A qualitative study on measuring patient‐centered care: Perspectives from clinician‐scientists and quality improvement experts

Abstract Background and aims Patient‐centered care (PCC) benefits patients, health‐care providers, and health‐care systems by providing delivery of care that addresses patient values and needs while improving provider experiences, and by decreasing health‐care expenditure. To improve PCC, health‐care systems need to measure it. Recently, we developed a PCC framework that is evidence based and patient informed. The purpose of this study was to gather the perspective of clinician‐scientists and quality improvement experts regarding the PCC domains included in the framework. Their perspectives were used to refine these domains, which ultimately will inform the development of PCC quality indicators. Methods Participants were recruited via expert and snowball sampling. Semi‐structured interviews were conducted with clinician‐scientists and quality improvement experts from Canada, the United States, and the United Kingdom from October 2017 to January 2018. With the use of an interview guide developed using the PCC framework, interviews were audio recorded and transcribed for a thematic analysis using NVivo qualitative data analysis software. Inductive thematic analysis was used to identify themes and subthemes. Results Sixteen semi‐structured interviews were conducted, which included four clinician‐scientists and 12 quality improvement experts. Twelve of the participants were from Canada, three from the United Kingdom, and one from the United States. From the thematic analysis, three major themes were identified: (a) measurability of PCC, (b) practical considerations for implementing measurement, and (c) policy and practice implications. Participants discussed barriers and recommendations to improve and increase the clarity of the PCC domains in health system reporting, resulting in several future directions to refine and target specific PCC domains. Conclusion Clinician‐scientists and quality improvement experts provided key recommendations for the measurement of PCC. The perspectives of key stakeholders in PCC measurement will inform strategies for the implementation and uptake of patient‐centered quality indicators in health‐care systems. The views of these key experts can lay the foundation for the development of standardized measures of PCC, to ensure monitoring and improvement of PCC.


| INTRODUCTION
Patient-centered care (PCC) is a model of care guided foremost by the needs and values of patients. 1 Patient-centered care is an increasingly well-recognized and highly sought-after model of care, reaching the height of its prominence in a report published by the Institute of Medicine, which listed PCC as one of the six most important dimensions of high-quality care 1,2 and defined PCC as care that is "respectful of and responsive to individual patient preferences, needs, and values, and ensures that patient values guide all clinical decisions." 1,3 Previous research has found that PCC has the potential to improve health outcomes [4][5][6][7][8][9][10][11][12][13][14] , and benefits health-care systems and health-care providers. Practice that goes against principles of PCC, such as failure to consider the patient's wishes in decisions related to care, has been associated with accusations of malpractice. 15,16 When a provider fails to consider a patient's needs and values, there is a risk for miscommunication. Additionally, health-care systems benefit from PCC in decreasing patients' length of stay, minimizing the need for unnecessary testing and procedures, and decreasing the cost per case, ultimately improving the efficiency of care. 5,9,14 Numerous frameworks have been developed for PCC, such as Mead and Bower's conceptual framework for patient-centeredness 17 and Scholl et al's integrative model of patient-centeredness. 18 However, most PCC frameworks have not focused on the practical implementation of PCC in health-care systems. Additionally, there is currently no systematic approach in place to measure the quality of the provision of PCC. 16,[19][20][21] Patient-centered quality indicators (PC-QIs) should be developed to measure PCC in a standard manner.

| How to practice person-centered care: A conceptual framework
A PCC framework was developed in collaboration with a patient partner, following a narrative review of the literature synthesizing evidence, recommendations, and best practice from existing frameworks, and case studies on the delivery of PCC. 22 This framework categorizes PCC into three components and a total of 13 domains (Appendix S1).
The Donabedian model for health-care improvement 23 was utilized to classify domains into the categories of structure, process, and outcome. The first component, structure, involves seven domains that focus on PCC at foundational and organizational levels, such as creating a PCC culture, codesigning the development and implementation of educational programs, and supporting a workforce committed to PCC. 22 The second component, process, involves four domains that focus on PCC from a patient and health-care provider level, such as cultivating communication and respectful and compassionate care. 22 The final domain, outcome, involves two PCC domains (access to care and patient-reported outcomes [PROs]) that focus on outcomes related to access and patient reports. 22 The PCC framework will be used to guide the identification, development, and classification of PC-QIs and will serve as a cognitive tool to ensure that the PC-QIs are consistent with the key dimensions of PCC.
While this framework provides a theoretical and empirical basis for measuring PCC, there is a need to engage clinicians and quality improvement experts, as the users of these measures, for monitoring and improving the quality of care. In particular, it is critical to ensure that the proposed areas of measurement are seen to be relevant to their work and feasible to implement in practice. Hence, the purpose of this study is to elicit the opinions of clinician-scientists and quality improvement experts regarding the proposed domains of PCC that will inform the development of PC-QIs. Specific objectives include the following: 1 to explore the views of clinician-scientists and quality improvement experts regarding proposed domains of PCC, and 2 to gain an understanding of current practices and opportunities for measurement of PCC at a health-care system level.

| METHODS
This qualitative study used semi-structured interviews to explore the views of clinician-scientists and quality improvement experts regarding PCC measurement, acceptability, and feasibility. The semi-structured interview guide was developed in collaboration with the PC-QI research team at the University of Calgary and is based on the 13 domains of PCC in the conceptual framework. 22 The interview guide aimed to refine the proposed definition of a PC-QI, the feasibility of the PCC domains, and impacts on policy, and to identify potential barriers and facilitators to implement, measure, and report PCC domains (Appendix S2). The interview guide was first piloted with local members of the PC-QI research team who were not involved in this project and was amended based on subsequent discussions between members of the research team. Interview audio files were transcribed verbatim and imported into NVivo (version 11.4) qualitative software for primary data analysis.
First, transcripts were analyzed for codes, with phrases within transcripts referring to specific topics, questions, actions, or perceptions 24-26 by two researchers. Once coding was completed, peer debriefing allowed both researchers to reach a consensus on the interpretation of the findings.
To conduct a thematic analysis, codes were organized based on repetition, material relating directly to research questions, or similarities and differences between participants, as per Ryan and Bernard's 27 suggestions for thematic analysis. In an open coding process, key phrases were organized into codes or topics of discussion. Codes were condensed and organized into subthemes under higher-order themes with the research team, with key quotations provided to support subthemes. Inductive thematic saturation was achieved with the interviews, defined as no additional codes or themes identified during data analysis. 28

| Trustworthiness measures
The quality of qualitative research is often assessed using trustworthiness measures. 24 Participants were asked only open-ended questions regarding the PCC domains. The current study also sought to enhance dependability by ensuring that all phases of the research process were recorded carefully and by utilizing other members of the research team as auditors of the research process. 24 Records of recruitment processes, interview transcripts, and data analysis decisions were carefully managed and made accessible to all members of the research team. Periodic team meetings allowed the research team to provide input regarding the research process, enhancing the dependability of the study.

| RESULTS
We conducted 16 interviews with clinician-scientists and quality improvement experts between October 2017 and January 2018, six of them face to face and 10 via phone. The roles of the participants included project lead in performance measurement and patient-centered care medical home director, and physician specialties were emergency medicine, respirology, and family medicine. Participant characteristics are summarized in Table 1.
Three overall themes were identified: (a) measurability of PCC, (b) practical considerations for implementing measurement, and (c) policy and practice implications.

| Measurability of PCC
Participants' overall conceptualization of person-centered care were aligned with the domains of the PCC framework such as codesigning educational programs for PCC, cultivating communication, and engaging patients in managing their care. Participants discussed the applicability of Donabedian framework for monitoring PCC, distinguishability of PCC domains, challenges in measuring subjective domains, and suggestions for the improvement of the PCC framework.

| Applicability of the Donabedian framework
Participants commented on the applicability of the Donabedian framework for PCC measurement. Most participants noted the Donabedian framework of structure, process, and outcome to be useful for organizing PCC domains and for monitoring quality of care. P 12(QI expert) : The focus on the reliance on the Donabedian framework is useful. It allows it to integrate it into a comprehensive framework in relation to the specific domains.
A clinician-researcher described how they distinguished between structure and process domains, indicating that structure domains such as "creating a PCC culture" and "educational programs for PCC" were tangible, foundational domains that can be enacted by the inclusion of process domains such as "cultivating communication," "being respectful," and "engaging patients."

| Challenges in measuring subjective domains
Some participants also discussed the notion of a "soft domain" and the difficulties in measuring such subjective domains. Patient-centered care researchers used the term soft domain to refer to domains that are subjective in nature, often referring to "Creating a PCC Culture" as an example of a soft domain. one is going to look at that dashboard. P 14(Clinician-scientist) : I would think that these would be feasible but it might be 5-10 years before you actually are able to get these integrated … the caveat is time.

| Stakeholder engagement necessary
Some participants also emphasized the need for stakeholder and community engagement in implementing the domains of PCC and, ultimately, designing PC-QIs. One clinician-researcher stated that moving PCC domains into practice in Canada depends on vigorous inter-professional collaboration and community engagement, for instance, bringing together different health-care professionals in the same room to discuss implementation of PCC. This participant believed that the confidence of stakeholders in the creation of PCC measures is a prerequisite for receptiveness once PC-QIs are disseminated.

| Defining quality indicator
In order to measure person-centered care using quality indicators, providers as an element to optimize the health system performance. 33 Therefore, indicators incorporating health-care provider experience should be developed and implemented.
One participant found the PCC domains to be inadequate in addressing the social determinants of health. For example, in the domain "access to care," there is no acknowledgment of the unique challenges to accessing care that rural or homeless populations may face. This finding is supported by past literature that has explored the understated principle of empowering patients in PCC. 34

CONFLICTS OF INTEREST
The authors report no conflict of interest.

TRANSPARENCY STATEMENT
The corresponding author (Maria Santana) affirms that this manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained.

DATA AVAILABILITY STATEMENT
This study analyzes qualitative data, and the participants did not consent to have their full transcripts made publicly available.