Background
A pragmatic public health approach to tuberculosis control has led to a decline in mortality from the disease worldwide with an estimated 37 million deaths prevented between 2000 and 2013 [
1]. However, the emergence of multi-drug resistant TB (MDR-TB) has threatened the progress made in TB control globally [
2]. MDR- TB is defined as resistance to Rifampicin and Isoniazid, the most effective first line anti-TB drugs [
3]. Globally, there were an estimated 480,000 cases and 190,000 deaths from MDR-TB in 2014 [
4]. In Nigeria, 2.9% of new cases and 14% of previously treated TB cases were estimated to have had MDR-TB in 2014 [
4].
The latest WHO guidelines stipulate a shorter MDR-TB treatment regimen of 9–12 months under specific conditions, as well as the conventional regimen of 20 months, and recommend models of care based principally on ambulatory/home based care rather than models of care based mainly on hospitalization [
5]. The intensive phase of conventional MDR-TB treatment regimen involves daily injections of aminoglycosides as well as orally administered potentially toxic and less effective anti-TB drugs for 8 months and oral medication is continued in the last 12 months of treatment [
6]. However, national guidelines for the management and control of MDR-TB in Nigeria are based on previous WHO guidelines and recommend admission of patients into specialised centres until sputum samples are ‘culture negative’ [
7]. In practice patients are admitted and confined to one of ten specialized treatment centres in Nigeria for the initial 8 months of intensive chemotherapy [
8].
The literature describes patient centred models of care, indicating that home or community based models are favoured by patients, family, community members and health workers [
9‐
11]. Indeed, community based care can be more cost effective [
2,
12], provide easier access to treatment, and patients are more likely to be able to seek support from their social network [
13,
14]. It enhances the psychosocial support available to patients and enables patients to continue earning a living [
10]. HIV/MDR TB co-infected patients might benefit most from this approach. They suffer from severe drug side effects and stigma and need substantial social and emotional support, as shown in India [
9].
The available literature suggests that hospital based care in Nigeria promotes adherence to treatment during the intensive phase of treatment [
8]. Perspectives of MDR-TB patients are shared in social media [
15] and rigorous research on patient views of MDR-TB treatment is limited to their experiences of home based care [
9‐
11]. There is a need to better understand views of hospitalized MDR-TB patients and providers and how this is linked to adherence and accessibility of care and support services for MDR-TB [
16], in line with international goals [
3].
We therefore aimed to (1) explore the physical, social and psychological needs of hospitalized MDR TB patients, (2) examine providers’ perceptions about the hospital based model and (3) discuss the model’s advantages and disadvantages from the patient and the provider perspective.
Methods
Study site
The participants for this study were recruited from the University of Port Harcourt MDR-TB treatment centre. The hospital is operated by the University of Port Harcourt Teaching Hospital with support from the Institute of Human Virology of Nigeria (IHVN) and the National TB Control Programme. The IHVN provides training and supplies with funding from Global Fund to fight AIDS, Tuberculosis and Malaria. MDR-TB patients are diagnosed at selected TB clinics across the country, kept on a waiting list and recalled for admission through phone calls or rarely home visits when bed space is available. Due to limited bed space, approximately 30 patients are enrolled in a treatment cycle. Once most of these patients are confirmed culture negative for MDR-TB in a treatment cohort, no new admissions are made due to concerns about reinfection of the cohort that have almost completed treatment. The head physician performs both administrative and clinical functions. Other doctors are on rotation from the teaching hospital and are not accommodated in the facility for on-call duty. Due to acute shortage of staff, only six nurses run shift duty in the hospital. The caterers are non-resident contractors hired by the administrators. Meals are prepared outside the facility and transported to the centre three times daily.
Design
We used a qualitative approach to understand patients’ and providers’ experiences of hospital-based care for MDR-TB patients, since this uncovers motivations and values which are shaped by everyday circumstances and culture and helps understand how these influence needs and health behaviour.
Discharged patients of different age, sex, marital status and ethnic origin were purposively selected from facility-based hospital records for two gender distinct focus group discussions (FGDs) comprising six men and six women and 11 in-depth interviews (comprising six men and five women). We included patients who were already culture negative for MDR-TB, had spent at least six months on hospital admission, and were discharged within the preceding 12 months. We selected four healthcare providers who played key roles in managing MDR-TB patients in the facility for key informant interviews. We recruited patients attending clinic for regular appointments. Following their specified appointment, the researcher handed patients an information sheet in a sealed envelope. Non-literate participants were taken to a safe and confidential location in a nearby health centre where the content of the information sheet was explained to them using Pidgin English. Participants were allowed to return home with the information sheet and were encouraged to seek clarification from the researchers through phone calls billed to the researchers’ accounts. Participants were further contacted through phone calls to confirm their willingness to participate in the study and to make appointments for data collection.
Data collection
Separate focus group discussions for men and women were held at a primary healthcare centre on weekends when the centre was closed for regular work. Each session, which lasted between 45 and 60 min, was audio recorded with a digital recording device. Informed consent to participate and to audio record the discussion was sought and obtained from the participants before the start of each focus group discussion. KB facilitated both focus group discussions using a series of core questions and probing questions when needed to gain understanding of the issues while an assistant researcher took notes and managed the recording device. Once the main topics were covered, the facilitator summarized and gave the participants an opportunity to discuss issues they considered important which were not covered initially. At the end of each focus group discussion, participants were offered refreshment and transport tokens before they returned home.
In-depth interviews were conducted with consenting participants from the FGD in the coming days after the first focus group discussion. However, three males and three females who were not involved in the FGD were also selected for the interviews. In-depth interviews allowed further exploration of issues raised in a more relaxed atmosphere, one to one, to allow participants express their own experiences and perspectives more freely. Furthermore, in-depth interviews provided an opportunity for researchers to explore individual experiences rather than social norms that evolved during treatment. Each interview lasted between 20 and 40 min at the health centre after regular working hours or at participants’ homes at their discretion.
Key-informant interviews were held following the completion of the focus group discussions and most of the in-depth interviews. This allowed the researchers to focus on issues raised by patients in the focus group discussions and in-depth interviews, and obtain provider perspectives and insights. Key informants were given the information sheet at least 48 h before the interview, were given the opportunity to ask questions or clarifications before deciding to participate. Informed consent to participate and to audio record the conversation was sought and obtained at the start of each interview. Each interview lasted for about 30 min.
Data collection took place between May and July 2014. Topic guides, initially developed from the study objectives, were field tested before data collection and modified throughout the data collection process to accommodate emerging themes. Data were collected in English and Pidgin English due to the diversity of local dialects in Nigeria. The population of MDR-TB patients enrolled on treatment in a facility is quite limited, and recruitment of recently discharged patients was challenging. Consequently our sample size is small, even for a qualitative study. However, we are confident that after the eleventh interview and second focus group discussion with this homogenous group, no new insights were forthcoming from participants.
Analysis
Data were transcribed word for word at the end of each session and the assistant researcher compared the contents with the audio recordings to verify the accuracy of the transcription. Data were analysed using the Framework approach for qualitative data analysis [
17]. Each transcript was read three times to identify and highlight key words or phrases across the data set. From the list of key words and phrases, we created an initial coding framework in NVIVO software [
18], and used this to label features of interest across all transcripts. Subsequently, coded sections were sorted into themes and sub-themes based on a hierarchy and constantly compared and contrasted to incorporate emerging features from the data set. Finally, data extracts from focus group discussions, in-depth interviews and key informant interviews were compared in a chart to understand the different dimensions of participants’ views and experiences for interpretation and reporting.
Discussion
To our knowledge, this is the first study that analyses insights of purposively selected patients and key-informants on hospitalized MDR TB care. As such, our findings provide a starting point for further research as they may only scratch the surface of the psycho-social impact of hospitalisation due to potential distortion by the indignity of disclosure and the tendency to present self as auspiciously as possible. The dislocation of patients from normal social networks and the detachment between healthcare providers and patients created the need for interdependence of patients on each other for emotional and physical support. The behaviour of healthcare providers, imposed by the fears of infection, contributed to discrimination of patients in the facility. However, healthcare providers seemed unaware of the stigma felt by patients in the facility as they rather attributed this to the attitude of non-MDR-TB healthcare providers towards patients when visiting other facilities for specialized services. There are potentially many other stories and striking incidents that were never discussed by participants.
This study illuminates how the fears of infection contributed to the stigma of MDR-TB, and in some cases discriminatory behaviour towards patients. At the community and family level, isolation of patients was either self-induced to protect family members or imposed by others. In both instances, stigma was rooted in ignorance of MDR-TB transmission mechanisms as the misperception of possible sexual and faecal-oral transmission of MDR-TB influenced attitude and behaviour. On interaction with healthcare providers, the ability to conceal their diagnosis was lost and patients felt increasingly vulnerable to stigmatization. Healthcare providers purposefully had minimal contact with patients, enforced mandatory use of facemasks in the facility, and patients took on the physical work of cleaning wards and fetching water because cleaning staff feared becoming infected; while most patients referred to these behaviours as ‘stigma’, they are deliberate actions to avoid MDR-TB patients and therefore more accurately described as discrimination. In a qualitative study in Nepal, TB patients also reported being stigmatized by healthcare providers [
19]. Furthermore, as HIV is also a stigmatizing disease [
9], HIV co-infected MDR-TB patients may have experienced more stigma from fellow patients and healthcare providers due to concerns about breach of confidentiality in the facility. The increased potential for occupational TB among healthcare providers who have prolonged contact with drug-resistant tuberculosis patients has been described [
20,
21]. More so, in areas with inadequate infrastructure and poor infection control, healthcare providers’ fears of infection may not be totally unwarranted or misplaced [
21]. However the impacts of these fears, often manifest in discrimination, on the emotional and physical wellbeing of patients could undermine MDR-TB control initiatives.
The substantial interdependence among patients to support each other emotionally developed naturally in our case but could be further exploited as an opportunity to improve outcomes. This finding resonates with a recent Medecins Sans Frontieres (MSF) project which identified the role of shared experiences via a social media blog in promoting social support for patients and ensuring adherence to MDR-TB treatment [
15]. However, social media blogging might be an insufficient strategy for providing substantial social support especially in low resource settings with mostly non-literate patients and poor internet facilities. Regardless, the prospect for social support should be distinguished from the discriminatory and physically stressful roles assumed by patients in our study, as they strived to mitigate the shortcomings in service delivery, helping each other to cope with strenuous physical activities, such as cleaning toilets, fetching water and washing beddings, at the expense of their physical wellbeing.
Prolonged isolation induced feelings of fear, anger, self-blame, depression and suicide in some of the patients. These feelings have been demonstrated in studies conducted on incarcerated individuals [
22,
23]. Indeed, it appears that our patients perceive their situation as imprisonment. However, second line anti-TB drugs may have the potential to cause similar psychological effects [
24]. In reality, the contribution of each to the patients’ mental status is difficult to distinguish. The dislocation of patients from normal social networks created the need for interdependence on others within a new social group. This new social identity with fellow patients was formed and reinforced through collective experience of illness enhancing understanding and similarity with new social relationships. Even though nurses are providers of emotional support for patients in hospitalized care [
25], mistrust arising from allegations of breaches in confidentiality and the stigma imposed by fears of infection limited their roles in the facility. Within the new social construct, the trust for confidentiality between younger and older females was perceived to be a mother and child relationship. In community-based care, trust for confidentiality may be vested in spouses [
9] and religious leaders especially in low resource settings. However, restriction of visits and the remote location of the facility undermined the role of these individuals in providing substantial emotional and psychological support for patients in hospitalized care.
Patients became actors to change their situation. They drew the attention of the authorities to perceived irregularities through organized group protests in form of hunger strike or refusal to take medications. This active and antagonizing role of patients is different from the passive role assumed by hospitalized cancer patients in Kenya [
26]. Possibly, MDR-TB patients, on symptomatic recovery did not view themselves as sick but as a community of individuals with similar needs that should be addressed. Communality of behaviour evolves naturally with the new value system of a stigmatized group. Transformation from unexpressive individuals in the larger society to highly vocal individuals occurs within their own social group and commonly, they express themselves in institutional terms [
27]. Although cancer patients are hospitalized, they are not isolated and as such receive emotional support from family members. Thus, their perception of services and their reaction to the treatment experience could be different from those of MDR-TB patients in hospitalized care.
With growing evidence that effective treatment expeditiously renders MDR-TB patients non-infectious [
28] there is little reason to keep people who are on treatment in hospital for months after they become culture negative. This is especially the case in Nigeria and other countries with limited specialist facilities that operate a waiting list for patients diagnosed with MDR-TB; those waiting to start treatment are highly infectious and can perpetuate community transmission of MDR-TB. In addition to this, the findings presented here show that patients suffer distress, discomfort and other psycho-social impacts when hospitalised for prolonged periods. Recent research from South Africa also suggests that hospitalisation increases treatment costs for MDR-TB [
29]. The current practice towards reduction in the duration of hospitalization with a target to discharge patients after culture conversion would reduce the psycho-social impacts of prolonged isolation and potentially reduce the risk of occupational TB from prolonged contact with MDR-TB patients. However, concerns about shortage of healthcare providers to sustain community/home-based care might prolong the proposed transition in Nigeria and similar settings. The identification and implementation of effective infection control interventions that reassure and protect health care providers and are acceptable to patients could dispel the fears of infections, reduce stigma and discrimination and possibly improve patients care. Building on shared experiences of illness and patients’ interdependence in this study, future research could focus on innovative patient-centred support mechanisms that are accessible and incorporate stigma reduction activities and emotional support for patients to improve treatment and support for MDR-TB patients. Further studies should explore the economic impacts of prolonged hospitalization.
Acknowledgements
We appreciate the contribution of staff of the University of Port Harcourt Teaching Hospital and the Rivers State Chest Clinic towards data collection. We thank Bliss Romanus Nwaiwu, for her role in facilitating data collection and transcription of audio-recorded data after field work.