Bulletin of the World Health Organization

Case study: South Africa

Karin Weyera

For the past decade, a concerted effort to reform TB control in South Africa has resulted in changes in case-finding and treatment policies, standardization of recording and reporting systems, and monitoring of the performance of control programmes using pre-defined indicators; these changes were all made in line with the internationally recommended DOTS strategy. The essential elements of the revised strategy,1 implemented in 1996 after TB was declared a national emergency, include bacteriological confirmation of disease, standardized first-line treatment regimens that are exclusively based on fixed-dose combination formulations and an electronic recording and reporting system. Expansion of the DOTS strategy followed rapidly: in 2003 there was complete coverage in all nine provinces, covering 183 health districts. Comprehensive programmatic management of patients with MDR-TB became national policy in 2000 and was implemented through a network of dedicated provincial MDR-TB referral centres.

Despite these efforts, however, TB incidence and case–fatality rates have increased threefold in South Africa over the ensuing decade.2 More than 400 000 cases of TB require treatment annually, but cure rates barely reach 50%,2 reflecting the classic mistake made in TB control of identifying cases but not treating them adequately. TB mortality is at an all-time high. There are some 10 000 incident cases of MDR-TB per year,3 representing the largest MDR-TB burden in Africa and further pointing towards a failure of TB control. Although a favourable outcome (cure and treatment completed) is achieved in more than 80% of MDR-TB patients who complete the full course of standardized treatment, deaths (up to 20% of patients who started treatment), defaulting from treatment (up to 25% of patients) and failure of treatment (around 10%) reduce the overall effectiveness of the programme to less than 50% (South African Medical Research Council, unpublished data, 2002–2004). Worryingly, patients with XDR-TB have been identified in each of the nine provinces over the past 18 months.

Determinants of the worsening TB epidemic in South Africa are diverse and multifactorial. Historically, there has been a legacy of neglect, poor management of patients and fragmented health services.4 Contemporary barriers to effective TB control in South Africa are similar to those elsewhere in Africa, and include an exploding HIV epidemic, deteriorating socioeconomic conditions among already vulnerable populations and constraints on human resources in the health-service sector. Although TB control has been fully integrated into primary health-care services and decentralized to district level, delivery is hampered by competing health priorities, slow district reform and deficient management capacity, especially at the level of implementation. Unemployment rates of up to 40%, as well as the resultant migration and massive growth in informal urban settlements, lead to failures in supervision of treatment and follow-up. Reasons for defaulting from TB and MDR-TB treatment include patients’ perceptions of negative attitudes among health-care workers, substance abuse and employment concerns.5,6

However, it is the lost opportunity for early, effective HIV intervention in South Africa that has brought the weaknesses in TB control into sharp focus. At least 60% of TB patients are estimated to be coinfected with HIV;2 this is most strikingly reflected in the excess and rising mortality. Up to half of patients categorized as treatment defaulters in the aforementioned research studies were subsequently found to have died, and the reason for death was often reported as being HIV-related.5,6 HIV-associated transmission of XDR-TB and the exceptionally high risk of mortality in HIV-positive people coinfected with XDR-TB7 amplify public health concerns over the threat of a virtually untreatable TB epidemic occurring within the context of HIV coinfection.

The view expressed in the base paper that drug-resistant TB poses a major threat to achieving global targets for TB control also holds true for South Africa. In addition, however, the 2005–2006 XDR-TB outbreak in KwaZulu-Natal7 serves as a serious warning that gains made in HIV care and treatment might be lost if drug-resistant TB is not effectively and rapidly addressed. Several epidemiological and genetic studies have confirmed both nosocomial and community transmission of drug-resistant TB in South Africa. Increased access to HIV treatment and care will inadvertently bring together highly vulnerable individuals with infectious cases of MDR-TB and XDR-TB, often in settings where large numbers of people congregate. The lack of adequate and appropriate infection-control measures in most public health settings, juxtaposed with an extremely high prevalence of HIV (both in patients and health-care workers), represent a public health emergency requiring much earlier detection of drug resistance, segregation of infectious patients, urgent improvements in infection control measures and a rapid, appropriate response to outbreaks.

Dire predictions of the impact of HIV on TB and MDR-TB in South Africa were made in 1999.4 Sadly, what had been mere assumptions at the time now seem to have come true. Substandard care, fertile conditions for transmission and the rapidly progressing HIV epidemic all impede the ability of South Africa to reach the required targets for TB control; they also contribute to establishing the endemicity and spread of drug-resistant TB. A dynamic and exceptionally strong collaboration between HIV and TB control programmes will be required to avert large-scale HIV-associated epidemics of drug-resistant TB. Failure to engage in such collaborations is bound to have devastating consequences. ■



  • TB Epidemiology and Intervention Research Unit, South African Medical Research Council, Private Bag X385, Pretoria, 0001 South Africa.