Assessing and promoting partnership between patients and health‐care professionals: Co‐construction of the CADICEE tool for patients and their relatives

Abstract Context Partnership between patients and health‐care professionals (HCPs) is a concept that needs a valid, practical measure to facilitate its use by patients and HCPs. Objective To co‐construct a tool for measuring the degree of partnership between patients and HCPs. Design The CADICEE tool was developed in four steps: (1) generate key dimensions of patient partnership in clinical care; (2) co‐construct the tool; (3) assess face and content validity from patients’ and HCPs’ viewpoints; and (4) assess the usability of the tool and explore its measurement performance. Results The CADICEE tool comprises 24 items under 7 dimensions: 1) relationship of Confidence or trust between the patient and the HCPs; 2) patient Autonomy; 3) patient participation in Decisions related to care; 4) shared Information on patient health status or care; 5) patient personal Context; 6) Empathy; and 7) recognition of Expertise. Assessment of the tool's usability and measurement performance showed, in a convenience sample of 246 patients and relatives, high face validity, acceptability and relevance for both patients and HCPs, as well as good construct validity. Conclusions The CADICEE tool is developed in co‐construction with patients to evaluate the degree of partnership in care desired by patients in their relationship with HCPs. The tool can be used in various clinical contexts and in different health‐care settings. Patient or Public Contribution Patients were involved in determining the importance of constructing this questionnaire. They co‐constructed it, pre‐tested it and were part of the entire questionnaire development process. Three patients participated in the writing of the article.


| INTRODUC TI ON
In recent years, patient engagement in direct care has received more attention in health care and is being implemented internationally. [1][2][3] Partnership in care takes patient engagement further, by considering patients as real partners in the care process and as full-fledged members of the health-care team. 4 This innovative approach to care is built on different approaches to care, including patient-centred care, patient empowerment and shared decision-making, but goes further by recognizing: (1) the experiential knowledge they have gained from living with a disease and from experiencing health-care services; (2) patients as full-fledged care team members, like the health-care professionals (HCPs); and (3) patients as capable of making the most appropriate decisions for themselves. 5 For many patients, their doctors' understanding of their disease is inconsistent with their experience with that disease. 5 Because patients understand the realities associated with their condition and the impacts of the disease and its treatment on their lives, their perspective enhances HCPs' expertise and should be considered in all care decisions. 4,[6][7][8] Patient partnership of care refers to the co-construction between health professionals and patients to assess the patients' participation in the decisions made regarding their care, and co-build intervention to support and accompany patients as they become increasingly autonomous in their decision-making and ability to influence their health. 4,9 However, patients differ in their ability to be autonomous, and their situation may change in the course of their care trajectory. Health professionals do not currently have the tools to fully perceive this reality and act based on patients' real needs.
Recently published frameworks for evaluating different aspects of patient engagement have underscored the importance of evaluating patients' involvement, not only at the organizational level, but also at the clinical level. [10][11][12][13] A tool has been developed to measure patients' contributions and public involvement at the organizational level, 14 thereby supporting the quality of public and patient engagement in health service organizations. The Public and Patient Engagement Evaluation Tool (PPEET) makes it possible to concurrently consider the points of view of members of the public/patients/family members, managers and board members/senior managers. However, the tools currently available at the clinical level cover only some of the aspects of this partnership as it is perceived by patients and health-care professionals. 8,[15][16][17][18] A review of the tools carried out when this study began and a complementary scoping review carried out in 2021 19 show four limitations in recent tools developed and used to assess patient engagement at the clinical level.
First, most of the tools evaluate dimensions related to one or two different concepts of patient engagement (patient-centeredness or empowerment or shared decision-making). [20][21][22][23][24][25][26][27][28][29][30][31] Accordingly, no tool offers an exhaustive assessment of all the core concepts and dimensions on which the new partnership in care approach is built. This article is therefore intended to bridge the gap in assessments of patient partnership at the clinical level by sharing the results of a three-year collaboration between patients, HCPs and researchers to develop a tool that could be used by a wide range of organizations in the health system. This tool, called CADICEE, is intended as a means to help patients and HCPs discuss the basis for the partnership relationship and identify potential partnership gaps. More broadly, it could also help create a database on how patients perceive the extent of their partnership with their HCPs.
This research project received ethics approval from the research ethics committee (CER-2017-018 and 18.045) of the Centre de recherche du Centre hospitalier de l'Université de Montréal (CRCHUM).
Step 4: Assessment of the tool's usability and exploration of its measurement performance.

| Methods
To be able to co-construct the tool, we decided to appeal to patients who understand the patient partnership context. We seek to hire expert patients on that topic regardless of their degree of literacy.
To do so, we asked the person in charge of the bank of patients participating in courses for future health-care professionals at the Université de Montréal on care partnerships. [36][37][38] From this bank, we targeted those who: (1) had at least one chronic disease, and (2) were capable of talking to students about their experiences with the disease and how they mobilize their experiential knowledge to become full partners in the care team, so that the team's decisions would be based on their life plan. 4,6,37 Drawing on the clinical sociology method, [33][34][35] patients selected were asked to write a three-to five-page account of their care experience in the context of a single consultation with an HCP. A qualitative thematic analysis was performed by the research team, composed of patient-researchers (n = 3) and researchers (n = 8), to analyse the accounts of these patients and highlight the various dimensions covered. 39 Each text was analysed using the NVivo (version 10) software. Two techniques were used in the coding: primary open coding, followed by thematic and selective coding, 40,41 as per the grounded theory approach. 42

| Findings
Of the 243 patients involved in care partnership courses, 85 were approached, 34 responded, and following interviews, 15 were retained with various characteristics. The 15 expert patients were mostly women (13/15; 87%), over 45 years old (10/15; 66%) (see Table 1) and with a variety of chronic diseases (including cancer, metabolic diseases, degenerative diseases and autoimmune diseases).
Development followed an iterative process in which the expert patients were asked to participate in the entire process of coconstructing the tool as a patient research advisory group that would  Table 2).

|
Step 2: Co-construction of the tool, based on the dimensions from Step 1

| Methods
After stabilizing the seven core dimensions of partnership in care, patients and researchers met to discuss the tool's structure and the questions to be asked, and determine measurement scales. In addition, to ensure that the wording used in the questionnaire would be TA B L E 1 Sociodemographic characteristics of the patients who participated in the development of the tool (steps 1 and 2) accessible to everyone who may potentially use it, a literacy expert was consulted to ensure that people with a Grade 6 level of literacy (children between 11 and 12 years old) could understand the tool, based on the Simple Measure of Gobbledygook (SMOG) readability formula. 44,45 The SMOG formula is based on the number of words containing three or more syllables and has been found exceptionally effective in the health-care setting.

| Findings
The research team and the patient research advisory group ex-

| Methods
The tool was sent to two different types of participants. One group consisted of new patient partners not previously involved in the TA B L E 2 Dimensions and items of the CADICEE tool project who were also teaching health professionals. Patients selected had to meet two criteria: (1) be a patient-as-trainer taking part in partnership courses given to students in the health sciences and psychosocial sciences; and (2) be a patient or caregiver having regular contact with health-care providers. The second group consisted of HCPs who had helped teach patient partnership to future HCPs. 6,37 A written invitation was sent to teachers who had participated in these courses for over three years (n = 23).
Participants from the two groups received the tool and an anonymous evaluation grid assessing their understanding of the objective/purpose of the tool, its overall readability, the appropriateness of the measured dimensions and the individual items (wording, relevance), as well as its ease of use and the appropriateness of the measurement scales. For each question in the evaluation grid, participants were asked to check a dichotomous Yes/No box and were encouraged to add a comment justifying their answer.

| Findings
A 14 patients were available to participate in reviewing the tool.
Patients who tested the tool had a variety of chronic diseases (including cancer, metabolic diseases, degenerative diseases, autoimmune diseases, respiratory and genetic diseases) and sociodemographic characteristics (see Table 3).
The research team synthesized the evaluation grids from both patients and HCPs. All of the patients and 90% of the HCPs found the tool relevant and useful. The HCPs mentioned that the tool led them to think about care relationship dimensions that they had not thought about before, such as patients demonstrating empathy for HCPs. In addition, all of the patients and 80% of the HCPs found the tool straightforward. Almost all of the HCPs (90%) were comfortable using the two measurement scales of the tool, whereas 71% of the patients were comfortable with these scales.
In a face-to-face meeting, the research team and the patient re-

| Findings
The three health and social service institutions and the two primary The sample was diversified in terms of gender (56% women) and covered a wide range of age groups, education levels and life experience with a chronic disease. 67% (n = 164) completed the questionnaire regarding a consultation with a doctor (see Table 4). response category was seldom used by respondents (mean use per item = 1.8%, SD = 1.5), and missing values were found ranging from 0% to 6.1% (mean = 2.8%, SD = 1.5).
Spearman's rank-order correlation (r s ) matrices were separately which concerned items about Confidence/trust and Empathy, which are related concepts.
We also conducted median tests to detect differences in the position and shape of response distributions for each CADICEE item depending on clinical setting, HCP consulted, or respondents' age, gender, education level and the presence of a chronic illness.
Several statistically significant differences were found (see Table 6).
However, we believe that these patterns were likely due to real differences between subgroups, in support of the construct validity of the CADICEE tool, rather than indicative of potential differential item functioning. For example, it is understandable for patient advisors to report having more useful expertise during the consultation compared with patients in other clinical settings (P = .035) and for them to have a greater need for their expertise to be recognized and considered by the HCP (P < .001). In addition, patients with an elementary level of education generally required more help in becoming more autonomous compared with others, and patients with a university level had a lesser need in this regard (P = .037).
Finally, we computed the gaps between experience of care and needs, for every relevant pair of items in the CADICEE tool. For example, reporting that the care received was tailored to personal context only (2 = very little) when one had indicated a prior need for it to be tailored (4 = very much) would lead to a negative gap of 2 ranks English, was adopted. 51

| Strengths of the tools
This CADICEE tool is, to our knowledge, the first developed in coconstruction with patients from development to validation 19,52

| Content validity and reliability
The CADICEE tool was designed and revised and involved numer-

| Key features that make the tool unique
Some of the CADICEE dimensions such as confidence/trust, information-sharing and empathy have been included in previous instruments measuring patient experience of care. 59 However, it is worth emphasizing that the CADICEE tool approaches these dimensions from an innovative angle of reciprocal relationships between partners, acknowledging that patients are more than passive recipients of care, which is manifested in items measuring, for example, the professional's level of trust in the patient, the information shared by the patient with the professional and the empathy that the patient shows for the professional. Other dimensions of the tool focus on aspects of care measured only by more recent instruments, such as patient activation or empowerment 28,60-62 or shared decision-making. 16,63,64 Finally, the dimension related to recognizing and using patient and professional expertise in the consultation has, to our knowledge, never been addressed in other instruments and represents an innovation. CADICEE effectively considers all these dimensions in a single and relatively short tool, from the perspective of patients-as-partners in their care with specific needs in each consultation, which represents an original contribution.
A recent scoping review 19 also emphasized that most of the tools evaluate dimensions related to one or two different concepts of patient engagement (patient-centeredness, or empowerment or shared decision-making), and a few tools simultaneously measure some dimensions related to all the different concepts of patient engagement. The CADICEE tool includes all patient partnership dimensions and was built in co-construction with patients from development to validation, which was not previously developed in the literature. 19 The tool can also be short due to general overarching statements rather than more numerous and specific items that could have referred more directly to some of their subcomponents or associated observable phenomena (eg behaviours), as recommended in psychometrics. The patient research advisory group was instrumental in ensuring that the definitions of dimensions remain sufficiently general and evocative to be easily understandable by patients. They preferred a tool that would provide a vaguely correct assessment of the partnership in care over one that is precise, but incorrect. Following a consultation, patients orient their conduct based on their subjective impressions, regardless of how these relate to a more objective or external assessment. This subjective perspective is what the CADICEE tool captures, and why it is intended as an instrument to promote further reflexive assessment, discussion and investigation by patients and HCPs.

| Limitations of the tool
The tool development process revealed some limitations. First, the CADICEE tool was developed with patients-as-partners who were familiar with the partnership concept. Although on this subject they were clearly experts among patients, their perspectives on the important dimensions of partnership in care may not represent the views of all patients. In addition, participants involved in developing the tool were mainly women. However, patient participants were of all ages and with various chronic diseases, and as such, the dimensions of the tool were developed on the basis of a diversity of experiences. Although CADICEE is intended to be a generic tool to be used for various diseases (chronic or acute) and in different clinical settings, we have not developed it for use in a paediatric setting; a separate version of the tool would most likely need to be developed for that purpose.

| CON CLUS ION
To our knowledge, CADICEE is the first tool developed in coconstruction with patients that evaluates the degree of partnership in care as perceived by patients in their relationship with their HCPs, and that can be used in different clinical contexts (hospitalization/ consultation; chronic/acute diseases). We encourage further applications of the tool in other Canadian provinces and countries and welcome their results to further refine the tool. An examination of the properties of the measurement scales will be presented in a future article. Moreover, we are currently developing a version for HCPs to measure the gap in perceptions between patients and professionals during the consultation.