Expanding beyond DOTS to stop TB

This article originally appeared in HIV & AIDS treatment in practice, an email newsletter for healthcare workers and community-based organisations in resource-limited settings published by NAM between 2003 and 2014.
This article is more than 18 years old. Click here for more recent articles on this topic

By Theo Smart

Last October in Paris, it was clear that changes were afoot in world of tuberculosis — that the usually conservative tuberculosis research and treatment community was in the process of making a break from its past, and trying to open itself up to new ideas, new voices and especially new resources. The International Union against Tuberculosis and Lung Disease (the Union) had asked Zachie Achmat, the HIV & AIDS (& now TB) activist to give the opening speech at its annual meeting World Conference on Lung Health (the World Conference). During his talk, he proceeded to describe a new way forward for TB control, by working with the patient as a partner — and a valuable resource to be nurtured — rather than treating people as public health cases.

He and other activists at the meeting attacked the very ‘vertical’ pillar of TB control strategy, DOTS (directly observed therapy, short course), as being paternalistic and antithetical to patient and community empowerment — which is needed to mobilise the political and funding support needed to bring TB control into the 21st century (see http://www.aidsmap.com/cms1001462.asp).



A specimen of tissue or other material taken from part of the body and smeared onto a microscope slide for examination. A Pap smear is a specimen of material scraped from the cervix (neck of the uterus) examined for precancerous changes.


Affecting the lungs.


active TB

Active disease caused by Mycobacterium tuberculosis, as evidenced by a confirmatory culture, or, in the absence of culture, suggestive clinical symptoms.


The part of the body below the chest, including the stomach, liver, intestines, kidneys, bladder, ovaries and uterus. The word ‘abdominal’ relates to pain or other problems in that area.


In a bacteria culture test, a sample of urine, blood, sputum or another substance is taken from the patient. The cells are put in a specific environment in a laboratory to encourage cell growth and to allow the specific type of bacteria to be identified. Culture can be used to identify the TB bacteria, but is a more complex, slow and expensive method than others.

The DOTS strategy, which has indeed limited TB’s spread in much of the world, had been introduced about a decade earlier and was built around five key activities:

  • Political commitment to provide the necessary resources to control tuberculosis
  • Passive detection of active infectious TB cases by sputum smear microscopy
  • A system to ensure regular drug supplies
  • A standard recording and reporting system that monitors and tracks cases as well as assessing treatment outcomes
  • A standard short-course of at least four drugs including rifampicin, administered by directly observed therapy

The DOTS strategy also set explicit targets of identifying 70% of tuberculosis cases and curing 85% of them. If these targets were met, WHO projected that there would be a 40% decrease in infected contacts, a 5-10% decrease in TB incidence and a decrease in the prevalence of TB. And indeed, in most of the world, DOTS has contained TB — but not where HIV has become well established.

“Unfortunately, HIV muddies the pitch,” said Prof. Anthony Harries at the recent Retroviruses Conference in Denver. 

DOTS’s failure in Malawi

Prof. Harries, who is the technical advisor in HIV Care and Support for the Ministry of Health for Malawi and an honorary Professor at the London School of Hygiene and Tropical Medicine, gave a very graphic description of how HIV has led to the failure of DOTS in Malawi (his entire talk on tuberculosis coinfection can be viewed here: http://www.retroconference.org/2006/data/files/retro2006_frameset.htm.

“Malawi has a very good "DOTS" tuberculosis programme [however] we are not winning the battle against tuberculosis,” he said. “HIV negatively impacts on TB control, both indirectly at a programme level and directly at a patient level.”

Between 1985 and 2003, the HIV prevalence among TB cases climbed from 26% to 75% and during the same period, notified TB cases in Malawi rose from 5000 cases per annum to over 25000 per annum.

The increased case burden forced a change in TB treatment policy at country level.

“Malawi, like many other African countries used to keep its TB patients in hospital TB wards for the first two months of the initial phase of TB treatment. Numbers climbed, wards became unbearable and seven years ago, we had to decentralise this care out to the communities and out to health centres. Increased cases mean increased logistics. You need more drugs, you need more stationary, you need more sputum containers and of course you need more money,” said Prof Harries.

He noted that stigma associated with HIV and TB is a particular problem. “If communities think that TB is linked with HIV, then patients may be reluctant to seek care, which results in poor or delayed case findings. And in Malawi — as elsewhere in Africa — people talk of the ‘old TB’ and people talk of the ‘new TB’ (TB/HIV) which the communities say is incurable.”

HIV also weakens healthcare delivery. “Staff are absent because they are sick, staff are absent because they attend funerals and staff die. Hospitals, and in particular medical wards, in sub-Saharan Africa, are dangerous places, not only for patients but also for healthcare workers, largely because of nosocomial TB transmission.”

He presented the results of a survey he carried out in 1999 in all the district and major mission hospitals in Malawi. Over the course of a year, around 62% out of nearly 3000 workers had died, nearly 50% from TB and about 40% died from HIV/AIDS. “HIV/TB was responsible for 90% of the deaths in our healthcare workforce,” he said. He also noted that AIDS was leading to the loss of many of the district TB Officer, the most important member of the TB control programme at peripheral facility level.

Prof. Harries also described some of the direct patient effects of TB/HIV coinfection, including the increased difficulty of diagnosis — particularly in smear negative tuberculosis (these issues have been described recently in HATIP, see http://www.aidsmap.com/cms1001489.asp.

The mortality rate is also shockingly high. Deaths in new smear positive pulmonary TB patients rose from 5% to over 20% per annum over a 12 year period. In one district, where a cohort of TB patients has been followed for ten years since 1995, the 12 month outcomes in new HIV-positive patients showed a death rate of 18% for smear-positive TB, almost 60% for smear-negative pulmonary TB, and almost 40% for extra pulmonary TB. “Out of the HIV-positive/TB patients put together, only 20% are alive, seven years after the diagnosis was made.”

Finally, coinfection leads to high rates of recurrent TB — largely due to reinfection but in some cases also due to re-activation. “This has a spin off on case notifications. Recurrent tuberculosis accounts for 11-12% of our total TB case notifications — therefore it adds to our TB case burden,” Prof. Harries said.

The failure of DOTS in South Africa, and some reasons why

Another presentation at the Retrovirus conference described skyrocketing rates of TB in a peri-urban community in South Africa. (see http://www.aidsmap.com/en/news/CE720DF3-2054-4C37-988C-409FE2B6B33E.asp). In 2005, with an HIV prevalence of 23%,the township had a TB notification rate of over 2,000 per 100,000. 62% of the cases were HIV coinfected.

This study, which has since been published in the April 1st issue of Clinical Infectious Diseases (Lawn 2005) was performed in a community with a well-implemented DOTS programme that achieved cures rates of more than 80% in smear-positive cases of tuberculosis. Nevertheless, tuberculosis rates increased 2.5-fold during the study period.

In an accompanying editorial, Dr Christopher Whalen of Case Western Reserve University writes that it is important to assess where DOTS is failing.  He writes “DOTS is not optimally designed to interrupt the spread of tuberculosis.” He states that there are essentially three factors involved in the transmission of m.TB

  • The number of contacts between an infectious person and susceptible contacts;
  • The probability of infection (or disease) per contact for transmission
  • The duration of infectiousness of the index case.

Treatment of disease only alters one of these key parameters by reducing the duration of infectiousness, but DOTS was not designed to address the number and frequency of contacts or the likelihood of disease after infection. Whalen notes that a single patient with TB is estimated to infect around 10 contacts before receiving treatment. Since people usually have a lifetime risk of developing active tuberculosis of around 10%, levels of the disease would be expected to remain static. But HIV changes the mix by greatly increasing the risk and speeding the developing active tuberculosis as well as reactivation of disease.

“The study from Cape Town illustrates the urgent need to develop novel strategies to act as companions to DOTS. These novel public health strategies should have as their primary goal to interrupt the spread of tuberculosis. Because of the potential effect of HIV infection on tuberculosis control, these strategies must also integrate with the HIV prevention and treatment programs,” concluded Dr Whalen.

STOP TB, the new strategic plan

A new plan was promised in Paris. Now focused on “enhancing and expanding” DOTS, WHO’s STOP TB department has produced a new global plan for tackling tuberculosis for 2006–2015 called Actions for life—towards a world free of tuberculosis (see http://www.stoptb.org//). Released at the World Economic Forum this past January in Davos, the plan is as much a fund-raising appeal as it is a strategic framework on how to meet Millennium Development Goals, to reduce prevalence of and deaths due to TB by 50% by 2015 (relative to 1990) and to eliminate TB as a public health problem (<1 case per million population) by 2050.

Perhaps since part of its purpose is to galvanise political will and secure financing, the plan may be forgiven some of its more optimistic projections.

The vision: by 2010, new drugs that are effective against antimicrobial resistance, that shorten treatment to two to three months, and that are compatible with antiretroviral treatment against HIV

The reality: The two leading new TB drugs (gatifloxacin and moxifloxacin) active against drug-resistant TB and compatible with HIV care are already available for other indications, however, we won’t know whether treatment duration can be shortened until the completion of large-scale studies. The other drugs? Well, time (and study) will tell.

The vision: The first in a series of new, safe, and effective vaccines is expected by 2015.

The reality: Like many vaccines, there is no guarantee that these will be effective in an HIV-infected or immune-compromised population.

The vision: For diagnostics: Rapid culture tests by 2006, a replacement for smear microscopy by 2008, and by 2010, simple, sensitive and rapid diagnostic tests that can be used by rural health workers should be available.

The reality: While some of these methods may indeed be on the cusp of being introduced, most are far from being field tested or shown to be cost-effective in low-resource settings and even further from being implemented on a wide scale in reference labs not to mention peripheral clinics. Affordable point of care tests in the bush? At present, this is wishful thinking.

This is especially true given the funding short-fall for developing new diagnostics. Despite a generous grant from the Bill and Melinda Gates Foundation,the current budget gap for tuberculosis diagnostics is more than 80% of what is needed (a $436 million gap).

So the need to promise deliverables appeal is understandable. However, to borrow a phrase from investor-seeking press releases from the financial services industry, the “forward-looking statements” regarding the diagnostics pipeline border on the insidious and could potentially detract from real-world efforts to improve diagnosis.

For example, when Dr Douglas Wilson’s recent paper on clinical algorithms for improved diagnosis appeared in the International Journal of Tuberculosis and Lung Diseases, it was instantly devalued by an accompanying editorial called “Admitting Defeat” by Perkins and Small, of FIND Diagnostics and the Gates Foundation. And yet, Dr Wilson’s paper is one of the few studies with results that could lead to improved diagnosis of TB, TODAY and not some time in the indefinite future.

Clinical algorithms for diagnosing tuberculosis in HIV-positive smear-negative cases

The study, conducted in 147 smear-negative HIV-positive patients with suspected TB enrolled at GF Jooste Hospital in Cape Town, found that all case definitions apart from constitutional symptoms offered a high positive predictive value for eventual culture-positive diagnosis of TB.

Table 1: Smear-negative tuberculosis case definitions

Pulmonary infiltrate

Cough for >21 days


pulmonary opacification or nodular infiltrate on CXR



Pneumocystis j i roveci (previously carinii) pneumonia excluded

(using CDC clinical case definition*18)



no resolution after treatment with broad-spectrum antibiotic(except in the case of patients with diffuse micronodular [miliary] infiltrate on CXR, who are started on antituberculosis treatment after cultures are sent)


Peripheral nodes

Significant asymmetrical peripheral nodes (long axis >3 cm)



fever >38°C on two occasions



drenching sweats for >2 weeks

Mediastinal or abdominal nodes

Visceral nodes (mediastinal/hilar or abdominal nodes seen

on imaging)



fever >38°C on two occasions



drenching sweats for >2 weeks


Pleural exudate

Lymphocytic exudates

Pericardial effusion

Effusion on ultrasound



fever >38°C on two occasions



drenching sweats for >2 weeks (aspirate reserved for patients

with haemodynamic compromise)


Ascitic exudates

Lymphocytic exudates



fever >38°C on two occasions



drenching sweats for >2 weeks


Constitutional syndrome

Wasting (BMI of <18.5)



documented weight loss of >5% body weight within a month



fever >38°C on 2 occasions



drenching sweats for >2 weeks


Table 2: Diagnosis by case definition and final diagnosis by TB culture

Case definition

Subjects meeting case definition


Subjects presenting with this as only case definition

Confirmed TB

Possible TB

Not TB

Positive predictive value

Pulmonary infiltrate

83 (56%)

24 (29%)

64 (77%)

12 (14%)

7 (8%)


Peripheral nodes

33 (23%)

11 (33%)

30 (91%)

1 (3%)

2 (6%)


Mediastinal nodes

69 (47%)

7 (10%)

59 (86%)

7 (10%)

3 (4%)


Abdominal nodes

17 (12%)

6 (35%)

13 (76%)

3 (18%)

1 (6%)


Pleural exudate

20 (14%)

3 (15%)

14 (70%)

3 (15%)

3 (15%)


Pericardial effusion

10  (7%)

5 (50%)

4  (40%)

4 (40%)

1 (10%)


Ascitic exudates

6 (4%)

3 (50%)

6 (100%)





11 (8%)

11 (100%)

4 (36%)


7 (64%)



Seventy-nine subjects (54%) met more than one case definition

Global plan strengths

However, the global plan has some definite strengths. For instance, it is pro-poor and recognises the need to improve healthcare systems overall and not just in the TB arena. It also demonstrates a growing openness in the TB community for less vertical programmes, more engagement in patient and community empowerment

Its important to note that, instead of clearly defining directly observed therapy as being necessary for the new “high-quality DOTS expansion and enhancement,” the strategy calls for “standardised treatment with supervision and patient support.” This suggests a growing flexibility to consider alternate forms of community-based adherence support.

And yet, by its own admission, the global plan will still not meet its targets in two regions - in African countries with a high burden of HIV, and in Eastern Europe/Russia, where drug resistant TB is out of control and HIV is on the rise as well.

Practical steps in the meantime

In his talk in Denver, Professor Harries discussed several possible approaches that should be attempted in the field now, starting with

I. The need to better integrate HIV and TB control programmes.

“We need to establish mechanisms for collaboration between the HIV world and the TB world. We need to set up co-ordinating bodies, we need to conduct surveillance of HIV prevalence in TB patients, do joint planning and do joint monitoring and evaluation. It's very easy to write this down, it's much more difficult to do in practice. Programme directors have to get together - they have to agree on how it should work. You have to share resources, you have to build joint capacity.”

II. The TB burden in people living with HIV/AIDS must be lowered (which will in turn lead to less risk of transmission). This can be done by:

a)     Intensified TB case finding— looking for TB cases where HIV & TB come together, finding them, diagnosing them, treating them quickly and thereby preventing transmission of infection. “We can do this in prisons, hostels, barracks, in household contacts and at VCT sites. It works well in the research arena - it does not work so well in the routine health system arena,” said Prof. Harries.

b) TB prevention control. One obvious way of reducing TB burden would be to avoid exposure. “TB infection control measures centre around environmental, personal and administrative protection,” said Prof Harries. Despite the existence of very good guidelines “in the real world, these guidelines are not really implemented.”

He highlighted one important concern: “HIV-positive staff should not really work on medical wards in Africa — these are very dangerous places.”

This could also be a weakness in a more community-based approach to TB control. HATIP panel advisor Chris Green noted: “We do need to be much more concerned over the risks to PLHAs in peer support environments. If someone comes to a meeting/drop-in centre coughing, what should we recommend? Should we have a stock of face masks to give to such people? Should we deny entry, and send them immediately to the clinic? Should we not at least have effective publicity material (posters) displayed? Should we not develop policies. Similar concerns may arise in community VCT centres, where PLHAs may be acting as counsellors.”

c) Isoniazid preventative therapy (IPT) , when given for 6-12 months in HIV positive persons, reduces the risk of TB by 40%; and IPT for 12 months in HIV positive patients who have completed TB treatment, successfully, reduces recurrent TB by 50-80%. Unfortunately, as discussed in a recent HATIP, IPT is beyond the capacity of most national TB programmes to deliver (http://www.aidsmap.com/cms1001472.asp

However, Prof. Harries pointed out that ART programmes (and/or national HIV/AIDS programmes) themselves can provide the infrastructure to deliver IPT. And indeed, the new Stop TB plan endorses this approach.


HATIP went to its panel of advisors to see how comfortable they are now with IPT as part of the basic package of care that HIV programmes or any HIV-treating clinician offer people with HIV.

Most agreed with Mark Harrington of the Treatment Action Group (TAG), who said “IPT should be part of the standard package of care for PLWHA. Six to nine months of IPT should be given to all HIV-positive people with latent TB infection as diagnosed 1) by tuberculin skin test (TST) or 2) without active disease (use of a symptom questionnaire should be sufficient).”

Dr Halima Dawood of King Edwards Hospital in Durban, South Africa added that “if the patient is chosen well, INH prophylaxis may be used as ‘training’ for ART and also provide the benefit prophylaxis.”

However, Dr Vijay Anthony Prabhu of International Training & Education Center on HIV in Chennai, India worries that “it is difficult to exclude active TB in patients who are severely immunocompromised, because of paucity of sputum positivity and clinical signs. Furthermore, while immune reconstitution with its attendant signs and symptoms are not specific for any one particular disease, it is extremely difficult in that instance to say that a patient has broken through prophylaxis and is actually experiencing active TB, while in fact, he or she may actually be suffering from a immune reconstitution.”

Dr Gerald van Osch of St. Maarten agreed that “the ‘general’ clinician on our island has very little knowledge of algorithms to check for active TB, so additional training might be necessary.”

Chris Green of the Spiritia Foundation in Jakarta, Indonesia added: “I think what we need is much better information, both for doctors and for 'patients', so that we can both achieve a consensus/speak with one tongue and also understand the options.” Green also wondered whether prophylaxis would have to be continuous and whether that is practical.

Although she prescribes nine months of IPT, Dr Dawood notes that “there is a problem with adherence,” though the well-supported patient can adhere to treatment. “Better adherence/treatment support would be essential,” added Harrington “and community education and TB literacy would also be helpful.”

And yet should a patient breakthrough or be poorly adherent, Harrington pointed out, “there is no data suggesting that IPT compromises later response to four drug treatment, or that it should not be used in areas where INH resistance is found.”

However, Dr Harries pointed out one serious problem with IPT in patients on ART — the potential for an interaction with d4T and an increase in peripheral neuropathies.

III. Decreasing the burden of HIV in TB patients by more aggressive HIV testing, provision of cotrimoxizole and ARV therapy.

Prof. Harries noted that another study presented at the Retrovirus Conference reported that long term ART could reduce the risk of TB to just 1% by year 5 — although it is still higher than the risk in HIV negative individuals (http://www.aidsmap.com/en/news/9E0101FB-11D2-47C3-A384-4F31995715D2.asp).

Nevertheless, ART represents a critical tool to interrupt the cycle of TB transmission — and if TB is to be controlled, ART needs to be widely available in any country with a high burden of coinfection.

New TB resources

On March 24, World TB day, two new excellent service standard guidelines became available:

The Patients’ Charter for Tuberculosis Care (The Charter), which outlines the rights and responsibilities of people with tuberculosis and sets out the ways in which patients, the community, health providers (both private and public), and governments can work as partners in a positive and open relationship with a view to improving tuberculosis care and enhancing the effectiveness of the healthcare process.

The charter was developed in tandem with the International Standards for Tuberculosis Care, which describes a widely accepted level of care that all practitioners, public and private, should seek to achieve in managing patients who have, or are suspected of having, tuberculosis.

See http://www.iuatld.org/full_picture/en/frameset/frameset.phtml

Last week, the World Health Organization launched a new strategy to fight tuberculosis called the "Stop TB Strategy" which underpins the Global Plan to Stop TB, 2006–2015. See its key points here: http://www.aidsmap.com/en/news/B4034F12-36F8-4BCC-AC67-B7FFF2F16F94.asp

Stop TB strategy paper http://www.who.int/tb/publications/2006/stoptb_strategy_en.pdf


Harries, A. HIV and tuberculosis. Thirteenth Conference on Retroviruses and Opportunistic Infections, Denver, 2006.

Lawn SD. Impact of HIV infection on the epidemiology of tuberculosis in a peri-urban community in South Africa: the need for age-specific interventions. Clinical Infectious Diseases 42: 1040–1047, 2006.

Whalen C. Failure of directly observed treatment for tuberculosis in Africa: a call for new approaches. Clinical Infectious Diseases 42:1048–1050, 2006.

Perkins, MD and Small, PM. Admitting defeat, The International Journal of Tuberculosis and Lung Disease, Volume 10, Number 1, pp. 1-1(1), 2006.

Wilson D et al. Diagnosing smear-negative tubeculosis using case-definitions and treatment response in HIV-infected adults. Int J Tuberc Lung Dis 10(1): 31-38, 2006.