Patients as team members: Factors affecting involvement in treatment decisions from the perspective of patients with a chronic condition

Abstract Background Active patient involvement in treatment decisions is seen as a feature of patient‐centred care that will ultimately lead to better healthcare services and patient outcomes. Although many factors have been identified that influence patient involvement in treatment decisions, little is known about the different views that patients have on which factors are most important. Objective This study explores the views of patients with a chronic condition on factors influencing their involvement in treatment decisions. Design Q‐methodology was used to study the views of patients. Respondents were asked to rank a set of 42 statements from the least important to the most important for active patient involvement in treatment decision‐making. The set of 42 statements was developed based on a literature search and a pilot in which two external researchers, 15 patients and four healthcare professionals participated. A total of 136 patients with one of three major chronic conditions were included: diabetes types 1 and 2, respiratory disease (i.e., chronic obstructive pulmonary disease and asthma) and cancer (i.e., breast cancer and prostate cancer). Data were collected in a face‐to‐face interview setting in the Netherlands. Results Four distinct views on the factors influencing active patient involvement were identified among patients with a chronic condition. (1) Enabled involvement: the extent to which patients are facilitated and empowered to participate will lead to patient involvement. (2) Relationship‐driven involvement: the relationship between patients and healthcare professionals drives patient involvement. (3) Disease impact‐driven involvement: the severity of disease drives patient involvement. (4) Cognition‐driven involvement: knowledge and information drive patient involvement. Discussion and Conclusion From the patients' perspective, this study shows that there is no one‐size‐fits‐all approach to involving patients more actively in their healthcare journey. Strategies aiming to enhance active patient involvement among patients with a chronic condition should consider this diversity in perspectives among these patients. Patient Contribution Patients are the respondents as this study researches their perspective on factors influencing patient involvement. In addition, patients were involved in pilot‐testing the statement set.


| INTRODUCTION
Active patient involvement has been high on the agenda in recent decades and is seen as an important feature for patient-centred care.
Patient involvement can take many forms, for example, by involving patients in governmental policies on healthcare, research or hospital policy. 1 Nowadays, patients are especially expected to be partners of healthcare professionals in the treatment decision-making process, as their experiential knowledge is seen as complementary to professionals' knowledge. 2,3 It is increasingly being recognized that patient involvement in decision-making can improve medical outcomes, patient satisfaction and quality and safety of care. [4][5][6][7][8][9][10][11] Many studies have therefore focused on identifying factors that may enhance patient involvement in treatment decision-making. 12,13 A wide range of studies have assessed factors that act as facilitators and barriers to patient involvement in treatment decisions. Next to the more general literature on patient involvement in treatment decision-making, 14 there is also literature that focuses on patient involvement within a specific disease group [15][16][17][18] or with respect to a certain outcome such as patient safety. 5,7 In addition, relevant factors can be found in the literature on shared leadership and teamwork between healthcare professionals, informal caregivers and patients, in which patients are considered as team members in terms of their involvement in treatment decisions. 19,20 Due to the fast-growing literature and the overwhelming number of identified influencing factors, it is currently difficult to paint a clear picture on what matters the most for patient involvement in treatment decision-making. Furthermore, people who participate in the care process seem to have different views on patient involvement. Research has revealed differences between the points of view of patients, informal caregivers and healthcare professionals, but also showed variety within these groups. 21-23 From the healthcare professionals' perspective, different studies have shown divergent views on their roles in self-management support. [24][25][26] For example, van Hooft et al. 24 distinguished four perspectives on the goals for selfmanagement support among nurses (i.e., the coach, the clinician, the gatekeeper and the educator perspective). 24 Been-Dahmen et al. 25 identified three divergent views among nurses on patient and nurse roles in self-management support (i.e., adhering to a medical regimen; monitoring symptoms; and integrating illness into daily life). 25 Also, Aasen et al. 26 identified three discursive practices among nurses on the participation of patients and their relatives in end-of-life decisions (i.e., the nurses' power and control, sharing power with the patient and transferring power to the next of kin). From the patient perspective, available studies often present the view of a specific group of patients.
Jedeloo et al. 27 identified different preference profiles for healthcare delivery and self-management among adolescents with chronic conditions (i.e., conscious and compliant, backseat patient, self-confident and autonomous and worried and insecure). O'Brien et al. 23 identified facilitators and barriers to patient involvement from the perspective of patients with early-stage breast cancer and their physicians. Although the variance in patient preferences to participate in treatment decisions is acknowledged, 7,28 there is a widespread belief that most patients wish to be involved in decision-making at least to some extent, either in terms of shared decision-making with their physicians and/or informal caregivers or by making their own decisions. 29,30 Considering the number of factors that may influence active patient involvement identified in previous literature, but also acknowledging that preferences may differ between patients, it seems important to understand the patient perspective on these factors better. Therefore, the aim of the present study is to explore the diversity of views among patients on factors influencing their involvement in treatment decisions.

| Setting
Studying patient perspectives on what is important for active involvement seems especially interesting for patient groups who could gain the most from involvement and have the most opportunities to be involved. 31 Chronically ill patients, who undergo long-term treatment, have to adjust to changes in their condition and/or their treatment over time and have long-lasting relationships with healthcare professionals are therefore the most suitable target group for the present study. 31 To study the views of patients with chronic diseases, we selected patients from three major, distinct types of chronic diseases: diabetes, chronic obstructive pulmonary disease (COPD)/asthma and cancer. Then, we looked at possible relevant subdivisions within these groups related to differences in care BULJAC-SAMARDZIC ET AL. | 139 trajectories. We decided to include both type 1 and type 2 diabetes patients, and both COPD and asthma patients. As there are many types of cancers, we decided to choose the two most common invasive forms of cancer for both men and women: prostate cancer and breast cancer. A Q-study usually has a sample size between 30 and 40 respondents, 32,33 but to capture the variety in views in this diverse sample, we decided to increase the sample size. We recruited around 20 patients from each (sub) type. The final study sample consisted of 136 respondents: 19 diabetic type 1 patients, 21 diabetic type 2 patients, 26 COPD patients, 25 asthma patients, 23 breast cancer patients and 22 prostate cancer patients. All of the respondents were living at home and received care from formal and informal caregivers. Depending on the severity of their condition (at that time), they interacted with primary care (GP and often a nurse practitioner) and/or secondary care (mostly a medical specialist from a general hospital). In case of asthma and diabetes, for example, respondents interacted most of the time with their primary care professionals, as is customary in the Dutch healthcare system.

| Q-methodology
We used Q-methodology to explore and compare the perspectives of patients on active involvement in their treatment decisions.
Q-methodology is frequently used to study the views, attitudes or perspectives of patients, professionals and other stakeholders in healthcare research, for example, views on medical leadership, patient-centred care, vaccination and effective teams. 34-38 Q-methodology is particularly useful for the purpose of this study because it combines aspects of qualitative and quantitative methods for an in-depth study of potentially complex and diverse subjective topics, such as patient perspectives. 39,40 The core of the data collection consists of respondents reading, evaluating and ranking a broad set of statements about the topic, usually between 40 and 60 statements, according to their perspective on the topic. The combination of this elaborate ranking exercise that is the same for all respondents and a follow-up interview that delves deeper into the individual perspective of respondents that they reveal through their ranking of the statements results in a rich data set, minimizing researcher bias and allowing respondents' voices to be heard in a unique way. [41][42][43] The study was conducted in four stages: (1) development of the list of factors that influence patient involvement (statement set); (2) selection of respondents; (3) data collection; and (4) data analysis. Because this study seeks to explore patient perspectives rather than confirm particular theoretical constructs, no formal hypotheses were formulated. However, we note that the development of the statement set is guided by existing research identifying factors likely to affect patient perspectives.

| Development of the statement set
The statement set should broadly cover the variety of factors that may influence active patient involvement. First, the literature was reviewed to identify the main factors related to patient involvement in treatment decision-making. In our endeavour to develop a comprehensive list of factors, we drew upon literature on patient involvement (and related concepts such as self-management and shared decision-making), shared leadership and teamwork. On the one hand, we studied literature reviews. 5,7,15,17,44 It is noteworthy that Dwarswaard et al. 13  availability of materials and the use of technology to support patients. 15,44,48 Illness-related factors reflect the prospective course, the stage, the severity and the impact of the illness, but also the presence of different treatment options and acceptable alternatives. 5,15,17 Next, in keeping with this, a comprehensive list of statements was compiled by the authors that aimed to cover the content of these five factors and this initial set of 60 statements was then pilot-tested.
To assess content validity, two external researchers with expertise in patient involvement were consulted. To assess face validity, 15 patients (i.e., two diabetic type 1 patients, three diabetic type 2 patients, two COPD patients, two asthma patients, two breast cancer patients, four prostate cancer patients) and four healthcare professionals engaged with patient involvement were consulted (i.e., one physician, three paramedics). In an interview setting, they were asked to rank the statements and reflect on them by judging the clarity of the statements and the completeness of the set. Based on this pilot test, 10 statements were deleted and eight statements were combined with other statements. This reduced the statement set to 42 statements. Thereby, multiple statements were rephrased from minor to major revision, for example, from 'how well a patient knows the healthcare system' to 'how well a patient knows where he/she can go with a request for help' or from 'how well a patient mastered the spoken and written Dutch language' to 'how well a patient mastered the Dutch language'. The final statement set consisted of 42 statements covering patient-, motivation-and opportunity-related factors in terms of active patient involvement (see Table 1).

| Data collection
In a face-to-face interview setting in the Netherlands, respondents were asked to rank 42 statements from the least important to the most important for active patient involvement in treatment decisions using a sorting grid ( Figure 1). These one-on-one sessions between the patient and the researcher were recorded, summarized and translated from Dutch to English. The data collection consisted of three steps and was conducted before the coronavirus disease pandemic of 2019. In the first step, respondents were instructed to first read all the statements; each statement was printed on a separate card. While reading each statement, the respondents were asked to divide the cards into three piles: important for patient active involvement, unimportant for active patient involvement and neutral. In the second step, respondents were instructed to read all the statements placed on the 'important pile' once again and to select the two most important statements, and then place them in the two spots on the extreme right side of the sorting grid (under 9). From the remaining statements in the 'important pile', respondents then selected the three statements that they found most important and placed these on the sorting grid (under 8), and so on until there were no statements left in the 'important pile'. This was repeated for the 'unimportant pile', which was ranked from the left side of the sorting grid, followed by the 'neutral pile', which was ranked in the remaining spots in the middle of the sorting grid. This resulted in a completely filled sorting grid. In the third step, respondents were asked to motivate their ranking of the statements. Respondents were asked to explain the reasoning behind the placement of the two statements they considered most important and then the reasoning for the two least important statements. In addition, respondents were asked to answer a number of questions about personal characteristics (i.e., gender, age, living situation, hours of informal care per week and relation to informal caregiver, years of experience with the disease and formal care). Respondents were also asked to rate their health status on a paper-based Visual Analogue Scale, which is a vertical line that represents a continuum from 0 (worst conceivable health) to 100 (best conceivable health).

| Data analysis
Analyses were conducted using the PQMethod 2.35 software package. 49 First, a correlation matrix between the rankings of the statements by respondents was computed to inspect the degree of similarity between the rankings. Second, the correlation matrix was subject to a by-person factor analysis (centroid factor analysis with a varimax ro-

| Four views on what is important for active patient involvement
The analysis revealed four views among patients with a chronic disease on what is important for them to act as a team member in their healthcare journey, as outlined below. Table 1 presents per view the ranking of statements from the least important to the most important. Figure 2 visualizes the ranking for the first view.

| View 2: Relationship-driven involvement
The core of this view is that the quality of the interaction between patients and healthcare professionals especially influences patient involvement. 'I had another internist before, but he didn't take me seriously. I noticed that this influenced my involvement:

| DISCUSSION
Modern healthcare increasingly calls for patients to be actively involved in their healthcare process. Especially as care is increasingly more organized around the patient in a holistic manner, the patient becomes the linking pin in a dynamic network of healthcare professionals from different organizations. Some even consider patients as being part of the healthcare team in which care is basically coproduced. 19,20 This study offers insights into the diversity in perspectives among patients with a chronic disease on what is important to be actively involved in decision-making about the content and process of treatment. This study may help to focus the efforts to enhance patient involvement. We identified four distinct views, which can be briefly summarized as follows: (1) Enabled involvement; these patients report that active involvement is influenced by the extent to which they are facilitated and empowered to be involved.
(2) Relationship-driven involvement; these patients state that active involvement mostly depends on the quality of the interaction and the factor, but about creating the right conditions that fit the views and characteristics of specific groups of patients. For example, many studies discuss patient portals as a tool to better involve patients and stimulate self-management. 50 Patient portals are secure internetbased platforms that offer patients the ability to view their personal health information. In practice, these portals are, however, only used by a select group of patients. 50 One could hypothesize that differences in use may be partly explained by the views that we identified.

| LIMITATIONS AND STRENGTHS
Q-methodology is well suited for exploring and describing the variety of views on a topic in a certain population in depth, but is less suited for making claims about the distribution of these views within the larger population or their associations with the characteristics of the