Qualitative outcomes

The following selected quotations were informative in driving changes to the proposed care pathway for remote monitoring of BP:

“You can spot trends, I wore a heart monitor, especially post-surgery

But the downsides are—how do you motivate people, is it anxiety provoking”—Participant 3

“I’m really interested in what stress and anxiety can do to your body but what about doing it everyday obsessively and a bit more than I need to. At what point should I be making contact with my GP.”—Participant 8

These quotes were indicative of the enthusiasm for remote monitoring that the group had and their inquisitiveness about how such their BP data would be shared with their usual care team. However, there was an emphasis additionally placed on safeguarding required to mitigate against over-measurement and the possible anxiety that elevated readings could create, though this was balanced with points made about elevated readings being positively viewed, in terms of being a motivating factor to engage in lifestyle behaviours such as exercise to reduce BP.

As a result of the group discussions, the following changes to the remote BP project were made:

• Adjusting the pathway to share BP data and feedback with more members of the care team
• Refinement of the data entry system to include app-based reminders and advice on when to recommend initiating further discussions
• A proposal to incorporate elements of counselling and support at the time of diagnosis

In terms of the research proposal, discussions focused primarily on the ethics and logistics of delivering a clinician-facing digital alert as an intervention within a research study. A selected quotation with regards to waiver of consent was:

“seems the most practical way of avoiding any behavioural effects that come from knowing you are in a research trial” Participant 8

This quote summarised the consensus view of the group in terms of how to balance the feasibility and safety of doing research studies of electronic alerts aimed at clinicians which seek to change their behaviour and downstream management strategies that can affect patients, In this regard, the group understood that thresholds for tolerating waived consent may vary case by case. There was a majority consensus for waived consent in the context of comparing existing non-pharmaceutical interventions such as fluid restriction versus no restriction, given the patient is the final arbiter of oral fluid intake. Support for waiver was not unanimous though and so additional review and communication strategies were sought by the trial management group because of the diverse and helpful patient feedback.

Quantitative outcomes

For the research proposal, different models of consent were rated on a 5-point Likert scale for suitability and responses were individually completed following discussion (Table 2). The average suitability for each consent model was: waiver of consent: 4.5/5; research registry: 3/5; verbal consent: 2.7/5. A significant amount of missing data was noted for participants who attended remotely.

Download:

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

Table 2. Patient views on different models of consent for the research proposal.

Participants (i) to (iii) were those who attended in person. First, participants were asked to rank the models of consent in order of preference (1 = most preferred, 3 = least preferred). Next, participants were asked to rate each model out of 5 for suitability (1 = not at all suitable; 5 = entirely suitable).

https://doi.org/10.1371/journal.pone.0278260.t002

Table 3 additionally summarises views collected on patient reported outcome measures (PROMs). Data was only available for the three participants who attended in person.

Download:

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

Table 3. Participant views on PROMs.

Participants were asked to what extent they agreed with specific statements on how patient reported outcome measures (PROMs) may be used (1 = not at all, 5 = very much agree).

https://doi.org/10.1371/journal.pone.0278260.t003

Feedback on PPI event and impact of context

The event was held on December 3^rd, 2021, and strict Covid-19 protocols were observed for those who attended in person. The hybrid format proved challenging in relation to timings observed for the day. Whilst the majority of participants (6 out of 9) were logged in remotely, ensuring an equitable distribution of time was difficult though partly helped by all participants having webcam access. Table 4 highlights feedback collected from 8 participants after the event with similar mean ratings between in person (overall rating 4/5, interactivity 4/5) and remote attendance (overall quality 4.5/5, interactivity 4.8/5).

Download:

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

Table 4. Summary of rated quality and interactivity of the day from participants.

https://doi.org/10.1371/journal.pone.0278260.t004

Comparison with literature

We are unaware of existing published reports of hybrid PPI events which not only examine digital tools as a theme, but which used such tools to allow the safe conduct of the day. Whilst reducing commute times and accommodating those who are risk averse in the pandemic era are advantages, there is the caveat of ensuring factors such as digital exclusion [7] do not reinforce biases and underrepresentation of key groups.

Thus systematic design and delivery of PPI will require dedicated focus to equality, diversity and inclusion, particularly as upstream PPI begins to shape both research (including improving recruitment to clinical trials [13]) and routine service design (such as increased participation in co-design).

When compared to the distribution of current PPI events, [14] it is unclear if participation levels are representative of patients affected by each condition or treatment studied, or if the distribution of known and published PPI events match up to patient priorities more generally.

Limitations

Our findings should be considered in light of several limitations. First, our sample size was small and there is the risk of ascertainment bias, given that recruitment of patients occurred through a limited number of sources. We mitigated this in part through a pre-event survey but nevertheless the views of the focus day may not be representative of the wider population, something that all PPI is at risk of [15].

Second, our event was time limited–at 4.5hours in duration including breaks, we sought to balance factors such as breadth and depth of coverage as well as risk of fatigue. Written feedback collected for the day noted that timing was tight and the afternoon session in particularly was abbreviated to adhere to time, ensuring that for in-person attendees would not miss their return travel home.

Third, the data collection form used to quantify views was a 1–5 Likert scale. We elected to forgo using a formal validated measuring instrument in favour of ease and speed of use. As a pilot event, we regarded the data collection exercise as more exploratory in nature.

Fourth, we did not use a formal framework to support our PPI day design. This was motivated by pragmatism, with existing evidence to suggest that most recognised frameworks have poor useability [16]. However, there remain questions regarding how best to harness PPI for implementation-focused clinical and research questions and previous work in the field has offered suggestions which require further exploration [17].

Future directions

The optimal way to harness PPI to help implement digital technologies and to inform how we test them as part of pragmatic research, [18] remain areas for further work. PPI efforts will be central to the maintenance of trust in their use. [19, 20].