Ambulatory care for multidrug-resistant tuberculosis: lessons learned in Addis Ababa, Ethiopia

Abstract

Setting: Ethiopia is one of the high multidrug-resistant tuberculosis (MDR-TB) burden countries. Efforts by the National TB Programme to control MDR-TB include expanding ambulatory care.

Objective: To investigate the opportunities and challenges faced by treatment follow-up health centres (TFCs) when managing MDR-TB patients, with greater focus on recording, TB infection control (IC) and supervision practices.

Methods: A facility-based cross-sectional study was conducted by reviewing the records of all MDR-TB cases in all 25 TFCs in Addis Ababa, Ethiopia. The TB focal point, pharmacy and laboratory heads were also interviewed.

Result: A total of 221 MDR-TB patients were registered; 157 (71%) patients had been referred from one of the two treatment initiating centres. While some TFCs oversaw up to 41 patients, others had just one patient. The majority of the TFCs (n = 21, 84%) followed standardised TB IC procedures. Poor documentation of patient information was observed at all sites; for example, human immunodeficiency virus and current treatment status was not indicated for respectively 86 (38%) and 41 (18%) patients.

Conclusion: The study revealed that infection prevention practices were largely adhered to. Documentation of patient-related information was a major challenge, and regular supervision of the TFCs should be emphasised. Record keeping is critical.

Multidrug-resistant tuberculosis (MDR-TB) is a threat to global TB control and is of major public health concern in several countries.¹ Ethiopia is one of the 27 high MDR-TB burden countries that together account for >85% of estimated cases globally.² Ethiopia had an estimated 2010 (range 1200–3000) MDR-TB cases in 2012, and was one of the 94 countries to notify at least one case of extensively drug-resistant TB (XDR-TB) by the end of 2012.¹ The MDR-TB programmatic response in Ethiopia began in 2009, when the Green Light Committee (GLC) of the World Health Organization (WHO) approved a pilot programme to treat 45 MDR-TB patients at St Peter's Specialised TB Hospital in Addis Ababa. A model of care based on hospitalisation was adopted for the full course of treatment. In 2010, this policy was changed to allow for more rapid expansion of services, whereby clinically stable patients are transferred to treatment follow-up health centres (TFCs) to continue their treatment on an ambulatory basis upon smear conversion.^3,4

The National TB Programme (NTP) is responsible for the programmatic management of MDR-TB in Ethiopia. A national MDR-TB policy and guidelines were developed, treatment initiating centres (TICs) and TFCs were set up, a monitoring and evaluation (M&E) system was developed, capacity of health care providers was created through training, and MDR-TB treatment services were rapidly scaled up from nine sites (one TIC and eight TFCs) in 2009 to 110 (10 TICs and 100 TFCs) in 2013. Treatment is available all over the country, with TICs in four of the nine regions and three additional centres in the two city administrations (Addis Ababa and Diredawa). To date, more than 940 MDR-TB patients have been enrolled on treatment. The treatment success rate currently stands at 70%.⁴ The NTP has clearly outlined the responsibilities for TICs and TFCs. The role of TICs includes the investigation of suspected cases, initiation of treatment among admitted MDR-TB patients, provision of mentoring and programmatic support to TFCs and quarterly reporting on progress to the NTP. Each TFC is required to manage all referred patients by providing daily medications to patients, referring patients with severe complications and drug side effects back to the TIC, and monthly standardised recording and reporting to the TIC.^3,5

In 2011, Ethiopia adopted the WHO ambulatory care model for MDR-TB management. Through this approach, MDR-TB patients are investigated as out-patients and their MDR-TB treatment initiated as out-patients as long as they are sufficiently well and do not require admission. Studies have shown that management of MDR-TB patients imposes substantial operational challenges in resource-constrained settings, such as patients lost to follow-up, development of severe drug side effects, the need to manage comorbidities, and stigma surrounding MDR-TB care both in the society and among health professionals.^6–8

The objective of the present study was to investigate the opportunities and challenges faced by TFCs when managing MDR-TB patients, with a focus on recording, infection control (IC) and supervision practices.

METHODS

A facility-based cross-sectional study was conducted at all 25 TFCs in Addis Ababa in November 2013. MDR-TB patient charts and health centre MDR-TB registers were reviewed using a structured data capture form. Socio-demographic and clinical data as well as programmatic information were extracted. If information was missing at the TFC, an attempt was made to collect the information from the TIC by record linking. In addition to each TFC, three staff members, the TB focal point, the person in charge of the pharmacy and the person in charge of the laboratory, were interviewed using a pre-tested structured questionnaire focusing on their respective thematic area. Questions were asked about the role they played in MDR-TB treatment follow-up, availability of a TB IC plan, health care worker (HCW) training on TB/MDR-TB, patient education about TB, supervision and mentoring received from TICs and the availability and use of personal protective equipment. In addition, detailed questions were asked about the management of second-line anti-tuberculosis drugs at the TFC level and about the involvement of laboratory units throughout the course of treatment at the TFCs.

Three nurses and one health officer from public health centres in Addis Ababa were trained as data collectors. Data collection was supervised daily by the investigators and two additional supervisors from the Addis Ababa Regional Health Bureau (RHB). Data capture forms were checked for consistency, completeness, clarity and accuracy on a daily basis. Data were double-entered using Epi Info™ version 3.5 (Centers for Disease Control and Prevention, Atlanta, GA, USA) and analysed using the STATA statistical package version 12 (Stata Corp, College Station, TX, USA).

Ethical clearance was obtained from the ethical review board of the Regional Health Bureau of the Addis Ababa City Administration and the ethical clearance committee of St Peter's Specialised TB Hospital.

RESULTS

Characteristics of MDR-TB patients and documentation practices

A total of 221 MDR-TB patients (115 male and 106 female) had been managed in the 25 centres since the start of their operations as TFCs. Nearly three quarters of the patients (n = 157, 71.0%) had been referred from St Peter's Specialised TB Hospital and 64 (29%) from the All African TB and Leprosy rehabilitation and training centre (ALERT) Hospital. Almost three quarters (71.1%) of the patients were aged between 15 and 34 years. Most patients (n = 185, 84%) were residents of Addis Ababa; however, residential address was not documented for 23 (10.0%) patients. The average number of patients treated in the TFCs was 8.8; most TFCs (15 of the 25) had 4–11 patients, 5 had only 1 or 2 patients, 4 clinics had 18–22 and one had 41 patients (Figure).

FIGURE.

Distribution of multidrug-resistant tuberculosis patients in the 25 tuberculosis follow-up centres (n = 221). HC = health centre.

Eighty-six (39%) patients did not have their HIV status documented in the registers kept at the TFC or the TIC (Table 1). For those with documented information, 135 (17.0%) were HIV-co-infected. At the time of data collection, current treatment status was not documented for 41 (18.6%) patients. For patients recorded in the registers, 132 (60.0%) were currently on treatment, 31 (14.0%) had completed treatment, 9 (4.1%) had been referred back to the TICs and 8 (3.6%) had died (Table 1). In addition, seven TFCs (28.0%) were using an MDR-TB treatment card that was different from the NTP format, and were instead using a card provided by a partner organisation supporting the national MDR-TB programme.

TABLE 1.

Characteristics of MDR-TB patients in treatment follow-up health centres, Addis Ababa, Ethiopia (n = 221)

TB infection control measures in TFCs

TB IC activities were assessed and classified according to the TB IC hierarchy.⁹ In this study, respectively 24 (96%) and 17 (68%) TFCs had a functional TB IC committee and a written TB IC plan. In terms of administrative TB IC measures, 20 (80.0%) TFCs triaged coughing patients by separating and fast-tracking them. However, there was no dedicated arrangement and appointment system for MDR-TB patients, and they were mixed with other, drug-susceptible TB patients when visiting the facility. In all the 25 TFCs, TB IC posters on the importance of cough hygiene and opening windows were displayed in the waiting areas. With respect to screening of staff for HIV infection and active TB, 23 (92%) of the TFCs did not offer confidential HIV counselling and testing services to their staff and 24 (96%) TFCs did not provide staff with symptomatic TB screening either (Table 2).

TABLE 2.

Practice of administrative infection control measures in the 25 MDR-TB treatment follow-up centres in Addis Ababa, Ethiopia (n = 25)

In terms of environmental TB IC measures, 21 (84.0%) TFCs had waiting areas with partial walls which provided good natural ventilation. In 20 (80%) TFCs, the TB/MDR-TB clinic had windows and doors that were on opposite sides, providing cross-ventilation. In addition, windows in the TB clinics were renovated, making them permanently open to ensure maximum natural ventilation at all times. The seating arrangements for staff and patients in most of the treatment rooms were categorised¹⁰ as ‘good’ (n = 16, 64%) or ‘good alternative’ (n = 6, 24.0%).

In terms of personal respiratory protection, 22 (88.0%) TFCs had a stock of FFP2 or N95 respirator masks for staff. Respirators were used by the TB focal person or by other staff while providing directly observed treatment (DOT) services for TB/MDR-TB patients (Table 3). However, none of the TFCs had a fit-testing procedure and none of the TFCs provided masks for patients.

TABLE 3.

Availability of environmental (seating arrangements) and personal protection measures (respirators) for TB infection control at the MDR-TB treatment follow-up centres, Addis Ababa, Ethiopia (n = 25)

Supervision

Twenty-three (92.0%) TFCs reported having received at least one supervisory visit by staff from the TICs since the start of their MDR-TB services, although quarterly visits should be conducted. Supervision had not been documented, and appeared to have been unplanned. No regular routine supervisory visits, except for irregular visits by different partner organisations, were carried out.

TFC staff interviews

A total of 75 staff were interviewed, three in each facility. Key findings from the staff interviews concerned the pharmacy: in all the TFCs visited, second-line anti-tuberculosis drugs were not integrated into the overall pharmacy management system but were being managed separately by the TB clinic in the TFC under the supervision of the TIC. For the laboratory, the involvement of laboratory staff in the management of MDR-TB at the TFC was not clearly outlined. Only 8 of the 25 TFCs (32%) said that they performed acid-fast bacilli testing for follow-up sputum smears of MDR-TB patients. The other 17 (68%) TFCs indicated that the laboratory workup of patients, both before and after diagnosis, was being done by the TICs as part of patient follow-up during the patients' appointments at the TICs.

DISCUSSION

The Ethiopian NTP recently introduced ambulatory TB care to accelerate access to MDR-TB treatment. At the time of the study, a total of 221 MDR-TB patients had already been referred from two TICs for ambulatory treatment and care to 25 TFCs in Addis Ababa. It is important to document experiences and learn from opportunities and challenges to improve the management of patients at the TFCs in case of rapid expansion. This study assessed all TFCs providing treatment and follow-up services for patients transferred from TICs in Addis Ababa.

The assessment showed that, overall, the TFCs in Addis Ababa had satisfactory TB IC strategies in place and were adhering to the national TB IC guidelines. However, screening of TFC staff for TB and HIV and protective measures for TB patients should be improved. Completeness of documentation of patient information was inadequate. For example, HIV status and current treatment status were frequently not recorded, and the treatment cards used were not uniform. This is potentially related to weak supervision practice of the TFCs by the TICs and the NTP as well as lack of knowledge about the importance of recording. Supervisory visits and support for the TFCs by staff from the TICs was weak and irregular. This could be because clinical staff at the TICs were over-burdened with other activities in their respective facilities. Furthermore, no joint monitoring and planning system has been established among the TFCs and TICs.

The NTP has recognised the importance of implementing TB IC measures in health care facilities and other congregated settings: a TB IC policy guideline was developed, health care providers were trained and renovation works, particularly in MDR-TB treatment settings, were carried out.^5,9,11 TB IC, especially in the context of drug-resistant TB, needs to be a leading priority of the programme, and it is important to protect health workers.^9,11 A study conducted among health care providers in South Africa working in MDR-TB clinics concluded that there was considerable fear of contracting MDR- or XDR-TB among staff. This could negatively impact the provision of quality patient care.¹² To minimise the risk of transmission to health care providers, a combination of different TB IC measures should be implemented.^3,5,9 However, implementation of the full package of TB IC measures may not always be feasible due to several barriers. A study conducted in Uganda on the implementation of TB IC measures at health facilities found that several barriers, including space, funds and social factors, were associated with facility-level implementation of all components of TB IC measures.¹³ Our findings showed that nearly all aspects of TB IC measures were implemented, although there was room for improvement, as 32% of TFCs did not have a written IC plan. Although personal protection was offered to staff, there was no fit-testing procedure and patients were not provided with masks. In addition, routine TB and HIV screening services were not offered to HCWs.

Accurate and complete recording and reporting are key activities in TB control. TFCs should properly document patient information and update the patient treatment follow-up charts. This study observed that key patient information was missing. For 39% (n = 86) of patients, HIV status was not documented, while current treatment status could not be obtained from the records kept at the TFC or the TIC for 41 patients (18.6%). This could lead to sub-optimal patient management, including diagnosis and treatment of comorbidities and drug side effects that may develop, patient loss to follow-up, and administration of inaccurate doses of drugs or even prescribed medicines. Programmatically, under-reporting of services undertaken by the TFCs could occur.

Coordination of patient management should be evidenced by regular supervision of the TFCs by staff from the TICs. Implementation was found to be weak, and supervision visits were few, irregular and not jointly planned by the two centres. Moreover, visits were not supported by written feedback or detailed follow-up. It is necessary to strengthen the working relations between the TICs and TFCs. One way to do that could be by holding regular catchment area meetings attended by representatives from the TFCs, TICs, the health bureau, health offices in the sub-cities and other relevant stakeholders. A uniform M&E system at all levels requires further strengthening and should include the provision of training for health care providers.^1,2

Limitations of the study were that no interviews were conducted with patients and only a limited number of staff of the TIC and TFCs were interviewed. Questions on patient support and side effect management were not asked during the interviews. This could have resulted in important information being missed. Furthermore, no interim outcomes for MDR-TB patients could be provided due to the incompleteness of patient records, which is a significant finding in itself. Despite these limitations, important lessons were learned for the further scale-up of MDR-TB ambulatory care.

A significant proportion of TFCs for ambulatory care in Addis Ababa had satisfactory IC practices, although staff screening and use of personal protective measures could be improved. Poor documentation of patient-related information was seen as the major challenge. Regular supervision of TFCs and the monitoring of data use should be emphasised. Strengthening of documentation practices in these centres is critical to guarantee proper case management. Improvements could be made in the scale-up of ambulatory care in Ethiopia.