AbstractSection Background

Capturing subjective user experiences is crucial for service design, assessment, and management. However, using metrics for such experiences remains an underexplored topic in the literature. This study introduces a novel survey tool to measure service-related subjective well-being (SWB) among oncology patients, encompassing the three dimensions of SWB: positive affect, negative affect, and life satisfaction. The hypotheses are that (H1) highly reliable service-related SWB metrics will be generated based on reliable tools from psychology, and (H2) perceived service experience will strongly predict positive affect and life satisfaction, but they will have less predictive power for negative affect. These hypotheses are supported by research suggesting that services can improve positive affect but may not fully counteract the negative affect evoked by cancer.

AbstractSection Methods

We conducted a cross-sectional survey of 288 oncology patients using convenience sampling. The online survey was divided into three segments: patient information, SWB measures, and perceived service experience. The SWB measures encompassed its three dimensions. The survey items were adapted from consolidated psychological research scales for this study. This theory-driven approach, known as the ‘construct method,’ involves generating hypotheses based on construct theory and testing them empirically. Participants completed the survey within seven days of hospital discharge. Our data analysis included reliability measures, exploratory factor analysis on the SWB items, and regression analyses to examine the predictive power of perceived service experience on SWB.

AbstractSection Results

The survey demonstrated high reliability (Cronbach’s Alpha = 0.946). Factor analysis indicated a two-factor model, explaining 78.01% of the total variance, with a separation of positive and negative items. Regression analysis revealed that perceived service experience substantially explained SWB. Positive affect and life satisfaction were well-predicted by perceived experiences with nursing, medical, and infrastructure, while negative affect was less well-predicted. H1 was confirmed by high-reliability metrics, and H2 was supported by the regression results.

AbstractSection Conclusions

The survey can identify areas for improvement, optimize services, and measure the relationships between perceived service experience and SWB, contributing to the literature on healthcare user experience and service design metrics.

Investigating patient experience can contribute to understanding their well-being and fostering service improvement initiatives in healthcare institutions. Services focused on cancer treatment are complex and need to adapt not only to the patient’s clinical conditions but also to their emotional state, preferences, and family support, among other factors [1]. Dealing with these patients requires the development of appropriate metrics to map their subjective experiences and, consequently, promote better service design^{Footnote 1}.

Cancer and its treatment profoundly affect patients’ quality of life, health, and family relationships, impacting their physical, emotional, psychological, and social well-being [2]. Numerous studies have examined how services influence subjective well-being (SWB), encompassing both positive and negative emotions and their effect on life satisfaction, the three components of SWB [3, 4]. While services often incorporate positive interventions, such as playful distractions during medical procedures [3], the inherent challenges of cancer treatment limit the ability to mitigate the negative impact of hospitalization [5]. Therefore, although interventions can alleviate psychological distress, some of the disease’s broader effects remain beyond control. This underscores the importance of understanding patients’ experiences.

Patient-reported outcome measures (PROMs) and patient-reported experience measures (PREMs) have become integral components of healthcare evaluation and service management. PROMs assess patients’ health status and functional outcomes, such as physical functioning, pain, and symptom severity, while PREMs evaluate their perceptions of care delivery and service quality [6]. PROMs and PREMs have been implemented in cancer care settings globally [7]. However, they are not designed to measure the broader emotional and cognitive dimensions of service-related SWB. Recognizing this gap, our study develops a survey tool that integrates established SWB measures with perceived service experience measures. This gap reinforces the need for a new approach that combines psychological insights with service design.

Service designers report difficulties collecting and analyzing quantitative measures and a lack of expertise in such methods [8], reinforcing the need for studies such as the one presented in this paper. Given this context, the objectives of this study were to develop a reliable service-related SWB survey tool for patients in the context of hospital oncology treatment and, based on this tool, evaluate the extent to which perceived service experience can predict users’ SWB. This study leverages consolidated metrics from psychology and aims to fill the gap in design-focused studies by developing a specialized SWB survey. The goal is to provide a tool that supports service design practices to improve patient experience. Its main purpose is not to describe the user experience with oncology services but rather to focus on developing metrics to assess service-related SWB.

To address these limitations and enhance our understanding of the relationship between perceived service experience and SWB, this study tests the following hypotheses: (H1) Highly reliable service-related SWB metrics will be generated based on reliable existing tools. (H2) Users’ perceived service experience will be strong predictors of positive affect, life satisfaction, and overall SWB; however, they will have a lower predictive power for negative affect. These hypotheses align with the evidence presented above. While purposeful efforts in service development may increase positive experiences, they may be limited in mitigating the negative impact of cancer on SWB.

The following subsections provide an overview of PROMs and PREMs, discuss SWB in oncology care, and present the measurement of SWB, grounding the study.

Patient-reported outcome measures (PROMs) and patient-reported experience measures (PREMs) have become essential tools for evaluating healthcare services from the patient’s perspective. While they are often discussed together, they assess different dimensions of healthcare quality and patient care. PROMs focus on health outcomes—how patients feel and function—whereas PREMs assess care delivery—how patients perceive the process and quality of care. Understanding both PROMs and PREMs is essential for improving oncology care, where patient well-being is influenced not only by clinical outcomes but also by the care experience itself.

PROMs gather direct feedback from patients about their health status, symptoms, and functional outcomes. They complement clinical and functional data by converting subjective experiences into measurable data, shifting the focus from disease-specific metrics to how patients feel and function in their daily lives. PROMs provide valuable insights into treatment effectiveness and help identify variations in healthcare outcomes across different patient populations [9]. They are typically based on structured, standardized scales—often using 5- or 7-point Likert-type formats—to ensure consistency across healthcare settings. Guidelines from initiatives such as OECD [6] and ICHOM [10] have established best practices for PROM development and validation, enabling consistent measurement and benchmarking.

Widely used PROMs not specific to cancer include the EuroQol 5-Dimensional family of tools (EQ-5D) [11, 12] and the 36-Item Short Form Health Survey (SF-36) [13]. The EQ-5D measures five dimensions of health (mobility, self-care, usual activities, pain/discomfort, and anxiety/depression) with response options ranging from three to five levels, depending on the version. The SF-36 measures eight health domains, including physical functioning, bodily pain, general health perceptions, and emotional well-being, offering a comprehensive view of health-related quality of life.

More recent developments in PROMs have focused on improving sensitivity and adaptability. The Patient-Reported Outcomes Measurement Information System (PROMIS^®) uses computer-adaptive testing to tailor questions based on previous answers, increasing precision and reducing response burden [14, 15]. PROMIS measures physical health (e.g., pain, fatigue), mental health (e.g., depression, anxiety), and social health (e.g., social function), offering a nuanced view of patient well-being.

Several PROMs are specific to cancer care. In a recent study [7] related to the Italian PRO4 All Project, which focused on developing a comprehensive library of PROMs and other measures of patient-reported outcomes in oncology, 356 PROMs were identified. Similar proportions were cancer type-specific (46.9%) and generic for cancer (45.2%), while 7.9% were originally developed for the general population but later recommended for oncology. In another publication from the PRO4 All Project [16], breast cancer was associated with 29 specific PROMs, further highlighting the degree of specificity that PROMs can reach.

Examples of PROMs in cancer care include the Functional Assessment of Cancer Therapy – General (FACT-G) [17], which is widely used to assess physical well-being, social/family well-being, emotional well-being, and functional well-being. The European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30) [18] is designed for cancer clinical trials and evaluates physical, role, cognitive, emotional, and social functioning, along with symptom scales for fatigue, pain, nausea, and vomiting. The MD Anderson Symptom Inventory (MDASI) [19] measures the severity and interference of cancer-related symptoms on daily functioning. The Patient-Reported Outcomes version of the Common Terminology Criteria for Adverse Events (PRO-CTCAE) [20] assesses the frequency, severity, interference, and presence/absence of symptomatic toxicities that include pain, fatigue, nausea, and cutaneous side effects from the patient’s perspective. The Edmonton Symptom Assessment System (ESAS) [21] is a tool that evaluates the intensity of common cancer symptoms such as pain, fatigue, and nausea, particularly in palliative care settings.

PREMs, on the other hand, capture how patients feel about the care they receive [22]. PREMs also rely on Likert-type scales or binary yes/no questions, with some using global rating questions. Like PROMs, PREMs undergo rigorous psychometric testing to ensure reliability and validity.

Widely used PREMs not specific to cancer include the Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) [23] and the Consumer Assessment of Healthcare Providers and Systems (CAHPS) [24]. HCAHPS assesses patients’ perceptions of hospital care, focusing on communication with nurses and doctors, responsiveness of staff, cleanliness and quietness of the environment, communication about medicines, discharge information, overall rating of hospital, and potential recommendation of the hospital. CAHPS, in contrast, is broader in scope, evaluating experiences with, and ratings of, a variety of healthcare settings, including also outpatient care and health plans. Additionally, the NHS Friends and Family Test (UK) [25] measures whether patients would recommend a service to friends and family if they needed similar care or treatment.

In cancer care, targeted PREMs have been developed to address the specific needs of oncology patients. The PREM-C (Patient-Reported Experience Measure for Cancer) [26] assesses respect for values, preferences and expressed needs; physical comfort; emotional comfort; involvement of family, and friends; coordination and integration of care; and information, communication, and education. The PREM-ECM (Patient-Reported Experience Measure for Experimental Cancer Medicine Trials) [27] focuses specifically on the experiences of patients in cancer trials, including factors such as information and understanding, decision-making (e.g., consenting to participate), support (e.g., support provided during the screening phase and during the trial), and how participation affects patients and how side effects are managed. Additionally, the PREM Item Bank for Outpatient Cancer Care [28] is a flexible tool that assesses outpatient care quality, including organization of healthcare, competence of professionals, communication, and patient empowerment.

PROMs and PREMs have been successfully implemented in oncology care settings worldwide. They provide distinct but complementary insights into patient care. However, despite their broad application, PROMs and PREMs are not designed to measure the broader emotional and cognitive dimensions of service-related subjective well-being (SWB).

Research and practice have introduced a wide range of constructs and measures to capture patients’ perspectives on healthcare [29]. These advances highlight the growing recognition that excellence in healthcare extends beyond clinical quality and safety, emphasizing the importance of patient-centered care [30].

Building on this understanding, healthcare service delivery has undergone a significant transformation, shifting from transaction-based models to user-centered approaches [31, 32]. In these models, providers collaborate to improve services and experiences through various stages: exploring users’ experiences and expectations, ideating future services, prototyping to test them with users, and implementing them. Consequently, patient engagement in service design became a reality [33, 34]. However, there is a substantial difference between desired services, where users choose what, how, and when to use, and undesired services, such as medical treatment [35]. In our study, user experience is examined through the lens of subjective well-being (SWB).

Discussions related to what is defined as SWB today are not new. They date back to ancient times, with thinkers such as Aristotle exploring the concept of eudaimonia, often translated as “flourishing” or “the good life.” In the 1980 s, Diener [36] defined subjective well-being as an individual’s cognitive and affective evaluations of their life, emphasizing the importance of self-assessment.

While objective well-being traditionally refers to the tangible aspects of life that contribute to a person’s quality of life—such as economic status and housing—research increasingly recognizes that subjective and objective well-being are interconnected rather than entirely separate constructs. Objective life circumstances, such as financial security or health status, can influence subjective states, while positive emotional and cognitive evaluations can, in turn, affect physical health and social functioning. This integrated perspective highlights the importance of addressing both subjective and objective dimensions of well-being in service design [37,38,39].

In healthcare, SWB represents a patient’s assessment of how their experiences impact their overall well-being, making it highly relevant for designing services rooted in care, ethics, and respect for the individuals involved. SWB goes beyond a direct assessment of these experiences, hinging instead on individuals’ subjective evaluations of life events, which makes it a self-report measure [40]. Thus, SWB is not solely a result of intuitive, emotional assessments as it encompasses cognitive aspects [41]. SWB is intertwined with three key components: positive emotions, negative emotions, and life satisfaction [36, 42,43,44].

Oncology treatment can influence patients’ SWB [45,46,47]. In the realm of positive emotions, for instance, feelings of hope, joy, and contentment may be experienced due to a supportive staff and a caring environment. However, oncology treatment can also be associated with negative emotions, such as anxiety, sadness, or frustration, due to medication side effects, surgical complications, and the hospital environment. Ultimately, for oncology patients, life satisfaction can be influenced by a multitude of factors, such as the availability of a robust support system and the effectiveness of their treatment. This highlights the imperative for a comprehensive approach to oncology care that encompasses both the medical and emotional needs of patients, thereby guiding the design of patient-centric services [48] targeting the enhancement of users’ SWB [3, 4].

Designing a healthcare service centered on SWB presents numerous challenges. Crafting experiences to evoke positive emotions or neutralize negative ones requires a profound understanding of those involved. Thus, designing services focusing on SWB aligns with pragmatic approaches that attempt to “predict” the potential effects of services on people and leverage this information in service development [49].

The debate regarding the best approach to conceptualize and measure the affective aspect of SWB persists. This underscores the potential for analyzing the multitude of emotions experienced during a cancer patient’s treatment journey. For this reason, this study focuses on developing a survey tool to measure the service-related SWB of cancer patients.

Even though there are no self-report measures of SWB in design research, other fields (e.g., psychology) have developed such measures. SWB is regarded as a self-perceived assessment of one’s life quality, encompassing both affective and cognitive evaluations. Affective components include emotions such as joy and sadness, while cognitive components involve life evaluations with respect to ideals. The measurement of SWB, especially its affective component, remains a topic of debate. The affective component is often assessed by how frequently emotions are experienced rather than their intensity. Item Response Theory offers a nuanced analysis of SWB measures. It considers the probability of responses as a function of an underlying trait, allowing for more precise reliability and validity estimates across different trait levels [50].

Among validated surveys available, two are of particular interest for this study due to their conciseness and one-of-a-kind excellent psychometric qualities: The Positive Affect Negative Affect Schedule – PANAS [51] – and The Satisfaction with Life Scale – SWLS [52].

The Positive Affect Negative Affect Schedule (PANAS) is a tool for measuring the affective aspects of SWB that has been used for many decades since its development. It consists of 20 items, divided into two parts with 10 adjectives to assess positive and negative effects. These adjectives were selected based on their strong correlation with one affect dimension and weak correlation with the other. The survey has demonstrated high internal consistency in various studies. Some studies suggest a two-factor model with positive and negative affect as uncorrelated or correlated factors, while others propose subfactors within these two main categories. There are short versions of PANAS with five items for positive affect and five for negative affect [53].

The Satisfaction with Life Scale (SWLS) is known as a key measure for the cognitive component of SWB. It was originally a pool of 48 items; however, the instrument was refined to just five items. In various studies, SWLS has shown high internal consistency and predominantly exhibits a single-factor structure [54,55,56].

In summary, while we have yet to establish service-related measures of SWB, the adoption of validated tools such as the PANAS and SWLS offers a valuable foundation for assessing the affective and cognitive dimensions of SWB. These tools can inform and enhance the human-centric approach to healthcare service development.

This study was conducted at the renowned philanthropic institution AC Camargo Cancer Center. The center’s ethics committee reviewed and approved the study, granting us access to patients for the development of metrics to assess service-related SWB in a practical setting. The study focused on patients who had recently undergone cancer care, as described in the following sections.

The AC Camargo Cancer Center, with around 400 beds, serves both public health system patients and private individuals with insurance. Since its inauguration on April 23, 1953, the center has gained international recognition for its multidisciplinary, integrated approach to cancer diagnosis, treatment, teaching, and research, mirroring the world’s leading cancer centers. Dedicated to advancing oncological knowledge and innovation, the center ensures patients are evaluated by a multidisciplinary team from diagnosis through rehabilitation. Each year, it provides over 328,000 outpatient services and 26,000 emergency services, including surgeries, chemotherapy, radiology, and various clinical tests, supported by a workforce of over 5,000 professionals. The institution’s hallmark is Personalized Oncology, utilizing advanced genomic and molecular testing for precise cancer diagnosis and monitoring. A specialized team collaborates to deliver tailored care throughout each patient’s unique journey, from prevention and diagnosis to treatment and follow-up, ensuring a highly individualized approach. The center’s commitment to holistic care is evident in its array of supportive and therapeutic services that address physical, mental, and emotional needs. Patients and their families benefit from spiritual guidance, animal-assisted activities, art programs, storytelling sessions, dance therapy, and music in waiting areas, all aimed at improving the hospital experience. Programs such as makeup and art workshops uplift patients’ spirits, while smoking cessation and psycho-oncological support have been integral parts of care since 1997. Additional support groups, yoga practice, and collaborations, such as providing wigs and prosthetics, highlight the center’s focus on patient well-being.

The sample consisted of 288 individuals who were hospitalized in the center. Data collection took place between November 30, 2022, and February 3, 2023. For the main data collection, a total of 1,028 patients who met the study criteria (hospitalized for oncological reasons during the specified data collection period) received an invitation to complete the survey. The survey link was automatically sent to all eligible patients upon discharge. Of these patients, 28% responded, resulting in a sample of 288 participants.

Demographic information for the respondents revealed that 174 (60.4%) were women. Ages ranged from 21 to 90 years (median: 51 years), and the length of hospital stay varied from 4 to 12 days (median: 6 days). In terms of education, 140 participants (48.6%) had at least a higher education degree.

Additionally, ten patients participated in the face validation of the instrument (described in the following subsection). These patients, admitted to the center, were selected based on the medical team’s judgment. The selection ensured diversity in terms of age and gender and took into account the patients’ emotional and physical readiness to verbally report on their current experiences at the hospital. No additional criteria were used to include or exclude patients.

To mitigate potential sampling bias, we ensured that all 1,028 eligible patients received a survey invitation, creating an inclusive approach to participation. By distributing the survey within seven days post-discharge, we minimized recall bias and preserved the accuracy of recent patient experiences.

The data collection instrument consisted of a survey developed specifically for this study (available as supplementary material), designed to measure a construct – SWB – using aggregated item scores. Adapting surveys for specific purposes is considered good practice; it enables greater fairness in the evaluation since the instrument assesses the construct based on existing theoretical and methodological perspectives. The approach for developing the survey, the ‘construct method,’ is theory-driven [57]. It begins by generating hypotheses based on a construct theory, such as those presented in this paper’s introduction, which are tested empirically.

The research instrument was an online survey distributed through the Salesforce Survey Tool platform, divided into three segments: (i) general patient information, (ii) SWB measures, and (iii) perceived service experience. In total, the instrument included 24 questions.

The first segment, referring to patient data, contained three questions about gender (male, female, or other), education, and age of respondents.

The second segment was composed of SWB items, including 15 statements. For the development of the items, we utilized the ten affective experiences measured by a short version of PANAS [53], as well as the five items from the Satisfaction with Life Scale [52]. To adapt them to the context of hospital services, items were crafted into concise sentences (rather than single words, as in the original PANAS), depicted in the results section in Table 1 and shown in detail in Appendix A (Tables 7 and 8). The respondents rated the extent to which they felt each described experience during their hospitalization using a 5-point scale (1 = A little or not at all, 2 = A little, 3 = Moderately, 4 = Quite a bit, 5 = Extremely). The specific prompt for this section was: “When I think about the last time I was hospitalized, I remember that the hospital staff had the following effects on me.” The 15 statements were presented randomized to mitigate potential order effects on the responses.

The third segment focused on patients’ perceptions of service experience, using the same 5-point scale as in the previous segment. Services considered were (i) nursing, (ii) medical, (iii) cleaning and hygiene, (iv) room infrastructure and comfort, (v) nutrition, (vi) access to entertainment, and (vii) administrative support for scheduling exams and consultations. The specific prompt for this segment was: “How much did each of the following services contribute to your well-being during your hospital stay?” The statements were also presented in a randomized order.

The instrument’s face validation was carried out before being applied to the final sample of respondents. For that, ten patients responded to a printed version of the survey and then reported their understanding of each question. The reports were gathered through individual semi-structured interviews with the patients whose profiles were described under ‘sample.’ The medical team indicated the time and place to conduct the interviews during hospitalization.

The choice of semi-structured interviews aimed to gain a deeper understanding of patients’ comprehension of each survey question. This method allowed patients to verbally express any confusion or misinterpretations, enabling researchers to identify and address specific issues. Conducting individual interviews provided nuanced insights into each patient’s understanding, facilitating refinements in survey language to ensure clarity and relevance before administering it to a larger sample. Additionally, this approach helped build a foundation of trust, allowing patients to share their thoughts in a supportive environment, which likely enhanced the validity of their responses. The authors initially planned for 10 interviews, with the flexibility to increase this number if significant issues with the instrument arose; however, the consistent results across these 10 interviews indicated that the survey was easy to understand, thus meeting data saturation requirements.

A researcher personally invited patients to participate. The research aim was explained during this interaction, and a comprehensive overview of the study procedures was provided. They received a consent form printed in two copies, signed by the principal investigator and the patient. The term ensured confidentiality and anonymity for participants, assuring them the right to withdraw from the research at any point or revoke their consent without incurring any negative consequences. The consent form stated the potential risks associated with participants’ involvement and clarified the absence of direct personal gains.

The interview script was structured in alignment with the survey questions. The interviews were audio-recorded for analysis. Insights from patients’ responses led to refinements in the survey, addressing minor issues in understanding certain questions.

The center’s team emailed potential respondents based on their discharge records. The email contained an invitation to participate in the study and a link to the survey, available through the Salesforce Survey Tool.

Before accessing the questions, participants completed an online consent form with content similar to the one described in the interviews. It was indicated that participants should download the form for their records before proceeding.

Data were exported from Salesforce in spreadsheet format for analysis. The analysis was conducted using the Statistical Package for the Social Sciences (SPSS Version 29) with several key objectives. First, we explored the characteristics of the sample (segment 1 of the instrument) to gain insights into the demographic and contextual factors influencing the data.

Next, we assessed the internal consistency of the instrument (segment 2 of the instrument). Our aim was to achieve very high internal consistency (Cronbach’s alpha α ≥ 0.9), reflecting a rigorous standard to ensure that the items on our instrument consistently measure the intended construct without redundancy. Internal consistency is typically assessed using Cronbach’s alpha, with reference values ranging from 0.7 (acceptable) to 0.9 or above (excellent). Achieving a high α is critical for validating the reliability of our instrument in capturing SWB.

We have chosen to employ a factorial model for dimensionality reduction of the SWB items in segment 2 of the instrument. This decision was based on a systematic approach to ensure its appropriateness. First, the Exploratory Factor Analysis procedure included calculating the KMO (Kaiser-Meyer-Olkin) measure of sampling adequacy, which yielded a result of 0.942. The KMO statistic ranges from 0 to 1, with values closer to 1 indicating that the sampling is adequate for factor analysis; a KMO value above 0.90 is considered excellent. The high KMO value suggested that a factorial model could effectively group survey items, indicating that the items shared sufficient common variance to be analyzed together [58]. Second, we conducted Bartlett’s test of sphericity, which resulted in 5083.432 (p <.01). The test evaluates whether the correlation matrix is significantly different from an identity matrix, where all variables are uncorrelated. A significant result (p <.05) implies that there are sufficient correlations among the variables, thereby validating the appropriateness of the correlation matrix for factor analysis [59]. Third, to enhance the interpretability of the factors, we applied a Varimax rotation, which simplifies the factor structure by redistributing variance among factors. The rotation helps variables to load highly on a single factor while maintaining low loadings on others, facilitating the identification of variables most closely related to each underlying construct. For instance, in our analysis, we expected one factor to represent aspects of positive affect, while another would reflect negative affect.

Finally, linear regression analyses were performed to examine the potential predictive role of perceived service experience (segment 3 of the instrument) on SWB (segment 2). In constructing the regression model, we assessed the hypothesis of linear relationships between variables by analyzing scatterplots that depicted the dependent variable (mean SWB) against their corresponding predicted values and residuals. These scatterplots confirmed the appropriateness of the linear model for our data and are presented in Appendix B. As shown in Figure 1 of the appendix, the dispersion model explains 62% of the variance in the model, indicating a strong linear relationship. Figure 2 presents the residual plot, which does not show any systematic pattern, suggesting that the original model is correctly specified.

Additionally, although the data exhibited some degree of positive skewness (e.g., means exceeding 4.0 on a 5-point scale for several items), linear regression was chosen due to its robustness with larger samples and its suitability for interpreting direct relationships between service experience and well-being outcomes. Despite the positive skewness, the large sample size (n = 288) allows the central limit theorem to apply, ensuring that the sampling distribution of the mean approaches normality. Additionally, linear models are known to be robust to moderate violations of normality, particularly when assumptions of homoscedasticity and linearity are satisfied — as confirmed by the residual plot (Appendix B). This supports the appropriateness of using parametric methods without requiring data transformation. In this study, we used 95% confidence intervals for the regression coefficients to assess the precision and statistical significance of the predictor variables.

In the analyses related to internal consistency, factor analysis, and regression, responses for items depicting negative affect were inverted to align with the overall structure of the survey, where higher responses corresponded to better SWB scores. To address potential recall bias, we collected data from patients discharged within 7 days. This short time frame is crucial as it minimizes the likelihood of memory distortion, allowing participants to reflect on their recent hospital experiences more accurately. By reducing the interval between discharge and survey completion, we aimed to enhance the reliability of the responses and the overall quality of the data collected.

Self-reported SWB indicated a strong tendency towards positive responses across both SWB items and perceived service experience, as displayed in Table 1. The average scores for positive items (positive affect and life satisfaction) are equal to or greater than 4.05, indicating that respondents generally have a favorable view of these aspects. Furthermore, low dispersion measures (i.e., SD values mostly below 1.0) suggest that there is little variability in responses; most participants responded similarly about their positive experiences. In contrast, the average scores for negative affect are 1.86 or lower, indicating that participants reported minimal negative affect. The dispersion measures for these negative items, however, are equal to or above 0.98, suggesting that there was more variability in the negative responses.

Additionally, perceived service experience was also positive for all services, with scores at or above 4.0 and dispersion measures ranging from 0.65 to 1.11. This indicates that responses were generally favorable, although there was some variation.

Data reliability was assessed using Cronbach’s alpha. The 15-item survey yielded a result of 0.946, indicating outstanding internal consistency. This high score suggests that the items on the survey are measuring the same underlying construct effectively, ensuring that the responses are reliable. As shown in Table 2, Cronbach’s alpha values for all subscales are equal or above 0.931, further confirming that each subscale maintains a strong level of consistency. Overall, these results demonstrate that our survey is a trustworthy instrument for assessing SWB.

Factor analysis was conducted to examine the underlying structure of the survey items, with communalities displaying an average of 0.78 (Table 3), indicating that each variable shares significant variance with the others and reflecting a strong interrelationship among the items. Although the survey items were developed based on three theoretical dimensions, the factorial model after the Varimax Rotation resulted in a two-factor model. This model accounts for 49.12% and 28.89% of the total variance, respectively. Together, the two factors explain 78.01% of the total variance across all survey items, as shown in Table 3, suggesting that the survey effectively captures the main constructs related to SWB.

The factor composition separated all positive items (five for positive affect and five for life satisfaction) from those associated with negative affect, as shown in Table 4. All loadings were well above the 0.45 threshold [60], underscoring the adequacy of the two-factor model. Such high loading values indicate that each item is meaningfully associated with its respective factor, enhancing the reliability of the survey in capturing the dimensions of SWB.

Hypothesis 1, which states that highly reliable service-related SWB metrics would be produced using established reliable tools, was confirmed with a Cronbach’s alpha coefficient of 0.946, indicating outstanding internal consistency. The high level of internal consistency suggests that the survey items are consistently measuring the intended constructs. Furthermore, the results of the factor analysis demonstrate a strong alignment with the theoretical constructs underlying the subscales, reinforcing the validity of the instrument in assessing service-related SWB.

A multiple linear regression analysis was conducted to assess the extent to which patients’ SWB could be explained by their perceived service experience, as shown in Table 5. The analysis used each respondent’s SWB global mean score as the dependent variable and their perceived service experience scores as independent variables. The model displayed a significant adjusted R² of 0.609, indicating strong predictive power and suggesting that a substantial portion of the variance in the SWB global results is explained by these assessments. Specifically, nursing, medical care, and infrastructure assessments showed a significant positive effect on SWB global mean scores. The positive regression coefficients associated with these scores imply that higher ratings correlate with improved overall SWB, reinforcing the role of services in promoting patient well-being. This finding underscores the importance of services to promote well-being.

Regression analyses were also carried out separately for each dimension of SWB, using each respondent’s mean scores for positive affect, negative affect, and life satisfaction as dependent variables, as shown in Table 6. Although factor analysis grouped positive affect and life satisfaction in the same factor, we analyzed the constructs separately to gain further insight into the results. This course of action was grounded on theoretical considerations: positive emotions are typically situational experiences, whereas life satisfaction pertains to individuals’ long-term appraisals of their lives. By separating these dimensions, we can better understand how different aspects of SWB are influenced by various factors, allowing for more targeted interventions to enhance patient well-being.

The Adjusted \(\:{R}^{2}\) values of 0.63, 0.222, and 0.562, respectively, suggest good predictive power for the positive subscales (positive affect and life satisfaction) and weak predictive power (although statistically significant) for the negative affect subscale. Perceived service experience with a significant impact on positive affect were nursing, medical, infrastructure, and administrative support. As for negative affect, significant effects were observed only for nursing and infrastructure service assessments, while nursing, medical, and infrastructure assessments significantly influenced life satisfaction.

Hypothesis 2 was supported by our regression analyses. It stated that users’ perceived service experience would strongly predict positive affect, life satisfaction, and overall SWB while having a lower predictive power for negative affect. Notably, nursing and infrastructure assessments significantly affected all dimensions of SWB, while medical services influenced only the positive components—positive affect and life satisfaction. Additionally, administrative support was found to impact only positive affect. These findings highlight the importance of focusing on specific service areas to enhance patient experiences and overall well-being.

Our study contributes to the state-of-the-art in service design, particularly in the realm of user-centered healthcare and service design metrics. By incorporating patient SWB into the design and evaluation of healthcare services, we facilitate more patient-centric approaches. This shift is increasingly relevant in a healthcare landscape that emphasizes user experiences and relationships with care delivery [31,32,33,34,35].

Our research utilized the Positive Affect Negative Affect Schedule (PANAS) [53] and the Satisfaction with Life Scale (SWLS) [52]. These instruments, recognized for their effectiveness in assessing the affective and cognitive dimensions of SWB [50], provided a solid foundation for our study, from which we developed a highly reliable self-report instrument aligned with SWB and its three components, demonstrating methodological rigor.

Our results highlight the potential impact of services on patients’ SWB in oncology settings, suggesting that hospital policies prioritizing patient experiences contribute to their SWB. This could involve advocating for policies that ensure adequate staffing levels, enhance communication between patients and providers, and support the integration of complementary therapies. For instance, in our study, assessments regarding essential services such as nursing, medical care, and infrastructure notably enhanced the means of all SWB dimensions, particularly positive affect and life satisfaction. In the center where this study was implemented, these are complemented by a range of supportive services aimed at patient well-being, such as Therapeutic Paws (animal-assisted activities), Citizen’s Song (recreational and cultural activities), and Love for Life (a support group for patients and their families). This approach aligns with literature emphasizing the importance of emotion frequency and intensity in SWB [51, 61] and reflects the center’s commitment to personalized, patient-centric care. It also aligned with previous work [3, 48] that advocates for comprehensive oncology care that caters to both the medical and emotional needs of patients.

While the objectives of our study were to develop a reliable measure for service-related SWB and assess the perceived impact of services on this measure, our results have implications for healthcare service management. The findings underscore the importance of targeted improvements in service delivery. The significant perceived impact of nursing, medical care, and infrastructure on SWB suggests that healthcare providers could consider staff training, communication protocols, and environmental upgrades to enhance patient experience. For instance, the strong predictive effect of perceived nursing care on SWB highlights the need for consistent, compassionate communication and emotional support during treatment. Hospitals could also implement structured feedback loops using this survey tool to continuously monitor and refine patient experiences in real time.

In line with this, it is worth mentioning that PROMs and PREMs are essential for evaluating healthcare services from the patient’s perspective. PROMs measure health status and functional outcomes—such as pain and mobility—while PREMs assess perceptions of care delivery, including communication and responsiveness. PROMs and PREMs have been widely implemented in oncology care. However, they are not designed to measure the emotional and cognitive dimensions of service-related SWB. Recognizing this gap, our study developed a survey tool that integrates SWB measures with perceived service experience, providing a more comprehensive understanding of how healthcare services influence patient SWB and offering valuable insights for improving oncology care.

Our findings reinforce the importance of complementing PROMs and PREMs with measures that capture SWB. PROMs primarily focus on health-related quality of life, whereas PREMs center on care delivery processes. However, SWB encompasses emotional states and life satisfaction, which reflect deep experiential factors. This highlights a gap that our survey tool aims to fill by integrating SWB measures with perceived service experience, thereby providing a holistic understanding of patient well-being.

Our results also hold implications for service design. We draw from the stages indicated by Pfannstiel and Rasche [32] for user-centered health design: (i) exploring users’ experiences and expectations, (ii) ideating for developing future services, (iii) prototyping services, (iv) testing them with patients, and (v) implementing them. Our survey can reliably capture patients’ responses in stages (i) and (iv).

In Stage (i) – exploring users’ experiences and expectations – our survey serves as a key tool for gathering quantitative data on perceived service experience, as well as SWB. Such data, commonly supplemented by qualitative methods [8], provides a panoramic view of patient needs. It reveals which services, as currently provided and assessed, contribute to SWB. Thus, the survey helps collect relevant and reliable information that guides ideation (stage ii) and service prototyping (stage iii). For instance, it supports brainstorming sessions by facilitating empathy mapping, user journey mapping, and service blueprint development. Additionally, engaging patients in co-designing services represents a significant shift toward a more patient-centered approach. By actively involving patients as co-designers, healthcare providers empower them to share their unique insights, experiences, and preferences, which can greatly enhance the relevance and effectiveness of services. This collaboration fosters open communication between patients and healthcare teams, creating an environment of empathy. Through co-design, patients can articulate their challenges, leading to the identification of tailored interventions that address their specific needs—whether that involves supportive therapies or peer support groups, such as those available at AC Camargo Cancer Center.

Based on these ideas, service prototypes (stage iv) can be developed, with our survey playing a crucial role again. Experimental research design can be implemented in this context, aligning with the field’s needs and future research trends [62]. This type of method enables the collection of feedback, leading to iterative improvements. Additionally, longitudinal data collection using our survey may facilitate tracking how changes in services correlate with improvements across SWB dimensions. Thus, our survey tool can support the establishment of structured feedback loops, allowing patient insights to inform service design directly. Even after implementation (Stage v), continuous monitoring remains essential to make necessary adjustments in services.

This comprehensive approach to user-centered service design is consistent with our objective of enhancing SWB. Our survey can reliably contribute towards this goal by prioritizing patient SWB and systematically addressing their needs at every stage.

This study has some limitations, which also present opportunities for future work. First, two potential response biases should be addressed. The first is social desirability bias. Patients tend to express gratitude to healthcare professionals [63]. Thus, they may agree that such professionals positively impacted their well-being as a sign of gratitude, calling for future studies that do not rely on self-report measures. The second type is recall bias. Even though all respondents were discharged from the hospital up to seven days before responding to the survey, which was an attempt to mitigate such bias, some might not be able to remember their experiences entirely due to factors such as the impact of traumatic experiences on memory. Thus, future studies could collect data in loco and immediately before or after discharge.

Second, this study adopted a non-experimental method and relied on regression analyses to assess the predicting power of services on SWB. Even though regression models are invaluable in assessing relationships between variables and making predictions, they can only identify associations between these variables, not prove causation. Moreover, the analysis may be influenced by potential confounders that were not accounted for, such as health status (e.g., disease stage and comorbid conditions), psychological factors (e.g., pre-existing mental health conditions and coping styles), and the level of social support (e.g., family and peer support). Although the regression model showed strong predictive power for positive affect and life satisfaction, the results should be interpreted with caution due to potential confounding effects from unmeasured variables such as health status, psychological traits, and social support. Future studies should explore these factors to refine the model’s explanatory power. These variables could affect both the perceived quality of services and patients’ SWB. Therefore, future research should consider incorporating experimental designs that manipulate service variables while controlling for these confounding factors. Additionally, longitudinal studies could be beneficial to track changes in SWB over time as service modifications are implemented, thereby providing deeper insights into the causal relationships between service quality and patient SWB. Also, expanding the study to cover other variables and patient profiles would potentially allow the exploration of non-linear dynamics between service quality and SWB. Qualitative methods, such as patient interviews or focus groups, could also enrich the understanding of how different services influence SWB by capturing the nuanced experiences of patients.

Building on the factors not considered in the study, the third limitation indicates that other variables and service attributes could be evaluated in future studies. For instance, unique service experiences in oncology (for instance, chemotherapy), digital health services, patient-provider communication, and the hospital’s built environment, to name a few.

Fourth, although our analyses did not attempt to generalize results to a larger population, using a single hospital might have limited the diversity of patient profiles. Thus, future studies may investigate different research contexts and among diverse patient groups.

Finally, we used adapted measures—PANAS and SWLS—in the survey. Even though we have advocated that this is a well-grounded methodological choice, aligned with the American Psychological Association guidelines since the 1950 s, future studies might develop and test new metrics, possibly in combination with ours, to assess the extent to which they relate.

Our research underscores the critical role of services in shaping patients’ SWB in oncology care and highlights the importance of capturing self-assessed user experiences for effective service design, assessment, and management. While PROMs and PREMs have become essential tools for measuring health status and care delivery, they are not specifically designed to capture the broader emotional and cognitive dimensions of patient experience. Our concise and reliable 15-item survey provides a practical tool for assessing SWB, distinguishing itself by its ease of implementation and adaptability across various healthcare settings. The tool not only captures perceived service experience but also offers valuable insights for creating patient-centered services.

The study identified key components of SWB—positive emotions, life satisfaction, and negative emotions—emphasizing the need to consider both positive and negative affect in assessing patient SWB. Regression analyses showed that perceived service experience significantly predicted SWB, particularly positive affect and life satisfaction.

While our findings are specific to the study setting, they reveal actionable areas for improvement in service delivery and underscore the importance of focusing on services that positively influence patient SWB. We encourage healthcare providers to adopt this survey tool as part of their routine assessments, enabling them to gather real-time feedback and continuously refine their services based on patient experiences.

Finally, surveys like the one presented here can support the design of more patient-centered, empathetic, and effective healthcare services by facilitating ideation, prototyping, and continuous improvement efforts. By leveraging this tool, healthcare organizations can create a more responsive and supportive environment that enhances patient SWB across diverse healthcare contexts.

SWB:

Subjective Well-Being

PANAS:

Positive Affect Negative Affect Schedule

SWLS:

Satisfaction with Life Scale

PROM:

Patient-reported outcome measure

PREM:

Patient-Reported Experience Measure

Download references

LMT contributed to the study conception, research design, data analysis, interpretation of data, and drafted the manuscript. CLPC was involved in the study conception, data collection, and data analysis. PBR contributed to research design, data analysis, and interpretation of data. CLA, LPK, and GBC participated in the study conception and interpretation of data. All authors contributed to earlier drafts of the paper, reviewed the manuscript, and approved the final version.

Correspondence to Leandro Miletto Tonetto.

This study was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki. All procedures involving human participants were approved by the Institutional Review Board of the AC Camargo Cancer Center (Protocol No. 5.465.066). Informed consent was obtained from all participants.

Not applicable.

The authors declare no competing interests.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions