Tuberculosis (TB) is a major public health concern and remains one of the most devastating diseases in the world. Although global TB incidence has been declining for the past few years, a new threat has emerged – multidrug resistant-TB (MDR-TB) – that is showing no signs of receding and is undermining progress in TB control. MDR-TB is defined as strains of TB not susceptible to two of the most potent first-line drugs, isoniazid and rifampicin.¹ Interrupted treatment of drug-sensitive TB is an important factor in the development of acquired drug resistance.² In 2017, the World Health Organization (WHO) reported the proportion of MDR-TB in new and previously treated cases of TB in Eswatini to be 6.8% and 16.0% respectively.¹ The country’s treatment success rate amongst MDR-TB cases remained at 68% (2015 cohort) and below the 2015 WHO target of 75% or higher.^3,4,5 So serious is the problem of TB in Eswatini that the government declared the epidemic already in 2011 a national disaster.

In response, the Eswatini Ministry of Health developed the National TB Strategic Plan (2015–2019)⁶ that is aligned within the framework of the End TB Strategy⁷ and the Sustainable Development Goals (SDGs).⁸ The plan aims to strengthen community drug-resistant TB management, increase access to treatment and ensure that 75% or more MDR-TB cases are treated successfully.⁶

In Eswatini, the standardised therapeutic regimen for MDR-TB consists of toxic, complex and expensive second-line anti-TB drugs that include a daily injectable for the first 8 months and treatment that takes up to 24 months to complete.⁹ The National TB Control Programme (NTCP) is in the process of scaling up the use of the WHO-recommended shorter MDR-TB treatment regimens (9 months instead of the usual 2 years) across the whole country. Patients with MDR-TB typically make daily visits to the closest primary health care (PHC) facility to receive directly observed treatment (DOT) and their injection from a professional nurse. DOT is a strategy recommended by the WHO that involves a health worker supervising patients with TB as they take each dose of their standardised medication.¹⁰

In Eswatini, rural areas carry a disproportionately high burden of MDR-TB. In addition, PHC facilities are often distant and geographically inaccessible from patients’ homes and are also frequently characterised by lack of front-line TB human resources for health (HRH).^11,12,13,14 In this context, front-line TB HRH refers to healthcare workers who are generally the community’s first level of contact within a PHC system for accessing TB services. To address the prevailing MDR-TB treatment access and HRH challenges in the predominantly rural Shiselweni region, in 2008, Médecins Sans Frontières (MSF) established a novel patient-centred community-based MDR-TB treatment strategy based on task-shifting within the existing NTCP. In this context, task-shifting refers primarily to the transfer of DOT and MDR-TB injection administration responsibilities, traditionally restricted to professional nurses, to trained lay community members. In extant literature, the task-shifting of highly differentiated clinical tasks such as injection administration to lay health workers has been limited to the delivery of contraceptives and injectable antibiotics in home-based management of newborn infections in rural and low-resource settings.^15,16

Under this model of care, patients with MDR-TB discharged from the MDR-TB treating hospital are linked to a lay community member (not a family member) of their choice from their neighbourhood, referred to as a community treatment supporter (CTS). This strategy allows patients with MDR-TB to receive DOT and their daily injections from CTSs in their (the patients’) homes without having to make daily trips to the nearest clinic. The selected neighbour must have sufficient literacy skills essential for reading the training manual in English and documenting administered injections. On inception, newly recruited CTSs receive hands-on demonstrations, mentored practice and theoretical training that focuses primarily on MDR-TB epidemiology, transmission, diagnosis, treatment or strategies (DOT), safe injection handling and adverse drug reactions. In addition, because of the potential risk of contracting human immunodeficiency virus (HIV) infection from needle stick injuries, newly recruited CTSs are routinely informed of the patient’s HIV status prior to undergoing training. The CTS’s responsibilities include patient confidentiality and procedures for patient referral to community MDR-TB nurses and other MDR-TB team members.

CTSs are directly supported by a team of professional healthcare workers made up of a doctor, community MDR-TB nurse, psychologist and social worker. Each CTS receives on-going supervisory visits from and at the discretion of the community MDR-TB nurse, to re-evaluate their skills and on-the-job training needs. CTSs accompany their patients with MDR-TB to the main health facilities for their monthly outpatient treatment review or unscheduled visits, in the case of worsening health condition. Although CTSs are typically recruited and trained to focus solely on MDR-TB, they are also important adjuncts in the provision of adherence support to HIV co-infected patients with MDR-TB on antiretroviral therapy. Core clinical decisions related to the MDR-TB treatment are taken by formal facility-based professional healthcare workers. Local administration of DOT and intra-muscular injections by a CTS was introduced to reduce some of the social and financial difficulties frequently faced by patients in accessing the often more distant centralised clinic-based MDR-TB treatment services.¹⁴

Despite the decade-long implementation of the task-shifting strategy in the Shiselweni region, the Ministry of Health is yet to officially endorse the strategy. Existing NTCP strategic plans and MDR-TB treatment guidelines only provide for task-shifting of certain professional nurses’ responsibilities such as DOT supervision and TB treatment defaulter tracing to lay health workers. Professional bodies in Eswatini have voiced ethical concerns about standards of care and safety risks for both patients and CTSs ranging from potential errors in dosing to transmission of infections through unsafe injection handling and inappropriate community MDR-TB infection prevention and control practices.¹⁷ These fears were heightened by the increased recognition of the role community transmission played in the publicised outbreak of extensively drug-resistant TB in the neighbouring KwaZulu-Natal province of South Africa.¹⁸

Authorities have also raised questions regarding the sensitive nature of aspects of care delegated to lay community members, in particular, whether the CTSs respect patients’ confidential medical information.¹⁷ This is important given the stigma that surrounds the disease. Despite many years of public health education, MDR-TB is presumed to be incurable.^19,20 CTSs learn about the medical condition and may view the intimate anatomy of patients as they administer and coordinate care. The CTS–patient encounter also takes place in the home setting, which is often not conducive for maintaining privacy and confidentiality.

Assessment of community-based MDR-TB services, specifically the task-shifting of care to CTSs, has often been restricted to clinical outcomes without accounting for patients’ satisfaction with health services received. There is strong evidence that patient satisfaction with healthcare services received can instrumentally affect treatment uptake, adherence and retention in TB care.^{21,22,23,24,25} The need to incorporate the patients’ perspective of TB care received is emphasised in the International Standards for Tuberculosis Care (ISTC)²⁵ and the Compendium of Best Practice Guidelines of the International Union Against Tuberculosis and Lung Disease (The Union).²⁶ Drawing from a myriad of existing WHO recommendations, the ISTC and The Union support the call for a patient-centred approach based on the patient’s needs and mutual respect between the patient and the healthcare service provider. By administering DOT and MDR-TB injections, CTSs perform a critical public health responsibility that carries a high level of obligation to the community and the individual MDR-TB patient. Thus, as frontline HRH in the provision of MDR-TB care, the interaction of CTSs with patients may considerably influence treatment outcomes.

Patient satisfaction with TB care refers to the match between the patient’s prior expectations of care and subsequent perception of services actually received.^19,20 Patient satisfaction may be seen to centre on the fulfilment of the following expectations: reasonable waiting time for care, respectful and considerate interaction with healthcare providers, accurate diagnosis and optimism of a good prognosis, continuity and coordination of care, affordability and the professional competence of service providers.^21,22,27,28

Despite the recognition of the importance of determining levels of patient satisfaction as a measure of the quality of TB services,^21,22 no study to date has evaluated patients’ satisfaction with receiving DOT and intramuscular MDR-TB injections from CTSs in the Shiselweni region. In view of the growing global experience of the potential role of patient satisfaction in optimising clinical outcomes, particularly with respect to TB services, this study was conducted to assess the level of patients’ satisfaction and their perceptions of the different dimensions of MDR-TB care delegated from professional nurses to CTSs in a rural region of Eswatini. Results will help in formulating recommendations to optimise the task-shifting approach in community-based MDR-TB management in resource-limited settings and contribute to the achievement of various MDR-TB control targets as outlined in the national TB Strategic Plan, End TB Strategy and SDGs.

Half of the participants (n = 39; 50.0%) were aged between 31 and 40 years (Table 1). Almost 9 in every 10 of the participants (n = 69; 88.5%) had received DOT and MDR-TB injections from a CTS for at least 4 months. Almost two-thirds (n = 47; 65.3%) were not married and more than half (n = 43; 55.1%) had low literacy and numeracy skills (primary education or lower). Most of the patients with MDR-TB interviewed (n = 65; 83.3%) were not employed. More than three-quarters of patients interviewed (n = 67; 85.9%) had an estimated monthly household income of less than ZAR1000 (= US$74.07 [exchange rate on 31 May 2017]). Most patients (n = 42; 53.8%) lived more than 20 km from the nearest PHC facility. A large majority (n = 75; 93.6%) relied on public transport to travel to the nearest healthcare facility.

The overall mean score of general patient satisfaction for receiving DOT and intramuscular MDR-TB injections from a CTS was 4.4 (standard deviation: 1.0). Table 2 illustrates patient socio-demographic factors influencing mean general patients’ satisfaction scores. All socio-demographic factors were considered as predictor variables in the adjusted model. The study revealed no significant association between mean general satisfaction score and the other patient socio-demographic characteristics.

A Spearman correlation coefficient was computed to examine the relationship between mean scores of general satisfaction and satisfaction with nine specific measures of MDR-TB services provided by CTSs (Table 3). There was a weak correlation between treatment costs, r (78) = 0.27, p = 0.015 and general satisfaction. Overall, the weak but positive correlation between treatment costs and general satisfaction scores suggests that this dimension assesses factors other than the patient’s general satisfaction.

Ordinal regression models were used to predict regression coefficients for the mean general satisfaction score for each of the nine dimensions of MDR-TB services provided by CTSs (Table 4). All nine dimensions of MDR-TB services provided by CTSs were included as potential explanatory variables in the adjusted model. In the adjusted model, adherence counselling (p = 0.019), confidentiality (p = 0.026), provider selection (p = 0.037) and treatment costs (p = 0.014) significantly predicted mean general patient satisfaction.

Approximately four-fifths (n = 62; 79.5%) of the respondents indicated that they were likely to recommend their friends or family to receive DOT and MDR-TB injections from a CTS. More than half (n = 44; 56.4%) of the respondents suggested broadening of the scope of services provided by CTSs to include treatment of minor ailments or side effects of the medication such as nausea, vomiting and loss of appetite. More than one-third (n = 27; 34.6%) of the respondents identified the need to adequately train CTSs on respecting the privacy and confidentiality of their condition and medical information. Just more than a quarter (n = 20; 25.6%) of patients highlighted the need for regular supervision of and sufficient incentives for CTSs as ways of improving MDR-TB services.

Previous studies evaluating community-based MDR-TB models of care have focused on clinical outcomes^32,33 without paying attention to patients’ satisfaction. As far as could be ascertained, this is the first study to assess MDR-TB patients’ views and the extent to which their perceptions of various aspects of care experiences provided by CTSs are associated with overall satisfaction with community-based MDR-TB treatment and care in the rural Shiselweni region. This study further contributes to the current national debate on task-shifting of DOT and injection administration to lay community members by providing the patients’ perspective and suggestions of ways to improve patient experiences in practice.

The overall mean score of self-reported general patient satisfaction in this study was fairly favourable: 4.4 based on a rating scale ranging from 1 (very low satisfaction) to 5 (very high satisfaction). There is a paucity of literature describing patients’ perspectives on receiving MDR-TB treatment, with most researchers limiting their assessment to patient satisfaction with drug-sensitive TB services. Patient satisfaction is considered to be a highly desirable outcome measure of clinical care, influencing both treatment adherence and continuity of care. Findings from studies on levels of general satisfaction with community-based TB services conducted in various African settings in diverse geographical locations, albeit limited in number, have often been inconsistent.^21,22,24,30 The mixed results are not unexpected because there are huge variations in how TB care is offered across different settings.

In studies conducted in South Africa²² and Nigeria,²⁴ patients’ socio-demographic characteristics such as age, gender and marital status have been reported to contribute significantly to patient satisfaction. Conversely, in this study patient characteristics were not significantly related to patient satisfaction, with the exception of the mode of transport. Patients with MDR-TB in this study also suggested the need to broaden the scope of services provided by CTSs to include treatment of minor ailments and side effects of the medication. Taken together, these findings provide evidence that patients in poor rural settings face many barriers in accessing PHC such as transport limitations in reaching the distant and geographically inaccessible clinics.^12,14 This is significant in the light of findings from a study in rural South Africa that reported time lost travelling for TB treatment as a potential cause of missed appointments and threat to treatment adherence and retention.³⁴ The findings from this study provide an impetus for policymakers to consider progressively increasing, where applicable, the geographical distribution of PHC facilities in rural areas. There is also a need for more research to understand the extent to which transport limitations impede access to MDR-TB services in rural settings.

In this study, four dimensions of care experience – adherence counselling, confidentiality, provider selection and treatment costs – were found to significantly predict overall patient satisfaction. These findings could assist authorities in framing specific strategic interventions to improve patient satisfaction and strengthen patient-centred community-based MDR-TB care. It was rather surprising to find that the other dimensions of MDR-TB care provided by CTSs were not related to general patient satisfaction because these aspects have been linked to satisfaction in prior studies.^{21,22,23,24,27,30,35}

MDR-TB treatment regimens are lengthy with a high daily pill burden carrying frequent side effects and requiring high levels of adherence to function optimally. Yet, the level of adherence to MDR-TB treatment varies significantly in different settings.^36,37,38 As such, adherence counselling is considered an integral component of a worthwhile care experience.^21,24 Consistent with results from previous research,²⁴ this study found adherence counselling to significantly influence patient satisfaction. These results suggest key targets to optimise the delivery of community-based MDR-TB care by CTSs. The WHO³⁹ and the Eswatini treatment guidelines⁴⁰ for drug-resistant TB recommend that healthcare workers should regularly assess, reinforce and support patient adherence to complex MDR-TB treatment plans because of its importance to successful treatment. Ongoing support for CTSs should recognise their critical role in adherence counselling and reinforce their training about the disease, treatment methods, common complications and skills essential to optimise adherence to care. On the basis of this finding, further research is needed to understand adherence counselling practices of CTSs that may guide the development of comprehensive MDR-TB treatment adherence interventions.

Findings from this study point to treatment costs as one of the dimensions of MDR-TB care that predicted general patient satisfaction. Although MDR-TB treatment is generally offered free of charge, a considerable share of the costs are incurred by patients and their households. Patients with MDR-TB are confronted with costs associated with visits to the healthcare facilities such as transport, food and accommodation and loss of income because of inability to work. These costs impose a financial strain on patients and their households, potentially creating barriers to treatment access and adherence and increased treatment interruptions.^41,42 In the present study, a majority of patients were unemployed and reported a low monthly household income. One of the main milestones of the End TB strategy is that no families affected by TB face catastrophic costs by 2020.⁶ As such, various levels and types of social protection support measures should be considered by policymakers to assist patients with MDR-TB and their families. Future studies should assist policy makers in identifying, understanding and mitigating the effect of MDR-TB treatment on the financial status of patients and their families.

This study found that provider selection influenced general patient satisfaction. Perrault⁴³ established that patients who purposely participated in choosing their healthcare provider reported higher satisfaction. These patients often preferred to choose providers with whom they felt they shared some similarity. Patients with MDR-TB in this study context purposely selected a CTS of their choice from their neighbourhood. Thus, the CTSs worked in the community they lived, typically sharing similar culture, language, religion, socio-economic status and life experiences with their patients with MDR-TB. The findings of the current study may suggest that provider selection empowered patients with MDR-TB to play a significant involvement in their own care by participating in arguably one of the most key decisions – the initial nomination of a DOT and injection provider. This is important given the recent calls by international bodies challenging national TB programmes to develop a broader interest in patient-centred care and systematically consider the perspectives of patients in the planning and provision of TB care.^25,26

Prior to decentralisation of MDR-TB treatment, few people other than professional healthcare workers were de facto custodians of patients’ health information, a situation that has changed with the task-shifting of clinical tasks to CTSs in the Shiselweni region. The observation of the confidentiality of patients’ medical information is recognised as a central tenet in engendering patient satisfaction with TB care.²² The present study found a link between the overall satisfaction of patients with MDR-TB and their perception of the maintenance of confidentiality as they interacted with CTSs. A breach of confidentiality can erode patients’ trust in their healthcare provider and may have grave repercussions such as marital difficulties, discrimination and loss of employment.^19,20 This observation is important in the light of an assertion made by Dapaah and Senah⁴⁴ in a study carried out in rural Ghana on the socio-psychological trauma patients experience in accessing HIV services. The authors reported that patients were actually more worried about the social consequences of treatment than the medical outcomes of accessing care.

The task-shifting of DOT and intramuscular MDR-TB injections to lay healthcare workers in the Shiselweni region has so far gone ahead despite reservations and apprehensions from policymakers and NTCP functionaries.¹⁷ Policymakers have raised ethical concerns related to the perceived inability of lay health workers to observe the privacy and confidentiality of patients and also recognise and report the adverse events of MDR-TB treatment such as hearing loss. A substantial proportion of patients with MDR-TB in this study recommended the need for adequate training on respecting patients’ confidentiality, regular supervision and appropriate incentives for CTS. Implicitly, these suggestions potentially represent concerns from participants about CTSs’ deficiencies in those aspects of care. Therefore, it is recommended that authorities consider formalising the task-shifting practice and facilitate the development of appropriate competency-based pre-service and on-the-job training, optimisation of supportive supervision of CTSs and active surveillance of treatment adverse events and breaches of confidentiality in the community based MDR-TB treatment strategy. Future studies could examine ways to improve CTSs’ respect for and maintenance of the confidentiality of patients’ medical information.

In this study, almost four in every five of the respondents indicated that they were likely to recommend to their friends or family to receive community-based MDR-TB care from a CTS. This is significant in the context of scaling up the community-based programme implemented in the rural Shiselweni region. Without discounting the significance of other factors, the fact that patients with MDR-TB would recommend to their significant others to receive care from a CTS is considered a valuable indicator of their personal experience of receiving services from a community-based healthcare provider.^45,46 The programme’s reputation (or level of former or current patients’ satisfaction) is likely to be the only information available to prospective patients with MDR-TB or their care givers to help them choose a preferred provider or model of care. Previous satisfaction is also important for former patients to return for further care to the same healthcare provider or programme in the future. The present study findings reveal new insights for programme managers and policy makers targeting new patients with MDR-TB, retaining existing ones and generally increasing access to MDR-TB care.

The use of patient perspectives to assess satisfaction with aspects of care by providers is not without problems. Tension exists regarding the subjective nature of patients’ evaluation and questions have been raised as to whether their views truly represent the quality of care or the healthcare workers’ interpersonal skills.²² Patient satisfaction is considered to reflect the service dimension rather than the technical aspect of quality of care.²² Participants’ responses may also reflect socially desirable answers rather than their true views and experiences. With this study mostly founded on self-reported satisfaction, social desirability bias may have occurred. However, this was partially countered by assuring respondents of the anonymity of the questionnaires that findings will be reported by using aggregate analysis and that the outcomes of the study would not affect their care in any way.

The sample for this study was relatively small (n = 78), and participants were selected purposively. Small samples may be inadequate to provide reliable correlations. Furthermore, the strength of association between general patient satisfaction and the dimensions of care that were significant was weak. Although the current study included a number of possible underlying dimensions of care influencing satisfaction, future studies with a broadened spectrum of dimensions of care may explain some of the differences in patient satisfaction reported in this research. In addition, satisfaction with healthcare services cannot be adequately assessed without accounting for a patient’s self-perceived health status. Literature suggests that the overall satisfaction rating is significantly influenced by the patient’s severity of illness.^47,48 Although the present study excluded patients with MDR-TB who were too ill or had profound deafness from medication toxicity, future studies on patient satisfaction with the community-based MDR-TB model should include a measure of the participant’s health status.

In spite of these limitations, the findings from this study provide valuable insights on how the services provided by CTSs are perceived by patients and what authorities may have to prioritise to improve patients’ satisfaction.

The findings provide strong evidence that patients were satisfied with the task-shifting of DOT and MDR-TB injection administration to CTSs in the study area. The study also provided some insights into the four dimensions of care grounded in adherence counselling, confidentiality, provider selection and treatment costs that significantly influenced patients’ general satisfaction. These aspects of MDR-TB care will need to be addressed to improve overall patient satisfaction. Provision of adequate training, regular supportive supervision, sufficient incentives and broadening of the scope of services offered by CTSs to include treatment of minor ailments were recommended as ways to optimise the delivery of community-based MDR-TB care.

The expansion of tasks delegated to lay health workers to include administration of important injection medication has great potential. Evidence from this study suggests the need for policy and regulatory support to formalise the task-shifting practice; developing interventions for active surveillance of treatment of adverse events; and improving patient experience of privacy and confidentiality. Future studies could further explore the barriers to and facilitators of the delivery of patient-centred MDR-TB care based on the task-shifting of clinical tasks to lay health workers.