Back to contents

HATIP #99, 17th January 2008

Published: 12 December 2007

Treating MDR TB in the community: what are the challenges?

By Keith Alcorn

National TB programmes should look at how treatment for multi-drug resistant TB can be carried out as far as possible in the community, according to doctors and treatment advocates from southern Africa meeting last November at the 38th World Conference on Lung Health in Cape Town, South Africa.

But moving MDR TB treatment into the community would represent a radical shift in approach, and some doctors and advocates are uneasy about the idea, fearing an increase in stigma and a rise in MDR TB due to poor infection control.

This edition of HATIP looks at the arguments for community-based treatment, and some of the obstacles that must be overcome if it is to be safe and effective.

Why is MDR TB managed differently?

MDR TB treatment is viewed differently from standard TB treatment because once an individual acquires MDR TB that is resistant to rifampicin and isoniazid, their treatment options become limited to a lengthy and toxic course of drugs. Although MDR TB is no more infectious than drug-sensitive TB, it has a far worse prognosis, especially in people infected with HIV.

And, if MDR TB treatment fails, there is a danger that individuals will have developed an even more drug-resistant strain, one that might even be extensively drug-resistant and virtually impossible to treat. That was how the epidemic of extensively drug-resistant TB currently affecting South Africa evolved, and similar potential exists in many other parts of the world.

MDR TB: key facts

Multi-drug resistant TB is tuberculosis that is resistant to two first-line drugs, rifampicin and isoniazid.

MDR TB is diagnosed after culturing TB bacilli and then testing for drug susceptibility. This process may take several months and is not available in many settings.

It is acquired either through transmission from another infected individual, or because of failure to take the full course of first-line TB treatment due to missed doses, drug stock outs, poor quality drugs or incorrect prescribing.

Some recent studies have shown that the majority of MDR and XDR TB cases are the result of transmission from an infected individual.

Extensively drug-resistant TB is tuberculosis that is resistant to rifampicin and isoniazid and also resistant to any fluoroquinolone and at least one of three injectable second-line drugs (i.e., amikacin, kanamycin, or capreomycin).

MDR-TB is most common in India, China, Russia, Eastern Europe and Central Asia, but an increasing number of cases are being detected in sub-Saharan Africa.

A major outbreak of XDR TB has been detected in South Africa; 481 cases had been identified by October 2007.

The prognosis of MDR and XDR TB in people with HIV is much worse than drug-sensitive TB.

The emergence and spread of MDR and XDR TB is judged to be particularly problematic in populations where there is high HIV prevalence. People living with HIV are more likely to develop TB, more likely to cluster together in health care facilities and more likely to experience rapid progression to disease once they become infected with a drug-resistant variant of TB, and much more likely to die from it before the diagnosis can be confirmed.

The example of Tugela Ferry in South Africa demonstrates just how dangerous MDR TB can become in a setting of high HIV prevalence. MDR TB had been present in the province of KwaZulu-Natal since the mid-1990s, but the transformation of a strain resistant to isoniazid and rifampcin into one that was extensively drug resistant led to an outbreak of XDR TB among HIV-positive patients at the Church of Scotland Hospital in Tugela Ferry. Patients who acquired XDR TB through nosocomial transmission often died before they could be diagnosed with drug-resistant TB. Fifty of the 51 patients first identified with XDR TB died within a median of 16 days of sputum collection at the hospital.

In South Africa at least, current policy in some provinces – but not all - is to hospitalise all patients diagnosed with MDR TB until intensive treatment is completed and the patient has a negative smear and TB culture – a period of at least six months.

This policy is intended to limit the number of contacts that an infected person has prior to conversion to smear-negative TB, and to ensure that the treatment regimen is adhered to.

However, lengthy hospitalisation is neither practical nor humane, in the view of doctors speaking at a satellite meeting during November’s World Lung Health meeting organised by the Treatment Action Campaign, Médecins sans Frontières, Partners in Health, AIDS Rights Alliance of southern Africa and the Open Society Institute.

Indeed, enforced hospitalisation is so resented by patients that they recently mounted a demonstration at Sizwe isolation hospital near Johannesburg. One patient was shot when police panicked.

"These were not activists, these were people demanding to be treated like human beings," said Dr Eric Goemaere, MSF head of mission in South Africa.

And just before Christmas 2007, 49 patients with MDR and XDR TB at an isolation unit in the Eastern Cape began to slip away after having cut through barbed wire fences surrounding the facility in Port Elizabeth, anxious not to be separated from their families during the holiday period.

In any case, says Dr Francois Venter of the Reproductive Health and HIV Research group at the University of the Witwatersrand, Johannesburg, this policy is not enforced in practice everywhere, particularly in the private sector in South Africa.

The policy is also unscientific, he pointed out in an article written with Mark Heywood of the AIDS Law Project in November 2007. “Generally, patients with TB are ill and infectious for many months before they are sick enough to be diagnosed. Once they start treatment, it usually takes months to diagnose MDR TB…A huge proportion of people with MDR TB die long before being identified, meaning those who actually end up on MDR TB treatment are only the tip of the iceberg.”

“Isolating someone for a brief period when they have been infectious for a long time is locking the door long after the horse has bolted.”

But hospitalisation may be necessary in some circumstances, for example if the patient is so sick they cannot walk, if infection control in the community is very difficult or if masks are unavailable. MDR TB treatment regimens also contain daily injectable drugs and administration may be difficult to arrange on an out-patient basis.

Referral to a specialist facility is also advisable, notes Dr Riitta Dlodlo, HIV programme director of the International Union Against TB and Lung Disease, because MDR TB drug regimens are complex and the treatment of patients should be supervised by centres of excellence – or at least centres with some experience in its management – and in centres with the necessary infrastructure, in order to ensure the best outcomes for patients.

But hospital facilities for MDR TB in South Africa are inadequate. Patients are referred to designated hospitals, often hundreds of miles from home, and must leave their families and livelihoods for six to eight months.

Yet the burden of MDR TB cases is now so high, these hospitals cannot cope. Khayelitsha township in the Western Cape province identified 109 cases in 2006, for example.

“The reality is that there are not enough beds in MDR facilities in South Africa to admit all those currently diagnosed with MDR TB, let alone the increased numbers that we will diagnose if we improve the coverage of drug sensitivity testing,” Dr Graeme Meintjes of GF Jooste Hospital in Western Cape province told HATIP.

“The current policy of initial inpatient treatment for all MDR TB cases results in patients with MDR waiting for a bed in an MDR facility while being treated in a general inpatient or outpatient facility. In most of these facilities infection control is currently poor and fellow patients, including those with HIV infection, are thus potentially exposed to MDR TB”.

Dr Jim Kim of Partners in Health told a press conference during the World Lung Health meeting: “We believe that this treatment does not require hospitals. In fact, you want to stay away from hospitals as much as possible and provide treatment on a community-based level [in order to reduce transmission within clinical facilities]”

In practice, some hospitals are having to treat MDR TB in the community. Dr Tony Moll of the Church of Scotland Hospital, Tugela Ferry, said that his clinic was managing MDR patients at local primary health clinics or at home unless they are very ill. A mobile injection team is visiting patients; as well as administering injectable TB drugs, the teams are also carrying out training in TB treatment literacy and infection control, side-effect monitoring and contact screening.

“Instead of…more hospital beds, money should be directed towards building MDR-TB community programmes inspired by existing antiretroviral ones…This has ceased to be a medical specialist issue; it has become a large-scale community problem,” Francois Venter and Mark Heywood wrote in the November 11 2007 edition of South Africa’s Sunday Times.

Experience of community management in Lesotho

The hospitals in sub-Saharan Africa that have gone furthest in treating recently diagnosed MDR TB in the community are those in Lesotho, where MDR TB in persons who have returned home from having worked in South African mines is a growing problem, according to Dr Pheello Lethola of the Scott Hospital, a 102-bed rural facility serving around 220,000 people in a mountainous area of Lesotho.

Hospital-based management of MDR TB is not the way to go, she told a session during the World Lung Health conference, because of airborne transmission within the facility, and it’s not realistic for rural areas of Lesotho, where people may have to travel many hours on horseback to reach a clinic.

Partners in Health has conducted a pilot study of community-based MDR treatment in Lesotho, reported Dr Hind Satti. She has now treated 40 patients with MDR TB, almost two-thirds of them HIV-positive, with an average of three to four TB regimen failures by the time they are diagnosed with multi-drug resistance.

The programme uses community treatment supporters to do DOTS home visits. The supporters are paid, and are required to bring the patient to the clinic once a month.

All patients diagnosed with MDR TB, their household contacts and any suspected cases are routinely offered an HIV test, with a view to starting antiretroviral therapy within a few weeks of starting TB treatment, regardless of CD4 cell count.

However the longest standing experience of MDR treatment in the community comes from Peru, where Socios En Salud, PIH's partner organisation in Peru, has been treating MDR TB in the slums of Lima for over ten years. SES trained and hired people from the community to accompany patients through the long and arduous course of treatment with second-line drugs, conducting directly-observed treatment.  In 2003 Carole Mitnick and Paul Farmer of Partners in Health, together with Dr Jaime Bayona and colleagues from Socios en Salud published results from their first 66 MDR TB patients in the New England Journal of Medicine. Despite being resistant to a median of six drugs, 83% had turned culture and smear-negative after four months and were therefore probably cured, five died and only one patient still had smear-positive MDR TB after six months.

By any national TB programme’s standards, these results are good. In MDR TB patients they are outstanding.

However, it is important to note that there were some clinical factors that may also have contributed to the good results. Therapy was more aggressive than in other studies of MDR TB treatment, using more drugs at higher doses for longer periods. Treatment was also individualised after drug susceptibility testing, allowing for the best possible response. Patients had fewer coexisting infections and were younger than patients in other cohorts, and their HIV coinfection status was not recorded (but is likely to be very low, given the low HIV prevalence in Peru).

Lack of access to MDR TB treatment

MDR TB treatment remains rare in many African countries despite a high presumed incidence, partly because laboratory facilities for drug susceptibility testing (DST) are lacking. Without DST the World Health Organization’s Green Light Committee is reluctant to grant access to the subsidised high quality second-line TB drugs that it guards, to ensure that use of the drugs in patients with reduced susceptibility does not result in increased resistance.

But the Green Light Committee is also reluctant to grant access to second-line TB drugs if there is a lack of political and administrative support for MDR TB treatment, and where there is a lack of capacity to ensure that patients take the drugs correctly.

The Global Fund to Fight AIDS, Tuberculosis and Malaria has told TB programmes in Africa to ask for more money in order to scale up laboratory capacity and institute MDR TB treatment programmes. Technical assistance is available from the Green Light committee to Global Fund recipients who have already received funds to provide an MDR TB treatment programme.

Drug susceptibility testing is likely to become more widely available outside South Africa as WHO’s Global Laboratory Initiative achieves its aims of upgrading laboratory facilities in order to improve TB diagnosis.

In most settings, laboratories first need to renovate and upgrade their infrastructure, streamline work-flow, hire and train more technicians and establish quality control for microscopy and solid culture and DST before attempting liquid culture. Additionally, appropriate bio-safety measures must be assured, and systems established to guarantee quick transportation of samples from the peripheral labs to the culture laboratory with rapid communication of results.

Another development that could increase the capacity to perform culture and diagnose drug resistant TB is the rollout of Mycobacteria Growth Indicator Tube (MGIT) automated liquid culture system systems in several countries. According to Dr. Richard O’Brien, FIND has negotiated a new cost structure with the test’s manufacturer, Becton Dickinson (BD) for low-income countries “that reduces the test cost to under US $3 a test — very comparable to solid culture costs,” he said.

Also, PEPFAR has announced a commitment to fund the rollout of MGIT systems in Côte d’Ivoire, Ethiopia, Kenya, Tanzania, Uganda, Malawi, Mozambique, and South Africa.

But some caution might be advisable before “air-lifting” such complex machinery — which will require ongoing maintenance — into lower-resourced settings, according to Dr. Ruth McNerney of the London School of Hygiene and Tropical Medicine. 

“Roll-out is being pushed for culture but we haven’t yet seen any data on the impact of MGIT and the liquid culture systems on patient care,” she said. “We’re pushing enormous amounts of resources into this in these areas, yet no one even knows if it’s going to make an impact. Two weeks is still quite a long time to get your results. Is that going to make a difference to your patients or will you already have them on treatment? If they’re not on treatment, will you ever see them again?”

But others think that the only way that targets for increased MDR TB treatment - 1.5 million treated by 2015 - can be met is by relaxing the exacting standards needed for MDR TB drug access.

“There’s no way that this target [for MDR treatment] is going to be met without risks being taken, such as presumptive diagnosis,” said Carole Mitnick of Partners in Health.

Infection control fears

The calls to move care of MDR TB into the community have been met with concern due to fears about the risks of MDR TB spreading further, particularly among people with HIV.

“We met with treatment activists from TAC and they said take these MDR patients away. They were scared,” said Dr Eric Goemaere of MSF, describing reactions when he first began talking about treating MDR TB in the township setting of Khayelitsha.

In Lesotho a similar reaction came from health care staff.

“When we first started treating patients with MDR-TB, we had to admit them into private wards,” said Dr. Lethola. “The nurses and all the workers who eventually learned that these were MDR suspects or cases did not want to have anything to do with them. No one wanted to go into the room including the people who bring the food or the cleaners,” she said.

So they performed trainings for health staff and patients at the hospital and community clinics on MDR-TB, infection control, and how to care for these patients.

“Once we had done this training, the stigma came down. I cannot say that there is no stigma, but people have become more accepting and more willing to work with these patients,” said Dr Lethola.

The reaction of TAC activists, said Dr Goemaere, “tells you how much work we have to do to build what I would call MDR-TB friendly communities - communities that through treatment literacy understand the disease, understand how it’s transmitted, understand how people are infectious, communities that would slowly accept that MDR TB patients are living among them. They’ve done that successfully in Peru and I don’t see why we wouldn’t be able to do it here because that is the only way, as we have learned from the ARV programme - it’s only by decentralising, it’s only by task shifting.”

“People are not cowering outside [Tugela Ferry hospital] and afraid to walk in," said Dr Gerry Friedland of Yale University School of Medicine, who has been closely involved with responding to the XDR TB outbreak. “The staff is working; people are being taking care of; there have been health care workers who got infected and died but there’s a very active attempt to provide infection control strategies - which we think may be working already.  And people are working in that environment, putting themselves at risk in the path of the worst MDR/XDR and doing their job.”

“And most of us in this room are healthcare workers, and we confronted situations of danger in our work and this is one of them. And I think that with appropriate education and understanding and maybe less sensationalism about it, there is no reason in the world why anyone in this room should not be working with people with HIV disease and MDR-TB.”

Infection control measures that can protect health care workers and other patients from known cases of MDR TB are only one side of the coin. Health facilities also need to protect against the cases of MDR TB that have yet to be diagnosed, by better infection control throughout the facility.

In Lesotho for example, Dr Pheello Lethola said that outdoor waiting areas are being promoted, together with separation of coughing and non-coughing patients. Patients are also being taught basic coughing etiquette – cover your mouth when coughing, don’t spit, turn away if someone is coughing near you.

Masks for health care staff are also a necessity, but often in short supply.

“Surgical paper masks are of no benefit to healthcare workers unless they put them on infectious patients who are coughing and unable to cover their mouths while doing so,” Dr Alasdair Reid of UNAIDS told HATIP. “Only special N95 respirator masks which are tight fitting and have special filters can protect staff, but they are hot and uncomfortable, often poorly fitted and expensive so they are not always available.”

Yet the risk of TB acquisition is highest when an individual spends virtually all their time in the presence of a person with active TB who is still sputum-positive. The most likely site for such exposure is in the home, or in the nearby beds in a hospital ward that is poorly ventilated.

This means separation of coughing and non-coughing in-patients, as Dr Lethola suggests, but it can also extend to building a separate hut for the MDR TB patient so that she can be treated at home, without infecting her family. In Peru, said Dr Salmaan Keshavjee, of Partners in Health, the cost of building an extra room onto a person’s home for the duration of their MDR TB treatment was less than $50.

“I think this issue of non-ventilation is a hopelessly under-rated thing,” said Dr Francois Venter. “Air flow through a room has been shown to be the most important issue in infection control – but in most hospitals, people refuse to open windows due to temperature discomfort; and we spend so much money on masks and UV light when simply improving the number of blankets and insisting staff wear more clothes would go a long way to making this option more palatable!”

But, points out Dr Alasdair Reid of UNAIDS, “the risk of infection goes down steadily the longer someone is on appropriate treatment. So the best way to achieve good infection control is to make the diagnosis earlier and get someone onto the best treatment as soon as possible rather than focus on isolating people with drug-resistant TB once they are on treatment”.

Supporting community members with symptoms of TB to access rapid diagnosis and treatment through intensified case-finding may have a significant impact on infection control in the community, he points out.

Patient education and side-effects

Successful community-based treatment will require more than simply permitting patients to receive treatment at home.

“No one explained to me how painful streptomycin was going to be for two months,” said Johanna Ncala, the Treatment Action Campaign’s national treatment literacy coordinator in South Africa.

Patient education about how TB is transmitted, why MDR TB is particularly dangerous, and the need for the full course of treatment to be completed will be necessary, together with preparation for the unpleasant side-effects of second-line treatment, advice on nutrition and guidance on infection control.

“From a patient point of view, adherence to a long and painful treatment being key for success, we believe that community-based adherence support mechanisms will enhance adherence in the same way it did for ARVs,” Dr Eric Goemaere of MSF South Africa told HATIP.

Community-based treatment supporters like those recruited for ARV patients may be an approach worth following, but as Dr Goemaere’s experience with TAC shows, there are still legitimate concerns about the risks to HIV-positive people that need to be addressed through community education before the approach can move forward.

 “I don’t hear enough about the human resource capacity, development and training that’s going to be necessary for these so-called treatment supporters,” said Mark Harrington, Executive Director of the Treatment Action Group (TAG). “I mean I’m HIV-positive, I’ve been doing HIV work for over 20 years and I will be frightened to be a treatment supporter for an MDR-TB patient. I would first want to know about how you do it.”

“What is the real risk for a lay person with HIV in South Africa? said Javid Syed of TAG. “We don't only need education and understanding but also aggressive efforts for infection control and intensified case finding, detection, and appropriate treatment.”

Dr Riitta Dlodlo, who is based in Bulawayo, Zimbabwe, takes a cautious view. Community treatment supporters “should ideally be persons who know their HIV status, which should be negative,” she told HATIP.

Instructing nurses how to give the injectable TB drugs will also be needed, as will training in how to recognise one of the most serious side-effects of second-line TB treatment, hypokalaemia.

Hypokalaemia, a very low level of potassium, is chiefly caused by the aminoglycoside antibiotics and capreomycin, and occurred in one-third of patients receiving MDR TB treatment in Peru. The condition occurred in two-thirds of those receiving capreomycin.

Early symptoms may be cramps, fatigue, nausea and irritability, but if not detected early hypokalaemia can result in seizures, tetany (severe cramping of the muscles in the hands and feet) and disturbances in the heart beat (cardiac arrhythmia) that can result in death.

The early signs can be subtle, and hypokalaemia may only be detected through measuring electrolyte levels (potassium below 3.5mEq/L is too low). In the Peruvian study low body weight was associated with hypokalaemia, and monitoring more than once-monthly may be advisable in these patients.

Hypokalaemia can be corrected by giving potassium chloride and magnesium sulphate. Once hypokalaemia was detected and treatment began, it took an average of six months for potassium levels to return to normal in the Peruvian study (the largest review to date in MDR TB patients).

A useful training module produced by WHO Europe on the side-effects of all TB drugs and their management can be downloaded here in pdf form.


“If we go in the direction [of community-based treatment] everyone is going to have to take risks – the doctors, the managers, the treatment supporters,” said Dr Eric Goemaere of MSF.

However the head of South Africa’s national TB programme, Dr Lindiwe Mvusi, was uncertain about this approach when she spoke at the World Lung Health conference in Cape Town last November.

“We’re being put under pressure to consider community care for MDR TB due to the shortage of hospital beds, but I’m not sure that this is the right stage to consider this, given that we don’t have an effective community care system for TB.”

“One approach to this problem would be to scale up the number of MDR beds in the country, but we know that such projects take years to budget and get done through Public Works,” comments Dr Graeme Meintjes. “We would always be playing catch-up with this approach and falling far short.”

“The more pragmatic approach is to say let’s develop separate areas in outpatient facilities with good ventilation where MDR patients can attend for their care, apart from other patients, and implement community-based interventions to address infection control in patients’ households and optimise adherence along the lines of the Peru model. Obviously those MDR patients who are so sick that they need inpatient care will still require admission to an MDR facility.”

But, says Mark Harrington, Executive Director of the Treatment Action Group, “the amazing progress in Lesotho... was because PIH, and Soros and all these donors, were there helping to set it up. But there aren’t cadres of Partners in Health who are able to go all around the world and treat 1.6 million cases of MDR-TB [by 2015].”

So additional partners must be identified and resources leveraged for other countries.

“One thing we hear a lot is from people that work on the Global Fund is that TB programmes aren’t asking for enough money,” said Harrington. “So one of the key messages from this meeting should be that TB programmes should ask for enough money to do the MDR scale up that they need, to build the labs and to ensure the purchases and supply chain management of the drugs and the training of the people that will be providing the services.”

“We need to be ambitious,” said Dr. Alasdair Reid of UNAIDS. “The TB community has always done what it can with the resources it has. We need to state clearly what resources we need, and that without them, we will fail.”

But the last word must go to Dr Riitta Dlodlo.

“Drug resistance is a man-made problem. We develop it - we health providers - by not adhering to the recommended standardised anti-TB treatment regimens and correct drug doses, and by not supporting our TB patients to adhere to this treatment. We could prevent more drug resistance by reviewing our practice in order to improve treatment success and cure rates, and by following up treatment interrupters and bringing them back into treatment as quickly as possible.”


Shin S et al. Hypokalemia among patients receiving treatment for multidrug-resistant tuberculosisChest 125:974-980, 2004.

Mitnick C et al. Community-based therapy for multidrug-resistant tuberculosis in Lima, Peru. N Engl J Med 348: 119-128, 2003.


  • Dr Alasdair Reid, HIV/TB Adviser, UNAIDS
  • Dr Riitta Dlodlo, Director: Health Systems Services Department - Bulawayo City Council, Zimbabwe, HIV Programme Coordinator, International Union Against Tuberculosis and Lung Disease
  • Dr Graeme Meintjes, GF Jooste Hospital, Cape Town, South Africa.
  • Dr Francois Venter, Reproductive Health and HIV Research Unit
    University of the Witwatersrand, Johannesburg, South Africa
  • Dr Eric Goemaere, MSF South Africa
  • Dr Haileyesus Getahun, WHO Stop TB Department

HATIP #99, 17th January 2008

This content was checked for accuracy at the time it was written. It may have been superseded by more recent developments. NAM recommends checking whether this is the most current information when making decisions that may affect your health.
Community Consensus Statement on Access to HIV Treatment and its Use for Prevention

Together, we can make it happen

We can end HIV soon if people have equal access to HIV drugs as treatment and as PrEP, and have free choice over whether to take them.

Launched today, the Community Consensus Statement is a basic set of principles aimed at making sure that happens.

The Community Consensus Statement is a joint initiative of AVAC, EATG, MSMGF, GNP+, HIV i-Base, the International HIV/AIDS Alliance, ITPC and NAM/aidsmap

This content was checked for accuracy at the time it was written. It may have been superseded by more recent developments. NAM recommends checking whether this is the most current information when making decisions that may affect your health.

NAM’s information is intended to support, rather than replace, consultation with a healthcare professional. Talk to your doctor or another member of your healthcare team for advice tailored to your situation.