Study setting

This study was nested within the “Adherence Support Coalition to End TB (ASCENT)” cluster-randomised trial in Ethiopia that aimed to evaluate the effectiveness of DATs with differentiated care to improve treatment outcomes for drug-sensitive TB. The trial was conducted in 78 health facilities in Addis Ababa and Oromia regions of Ethiopia [14]. Using the Ethiopian Government’s definition of urban and rural, five of the 78 health facilities were located in rural areas, all of which were in Oromia.

Two DATs were being evaluated; smart pillboxes and medication labels with an associated adherence platform. Smart pillboxes had a daily audio-visual reminder for the PWTB to take treatment and were equipped with sensors and connectivity features that recorded when the pillbox was opened on the platform. The medication label, affixed to the fixed-dose blister-packaged TB medication, displayed a unique code that was valid for one week. Patients sent a text with these codes every day to confirm their daily dose was taken, with the platform logging data and sending reminders if doses were missed. The adherence platform was used by health providers in health facilities to monitor patients’ adherence (using proxies of pillbox opening or SMS sent) in real time through a web-based dashboard. This platform enabled providers to track treatment progress, send reminders, and follow up with patients who miss doses, aiming to improve adherence and provide timely support [14].

Study design and participants

Participants for the qualitative study were enrolled from purposively selected health facilities implementing the smart pillbox or medication label interventions from both regions (S1 Appendix). In-depth qualitative interviews were conducted with four categories of participants: 1) people with drug-sensitive TB (12 facilities), 2) people with multi-drug resistant TB (six facilities), 3) healthcare workers (20 facilities), and 4) other key stakeholders (policymakers, community members, and implementing partners). People with multi-drug resistant TB (MDR-TB) were not part of the randomised trial, though they did receive the smart pillboxes, and followed up for treatment outcomes. They were included in the interviews to better understand their perspectives in terms of the acceptability of DATs and their usefulness in improving adherence in MDR-TB settings. Interviews were conducted between March 2022 and February 2023.

Specific inclusion criteria were used to select PWTB and HCWs including adherence profile, age, sex, geography, and length of DAT use. Key stakeholders were purposively selected from the national TB program, implementation partners and the community considering their engagement and experience in this study and the national TB care policy in Ethiopia. Separate semi-structured interview guides were developed for each group of participants by the research team (S2 Appendix). The interviews with PWTB and healthcare providers were conducted in private rooms at health facilities, while stakeholders were interviewed at their offices. Saturation was declared when no new information was obtained.

Interview process

In-depth interviews using semi-structured topic guides were conducted by two research assistants (male and female) who have BSc degrees in public health. The research assistants received one day of training on qualitative interview techniques, the consenting process, and the contextual factors that interviewers need to consider to ensure privacy and cultural appropriateness while greeting and talking to persons with TB. At the end of the training, the research assistants conducted practice interviews among themselves to practice and refine their skills. Feedback was provided by the trainers, ensuring that any misunderstandings or errors were promptly addressed. This was followed by several discussion sessions and debriefings with a senior qualitative researcher throughout the data collection process to fine-tune the topic guides and ensure data quality. PWTB were interviewed at health facilities during their refill visit based on their treatment follow-up appointments. PWTB were pre-informed by TB focal health professionals about the interview and invited to voluntarily participate in the study.

Data analysis

All interviews were digitally audio-recorded and were conducted in local languages (Amharic and Afaan Oromo), translated and transcribed verbatim to English by experienced researchers who were recruited for this purpose (PhD students in public health). The primary author evaluated all original audio recordings against transcripts to ensure the accuracy and completeness of translations. Analysis initially included familiarisation and discussion of the interviews. The researchers coded interview transcripts inductively in NVivo and agreed on the preliminary coding framework. To ensure inter-rater reliability, we used double coding of the first six interviews (ZS, LDG), followed by discussion and consensus meetings to resolve any discrepancies. The first author coded all the remaining transcripts, and a final coding framework was developed. A codebook was generated and shared with the entire research team and key themes and interpretations were identified.

The interview guides and the qualitative analysis were informed by key constructs of the Unified Theory of Acceptance and Use of Technology (UTAUT) framework to examine the acceptability and uptake of DATs. The UTAUT framework has been used in previous studies and posits that acceptance of a new technology can be determined by the effects of performance expectancy (usefulness), effort expectancy (easy to use), social influence and facilitating conditions [15–17].

Sequential numbers and the category of respondents were used to save transcripts and audio recordings. To protect the identity of respondents, the names of participants, institutions, and individuals mentioned during the interviews have not been reported. Research interviews and transcripts were digitally stored securely on a password-protected network drive to protect the confidentiality and privacy of participants. Research interviews and transcripts will be deleted after ten years according to the data retention policies specified in the research protocol.

Direct quotes extracted from interviews are identified using the category of the respondent and numbers to avoid accidental breach of anonymity (e.g., PWTB #1, HCW #2, stakeholder #3, etc.). All personal identifiers were removed from all records to ensure confidentiality.

Ethical consideration

Ethical clearance for the ASCENT study which included the trial and qualitative sub-studies was obtained from the London School of Hygiene & Tropical Medicine Ethics Committee, United Kingdom (19120–1) and the WHO Ethical Review Committee, Switzerland (0003297). In addition, ethical approval was granted from the two regional health bureaus in Ethiopia where the study was undertaken: Addis Ababa City Administration Health Bureau (A/A/H/B/1/978/227) and Oromia Regional Health Bureau (BEFO/HBTFH/1-16/10322). All participants provided both written informed consents to participate in the study and for the interviews to be recorded. Those who could not read and write provided a thumbprint as a form of signature.

Characteristics of participants

In total, 57 participants were interviewed across four categories: 25 people with drug-sensitive TB, eight people with MDR-TB, 20 healthcare workers, and 4 national stakeholders in TB care (S1 Appendix). Of the 25 participants with drug-sensitive TB, 13 were using medication labels, and 12 were using the smart pillbox. Nearly equal number of female and male participants (13/12) participated, with an average age of 34 years. People with MDR-TB (n = 8; 4 female and 4 male) were recruited from six health facilities in Addis Ababa and all were receiving the smart pillbox intervention. Healthcare workers (n = 20) were selected from health facilities who were implementing DATs (10 participants each from the label and smart pillbox arms). The majority of healthcare workers were female nurses (n = 15). The four national stakeholders were purposively selected from the Ministry of Health (MOH), implementing partners and TB survivors, considering their level of engagement in the Ethiopian TB program.

Facilitators and barriers to uptake of DATs

Overall, responses from all groups of participants ascertained that DATs were highly acceptable despite the existence of critical implementation barriers.

Informed by the UTAUT framework, the perspectives across all types of participants identified four broader themes that capture the facilitators for the uptake of DATs in the current Ethiopian context (Fig 1). First, DATs were considered very useful for both PWTB and HCWs (Performance Expectancy). Second, both PWTB and HCWs indicated that, practically DATs were easier to use than their pre-expectation before the intervention was started (Effort Expectancy). Third, PWTB revealed that the responses they received from important others about DATs were positive and encouraged them to use it (Social Influence). Fourth, there are promising policy opportunities and initiatives to increase the uptake of DATs in Ethiopia including strong TB care structures and digitization initiatives by MOH, and expansion of the telecom services (Facilitating Conditions). On the other hand, barriers to DAT uptake were constructed into three themes containing the critical factors limiting the implementation of DATs. These key uptake challenges include network issues, lack of inclusivity, and perceived stigma (Fig 1). The overall findings highlighted that there were no differences by age and sex in the context of experiences and perspectives of using DATs among the study respondents.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Facilitators and barriers of DATs uptake in Ethiopia (adapted from Thomas et.al 2021).

https://doi.org/10.1371/journal.pdig.0000667.g001

Facilitators of DATs uptake

Four themes were identified on the facilitators of DATs uptake.

Theme 1: Usefulness of DATs (facilitator)

Participants noted that DATs were useful in reducing the burden of daily health facility visits (saving time and reducing cost), improving treatment adherence and flexibility, and enhancing patient-provider relationships. From the healthcare perspective, DATs were useful in reducing their workload, work-related TB exposure, enabling them to provide differentiated responses (person-centred care), and increasing their motivation in providing care to PWTB.

Theme 2: Easy to use (facilitator)

The simplicity of using DATs was reflected repeatedly in responses from PWTB and HCWs. PWTB reported that taking their medication through the use of DATs did not pose major challenges.

It’s easy. Is not difficult for me, charging it also simple, it has no difficulties. (PWTB #18, Smart Pillbox Arm)

First, I it is very simple and helpful for me. (PWTB #18, Smart Pillbox Arm)

Similarly, HCWs revealed that practical implementation of DATs was easier than their expectations during the initial training they received.

DAT is good. Health professionals monitor patients from home while at the health center. Healthcare professionals can easily call patients and invite them or do follow-ups. (HCW #8).

I thought that it [DAT] was hard to implement. I even said an impossible thing is coming during the training for the first time. But gradually it becomes easy for me. I was considering it to be challenging and complicated for patients but now, it becomes an important thing (HCW #9).

The initial orientation delivered by health professionals played a vital role in making use of DATs simple to PWTB. Overall, PWTB were satisfied with the initial orientation provided by HCWs about how and why to use the DAT.

When I first arrived here [health facility], I met the health professional working at the TB clinic. she immediately explained to me the general description of the box that was completely new and that it was for testing. The explanation is enough for people like me [educated patients]. The information provided by the nurse was very helpful to me (PWTB #1; Smart Pillbox Arm).

It [the information provided during the orientation session] was enough. What she [the Nurse] told us was about texting, she adjusted my phone and I started texting. It was not very difficult. When you text, you immediately receives texts within seconds. so it was not very difficult to send or receive. it’s very clear (PWTB #5, Medication Label Arm)

While there were PWTB who did not fully understand the procedure during the initiation phase, they reported that healthcare providers were able to assist by providing continuous information to clarify issues.

It is easy to use but I was worried about the battery because I was not told about it, but I finished my treatment before the smart pillbox run out of charge. If they told me the color of the light will be changed when it runs out of charge, I will be cautious (PWTB #6, Smart Pillbox Arm).

Indeed, she has told me many things. But I don’t have clue how to send the message by the time [when the orientation was delivered]. I asked her [the nurse] how to send the message and she oriented me [the next day], then I started using [the label] the next day (PWTB #2, Medication Label Arm)

PWTB using the smart pillbox also identified that DAT made it easier to properly keep their medicine at home and workplace.

It [the smart pillbox] helps me put the medicine properly. It helps for the proper handling of the medicine. if you are using the medicine without the smart pillbox, you worry about where to put and the temperature. It has a potential to preserve the medicine. (PWTB #11, Smart Pillbox Arm)

It [the smart pillbox] protect the medicine from becoming dirty because you put it inside the box. (PWTB #24, Smart Pillbox Arm)

Theme 3: Positive feedback from important others (facilitator)

PWTB reflected that responses they received from important significant others (families, friends, colleagues and neighbors) were positive and encouraging for them to use DATs. While there were some reservations, most participants indicated that they were not afraid of telling other people about DATs and how they are useful in their TB treatment.

Honestly, I don’t have that complicated feeling to tell people [about DATs]. I freely tell to whosoever. People traditionally used to follow their treatment by going to the health center every day, but this medication label is quite different from that and I found it relevant. People appreciate the service for the reason that I don’t have to go to health center every day (PWTB #2, Medication Label Arm)

Theme 4: Policy opportunities and initiatives (facilitator)

Stakeholders identified that there are several initiatives and opportunities which can increase the uptake of DATs in the Ethiopian context. For instance, the recent expansion of the telecom infrastructure in Ethiopia following the new agreement between an international provider and the Ethiopian government is anticipated to increase access to mobile phone networks. In this study there is no cost to the PWTB for the SMS sent in the label arm.

Communication institutions like Safaricom are entering to our country and there will be cost minimization. So, through DAT can be scaled-up. The patient may be charged maximum 10 cents to send a message per day. (STK #4).

There is an expanding telecommunication through Safaricom. The CEO of Ethio-telecom is going to be an ambassador of TB program in the near future and this will be a good opportunity. Safaricom may reduce the cost of messaging. (STK #1).

Stakeholders also identified that existing TB care structures and coordination platforms within the ministry of health (MOH) is vital to expand DAT implementation.

There is a TB technical working group at national and regional level. If there is a focal person for DAT [within the technical groups] it will be more helpful [to integrate into national programs] (STK #1).

The new national healthcare digitization initiatives was also seen as major opportunities that enhance uptake of DAT implementation in Ethiopia. A participant mentioned that a taskforce was established at MOH to lead the digitisation initiatives which will be important to support the feasibility and scale up of DAT implementation.

The ministry of health also supports the digital technology implementation. To lead this digitisation, a task force in the ministry is established to be led by the state minister. So, bringing all the fragmented activities in the same pool is one opportunity for scaling up of technologies like DATs (STK #4).

Barriers to uptake of DATs

Responses across all groups of participants indicated that lack of adequate infrastructure including poor telecom signals and electricity supply are the major challenges for DATs implementation. Providers and stakeholders were also concerned that DATs may not be used by all PWTB. For example, PWTB with low literacy, those who did not have phones, and people who are homeless and prisoners may not have the opportunity to use DATs. Stigma was also indicated as a potential barrier to the implementation of DATs. HCWs have reflected their concern (perceived) that use of DATs may lead to unintended disclosure of TB status (e.g, during use of DATs at workplace or home visits by health workers) which in turn can result in the fear of stigma among PWTB. However, responses from PWTB did not support this concern (either perceived or experienced stigma) related to use of DATs.

Theme 1: Network and electricity issues (barrier)

As both the smart pillbox and medication label adherence technologies are built to be linked with phone devices, poor telecommunication network and frequent power outages in Ethiopia posed a wide range of challenges in the effective utilisation of DATs. As a consequence of poor network or power outages, patients identified several types of challenges they were experiencing in the use of DATs. For example, patients experienced a delay in receiving text reminders or alarms because of network or power issues and their time to take the medication may become overdue.

Sometimes it is quiet, does not give alert even if time is up, I always watch my phone for time, so I take my medicine timely even if the smart pillbox does not give any alert. [PWTB #6, Smart Pillbox Arm)

Sometimes there is a problem of network and light in our area. I can take the medicine without the alarm by checking the time from my watch. When I check its alarm [alarm of the smart poll box], the time is up [but not alert], I come to know it is the problem of network. I came here [health facility] and reported this for them [health professionals]. (PWTB #11, Smart Pillbox)

In other cases, a repeated or random alarm from the smart pillbox prompted multiple opening of the device.

One day, after taking the medicine, it started alarming, so I opened it more than one time that day in order not to be disturbed. But it didn’t stop and shout again and I opened it again. This happened only one day. (PWTB #8, Smart Pillbox Arm)

Other PWTB also experienced a random alarm from the box before or after the scheduled time of taking their medication.

For some time, it alarmed at 7:00 am which is before the adjusted time [the schedule was at 10:00am]. This was not right. If it was reminding me at 9:40 or 10 minutes later, it could be useful for me. But in this case, it is only disturbing. (PWTB #9, Smart Pillbox Arm)

Due to network issues, PWTB in the label arm reported delays when receiving a confirmation message after they took their medications and submitted a text to the platform. As such, PWTB in the medication label arm were sometimes forced to send multiple messages when they were unable to receive immediate confirmation. This has created emotional discomfort and annoyance in some PWTB. However, during their next health facility appointment, PWTB were able to discuss the issue with HCWs who became aware of the problem.

For I didn’t receive a feedback message, I kept frequently sending of a message by assuming the message did not reach them [health professionals]. While I came here [in the health facility] on the 3rd month, I asked them why sometimes their message not sent back to me? (PWTB #22, Medication Label Arm)

There were issues I faced sometimes. I faced things like when I text, I wouldn’t get a reply. Sister [name of focal nurse deleted) said, “it was a network problem”. (PWTB #5, Medication Label Arm)

There are PWTB who received multiple follow up text messages despite medication had been taken on time. When there were network issues, the platform is unlikely to receive the notification that was sent when the smart pillbox was opened or when patients sent the “code” in the medication label arm. As such, PWTB often received reminder text messages notifying them that they had forgotten to take their medication. This led to frustration in PWTB because they feared the health professional may consider them to be non-adherent. Some even decided to come to the health facility before their appointments to explain the situation to the health professionals.

Sometimes I receive a message that says, “you did not take your medicine today”. At that time, I came here [to the health facility] and explained that I have taken the medicine, but my house has a network problem. This inconvenience happens because of network, not because I didn’t take it. (PWTB #11, Smart Pillbox)

There are four or five messages in my phone saying "please take your medicine today” while I already have taken it. I don’t know if they can fix these things with the box. But I did not miss it. (PWTB #1, Smart Pillbox Arm)

Theme 2: Lack of inclusivity (barrier)

As a barrier, HCWs and stakeholders identified that DAT lacks inclusion of all categories of PWTB. For example, those with no literacy, did not have phones, homeless and those in prison may not have the opportunity to use DATs.

When I was visiting [health facilities], patients who are using directly observed therapy (DOT) request to join DATs intervention. But I didn’t face DAT user requesting to use DOT. To make it fair, people who request the technology must have an access (STK #1).

There are patients who cannot text, but we talk to their relatives at home to show them how to text. Most patients get used to it after a few days. Those who are unable, they have to visit the health facility (HCW #17).

Patients who don’t have accommodation can’t stay for a year, so I don’t include them in the DAT program. However, they want to be part of the program. I always wonder how to handle such cases. (HCW #2).

It is good to involve more partners, to mobilize more resource, to think more about scale up and inclusivity, to try it in other regions, and on different settings. It is important to try it in prison, to try it in pastoral communities, and try it in different settings (STK #3).

Theme 3: Fear of stigma due to unintended disclosure of TB status while using DATs (barrier)

Our study identified mixed perspectives about stigma related to the use of DATs. While stigma was not a concern for PWTB, HCWs reflected their fears (perceived stigma).

I am not afraid of what people might say about it [smart pillbox]. I am not afraid, usually I hold it in a plastic bag and sometimes I hold it as it is, but I do not mind what others think about it even if they know my TB status. It’s not a concern for me. (PWTB #6, Smart Pillbox Arm)

Mostly people perceive it [the smart pillbox] as a lunch box. But if they ask me about it, I tell them that it is my medicine box for my treatment follow-up. When I come here [to the health facility] to receive the medicine, individuals see the smart pillbox and ask me “are you travelling with your lunch box?” But I tell them that it is a smart pillbox which helps me to follow-up the medication or treatment. (PWTB #9, Smart Pillbox Arm)

According to HCWs’ accounts, there were PWTB who were not comfortable when they were being home visited due to fear of stigma.

Patients do not want home visits by healthcare workers because their families may not be unaware of their TB status (HCW #19)

Implications

The findings of the study suggest that DATs are highly acceptable among PWTB, HCWs, and policy stakeholders in Ethiopia. DATs were found to improve treatment flexibility, reduce the burden of frequent health facility visits, and enhance patient-provider relationships. However, key barriers to the implementation of DATs include poor infrastructure, lack of inclusivity for all PWTB, and concerns about stigma. Despite these challenges, several opportunities and initiatives, including the presence of TB care coordination platforms at national and regional levels, health digitization initiatives by the MOH, and the MOH’s commitment to funding new TB programs, could support the feasible implementation, adaptation, and scale-up of DATs in the Ethiopian context. Future research should focus on developing strategies to overcome these barriers, particularly by addressing infrastructure challenges and ensuring the inclusivity of DATs for all PWTB, to maximize their effectiveness and sustainability in different settings.

Strengths and limitations

A strength of our study includes the capturing of comprehensive perspectives of participants by engaging diverse groups of respondents including people with drug-sensitive TB, people with multi-drug resistant TB, healthcare workers and policy stakeholders in Ethiopia. Another strength of this study is that it adds more insights to the emerging body of knowledge regarding the role of acceptability in digital adherence technologies that can be used to help improve adherence to TB treatment in the future. However, findings in this study need to be interpreted in the context of some key limitations. As with all qualitative research, the findings cannot be considered representative of the broader views of other PWTB in Ethiopia. For instance, some groups of PWTB were excluded from the trial including homeless people and people with extrapulmonary TB. All health facilities were in urban areas, similar to the trial setting, so we cannot generalise to experiences of PWTB and HCWs from rural settings. Social desirability may have influenced PWTB and HCWs while reporting on the usefulness of DATs. In addition, only a small number of respondents were interviewed among key policy stakeholders which may limit our ability to highlight the national opportunities to integrate and scale-up DATs in the Ethiopian context. A further limitation of this study is the absence of input from relatives of persons with TB, who could have provided additional insights.

Conclusions

In general, findings from this qualitative study have provided a rich set of perspectives relevant to policymakers in identifying the facilitators and barriers to the uptake of DATs in Ethiopia. The overall finding suggests that DATs are highly acceptable among PWTB, HCWs and policy stakeholders. PWTB and HCWs identified several facilitating factors that increased the uptake of DATs including their usefulness and simplicity of use. On the other hand, critical barriers that limit the uptake of DATs were identified as poor infrastructure, lack of inclusiveness, and fear of stigma. Despite those barriers, key policy stakeholders believe there are several opportunities and initiatives that could support the feasible implementation, adaptation, and scale-up of DATs in the Ethiopian context, including the MOH’s commitment to expanding health digitization efforts and funding new TB programs.