Areas of focus
Several points along the continuum of comprehensive care present opportunities for applying the lessons learned between HIV and TB services (Table
1):
With respect to
HIV counselling and testing, TB services present a readily identifiable cohort for HIV provider-initiated testing (PIT). Service integration in this cohort will undoubtedly increase HIV screening for this high risk cohort. Studies in other African countries have shown that patients are willing to attend integrated services and that, if provided, uptake of voluntary counselling and testing(VCT) or PIT is high[
15,
22]. Given the 15% annual risk of developing TB in HIV infected individuals[
4], a more intimate connection between TB and HIV services would aid early diagnosis and management of TB. PIT in TB clinics can focus on the clinical requirement to know a patient's status in order to provide appropriate medical care[
23]. In addition to PIT, point of care CD4 count testing can be conducted to minimise the delay between HIV testing and enrolment into HIV care. The proportion of notified TB patients who have a known HIV status at the end of their outpatient TB treatment at our urban TB clinic using conventional VCT was 21.7% in 2007, having increased from 8.8% in 2004 (p < 0.0001) (Table
2). Amongst patients admitted for treatment of multi-drug resistant(MDR) TB, 51.5% of patients had a known HIV status in 2003, having decreased from 65.7% in 2000 (p < 0.002) (Table
3). It is likely that these patients would have had multiple encounters with health care providers before and during admission, yet the proportion of patients with a known HIV status remains unacceptably low. With the expansion of outpatient VCT facilities in many hospitals and an emphasis on privacy and confidentiality, the wards may increasingly be viewed as unsuitable for VCT. In addition, dually infected inpatients may be perceived to be too ill for VCT or unprepared to deal with a second diagnosis. Given the ambivalence regarding the timing of HAART initiation in TB patients, clinicians may not see the urgency in establishing an HIV diagnosis in an ill TB patient. While randomized studies examining the timing of HAART initiation in TB patients are still underway, preliminary data from our centre suggests a reduction in mortality in excess of 50% if TB treatment and HAART are integrated[
24]. The downward trend in the proportion of inpatients that have a known HIV status is concerning, and illustrates a sizable missed opportunity in this setting. A patient-centred approach to VCT in HIV services has evolved successfully. The perceived lack of respect for confidentiality at TB facilities may have rendered it unattractive and unsafe as site for HIV counselling and testing[
22]. Combating the stigma associated with HIV, whilst simultaneously preserving respect for confidentiality by ensuring privacy in the setting and communication is essential. To enhance the uptake of counselling and testing, patients in the various waiting areas in our TB clinic receive simple key messages from health care workers on TB and HIV. Initially this intervention was directed at large groups of patients but has evolved into smaller group discussions of 4–6 individuals to improve exchange of information between patients and health care workers. Efficient use of patients' time in the TB waiting area is maximized at our centre by delivering health care messages to a captive audience without interfering with the procedures conducted and flow of patients at the clinic. A similar commitment to patient education has not, however, diffused into the HIV care setting. Rarely, if ever, are formal patient education programmes focused on TB presented to this high risk group. Such programmes may assist the health system in early identification of TB cases by promoting self-reporting by patients.
Table 2
HIV status of patients on TB treatment at a TB referral facility in Durban, South Africa
2004 | 8716 | 770 (8.8%) |
2005 | 8567 | 1145 (13.4%) |
2006 | 8180 | 1392 (17.0%) |
2007 | 7202 | 1564 (21.7%) |
Table 3
HIV status of patients on treatment for Multi-Drug Resistant Tuberculosis(MDR-TB) at an MDR referral facility in Durban, South Africa
2000 | 204 | 134 (65.7%) |
2001 | 266 | 159 (59.8%) |
2002 | 384 | 228 (59.4%) |
2003 | 435 | 224 (51.5%) |
The WHO and UNAIDS have identified TB facilities in settings of high HIV prevalence as important sites for additional HIV surveillance. Integrated TB-HIV services will contribute to our knowledge of both epidemics, and facilitate the estimation of the burden of coinfection. As coinfected patients are symptomatic from HIV infection, surveillance at integrated clinics will provide a good estimate of the burden of HIV disease and mortality. This is substantially different from HIV surveillance at antenatal clinics where patients are generally asymptomatic from HIV infection, and only an estimation of population level infection is obtained. Estimates of HIV disease burden in coinfected patients will provide crucial direction to resource allocation and future health care planning. The active monitoring of patients in HIV clinics for the development of TB is an essential element of closer linkages between TB and HIV centres. Routine symptom check lists and clinical cues may be employed at HIV clinics to increase vigilance and identification of TB cases. This will serve as a method of TB case finding in high risk group, as well as give expression to the intimate relationship between TB and HIV to clinicians and patients.
Adherence support presents opportunities for enhancing patient support and improving resource utilization. Integrating HAART and the TB DOTS strategy in a manner suitable to the local conditions will initially require investment in health worker training and service reorganization. However, long-term benefits will include efficient usage of health workers' time and a more easily navigable experience for patients, ultimately ensuring increased TB treatment completion rates and MDR-TB prevention. DOT for TB originated almost half a century ago as a method of ensuring adherence to TB drugs and completion of treatment, and is the most successful and well studied of adherence interventions[
25]. DOTS programmes for TB could provide the necessary infrastructure through which DOT for HAART could be conducted. As direct observation evolves into a more supportive, empowering experience for patients, adherence to both TB drugs and HAART will be bolstered. This Modified Directly Observed Therapy(in which not all medication doses are administered under direct observation) may also present an opportunity for more closely monitoring adherence to HAART. A pilot study at our TB clinic revealed that a modified DOTS approach, in which the administration of TB medication and HAART was directly observed on weekdays and self-administered on weekends, resulted in a well-tolerated, acceptable and therapeutically beneficial treatment programme[
25]. Such a programme may provide an initial structured and supervised HAART experience from which transition may be made toward greater self-administration. Currently at the adjoining HIV clinic, patients self-administer their HAART and receive adherence counselling at five strategic points in their care. These sessions are used to familiarize patients with their diagnosis, the drug therapy they will be exposed to, relevant time-frames, the importance of adherence, and the consequences of non-adherence. These interactions with patients focus on empowering patients and promoting autonomy and commitment to treatment, while emphasising the benefits of adherence. These patients are also counselled on the risk of acquiring TB as well as the symptoms they should be vigilant for. At the TB clinic, patients receive minimal counselling, and few are familiar with their treatment plan or the consequences of non-adherence. In a health systems assessment of KwaZulu Natal, it was found that merely 29% of TB patients knew that their full course of treatments should be completed, and only 2% knew that TB had to be treated for a minimum of 6 months[
8]. These lessons from ART should be integrated into TB services to improve understanding of and adherence to treatment, stem the high TB treatment interruption rates and sufficiently educate patients in protecting themselves from re-infection and infecting others. An often neglected element in improving adherence is the commitment and attitude of health care workers who may not necessarily be a part of the formal adherence support programme. A supportive, sensitive, respectful and encouraging attitude from all healthcare workers may promote adherence by making clinic visits more pleasant[
8]. The growing call for increased community participation and patient-support as part of the STOP TB Strategy, can be gainfully directed to a less imposing and more empowering treatment programme[
4]. This approach will be more respecting of patients' rights to dignity and choice, as set out in the The Patients' Charter for Tuberculosis Care[
4]. A single strategy for bolstering adherence to both TB and ARV therapies makes more programmatic sense, and will be more logical and practical to patients, who, unlike the healthcare system, do not compartmentalize their illnesses. A less fragmented approach will advance the paradigm of 'treating the patient and not just the disease'.
Until recently, little attention has been paid to TB
infection control measures in the developing world. The growing evidence of nosocomial transmission of the infection in both inpatient and outpatient settings has failed to arouse a compelling attempt to address the issue[
26,
27]. In the era of the dual epidemics of TB and HIV, even less attention has been paid to the transmission of TB within HIV facilities. The risk of TB transmission to health care workers and other HIV infected patients is an understandable challenge to HIV services which articulate intimately with TB services[
26]. As a result of the increased risk of negative acid-fast bacilli sputum-smear results in TB-HIV coinfected individuals, there is often a delay in the diagnosis of TB, with a consequently increased exposure to other HIV service users and health care workers[
28]. The contribution of smear negative TB to ongoing transmission may have been underestimated, particularly in terms of transmission to immuno-compromised contacts[
28]. TB infection control measures in HIV services is based on a three level hierarchy of controls, including administrative or workplace controls, environmental controls, and personal respiratory protection[
2]. Low cost interventions include proper and innovative triaging of patients, with expeditious attention to coughing patients. Preventing re-infection of patients in the continuation phase of treatment may include scheduling new and follow-up patients on separate days. Open, well-ventilated, sheltered waiting rooms may be a simple alternative to more sophisticated, resource requiring infrastructure. Without making structural changes to our TB clinic, large waiting areas have been redesigned in favour of multiple smaller waiting areas with open air cover. In addition, a separate well ventilated area is used for known cases of drug resistant TB. Existing structures in established primary health care clinics, at which the majority of TB patients receive care, can be economically modified to improve infection control. In a model of nosocomial transmission of extensively drug-resistant TB in this setting, it was found that improvements to natural ventilation could prevent 33% (8–35%, 95% CI) of future XDR-TB cases[
27]. Personal respiratory protection is usually uncomfortable and when made available is ineffectively used in most health settings[
29]. Nonetheless, all patients are requested to use surgical masks, while health care workers are provided with personal respirators. However, there is currently no assessment of air quality changes, and staff often complain of a poor face seal of the masks as well as the discomfort of using the mask during the high temperatures experienced in this part of the world. While TB among health care workers constitutes a relatively small proportion of all TB cases, the criticality of staff availability and occupational safety warrants our attention. Promoting HIV testing among health care workers, redeploying HIV infected workers away from the TB unit and ensuring strict adherence to mask use may prevent more than two thirds of staff XDR-TB infections, and possibly offer similar benefits in drug sensitive TB[
27]. In addition, early diagnosis and treatment of TB as would be possible in an integrated service would in itself represent a public health infection control effort. Isoniazid prophylaxis remains an underutilized measure to reduce TB transmission in HIV services, especially among patients with a past history of TB[
2,
30].
Finally, HIV has propelled the innovative approach of
positive prevention for intervening in both the transmission of infection and the development of illness. The strong prevention arm of HIV programmes is only vaguely present in the TB programme, and is usually confined to isoniazid prophylaxis for contacts of TB patients. Key lessons and opportunities for positive prevention in TB have not been fully exploited. A more patient-centred approach to TB care may be able to recruit the active participation of TB patients in positive prevention efforts, which could include maximizing personal infection control, limiting exposure of social contacts to TB during the intensive phase of treatment, advocating isoniazid prophylaxis within the home and patient-centred education efforts to reduce overall transmission. Disclosure of HIV status remains a significant challenge to HIV prevention. While a pilot study at our clinic revealed that approximately two thirds of patients disclosed their status to one other person with the hope of receiving support from them, only few patients disclosed it to their sexual partner[
31]. The common law of South Africa places an ethical and moral duty on the health system to inform vulnerable uninfected partners of TB and HIV infected patients, after impressing upon the patient the importance of disclosure to their sexual partners. Importantly, these partners have a reciprocal moral and ethical duty to receive such information. Partner notification for HIV may subtly and opportunistically piggy-back on contact tracing efforts in TB. Even without this, contacts of TB who present to health services in response to TB contact tracing can be considered a high-risk, readily identifiable cohort for HIV testing. Every patient, therefore, serves as an entry point to a network of vulnerable individuals[
17]. The assessment of vulnerability in our context is a complex issue and undoubtedly requires a thorough evaluation of the pros and cons of facilitating such a disclosure to sexual partners of the patient.