For the development of the PAOLI, the EQUATOR (Enhancing the QUAlity and Transparency Of health Research) toolkit for developing a reporting guideline (Moher et al., 2010) was followed. This involved the following stages: (1) gathering evidence from a scoping review [7], (2) conducting interviews with PWA and stroke on the topics to be included in the guideline, (3) undertaking a thematic analysis of the in-depth interviews [9], (4) a two-round Delphi survey for item/statement selection and (5) an experts’ consensus meeting to finalize the guideline. PAOLI was registered as a ‘reporting guideline under development’ on the EQUATOR Network website in June 2022 (https://www.equator-network.org/library/reporting-guidelines-under-development/reporting-guidelines-under-development-for-other-study-designs/ [accessed 10 December 2022]).
The current work builds on the scoping review evidence which previously identified the need for the creation of a functional methodological framework to facilitate patient-centred research design for PWA [7]. Specifically, twenty published studies, from the review, on the involvement of PWA in the creation of quality-of-life assessment tools, revealed significant limitations around the reporting of the context, and the process of PPI, deterring the interpretation of PPI impact [7]. The absence of a standardized approach for conducting and reporting PPI in aphasia qualitative research prompted the creation of the PAOLI reporting guideline, in close collaboration with PWA.
To begin the PAOLI development process, the Dialogue Model [10], a multi-phased PPI scheme, based on the methodology of the Responsive Evaluation [11, 12] and the Interactive Learning and Action approach [13], was selected. The rationale for selecting the Dialogue Model for research agenda-setting, was that it enables an equal partnership to be created between all stakeholders (including patients) during the research process. The first author (MC), the lead investigator, served as the facilitator. The Dialogue Model has six phases: Exploration, Consultation, Prioritization, Integration, Programming, and Implementation. These phases are presented in Table 1 with a description of the actions related to the development of the PAOLI guideline.
Table 1
Overview of the six phases of the Dialogue Model as implemented during the PAOLI development.
Phase
|
1. Exploration
|
2. Consultation
|
3. Prioritization
|
4. Integration
|
5. Programming
|
6. Implementation TBA
|
Aim
|
To create good communication conditions and to gain the first understanding of the participants’ issues.
|
To identify topics from patient participants
|
To prioritize topics for the pilot draft of the PAOLI
|
To discuss results with stakeholders
|
To keep stakeholders engaged
|
To implement the PAOLI guidance on a research project
|
Research agenda
actions
|
1.Made contacts with patients’ organizations:
• Association Internationale Aphasie
• Stroke Alliance for Europe
• World Stroke Organization
• Aphasie Suisse
• Cyprus Stroke Association
• AVC Portugal
|
1.Semi-structured Interviews:
8 people with stroke
• 4 with aphasia
• 4 without
from 6 European Countries: Cyprus, Denmark, Greece, Switzerland, France, and Portugal
The research team involved a PPI partner (AK)
|
Creation of the pilot draft of the PAOLI:
• Phases of the Patient and service user engagement
(PSUE) framework.
• Key topics from the thematic analysis
• Recommendations from European Alliance of Associations for Rheumatology (EULAR) guidance
• Stages of the research cycle from the Enhancing the Quality of User Involvement in care Planning (EQUIP) study
• Items from the Guidance for Reporting Involvement of Patients and the Public (GRIPP2) checklist
|
1.Experts’ consensus meeting
(n = 12 participants)
• PPI Partners (AK & JRS)
• stroke survivor without aphasia
• PPI Aphasia Academic Experts
• Academic Aphasia Researcher
• Aphasia Rehabilitation Clinician
• Stroke Advocate from WSO
• Policy Maker SAFE
• Author of EULAR PPI guideline
• MC
|
1.The programming phase has already been started in previous phases (1–4)
|
1. It is recommended that PAOLI is tested in a future PPI stroke and aphasia qualitative study
|
|
2. Research team established: Involved leading investigator (author MC), PPI partners (AK & JRS) and communication partners (family members & speech and language therapy students)
|
2. Thematic Analysis resulted in four themes:
• Restrictions: the kinds of restrictions that make involvement in research difficult
• Involvement: the preferred levels and ways of involvement during the research process
• Support: the support required for active and collaborative involvement
• Impact: Impact of their involvement and how it benefits the study’s outcomes.
|
2. Cognitive Interviews: followed Consensus-based Standards for the selection of health status Measurement Instruments (COSMIN) guidelines to test the content validity on:
• Comprehensibility
• Comprehensiveness
• Relevance and importance of the items
n = 2 people with chronic aphasia
Made suggestions on:
• Vocabulary simplification
• Reducing the complexity of the statements
• Selection of pictograms
• Addition of a statement
|
2. Consensus
• n = 66 statements were confirmed
• n = 4 statements were discarded
• AK & JRS expressed their views and gave feedback on their involvement
|
2. MC was updating stakeholders with controlled feedback reports and email
|
2. Create a shorter version of the PAOLI to be used when patient involvement is secondary to the scope of the study
|
|
3. Informal conversations and preliminary discussions with patient partners and the representatives of the stakeholders:
• Informed and motivated potential participants of the PAOLI project
• Creation of commitment to be involved right from the start and signed informed consent
|
3. Published open-access peer-reviewed article on thematic analysis with AK as co-author
|
3. Pilot tested the Delphi questionnaire online format with a person with chronic aphasia.
Made suggestions to:
• Provide more simplified instructions on how to access, complete and submit the questionnaire
• Indicate a clearer deadline for submission
|
3. Validation of the PAOLI
The final form of the guideline was examined in collaboration with the AK & JRS
n = 3 statements were re-worded
|
3. MC drafted the article with AK & JRS as co-authors
|
3. Patients Organizations (Cyprus Stroke Association, SAFE, Aphasie Suisse) to disseminate the PAOLI guidance in meetings and informal networks
|
|
4. The PAOLI Consortium was established:
• People with aphasia after a stroke
• Stroke survivors without aphasia
• PPI Aphasia Academic Experts
• Academic Aphasia Researchers
• Aphasia Rehabilitation Clinician
• Clinical Psychologist with PPI experience
• Stroke Policy Maker
• Aphasia Advocate
• Stroke Advocate
• PPI Guidelines Author/ Policy Maker
• Research Funding Agent
|
4. MC, AK and her communication partner (SLT student) searched the literature on published PPI frameworks, the internet fora and environmental scan.
|
4. Delphi survey
participants: PAOLI Consortium
(n = 23 panellists)
2-round Delphi survey
Result: n = 66 statements + n = 17 topics voted at an agreement level of > 80%
Consensus levels were the highest to the need of involving people with aphasia in the research team from early on and identifying topics using co-design methods
|
4. It is recommended that PAOLI is also reviewed and approved by an external board
|
4. MC and AK presented preliminary data at the European Life After Stroke Forum of the Stroke Alliance for Europe
|
4. The research team to co-present the PAOLI guidance in scientific conferences (World Stroke Conference, International Aphasia Rehabilitation Conference, World Congress of Neurorehabilitation)
|
Phase 1. Exploration
The Exploration phase involved contacting patient organizations, such as the Association Internationale Aphasie, the Stroke Alliance for Europe, the World Stroke Organization, to inform potential stakeholders about the PAOLI project, recruit participants, build the PAOLI consortium (see Τable 1) and establish the research team.
The research team
The research team consisted of 7 individuals: a leading investigator, 2 PPI partners and 4 communication partners. The first author (MC), a senior speech and language therapist practising in aphasia rehabilitation, with previous PPI experience in aphasia research1 served as the lead investigator [7, 9, 14]. The two PPI partners are co-authors AK and JRS. The term ‘PPI partner’ will be used to reflect the constant commitment and active involvement of the two people with chronic stoke-induced aphasia throughout the research process. The aim for collaborating with AK and JRS was to enhance the quality and transparency of the PPI evidence in this study, by involving them from the beginning and in all stages of developing PAOLI. AK is a 36-year-old female with mild-moderate aphasia and a right hemiparesis, both resulted after a hemorrhagic stroke 8 years ago. AK holds a Master of Science (MSc) in Social Research Methods from the University of Sussex and was enrolled in doctoral studies in Social Care but dropped out after her stroke event. She has previous experience as a PPI partner on research projects, and has co-authored publications from the research [9, 14]. JRS was Professor of English Linguistics at the University of Lausanne and is now retired. He suffered a stroke in 2009 and presents with mild-moderate aphasia and a right hemiplegia. JRS holds a PhD in Linguistics from the University of Cambridge, has a vast background in research as primary investigator, on several projects, and has authored aphasia-related research publications [15–17]. Both AK and JRS were actively involved in all stages of this study, from the conceptualization of the study to patient recruitment and to the final validation of PAOLI. Finally, the research team included four laypeople the so-called communication partners (family members, speech-language therapy (SLT) students) who supported PWA from the onset during the informed consent procedures, the completion of the cognitive interviews, the Delphi survey, and the validation phase.
Phase 2. Consultation
The Consultation phase involved three steps: (1) conducting 8 semi-structured interviews with people with chronic stroke, four with aphasia and four without, (2) a thematic analysis of the in-depth interviews obtained for the identification of key topics to be included in the pilot draft [9] and (3) a literature search on published PPI frameworks [8].
Phase 3. Prioritization
The research team (RT) consulted several published guidelines [8, 18] to become familiar with recommendations from previous research on the conceptual elements required to build a new guideline. Specifically, the RT accessed the following:
-
the patient and service user engagement (PSUE) framework [19] on how to develop the research phases.
-
the European Alliance of Associations for Rheumatology (EULAR) guidance [20] to determine topics on conceptualization, researching consensus and co-design methodology.
-
the research processes from the Enhancing the Quality of User Involvement in care Planning (EQUIP) study [21, 22] to gather information on patient recruitment and induction procedures.
-
the Guidance for Reporting Involvement of Patients and the Public (GRIPP2) checklist [6] on selected material for the proposal development and impact.
All this information combined, resulted in the concept development of the PAOLI leading to the adoption of 99 statements and the creation of the first pilot draft.
Pilot Draft
The pilot draft of the PAOLI encompassed the following components:
-
Four phases: Foundation, Development, Translational and Ongoing Processes.
-
18 items [9].
-
33 statements in Phase I: Foundation, on establishing collaborations, recruiting patients, gaining informed consent, organizing induction meetings, training patient partners, creating communication links and engaging communication partners.
-
34 statements in Phase II: Development, on conceptualizing topics, establishing research priorities, reaching consensus, working with co-design methods, and developing research proposals.
-
18 statements in Phase III: Translational, on data analysis, assisting with the dissemination of the results and promoting the implementation of the outcomes.
-
14 statements in Phase IV: Ongoing Processes, on supporting patient partners and promoting self-evaluation, monitoring progress, and assessing the impact of the patient involvement.
After organizing the conceptual constructs, the research team formulated questions using a 5-point Likert Scale by asking, “How important is it to…” (1 = not important to 4 = very important 5 = extremely important) using an aphasia-friendly format [23, 24] selected by AK as appropriate for PWA.
Cognitive Interviews and Pilot Test
After the first draft was created, two rounds of cognitive interviewing and a pilot test to explore how patient partners with aphasia understood the statements were completed. The cognitive interview study aimed to improve the validity and acceptability of the questionnaire and followed the Consensus-based Standards for the selection of health Measurement Instruments (COSMIN) guidelines [25]. Participants for this part of the study were stroke survivors with aphasia outside of the research team who were involved only in the piloting stage of the guideline. The first cognitive interview was performed with a male stroke survivor (SO), a retired engineer. SO had suffered a left hemispheric stroke three years prior to the study and presented with a moderate anomic aphasia and right hemiplegia. During the first cognitive interview he completed the questionnaire under the guidance of MC, and with occasional assistance from his communication partner, an SLT student. On completion, SO proposed modifications to 11 statements in terms of simplifying the vocabulary (e.g., replace ‘tailor made’ with ‘personalized’) and the complexity of the statements (e.g., replace ‘PWA to form a steering committee to check, and feedback, on the progress of the research’ with ‘PWA to form a committee to check the progress of the research’). Once modifications were made to the first draft of the questionnaire, the second cognitive interview was undertaken with an unemployed female (MG) who had suffered a left hemispheric stroke five years prior to the study and presented with mild-moderate anomic aphasia. She reviewed the revised questionnaire, approved the content, and proposed an additional statement as follows: ‘PWA to designate as a communication partner another person with aphasia’ as she felt that PWA can support each other very effectively during group discussions. The pilot questionnaire with the 18 items and the 100 statements is reported in Appendix 1.
Since the first two pilot tests were face-to-face, a third pilot was completed to test the method online. During the final pilot, a female presenting with stroke-induced expressive aphasia (DT), was emailed the instructions and completed the questionnaire online. DT recommended that instructions in the email be simplified regarding accessing, completing, and submitting the questionnaire. These issues were addressed before the initiation of the Delphi study.
Delphi study
The Delphi technique is widely used to achieve reliable consensus from a group of subject experts on a particular issue [26]. For this study, the Delphi technique
-
supported the anonymity of the participants and the confidentiality of their responses;
-
brought together a geographically dispersed international panel of experts;
-
encouraged honest opinion;
-
allowed a structured/organized group communication process;
-
facilitated endorsement of the computer-based survey by the PPI partners (AK and JRS) as simple to use;
-
was a method free from group pressure, especially for patient participants [27].
The “Recommendations for the Conducting and Reporting of Delphi Studies” (CREDES) [24] was followed (see Appendix 2 for the CREDES checklist). To ensure a wide scope of opinion the expert panel was categorized as follows:
-
PPI and aphasia experts in academia, scientific publications, and training;
-
“Experts by experience” who were PWA and stroke survivors without aphasia (SswoA: we aimed for the opinion of people with stroke and unaffected language skills but were experiencing other symptoms of stroke i.e., fatigue, hemiplegia etc);
-
aphasia rehabilitation clinicians;
-
stroke and aphasia advocates;
-
stroke policy makers;
-
the author of previous PPI guidelines; and a
-
research funding agent.
The authors (MC, JMA and MK) identified experts in the field of PPI, stroke and aphasia research and rehabilitation through a nomination process. People with stroke and aphasia were also nominated and recruited by MC and JRS in Europe through a snowball effect. Individuals were considered eligible to be invited to participate in the expert panel if they had relevant clinical and or/academic backgrounds and experiences concerning PPI in stroke and aphasia research and/or rehabilitation and could contribute to the topic of the study. The criteria for the patient partners were as follows: (1) to be a stroke survivor, (2) to be in the chronic stage of stroke (> 6 months post-stroke), (3) to be able to speak, understand, read, and write in English post-stroke, (4) to be socially active as confirmed from case history, (5) to have at least one academic qualification (Bachelor), and (6) to have had previous research experience, either as a student or as a researcher. Furthermore, for PWA evidence from case history interview of mild-moderate chronic aphasia was a criterion. The Aphasia Severity Rating Scale (ASRS), of the Boston Diagnostic Aphasia Examination [28] was used to rate the severity of the observed language difficulties. Spontaneous speech samples were elicited during a 15-minute semi-structured interview that comprised of four topics: the illness, previous/current occupation, family and housing, hobbies [29]. Aphasia severity was assessed by MC using the ASRS to allow a classification based on fluency and intelligibility.
Participants
The panel consisted of 23 participants deemed sufficient for the survey [30]. Geographical diversity was achieved by recruiting panellists from various organizations around the world. Demographic characteristics of the panellists are reported in Table 2.
Table 2
The demographic characteristics of panellists in the Delphi study
|
Number of Panelists
|
Characteristics of the Panelists
|
Round One
N = 23 (%)
|
Round Two
N = 20 (%)
|
Gender
|
|
Female
|
16 (70)
|
13 (65)
|
Male
|
7 (30)
|
7 (25)
|
Country
|
|
Argentina
|
1 (5)
|
-
|
Australia
|
2 (8)
|
2 (10)
|
Cyprus
|
5 (20)
|
5 (25)
|
Denmark
|
1 (5)
|
1 (5)
|
Estonia
|
1 (5)
|
1 (5)
|
France
|
1 (5)
|
1 (5)
|
Germany
|
1 (5)
|
1 (5)
|
Greece
|
1 (5)
|
-
|
Ireland
|
1 (5)
|
1 (5)
|
Norway
|
1 (5)
|
1 (5)
|
Portugal
|
2 (8)
|
2 (10)
|
Switzerland
|
2 (8)
|
1 (5)
|
United Kingdom
|
4 (16)
|
4 (20)
|
Roles and self-reported job titles*
|
|
People with aphasia after stroke
|
5 (24)
|
4 (20)
|
Stroke survivors no aphasia
|
4 (18)
|
3 (15)
|
PPI Aphasia Experts
|
3 (12)
|
3 (15)
|
Academic Aphasia Researchers
|
4 (18)
|
3 (15)
|
Aphasia Rehabilitation Clinician
|
1 (4)
|
1 (5)
|
Clinical Psychologist on PPI
|
1 (4)
|
1 (5)
|
Stroke Policy Maker
|
1 (4)
|
1 (5)
|
Aphasia Advocate
|
1 (4)
|
1 (5)
|
Stroke Advocate
|
1 (4)
|
1 (5)
|
PPI Guidelines Author/ Policy Maker
|
1 (4)
|
1 (5)
|
Research Funder
|
1 (4)
|
1 (5)
|
Years of experience in research
|
|
Less than 10
|
6 (25)
|
5 (25)
|
10 to 25 years
|
8 (35)
|
7 (35)
|
More than 25 years
|
9 (40)
|
8 (40)
|
Years of experience in PPI research methods
|
|
None
|
12 (50)
|
9 (45)
|
Less than 10
|
2 (10)
|
2 (10)
|
10 to 25 years
|
9 (40)
|
9 (45)
|
*Note that some panellists held multiple roles in addition to their principal job title
|
Patient partners were recruited from the Cyprus Stroke Association, the French Association S’ Adapter- AVC et Aphasie, the Portugal AVC Stroke Association, the Norwegian Stroke Association, Aphasie Suisse, the British Stroke Association and Say Aphasia UK. Nine people with chronic stroke, five with chronic aphasia and four without, met the inclusion criteria. Patient partners were aged between 27 and 70 years old, with a range of education of 15–22 years. All patient partners had completed a research project during their studies or work commitments prior to the stroke. Specifically, PWA 3 was the primary investigator in several projects throughout his academic career and two SSwoA (1 and 4) are now the primary investigators in studies in their perspective fields. The remaining patient partners were familiar with the research process because of prior experience from the completion of thesis work while studying. The demographic characteristics of the PWA and the SSwoA are reported in Table 3.
Table 3
Patient partner demographics
|
Gender
|
Stroke Type
(Hemiplegia)
|
ASRS*
(0–5)
|
Completed Education
|
Research Experience
|
Premorbid Empl/ment
(RtW*)
|
People with aphasia (PWA)
|
1
|
Female
|
Haemorrhagic LH*
(Yes)
|
4
|
Doctoral
|
Thesis Completion
|
Teacher
(No)
|
2
|
Male
|
Ischemic LH*
(Yes)
|
4
|
Masters
|
Thesis Completion
|
Lawyer
(No)
|
3
|
Male
|
Ischemic LH*
(Yes)
|
5
|
Doctoral
|
Primary Investigator
|
Academic
(No)
|
4
|
Female
|
Ischemic LH*
(No)
|
5
|
Masters
|
Thesis Completion
|
Admin
(No)
|
5
|
Male
|
Ischemic LH*
(Yes)
|
4
|
Bachelor
|
Thesis Completion
|
Businessman
(Retired)
|
Stroke Survivors without Aphasia (SSwoA)
|
1
|
Male
|
Ischemic LH*
(Yes)
|
N/A
|
Doctoral
|
Primary Investigator
|
Academic
(Yes)
|
2
|
Female
|
Ischemic LH*
(No)
|
N/A
|
Masters
|
Thesis Completion
|
Nurse
(Yes)
|
3
|
Female
|
Ischemic LH*
(No)
|
N/A
|
Bachelor
|
Thesis Completion
|
Unemployed (No)
|
4
|
Female
|
Ischemic LH*
(No)
|
N/A
|
Doctoral
|
Primary Investigator
|
Academic
(Yes)
|
*LH: Left Hemisphere; ASRS: Aphasia Severity Rating Scale: 0 = limited verbal output and comprehension, 5 = mild word finding difficulties; RTW: Return to Work |
Prior to obtaining informed consent from the participants MC had individual meetings with all patient partners, aphasia and stroke advocates, the policy maker, the PPI guidelines author, and the research funder to explain the scope of the study. Informed consent to participate was obtained by all panellists prior to their involvement in the study [29]. Furthermore, AK and MC co-designed the aphasia-friendly consent form (see Appendix 3).
Delphi survey
A consensus level of 80% was selected as this percentage marks a clear majority opinion [26] and was used in previous PPI framework development research using the Delphi method [6]. Therefore, a statement was deemed to be ‘very important’ or ‘extremely important’ (it was considered as the most useful for the involvement of PWA as research partners for the PAOLI guideline) if it had been rated as either 4 or 5 on the Likert scale by at least 80% of respondents. In recognition that the 80% cut-off criteria selected is a strict and somewhat arbitrary definition of consensus, the statements that at least 70% of the panellists scored 4 or 5 on the Likert scale were highlighted in the findings as statements that were ‘nearing’ the pre-set cut off point and were voted on again in round two. Statements ‘nearing’ consensus from round one, that did not receive at or more than 80% consensus in round two, were eventually discarded.
In round one participants received an electronic invitation, with an attached link and simplified instructions. While rating all statements in each section, panellists were also asked to comment on each statement, if they wished, in the space provided. Each participant was allowed 4 weeks to respond to the survey questionnaire [30]. After receiving panellists’ responses, the new information collected was used to modify the second version of the questionnaire. The panellists received a personalised report showing quantitative responses to the round one statements. MC examined the anonymised quantitative scores and qualitative comments for each statement and generated round two.
In round two each panellist was asked to review the statements summarized based on the information provided in round one and to rate them again using the Likert scale. As a result of round two, there was a high level of agreement for most statements and consensus was achieved. Again, after round two MC send out a summary report with the revised set of statements to the panel.
Phase 4. Integration
During the Integration phase the aim was to discuss the results of the survey with the stakeholders in an expert’s consensus meeting [10]. Before the experts meeting, each panellist received a report that included all statements and ratings from the two rounds. The meeting was held online, and participants were given the opportunity to discuss both the information generated and the structure of the questionnaire. The meeting provided a final opportunity for experts, patients, and PPI partners to express their viewpoints on the items and the statements. For this Delphi study two iterations and one experts’ meeting (March 2020- Dec 2020) were sufficient to collect the information needed to reach consensus [29]. See the flowchart for the Delphi procedure in Appendix 4. This was then followed by the validation phase of the guideline initiated by PPI partners AK and JRS.
Phase 5. Programming
During Programming, the ongoing phase that was instigated from phase 1, MC was updating continuously all stakeholders with controlled feedback reports and email to sustain engagement [10]. Also, during Programming, MC drafted the manuscript with AK and JRS as co-authors and presented the preliminary data of the study at the European Life After Stroke Forum of the Stroke Alliance for Europe (March 2023).
Data analysis
The statistical measures used for analysis were measures of central tendency (medians and mode) concerning the collective judgments of the respondents [30]. Measures of central tendency were used as they are widely applied in Delphi studies [24, 27]. All statistical analyses were undertaken with the jamovi (version 1.6) statistics computer software [31, 32]. To prevent bias, an independent researcher was employed to statistically analyse the results of the Delphi survey.