Patients' perceptions and experiences of directly observed therapy for TB

Tuberculosis (TB) is caused by a bacterium (Mycobacterium tuberculosis). When people with active TB in their lungs or throat cough, sneeze or spit, they can propel TB germs into the air which can be breathed in by others. A person need only inhale a single bacterium to become infected. TB usually affects the lungs, causing consolidation and tissue destruction, but can spread from the lungs to other parts of the body such as the spine, other organs and lymph nodes, causing serious infection. People with active TB in their lungs can infect 5–15 other people through close contact over the course of a year. Without proper treatment, 45% of people with active TB who are HIV negative and nearly all people with active TB who are HIV positive will die (World Health Organization (WHO), 2021).

Drug-sensitive TB is treated with a combination of antibiotics, usually 6 months of rifampicin and isoniazid augmented with ethambutol and pyrazinamide for the first 2 months. Severe cavitary pulmonary disease may need a longer course of treatment (9–12 months) and nearly all patients will experience undesirable side effects ranging from fatigue, nausea and vomiting through to rashes and liver toxicity. Drug-resistant strains of TB, however, require longer courses of treatment of between 9 and 24 months and include the use of second-line antibiotics with profound side-effects such as ototoxicity, gastrointestinal and central nervous system disturbances, dermatological changes and arthralgia.

Adherence to TB treatment is crucial to achieve a cure, minimise transmission to others and prevent drug resistance (National Institute for Health and Care Excellence (NICE), 2016). People's ability to adhere to TB treatment is dependent on a range of complex socioeconomic, structural and personal factors (Craig and Zumla, 2015). In 2020, 12.7% of people notified with TB had a social risk factor (SRF) such as homelessness, drug and alcohol addictions, mental health problems or unemployment. Some 22% of UK-born cases had at least one SRF recorded, compared with 9.1% of non-UK-born cases. SRFs are associated with lower treatment completion rates and people with a combination of TB and an SRF are more likely to die or be lost to follow-up (UK Health Security Agency (UKHSA), 2021).

Directly observed therapy (DOT) is a key tool in promoting adherence to TB treatment (WHO, 1999). It requires the patient to swallow their TB medication in front of a nominated supervisor—usually on a daily basis. Although some countries use legislation to enforce DOT, within the UK, TB teams assess their patients' ability to adhere to treatment and, where a risk of non-adherence is identified (including but not limited to those with SRFs), national guidance advocates DOT rather than requiring it by law (NICE, 2016; Public Health England (PHE), 2019).

Anderson et al (2014) highlighted variations in the provision and uptake of DOT across England, with fewer than half those who were eligible receiving DOT. A systematic review by Wright et al (2015) suggested that DOT does not provide a panacea to poor adherence; instead, greater emphasis is needed to address the wider barriers to adherence such as mental health problems, addictions and housing issues. Greater provision of resources to provide support in locations and at times convenient to patients is equally needed, alongside evaluation of strategies that motivate patients and staff.

It is known that the uptake of DOT can be affected by:

• Inequitable service provision (PHE, 2015) and
• Patients refusing or defaulting from DOT because they feel supervision is unnecessary and/or could adversely affect their social schedule (Gupta et al, 2004).

There remains a paucity of research from high income countries with a low TB incidence (including the UK), to distinguish which elements of DOT patients consider supportive or unsupportive. Although DOT remains an integral component of TB control, policymakers, commissioners and service providers share responsibility in understanding the patient experience and using this knowledge to develop patient-centred support.

The aim of this study was to understand patients' perceptions and experiences of DOT used in TB treatment within the UK. The inclusion of participants with SRFs from high and low TB incidence areas, as well as participants placed onto DOT as a result of TB drug resistance, altered mental or physical capacity and prior non-adherence, would suggest that this research is of particular relevance to TB service commissioners and service providers from both these incidence areas.

Aims

The aim of the study was to understand patients' perceptions and experiences of DOT in tuberculosis treatment, with the following objectives, to:

• Explore what it felt like to receive DOT
• Explore patient descriptions of how DOT impacts on their lives
• Understand how patients fit DOT into their everyday routines/lives, or not
• Identify any perceived barriers to patients undergoing DOT and how these were overcome, if at all.

Study population and methods

Study design

Qualitative research methods add to our understanding of complex social worlds, enabling service users to contribute their views and experiences to the evaluation of healthcare interventions (Rapport et al, 2013).

This study was undertaken from a ‘subtle realist’ position (McLaughlin, 2012), enabling the authors to explore participants' complex social worlds, their views and experiences of DOT. Alternative epistemological approaches such as positivism, could have been used but the positivist distinction between science and personal experience, fact and value judgement risked structural limitation to the analysis (Ritchie et al, 2013).

Audio-recorded semi-structured interviews were analysed to explore the four specific objectives. Data were arranged using a framework approach, which enabled the formation of conceptual themes (Table 1). Themes are discussed in relation to published literature from which conclusions and recommendations for service provision are drawn.

The study was undertaken in part-fulfilment of a Masters of Research (MRes) degree funded by the National Institute for Health and Care Research. Ethical and governance approval was obtained from the West Midlands Research Ethics Committee and Health Research Authority prior to study commencement. Due to the vulnerable circumstances of the participants (underserved populations), ethical and substantive validations (Angen, 2000) were additionally considered. Owing to a limited number of published studies into the use of DOT within the UK, understanding patients' experience was considered imperative.

Eligibility, sampling and recruitment

A purposive sample of up to 10 participants from high and low TB-incidence areas was sought. All TB service leads from the south, south west and south east of England were contacted and asked to apply eligibility criteria and to offer potential participants a study pack, which included participant information and consent forms. Participants were not known to, nor were they recruited from, the first author's (DT) clinical catchment. Participants were invited to speak to their TB service lead if they had questions or wished to participate in the study. Written informed consent to be interviewed and for their data to be used within the results was obtained for all participants before their interview.

Data collection

Face-to-face, semi-structured interviews were audio-recorded and transcribed verbatim. An interview guide developed by DT was used to maintain consistency between interviews. All interviews were conducted at a location of the participant's choice. Participants were invited to bring a chaperone, and none chose to do so. Interviews lasted between 30 minutes and 1 hour and were not repeated.

Data analysis

Data analysis began after the first interview. Data were organised using NVivo 11 Pro software and analysed using Clarke and Braun's (2013) thematic analysis. Lincoln and Guba's (1985) quality indicators were used throughout the research process, The second author (RS) second coded the first interview transcript to check the coding framework and provided peer review of developing codes and themes throughout analysis.

The participants

Eleven potential participants agreed to be contacted. One withdrew due to time constraints and two felt their mental states were too fragile for them to take part. Two female and six male participants, median age 46, from six ethnicities (Table 2) were interviewed. Participants resided in different inner-city, mid-size and large towns in southern England and had a range of employment histories. Participants presented with one or more risk factors for non-adherence to treatment (Table 3) and experienced a range of supervision styles (Table 4). All participants had either completed their treatment under DOT or were expecting to do so within 6 months.

Results and discussion

Theme 1. Adherence versus concordance

Participants reported a lack of concordance in their supervision. This lack of concordance experienced by some patients meant they did not feel their views or treatment wishes were taken into consideration. Rather, the nurse or doctor told them what was going to happen. Initially, some patients felt they were not trusted (were spoken to rather than included in the decision-making process). The lack of concordance may have contributed to eventual non-adherence for some patients. This was counter-balanced by perceived benefits such as practical support, social connection and, for a few, a degree of empowerment, where DOT was viewed as a vehicle to retaking control of their health.

‘At first I felt as if I'm not being trusted. I was a bit upset but, on the other hand, I said okay I am in the right place, I get support because these people come in, even in the flat. I feel that I am being looked after.’

Participant 6

‘I think they sussed I wasn't taking the medication and that I was drinking. Basically, [staff name] and the other nurse sat me down—I felt like a naughty schoolboy [laughs]. I appreciate the fact that they realised I had a problem and they were helping me, and the supervision helped a lot. It put me back on track and it helped me focus on getting rid of it [TB].’

Participant 1

Healthcare workers' failure to listen during interactions was described by a few of the participants. For example, following a police-supported readmission to hospital (having run away), Participant 7 described the healthcare staffs' ongoing failure to listen to him.

‘That was another thing that annoyed me about the doctors and nurses when they came around. “How are we feeling today and are you in any pain?” I wasn't in any pain the whole time I was there and I didn't feel ill the whole time I was there. They [the medical team] didn't listen.’

Participant 7

Similarly, Participant 8 expressed frustration at perceived wasted time during outpatient appointments where he had to re-explain his diagnosis and treatment progress to new doctors at each appointment.

‘Every time I come up it's been the same old spiel, same old thing again and again. I said why can't we just do this on the phone?’

Participant 8

Reasons for these failures may lie with ritualistic (but well intended) questions within ward rounds and failures of staff continuity during outpatient appointments. Nurse-led TB clinics and the use of telemedicine could help to negate this problem and provide greater continuity.

Failure to listen and respond to lay knowledge has been linked to patients with TB defaulting from their treatment (Zwarenstein et al, 1998). Despite this, Bayer and Dupuis (1995) suggested that patients experience difficulty in questioning the way TB treatment is organised. These findings are supported by Sagbakken et al (2012) whose Norwegian research revealed there was little room for patients to negotiate individually adjusted treatment. Instead, patients' perceived that they were subject to surveillance and strict control, which led to feelings of humiliation. Within their study into patients' experiences of DOT in Iran, Behzadifar et al (2015) reported healthcare workers' failure to trust patient knowledge and a perceived lack of time and resources to facilitate patient inclusion in care planning.

A ‘one-size-fits-all’ approach to DOT does not promote patient concordance. DOT has been described as invasive of privacy, time and dignity (Reichmann and Hopkins-Tanne, 2002). Implemented without achieving concordance, it can impede patients' social and employment schedules, risking disengagement from treatment. Achieving concordance with DOT requires expert use of patient inclusion, behavioural change strategy, ongoing support and responsible use of motivators to overcome barriers. Fundamentally, adequate resources are required to facilitate reasonable flexibility in meeting patients' needs rather than expecting them to adhere to limitations within service provision.

Theme 2. Engagement and disengagement with DOT

An estimated 25%–50% of patients undertaking prescribed TB treatment are non-adherent (Mekonnen and Azagew, 2018). According to the Centre for Pharmacy Postgraduate Education (2015), complex facilitators of non-adherence fall into two overlapping categories: intentional, where patients choose not to follow instructions because of beliefs, preferences or practicalities, and unintentional, where patients want to adhere but are prevented from doing so. For example, misunderstanding instructions, forgetting instructions or encountering barriers.

A few participants were placed on DOT following intentional non-adherence to self-administered therapy (SAT). Participant 4, for example, claimed full adherence to SAT until a home visit by an experienced TB specialist nurse revealed several weeks of untaken medicine. Re-starting his treatment from scratch using DOT achieved engagement but failed to fully prevent intermittent (intentional) non-adherence.

‘Yes, sometimes, I still miss one day, I no take medicine and drink beers. I am missing them 1, 2, 3 times, or maybe more, 4 times I think.’

Participant 4

Participants made daily behavioural health choices influenced by their employment, domestic situation and personal experience of TB. Similar to Participant 4, Participant 1 (also unemployed, single and had not experienced a lengthy TB-related hospital admission) remained (intentionally) non-adherent.

‘Occasionally, I didn't [attend for DOT] because maybe I had a job interview or like when my mum was ill, so I missed a couple but it was always rearranged, she [healthcare worker] always checked on me.’

Participant 1

In contrast, Participant 2 negotiated a goal of seeing his young child (living overseas) and used his experience of TB as motivation to endure daily disruption to his work and debilitating treatment side-effects.

‘I don't want to go back to the hospital and stay like for months. I don't want coughing blood again. I no want to stay away from my family again. This time I keep going, pushing myself to take the tablets.’

Participant 2

One participant utilised her knowledge of behavioural change gained through her qualification and experience of working as a nurse.

‘So I had to decide, it's medication or it's my life. Then I had to be strong, I had to accept it. I should not cheat, yeah, because cheating is going to mislead the medical team on what is going on, as if the medication is not responding.’

Participant 6

Arlinghaus and Johnston (2018) suggested that knowledge by itself can be insufficient to change behaviour and healthcare workers need to educate patients about why behavioural changes are needed and how to achieve tasks through the promotion of skills and confidence.

Paterick et al (2017) highlighted how a sense of self-efficacy, defined by Bandura (1977) as a person's belief in their ability to accomplish tasks or succeed in specific situations, can help patients approach personal goals, tasks and challenges. Wagner et al (2001) reported the clinical benefits of promoting self-efficacy. In an era of rising health costs and where a public health need for adherence exists healthcare workers need to reinforce behavioural change education with ongoing patient support. This point was emphasised by Atkins et al (2011) in their South African study, which reported that additional training for community TB health workers, augmented by increased patient education, failed to improve treatment outcomes. They hypothesised that the failure was related to the intensity of patient support, which diminishes throughout the longer ‘continuation phase’ of TB treatment.

Theme 3. Threats to, and the opportunity for, social connection

DOT: a mechanism for disclosure

Within their systematic review of cultural variations in knowledge, attitudes and health responses to stigma, Chang and Cataldo (2014) reported that negative attitudes towards TB are influenced by beliefs, misunderstanding of TB transmission and fear of contagion and stigma. Discrimination, whether actual or perceived, is an important concern for patients, affecting mental health, social connection and treatment outcome (Cremers et al, 2015). Fuelled by perceptions of discrimination, a few participants viewed DOT as a mechanism for disclosure, which led to secretive behaviour.

‘Some people they don't understand things. It makes moves inside their brains. You know what I mean? This is why I want to keep it [TB and need for DOT] away from my work. This is my business.’

Participant 2

‘I was really sick, like, it was really hard for me go to work and I couldn't tell them what was the reason that I can't work for them. They really no understand.’

Participant 4

‘Yeah I felt that I was being discriminated at work as well. I know I'm not at risk of infecting somebody, at the same time I don't want to destroy my relationships, I'm a nurse. I don't want people to feel unsafe while they are with me.’

Participant 6

Participant 6 additionally perceived discrimination from within her family.

‘When my son came in I looked like a ghost. He did not want to see me. He ran away, disappeared. I was having the hope that he was coming to see me to help me. He wouldn't even stand here; he would be standing by the door because he did not understand about the mode of transfer. He wouldn't eat the food that I'm making. He didn't use the toilet, he just disappeared. Then I thought okay this is discrimination [because] of TB.’

Participant 6

Participants displayed resilience however; Participant 2 completed his DOT and Participant 6 attempted to educate her son about TB.

‘I sit down with him and explain to him the spread and that I am not at risk. He didn't much understand. Family members have to get involved and to get education about this thing because, if they don't know, they think that they are going to be infected too. So to be neglected by your own family is very difficult.’

Participant 6

In contrast, Participant 3, despite perceiving that he had lost his job and accommodation through his employer's fear of TB, still viewed DOT as a bridge to social connection.

‘I came in the middle of the night and they [employers/landlord] had locked the door. I am feeling hungry. It was difficult, I did not [inaudible] know what to do, I don't eat and can't claim the money through.’

Participant 3

Following an emergency social housing placement, Patient described how DOT visits enabled donations of food, basic cooking equipment and furniture from his TB team, as well as help with benefits applications.

‘I can't believe [healthcare worker's name] do lots for me, she come to my room, looks after me, tell me what to do, she say I can talk to the council and bring letters. The nurse cares this much, you know.’

Participant 3

Understanding cultural drivers of discrimination can help staff balance the appropriateness of support activities against the risk of disease disclosure (Macq et al, 2008). Data revealed that a Nepalese participant left his job rather than disclose his TB status and need for DOT. Providing a possible explanation, Baral et al's (2007) research into TB discrimination within Nepal identified a fear of perceived infectivity, fear of association with low-caste poverty, and a perception that TB is somehow linked to disreputable behaviour and is even considered by some to be a divine punishment. Mindful of this, Participant 4's actions may have been driven by a perception that DOT might bring about social discredit to himself and his family.

Threats to social schedule

DOT impacted participants' social schedules and employed participants found the time demands most onerous. In some cases this led to feelings of being trapped.

‘You don't feel free, you feel like you're inside a prison. You want to go out this prison but you can't. You have to take medicines. Do the same things every day (leaving work every morning for 2 hours to receive DOT injections). Everything in your mind changed, everything around you change. Your life change, it's complicated.’

Participant 2

‘I can do the medicine but it's like a prison, the nurse says I can't stop.’

Participant 3

In contrast, unemployed participants adapted to DOT more smoothly. Participant 1, for example, despite finding DOT laborious, derived comfort from its rigid routine.

‘People might be making me do stuff, but it's for my own good. I appreciate the effort so, yeah, I didn't mind it (DOT). I appreciated that they were making the effort, that I was doing what I was meant to be doing, which in a funny way was quite comforting, really. It was nice to know there was a set routine and I was getting forced to do it, so I preferred that regime to the other (SAT).’

Participant 1

Valuing social support

Participants gained social connection and psychological support through DOT, and some found motivation to engage in personal hygiene and housework prior to the healthcare worker's visit.

‘That's one of the things, if it hadn't been for people like [healthcare worker's name], it would have been a hell of a lot worse. I'd say 100% worse. If you had a query or was feeling peed off about something, you could just phone her up and she would sort you out or she would say “I'll come out”.’

Participant 8

‘Every Friday [healthcare worker's name] comes around and tops up my [medicines] box. Which I can do myself but it's nice to have a chat with someone who knows what they are doing.’

Participant 7

Social support has been shown to positively influence people's ability to adjust to illness and is important in recovery and maintenance of wellbeing (Helgeson and Cohen, 1996). Tian et al's (2014) systematic review found no convincing evidence that clinic-based DOT was any more effective than self-administration in terms of providing ongoing patient support. Certainly, in the cases of Participant 1 and Participant 3, clinic-based DOT failed to prevent intermittent non-adherence. Conversely, Craig and Zumla (2015) demonstrated positive correlations between adherence, community-based DOT and social support, demonstrating that patients appear more able to sustain a course of treatment when their environment is kept stable.

Theme 4. Mental health: an overlooked issue

Several participants experienced symbiotic and multidirectional mental health problems prior to and as a result of their TB diagnosis. This is unsurprising given the higher prevalence of mental health issues in people with TB compared with the healthy adult population (Doherty et al, 2013). Perceiving isolation and an inability to escape DOT left participants feeling trapped. This fuelled negative associations and, for some, destructive behaviours. Despite this, only Participant 2 reported being offered access to psychological assessment/therapy as an inpatient, which he considered helpful.

‘In the hospital I asked to speak to a psychologist. You don't feel free, you feel like you're inside a prison. You want to go out this prison but you can't.’

Participant 2

This contrasted with his later experience as an outpatient where psychological support was not offered and, instead, he was advised to rely on inner strength.

‘I speak with the doctors about the feelings but they said they can do nothing. Just carry on and take the tablets. They say they can't do anything, you need to be strong and carry on.’

Participant 2

Providing insight into a consequence of overlooking mental health, Participant 1 felt his medical team's failure in recognising his deteriorating mental state contributed to a downward spiral of untreated depression, alcoholism and non-adherence.

‘At the time, they didn't realise the emotional psychological things I was going through because of the TB, because of losing my job, because of losing my accommodation. There should have been maybe some help in dealing with that, but then I wouldn't have had to have gone through everything [treatment] twice. I think that maybe that's a thing that could be looked at, you know, to have some psychological help, a counsellor or somebody because it's a big thing, you know. That's a big thing to get TB.’

Participant 1

Comorbid mental illness presents a significant challenge to TB management. Both illnesses share common risk factors, including social disconnection, alcohol/substance abuse, smoking and homelessness; all of which can lead to additional comorbidity. People living with TB and mental illness are at greater risk of poor health-promoting behaviour, poorer adherence and have worse treatment outcomes (Mason et al, 2016). Economic modelling shows that for every US$1 invested in treating mental disorders, up to US$5.70 is saved in economic cost and health returns (Chisholm et al, 2016).

Published examples of successful ways in which mental health among TB patients is being addressed remain scarce. But, internationally, these include a psychosocial support group for people with multidrug-resistant TB in Peru (Sweetland et al, 2014), a prospective controlled study in India that concluded that TB patients receiving individual psychotherapy were significantly more likely to complete TB treatment (Janmeja et al, 2005), and rural support groups for people with TB in Ethiopia (Demissie et al, 2003). UK TB guidance has begun to highlight the need for psychosocial management (Royal College of Nursing, 2019; PHE, 2019), which offers widespread opportunities for patient-centred initiatives to be trailed and evaluated.

Limitations

Data saturation where data was being repeated by different participants was not encountered. This suggests a future larger study may be useful in uncovering new themes.

Conclusion

Participants' level of acceptance of DOT for TB was in accordance with their socio-cultural identity, mental health, employment status and social connection. Although a lack of concordance left participants feeling they were not trusted, some showed understanding in appreciating DOT's benefits.

Employed participants perceived a risk that participating in DOT may disclose their TB status to others. Fuelled by stigma, this led to secretive behaviour and internalised discrimination. Understanding individual and cultural drivers of discrimination could help healthcare workers balance the appropriateness of DOT interventions against the risk of disclosure and social discredit for the patient.

Face-to-face community-based consultations with experienced TB key workers provided an antidote to participants' social isolation and encouraged motivation. Weekly home visits were particularly valued as a gateway for psychosocial connection and support.

Adherence remained at the mercy of circumstantial drivers such as a fear of discrimination and loss of control, insecure food and shelter, comorbid mental illness and addictions. The overwhelming influence of coexisting barriers implies it is unrealistic to place sole responsibility for adhering to treatment on either the patient or their treatment supervisor. Effective TB control, particularly among underserved populations, will require multi-agency ownership of responsibility and collaboration.

Implications for practice

Concordance in TB supervision can be achieved through the use of flexible supervision offered at the start and throughout treatment. This should include but not be limited to offering DOT. Whereas multi-agency collaboration promoting concordance needs to be endorsed nationally, commissioned and evaluated, opportunities also exist for TB teams to increase patient support options themselves. For example, promoting concordance during the initial patient discussion could include discussing a range of support options promoting informed decisions. Telemedicine may offer greater flexibility and improved access to appointments. Using telephone appointments during COVID-19 lockdowns, and continuing to date the first author's local TB clinic's non-attendance rate has been cut by 74%; patients no longer need to take several hours off of work to attend or incur the hassle and expense of travelling to the hospital for routine appointments. Nurse-led clinics have additionally freed up respiratory physician time and enabled longer appointments and an ability to explore the individual patient's barriers to concordance.

KEY POINTS

• Participants in this study accepted directly observed therapy (DOT) for tuberculosis (TB) according to their socio-cultural identity, mental health, employment status and social connection
• Face-to-face community-based consultations with experienced TB key workers provided an antidote to participants' social isolation and encouraged motivation. Weekly home visits were particularly valued as a gateway for psychosocial connection and support
• Concordance with TB supervision can be achieved using flexible supervision offered at the start and throughout treatment. This should include but not be limited to offering DOT
• Effective TB control, particularly among underserved populations, requires multi-agency ownership of responsibility and collaboration

CPD reflective questions

• What opportunities exist for TB teams to increase patient support options?
• How would you feel if a healthcare worker told you they were going to watch you swallow your TB medicines daily? Considering this, how would you support a patient in these circumstances?
• How do you think DOT might impact on a patient's work and social schedule?
• Is it realistic for a TB nurse to be able to gain a patient's concordance to a long course of treatment or does there need to be a multidisciplinary or multi-organisational approach?