Driving Patient eWOM: The Role of Perceived Value in Health Care Services

Abstract

Due to the health information asymmetry, the upsurge of Patient Online Communities (POCs) and Patient Social Media groups has increased the importance of electronic word-of-mouth (eWOM) in health care, influencing individuals’ health decisions, as well as a medical organization’s image. This study investigates the association between the multidimensional perceived value of patients and their eWOM intentions in health care services, based on Art Weinstein’s adapted Perceived Value framework. According to this framework, perceived value comprises perceived quality, perceived service outcome, non-monetary costs, and organizational image. Data were collected from 210 Cardiology patients and analyzed using Partial Least Squares Structural Equation Modeling (PLS-SEM). Findings of this study revealed that perceived value is positively associated with eWOM intentions within this sample, which highlights the practical importance of enhancing patient experience. As perceived value improves, it may be associated with increased patient-generated content in the form of eWOM. This study provides practical insights and contributes to the understanding of the patients’ perceived value in engaging in health-related eWOM.

1. Introduction

Understanding how patients perceive and assess health care services has become a major focus of both academic research and managerial practice in competitive and consumer-driven health care markets. The digitalization, the necessity of greater transparency in medical information, patient empowerment, and increased competition brought significant changes in modern health care systems [1,2,3,4]. These advancements have transformed patients’ roles from passive receivers of treatment to active decision-makers. Such transformation was visible in the patient-generated content, in the shape of electronic word-of-mouth (eWOM), which brought also notable contributions to the digital health ecosystems [5,6,7].

Health care services differ from other types of services due to their high credibility characteristics, strong emotional involvement, high perceived risk, and a substantial knowledge asymmetry between providers and patients [8,9]. Even after the health care consumption stage, patients often lack the technical know-how required to objectively assess clinical outcomes [10]. As a result, while making decisions about health care providers, be they specialists or organizations, patients mainly rely on subjective assessments of their service experiences, namely, the functional attributes [11]. This change has been supported by the growth of POCs, health forums, and social media platforms, which allow people to share opinions and experiences that may impact other persons’ health care decisions [12,13]. Thus, eWOM can be considered as a social process embedded within a health care system [14], which may contribute to strengthening the trust in health care services, enhanced perceptions of health service quality or satisfaction, and patient empowerment.

Perceived value has been acknowledged to be a core concept in Marketing and Services Research, which refers to a consumer’s overall evaluation of a service’s usefulness, based on the perceived benefits in comparison to the sacrifices made [15]. Subsequent research extended this conceptualization by emphasizing how value perceptions are subjective, experiential, and multifaceted [16,17]. Therefore, perceived value is a cognitive mechanism used to interpret experiences, defend choices, and decide on future behavioral intentions, especially in health care services, when patients frequently cannot directly assess technical outcomes, due to information asymmetry [18]. Building on this, Weinstein developed a comprehensive framework that conceptualizes perceived value as being made-up of costs for sacrifices, while benefits were defined by perceived quality, perceived service outcome, and organizational image [19]. This framework may be applicable in health care services because patients assess more the functional attributes, such as empathy, accessibility, safety, waiting time, emotional support, and tangible elements [20,21].

Moreover, perceived value has long been known to predict loyalty and satisfaction [22,23,24,25], but its relationship to eWOM has attracted increasing scholarly interest, due to its contributions at multiple levels in the health care field: the applicability of eWOM at micro level refers to the patient shared experience, the eWOM application at meso level may refer to building an organization’s online image or online reputation, whereas the applicability of eWOM at macro level, namely, at system or society levels, may refer to shaping regulations, rules or policies, based on the patients’ needs and perceptions. Similarly, according to the Services Research field, higher perceived value tends to encourage favorable online reviews, and recommendations [26,27], but in health care services, there are still significant gaps in understanding how patients form their perceived value and if there is an association with eWOM engagement. Previous studies have mostly examined individual factors as predictors of patient intentions, such as satisfaction, trust, or service quality [28,29]. Although these studies offer insightful information about health care service experiences, they frequently consider value-related constructs as separate variables rather than considering perceived value as a comprehensive, multifaceted evaluation process in which patients concurrently assess the advantages and disadvantages of health care services [19]. Additionally, although the Marketing scientific literature acknowledges perceived value as a key factor for influencing consumer behavior, empirical studies examining how it directly influences eWOM, especially in health care settings, are limited. Previous studies on perceived value and eWOM have mostly focused on travel, retail, and aviation [26,27,30], whereas studies on health care mostly examined online reviews [31] or patient satisfaction [32] separately rather than linking them into a multidimensional concept.

In an era of digital transparency, where online patient opinions can impact organizational reputation and image [33], and future patients’ decision-making [34], experts must understand this association [35,36]. The current study conceptualizes perceived value as a multidimensional construct comprising perceived quality, perceived service outcome, organizational image, and non-monetary costs, and it aims to investigate the role of perceived value as an antecedent of patients’ intentions to engage in eWOM about health care services. The objectives of this study are twofold as follows: (1) to apply Weinstein’s adapted Perceived Value framework, including perceived quality, perceived service outcome, non-monetary costs, and organizational image in health care services, and (2) to investigate how perceived value in health care services is associated with patients’ intentions to engage in eWOM.

2. Theoretical Background

A theoretical background for understanding online patient behavior in health care services is provided by the combination of Weinstein’s adapted Perceived Value framework with Social Identity Theory [14]. This combination highlights the relational and experiential processes that trigger online patient-generated content [1,37], clarifies the association between the organizational image and eWOM [36,38], and enables researchers to distinguish how different value components shape online communication behaviors [15,39]. In the health care field, where services are mostly credence-based, and patients find it challenging to evaluate clinical quality (technical quality) [40], even after consumption, this approach becomes highly important [41].

Patients’ evaluations of clinical effectiveness, safety, diagnostic accuracy, and treatment efficacy are reflected in perceived quality [42,43,44]. Although technical quality is essential, patients frequently lack the knowledge necessary to assess clinical outcomes directly [45]. As a consequence, patients use functional service attributes, including trustworthiness elements, experts’ communication skills, the presence and accessibility to medical equipment, and organization image, in order to determine the functional quality [46,47,48]. These functional quality cues may affect the likelihood of patients to share online experiences and opinions, implicitly, and engage in eWOM [5,49].

Moreover, the potential results of patient care experience are captured by the perceived service outcome [1,42,50]. Patients’ desire to engage in positive eWOM is often influenced by their overall assessment of care [14,51,52].

According to Weinstein’s Perceived Value framework, costs are perceived as sacrifices [19]. In this study, this perspective was adapted by referring to positive non-monetary costs as benefits. More exactly, in the health care field, non-monetary costs, such as psychological, temporal, and physical sacrifices, are important [15,16,53]. Patients often experience worry, administrative complex accessibility, travel burdens, and waiting times, all of which may lower perceived value [9,54,55], and engage in negative eWOM [5]. If these are being limited or managed properly, then they may become benefits, which would lead to positive eWOM.

A health care provider’s credibility is reflected in its organizational image [36,56,57]. Strong branding increases perceived benefits [40,51,58], and triggers positive eWOM [59].

Patients frequently rely on online reviews and eWOM to reduce ambiguity and make informed decisions because it is challenging to evaluate health care services’ trustworthiness [60,61]. According to research [12,62], patients are encouraged to share their experiences online, not just when they are satisfied, but also when emotional, relational, or process-related aspects of value are dominant [63]. By combining these concepts, perceived value can be considered an important antecedent of eWOM.

Based on Weinstein’s adapted Perceived Value framework, the current study proposes a model in which patients’ eWOM intentions are associated with perceived value, defined as a multidimensional construct comprising perceived quality, perceived service outcome, non-monetary costs, and organizational image (Figure 1).

3. Materials and Methods

3.1. Study Design and Participants

A cross-sectional design was used to examine the relationship between perceived value and eWOM engagement in a Romanian Cardiology health care private organization. The health care private organization is a medium-sized business practice and is located in Bucharest.

The target population consisted of patients, who needed Cardiology care during March–April 2025. Participants received an envelope containing two predefined forms: an informed consent form and the data collection questionnaire. The informed consent form included the study’s aim, the voluntary condition of participation, the confidentiality of the collected and processed data, and the use of anonymized data for research purposes.

The inclusion criteria of the participants were: (1) aged 18 years or older, (2) having used the organization’s services at least three times because it may help in filtering out initial first-time opinions which do not accurately reflect the service experience, and (3) having access to the Internet or Social Media. Patients with cognitive impairments, who were unable to provide informed consent, and required emergency consultations were excluded. A non-probabilistic, purposive sampling technique was used to select participants, as it allowed for the inclusion of individuals with relevant health care experiences [64].

Initially, the questionnaire was translated into Romanian by one independent translator, and the resulting translation was back-translated into English to verify accuracy and identify potential linguistic inconsistencies. Minor improvements were made to one item of the questionnaire.

Before the main data collection, a pilot study was conducted with five participants in order to evaluate the clarity, relevance, and the comprehensibility of the items of the questionnaire, and to ensure the cross-cultural adaptation of the research instrument [65]. The pilot study participants were asked to complete the questionnaire and provide feedback regarding the wording and the structure of the items [66]. Based on the received feedback, minor revisions were made to improve the clarity of four statements. The participants involved in the pilot study were not included in the main study.

In the main study, a total of 350 respondents were approached, of which 210 completed questionnaires (response rate = 60%) were retained for analysis after screening for completeness of the questionnaire and informed consent. This sample size meets the minimum requirements for structural equation modeling (SEM) [67], ensuring sufficient statistical power (power = 0.70, α = 0.05). The sample consisted of 58.1% women and 41.9% men, with the mean age of 57.62 (SD = 15.49). A total of 46.2% of respondents had a university degree, 59.5% reported being married, and the majority of patients’ motivation was the monthly usual consultation (43.3%).

The socio-demographic profile of patients is depicted in

Table 1. The socio-demographic profile of the respondents in the main study.

Variable	Frequency	Percentage (%)
Gender
Female	122	58.1%
Male	88	41.9%
Marital status
Unmarried	24	11.4%
Married	125	59.5%
Divorced	27	12.9%
Widower	34	16.2%
Education level
Primary school studies	10	4.8%
High-school studies	55	26.2%
Undergraduate studies	97	46.2%
Postgraduate studies	48	22.9%
Motivation for consultation
Usual monthly consultation	91	43.3%
Diagnosis consultation	68	32.4%
Treatment prescription	47	22.4%
Other	4	1.9%

.

3.2. Instruments

Data were collected using a self-administered questionnaire on paper, developed based on previous research. The self-administered questionnaire consisted of 6 sections, as follows: (1) the first section collected socio-demographic information about the patients, (2) the second section comprised the subdimensions of perceived quality, (3) the third section comprised the subdimensions of image, (4) the fourth section referred to the perceived service outcome, (5) the fifth section collected information about the non-monetary costs, and (6) the sixth section comprised the items of measuring eWOM engagement intention. Patients agreed with the statements of the questionnaire on a 5-point Likert scale, which varied from 1-Strongly Disagree to 5-Strongly Agree.

Table 2. The sources and the adapted measurement instruments.

Construct	Source	Items
Perceived Quality
Empathy	[68]	EM1: The physician had a polite attitude. EM2: The physician offered detailed information to me. EM3: The physician understood my situation. EM4: The physician was friendly.
Tangible Elements	[68]	ET1: The medical organization has advanced equipment. ET2: In the medical organization, consultations are provided by well-qualified staff. ET3: The medical organization has a comfortable waiting room. ET4: The medical organization has a comfortable consultation room. ET5: In the medical organization, the employees’ clothing is clean. ET6: The medical organization maintains proper cleanliness of the facility. ET8: The consultation room ensures confidentiality and privacy. ET9: The consultation room is spacious and clean.
Safety	[68]	SI: The medical organization provides a comfortable and safe environment for patients. SI2: I believe that my physician established an appropriate diagnosis for me. SI3: I believe I can trust my physician.
Image
Social Responsibility	[69]	RS1: The medical organization supports the social causes in the community it activates. RS2: The medical organization supports social cause campaigns. RS3: The medical organization seems to be environmentally friendly.
Performance	[69]	PER1: I believe my medical organization is strong, given its strong history. PER2: I believe my medical organization tends to outperform other medical organizations. PER3: My medical organization has strong prospects for future growth.
Perceived Service Outcome	[68]	PS1: I believe the medical organization provided an appropriate service to me. PS2: I believe that, following the medical service, I was prescribed appropriate treatment. PS3: The improvement in my health is a result of the efforts and treatments prescribed by the medical staff. PS4: My health has improved after the medical consultation. PS5: The employees of the medical organization provided information on prevention.
Non-monetary costs	[70]	CNM1: The time spent in the waiting room was appropriate. CNM2: I was able to get an appointment with my physician promptly. CNM3: I had enough time to present my investigation results and analyses to my physician. CNM4: I easily accessed other medical services in addition to the consultation. CNM5: The noise in the waiting room was tolerable. CNM6: The smell in the waiting room was appropriate. CNM7: The location of the medical organization is geographically accessible.
eWOM	[71]	eWOM1: I intend to recommend the medical service online. eWOM 2: I will recommend the medical service online. eWOM3: I will definitely recommend the medical service online.

summarizes the sources of the instruments and their adapted items from the literature.

3.3. Statistical Analysis

The analyses consisted of two stages, as follows: (1) the data were collected and processed in SPSS version 24 (IBM Corp., Armonk, NY, USA), and frequencies and percentages were used to describe the socio-demographic characteristics of the main study’s sample; and (2) the main study was analyzed in both SPSS version 24 and SmartPLS version 4 (Hamburg, Germany). An Exploratory Factor Analysis was conducted in SPSS, followed by reliability and validity testing and hypothesis testing in SmartPLS. Exploratory Factor Analysis (EFA) was conducted to examine the underlying structure of the questionnaire items, identify coherent factors, and remove any items that did not meaningfully contribute to their intended constructs, thereby ensuring the reliability and validity of the measurement model [72].

To address the potential influence of the Common Method Bias (CMB), which may occur when data on independent and dependent variables are collected from the same source, multiple tests for checking CMB were conducted. First, Harman’s single-factor test was performed in SPSS version 24 to check if a single factor accounted for the majority of the variance. If the test shows that no single factor explains more than 50% of the total variance, then CMB would not be a potential concern [73]. Secondly, the Variance Inflation Factors (VIFs) were calculated in SmartPLS for the formative constructs to examine multicollinearity [67].

Consequently, PLS-SEM was employed to conduct a more in-depth analysis and test the association between perceived value and eWOM. PLS-SEM is a variance-based structural equation modeling approach suitable for theory development [74]. It is commonly applied in prediction-oriented research and in testing theoretical relationships, as it offers flexibility for analyzing complex models [75]. Specifically, PLS-SEM was used for this study because the model had a complex structure, including multiple constructs, higher-order constructs, and indicators that were both formative and reflective [67]. This analysis followed a two-stage approach [67]: (1) the assessment of the measurement model’s reliability, by assessing the internal consistency with Cronbach’s Alpha Coefficient and the Composite Reliability (CR), while the validity of the model was examined with both Convergent Validity through the Average Variance Extracted (AVE), and the discriminant validity was tested using cross-loadings, the Fornell–Larcker criterion and the Heterotrait–Monotrait ratio (HTMT), and (2) the structural model which assessed the association between the latent variables, Perceived Value and eWOM, which included the evaluation of the path coefficient (β), Coefficient of Determination (R²), the predictive relevance (Stone–Geisser Q²), and the effect size (f²). A p-value ≤ 0.05 was considered statistically significant.

4. Results

Following the pilot study, the Exploratory Factor Analysis (EFA) was assessed.

4.1. The Exploratory Factor Analysis

The suitability of the data for EFA was determined by two tests: (1) the Kaiser–Meyer–Olkin (KMO) measure of sampling adequacy, which had a value greater than 0.6 (KMO = 0.94), and (2) Bartlett’s Test of Sphericity, which revealed that the items were correlated (p < 0.05). The two tests emphasized that the study data were suitable for the EFA.

In the EFA, the Principal Component Analysis with Varimax rotation and factor extraction was guided by a fixed number of factors (i.e., 8), and only those with loadings above 0.40 were retained. Initially, all items were included, but several were dropped due to low factor loadings (<0.50) or cross-loadings, indicating that the following items were dropped: EM3, ET6, PS4, PER3, ET4, ET8, SI1, PS3, RS2, CNM4 and CNM6. After item refinement, the final EFA yielded an 8-factor solution with an explained variance of 84.60%, and all retained items loaded on the assigned factors, demonstrating a clean structure and a suitable basis for proceeding with the PLS-SEM analysis.

4.2. Common Method Bias Analysis

In what concerns Harman’s single-factor test, it was revealed that an unrotated factor analysis with a single factor accounted for 48,81% of the total variance. Despite obtaining a value below the threshold of 50%, taking into consideration the limitations of the test, potential method variance can still be a potential problem [73]. As a complementary approach [76,77], to strengthen the assessment, full collinearity VIFs of the formative constructs were also evaluated.

4.3. PLS-SEM Analysis

The PLS-SEM model was a formative–reflective third-order (higher-order) construct. The formative second-order construct, Perceived Quality, was composed of the reflective first-order constructs, Empathy, Tangible Elements, Safety, and the second-order construct, Image, was composed of the reflective first-order constructs, Social Responsibility and Performance. Perceived Value was a formative third-order construct, which comprised the formative second-order constructs Perceived Quality and Image, and the reflective first-order constructs Perceived Service Outcome and Non-monetary costs. The eWOM construct was also a first-order reflective construct.

Following the recommendations for higher-order constructs in PLS-SEM [67], the measurement model begins by assessing the reliability and validity of the first-order constructs, then assessing the validity of the second- and third-order constructs. After assessing the measurement model, the structural model is determined.

4.3.1. Measurement ModelReliability and Validity

Reliability and Validity

The reliability of the first-order constructs was evaluated using Cronbach’s Alpha Coefficient and Composite Reliability (CR) values. The outcomes of the factor loadings after EFA, along with the reliability and validity of the constructs, are presented in

Table 3. Item loadings, reliability, and validity.

First-Order Construct	Item Loadings	Alpha	CR	AVE
Empathy		0.92	0.93	0.87
EM1	0.93
EM2	0.92
EM4	0.94
Tangible Elements		0.90	0.91	0.72
ET1	0.85
ET2	0.86
ET3	0.88
ET5	0.79
ET9	0.86
Safety		0.95	0.95	0.95
SI2	0.97
SI3	0.97
Perceived Service Outcome		0.91	0.91	0.85
PS1	0.95
PS2	0.94
PS5	0.88
Social Responsibility		0.95	0.95	0.95
RS1	0.97
RS3	0.97
Performance		0.91	0.91	0.92
PER1	0.95
PER2	0.96
Non-monetary costs		0.91	0.92	0.75
CNM1	0.89
CNM2	0.88
CNM3	0.90
CNM5	0.79
CNM7	0.85
eWOM		0.87	0.87	0.79
eWOM1	0.91
eWOM2	0.84
eWOM3	0.92

Note: Alpha-Cronbach Alpha Coefficient; CR-Composite Reliability; AVE-Average Variance Extracted.

. All Cronbach Alpha Coefficient and CR values were higher than the recommended threshold of 0.70 [67]. In addition, the Average Variance Extracted (AVE) values, which measured Convergent Validity, were greater than 0.50, indicating that more than 50% of the variance in the indicators was captured by the construct [67].

Discriminant validity was assessed through cross-loadings, the Fornell–Larcker criterion, and the Heterotrait–Monotrait Method (HTMT).

Table 4. Discriminant validity—cross-loadings.

	Empathy	Non-Monetary Costs	Perceived Service Outcome	Performance	Social Responsibility	Safety	Tangible Elements	eWOM
EM1	0.935	0.453	0.725	0.373	0.377	0.787	0.275	0.477
EM2	0.925	0.563	0.760	0.495	0.525	0.727	0.434	0.554
EM4	0.940	0.506	0.775	0.429	0.446	0.762	0.342	0.521
CNM1	0.393	0.897	0.414	0.500	0.414	0.351	0.549	0.455
CNM2	0.468	0.889	0.487	0.485	0.542	0.430	0.495	0.469
CNM3	0.631	0.904	0.626	0.466	0.433	0.545	0.533	0.521
CNM5	0.335	0.793	0.424	0.404	0.368	0.358	0.535	0.373
CNM7	0.510	0.851	0.446	0.433	0.429	0.374	0.397	0.401
PS1	0.742	0.546	0.950	0.478	0.524	0.784	0.492	0.590
PS2	0.763	0.481	0.940	0.404	0.447	0.833	0.404	0.588
PS5	0.738	0.521	0.883	0.481	0.543	0.684	0.400	0.574
PER1	0.432	0.492	0.448	0.959	0.680	0.390	0.552	0.560
PER2	0.464	0.523	0.494	0.960	0.682	0.406	0.541	0.571
RS1	0.477	0.498	0.544	0.686	0.976	0.452	0.520	0.598
RS3	0.472	0.490	0.523	0.701	0.977	0.437	0.528	0.606
SI2	0.787	0.457	0.817	0.419	0.471	0.978	0.429	0.579
SI3	0.798	0.482	0.804	0.391	0.416	0.975	0.415	0.554
ET1	0.277	0.474	0.380	0.559	0.573	0.333	0.853	0.580
ET2	0.489	0.484	0.559	0.542	0.506	0.510	0.865	0.625
ET3	0.199	0.539	0.308	0.486	0.431	0.245	0.883	0.537
ET5	0.310	0.424	0.352	0.410	0.297	0.356	0.792	0.478
ET9	0.319	0.536	0.369	0.417	0.448	0.377	0.863	0.594
eWOM1	0.518	0.468	0.567	0.513	0.558	0.517	0.623	0.913
eWOM2	0.412	0.439	0.511	0.505	0.525	0.443	0.544	0.841
eWOM3	0.551	0.475	0.608	0.560	0.565	0.585	0.611	0.920

Note: The values in italics are the loadings of the items on their assigned constructs.

depicts the cross-loadings of all the items. It can be observed that all factor loadings exceed their cross-loadings, suggesting discriminant validity. The results of the discriminant validity test using the criterion of Fornell–Larcker and the Heterotrait–Monotrait Method (HTMT) are illustrated in

Table 5. Discriminant validity using the criterion of Fornell–Larcker and the Heterotrait–Monotrait Method (HTMT).

	Empathy	Non- Monetary Costs	Perceived Service Outcome	Performance	Social Responsibility	Safety	Tangible Elements	eWOM
Empathy	0.933	0.581	0.877	0.504	0.513	0.866	0.404	0.614
Non-monetary costs	0.546	0.868	0.603	0.576	0.539	0.508	0.633	0.572
Perceived service outcome	0.808	0.558	0.925	0.537	0.585	0.890	0.508	0.706
Performance	0.467	0.529	0.491	0.960	0.761	0.444	0.623	0.661
Social Responsibility	0.486	0.506	0.546	0.710	0.977	0.478	0.570	0.677
Safety	0.812	0.480	0.830	0.415	0.455	0.976	0.461	0.634
Tangible Elements	0.380	0.578	0.468	0.570	0.537	0.432	0.852	0.743
eWOM	0.557	0.516	0.631	0.590	0.616	0.580	0.665	0.892

Note: Elements in italic are the square roots of the AVE. Below the diagonal, elements are the correlations between the constructs. Above the diagonal, elements are the HTMT values.

. According to Table 5, the AVE values are greater than their correlations with the other constructs of the model, and the HTMT values below 0.90 in lenient contexts indicate satisfactory discriminant validity [76,77].

Validating Higher-Order Constructs

Perceived quality and Image were formative second-order constructs, and perceived value was a formative third-order construct. Perceived quality was built on the lower-order constructs—Empathy, Tangible Elements, and Safety—while Image was built on the lower-order constructs—Social Responsibility and Performance.

To assess multicollinearity among the higher-order constructs, Variance Inflation Factors (VIFs) were computed. In formative HOCs, although VIF values ≤ 5 are preferred [76], moderate multicollinearity (VIF > 6) is acceptable if the outer weights are statistically significant [78,79]. In the present study, Empathy and Safety constructs showed moderate collinearity, but were acceptable because their outer weights were significant (Empathy construct: w = 0.45, t = 21.08, p = 0.001; Safety construct: w = 0.47, t = 21.92, p = 0.001) (

Table 6. Second higher-order constructs validation (Perceived Quality and Image).

Second Higher Order Construct	LOCs	VIF	Outer Weights	t Statistics	p Values	Outer Loadings	p Values
Perceived Quality	Empathy	6.20	0.45	21.08	0.001	0.80	0.001
	Tangible Elements	1.37	0.27	6.67	0.001	0.48	0.001
	Safety	6.39	0.47	21.92	0.001	0.84	0.001
Image	Social Responsibility	1.66	0.54	20.21	0.001	0.75	0.001
	Performance	1.59	0.57	19.46	0.001	0.79	0.001

Note: LOCs—Lower-order constructs; VIF—Variance Inflation Factor.

). In addition, they were retained because they were theoretically justifiable, and that they should rarely be removed because the formative measurement theory requires that indicators collectively represent the full domain of the construct, as defined in the conceptualization stage [67]. Hence, the second formative higher-order constructs were validated.

In what concerns the formative third higher-order construct, Perceived Value, the VIF values of the constructs Perceived Quality and Perceived Service Outcome showed moderate multicollinearity (VIF > 6), resulting in non-significant outer weights (Perceived Quality construct: w = 0.09, t = 0.38, p = 0.69; Perceived Service Outcome construct: w = 0.40, t = 1.61, p = 0.10) (

Table 7. Third higher-order construct validation—Perceived Value.

LOCs	VIF	Outer Weights	t Statistics	p Values	Outer Loadings	p Values
Perceived Quality	6.63	0.09	0.38	0.69	0.73	0.001
Image	1.18	0.63	8.03	0.001	0.85	0.001
Perceived Service Outcome	6.78	0.40	1.61	0.10	0.77	0.001
Non-monetary costs	1.17	0.14	1.30	0.19	0.59	0.001

). However, if the corresponding outer loadings are sufficiently high (>0.50) and are statistically significant (p < 0.05), then the constructs should be retained, especially if they have theoretical justification [80]. Thus, Perceived Quality and Perceived Service constructs were retained because of their high statistically significant outer loadings (i.e., Perceived Quality construct: λ = 0.73, p = 0.001, Perceived Service Outcome construct: λ = 0.77, p = 0.001). Moreover, the Non-monetary costs construct presented a non-significant weight (w = 0.14, t = 1.30, p = 0.19), but significant substantial outer loading (λ = 0.59, p = 0.001). Therefore, all four dimensions were retained because they contributed to the validation of the Perceived Value construct (Figure 2).

4.3.2. Structural Model

The next step in the analysis was to assess the association between perceived value and eWOM, which was supported, showing a significant positive association between the constructs (β = 0.62, t = 16.78, p = 0.001) for this sample (Figure 3). In this single-site context, the Coefficient of Determination (R² = 0.39) and the effect size (f² = 0.64) suggested that perceived value explained a moderate proportion of variance in eWOM, while the predictive relevance (Q² = 0.35) revealed that the model was acceptable [77].

5. Discussion

The findings of this study indicated that there may be a positive association between perceived value and the engagement in eWOM (β = 0.62, t = 16.78, p = 0.001), suggesting that within this sample, higher perceived value, comprising perceived quality, perceived service outcome, non-monetary costs, and organizational image, was associated with a greater patients’ eWOM engagement. Although perceived value appears to have a relatively large effect size in association with the explained variance of eWOM (f² = 0.64) [76], it is highly recommended to interpret this finding within the context of the specific sample. Thus, the findings of this study may be a starting point for future research, however they should be interpreted with consideration to the study-specific context and sample characteristics [42].

5.1. Theoretical Contributions

Firstly, it may offer potential preliminary support for the Weinstein’s adapted multidimensional perceived value framework in the health care context [19]. Accordingly, while prior research has explored perceived value in other fields, this study extends its application to credence-based services, such as health care services [81].

Secondly, the findings may contribute to the existing body of literature about eWOM, perceived value, and health care services. Prior research has shown that patients’ online evaluations and service perceptions are associated with their eWOM intentions and health-related decisions [13], while broader eWOM literature suggests that higher perceived value may be associated with greater patient eWOM engagement, often through mediating factors such as satisfaction and attitude [82].

Thirdly, by integrating non-monetary costs, the study extends previous research by emphasizing that positive patient costs, such as reduced waiting time, lower effort, and decreased psychological burden, may also play a role in eWOM engagement. In this study, this finding reinforces the importance of considering both perceived benefits and the positive aspects of patient sacrifices, when examining patients’ perceived value and its potential association with eWOM in Cardiology services [16,54].

5.2. Practical Contributions

From a managerial perspective, within this study, it was highlighted that perceived value may be associated with greater eWOM engagement. Improving perceived value requires attention to the multiple facets of health care services, including service quality, organizational image, perceived service outcome, and non-monetary costs, all of which may be associated with the patients’ eWOM engagement [83].

Service quality becomes essential across both technical and functional attributes of health care services. Health care organizations need to provide high-quality medical care and ensure that tangible elements, such as modern facilities and equipment, are reliable and available. These tangible aspects, combined with functional quality, enhanced patients’ perceived value and comfort during their care [42,81]. At the same time, how staff interact with patients, by showing empathy and reassurance, makes patients feel understood and cared for, which may be linked to more positive reported experiences [41,52,55].

A health care organization’s image may have a significant role in shaping perceived value. A strong image, built on both performance excellence and social responsibility, fosters patient trust and confidence, making patients more likely to engage in eWOM [56,57]. At the same time, reducing non-monetary costs such as long waiting time, complex administrative processes, and psychological stress further may be linked to perceived value. Health care organizations can address these challenges by simplifying administrative processes, implementing online appointment systems, or interacting with patients by email or online consumer relationship management programs, thereby reducing both effort and stress while improving overall patient perceived value [55,83]. Moreover, health care organizations may need to implement active online image and reputation management systems, so as to monitor the patients’ generated eWOM content [84] and to strategically use it in the co-creation value experience of patients [85]. By coordinating improvements in clinical quality, perceived service, non-monetary costs, and organizational image, health care organizations may increase the perceived value, foster loyalty, and encourage positive eWOM, which is a core factor in patient decision making and health care organization reputation [4,86].

At a broader level, the association between perceived value and eWOM engagement may be interpreted as a potential shift toward more participatory and patient-centered health environments [87,88]. However, it should be emphasized that the empirical model of this study is situated at the individual level; therefore, any broader societal implications remain theoretical, as they are not directly examined within the present analysis. From this perspective, based on the fact that eWOM actively contributes to building the aggregated blocks of user-generated content, which accumulates online and remains accessible over time for other users [89,90], in health care, it may become a distribution infrastructure of information. Aggregated eWOM may contribute to the reputational governance mechanisms in the health care field by increasing visibility of patient experiences, which, in its turn, may stimulate improvements in service quality and non-monetary costs. In addition, eWOM information may complement already formal patient satisfaction monitoring systems, offering policymakers and health care specialists real-time insights into the service performance or patient concerns [89].

5.3. Limitations

Despite its contributions, the findings of this study should be interpreted with caution. Firstly, this study extended the application of Weinstein’s Perceived Value framework in an adapted format for the health care context, in a single-site, using a cross-sectional design. This limits the establishment of causal relationships, and the direct transferability of the findings to other health care systems, populations, or cultural settings, and the generalizability of the findings. Longitudinal research would provide more insights into how perceived value and eWOM intentions develop over time in the health care field.

Secondly, in this study, purposive sampling was used to target patients with internet access, which may also limit the generalizability of the findings, particularly to rural populations or those with low electronic health literacy (eHL) [91]. The sampling method might have also introduced selection and volunteer bias, excluding the patients who had discontinued care. Moreover, the study focused exclusively on Cardiology patients from Romania, which may limit its applicability and the generalizability to other health care settings, such as other medical specialties or long-term care, where patient experiences and eWOM engagement may differ substantially [1].

Thirdly, the reliance on self-reported measures may be subject to social desirability or recall bias, and reported intentions might not fully reflect actual online intention. Future research can rely on other online channels, such as Social Media, online reviews, or Patient Online Communities, to provide a more comprehensive overview [83].

Finally, the study did not consider potential moderating and mediating factors. The moderating factors, such as age, health status, prior experience with digital tools, or cultural differences, may influence the association between perceived value and eWOM intentions [92]. Future research could also explore the mediating factors that explain how perceived value is associated with eWOM [93]. Patient satisfaction is among the most frequently mentioned and researched mediators, as perceived value is a key antecedent of satisfaction in the Services field [4]. When patients perceive that health care services provide significant benefits in comparison to the efforts or costs involved, they are more likely to be satisfied and subsequently share eWOM-type messages. Other potential mediators include trust in health care providers and patient loyalty, both of which have been shown to encourage recommendation behaviors and eWOM in health care services.

5.4. Future Research Directions

Despite the limitations, the study offers useful theoretical and practical insights and provides a starting point for future research in order to extend or assess the association between perceived value and eWOM engagement across different other health care contexts. Firstly, health care managers and administrators may use these results to design strategies to enhance perceived value and encourage positive eWOM. Future studies could, therefore, investigate how managerial strategies, as for instance, improvements in service quality, patient communication, and non-monetary costs, translate into positive eWOM among their patients.

Secondly, health care professionals, including physicians and nurses, could use these insights to improve interactions. Future research may examine how interpersonal factors such as empathy, trust, and patient-centered communication influence perceived value and patients’ eWOM intentions.

Thirdly, health policymakers and public health institutions may benefit from understanding how perceived value shapes eWOM communication in the medical field. Future studies could explore how patient experience initiatives, health reforms, or digital health policies support building perceived value and eWOM intention and behavior across different health systems.

Fourthly, developers of digital health technologies, such as telemedicine platforms or patient portals, may shape how patients evaluate health care services, and encourage them to share their experiences online, by designing more user-friendly communication channels that enhance the perceived ease of use and usefulness among patients. At the same time, it is also important to design and implement feedback and rating systems for health care organizations and for the health care information that would facilitate credible eWOM exchange. For health care organizations and for a community, experts of advanced technology may use different analytic methods to identify the antecedents of perceived value and eWOM engagement as aggregated discussions. Further studies may be conducted in order to investigate the use of AI-oriented emotion analysis for monitoring positive or negative eWOM.

Marketing specialists may also benefit from the study’s findings. Future research could examine how marketing strategies that highlight patient value can trigger more positive eWOM. Health care organizations can use patient-centered communication campaigns, by sharing success stories, emphasizing quality of care, and promoting safety and empathy during patient–physician interactions, to build trust and encourage patients to share their experiences online. With the help of AI, predictive analytics, based on eWOM messages, may be implemented in order to anticipate the patient needs, as well as strengthen the online image or reputation of the health care provider.

Digital engagement strategies, including patient portals, Social Media campaigns, and interactive health applications, can provide patients with convenient access to information and support, increasing satisfaction and the likelihood of positive eWOM. Additionally, transparent feedback can reinforce a culture of openness, making patients feel that their eWOM messages are useful.

6. Conclusions

By using Weinstein’s adapted Perceived Value framework, in this study, patients’ perceived value may be associated with greater eWOM engagement in health care services. Thus, in this specific context, greater perceived value in Cardiology services may be linked to improved patient experiences and eWOM engagement.