Co-designing health services for people living with HIV who have multimorbidity: a feasibility study

As a result of combination antiretroviral therapy (ART), HIV has become a long-term condition with an excellent prognosis (Kooij et al, 2015). People diagnosed with HIV now have a near-normal life expectancy, provided that ART is initiated promptly and a high level of adherence to treatment is achieved and maintained (May and Ingle, 2011).

Increasing numbers of people with treated HIV are now being diagnosed with comorbid conditions associated with ageing, including cardiovascular, renal, hepatic, bone, neurological and pulmonary diseases (Petoumenos et al, 2011; Schouten et al, 2014; Langebeek et al, 2017).

Currently, HIV care is usually provided in specialist clinics, with a focus on the prescription and management of ART. Despite excellent HIV outcomes, it is not yet known how healthcare for people living with HIV should be provided in the future to sustain these outcomes and to address the broadening spectrum of HIV infection as the cohort continues to age (Baylis et al, 2017). There is a continuing need for an evidence-based approach to future HIV healthcare that incorporates multimorbidity models and is developed in tandem with the recently formed integrated care systems.

Increasingly, healthcare providers are seeking to develop services in ways that are informed by patient experiences. Experience-based co-design (EBCD) is an established service improvement approach that enables service users and providers to work closely together in the redesign of services with the explicit goal of improving experiences and care quality (Bevan et al, 2007; Bate and Robert, 2007). Although EBCD has been used to improve a number of healthcare areas including cancer services (Tsianakas et al, 2012), emergency departments (Piper et al, 2012) and medical outpatients (Wolstenholme et al, 2010), to the authors' knowledge it has not to date been used in the development of services for people living with HIV or applied to people with comorbid conditions who are accessing a number of different healthcare services.

The participatory approach of EBCD as a service improvement methodology lends itself well to a clinical specialty that has historically worked in partnership with patients. This approach is like participatory research where subjects are involved in both generating and analysing data and are integral to the research process (Vaughn and Jacquez, 2020).

EBCD is typically conducted in six stages (Robert et al, 2015; Point of Care Foundation, 2023):

• Stage 1. Project set-up
• Stage 2. Gathering and understanding staff experiences through non-participant observation and discovery interviews, and feeding back the findings
• Stage 3. Filmed patient interviews that capture patients' journeys through healthcare services from the point of diagnosis onwards, and enable patients to reflect together on their shared experiences
• Stage 4. Bringing staff and patient participants together where this ‘trigger’ film is shown and used to jointly identify service improvement priorities
• Stage 5. Co-design groups comprised of staff and patients work on agreed improvement priorities
• Stage 6. Final celebration and review event.

The filmed patient interviews of stage 3 are a distinctive component of EBCD; these are analysed to identify emotionally significant touchpoints in the patient pathway (Robert et al, 2015). Emotional touchpoints have been defined as positive or negative emotional responses to care across the patient pathway (Robert et al, 2015). Individual patient films are edited to emphasise these touchpoints and are developed into a composite film that includes excerpts from all patient interviews. This composite film is later used as a trigger for change and service improvement (Robert et al, 2015).

EBCD has not previously been attempted in the HIV setting or with patients with multimorbidity. It was therefore not known whether filmed interviews would be acceptable in this context. Moreover, there was no established predefined pathway for people living with HIV with additional long-term conditions.

Project aims

This aims of this study were to:

• Determine the feasibility of using the EBCD approach with people living with HIV who had one or more comorbid condition and with staff from identified healthcare services (HIV, other medical specialties and general practice) involved in their care
• Identify shared staff and patient priorities for future service redesign.

This project was intended to test feasibility and identify priorities for change and did not include co-design working groups (stage 5) or a celebration and review event (stage 6), which are usually undertaken in a full cycle of EBCD.

Methods

This feasibility study was conducted at Brighton and Sussex University NHS Trust from October 2014 to July 2015. Ethical approval was obtained from the NHS Health Research Authority (reference 14/LO/1479). Approval to approach GPs in Brighton and Hove was obtained from the Sussex NHS Research Consortium. The study was conducted in three phases: gathering patient and staff experiences (stages 2 and 3); data validation in individual staff and patient groups (stages 2 and 3); and a workshop with people living with HIV and staff to identify priorities for future service planning (stage 4).

Gathering patient and staff experiences

Staff and patient experiences of services were examined using non-participant observation of clinical areas (Bate and Robert, 2007), patient diaries (Bernays et al, 2014) and discovery interviews with patients and staff (Bridges et al, 2008. Inclusion and exclusion criteria for staff and patients are shown in Table 1.

Non-participant observation of clinical areas and consultations

Six clinical areas (cardiology, renal, rheumatology, liver/digestive diseases, HIV and general practice) were selected. Lead clinicians in each area were approached by the research team and written informed consent was obtained to undertake informal visits within the departments and to approach study-eligible clinical colleagues to participate in observed consultations.

Hospital and clinic databases were then searched to identify patients with HIV with one or more comorbidity who had appointments in one of the five hospital services in the next 3 months. Clinical staff providing care to patients with HIV at one of these appointments were then approached and invited to participate in the study. Once written consent had been obtained from staff, identified patients were approached by their clinicians and invited to participate. Appointments in general practice were identified on an ad-hoc basis by asking study-eligible patients whether they had upcoming appointments and inviting them to take part in an observed GP consultation. Consent was sought from staff and patients for the research team to observe up to two clinical consultations.

Non-participant observations of the clinical areas were undertaken by two researchers (EN and EY) (Tsianakas et al, 2012). A notice was placed in waiting areas informing patients of the research and inviting them to wait elsewhere if they did not wish to participate. The researchers sat in patient waiting areas and made written notes on the environment, patient flow and staff and patient interactions. Since staff working in the HIV department were known to EN, observations in the HIV department were undertaken by EY; all other observations of clinical consultations were undertaken by EN. The process (eg structure of the consultation, medical notes, results and follow-up) and transactional elements (eg style, communication and relationship) of the consultation were documented at the time of observation.

Patient diaries

Patient diaries were used to capture a broader appointment experience in real time with minimal researcher influence (Bernays et al, 2014). Clinic and hospital databases were searched for people living with HIV who had two or more appointments booked in any hospital setting over the subsequent 3 months.

Eligible patients (Table 1) were then approached by an HIV health professional involved in their care and invited to participate. Written informed consent was sought for keeping a diary on up to four appointments. Patients were asked to keep an electronic, audio or paper diary of their healthcare appointments over a 4-8 week period. They were given a template indicating the type of information to be recorded (such as making and travelling to the appointment), experiences during the consultation as well as their thoughts and emotions during these consultations. Participants were contacted every 1-2 weeks by telephone by a member of the research team to deal with any queries or concerns.

Staff and patient interviews

Interview guides were developed based on the analysis of the clinical observations and the diary study. Nurses and doctors were purposively selected for the discovery interviews from the six clinical service areas and invited to participate in a semi-structured, digitally recorded interview.

All staff interviews were conducted by an experienced qualitative researcher (EN) who was trained in EBCD, and took place in a private room in a hospital or clinic. Interviews were audio-recorded and transcribed verbatim by an independent transcription service. Staff interview transcripts were labelled as HIV specialist or non-HIV specialist to provide anonymity.

Patients who had one or more comorbidity were identified through searches of clinical and hospital databases and forthcoming appointments in the HIV clinic as well as through team and multidisciplinary meetings and invited to participate in a filmed interview. Those who agreed to participate gave written consent for their face to be visible on film and for the film to be shared with other patients or health workers for education or health service planning.

Patient interviews were filmed in a private research room at the hospital by EN, who had been trained to conduct filmed interviews using a static camera. Following the interviews, each film was sent by courier to an external editor who produced an individual patient film. This was couriered to each participant, along with a consent form for release of the footage for use in a composite film.

Data analysis and validation

Data analysis was undertaken by VC, EN and EY on the clinical observation notes, patient diaries and staff and patient interview transcripts using inductive thematic analysis (Braun and Clarke 2006). This involved data familiarisation, initial coding, searching and reviewing themes and the final definition and naming of themes using a realist approach.

The data were initially analysed independently by VC, EN and EY. The combination of one clinical researcher (EN) and two academic researchers (VC and EY) enabled an inductive approach and coding was then reviewed as a group and developed into themes. Final naming of the themes was undertaken in data validation as described below.

In addition to the thematic analysis, individual patient films were viewed by three researchers (EN, EY and VC) to identify emotional touchpoints in patients' experience (Point of Care Foundation, 2023). These touchpoints were used by the research team and film editors to develop a composite 30-minute film that illustrated each touchpoint using extracts from individual patient films (Tsianakas et al, 2012; Robert et al, 2015). Film participants were identified by pseudonyms.

Data validation was undertaken in separate patient and staff events. Patients were invited to attend a feedback session to watch the composite film and discuss the identified emotional touchpoints. A visual patient pathway was created by the research team and displayed at the meeting. Patients were asked to place sticky notes on the visual pathway to indicate their emotional responses and experiences at each step along it. The group then discussed, agreed and ranked their priorities for future service improvement.

The thematic analysis of staff interview transcripts was presented to staff participants at a separate meeting. Each identified theme was discussed and agreed, and a consensus reached over a list of priorities for the future development.

Workshop for setting future priorities

Staff and patient participants were invited to take part in a joint stakeholder event, facilitated by EN. Key staff members from the six clinical areas, the appointments service manager, an IT systems manager, the deputy chief nurse and the medical director were also invited.

The group was presented with the findings from the staff interviews then watched the composite patient film. After discussion of the findings, six groups were formed to work together to identify specific actions to address each of the priority improvement areas identified.

The groups then reconvened to reach a consensus on which priorities had the most potential to improve patient and staff experiences. The three most popular priorities were selected to be taken forward for future co-design work.

Results

Patient engagement in EBCD and participation in filmed interviews

In total, 616 patient records were reviewed to identify eligible patients with a booked appointment in one of the designated hospital clinical areas. Seventy-seven patients (12.5%) met the study inclusion criteria (Table 1).

Of those, eight were approached to participate in clinical observation and seven (87.5%) were recruited; seven were approached for the diary study, of whom five (71.4%) were recruited.

Twenty-four were approached for the filmed interviews, of whom 15 (62.5%) were recruited. Reasons for declining to take part in the filmed interview were: not wanting their face to be visible because of concerns about confidentiality or body image (6); or lacking interest in the study (3). Of the 15 patients recruited to the filmed interviews, only 10 participated. Reasons for this were a decline in physical health (one patient was admitted to hospital and another became acutely unwell) and psychological health issues (bereavement: n=2; anxiety: n=1). The overall response rate for recruitment to the filmed interviews was 41.6%. Although patients had given consent for filmed interviews for education and health planning purposes, some who participated (40%) expressed concern relating to public disclosure of their HIV status through the internet or wider access to the film. This led to additional precautions to safeguard patient confidentiality, including the use of pseudonyms and ensuring secure mechanisms to transport individual films.

Of the 22 patient participants, the majority were male (20; 90.9%), 20 (90.9%) were white British and 21 (95.4%) had an undetectable viral load; a total of 110 comorbid conditions were identified (Table 2). The median age was 55 years (range 31–91 years) and the median time since diagnosis was 16.5 years (range 7–25 years).

Five of the 10 film participants (50%) attended the patient feedback meeting. Of all patient participants, 10/22 (45%) attended the final workshop. Reasons given for non-attendance were ill health and having other pre-planned commitments.

Staff engagement in EBCD

All 10 staff who were approached for interview agreed to participate. Four were medical consultants from renal, rheumatology, liver and cardiology departments, two were GPs and four were HIV health professionals who had participated in joint clinics (three medical consultants and one nurse specialist). One of the GPs provided a commissioned local enhanced service for people with HIV.

For the clinical observations, 10 staff were approached, of whom seven (70%) were recruited (three did not respond to emails). Five staff (four medical consultants and one nurse specialist from renal, cardiology, HIV and rheumatology, including one interview participant) took part in seven consultations; two staff consented but could not participate because the patients could not be contacted before the consultation appointment. No paired clinical observations took place in general practice. Staff recruitment was easier in areas where relationships had already been established through joint working.

The joint co-design event was attended by 20 staff who had participated in the clinical observations and/or interviews or who were representatives of the services provided in the study or the priorities raised.

Overall patient and staff feedback on their experience of the co-design process was positive (Box 1).

Box 1.Feedback on the co-designed film and process

Clinical observations

In total, 12 hours of observations were undertaken in four of the six clinical areas, of which four hours were observed consultations. The observation component was pivotal in building relationships with staff and understanding processes in the individual clinical areas. The clinical consultations highlighted issues with separate medical notes for GP, hospital and HIV clinics that were not visible to staff in different clinical areas because of confidentiality, and increased researcher insight into the expectations of HIV patients within other departments.

Diary study

Of the five patients recruited to the diary study, four completed records of 20 (range 3–9) appointments over a 2-month period. One patient had repeated hospital admissions following recruitment and was therefore unable to participate.

The diary study captured patient experiences across a wide range of healthcare settings, including unscheduled care. The diaries highlighted the importance of relationships with practitioners, the expertise of health professionals, access to appointments and interdepartmental communication. Negative experiences included the sheer volume of appointments and staff attitudes towards participants.

Key touchpoints for people living with HIV with comorbid conditions

Analysis of the individual patient films, diaries and non-participant observations identified eight touchpoints (Table 3). These were: managing multiple appointments; telling my story again; HIV services as the benchmark; expertise and knowledge; working alongside healthcare workers; shared medical records/results; communication; and care co-ordination. The composite 30-minute film illustrated and was structured around these touchpoints.

The volume of appointments was a significant issue for patients and many described difficulty in attending different appointments on the same day. Patient relationships with the HIV service and its staff were an overriding theme and seen in all data collected. This appeared to be based on familiarity and feeling welcome and safe in the HIV department which, in turn, made it more difficult for patients to develop relationships with staff in other departments or settings. This was compounded by a perceived lack of knowledge and expertise in HIV in non-specialist staff, particularly with regard to the interplay between HIV and comorbid conditions. Patients strongly valued partnership working with their HIV team and participating in their own care and found this hard at times to replicate elsewhere.

Having separate medical records for HIV was perceived as a barrier to joined-up care compounded by inadequate communication between and within services about individual patient health issues. Patients reported that this placed them in the position of having to co-ordinate their own care, sometimes when they were not well enough to do so.

Improvement priorities for staff and patients

Staff identified three priorities for future service development for people living with HIV who have comorbid conditions (Table 4):

• Care co-ordination was identified as key to future service delivery to support HIV patients with comorbidities to navigate across multiple services
• Access to shared medical records across primary, secondary and tertiary care was considered a priority to improve communication and continuity of care across healthcare teams
• The delivery of streamlined pathways across services and joint clinics was proposed to avoid duplication of investigations and promote easy access to shared test results and clinical expertise.

Patient participants identified six priorities (Table 4). These were: care co-ordination; shared medical records and results; improved communication across sections in primary, secondary and tertiary care; streamlining referral processes to and between specialities; managing multiple appointments; and a holistic approach to managing multiple health conditions.

At the workshop for setting future priorities, staff and patients agreed the following shared priorities:

• Medical records and results systems/information sharing
• Managing appointments
• Care co-ordination and streamlining services.

Discussion

This study aimed to examine the feasibility of using EBCD to assist in the planning of future services for people with HIV who have comorbid conditions. The findings demonstrate acceptable staff and patient engagement with the process and an alignment of staff and patient priorities for future service development. They indicate that the EBCD approach can be applied to the HIV setting and across a comorbid disease pathway.

The aim of EBCD is to capture the ‘seldom heard voice’ of the patient (Tsianakas et al, 2012; Point of Care Foundation, 2023) and one of the key questions in this study was whether people living with HIV would choose to participate in a filmed interview. Although some modifications were made to the traditional form of EBCD, minimal difficulties were encountered in recruiting patients who were willing to be filmed for educational and health service planning purposes. However, there were some reservations about the distribution of the films to the wider public domain.

The concurrent diary study provided a unique opportunity for patients to record their experiences in real time. While the themes identified in the diaries were similar to those identified in patient interviews, experiences of healthcare reported in the diaries were generally more negative and may represent a more accurate lived experience.

The patient voice has historically been integral to the development of HIV services and there are many examples of effective HIV patient advocacy groups and patient involvement models (Namiba and Anderson, 2011; Cairns, 2012; Positively UK, 2017). Studies on patient experiences and values in HIV care are prevalent in both grey and published literature (Weatherburn et al, 2013; Cooper et al, 2016; Miners et al, 2017; Pollard et al, 2018). Health services research by The King's Fund drew upon the experiences of both health workers and patients to inform recommendations for future HIV care provision (Baylis et al, 2017). This EBCD study adds to the HIV co-production field a systematic service design approach to capturing staff and patient experiences of providing and receiving care within the context of multiple comorbid conditions and using these experiences to define joint priorities for service improvement. The composite film was a powerful tool for generating discussion of the patient experience.

In the continued absence of a defined and evidence-based model of HIV care that can sustain current treatment outcomes and quality of care, EBCD can be applied at a local service level to identify and co-design improvements in care and has the potential to inform the design principles for a national model of care for HIV patients who have more complex needs.

Although this study was completed 6 years ago, there has been no cohesive change in the way in which HIV healthcare is commissioned for people living with the virus who have comorbid conditions. The EBCD methodology and priorities for future HIV healthcare identified in this study remain valuable for current and future service configuration.

Previous studies using EBCD have focused on single disease areas or departmental pathways (Wolstenholme et al, 2010; Piper et al, 2012; Tsianakas et al, 2012). This research expanded those parameters by including six departments concerned in the care of HIV patients who have comorbid conditions. It involved developing a multimorbid disease pathway that incorporated access to and care provided by a number of clinical services. Identifying eligible patients with comorbid conditions was initially cumbersome because of a lack of coding systems for multimorbidity at that time.

This feasibility study was conducted at a single site with a caseload of predominantly men who have sex with men. Research with more diverse caseloads may result in different service improvement priorities and this highlights the importance of identifying needs at a local as well as at a national level.

This feasibility study highlighted that EBCD methodology is costly in term of time and resources but the data yield is high with concrete outputs for service improvement and investment from both staff and patient groups. Future consideration could be given to using an accelerated form of EBCD that uses patient films of national as opposed to local narratives to generate service improvement themes in individual places (Locock et al, 2014).

Conclusion

The findings of this study suggest that EBCD methodology is acceptable to people living with HIV with comorbid conditions and staff from across the patient pathway.

Lessons learned from using this methodology could facilitate the wider use of EBCD to implement service improvements in other HIV settings. Patient and staff priorities for future HIV healthcare were focused on a joined-up approach to care, information sharing and the streamlining of services across HIV, general practice and other specialist areas.

KEY POINTS

• Multimorbidity is common in people living with HIV and new approaches to care are required to meet their needs and sustain excellent HIV outcomes
• Patients living with HIV want joined-up care and streamlined services
• Experience-based co-design is a service improvement approach that can be adapted to people living with HIV who use multiple healthcare services
• Shared medical records and shared access to results would greatly improve the experience of people with HIV who have multimorbidity

CPD reflective questions

• Reflect on how you involve or consult your patients on improving their healthcare services
• Consider how you work in partnership with other agencies to provide a joined-up approach to care
• Reflect on how you can best support patients with multimorbidity within your role