Patient-centered care in practice: hospital and online primary care settings Open Access

Health-care is a complex, high-stakes vital service (Berry, 2019) that should be centered on the patient (Danaher et al., 2024), “providing the care that the patient needs in the manner the patient desires at the time the patient desires” (Davis et al., 2005, p. 953). Importantly, patient-centered care (PCC), views health-care as:

The Institute of Medicine (IOM) advocates PCC as critical to improving health-care [Institute of Medicine (IOM), 2001].

Yet, while PCC, as an ideal approach to health-care (Olson et al., 2021), has been adopted by many health-care organizations (Handley et al., 2021), including in hospital and primary care settings (Kuipers et al., 2021), and entered into the lexicon of health-care institutions, including their public relations department and that of a congressional representative (Epstein and Street, 2011), Edgman-Levitan and Schoenbaum (2021, p. 11) contend that a significant number of health-care organizations around the world still “tend to focus on the needs of their physicians and staff, rather than on their patients”, failing to understand that “patients need to be a partner and co-designer of all improvement activities, a concept many leaders and clinicians still find threatening or unnecessary”.

PCC is multidimensional, extending the “bedside manner” which traditionally has focused on giving comfort, listening attentively, showing respect and providing an empathetic response (Modic et al., 2020), and more recently, exchanging information, fostering relationships, shared decision-making and enabling self-management (Alpert et al., 2021). Indeed, the Picker Institute proposed eight general principles (Cramm and Nieboer, 2018; Hudon et al., 2012) that the Institute of Medicine (IOM) (2001) adopted emphasizing that “health-care in the United States must be improved, and that patient-centeredness is a key objective for improvement” (Rathert et al., 2012, p. 352).

Yet despite significant prior recognition of the importance of taking a PCC approach in health-care (McColl-Kennedy et al., 2017), as it contributes to high-value care (Handley et al., 2021), through improved clinical outcomes (Agency for Health Research and Quality, 2018), improving health and well-being (Edgman-Levitan and Schoenbaum, 2021) and increased service satisfaction (Tavares Alze Pereira dos Santos et al., 2017), much remains to be done to better understand the key principles, the relationships between the principles and key outcomes including subjective well-being and service satisfaction and importantly how to implement PCC in practice (Gallan et al., 2019) in different settings (Epstein and Street, 2011), including in traditional hospitals as well as online settings.

Accordingly, our study aims to address this critical gap by investigating three research questions:

1. how are the key principles of PCC related;

2. how do the relationships between key PCC principles and outcomes (subjective well-being and service satisfaction) vary depending on the channel providing the care (hospital/online primary care); and

3. what differences are placed on the involvement of family and friends in these different settings by health-care customers.

We make at least three key contributions. First, we model PCC as a multidimensional construct comprised of service providers respecting health-care customers’ values, needs and preferences; collaborative resources of the multi-disciplinary care team; health-care customers actively collaborating with their own resources; and health-care customers involving family and friends, identifying which specific key principles of PCC enhance subjective well-being and positively impact service satisfaction in different settings – hospital and online primary care settings. Second, and counterintuitively, we find that family and friends are not always viewed by health-care customers as being helpful in hospital and online primary care settings. In the online setting, involving family and friends was considered important for well-being but not for service satisfaction, pointing to the importance of recognizing differing opportunities for value cocreation in online settings compared to hospital settings and hence the need to provide digital resource options. Third, we show that there is an important difference in the role of family and friends for online primary care settings, depending on the online channel, i.e. if video or voice is used. These findings have important policy and managerial implications for health-care providers adopting PCC.

Clinician-led health-care focuses on delivering care to patients, with little regard for the patient’s values, needs or preferences (Danaher and Gallan, 2016; Edgman-Levitan and Schoenbaum, 2021; Gallan et al., 2019), and often little recognition is given to health-care customers actively collaborating in their health-care with their own resources, and the involvement of family and friends (McColl-Kennedy et al., 2012; Sweeney et al., 2015). A patient-centered perspective encourages efforts to understand the patient’s lived experience of the illness and the psychosocial context (Stewart et al., 2007), rather than the focus being on what is best for the provider (Vogus et al., 2020).

PCC as a philosophy is being adopted in hospitals and primary care clinics (Olson et al., 2021), with several service providers taking steps in practice to implement a patient-centered approach indicating the potential benefits of PCC including improved clinical outcomes (Epstein et al., 2005), improved service satisfaction (Agency for Health Research and Quality, 2018) and increased patient participation in their own care (Kaziunas et al., 2016).

Review of the nursing and medical literature underscores early evidence of PCC themes (Balint et al., 1969; Leino, 1949), incorporating this approach into medical training curricula in the USA and UK during the 1980s, as a sign of a growing acceptance of the philosophy of PCC (Luban-Plozza, 1995). In 2001, the US IOM revitalized interest in PCC to address health-care system improvement and hospital quality. The Institute of Medicine (IOM) (2001, p. 7) defines PCC as:

Eight general principles of PCC were developed with researchers from the Harvard Medical School in collaboration with the Picker Institute [Institute of Medicine (IOM), 2001]. Initially, drawn from Gerteis et al. (1993), IOM endorsed six PCC general principles: respectful of patients’ values, expressed needs and preferences; coordinated and integrated care; providing information, communication and education; ensuring the patient’s physical comfort; providing emotional support; involving family and friends; and further two principles were added by the Picker Institute, namely, continuity and transition and access to care [Institute of Medicine (IOM), 2001]. These eight general principles of PCC are consistent with Berghout et al. (2015), Handley et al. (2021) and Kuipers et al. (2019). Work by Langberg et al. (2019) identified sharing power and responsibility, which aligns with the latest development of the Picker principles of PCC, namely, involvement in decisions (Picker Institute, 2020).

Yet, while the philosophy of PCC has been adopted and shown to enhance clinical service improvement and hospital quality (Epstein et al., 2010), with positive effects on patient outcomes, service efficiency and quality of life, adequately measuring and implementing PCC has proven difficult in practice (Cramm and Nieboer, 2018; Epstein and Street, 2011; Langberg et al., 2019; Vogus et al., 2020). Rather than viewing PCC as multidimensional, many studies tend to focus on a single dimension such as shared decision-making (Scholl et al., 2018) or patient–doctor communication (Alpert et al., 2021; Danaher et al., 2023; Stewart et al., 2013). We admire Cramm and Nieboer’s(2018) attempt to address the lack of a validated instrument. However, their instrument is based on a sample of 216 respondents, and it was not developed using established protocols. Further, their analysis is based on correlation analysis, and they did not let the different dimensions compete in explaining the outcomes.

Web Appendix A provides a matrix of the general principles espoused in PCC theory and measurement instruments. Building directly from these general principles, and evidenced through our empirical studies, we view PCC as holistic care with service providers respecting health-care customers’ values, needs and preferences, in multi-disciplinary care teams and health-care customers actively collaborating with their own resources, involving family and friends.

Value cocreation is important to service research, and especially to health-care (Amorim and Ventura, 2023; Danaher et al., 2024) and well-being (Ostrom et al., 2021). It is also applied more broadly in consumer behavior research (e.g. Arnould et al., 2006) and in marketing (e.g. Epp and Price, 2008). Central to value cocreation is the view that value is cocreated from the integration of resources from a range of sources (Vargo and Lusch, 2008; Payne et al., 2008; Pham et al., 2021), not only from the service provider but from, for example, personal sources such as an individual’s own resources, e.g. knowledge, skills and abilities, with other resources of other actors in a given service ecosystem (Vargo and Lusch, 2014), such as another service provider/or service providers, even the customers’ family and friends or other customers (McColl-Kennedy et al., 2012; Vargo and Lusch, 2016; Vogus et al., 2020).

Rather than viewing a service provider as the only actor to have resources to provide to a customer for value to be realized, value cocreation takes a broader perspective, going beyond the dyad (service provider to customer) to a potentially much wider array of actors. Customers may cocreate value with other actors in addition to the focal firm, such as a hospital, clinic or with online health-care services. That is, health-care customers could potentially cocreate value with complementary medicine providers or therapists (McColl-Kennedy et al., 2012; Sweeney et al., 2015) or potentially more broadly still with community sources (Arnould et al., 2006; Gallan et al., 2019). Central to value cocreation is the notion of actors sharing in decision-making highlighted by McColl-Kennedy et al. (2012), one of the first empirical studies of value cocreation in health-care settings.

Consistent with value cocreation (McColl-Kennedy et al., 2012; Sweeney et al., 2015; Vargo and Lusch, 2014), we conceptualize PCC as based on the activities and integration of resources from the health-care system, comprising health-care service providers multi-disciplinary care teams in hospitals and online health-care services, as well as the health-care customers’ own resources, and involving their family and friends. As such, we argue that the PCC involves resources in the health-care system and health-care customer resources which include themselves and their family and friends. Figure 1 depicts our conceptual framework, which features our predictions about the internal relationships between the constructs of PCC, and relationships between PCC and subjective well-being and service satisfaction. Consistent with our definition and conceptualization, the featured key principles meet several criteria: they relate to the health-care professional (service provider) but also incorporate the health-care customer; they are purposeful and they focus on beneficial outcomes.

Accordingly, we specify four key principles of PCC:

1. service providers respecting health-care customers’ values, needs and preferences;

2. collaborative resources of the multi-disciplinary care team;

3. health-care customers actively collaborating with their own resources; and

4. health-care customers involving family and friends.

The first two sets of principles use resources from the health-care system, which means that they are principles that actors in the health-care system are responsible for conducting. The other two sets of principles are built on health-care customer resources. Mickelsson et al. (2022) argue that health-care customers use these resources to create their own ecosystem for care. The conceptual framework includes two important outcome measures of patient-centered care, subjective well-being and service satisfaction (Kuipers et al., 2019). In the following section, we outline the hypotheses in our conceptual framework.

Health-care providers respecting patients’ values, needs and preferences include showing concern, showing respect and being treated as an individual rather than doing “what can be done” for the patient (Berry et al., 2022, p. 164). Because health-care contexts feature substantial uncertainty and risk (Berry, 2019; Danaher and Gallan, 2016; Vogus et al., 2020), such interactions are necessary to build trusting relationships (Vandenbosch and Dawar, 2002; Gallan et al., 2019; Berry et al., 2022). In line with value cocreation, these interactions between technically competent professionals and patients should encourage a sense of social connectedness (Deci and Ryan, 2002; Ryan and Deci, 2000; Vogus et al., 2020) and result in the delivery of PCC and improved outcomes (Rathert et al., 2024). This social connectedness should function as a prerequisite for enacting better use of the health-care systems’ resources and the health-care customers’ resources (Rathert et al., 2022; Suarez-Alvarez et al., 2021). In line with patient-centered care, respecting health-care customers’ values, needs and preferences should enable the integration of resources from the health-care system, as well as the health-care customers’ own resources. Thus, the following hypothesis is proposed:

The involvement of multi-disciplinary teams is expected to further enact better utilization of the health-care customers’ own resources. Activating resources of the multi-disciplinary care team means that different actors in the health-care system would work respectfully with other members of the health-care team, sharing power and creating a cooperative atmosphere (Vogus et al., 2020). These multi-disciplinary care teams are important to patients, providing a positive and supportive culture during care that is mindful of personal dignity and integrity (Berry, 2019; Berry et al., 2022). Having access to resources from the health-care system should enable health-care customers to feel competent and able to make informed decisions despite information asymmetry (Shay and LaFata, 2015; Berry et al., 2022), while listening to the advice of multi-disciplinary care team members. Recognizing the role of multi-disciplinary teams and their importance to patients (Kitson et al., 2013), we hypothesize that having multi-disciplinary care team members working collaboratively sharing power, understanding what other team members can do and establishing a sense of trust among the team members provides a basis for health-care customers to use their own resources so that together they can achieve beneficial outcomes:

In line with the view of an active participant, health-care customers can undertake activities themselves related to their health, such as monitoring their diet, exercising, taking medications, actively researching their condition, bringing information to health-care service providers and coordinating support teams around their individual needs (McColl-Kennedy et al., 2012; Sweeney et al., 2015). Taking advice from others, such as family, friends and caring health-care professionals may be especially valued in complex health situations, rather than health-care customers applying solely their own health-care knowledge based on limited and/or incomplete resources. However, prior research suggests, that patients in oncology clinics who used their own resources to seek health information experienced a higher quality of life McColl-Kennedy et al. (2012), and increasingly patients wish to play a more active role in their health-care (Keeling et al., 2021). Danaher et al. (2024) show that, at least in a cancer setting, health-care customers investing their own resources are willing to strive toward a goal even if the associated activities are challenging and time-consuming. The benefits of getting customers to invest their resources are improved subjective well-being and service satisfaction. Therefore, we predict that health-care customers’ active collaboration will likely result in positive well-being and service satisfaction. Formally, we propose the following hypothesis:

This key principle involves activities that family and friends undertake to support the health-care customer during the health-care service experience, such as participating in decisions about care, aiding with treatment or care plans and engaging in discussions about the patient’s recovery. In line with prior research, patients can integrate resources from a range of sources not only through the health-care service provider–customer dyad but also through involving family and friends (Sweeney et al., 2015; Frow et al., 2016). Having access to these resources – from family and friends – is likely to foster a sense of internal harmony, feelings of coping from this connection with others (Deci and Ryan, 2002).

In particular, family and friends are now considered primary stakeholders in service experience by being involved in health-care decision-making cocreation given their contributions of support including emotional and instrumental support as well as counseling (Amorim and Ventura, 2023; McColl-Kennedy et al., 2012), thus enabling the health-care customer to have greater levels of satisfaction from the service experience. In qualitative work, Lam and Bianchi (2019) showed that involving family potentially has positive effects on well-being, but they did not investigate the nature of this relationship. Health-care customers who involve family and friends will likely see positive effects on subjective well-being and experience satisfaction with the service. Thus, we hypothesize the following:

Several research studies have provided evidence that there are systematic differences between face-to-face and online service provision. For example, in a retailing context, Bolton et al. (2022) show that online customers weighed cognitive and behavioral qualities more heavily than in-store customers. Lumivalo et al. (2024) argue that health-care customers using online health-care services might not be able to perform some activities online because of a lack of resources (e.g. sufficient medical knowledge). Based on this argument, when using a service online, a health-care customer is likely to be more dependent on using their own resources, hence we expect that health-care customers with better access to resources would have higher subjective well-being and service satisfaction. This suggests that in comparison to face-to-face health-care, we would expect that for online health-care, health-care customers actively collaborate with their own resources and health-care customers involving family and friends would have higher subjective well-being and service satisfaction. Thus, we hypothesize as follows:

To answer our research questions and provide a test of the conceptual model and our hypotheses, we collected an extensive amount of data in hospital and online primary care settings. First, we undertook a pre-study conducting exploratory qualitative interviews which aimed to better understand the health-care context with 47 respondents: 11 doctors, 16 nurses, 9 allied health professionals and 11 patients and their family/friends, associated with a large, tertiary public hospital and to assist with developing our survey instrument. Interviewees were encouraged to share their views and experiences of implementing PCC in practice. The interviews were recorded and transcribed and ranged from 30 to 90 min in length. All authors, except one, undertook face-to-face interviews providing pertinent details of what PCC meant in practice to the various respondents. We analyzed the interviews according to hermeneutic circle principles, until an integrated and comprehensive “account of the specific individual elements, as well as the text as a whole emerged” (Arnold and Fischer, 1994, p. 63).

Following our exploratory qualitative interviews, we collected four data sets of health-care customers (Table 1). In line with current practice (Cenophat et al., 2024), we wanted to ensure that the measurement tool captured the theoretical constructs and demonstrated adequate psychometric properties regarding the key principles of patient-centered care. The first two samples of health-care customers were recruited through an online national panel. Data collection for each sample took place over a three-week period. In the first sample, the health-care customers had experienced admission to a hospital for at least one night within the past six months (Sample 1, n = 272). As Table 1 details, this sample included 51.3% females, with an average age of 42.9 years and an average hospital stay of 4.8 days. The most common reasons for hospital admission were medical events (34.1%) or surgical procedures (32.7%). The second sample (Sample 2, n = 278) was based on the same selection criteria and included 54.5% females, with an average age of 43.5 years and the average length of stay in the hospital was equivalent to that in Sample 1. The reasons for their admissions were mainly medical events (26.3%) and surgical procedures (36.9%).

The basic premise for our research study was to test the conceptual model for health-care customers in a hospital (Sample 3) and compare the results with health-care customers in online primary care (Sample 4). For Sample 3, we administered a paper-based questionnaire in a large, metropolitan tertiary hospital and invited patients admitted to medical wards to participate after their admission. Data collection was undertaken over a 6-week period. Sample 3 (n = 275) consisted of 45.2% females, the average age was 54.2 years and the average length of stay was 7.3 days. The most common reason for hospital admission was for a surgical procedure (67.6%). Sample 4 (n = 297) consisted of 56.6% females; the average age was 45.6 years. The most common reasons for using online primary care were diabetes (14.5%) and hypertension (14.5%). Respondents were recruited through an international market research firm over a 3-week period. In addition, of the health-care customers using online primary care, 26.3% used voice-only communication, while the rest used a video service. The online service ranged from 5 to 145 min, with the average duration being 30 min.

Dependent variables. In our conceptual model, we focused on two important outcomes for health-care customers, subjective well-being and service satisfaction (Kuipers et al., 2019). We used widely used measures in health-care and marketing. For subjective well-being, we used an established scale with four items (Fox, 2004; Yun and Sim, 2021), and for service satisfaction, we used a three-item scale (Dagger and Danaher, 2014).

Independent variables. We conducted a comprehensive search using PubMed, Medline, CINAHL, PscyhInfo, Web of Science and Scopus databases for measurement scales related to “patient centered care” or “person centered care”, “patient centeredness”, “health-care customer-centered care” and “health-care customer centered practices” as keywords. We used established sub-scales from the literature to measure the eight general principles of PCC. Next, we undertook exploratory factor analysis (EFA) to investigate the dimensionality of PCC, and then confirmatory factor analysis (CFA) to validate the measurement model with different samples and discriminant validity tests (Böttger et al., 2017; Homburg et al., 2015). Finally, we conducted nomological validity checks and tested the scale for predicting subjective well-being and service satisfaction, using partial least squares (PLS) (Hair et al., 2011).

We identified 21 sub-scales that relate to our definition of PCC derived from our extensive search of the literature. Web Appendix B provides the sub-scales used and indicates which of the items were retained after analysis. Based on the identified theoretical constructs, we used measures from sub-scales for respecting patients’ preferences, needs and values (Dagger et al., 2007; Stewart et al., 2007), adopting appropriate communication styles (Campbell et al., 2007; Galassi et al., 1992), shared decision-making (Elwyn et al., 2003, 2013; Kriston et al., 2010; Simon et al., 2006), involving family and friends in care (Westbrook et al., 2015), empowering patients, (Johnson et al., 2012; Lerman et al., 1990; Small et al., 2013). Finally, two sub-scales measure involving multi-disciplinary care team members (Kenaszchuk et al., 2010; Orchard et al., 2012).

Next, because we model PCC as a multidimensional construct, we sought to develop multiple measures of the general principles of patient-centeredness, while minimizing overlap among factors. In line with prior theory and recommendations (e.g. Hudon et al., 2012), we include items adapted from existing sub-scales, using conventional approaches: expert judgments and statistical scale purification (Churchill, 1979) to reduce the number of items. Further, we asked an expert panel of 12 patients, health-care professionals and service researchers to rate the face validity and content validity of the initial pool of items and thereby eliminating redundant items. The resulting pool of 51 items entered an EFA.

Preliminary test of the scale involved a national sample of patients obtained through a national online panel (Sample 1, n = 272). An EFA with oblique rotation was conducted in SPSS v23. The statistical criteria for item retention were item-to-total correlations above 0.35 and standardized factor loadings above 0.5. We also considered item clarity, face validity and construct validity (Bearden et al., 2001). This process eliminated 23 items, owing to low factor loadings or cross-loadings on other factors. As Table 2 reveals, the EFA supports the four-factor solution (28 items) for PCC, comprised service providers respecting health-care customers’ values, needs and preferences with nine items (factor loadings 0.72–0.82, Cronbach’s alpha [CA] = 0.97), health-care customers involving family and friends with seven items (factor loadings 0.72–0.85, CA = 0.96), health-care customers actively collaborating with their own resources with five items (factor loadings 0.68–0.85, CA = 0.87) and resources of the multi-disciplinary care team with seven items (factor loadings 0.69–0.89, CA = 0.92). The standardized factor loadings and CAs thus exceed the acceptable thresholds. The average variance explained (AVE) in the EFA for Sample 1 is 76.1%.

To further refine the scale, we used one more sample: another online national panel (Sample 2). To validate the measurement and structural models, we performed CFA in AMOS v. 23, to determine the model fit using the analysis. As the final factors correlate, we used oblique rotation with maximum likelihood. The range of descriptive statistics suggests that non-normality is not excessive (Byrne, 2016). Therefore, we assessed each measurement factor as a congeneric latent variable. Following established practice (Byrne, 2016; Hair et al., 2017), we determined the measurement model fit using the comparative fit index (CFI), incremental fit index (IFI), normed fit index (NFI), Tucker–Lewis index (TLI) and root mean square error of approximation (RMSEA). The final model fits the data well (see Table 3). In Sample 2, we find χ² = 237.3, df = 113 (CFI = 0.97, IFI = 0.97, NFI = 0.94, TLI = 0.96, RMSEA = 0.06). We also confirm convergent validity, in that all AVE values are greater than 50%. In addition, the AVEs exceed the squared correlations between factors (Bearden et al., 2001; Fornell and Larcker, 1981), suggesting good discriminant validity. The final model contains 17 items, with four distinct dimensions corresponding to the key principles of PCC identified in our conceptual framework (Table 4).

The hypotheses in our conceptual model were tested through PLS using SMARTPLS (Ringle et al., 2022). The use of PLS for testing the hypotheses is motivated by the relatively small data sets, and the objective to explain how PCC affects subjective well-being and service satisfaction (Hair et al., 2011). Before performing the analyses, the data were checked for systematic missing values in key variables. Some cases were deleted based on too many missing values; or no variation in responses (e.g. had failed to detect “attention checks” in the questionnaire). The reliability and validity of the measurement model were tested using the recommended procedures (Fornell and Larcker, 1981; Hair et al., 2012). All items were loaded appropriately on the intended constructs, and all loadings reached the recommended level of 0.70 (Hulland, 1999). In addition, Cronbach alphas and composite reliabilities were higher than the recommended threshold of 0.70 (Hair et al., 2011), and the AVEs higher than the suggested threshold of 0.50 (Fornell and Larcker, 1981). To ensure the discriminant validity of the constructs, the AVEs of the latent constructs were compared to the square of the correlations among them (Chin, 1998). One loading did not reach the recommended threshold in each model but was kept, enabling comparison between the two samples. Discriminant validity was examined using Heterotrait–Monotrait (HTMT) (Hair et al., 2022). All the ratios for this study are below the recommended thresholds in HTMT guidelines. To assess the statistical significance of our proposed relationships, we applied a nonparametric bootstrapping procedure using 5,000 subsamples with 275 cases and individual sign changes (Henseler et al., 2009). Moreover, the Stone–Geisser Q² values for each outcome variable, computed using PLS predicted, range between 0.01 and 0.43 (i.e. are well above zero), in further support of the predictive power of our model. The analyses for Sample 4 followed the same procedures. See Table 5 for descriptives. See also Web Appendix C for loadings and cross-loadings for Sample 3.

The hypotheses were tested through the operationalization of our conceptual model using Sample 3 (Hypotheses 1–4) and Sample 4 (Hypothesis 5). To provide a strong test of the hypotheses, Sample 3 was collected for patients who currently were in the hospital. First, service providers respecting health-care customers’ values, needs and preferences had a positive impact on resources of the multi-disciplinary care team (⁠$β1a = 0.66, p<0.01)$⁠, and on health-care customers actively collaborating with their own resources (⁠$β1b = 0.33, p<0.01)$ supporting H1. Second, resources of the multi-disciplinary care team had a positive impact on health-care customers actively collaborating with their own resources (⁠$β2a = 0.28, p<0.01)$⁠, and health-care customers involving family and friends (⁠$β2b = 0.39, p<0.01)$⁠, supporting H2. The support for H1 and H2 shows that using resources of the health-care system contributes to health-care customers using their own resources.

Third, in support of H3, health-care customers actively collaborating with their own resources has positive impacts on (a) subjective well-being (⁠$β3a = 0.15, p=0.053)$ and (b) service satisfaction (⁠$β3b = 0.40, p<0.01)$⁠. It should be noted that the effect on subjective well-being is marginally significant, hence why we have chosen to display the exact p-value (McShane et al., 2024). Fourth, health-care customers involving family and friends have positive impacts on subjective well-being (⁠$β4a = 0.14, p<0.01)$ and satisfaction with the service (⁠$β4b = 0.15, p<0.01),$ supporting H4. For consistency, we tested the model for Samples 1 and 2, which displayed similar results. The test of H1 through H4 provide support for the structure of our conceptual model and that PCC has a strong influence on subjective well-being and service satisfaction. Examining the f² values, it should be noted that the model does a better job of predicting service satisfaction than subjective well-being. We argue that this to a large extent can be explained by what kind of illness a patient is being treated for, which has a large influence on subjective well-being.

Finally, the last hypothesis concerns health-care customer resources (health-care customers exercising their own resources and health-care customers involving family and friends). We formally tested H5 through a multi-group analysis using permutation (Hair et al., 2018). This enabled us to test for differences in the conceptual model for different groups. Multi-group PLS-SEM is a type of moderator analysis, where online primary care compared to hospital health-care service is used as the grouping variable (Henseler and Fassott, 2010). A detailed account of the results can be found in Table 6 (last column), where significant differences (p < 0.05) between the two samples are displayed. The results show that there are significant differences between online primary health-care and hospital health-care service concerning how health-care system resources influence the use of health-care customer resources; service providers respecting health-care customers’ values, needs and preferences → health-care customers actively collaborating with their own resources, collaborative resources of the multi-disciplinary care team → health-care customers actively collaborating with their own resources. While these differences were not predicted in our hypotheses, we will discuss them in the implications section.

Regarding the key relationships for H5, the results of the multi-group analysis show that there are significant differences between hospital and online primary care concerning two of the four relationships on how health-care customer resources influence outcomes. The relationship between health-care customers involving family and friends and subjective well-being follows the prediction in H5 (0.40 > 0.14, p < 0.05), but the other two relationships do not reach statistical significance in the hypothesized direction. In one instance, the relationship between health-care customers involving family and friends and service satisfaction, there is a significant relationship in the opposite direction than predicted (0.15 > −0.01; p < 0.05). Comparing the four potential paths, we can see that for three of them the influence is lower for the online sample, hence H5 is rejected.

To further investigate why health-care customer resources in online primary care do not have as large an effect as expected, we performed a post-hoc analysis of the sample for online primary care. In our sample of online primary care, 26.3% of the patients received care through voice only, while the others received care via video. We again performed a multi-group analysis in PLS, comparing the conceptual model between the two types of online care. The moderating effects of voice only/video were analyzed through a multi-group analysis, as the voice only/video was categorical (Henseler and Fassott, 2010).

The multi-group analysis can be found in Table 7. The key result shows that there is an effect of voice only/video on the relationship between involving family and friends and service satisfaction. This effect is stronger for video, in comparison to voice only (⁠$deltaβ = 0.25 p<0.05)$⁠. Taken together, there is a positive effect for family and friends on satisfaction with online primary care, but only if care is provided through video. The effect of involving family and friends is however not larger than it is for care in the hospital, but what becomes evident is that how care is provided in online primary care determines to what extent the health-care customer uses their own resources.

We empirically test our hypotheses about the impacts of PCC on service satisfaction and subjective well-being; leveraging value cocreation theory which underscores the importance of multiple actors integrating resources from a range of sources (Danaher et al., 2024; McColl-Kennedy et al., 2012). As such, our study provides a foundation for subsequent integrative theoretical models that pursue beneficial outcomes for multiple stakeholders. Increasingly, organizations are concerned about service experiences in a health-care setting, particularly what health-care customers think and feel (McColl-Kennedy et al., 2019), the role of digital technology in service provision (Rosenbaum et al., 2017; Ostrom et al., 2021) and their well-being (Field et al., 2021). Our research shows that especially in hospital, our conceptual model better explains service satisfaction than subjective well-being. This shows that although satisfaction with the care in a hospital is determined by how a patient is treated, the multifaceted nature of well-being is more complex and does not appear to be influenced directly by the care provider.

The conceptual model provides a novel approach to PCC delineating that resources in the health-care system are used to activate the health-care customer resources toward greater subjective well-being and service satisfaction. While the key dimensions of PCC may not positively impact on how a patient feels physically due to the illness they have and/or are recovering from during hospitalization or online primary care, these dimensions have positive implications for how patients feel about themselves and their lives when reflecting on their health-care experience as measured by the subjective well-being scale. This finding is important. Respecting patients’ values, needs and preferences cannot treat an illness or reduce the time needed for the body to heal from surgery. Understandably, the patient is still likely to feel pain, nausea, tiredness and remains unable to undertake desired activities in the immediate short term while ill or still recovering. However, and importantly, the patient’s active involvement in health-care activities helps them feel better in terms of feeling positive about their life (Ryan and Deci, 2000), contributing to their subjective well-being.

Our findings of the beneficial outcomes of collaborative activities are in sharp contrast to much of traditional health-care where the focus has been on the provider (Edgman-Levitan and Schoenbaum, 2021; Vogus et al., 2020). As Edgman-Levitan and Schoenbaum (2021, p. 11) point out, many health-care organizations around the world still “tend to focus on the needs of their physicians and staff, rather than on their patients”, failing to understand that “patients need to be a partner and co-designer of all improvement activities.” In line with this view, patients were frequently excluded from decisions regarding their care plans with health-care professionals working independently, not within a collaborative multi-disciplinary care team.

Many customers want to play a more active role in their health-care (Keeling et al., 2021; Danaher et al., 2024). To fulfill their desires for autonomy and competence, patients need to be empowered, such as by providing information about their health to health-care professionals, refusing suggestions regarding treatment, realizing that they may change their mind about care plans and being able to talk to a health-care professional further if they do change their mind. Another critical component, consistent with the desire for relatedness especially to a significant person important in the individual’s life, involves family and friends and all multi-disciplinary care team members. Creating a cooperative, inclusive dynamic with the patient and among care team members, while addressing the patient’s situation, sharing power, acknowledging boundaries and establishing a sense of trust enhances a patient’s ability to cope and remain positive while experiencing a stressful situation. In summary, our work highlights the importance of taking a customer-centric cocreated approach integrating resources from the health-care system with the customer’s resources – a radical shift in thinking and behavior from the health-care provider-led perspective (Danaher et al., 2024; Vogus et al., 2020).

We make the following theoretical contributions. First, to the best of our knowledge, we are the first to model the four key principles of PCC:

1. service providers respecting health-care customers’ values, needs and preferences;

2. collaborative resources of the multi-disciplinary care team;

3. health-care customers actively collaborating with their own resources; and

4. health-care customers involving family and friends, explicating which principles of PCC have positive effects on subjective well-being and service satisfaction in different settings.

Consistent with value cocreation in health-care (McColl-Kennedy et al., 2012), we demonstrate that patient well-being is best served when PCC uses resources in the health-care system with the health-care customers’ resources. Our analyses validate claims that PCC has widespread benefits for individual patients and reflects the contemporary view of patients as empowered, active value cocreators (Alpert et al., 2021; Gallan et al., 2019; Danaher et al., 2024), not merely passive recipients of what can be done for them (Berry et al., 2022; Edgman-Levitan and Schoenbaum, 2021).

Second, and counter to what we expected, we find that for online primary care, the influence of health-care customer resources on subjective well-being and service satisfaction is not higher than for care provided in a hospital setting. The underlying assumption is that when a health-care customer is receiving care online from their home, their resources would be key in their satisfaction, but our results do not support this. Interestingly, family and friends were not always viewed by health-care customers as being helpful in hospital and online primary care settings. In the online setting, involving family and friends was considered important for well-being, but not for service satisfaction. It is important to understand the drivers of both well-being and service satisfaction and understand that even if the health-care customer is using their own resources, the use of these resources might have different effects on well-being and satisfaction with the service encounter, see McColl-Kennedy et al. (2017).

While it may appear at first that a value cocreation perspective (McColl-Kennedy et al., 2012) – which emphasizes the integration of resources from various sources – does not provide a good explanation across settings, it is important to recognize that some resources may be more important than others in different settings. For example, in the online setting, the duration of the video/phone call was significantly less than the time spent in hospital and so the health-care customers could not easily “view” the integration of resources in the health-care system, such as how team members work together, sharing power and so on, nor did the health-care customer have the opportunity to use their own resources to the extent they could in hospital where they could discuss treatment options more fully, take information to the health-care professional and involve family and friends in making decisions about care.

Alternative theories such as construal level theory (Bolton et al., 2022) or psychological ownership (Chen et al., 2021) might be better placed to explain the identified differences regarding satisfaction and well-being, respectively. As an example, Bolton et al. (2022) suggest that customers use the original channel as their referent channel and that this influences how customers evaluate their care. Accordingly, online care is likely to perform worse on several core characteristics of care.

Third, our post-hoc analysis shows that there is an important difference in the role of family and friends in online primary care settings, depending on whether video or voice only is used. This finding highlights the importance of building the right prerequisites for online primary care not only inside the health-care system, but also in the health-care customer ecosystem (see also Snyder et al., 2019). This supports and extends research by Lumivalo et al. (2024), who argue that the lack of sufficient medical knowledge by health-care customers limits the potential benefits of health-care. Our research shows that there is a need to build digital resources for health-care customers to be able to co-create appropriate care (Orsingher et al., 2024). Critically, we find there is a difference in outcomes if the health-care customer can only access care through voice and not through video, and this limits the use of involving family and friends in their care. This finding highlights the need to deeply understand the implications of using digital technology in the delivery of health-care, especially the limitations and modifications that are required for effective delivery (Green et al., 2016).

Practically, our work identifies specific PCC dimensions that not only can improve service satisfaction but can enhance patients’ well-being, which is foundational to better health-care (Kuipers et al., 2019). In particular, we offer a parsimonious practical tool designed to enable managers to assess the extent to which PCC is adopted by health-care professionals overall and which particular principles are being implemented, and importantly, where to focus resources to improve subjective well-being and satisfaction.

Respecting patients’ values, needs and preferences should focus on personalized and authentic interactions that support high-quality care, (Vogus et al., 2020; Hunter-Jones et al., 2020). Involving family and friends has widespread benefits for patients’ well-being, enhancing their feelings of social connectedness while providing support and encouragement during times of stress. Our work highlights the criticality of co-designing health-care with patients which takes time and effort. Health-care professionals can encourage their patients to learn more about their disease and collaborate with them to address complex health-care problems and be actively involved in co-designing their care plans (McColl-Kennedy et al., 2012; Danaher et al., 2024).

Likewise, involving all multi-disciplinary care team members nurtures a culture of trust, where individuals are prepared to use their own initiative and solve problems. In different health-care contexts, it is especially important to understand which specific activities contribute most effectively to better health-care outcomes. This is even more important in online health-care settings, where it is the use of health-care resources that activates the health-care customer to use their own resources in their care. Snyder et al. (2019) show that the customer has the best knowledge about care in their own context, thus why they can contribute with solutions in their daily life that are both creative and valued. But for this to happen, health-care customers need to see that the health-care system invests their resources in taking care of the health-care customer.

Policymakers should leverage key insights promoting activities that focus on the patient, as they can benefit well-being and lead to improved health-care outcomes. As prior work shows, better patient experiences improve the clinical effectiveness of treatments for a wide range of diseases, with benefits for many people (Jeon et al., 2010). Our study goes a step further - we identify which PCC activities influence the health-care customer (patient) experience most positively in different settings. Accordingly, our results highlight where policymakers should invest in – training – to ensure that staff truly respect patients’ values, needs and preferences, as well as encouraging patients to involve family and friends and use their own resources thus helping to reduce barriers to PCC implementation such as having an employee rather than patient focus, prioritizing provider incentives, focusing on standardizing care and organizational risk (Vogus et al., 2020).

Further, our work demonstrates the importance of encouraging care team members to spend more time working with patients to effectively coordinate and integrate care and making practical provisions for family and friends to be involved during hospital and online primary care to achieve beneficial outcomes for health-care customers, service providers and society. In the hospital settings, our results suggest that a health-care provider that enables the health-care customer to involve family and friends contributes to higher service satisfaction and subjective well-being. For patients going through treatment for major illnesses, this may be a key to enable them to have the energy needed to manage the often stressful experience associated with a major illness. In online settings, the effects are different, as involving family and friends only contributes to subjective well-being. However, having appropriate digital assets are important to video and other newer digital technologies such as AI assistants and should be offered to provide more detailed knowledge of the treatment options available, as well as knowledge of what the care team is doing beyond the video/phone call itself.

While Doherty et al. (2020) point out that health-care service providers see it as a challenge to let go of their traditional way of providing health-care and change to partnering with patients, we have shown that it is important, and that we can expect PCC to have positive results. Although our research scratches the surface of this critical issue, it suggests an important direction for future research to further investigate implementing PCC in practice and to recognize and limit barriers to implementation.

While we undertook a series of studies comprising four surveys across hospital and online primary care settings, the limitations of our work suggest several important areas for further investigation. First, our way of measuring PCC should be tested in additional health-care contexts, across geographies and cultures, such as third-world countries where PCC may be viewed differently (Field et al., 2021). In addition, testing other service contexts where well-being is relevant such as in vulnerable, disadvantaged and marginalized customers and communities including for instance, in aged care settings and with people with disabilities, would be a worthwhile extension (Field et al., 2021).

Second, our study captures PCC principles at a specific point in time. Longitudinal investigations of PCC and how their relative importance changes over time during a health episode could be informative as well, as demonstrated by Danaher et al. (2024) in cancer settings. For example, principles that support relatedness with others may be especially important when patients experience high levels of stress and strong emotions (Berry, 2019), such as when patients are about to undergo a surgical procedure. In addition, our research suggests differences in the importance of various activities on well-being that depend on the timing of the patient journey. Exploring the patient journey before, during and after hospitalization may highlight important differences in the significance of each of the principles and their respective impact on well-being.

Third, future research should investigate the service design of online health-care given that it is increasingly viewed as a valid alternative to face-to-face health-care, especially since the pandemic (Smith et al., 2020). Our post-hoc analysis provides initial evidence that the design of the online channel can influence how customer resources influence the outcomes. Future studies should explore more in-depth differences in experiences when video versus voice only is used, both from the health-care customer and service provider perspectives, to determine how health-care could be provided in an online channel to have a substantial effect on health-care customers. To further explore implications of the design of the online channel, future research could explore differences resulting from chatbots and virtual service robots. An additional challenge is to find identical patients who are treated in the hospital and online. Our samples differ in the medical conditions of patients which could influence results.

Finally, we undertook four surveys designed to capture PCC and perceptions at discrete points in time. Alternative approaches could reveal additional insights about the most beneficial specific PCC activities in real-time. Digital technology-based methods, in particular, could monitor the patient experience in real time. For example, tracking devices such mobile applications (e.g. Orsingher et al., 2024), could capture real-time data about how customers feel, their well-being, their sense of being in control over time (Baxendale et al., 2015). In health-care settings, such tracking digital technology may prove especially valuable to enable appropriate ways for sharing test results with patients and their families or ensuring that the multi-disciplinary care team is involved in patients’ treatment and discharge plans. Further, social media platforms could allow patients and health-care professionals to report their reflections on different patient-centered activities in specific contexts and in real time. We encourage future research along these promising avenues.

The authors greatly acknowledge research assistance provided by Dr Nazila Babakhani, Dr Fiona Willer and the financial support provided by the Australian Research Council Linkage Projects Grant #LP150100629. In addition, the authors also greatly appreciate the helpful comments and suggestions provided by the editor and the review team.

The supplementary material for this article can be found online.