Study design and population

We conducted an exploratory qualitative study nested within the EAPOC VL trial (ClinicalTrials.gov identifier: NCT05048472; https://clinicaltrials.gov/study/NCT05048472) from January 2023 to January 2024. The study employed a combined longitudinal and cross-sectional design, which allowed follow-up of participants over time while also recruiting new participants at later time points.

Participants in the study included laboratory technicians, nurses, doctors, clinical officers, data clerks, pharmacists, peer educators, and counsellors. Recruitment of participants began on 11 February 2023 and ended on 28 January 2024. Data were collected through semi-structured, in-depth interviews with HCWs who had direct contact with clients in care and treatment centres (CTCs), as well as semi-structured observations.

Study area and setting

We conducted this study in the Kilimanjaro region in the north of Tanzania and the Dar es Salaam region on the east coast of Tanzania. The study was conducted at five CTCs across the country, including a Referral Hospital (Mawenzi in Kilimanjaro), two district hospitals (Sinza in Dar es Salaam and Hai in Kilimanjaro), and two health centres (Pasua and Majengo, both in Kilimanjaro). Two sites (Sinza and Hai) performed nPOC HIV VL tests at on-site hospital laboratories with no transport required, while three sites (Mawenzi, Pasua, and Majengo) transported samples by boda-boda (motorcycle courier) to the off-site Kilimanjaro Clinical Research Laboratory for testing. All testing was conducted using GeneXpert machines, which can process up to four samples at a time and provide results in approximately 90 minutes. All interviews were conducted in private rooms within the clinics to ensure confidentiality and no third parties were present during interviews.

Sampling of participants

For the interviews, the HCWs were selected purposively from those involved in implementing nPOC HIV VL monitoring at the intervention sites; that is, HCWs (1) attending PLWHIV at CTCs in one of the five selected health facilities, (2) who were involved in VL monitoring, (3) who were part of the EAPOC-VL study, (4) who were providing healthcare treatment and advice based on formal training and experience, such as doctors, nurses, and laboratory technicians and (5) who were not formally trained in providing healthcare, like data managers, peer educators and counsellors. We aimed for maximum variation, including a broad range of the professional cadre involved in implementing nPOC HIV VL monitoring. The chief physician of the selected intervention health facilities assisted in identifying locally practising HCWs who were currently or were previously involved in HIV client diagnostics and management and staff involved in the EAPOC-VL study. We initially planned to conduct an average of 25 interviews (five per site) across T0, T1, and T2, for a total of 75 interviews, based on the saturation principle [15,16]. All HCWs invited agreed to participate and no participant withdrew from the study. However, not all HCWs enrolled at baseline were available for follow-up interviews at T1 and T2 due to annual leave, staff transfers, and changing clinical responsibilities. To ensure sufficient data, new participants were recruited consistently with the study’s combined longitudinal and cross-sectional design.

Theoretical framework

To investigate implementation bottlenecks, we focused on the feasibility of nPOC HIV VL monitoring, and we used the MIDI theoretical framework [17]. The MIDI framework is designed to help implementation researchers improve their understanding of the critical determinants that may affect the implementation of an intervention in order to better target the implementation strategy. The framework consists of 29 subdomain constructs divided into four groups of determinants (domains) associated with (1) the intervention, (2) the provider, (3) the implementing organisation, and (4) the social, political, and cultural context. The operationalisation of each domain and its subdomains is presented in Table 1.

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Items and constructs of the adapted MIDI for nPOC HIV VL monitoring.

https://doi.org/10.1371/journal.pone.0351304.t001

Data collection procedures

Before starting the interviews, we provided all HCWs with a thorough explanation of the study, including its objectives and procedures, and obtained written informed consent. The interviews were conducted by trained researchers with backgrounds in qualitative research, comprising PM (MSW, PhD candidate), AM (MPH), RM (MPH), and RA (MD). All Tanzanian health researchers are familiar with HIV service contexts. All interviewers had prior experience working with HCWs in HIV care settings and none had a prior relationship with the participants before recruitment. Researcher reflexivity was considered throughout the study. PM, with a background in social science and PhD candidate, played a role in coordinating data collection. PM and other research assistants conducted clinic observations and in-depth interviews with HCWs at the selected sites. Then, PM led regular debriefing sessions with the research assistants for emerging themes and ensured the use of neutral probing language. HCWs were interviewed at three time points (months 0, 1 and 6).

Saturation was assessed by continuously reviewing the data throughout the interview process to determine whether new themes were emerging. While saturation was reached across most of the MIDI domains, the themes varied slightly by time point: baseline interviews focused more on provider knowledge, self-efficacy and compatibility with existing practices, while post-implementation interviews introduced new perspectives on organisational coordination, workflow, and client responses to nPOC HIV VL testing. At month six of the interviews, no new themes emerged, indicating that full saturation had been achieved across sites and domains.

Data analysis

We conducted a deductive thematic framework analysis using the MIDI tool, with an inductive approach added. All transcripts were imported into NVivo 12 for coding and organisation. Two researchers (PM and AM) double coded the first six transcripts using a preliminary codebook based on the MIDI framework. During this phase, inductive codes were generated to capture data that did not fit within the redefined MIDI categories. Analytic memos were created during this stage and discrepancies were resolved through discussion and consensus leading to refinement of the coding framework. PM applied the refined combined deductive (MIDI-based) codes with inductively derived subthemes to the remaining 69 transcripts. For consistency, we repeatedly reread the transcripts during analysis and held ongoing discussions within the research team to confirm interpretation. We used NVivo 12 to organise the codes into nodes, highlighting them with coding stripes for consistency, and to track analytic decisions via memos, making the process clear and easy to follow. We created a summary table linking codes to quotes across the four MIDI domains to clearly show how the data aligned with the framework. Credibility was enhanced through: (1) triangulation of IDIs with structured clinic observations; (2) peer debriefing with co-authors to make sure our interpretation makes sense; and (3) use of the MIDI framework as a guide to organise and check our coding of the data. All participants shared similar experiences, and no conflicting or contradictory responses were identified during analysis. The full coding tree and a quote code matrix are provided as supplementary files, S3 Fig and S4 File, respectively.

Ethical clearance

The study obtained ethical approval with the certificate NIMR/HQ/R8.a/Vol. IX/4140 from the College Research and Ethics Review Committee at Kilimanjaro Christian Medical University College, as well as the National Health Research Ethics Committee of the National Institute for Medical Research. Written informed consent was obtained from each study participant prior to enrolment. Participant information sheets and informed consent forms were translated from English into the Swahili language most commonly used in Tanzania. Each participant (or parent/guardian, where applicable) signed two copies of the informed consent form: one copy was retained by the participant and the other by the research team for documentation. For individuals unable to read or write, consent was provided using a thumbprint, witnessed by an impartial witness chosen by the participant. For participants under 18 years who were not mature or emancipated minors, parent or guardian consent was obtained in addition to the child’s assent (for those aged 8–17 years). Mature or emancipated minors, such as married adolescents, parents, or those heading households, were permitted to provide their own consent, though they were encouraged to involve parents or guardians if safe to do so. This study is reported in accordance with Consolidated Criteria for Reporting Qualitative Research (COREQ) guidelines provided as S5 Checklist.

Overview of results and analytical framework

The results are presented in alignment with the study objectives: (1) to investigate factors that contributed to the successful implementation of nPOC HIV VL monitoring and (2) to explore challenges faced by HCWs during implementation. The implementation process of nPOC HIV VL monitoring was guided by the MIDI framework, with findings organised into its four main themes. Within each theme, subthemes were identified based on the 29 MIDI subdomains. To provide context for the findings, we first describe the characteristics of participants and the pre-implementation context. Finally, we describe the key facilitators and challenges in two sections, each organised by the MIDI domains. All quotations are identified by participant cadre, gender and age (e.g., Nurse, Female, 34 years). While most participants’ accounts aligned with major themes, some highlighted challenges they encountered, reflecting variations in perspectives.

Participant characteristics

A total of 75 interviews were conducted among 43 HCWs across three time points using a design of longitudinal follow up and cross-sectional sampling. At baseline (T0), 33 HCWs participated in interviews. At one month (T1), 25 participants were interviewed, this included 19 participants returning from T0 and 6 new participants. Of the 33 participants at T0, 14 did not attend T1, of whom 7 later returned at T2. Among the 25 participants interviewed at T1, 8 participants were unavailable for further follow-up due to staff transfers, annual leave, or competing clinical duties. By the sixth month of follow-up (T2), 17 participants were interviewed, comprising 6 participants returning from T1, 4 new participants, and 7 participants from T0 (see S1 Table).

Most participants at baseline were female (82%) and most comprised individuals aged 30–39 years (37%). Nurses represented the largest professional group at baseline (49%). However, at six months, the participant profile shifted, with a higher representation of doctors (53%) and a more balanced gender distribution (53% female). The socio-demographic characteristics for each time point are detailed in Table 2 below.

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Demographic characteristics of healthcare workers: Before, one month and six months after nPOC HIV VL monitoring implementation.

https://doi.org/10.1371/journal.pone.0351304.t002

Pre-implementation context

Before the implementation of nPOC HIV VL monitoring, we conducted observations at the clinic and interviews with HCWs on the standard operating procedures (SOPs). Findings confirmed the standard operations as described in the Tanzania national HIV VL monitoring guidelines. HCWs explained to the client why the VL test was being conducted. Blood was drawn and sent to the central laboratory for processing and testing. The results usually took two weeks to one month to return to the clinics and the HCWs would inform the client of the results during a follow-up visit. Clients with a suppressed viral load (VL) were encouraged to continue taking their medication as prescribed. Those with high VL levels received counselling and were enrolled in more intensive adherence support programs.

Factors contributing to successful implementation of nPOC HIV VL monitoring

In all four MIDI domains – the intervention, the provider, the organisation, and the socio-political context – HCWs mentioned factors that contributed to the successful implementation of nPOC HIV VL monitoring.

Regarding determinants associated with the intervention, the introduction of nPOC HIV VL monitoring gave HCWs a new responsibility: for clients whose VL was high (≥ 1000 copies/mL), the HCWs needed to clearly explain the new testing process, before taking a new sample for an nPOC test for confirmation, the results of which would be available within approximately two hours. If the VL was high, this allowed for same-day targeted counselling to address barriers to adherence and decide about the next steps (procedural clarity). HCWs trusted the nPOC HIV VL test results, considering them reliable for assessing client treatment progress and VL status. A doctor had this to share:

“We interpret the results correctly by understanding whether a client’s viral load is suppressed or high, which helps us assess treatment progress: a suppressed viral load indicates effective treatment, while a high viral load signals the need for enhanced adherence counselling and further monitoring.” (Doctor, Male, 28 years) (correctness)

Complete information on the procedures and steps of nPOC HIV VL monitoring was available through the SOPs, which ensured quality and continuity of care. The SOPs supported the HCWs in adhering to professionally accepted standards when providing nPOC HIV VL monitoring services. A lab technician reported that:

“We cover everything according to the standard operating procedures; from sample collection and methods, we use to ensure accurate results. There are many aspects to consider.” (Lab technician, Male, 37 years) (completeness)

HCWs highlighted that nPOC HIV VL monitoring aligned with the values and working methods in place and that the same-day results made it easier to observe client outcomes and take timely actions. A doctor reported that:

“Near POC HIV VL monitoring is the same as the existing viral load monitoring system in terms of procedures; the only difference is that with POC, the results come back in a short time.” (Doctor, female, 24 years) (compatibility)

The HCWs mentioned that receiving their clients’ results on the same day instead of waiting weeks was helpful for them. It enabled them to understand the clients’ condition more quickly and decide on the next steps and has made it easier to comprehend the clients’ progress more thoroughly. A doctor noted as:

“When the client’s results are ready, I receive them from the laboratory. The same day results make it easier to know the client’s results and undertake the necessary management.” (Doctor, female, 24 years) (Observability)

Regarding the provider, the interviews with HCWs noted that the implementation of nPOC HIV VL monitoring brought several benefits. HCWs highlighted that, instead of transporting the sample to the central laboratory, they could now collect and test samples at the nearby lab using nPOC machine, which made them feel more in control of the process. A doctor reported:

“Our regular procedure takes much longer to get results compared to near point of care VL monitoring, where results come on the same day, which in turn helps us to do early counselling.” (Doctor, female, 24 years) (personal benefits)

Where possible, HCWs included an explanation of what a high or low VL meant as part of their health counselling to promote treatment adherence. The timely delivery of results, along with accurate visibility of low or high VL, was seen as motivating to the clients and rewarding to the HCWs, as it realised the fruit of their work. A nurse highlighted that:

“It is advantageous for the client to receive their results on the same day. This allows them to understand their next steps and receive any necessary counselling.” (Nurse, female, 35 years) (outcome expectations)

HCW felt that the implementation of nPOC HIV VL monitoring not only improved their efficiency but also gave them a chance to do what they feel responsible for providing accurate results to their clients during same-day targeted counselling. A doctor reported:

“When you see a client has a high viral load, you cannot be happy, because your work requires you to make sure the person has received the POC service and is suppressed. So, you must start counselling and give him new education immediately…” (Doctor, male, 42 years) (professional obligation)

Strong teamwork was key within the organisation. HCWs described how they supported each other by dividing their tasks; if one took a sample, another would process it and the third might provide counselling. A nurse explained:

“In our clinic, we work together as a team. If I am occupied with another client, a colleague will step in to collect the sample or complete the forms. This kind of support makes it possible to keep the service running smoothly even on busy days.” (Nurse, female, 35years) (social support)

Observation at the clinic confirmed that HCWs helped each other, which made things easier in every procedure of nPOC HIV VL monitoring.

HCWs expressed confidence in their ability to effectively conduct nPOC HIV VL monitoring. With the recent advancements and improved protocols, they are now well-equipped to carry out the nPOC HIV VL monitoring process accurately and efficiently. It was noted that:

“In the beginning, we did not understand the POC processes very well. But after receiving the training and working with the system over time, we have gained more confidence and now we understand how to manage the procedures better.” (Doctor, female, 42 years) (self-efficacy)

Having the understanding required to effectively perform nPOC HIV VL monitoring was noted by HCWs. This included familiarity with the technology, comprehension of its features and the ability to apply relevant skills to make full use of the innovations. A doctor highlighted:

“I know how the viral load test works, the importance of collecting a good sample and how to store and transport it to ensure accurate results.” (Doctor, female, 58 years) (knowledge)

In general, HCWs were eager to learn more about nPOC HIV VL monitoring and a few of them mentioned requiring refresher training.

HCWs felt they carried a big responsibility for the implementation of nPOC HIV VL monitoring. They reported that they initially lacked awareness of nPOC HIV VL monitoring, but that the training they had received helped them understand the contents, follow the procedures and interpret the results. A lab technician confirmed this:

“Before implementation of POC, we had no idea about doing viral load tests ourselves. We used to send everything to [the]central laboratory. It was new for us.” (Lab technician, female, 36 years) (awareness)

Regarding determinants associated with the organisation, after the HCWs received training on nPOC HIV VL monitoring, they reported receiving formal commitment by the management of the organisation, enabling employees to implement the intervention well. A nurse confirmed that:

“…the management supported us in implementing the point-of-care viral load testing. They allowed us to adjust duties so that someone is available to run the nPOC test machine and attend to clients. Without that support from management, it would have been difficult to integrate it into our daily routine.” (Nurse, female, 34 years) (formal ratification by management)

HCWs had dedicated time to incorporate nPOC HIV VL activities into their daily schedules, as observed during clinic observations and informal conversations. This helped ensure that the nPOC HIV VL monitoring service did not interfere with other clinic activities, such as HIV care and treatment services (time available).

HCWs noted that it was easy to find information regarding the intended procedures for conducting nPOC HIV VL monitoring at the facility. The accessible information included detailed guidelines that outlined best practices and effective implementation strategies. Access to information ensured HCWs had a clear understanding of how to apply the innovation effectively. A nurse highlighted that:

“When we need to check guidelines or confirm the correct procedure for viral load testing, the information is available in the clinic in printed form. This makes it easy to refresh our memory and follow the correct steps without delay.” (Nurse, female, 34 years) (information accessibility)

The organisational leadership showed a significant contribution to the implementation of nPOC HIV VL monitoring by collaborating well with staff and providing regular performance feedback on the progress of nPOC HIV VL monitoring. The feedback was frequent and aimed to show progress in the implementation of nPOC HIV VL monitoring, including the delivery of nPOC results and improvements in treatment adherence counselling. A lab staff member noted:

“When we submit nPOC viral load testing reports, the management and sometimes the national program give us feedback on our performance. This helps us understand where we are doing well and where we need to improve.” (Lab technician, male, 38 years) (performance feedback)

Certain HCWs commented positively on the socio-political context of the intervention. They viewed the implementation of GeneXpert as a most welcome inclusion, in line with best international practice as promoted by the WHO, even though national guidelines for practice had yet to accept such innovation. From a policy perspective, a nurse argued:

“I don’t think the POC test violated any of the Ministry’s guidelines. This is because the viral load test is done in the same way as it’s usually done. The only difference is that with point-of-care, we get the results on the same day. While if we follow the entire standard procedure [of centralised testing], we end up delaying the client. I think this method is very good.” (Nurse, female, 49 years) (legislation and regulation)

Challenges in implementing nPOC HIV VL monitoring

Despite the positive intentions, several challenges arose and affected implementation. Regarding the innovation, at the first stage of the intervention, HCWs encountered difficulties using the GeneXpert machine. The machine requires precise amounts of blood and if there is too little blood, it causes an error and gives the wrong results (complexity).

HCWs also raised the challenge that they face in delivering timely results to clients. To address this, they adapted their service delivery by offering flexible options to accommodate clients’ schedules and minimise inconveniences. A nurse noted that:

“Because of delays and clients having other commitments, we give them the option to go home and then we call them when their results are ready.” (Nurse, female, 42 years) (relevance for client)

Regarding determinants associated with the provider, due to too few HCWs available, many had multiple responsibilities, resulting in delays in clinic services and increased pressure on the team. A doctor highlighted:

“We don’t have enough staff to cover all the services at the same time. Sometimes one person is handling the pharmacy, counselling, and sample collection, which makes it hard to keep up with all the clients.” (Doctor, female, 42 years) (personal drawback)

Furthermore, through informal conversations with HCWs, they reported that clients were not satisfied with the nPOC test as counselling did not always occur when results were available, particularly when the client had other activities and could not remain on-site to wait for their results. At some participant sites, samples had to be transported to the nearest laboratory by motorbike couriers (boda-boda), which caused delays in obtaining results at three sites Mawenzi, Pasua and Majengo). Based on field observations, transport typically took 20–40 minutes each way, with additional queue time at the receiving laboratory of approximately 30–60 minutes, extending total turnaround time to 3–4 hours or more compared to approximately 90–120 minutes at on site nPOC laboratories (Sinza, Hai). This distinction affected same day counselling feasibility at transport dependent sites (client satisfaction). When this happened, HCWs called clients to share their test results and expected the clients to cooperate further. If the results showed a high VL, clients were expected to return for a separate counselling session (client cooperation).

Regarding determinants associated with the organisation, in some facilities, only a small number of staff were directly involved in nPOC HIV VL monitoring, leading to gaps when a staff member left. Regarding continuity, a doctor reported:

“If someone trained in POC leaves, we don’t have a structured plan to immediately replace them with someone equally trained. That delays implementation.” (Doctor, female,42years) (replacement when staff leave)

HCWs highlighted a limited number of staff within their facilities. They emphasised the need for an adequately trained workforce to effectively utilise the nPOC HIV VL monitoring system and ensure its integration into routine clinic operations.

A doctor had this to share:

“At one centre, there are only a few staff members involved in POC testing. This limited staffing means that if one person leaves or is unavailable, the workload becomes challenging to manage, although the remaining staff try to continue the work.” (Doctor, female, 58 years) (staff capacity)

The sustainability of the nPOC HIV VL monitoring system was a common concern among HCWs, largely due to limited financial support. HCWs did not identify any financial resources allocated for the long-term implementation of nPOC HIV VL monitoring. A doctor explained:

“The nPOC HIV VL monitoring equipment is expensive and challenging to obtain; with funding, it is possible to run it. If those resources were not there, we could not afford to maintain [it] ourselves.” (Doctor, Male, 42 years) (financial resources)

Infrastructure also causes problems; there are often too few service rooms. One doctor highlighted:

“For my part, I think the rooms are not enough for nPOC implementation. If three or two rooms are present, then the working conditions will be comfortable.” (Doctor, male, 53years) (material resources and facilities)

In addition, regarding the presence of material resources, a doctor noted that:

“There are times when we lack gloves and other laboratory supplies, and we have to wait until they are brought from the main hospital.” (Doctor, female, 24years) (material resources and facilities)

Unreliable electricity and a lack of generators also hindered sample processing. Transportation delays to the nearby laboratory led to sample rejection or result delays. A nurse reported:

“The biggest challenge is unreliable electricity; we do not have a generator, so sometimes the tests are delayed, or samples get spoiled.” (Nurse, female, 28 years) (material resources and facilities)

Overcrowded waiting areas and non-client-friendly counselling spaces were observed, which compromised the quality of the service. On the coordination of some of the facilities, a doctor noted that:

“Sometimes you find the coordination (at the clinic) is not smooth because everyone has their own work. The lab can be busy with other tests, and we are waiting for results. You may have to follow up yourself to make sure clients are not delayed.” (Doctor, female, 42 years) (coordination)

Furthermore, HCWs reported that the implementation of nPOC HIV VL monitoring took place within an unsettled organisational structure, which added additional coordination needs, requiring regular morning meetings to assign responsibilities. A doctor explained:

“Roles and responsibilities are not clearly assigned, leading to confusion. Every morning, we have a nPOC HIV VL monitoring meeting, followed by a hospital meeting.” (Doctor, male, 27years) (unsettled organisation)

Clinic observations revealed several challenges associated with the socio-political context regarding the alignment of nPOC HIV VL monitoring with existing regulations in Tanzania. Consistent with WHO’s 2021 policy, the Tanzania national guidelines used a threshold of >1000 copies/mL to distinguish stable from unstable clients. According to WHO framework, a viral load >1000copies/mL signifies virological failure, while results falling between >50 and ≤ 1000 copies/mL signify low-level viremia, which necessitates enhanced adherence counselling and repeat test after three months. Results ≤ 50 copies/mL are classified as undetectable (target not detected). On top of that, in northern Tanzania, at our study sites, we used 50 copies/mL as imposed by Elizabeth Glaser Pediatric AIDS Foundation (EGPAF). It is important to note that <50 copies/mL represents a laboratory sensitivity benchmark for ‘undetectable’ viral load and is not WHO’S policy threshold for defining suppression (legislation and regulation).

Another challenge involved transitioning from mPIMA to GeneXpert. It was noted that not every POC test can be used for VL monitoring in each setting, depending on the guidelines/practices. GeneXpert nPOC HIV VL monitoring, with its high sensitivity (detection limit approximately 50 copies/mL), was viewed as the best option for providing clinically meaningful results and aligning with future national guideline updates, particularly given the global trend towards more sensitive VL monitoring.

Framed in terms of the determinants of the MIDI framework, HCWs did not experience many problems regarding determinants related to either the intervention or the providers. All subdomains related to the intervention were positively judged, as the procedures for nPOC HIV VL monitoring were clearly described (procedural clarity), based on factually correct knowledge (correctness); the activities described were complete (completeness), aligned with the existing values and working methods in place (compatibility), and the produced outcomes were visible to the user (observability). However, usability of the machine was reduced due to complexity, and relevance to clients was limited by delays in turnaround time for the results.

In terms of provider determinants, most subdomains show enabling factors rather than obstacles: HCWs perceived personal benefits from using nPOC HIV VL monitoring, and believed it helped them achieve their objectives (outcome expectations), aligned with their professional responsibilities (professional obligation), and they expected client’ cooperation. They also reported receiving support from their colleagues (social support), felt confident in their ability to perform the required tasks (self-efficacy), and demonstrated adequate knowledge and awareness of the intervention. Some problems were reported, including personal drawbacks associated with using the system, lower-than-expected client satisfaction, and reduced cooperation. Additionally, expectations of important others (subjective norms) and colleagues’ observed behaviour (descriptive norms) were not clearly mentioned.

At the level of organisational determinants, some subdomains were mentioned as enabling implementation, including formal ratification by management, the time available for nPOC HIV VL monitoring, replacement of staff when they leave, access to information and feedback to HCWs about their use of the nPOC HIV VL monitoring process (performance feedback). However, in the domain of the organisation, many problems were noted, including a lack of replacement when staff left; limited staff capacity, financial and material resources, and facilities, as well as problems with coordination and an unsettled organisation.

On determinants related to the socio-political context, legislation and regulations appeared as a positive subdomain; however, integrating the intervention into national policy proved to be a challenging task.

Overall, the findings show that nPOC HIV VL monitoring improved turnaround time for results and enabled same-day counselling, supported by training, teamwork, and leadership. HCWs adapted well, but faced challenges such as machine errors, staffing shortages, infrastructure issues, management and coordination problems, and a lack of national guideline alignment. Despite these challenges, HCWs remained motivated and viewed the innovation positively.

Recommendations

Successful implementation of nPOC HIV VL monitoring in Tanzania requires a multi-faceted approach addressing organisational capacity, policy integration, and system strengthening. While HCWs demonstrated a readiness to adopt nPOC HIV VL technology, sustainable implementation depends on addressing fundamental health system constraints, including staffing, infrastructure, and policy support.