resources, history and current practice.

Diagnosis of hiv in patients with active tb

The availability and offering of HIV counselling and testing to patients with TB is an essential starting point for effective management of both conditions [10]. However, it is not clear how best to provide this service. Barriers to delivering this exist at both the national and service delivery levels including programmatic, provider-based and patient-related barriers.

National programmatic barriers include the emphasis on stand-alone HIV voluntary counselling and testing (VCT) sites by some health ministries and a reluctance to provide access to HIV counselling and testing through other sites. Although access to HIV counselling and testing is increasing, with the establishment of HIV counselling and testing services in health care settings, its continued general lack has delayed or blocked TB patients' access to HIV treatment. As access to HIV counselling and testing slowly expands, TB treatment sites face site-level programmatic barriers to providing this service to their patients including lack of space to provide confidential counselling and testing in most TB treatment sites and poor communication between TB treatment sites and off-site HIV counselling and testing services.

Provider-related barriers include the lack of appreciation of the importance of HIV status on their patients' outcomes, lack of skills in the management of HIV disease, and a reluctance to discuss HIV, another stigmatizing disease, with their patients. TB providers who do not already provide HIV services need to be educated about the relationship between TB and HIV and involved in the delivery of these services through a multi-disciplinary team approach to care. On the other hand, barriers among patients with TB include the fear that, if HIV infected, they would face further discrimination and, until recently, in the absence of treatment for HIV disease, their perception of no benefit of knowing their status. As access to HIV care and treatment becomes more widely available in countries where HIV and TB prevalence is high, it is critical to address programmatic, provider and patient barriers to HIV testing in order for co-infected patients to benefit from the scale-up of HIV programmes. Box 1 highlights some of the unresolved challenges in the collaboration and integration of TB and HIV programmes and services.

Introducing HIV counselling and testing for TB patients in settings where antiretroviral therapy (ART) is available has been shown to be feasible with a high uptake of testing by TB patients [11,12], but other strategies to promote uptake of testing among TB patients need to be evaluated, especially in settings where ART is not yet available. Determining what measures will ensure the highest acceptability and uptake of HIV counselling and testing by TB patients is a vital area of study. How to best implement HIV counselling and testing effectively on a wide scale at TB clinics remains unclear. For example, should TB programmes integrate HIV counselling and testing as a component of the services they provide, as has been advocated, or should patients be referred elsewhere for this service? Referring patients elsewhere can result in low uptake of testing, but in some settings, especially those without adequate counselling space or staff, this, along with increasing communication between TB and HIV service providers, may be the only feasible option. Operational research efforts that assess strategies to implement universally available HIV counselling and testing for all TB patients are needed.

If HIV counselling and testing is to be performed at TB treatment sites, should TB providers themselves be trained to provide counselling and testing to their patients or should national TB programmes or VCT sites provide specific counsellors for this particular service to take place at TB treatment sites? There may not be one answer to this question, with the best method varying depending on the volume of patients a site cares for and the number of TB and other staff available.

HIV counselling and testing through an ‘opt out’ as opposed to an ‘opt in’ strategy has been advocated in TB treatment sites [13]. In ‘opt-in’ testing, patients are offered counselling and testing and must actively provide consent. In contrast, counselling and testing is part of routine care in ‘opt-out’ testing, and patients are informed of the right to refuse the test. A study of counselling and testing in antenatal clinics has demonstrated that an ‘opt-out’ strategy tests a greater percentage of patients [14]. Despite this, some national TB/HIV coordinating bodies are not utilizing the ‘opt out’ approach, fearing possible infringement of individual human rights. In this case, the ‘opt in’ strategy may be the only choice and methods to maximize uptake of testing using this strategy need to be studied. Studies to assess HIV knowledge and attitudes among TB providers and patients, and the impact of different attitudes on utilization of these services, are also needed.

The components of care which follow a positive HIV test in a TB patient and how they are best delivered will need to be studied using operational research, including: when and by whom HIV clinical staging is best done, where the collection and measurement of CD4+ T-cell counts and HIV viral loads (if available) are completed, who begins the administration of cotrimoxazole preventive therapy and how is it continued after completion of TB treatment, and what mechanisms for referral into HIV care and treatment work best in which settings to minimize loss to follow-up. Outcomes research will be necessary to determine which of the possible service delivery methods of these components are associated with the greatest reductions in morbidity and mortality. This should be accompanied by cost-effectiveness research of the various programmatic components to determine if the components are within the resource capacity of the local health care delivery site and national programmes.

Diagnosis of tb and latent tb infection in hiv-infected patients

HIV co-infection complicates clinical presentations and diagnosis of active TB and limits the performance of the acid fast bacilli sputum smear, the most widely available TB diagnostic method in resource-limited settings [15,16]. As a consequence of the high rates of sputum smear negativity, diagnosis of TB may be missed or delayed by health care providers at HIV treatment sites. This increases the risk of morbidity and mortality for individual patients as well as the risk of transmission of TB to other patients, staff, household and community members. Sputum culture, while useful in patients with smear negative disease is not routinely used in resource-limited settings due to cost constraints. In addition to the increased complexity of diagnosing active TB, HIV-infected individuals who have latent TB infection (LTBI) also have a higher risk of reactivation than HIV-uninfected individuals with LTBI [17,18] due to the reduced sensitivity of the tuberculin skin test (TST), especially in those with lower CD4 T-cell counts [19,20], resulting in missed opportunities for providing preventive therapy.

Once diagnosed with active TB or LTBI, the difficulties do not end for the health care provider. To ensure treatment for the HIV-infected patient with newly diagnosed TB, a mechanism of referral to the TB treatment site must be in place with a strong communication system and contingencies for finding those patients lost to follow-up. Establishing methods to prevent loss to follow-up is essential and will vary depending on many factors including local geography and community structure. Similarly, treating the HIV-infected patient found to have LTBI will require programmatic support for the implementation of a new service requiring 6–9 months of therapy with isoniazid. Whether isoniazid preventive therapy (IPT) is best administered by the HIV care and treatment site or the TB treatment site is an open question and may vary depending on individual capabilities and resources. The timely answers to these questions are not likely to come from traditional clinical research methods but more readily through site-specific operational research.

Generally speaking, current methods for diagnosis of TB and LTBI in patients with HIV disease are suboptimal and there is great need for new rapid, effective and affordable diagnostic methods. That said, there are major discrepancies between the technologies used in well-resourced as opposed to resource-limited settings; the former having access to culture and nucleic-acid amplification technologies and the latter dependant on smears. These differences are not only due to the prohibitive cost of some technologies but also, in part, reflect a lack of commitment on the part of health ministries and NTPs to introduce modern technologies and upgrade laboratory systems in general.

There have been some advances in diagnostics for active TB and LTBI including detection of TB-specific antigens or antibodies [21–23] and tests based on interferon-γ production by T-cells when stimulated with TB specific antigens [24,25]. However, research is needed to determine the utility and cost-effectiveness of these and other new TB diagnostic tests in resource-limited settings with high HIV seroprevalence. Progress in this area has been hindered by the fact that, to date, no internationally accepted guidelines for the assessment and evaluation of TB diagnostics are available [26]. Until new tools are widely available, ways to maximize the effectiveness of old tools must be studied and then implemented. For example, the use of a screening questionnaire and sputum smear have been shown to effectively rule out active TB in HIV clinics [27]. Despite these types of studies, little attention has thus far been paid to implementing active TB case finding by HIV programmes. Similarly, screening for LTBI and use of IPT is not a component of most HIV care and treatment programmes despite its proven clinical effectiveness [28] and probable cost effectiveness [29]. National programmes, with the exception of ongoing efforts in Botswana, have not embraced this intervention, often because of concerns about initiation of treatment for LTBI with isoniazid in patients who actually have underlying active TB disease, understandable given recent reports of high rates of sub-clinical TB in HIV-infected patients [30]. The combined application of clinical and operational research is needed to determine what are the most efficacious and cost-effective TB screening algorithms in combination with new and old diagnostic tools as well as the methods for providing full access to IPT for HIV-infected patients found to have LTBI. These algorithms and tools then need to be modified to fit individual site circumstances.

Treatment of hiv-infected patients with tb

Concurrent treatment of TB disease and HIV disease presents practical and logistic questions including: 1) when in the course of TB treatment should antiretroviral (ARV) therapy, if indicated, be started; 2) what ARV regimen should be used; and 3) how should these patients be monitored for toxicity and clinical outcomes?

It is universally recommended that antituberculous treatment be promptly initiated upon diagnosis of TB in an effort to decrease mortality and risk of transmission of M. tuberculosis to others [31,32]. Additionally, emphasis has been placed on starting ARV therapy, when indicated, during the TB treatment phase due to the high morbidity and mortality in patients with HIV-related TB co-infection even when receiving appropriate antituberculous treatment [33]. WHO guidelines, based on expert opinion, indicate that the timing of initiation of ARV therapy in these patients should be based on the degree of immunosuppression as defined by CD4 cell count [34]. Although the proposed algorithm provides a reasonable approach, data supporting the current guidelines are lacking regarding the risks, including drug toxicities, and the benefits of starting ARVs at different levels of immunosuppression.

Several factors may complicate concurrent use of antiretroviral (ARV) and antituberculous drugs including: additive toxicities of medications, drug interactions, risk of immune reconstitution events, and difficulty in adherence with multiple medications [35]. The choice of ARVs is severely limited by the need for concurrent use of rifampin due to the latter's effect on levels of protease inhibitors and non-nucleoside reverse transcriptase inhibitors (NNRTIs). Nevirapine-based regimens are the primary ones used in most national antiretroviral treatment programmes. The safety and efficacy of concomitant use of nevirapine-based regimens in conjunction with rifampin requires further study. Efavirenz-based regimens have been more extensively studied and are recommended in this scenario [31,32]. However, the precise dose of efavirenz in this scenario remains to be determined due to rifampin's effect on efavirenz metabolism [36].

Both pharmacokinetic and clinical studies are urgently needed in patients receiving concomitant antituberculous and antiretroviral drugs, particularly studies focused on antiretroviral drugs that are currently recommended and widely available in resource-limited settings. A few studies with limited numbers of participants have assessed co-administration of these agents with encouraging preliminary results [6,7,37]. However, studies with larger numbers of participants are needed to evaluate the concomitant use of these various regimens and to assess their effects on laboratory and clinical outcomes [38]. The feasibility and effectiveness of rifabutin in resource-limited settings needs urgent evaluation [39]. Although the rifabutin dosage needs to be adjusted when administered with ARVs, its use does not preclude the utilization of many of the currently available ARV treatment options as does the use of rifampin.

Management of the tb/hiv co-infected patient

Current national TB and HIV programmes remain largely separate with varying levels of interaction and communication. However, these programmes are often situated within the structure of the ministry of health. Thus, ministries of health can have an important role in ensuring communication and collaboration between the two programmes. Collaboration at the central programmatic level will facilitate similar collaboration and communication at the service delivery sites, resulting in better services for co-infected patients. Fig. 1 depicts stylized representations of two different paradigms for interactions between HIV and TB programmes and service delivery sites: an all too common paradigm and an alternative paradigm. The common paradigm is characterized by separate and distinct programmes with little coordination or overlap. However the alternate paradigm is focused on the co-infected patient rather than on the two diseases in isolation, which would improve the care and treatment of TB/HIV co-infected patients through increased collaboration and communication between programmes and integration of services. Achievement of this alternative paradigm requires intensive assessment of various models of collaboration and integration between TB and HIV programmes and services. These models may range from maintenance of separate programme services with enhanced communication and referral mechanisms to partial or full integration of programmes and the services they provide.

Figure 1

Common and alternative TB and HIV programme paradigms. The alternative paradigm features increased communication and collaboration at national level, and partial, site appropriate, integration of services among TB, HIV and primary care services.

Box 1 Unresolved questions in the collaboration and integration of TB and HIV programmes and services

Availability and uptake of HIV counselling and testing by TB patients

How should HIV counselling and testing services be linked with TB diagnosis and treatment sites?
How should HIV counselling and testing for TB patients be offered?
- Routinely?
- Provider based?
- Opt out vs. opt in?

Diagnosis of TB and LTBI in HIV-infected patients

What are the most efficacious algorithms for diagnosis of active TB and LTBI in HIV-infected patients?
- Site-specific and cost-effective
- In screening for LTBI
- In active case finding for TB
What combination of new and old technologies will most rapidly diagnose TB and LTBI in HIV-infected patients?

Treatment of patients with TB/HIV

What is the optimal time to start antiretroviral therapy in a patient with active TB?
Where should co-treatment of TB and HIV disease be delivered?
- TB treatment sites?
- HIV care and treatment sites?
- Primary care sites?
Who should provide co-treatment of TB and HIV disease?
- The role of non-physician health care workers
- The role of community and family
What are the training needs for TB and HIV and primary care staff?
What are the best mechanisms to support adherence?
- What is DOT's role in the treatment of TB/HIV co-infected patients?
What is the optimal antiretroviral regimen to use during co-treatment for TB?
- In rifampin based TB regimens
- In rifabutin based TB regimens
What is the proper dose of efavirenz when used in conjunction with rifampin?

An example of a collaborative model is one where ARVs are initiated in eligible patients during TB treatment at the TB clinic with subsequent transfer of the patient to the HIV programme after completion of antituberculous therapy. This model would allow for reinforcement of the importance of adherence with ARV therapy through the use of directly observed therapy (DOT) services at the TB clinic and through utilization of other supportive and outreach efforts available through these programmes. Superior outcomes have been noted among patients with HIV-related TB who received their antituberculous treatment through DOT [40]. Availability of the DOT infrastructure at TB treatment programmes allows for its use with ARVs and thus creates a unique opportunity to achieve high rates of adherence with ARV therapy and may provide patients with the skills to maintain self-administered lifelong adherence with ARVs once TB therapy is completed [41]. In addition, close follow-up of patients through the frequent contact required for DOT allows for more careful monitoring of patients to promptly identify adverse effects of ARVs, particularly during the early phase after initiation of these medications. This approach will require training of TB programme staff on management of HIV and its complications, access to ARVs, and availability of laboratory assays such as the CD4 cell count and those needed for monitoring toxicity of the drugs used.

From the perspective of the HIV programmes, a focus on the co-infected patient also requires establishing models of collaboration. For patients with HIV who develop TB while in follow-up at HIV care and treatment programmes, these programmes may also need to expand their services to include provision of TB treatment and use of DOT. Similar to the scenario in the TB clinics described above, this approach requires training of HIV staff on TB diagnosis and management as well availability of antituberculous medications and other TB-specific resources.

Training of providers in both HIV and TB diagnosis and management can assist in the effective use of clinical staff. However, the scope of the HIV and TB epidemics requires assessment of the feasibility and effectiveness of new cadres of healthcare workers, particularly in settings with a paucity of nurses or physicians such as DOT workers and treatment supporters [42]. Similar efforts are needed to determine how to provide community care models for the co-treatment of TB and HIV. The use of community care models for delivery of TB treatment to HIV-negative and HIV-infected patients has been studied and shown to be effective [43,44]. In many resource-limited settings, patients' families and communities provide resources not available to many HIV patients in developed countries [45,46], and some early data suggest that use of these facilitators may be associated with favourable clinical and virologic outcomes in patients with both TB and HIV disease in need of treatment [7,47].

Thus, in order to identify the optimal models and strategies for managing patients with HIV and TB, studies are needed to compare the feasibility, outcomes and cost effectiveness of specific collaborative and integrative approaches. It is however clear that one model may not fit all countries. Prevalence of HIV and TB, demographic characteristics of the patients, available resources, existing programmatic structures, types of provider available and levels of expertise vary, and are all factors which influence the choice of model of care. Nor, for that matter, may one model fit each country, as differences in rural and urban sites of care, needs of special populations such as intravenous drug users, and varying levels within the healthcare system from primary to specialty clinics and tertiary care sites may require different degrees of collaboration and integration. Different models and strategies will need to be developed and evaluated, and their collaborative and integrative approaches compared.

Making it happen

It would be naive to believe and argue that a collaboration strategy could be uniformly and rapidly accomplished throughout all areas in which HIV and TB epidemics overlap. Indeed it is important to recognize that, in most areas, TB programmes have insufficient structural and human resources in the face of rising caseloads driven by the HIV epidemic. Furthermore, national TB programmes in many cases do not have the authority to enforce high-level collaboration. Thus successful collaboration requires a commitment by national leadership in ministries of health and a devotion to achieving this goal. Infusion of resources is needed to strengthen the TB programmes, to expand the availability of trained TB providers and DOT supporters, and to improve facilities for diagnosis, treatment and integrated care for TB/HIV co-infected patients. Similarly, efforts must also focus on HIV programmes. These programmes are inundated with patients seeking care and treatment. They face enormous obstacles including the need for trained staff, extensive laboratory resources, and availability of ARVs, as well as the challenge of achieving excellent lifelong adherence with medications by their patients. Thus there is a specific urgency to undertake studies that address TB-related issues within HIV care and treatment programmes, lest in the understandable haste to establish these programmes, the opportunities to determine the most effective means for TB and HIV collaboration and integration are missed. Fundamentally, without a general strengthening of the primary healthcare system, the best intentioned efforts to improve collaboration and service delivery for TB and HIV co-infected individuals may fail.

Collaboration between HIV and TB care and treatment programmes has been hampered by their separate traditions and practices. TB programmes are characterized by their public health approach, with firmly established algorithms and standardized measures and outcomes. On the other hand, HIV care and treatment derives from a focus on the individual patient and is characterized by rapidly evolving treatment paradigms that must be rapidly incorporated into treatment guidelines. The nuances, subtleties and added complexities of TB diagnosis and treatment in the context of HIV co-infection must be recognized, as must the need for a large-scale public health approach for the management of HIV in resource-limited settings. Each discipline needs to accommodate the other. For the TB world, HIV should no longer be seen as an intruder and must be accepted as part of the current and future reality. For the HIV world, the accumulated experience acquired over the longer history of TB must be valued and can serve as source of important lessons. How best to harmonize these two basic approaches is at the core of the proposed research priorities delineated in this communication. Addressing the outstanding unanswered questions in the management of the two diseases in a spirit of accommodation and collaboration will greatly benefit patients with TB and HIV and establish a new paradigm for the future.

References

1. Dye C, etal (1999) Consensus Statement. Global burden of tuberculosis: estimated incidence, prevalence, and mortality by country. WHO Global Surveillance and Monitoring Project. J Am Med Assoc 282: 677-686.

2. World Health Organization, (2005) WHO Report 2005. Global TB control - surveillance, planning, financing Geneva, Switzerland:World Health Organization . WHO/HTM/TB/2005.349

3. World Health Organization, (2004) Country profile South Africa. Global tuberculosis control - surveillance, planning financing Geneva, Switzerland:World Health Organization . WHO/HTM/TB/2004.331

4. Wilkinson D, Davies GR, (1997) The increasing burden of tuberculosis in rural South Africa - impact of the HIV epidemic. S Afr Med J 87: 447-450.

5. World Health Organization, (2004) Interim policy on collaborative TB/HIV activities Geneva:Stop TB Department and Department of HIV/AIDS, World Health Organization . WHO/HTM/TB/2004.330

6. Jack C, etal (2004) A pilot study of once-daily antiretroviral therapy integrated with tuberculosis directly observed therapy in a resource-limited setting. J Acquir Immune Defic Syndr 36: 929-934.

7. Gandhi N, etal (2004) Initiating and providing antiretroviral therapy for TB/HIV coinfected patients in a rural tuberculosis directly observed therapy program in South Africa: the Sizonqoba study. Medscape general medicine 6: . conference abstract number MoOrB1014

8. Chimzizi RB, etal (2004) Counselling, HIV testing and adjunctive cotrimoxazole for TB patients in Malawi: from research to routine implementation. Int J Tuberc Lung Dis 8: 938-944.

9. Currie CS, etal (2003) Tuberculosis epidemics driven by HIV: is prevention better than cure?. AIDS 17: 2501-2508.

10. World Health Organization, Key conclusions and recommendations: 4th meeting of the Global TB/HIV Working Group of the Stop TB Partnership, 2004, Addis Ababa, Ethiopia . (WHO/HTM/TB/2005.348). Available from: http://www.who.int/tb/publications/aa_meeting_report.pdf

11. World Health Organization, (2003) Report of a “lessons learnt” workshop on the six ProTEST pilot projects in Malawi, South Africa and Zambia Durban, South Africa:WHO . (WHO/HTM/TB/2004.336). Available from: http://whqlibdoc.who.int/hq/2004/WHO_HTM_TB_2004.336.pdf

12. Godfrey-Faussett P, etal (2002) How human immunodeficiency virus voluntary testing can contribute to tuberculosis control. Bull World Health Organ 80: 939-945.

13. UNAIDS, (2004) UNAIDS/WHO policy statement on HIV testing . Available from: http://data.unaids.org/una-docs/hivtestingpolicy_en.pdf

14. Tay E, Emmanuel S, (2003) Improving the uptake of human immunodeficiency virus (HIV) antenatal screening in a primary healthcare setting. Ann Acad Med Singapore 32: 92-95.

15. De Cock KM, etal (1992) Tuberculosis and HIV infection in sub-Saharan Africa. J Am Med Assoc 268: 1581-1587.

16. Colebunders RL, etal (1989) HIV infection in patients with tuberculosis in Kinshasa, Zaire. Am Rev Respir Dis 139: 1082-1085.

17. Selwyn PA, etal (1989) A prospective study of the risk of tuberculosis among intravenous drug users with human immunodeficiency virus infection. New Engl J Med 320: 545-550.

18. Daley CL, etal (1992) An outbreak of tuberculosis with accelerated progression among persons infected with the human immunodeficiency virus. An analysis using restriction-fragment-length polymorphisms. New Engl J Med 326: 231-235.

19. Johnson MP, etal (1992) Tuberculin skin test reactivity among adults infected with human immunodeficiency virus. J Infect Dis 166: 194-198.

20. Portu JJ, etal (2002) Tuberculin skin testing in intravenous drug users: differences between HIV-seropositive and HIV-seronegative subjects. Addict Biol 7: 235-241.

21. Singh KK, etal (2001) Antigens of Mycobacterium tuberculosis expressed during preclinical tuberculosis: serological immunodominance of proteins with repetitive amino acid sequences. Infect Immun 69: 4185-4191.

22. Singh KK, etal (2003) Combined use of serum and urinary antibody for diagnosis of tuberculosis. J Infect Dis 188: 371-377.

23. Perkins MD, etal (2003) Serologic diagnosis of tuberculosis using a simple commercial multiantigen assay. Chest 123: 107-112.

24. Mori T, etal (2004) Specific detection of tuberculosis infection: an interferon-gamma-based assay using new antigens. Am J Respir Crit Care Med 170: 59-64.

25. Mazurek GH, etal (2001) Comparison of a whole-blood interferon gamma assay with tuberculin skin testing for detecting latent Mycobacterium tuberculosis infection. J Am Med Assoc 286: 1740-1747.

26. O'Brien R, (2004) Perils and pitfalls in clinical trials of new TB diagnostics. Int J Tuberc Lung Dis 9: S19.

27. Mohammed A, etal (2004) Screening for tuberculosis in adults with advanced HIV infection prior to preventive therapy. Int J Tuberc Lung Dis 8: 792-795.

28. Woldehanna S, Volmink J, (2004) Treatment of latent tuberculosis infection in HIV infected persons. Cochrane Database Syst Rev CD000171.

29. Bell JC, Rose DN, Sacks HS, (1999) Tuberculosis preventive therapy for HIV-infected people in sub-Saharan Africa is cost-effective. AIDS 13: 1549-1556.

30. Mtei L, etal (2005) High rates of clinical and subclinical tuberculosis among HIV-infected ambulatory subjects in Tanzania. Clin Infect Dis 40: 1500-1507.

31. World Health Organization, (2003) Treatment of tuberculosis: guidelines for national programmes . WHO/CDS/TB/2003.313

32. Blumberg HM, etal (2003) American Thoracic Society/Centers for Disease Control and Prevention/Infectious Diseases Society of America: treatment of tuberculosis. Am J Respir Crit Care Med 167: 603-662.

33. World Health Organization, (2004) TB/HIV: a clinical manual Geneva:World health Organization . WHO/HTM/TB/2004.329

34. World Health Organization, (2002) Scaling up antiretroviral therapy in resource-limited settings: guidelines for a public health approach. IAPAC Mon 8: 168-175.

35. Dean GL, etal (2002) Treatment of tuberculosis in HIV-infected persons in the era of highly active antiretroviral therapy. AIDS 16: 75-83.

36. Lopez-Cortes LF, (2002) Pharmacokinetic interactions between efavirenz and rifampicin in HIV-infected patients with tuberculosis. Clin Pharmacokinet 41: 681-690.

37. Dheda K, etal (2004) Outcome of HIV-associated tuberculosis in the era of highly active antiretroviral therapy. J Infect Dis 190: 1670-1676.

38. Friedland G, etal (2004) Utility of tuberculosis directly observed therapy programs as sites for access to and provision of antiretroviral therapy in resource-limited countries. Clin Infect Dis 38: S421-S428.

39. Centers for Disease Control and Prevention, (2000) Updated guidelines for the use of rifabutin or rifampin for the treatment and prevention of tuberculosis among HIV-infected patients taking protease inhibitors or nonnucleoside reverse transcriptase inhibitors. MMWR 49: 185-189.

40. Alwood K, etal (1994) Effectiveness of supervised, intermittent therapy for tuberculosis in HIV-infected patients. AIDS 8: 1103-1108.

41. Jack C, etal (2004) Long term follow-up of pilot project to integrate antiretroviral therapy into an existing tuberculosis directly observed therapy program in a resource poor setting. Medscape general medicine 6: . conference abstract number MoPeB3355

42. El-Sadr W, Medard F, Dickerson M, (1995) The Harlem family model: a unique approach to the treatment of tuberculosis. J Public Health Manag Pract 1: 48-51.

43. Sinanovic E, etal (2003) Cost and cost-effectiveness of community-based care for tuberculosis in Cape Town, South Africa. Int J Tuberc Lung Dis 7: S56-S62.

44. Miti S, etal (2003) Integration of tuberculosis treatment in a community-based home care programme for persons living with HIV/AIDS in Ndola, Zambia. Int J Tuberc Lung Dis 7: S92-S98.

45. Farmer P, etal (2001) Community-based treatment of advanced HIV disease: introducing DOT-HAART (directly observed therapy with highly active antiretroviral therapy). Bull World Health Organ 79: 1145-1151.

46. Gandhi NR, etal (2004) Developing strategies to implement HAART in high TB prevalence communities in resource constrained settings in rural South Africa. Medscape general medicine 6: . conference abstract number TuPpE2055

47. Koenig SP, Leandre F, Farmer PE, (2004) Scaling-up HIV treatment programmes in resource-limited settings: the rural Haiti experience. AIDS 18: S21-S25.