Data analysis.

A thematic analysis was conducted on focus groups and personal interview transcripts following the seven stages of the Framework Method [27] to identify codes and construct main themes. Two researchers, AEO and ERS, reviewed the transcripts and agreed on emerging themes, which were later discussed with the research team. Two main classifications were defined prior to the analysis: (1) barriers and challenges, and (2) facilitators and strategies to streamline the implementation of BP@home in the real-world setting. Within these groups, we later identified four sub-categories derived from the data: (i) project management, (ii) logistics (including technical and IT issues), the recruitment and engagement of (iii) PCNs and practices, and (iv) patients. We contextualised our findings using the second iteration of the Consolidated Framework for Implementation Research (CFIR) published in 2022 [28] and proposed evidence-based recommendations to help streamline the implementation of the BP@home initiative in London’s primary care setting.

Project management.

Some respondents expressed confusion and frustration due to shifting directives in the project, which they thought has led to a waste of time and resources:

“It started with being a project about working with UCLP. It then sort of got honed down to a project about hypertension. So a lot of the time, actually the funded time was lost doing other things before somebody decided this is what they wanted us to do” [PM]

Further, the original BP@home national SOP, while useful, was also considered by some respondents as “insufficient” and still requiring substantial work and “time to drill down into the logistics of how specifically this is going to work.” [PM]

Respondents also commented on the impact of the high rate of staff turnover, which required constant training of new recruits and contributed to “the organisational memory loss”. Finally, the BP@home program was perceived, especially by HCPs involved at the practice level, as being too “top-down” and not responsive enough in its approach:

“I feel that I’ve had very little ability to influence things at an ICS level. You know, we’ve had lots of frustrations with the delays and the BP machines and no matter how many times we’ve raised those concerns, nothing has really changed. [GP]

This statement was further reinforced by one of the clinial pharmacists:

“I think I kind of echo what what was said at least in terms of influence of the [local PCN] level, I you know, we received some communications every now and then that was that was it” [CP]

Logistics.

A central logistical barrier identified by almost all respondents related to the reception, storing, distribution and tracking of BPMs, reinforced by the absence of a unified management template and leading to the potential waste of BPMs as a result.

In one of the ICSs, the delivery of 4,500 BPMs created "a bit of a panic" [PM]. The BP@Home initiative was described by this respondent as "pretty chaotic" with the distribution deemed as "unprofessional". Another respondent reported a significant error with the application for BP monitors which left their ICS without any monitors for the third wave of allocation. In addition, respondents reported an insufficient number of XL cuffs, which were sometimes limited to only one per practice. Accessing the BP monitors "was complicated and time-consuming" with some primary care sites still waiting for the machines while "there’s now like two office rooms at the federation that are just chock full of monitors." [PM]

For practices, especially the smallest ones, the lack of time and human resources, exacerbated during the Covid-19 pandemic, meant that they had to adapt the type and amount of information they could record and track since they "do not have the manpower or infrastructure to keep an accurate log of [BPMs]" [GP]:

We initially requested the serial numbers [of BPMs] but then decided not to record them—it is very time-consuming and [we’re] uncertain about the clinical relevance. It was a lot of work for a small practice. So we opted instead to record the cuff sizes. [PM]

Tracking and management issues were further reinforced by the absence of a unified template from NHSE to support the tracking of monitors and patients. Respondents also highlighted that most had little awareness of existing IT tools to help effectively track monitors and patients.

Finally, while respondents were mostly satisfied with AccurX which was the most commonly used tool to request and receive blood pressure readings from patients, some limitations were emphasised, including:

the problem with AccurX is that is sounds as if it is text-based but it actually requires a smartphone (…) And we find that quite a lot of our AccurX requests fail either because the patient doesn’t want to bother taking their BP for four days or because they just can’t manage the tech. [PM]

One respondent complained about the links sent to patients using SMS requesting the readings expired automatically after a few days, and because this added a time constraint for patients, it may have further reduced patient compliance and the number of readings communicated to HCPs. Another respondent noted that the lack of automated feedback from AccurX may have limited the activation of patients using reminders, with the follow-up being reliant solely on the availability of HCPs.

Recruitment and engagement of PCNs & practices.

The lack of adequate, dedicated funding to support the recruitment and implementation of the BP@home program was perceived as an important barrier to the engagement of PCNs and practices:

The only financial incentive for this is actually getting the monitors. There’s no financial gain for a PCN, and that’s why it’s such a huge sort of issue. [PM]

This was particularly problematic due to the substantial additional workload associated with participating in BP@home. When approached, numerous practices saw this as: "an extra piece of work" and "too complicated" [PM], especially in the context of the ensuing Covid-19 pandemic and national lockdown which severely disrupted all aspects of daily life.

Broadly speaking, the initiative was highly welcomed. However, due to pressure from implementing vaccine rollout for Covid-19, resources were redeployed [PM]

This further emphasised the existing differential capacities of PCNs and practices. One respondent described PCNs as "very variable beasts" with some "working very, very closely together" where you can "get a whole program out across the PCN" while others described practices as "forced to work together (…) but not actually really working together" [PM]. Similarly, another HCP noted that "skill sets vary from practice to practice" [PM] thus affecting their capability to participate in this type of program.

Finally, these engagement issues led to equity concerns, in particular due to the unequal distribution between PCNs which might exclude or disenfranchise the most vulnerable patients:

There are PCNs that have had lots and lots of monitors and their patients are benefiting, and you’re having these populations who have high levels of deprivation, who again have missed out on these BP monitors. [PM]

Recruitment and engagement of patients.

Similar issues were reported regarding how patients were stratified or recruited into the BP@home pathway. While interviewees reported a general ease of use of BPMs by patients, they found that many patients did not follow the recommendations:

“Patients didn’t need much help with using the BPM, they were quite comfortable using them. But many did not follow the request to send 4 readings a day, even after being sent a reminder and being provided a diary in the letter." [PM]

While the IT limitations with AccurX might have played an important role, respondents also highlighted other IT-related challenges that could explain some of the challenges encountered. Several HCPs reported some reluctance on the part of their patients due to their fear of losing face-to-face contact, especially in a context of increasing remote socialising, due in part to the Covid-19 lockdown but also to the already existing digitalisation of services. Another explanation for the limited engagement of patients was their immediate perception that their hypertension was not a priority in the context of the ensuing Covid pandemic where "getting a jab” [PM] and limiting physical contact were paramount. The prolonged duration of the BP@home initiative was also identified as a potential issue, along with the "experimental" nature of the program:

It went on for too long and patients lost interest. It might have helped to have a timeline. If we don’t dress it up as a trial, patients might take it more seriously. [PM citing feedback received from other care coordinators in the local PCN]

Some of us [HCPs] actually believed this was a pilot—we were told this was a pilot- but it turns out general practice will now be doing this [BP@home] for ever [PM].

This lack of engagement was reported to be particularly acute among patients with multiple conditions and high levels of deprivation with whom engagement with primary care was already limited. There was also the impression that patients with high BP might feel discouraged by their readings and thus prefer not to measure, or communicate the readings, since those communicating their results the most or more frequently were the patients needing the least care as their hypertension was not impacted by multimorbidity. Another issue raised by a respondent was that once reassured that their BP is good, patients might not feel the need to keep self-monitoring and/or submitting readings.

Project management.

One programme manager described that communication regarding BP@home was "very good throughout the initiative" and "ongoing" [PM]. The regular meetings with NHS England and at the ICS level were good opportunities to share information, discuss processes and systems and increase efficiency.

Other respondents emphasised the importance of task-sharing and coordination centred on long-term conditions (LTCs). Reflecting on the ICS one respondent was allied with:

There is someone being appointed as the [ICS] LTC clinical lead and their job is basically going to be to go to practices and try and sort of establish what the problems are and what support they need, what resources they need, what’s lacking in certain practices and PCN [PM]

Such centralised coordination at the ICS level was also recommended for practices that, according to several respondents, should all have a named person in charge of the BP@home, and that this local “champion” should be trained and funded:

… because it takes time, and the practices that are successful with BP at home are the ones who have a person who has allocated time to be the responsible [GP]

Whereas one respondent previously countered this proposal, one participant recommended the incorporation of BP@home into routine general practice and that it should be considered business as usual:

It is here to stay. It is a priority as part of pushing proactive care. I do think it can become more and more sustainable, but we just need to keep tweaking and refining how we integrate it into our everyday approach to patients with long-term conditions and also how we support patients in their engagement, especially for groups who have been hard to engage [PM]

In order to support this integration into daily practice, processes should be based on NHSE recommendations but still be adapted locally so that they work in the real world even if they are tailored to be practice-specific:

Systems for managing results have to be developed for each practice as all practices work differently and it has to be decided who will respond to results etcetera. Some of these resources we have shared with practices existed on the NHS futures website and have been adapted for use locally [PM]

Logistics.

A central, recurrent recommendation to the logistical challenges encountered in terms of storing, distribution and tracking of BPMs was to provide the devices on prescription. This was also presented as a way to deliver more equitable access to the machines. One respondent shared their ICS plan of recycling BPMs 3 to 4 times, after a few weeks with each patient. This ‘recycling’ was in line with the BP@home SOP to promote equity by avoiding a first-arrived, first-served logic which felt "irresponsible" and to limit waste and underuse. Respondents provided examples of how some PCNs intended to lend the monitors for a week to tier or priority group 3 patients and for 8/10 weeks and up to 6 months for tiers 1 & 2 patients. With a similar concern towards addressing inequalities, another respondent explained that the allocation of BPMs to practices was weighted based on deprivation scores and demographics.

Another central driver to the successful implementation of the BP@home initiative was the development of practice-specific processes since:

There are a lot of processes to be thought about and developed. And it still does for any practice because it happens at practice level: each practice needs to have systems for managing things they didn’t use to manage, which is patients sending up results from home, and also onboarding them, giving equipment, etc. [PM]

In parallel, some respondents also emphasised the need for simplified, unified IT tools and tracking templates:

"we’ve brought together quite a lot of resources, but actually it would be quite nice to have them much more streamlined and easily available. Maybe adapting the template and simplifying it. [PM]

Another recommended strategy was to involve pharmacists in the planning of logistics due to their privileged relationship with patients as well as their experience in tackling logistical issues:

Pharmacists are brilliant at motivating patients as well as making sure that data is being recorded in the right way so that it’s useful for them, for your IT people to pick up things on their searches and things like that [PM]

Recruitment and engagement with PCNs and practices.

Several drivers to support the recruitment and engagement with PCNs and practices were identified. A central recommendation referred to increased incentives, including the use of clinical targets, such as QoF or local hypertension goals as part of broader LTC plans. In one ICS, a conditional financial incentive was provided:

[we] set up an incentive scheme as we recognised that money would be helpful. 130,000—so about 15,000 per borough for the eight boroughs, about two or three thousand per PCN, so very little per practice but it smoothed an awful lot of feathers and was very helpful. It also gave an opportunity to provide a plan: they had to put up a PCN team, (…) and confirm that ten unique patients had submitted readings remotely to receive the incentive [PM]

All respondents agreed that adequate funding was essential to ensure the sustained and efficient implementation of the BP@home programme since "a lot of the work that’s expected to be done is sort of out of an already stretched system that doesn’t have actually any additional capacity". [GP]

Another driver to support engagement with practices was the implementation of a needs assessment plan -–including training, role-playing, feedback and standardisation of materials -–to identify and address differential capacities within participating practices.

Recruitment and engagement of patients.

One of the drivers supporting patient recruitment was the quality of the onboarding material provided by NHSE, praised by respondents as "fantastic, extremely clear, a lot of support with IT, how to explain to patients that it would benefit them, and how" thus leaving this respondent with "no unanswered questions, we didn’t feel like we were in the dark.” [PM]

In addition, the identification of eligible patients was facilitated by the integration of the UCLP priority list onto EMIS and SystmOne, "which breaks patients down into different groups" [GP]. However, this procedure usually yielded many more eligible patients than available monitors. Most practices described adopting a pragmatic or opportunistic approach to prioritising patients, such as those described below:

If we have only 20 machines, there is no point sending 200 texts: only send 20 texts saying “We identified you to benefit from this program”. If the patient sends back “yes”, call them back. (…) Pragmatically, we simply called back those who responded yes [GP]

In another practice, the team decided to consider the search for eligible hypertensive patients identified using the UCLP Proactive Care stratification tool who were known diabetics.

We looked at the priority list and then matched them with high-risk patients who have pre-existing diabetes. Then we sent en masse text messages to all the patients in the cohort to identify those who have a BPM at home. (…) then started to engage in terms of creating a management plan that is unique to our practice in that it included information about BP, the medication they were taking, along with instructions about how to monitor BP at home [CP]

The adoption of a more person-centred approach, which prioritises the patient’s needs, understanding and acceptance of the initiative, was highly encouraged:

Patients have to understand why we’re doing it and that is not just a thing the practice is asking for the benefit of the practice. They need to see that it’s personally helpful for me to understand my blood pressure and to have this engagement and interaction with the clinician [PM]

The same respondent suggested one of the ways to better adapt to each patient’s needs was to reduce the number of requests in case of satisfactory readings. In one practice for example:

If their BP was to target then we said OK, you don’t need to do this for another 12 months and we automated it (…) because we don’t want patient fatigue. [PM]

Finally, numerous respondents emphasised the necessity to maintain non-digital options for communicating results for patients with limited digital literacy and/or access in order to avoid excluding patients. Several practices reported using a multiplicity of channels as exemplified in the quotes below:

We use four different methods of sending in their BP readings: they can hand them at the reception by hand, they can email them to the practice email address, they can email the Excel spreadsheets that averages them for us [GP]

the remote monitoring requires the patient to at least have a mobile phone. So we have some older patients (…) or actually patients who maybe are from the slightly more deprived backgrounds who have mobile phones but don’t have Internet access on their mobile phone. So that was OK because we sent the initial messages just as an SMS text message with the information and they could reply to the message if they wanted to bring their results in on paper. So we did have a cohort of patients who don’t want to use Floreys* for whatever reason. They either got confused using a Florey* or they didn’t have Internet access or their mobile phone was a more basic mobile phone without Internet access. So depending on the area, but it was usually not more than 10% of patients were still bringing paper copies in, but we still use the same protocol, the same SOP, to monitor them and so on.

Future: An inevitable shift that requires adjusting.

Finally, it is important to note that remote monitoring of blood pressure was perceived by most respondents as inevitable, and even desirable, despite all its challenges:

It is definitely hard for GPs but in the long run, most would agree that it is a much more sound way of tracking people’s BP due to the "white coat" syndrome [GP]

However, as noted by one respondent, this type of remote monitoring is still perceived by some HCPs as an alternative, additional option and not a replacement to face-to-face care and consultations:

It is going to be interesting to see what happens with community pharmacy in terms of task-shifting. Will it increase the number of people taking care of their health? This would be great though it does not replace a visit to the GP. It is not either or. Hypertension reviews will still be happening. [PM]

Contextualisation of findings using the Consolidated Framework for Implementation Research (CFIR).

Our study’s findings, examined through the lens of the second iteration of the Consolidated Framework for Implementation Research (CFIR) published in 2022 [28], offer comprehensive insights into the implementation of the BP@home initiative in London’s primary care setting.

Intervention characteristics. The BP@home initiative which is a complex intervention with multifaceted components primarily encountered challenges in its implementation due to resource constraints. These constraints align with the ’Intervention Complexity’ and ’Resource Availability’ constructs of the CFIR. The initiative’s design, which required significant IT, human and financial resources, was also a critical barrier, reflecting CFIR’s emphasis on both the intervention’s adaptability and cost. However, the integration of eligibility criteria into electronic health records, which is a feature of the ’Design Quality and Packaging’ construct, emerged as a facilitator. This enhanced the initiative’s uptake but more work should be done at a national level to ensure that the distribution of blood pressure monitors does not exacerbate extant inequalities [31].

Outer setting. Our findings highlight the impact of ’Patient Needs and Resources’ and ’External Policies and Incentives’, whereas the pandemic’s urgency necessitated rapid adaptation of healthcare practices, aligning with the ’Cosmopolitanism’ construct. However, disparities in resource distribution, a component of the ’Outer Setting’, exacerbated equity concerns, highlighting the need for policies that ensure equitable access to healthcare innovations.

Inner setting. The ’Inner Setting’ of primary care practices significantly influenced the implementation. Resource inadequacy, aligning with the ’Readiness for Implementation’ construct of CIFR, and the additional workload posed challenges. However, CFIR’s ’Networks and Communications’ construct was evident in the collaborative efforts among PCNs and practices, which facilitated smoother implementation. This collaboration, alongside CFIR’s ’Culture’ and ’Implementation Climate’ constructs, will be pivotal in addressing logistical issues and promoting a supportive environment for the BP@home initiative.

Characteristics of individual. HCPs’ perceptions and engagement played a crucial role as emphasized by CFIR’s ’Knowledge and Beliefs about the Intervention’ and ’Self-Efficacy’ constructs. HCPs’ pragmatic approach to patient onboarding and their adaptability reflect the ’Individual Stage of Change’ and ’Personal Attributes’, and highlight the importance of engaging and training HCPs to foster positive attitudes towards new health interventions.

Process. The ’Planning’ and ’Engaging’ constructs were also critical in success of the BP@home initiative. Effective project management and patient engagement strategies aligned with these constructs, and both are recommended to overcome barriers. Additionally, the ’Executing’ and ’Reflecting and Evaluating’ constructs of CFIR suggest the need for continuous monitoring and adaptation of the initiative based on feedback from HCPs and patients. Our experience highlights that there was little time to reflecv and evaluate in real time during the roll-out of the BP@Home programme at a time of national crisis.

The CFIR framework provides a comprehensive lens to understand the complexities and challenges in implementing the BP@home initiative. Our study highlights the importance of considering all aspects of the CFIR framework–from the intervention’s design to the external environment and individual characteristics–to ensure successful implementation. Future research should focus on exploring these dimensions in other regions and healthcare settings, contributing to a more nuanced understanding of implementing remote healthcare initiatives in the post-covid era.