Improving the experience of older people with colorectal and breast cancer in patient‐centred cancer care pathways using experience‐based co‐design

Abstract Background Patient and public involvement (PPI) in quality improvement of oncological care pathways for older patients are rare. Objectives Improve the care pathway experience of older cancer patients and explore lessons learned regarding how to engage this vulnerable group. Design Experience‐Based Co‐Design. Setting and participants: Older cancer patients, their caregivers and healthcare professionals within colorectal and breast cancer care pathways. Interventions: Co‐design quality improvement teams. Main outcome measures: Colorectal cancer care pathway touchpoints were (a) availability of a contact person during diagnostic, treatment and aftercare phases; (b) collaboration between physicians and different hospital departments; (c) continuous relationship with same physician; (d) respectful treatment; (e) and information transfer with primary care. Breast cancer care pathway touchpoints were (a) comprehensive information package and information provision, (b) care planning based on patient preferences, (c) continuity of patient–professional relationship and (d) specialized care in case of vulnerability. Challenges related to PPI included (a) ability of older cancer patients to be reflective, critical and think at a collective level; (b) gaining support and commitment of professionals; (d) overcoming cultural differences and power inequalities; and (e) involving researchers and facilitators with appropriate expertise and position. Conclusion This multidisciplinary quality improvement project revealed several challenges of PPI with older cancer patients and their caregivers. Research teams themselves need to assume the role of facilitator to enable meaningful PPI of older cancer patients. Patient or Public Contribution Patient and caregiver representatives and advocates were involved in the design, conduct, analysis, interpretation of the data and preparation of this manuscript.


| INTRODUC TI ON
The total number of patients with cancer is increasing, in part due to earlier diagnosis and improved cancer treatment. 1 In the Netherlands, for example, the 5-year cancer survival rate has increased from 45% in 2000 to 65% in 2018. 2 Cancer patients are an ageing population: in 2019, half of all new cancer patients in the Netherlands was 70 years or older. 2 Most cancer patients in the Netherlands receive their care through multidisciplinary oncological clinical care pathways. In these pathways, patients are treated by a multidisciplinary oncological team according to national guidelines, to improve coordination and continuity of patient-centred care. 3 Care pathways have traditionally been developed by healthcare specialists, from a disease-based perspective. 4 Patient involvement in the development of care pathways is increasing. 3,5 Patients have been acknowledged for their unique experiences that can contribute to research-based quality improvement. 6,7 Funding bodies require patient and public involvement (PPI) such as Control in Australia. 8 In this study, we define PPI as the contribution of patients and their family caregivers in improving healthcare services by active involvement in a range of activities that combine experiential and professional knowledge. 9 In the Netherlands, approximately 200 disease-specific patient organizations are united within the Dutch Patient Federation. 10 Patients, then, assume the role of patient representatives, telling a collective story to bring forth their concerns. A large number of patients, however, do not organize themselves, such as older cancer patients.
To date, several reviews have been published on PPI of older people. [11][12][13][14][15] Baldwin et al 15 [16][17][18][19] ; however, older cancer patients have primarily been involved in oncological end-of-life research. 20 Bombard et al 6 conducted a systematic review with the objective to identify the strategies and contextual factors that enable optimal engagement of patients in the design, delivery and evaluation of health services. Only two out of the 48 studies investigated cancer services. Overall findings show that low-level engagement, mainly consultative unidirectional feedback, had an impact on discrete products (eg policy and planning documents and governance). Highlevel engagement, mainly partnership approaches or co-design, had an impact on care process or structural outcomes (eg collaboration and mutual learning and negotiating for service change). One such partnership approach is experience-based co-design (EBCD): patients, family carers and professionals share their experiences, identify and agree on improvement priorities and work together to achieve them. 7 EBCD has been applied successfully to improve several cancer services in acute care, including breast, 21,22 lung, [21][22][23][24] head and neck, 25 and gynaecological cancer 22 ; in acute and community care for adolescent and young adults with cancer 26 ; and in community care in an outpatient oncology care centre. 27 Two studies described palliative care of older patients with cancer in emergency care. 28,29 All these studies show that there remain substantial areas for improvement on areas such as communication and interpersonal skills, patient and caregiver information, intra-and interorganizational continuity of care, collaboration among (medical) specialists, age-specific subjects including fertility, cancer-specific subjects such as cosmetic concerns and service-specific subjects including day care. Despite the increasing number of published studies on PPI using partnership approaches, currently, studies involving older people with cancer are limited, 30,31 despite older patients constituting a majority in current clinical practice. 14,15,32 The aim of this study was twofold: first, to improve the cancer care pathway experience of older cancer patients and, second, to explore lessons learned regarding how to involve this vulnerable group.

| Design
We used a participatory approach, where PPI is applied as partnership approach using the experience-based co-design (EBCD) methodology. 7,33 In this qualitative study, the subjective experiences of older cancer patients, caregivers and healthcare professionals are the starting point for quality improvement.

| Setting
The study was conducted at Zuyderland Medical Centre (MC) in Heerlen, one of the largest urban teaching hospitals in the Netherlands. This hospital is located in one of the most ageing regions in the Netherlands, where 21% of the population consists of people older than 65 years (17.4% is the Dutch average). In 2004, Zuyderland MC signed a manifest called 'contract with society' as a promise to invest more in patient-centred health care. 34 We chose the colorectal and breast cancer patient pathways because these are

| Participants
In this study, there were three target groups: older people with colorectal or breast cancer (n = 24), their caregivers (n = 24) and healthcare professionals (n = 32) ( Table 1). Inclusion criteria for older people with cancer were age 65 years or older, receiving treatment in the hospital's colorectal or breast cancer pathway (in the diagnostic, treatment or aftercare phase), a life expectancy of more than 1 year and somehow vulnerable, such as having a small social network. We did not apply specific exclusion criteria because of the participatory nature of this study. Older people with cancer were purposively sampled and approached by the nurse (case manager) or an oncologist. All older people with cancer were asked to identify a caregiver-their spouse, relative or a friend-most involved in their care. The mean age and age range of the older people with breast cancer and colorectal cancer was 71.9 [65-80] and 73.9 [65-88] yrs., respectively. In the colorectal group, eight partners, one son and three daughters participated, and in the breast cancer group, there were six partners, three daughters, one son, one brother and one friend.
Healthcare professionals included 12 physicians (oncologists, surgeons, radiotherapists and general practitioners), 10 nurses (clinical nurse specialists, nurse practitioners, nurses from various departments and home care nurses) and 10 allied medical professionals (physiotherapists, dieticians, pharmacists and psychologists).
The single inclusion criterion was involvement in the colorectal or breast cancer care pathways. Excluded were those who were not mentioned by the patients in the interviews, as they were likely less significant in terms of the experience of older people with cancer.

| The research team and PPI
Our research team consisted of two project leaders: a nurse researcher (AM) and a haemato-oncologist (KSJ), one patient and caregiver advocate (ED), two patient and caregiver representatives (MYV, ES) who were staff members of Zuyderland MC, a patient umbrella organization covering the southern region of the Netherlands, and two junior researchers (IM, AH). The patient and caregivers' advocates and representatives acted as patient facilitators to support older cancer patients in their involvement as needed. In addition, also involved was an advisory board consisting of a representative of

| Phase 1
In Phase 1, we gathered the stories of the patients and their caregivers at the participants' homes, using in-depth discovery interviews, each lasting between 45 and 120 minutes 7 ( Figure 2 diagnosis, treatment and aftercare. These video and voice recordings were used in phase 3.

| Phase 2
In Phase 2, we gathered the positive and negative experiences of the healthcare professionals involved in the two care pathways by focus group discussions (

| Phase 3
In Phase 3, older people with cancer, caregivers, and professionals prioritized the various touchpoints: first within each of the three groups separately (older people with cancer, caregivers, and professionals), and then with all groups together. Each prioritising meeting started with a presentation of the various touchpoints, illustrated by written or audiovideo taped quotations from the previous phases. Subsequently, one collective experience map was generated. After this, the participants prioritized the touchpoints which needed to be improved, individually with adhesive notes on the wall so that they became 'collective' touchpoints within the group. Each participant could assign three, two or one adhesive notes, assigning three to what they considered the most important touchpoint. The items with the most adhesive notes were prioritized. The last meeting ended with a consensus shortlist of touchpoints which needed to be improved. We also applied the threetwo-one adhesive note system in the mixed groups, and the shortlist was composed of the items that received the most adhesive notes. The facilitators and barriers to patient engagement were noted in detail.

| Phase 4
In Phase 4, we formed five co-design quality improvement teams involving older people with cancer and caregivers together with healthcare professionals to design and implement quality improve- The total numbers of participants per phase could vary because the number of patients, carers and healthcare professionals in each phase was different.
facilitated the teams. In each quality improvement team, at least two older people with cancer and two caregivers participated. We made extensive notes on the challenges and facilitating factors in the codesign quality improvement teams.

| Data analysis
All interviews and focus group discussions were video-or audiotaped and transcribed verbatim. Field notes were written down during the observations of all meetings. We analysed all data using open and axial coding strategies and comparative analyses. 38 The data were read to identify text passages relevant to the research question. Each relevant line of the interview data was coded using open codes that were often descriptions used by the respondents. Next, codes were grouped into subcategories and categories that best characterized the data collected. Throughout the analysis process, the codes, subcategories and categories were constantly compared and contrasted within and among the data. Memos were written about these codes and (sub) categories and of the analysis process itself. These were examined for links and connections to further the analysis. The analysis resulted in touchpoints, improvement priorities and outcomes and the four lessons learned. We derived categories grounded in the data, and also based on the six topics required by the funding body. Three members were involved in the analysis. In the event of different interpretations of text fragments and different codes assigned, the original quote was reviewed to capture the intended meaning. After discussing the meaning, an agreed-on code was assigned. Qualitative data analysis software Nvivo (version 10 for Windows) was used to process the data.
We stopped the analysis (and data collection) after we reached data saturation. We determined data saturation by the degree to which new data repeat what was expressed in previous data meaning that data were replicated in the interviews. 39

| Trustworthiness
We sought to safeguard credibility and transferability. 40 Credibility was ensured using various kinds of data (cancer patients, caregivers and healthcare professionals), methods (discovery interviews, focus group discussions and field notes) and investigator triangulation (see research team section above). In addition, we member-checked the transcript. Furthermore, 'thick descriptions' of the context of the particular patients, and the social settings in which the data were gathered, were used to inform other researchers about the extent to which the findings are transferable to other contexts.

| Ethics approval and informed consent
The project was approved by the Ethics Commission of Zuyderland MC. Participants received information before the start of the study were allowed to ask questions and signed an informed consent form.

| FINDING S
We gained rich data and reached data saturation, with a sample size that is consistent with several other studies using EBCD. 26,27 The number and characteristics of participants and the background of the healthcare professionals involved (per study phase) are shown in Table 1 and Table 2. In the following subsection, we present the key touchpoints, improvement priorities and outcomes for each cancer care pathway and our reflection on four challenges of involving older people with cancer as equal partners.

| Touchpoints, improvement priorities and outcomes: colorectal cancer care pathway
Older people with colorectal cancer, caregivers and healthcare professionals identified and prioritized several touchpoints. What we do is that we call the family practitioner: "Your patient has had surgery and will be discharged in a few days." But we do not have a well-designed procedure or discharge-service. Because, imagine Respectful treatment Interpersonal skills and service development 1. Establishment of shared outpatient clinics between surgeon and oncologist to support access to the same doctor. 2. Current information for older people with cancer and caregiver about the waiting time in the outpatient clinic. 3. Implementation of a working process that allows the doctor to prepare the patient case beforehand and not during the outpatient contact. 4. Establishment of a rapid-response work procedure for nurses at the oncology unit to respond timely to emergencies 5. Weekly lunch meeting at the oncology unit to reflect on current bottlenecks and facilitate instant quality improvements. 6. Implementation of two fixed time slots for discharge per day to appoint precise pick-up times for caregivers. 7. Involvement of caregivers in the intake procedure at the oncology unit to provide relevant background information.
Information transfer with primary care Information transfer from and to primary care 1. The discharge information of the cancer unit is also forwarded to the home care nurse (next to family physician). 2. Introduction of a checklist for nurses in addition to discharge information: older people with cancer and caregivers will receive a copy of the checklist 3. Introduction of an information sheet where family physicians provide patient-relevant information (eg Do not resuscitate wish) at the point of admission. This information is used in the multidisciplinary team meetings on individualized treatment decisions. Healthcare professionals expressed the desire to obtain patientspecific context information before making medical decisions, such as the availability of the social network or do-not-resuscitate orders.
Based on the prioritization, four co-design quality improvement teams were formed. The prioritized touchpoints 'collaboration between physicians and different hospital departments' and 'continuous relationship with the same physician' were combined in one co-design quality improvement team because the focus was similar.
In total, 14 co-design quality improvement team meetings were held, and together they worked on 16 outcomes, which were incorporated into care processes (Table 4).

| Touchpoints, improvement priorities and outcomes: breast cancer pathway
Older people with breast cancer, caregivers and healthcare professionals identified and prioritized several touchpoints ( Table 3).

| Comprehensive information package and information provision
Older people with breast cancer, caregivers and professionals experienced that information material was not always up-to-date, and some parts were missing, such as information regarding aftercare.
Family members experienced that they sometimes received information twice or some necessary information, not at all. In addition, professionals in the care pathway provided individual information packages as stand-alone packages. Information was provided in multiple leaflets. There was a need to streamline the information with respect to content and timing, for themselves and the caregivers. Professionals were mostly concerned about providing the necessary information about the treatment they provided, and they perceived as relevant. Professionals said that it was hard to provide patient-centred information just-in-time across the pathway, especially for older people who suffered from multi-morbidity. They perceived that older people with breast cancer did not get a comprehensive care plan which took into account their personal preferences and needs.

| Continuity of patient-professional relationship
The patients and professionals stressed the importance of having the same healthcare professional during their entire treatment or informing the patient beforehand when they would see another professional during their patient journey.
A personal relationship, because the doctor knows more about you as a person, than what can be found on the electronic health record. That gives you more confidence, because you are sure that you are being treated in a way that suits your personal situation.

(Patient 15)
In a multidisciplinary setting, the professionals stressed that it was inevitable that the patient would see different physicians during their patient journey (ie a surgeon, a radiotherapist or an oncologist).

| Specialized care in case of vulnerability
Professionals especially experienced a lack of care for the vulnerable older patient with cancer. They acknowledged that they increasingly were treating a new patient group aged over 80, and missed dedicated geriatric services. In the breast cancer pathway, the prioritized 'collective touchpoints comprehensive information package and information provision' and 'care planning based on preferences of patients' were combined in one co-design quality improvement team. Older people with cancer, caregivers and professionals perceived that these two areas for improvement were interrelated. For the touchpoints 'continuity of patient-professional relationship' and 'specialized care in case of vulnerability', no co-design quality improvement teams were organized because at that time the breast cancer pathway-team lacked sufficient manpower, time and suffered from heavy workload to establish another quality improvement team. In total, three co-design quality improvement team meetings were held, and together they worked on two outcomes, which were incorporated into care processes (Table 4).

| Reflection on main challenges
3.3.1 | The ability of older people with cancer to be reflective, critical and think at a collective level It was striking to notice how the older people with cancer struggled to think at a collective level, to adopt a helicopter view and to go beyond their personal experiences. Several older people continued to share their personal experiences, without being able to translate them into general arguments, which impeded the progress of the meeting.
Although we thought that giving them preparatory homework in advance of the co-design quality improvement teams might better prepare them, their unfamiliarity with reading documents made them lag even further behind the professionals. Eventually, the facilitator decided to prepare the patients face-to-face without written documents, half an hour before the start of the meeting with the professionals.

| Gaining support and commitment of the professionals
The approval and cooperation of the professionals were needed to make PPI a success. Although several nurses and the general prac- This inequality between participants was further reinforced when the professionals wore their white uniforms. In the co-design quality improvement teams, professionals found it difficult to avoid predetermined solutions, rather than creating improved methods with the patients. When the team discussed the lack of a central contact point for the patients, the professionals initially saw no need to discuss this point anew with the patients, as they already had found a predetermined solution: assigning a dedicated team of nurses as a central contact point. The patients' solution, however, was economically more advantageous, as they just wanted a business card with phone numbers for consultation. This was surprising for the professionals, as they had never understood the 'real' problem from the older people's perspective, although they had been convinced they did.  Notably, the question of inequality in partnerships did not seem to be a problem for older people with cancer. They considered much more important than equality in the teams was the extent to which they were able to form a team and a community feeling based on mutual respect. Our participants deemed it essential that they identified each other, realized that they needed each other and that they had a common goal on which to focus. We should, therefore, focus more on discovering the individual personal and professional strengths of all participants to better 'match older people's skills, expertise and motivations to appropriate roles' and use those strengths at the right place and time in research projects. 15  and personal attributes that they will need to have to fully make use of the potential of PPI. It also helps to build a relationship, reflect on roles, responsibilities and expectations and to have a dialogue on their participation preferences. Our findings show that getting a close match between the requirements of the task and the individual performing the task is really important.

| D ISCUSS I ON AND CON CLUS I ON
Although we wanted to create a mindset of equal partnership, which is an enabler of PPI, 15 we did not manage to escape the disciplinary power of the professionals. We realized along the way that we researchers had to adapt to the social norms of the healthcare professionals, as their involvement-being the traditional power holders-was essential to ensure PPI success. Although PPI has given more power to older people with cancer, its success depends entirely on the commitment of the professionals as the traditional power holders. The Dutch National Fund for Health Research, ZonMw, has recently made PPI a prerequisite for funding (ZonMw, 2019). The external motivation of PPI, however, might result in insincere assurances. 15 What we did notice is that once professionals had become involved in the prioritization phase (and beyond), they were less likely to drop out along the way. Hearing and seeing the emotions of the older people with cancer and their caregivers on video or in voice recordings 7 made them committed witnesses. Professionals, mostly nurses, who already wanted to initiate change among hospital management, now have the patients' and caregivers' stories as testimony. In this way, these stories enabled them to become change agents for better care. 41 Another limitation is that in the relatively short timeframe of the project we were not able to set up all co-design teams. In the breast cancer pathway for two touchpoints, no co-design quality improvement teams were organized. This limitation has impeded observing the quality improvement cycles and studying facilitators and barriers in this specific moment in the implementation of quality improvement. We also did not further explore how to deal with staffing and high workload and its impact on the EBCD approach.
Although we wanted to involve a group that is vulnerable, hard to reach and underrepresented, 33 older people with cancer such as those aged over 90 years, those with multiple disabilities or the very ill, did not want to participate or were simply unable to do so. Their voices remain unheard.
In conclusion, the project revealed several challenges of PPI for older people with cancer, caregivers and professionals in multidisciplinary quality improvement. Future initiators of participatory research projects should not let the inherent inequality among them detract from implementing PPI to improve the quality of cancer care pathways.
Instead, they should strive for meaningful PPI, to better utilize the older people's personal skills, expertise and motivations for appropriate roles. Research teams (which should include patient and caregiver advocates and/or representatives) need to take the role of facilitator to enable meaningful PPI of older cancer patients and caregivers.

ACK N OWLED G EM ENTS
Our acknowledgements go to the patients, their caregivers and healthcare professionals who participated in this study.

CO N FLI C T O F I NTE R E S T
The authors declare that there is no conflict of interest.

AUTH O R S ' CO NTR I B UTI O N S
AM and KSJ had significant involvement in the design, analysis and interpretation of data and writing the manuscript, IM was involved in the acquisition, execution, analysis, interpretation of data and writing the manuscript; MV, ES and ED contributed as patient and caregivers representatives (MV,ES) and patient advocate (ED) and had a major role in the execution of the study, analysis and interpretation of data. All the named authors agree to take accountability for the integrity and accuracy of the work and have read and approved the final manuscript.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data available on request due to privacy/ethical restrictions. Kon-Siong Jie https://orcid.org/0000-0002-7471-0094