Health, Spirituality and Medical Ethics

Introduction

atients with severe or chronic diseases often face challenges in advocating for their rights and needs, making them vulnerable to inadequate care [1]. Vulnerable patients are considered to be unable to fully express and defend their rights and needs. Therefore, they need someone to advocate for them [2]. Bazmi et al. and Rowaghi et al. found that most hospitalized patients have little knowledge about their rights. Effective patient advocacy improves the quality of care, reduces treatment costs, assists patients in covering treatment expenses, and strengthens the nursing profession [3, 4]. Therefore, the failure to fulfill this role can diminish the power of nursing and lead to nurses leaving the profession.

Nurses have the most direct contact with patients among other members of the treatment team [5]. They constitute the largest group in the patient care team [6] and play an important role in promoting the quality of healthcare [7]. Numerous care opportunities provide nurses with conditions conducive to patient advocacy, which enhances the public image of nursing and fosters trust between patients and nurses. Consequently, they can act more effectively in patient advocacy [8].

The quality of nursing care is a key component of healthcare management, and accurate measurement of this quality is essential for improving health services [9, 10]. Evaluating health service quality requires accurate, valid, and reliable instruments [11]. Patient advocacy is a central role of nurses, but assessing their performance in this area requires precise and reliable scales [12-14].

Previous studies have developed scales to assess various aspects of patient advocacy, such as protecting patient autonomy, acting on behalf of patients, promoting social justice [15], nursing advocacy beliefs and actions from a protective perspective [16], and social support for self-care in patients with certain diseases [17, 18]. While these scales have provided valuable insight into patient advocacy, they have certain limitations, such as being focused on certain diseases or concentrating on a particular aspect of advocacy. Therefore, a more comprehensive and adaptable scale is needed to evaluate the overall performance of nurses in patient advocacy. Therefore, a more comprehensive and adaptable scale is needed to evaluate the overall performance of nurses in patient advocacy.

Previous scales developed by Bu and Wu [15] and Hanks [16] were based on a definition of patient advocacy established prior to 2010 and may not fully capture the evolving concept of patient advocacy. Similarly, Naderimagham et al. [17] and Garvey et al. [18] focus on specific aspects of patient care and support, making them unsuitable for comprehensively assessing patient advocacy. Therefore, there is a need for a new scale that can measure patient advocacy more comprehensively. This study aimed to develop the patient advocacy scale (PAS) and evaluate its psychometric properties, including content validity, construct validity, and reliability, among nurses in intensive care units.

Methods

Design and setting

This study used an exploratory sequential mixed-methods design conducted between April and September 2023. The study consisted of two phases: (1) a qualitative approach to generating items and designing the scale, and (2) a cross-sectional study to assess the psychometric properties of the scale.

Phase 1: Item generation and instrument design

In the first phase of the study, we employed Schwartz-Barcott and Kim’s Hybrid Model, which includes theoretical, fieldwork, and final analytical phases [19].

Theoretical phase

In the theoretical phase, we conducted a comprehensive review of the literature on patient advocacy, including a concept analysis by Abbasinia et al. [8]. Based on this review, we identified key dimensions of patient advocacy, including safeguarding, apprising, valuing, mediating, and championing for social justice. In the fieldwork phase, we used guided content analysis to develop a pool of potential items for the instrument, based on the identified dimensions and sub-dimensions of patient advocacy.

Fieldwork phase

In the fieldwork phase, a conventional qualitative content study was conducted to explore nurses’ experiences with patient advocacy. Participants were selected using purposeful sampling to maximize variation in age, gender, education, work experience, workplace, and organizational position [20]. The inclusion criteria were having a bachelor’s degree or higher in nursing, at least three years of work experience as a nurse, the ability to speak Persian, the ability to communicate personal narratives in a one-on-one interview setting, and consent to participate in the study. In-depth, face-to-face, semi-structured interviews were conducted with 21 nurses from Shahid Beheshti and Khairin Salamat Hospitals in Qom, Iran, until data saturation was reached. The interviews were conducted after the nurses’ work shifts, in the staff break room. The collected data were analyzed through qualitative content analysis and inductive analysis methods, utilizing MAXQDA software, version 10. The interviews were conducted individually, recorded, and transcribed verbatim by the first author. The transcripts were analyzed using a content analysis approach, with initial codes grouped into subcategories and categories based on similarities and differences [21]. The study found several key dimensions of patient advocacy, including safeguarding patients’ rights, promoting patients’ autonomy, and advocating for social justice, which were used to develop items for the PAS.

Final analytical phase

The final analytical phase of the study involved comparing, working, and writing up the findings from the theoretical and fieldwork phases. A synthesis was conducted to determine the characteristics of patient advocacy in the intensive care units, with a focus on identifying similarities and differences between the theoretical and fieldwork findings. The researcher addressed questions related to the differences and inconsistencies between the literature and the field, as well as identifying unique characteristics of patient advocacy in the field. The categories and classes identified in the theoretical and fieldwork phases were combined or reported separately based on their similarities and differences. The synthesis of findings led to a change in the concept and the development of a comprehensive definition of patient advocacy in the intensive care units.

Instrument design

Based on the attributes of patient advocacy identified in the Hybrid Model, a pool of 63 items was generated. A total of 37 items were extracted from the literature review, and 26 items were derived from the qualitative phase. The research team reviewed and refined the items, removing overlapping or similar items, resulting in a preliminary version of the instrument with 51 items into 5 subscales.

Phase 2: Psychometric evaluation of the instrument

To determine the psychometric properties of the scale, face validity, content validity, construct validity, reliability and responsiveness are examined.

Face validity

Both quantitative and qualitative methods were used to evaluate the face validity of the scale. The quantitative method involved administering the instrument to 10 nurses, who rated each item on a 5-point Likert-type scale of necessity and importance. The IIS was calculated for each item, with a score of 1.5 or higher considered appropriate [22]. The qualitative face validity was assessed through face-to-face interviews with nurses selected through purposive sampling based on their education, work experience, and workplace. The nurses were asked to evaluate the difficulty, relevance, and ambiguity of each item. This step helped to refine the wording, improve the comprehensibility, and enhance the overall appropriateness of the scale items for the target population.

Content validity

The content validity of the PAS was evaluated by a team of ten clinical and educational nursing specialists using both quantitative and qualitative methods. The quantitative content validity was assessed using the content validity ratio (CVR), item content validity index (I-CVI), and scale-level content validity index/averaging calculation method (S-CVI/Ave) [22]. The experts rated each item based on its essentiality, usefulness, and necessity, with a CVR score of 0.62 or higher considered acceptable according to Lawshe’s table. The I-CVI was calculated using a 4-point Likert scale, with scores of 0.78 or higher considered acceptable, and the S-CVI/Ave was calculated using the average CVI, with scores of 0.90 or higher considered acceptable [23]. The qualitative content validity was assessed through expert evaluation of item clarity, simplicity, ambiguity, placement, appropriateness, and scoring. Appropriate corrections were made based on expert feedback.

Item analysis

Item analysis was conducted by administering the PAS to 30 nurses randomly selected using a random numbers table from Shahid Beheshti and Khairin Salamat Hospitals in Qom, Iran. Internal consistency was assessed using cronbach’s α coefficient and item-total correlation for each item. A cronbach’s α coefficient of 0.7 or higher and an item-total correlation of 0.3 or higher were considered desirable [23].

Construct validity

Construct validity of the PAS was assessed using exploratory factor analysis (EFA) with a sample size of 214 nurses from Shahid Beheshti and Khairin Salamat Hospitals in Qom, Iran, who met the inclusion criteria of having at least three years of work experience and willingness to participate [24]. The KMO and Bartlett’s test of sampling adequacy were used to assess the suitability of the data for factor analysis. Maximum likelihood and Varimax rotation methods were used to extract factors, and criteria, such as Kaiser’s criterion, scree plot, and cumulative variance were examined. A cutoff value of 0.3 for item factor loading was used [23]. Finally, the extracted factors were named.

Reliability

The reliability of the PAS was assessed using internal and external measures. Internal consistency was evaluated using the cronbach α test with a sample of 50 nurses, with values of 0.60 or higher considered satisfactory. External reliability was assessed using the intraclass correlation coefficient (ICC) and standard error of measurement (SEM) with the same sample of 50 nurses. In both the internal and external phases, nurses were randomly selected using a random numbers table from Shahid Beheshti and Khairin Salamat Hospitals in Qom, Iran. Correlation coefficient values of 0.8 or higher were considered indicative of good reliability [24].

SEM, minimal detectable change (MDC), minimal important change, and interpretability

Responsiveness or sensitivity to change was evaluated using the SEM and MDC statistics. SEM was calculated based on data from the reliability phase using the Equation 1:

1. SEM=SD×(√[1-ICC])

where SD is the standard deviation of the sum of test and retest values. MDC was calculated using the Equation 2:

2. MDC=SEM×√2×1.96

with a percentage MDC lower than 30% considered acceptable and values lower than 10% considered excellent [25]. The minimal important change (MIC) was calculated using the Equation 3:

3. MIC=0.5×SD

of the Δ Score, where SD is the standard deviation of changes between test and retest, and an effect size of 0.5 was used. Interpretability was assessed by evaluating ceiling and floor effects in the construct validity phase, with effects occurring when 15% or more of participants achieve the lowest or highest possible score [26].

Feasibility

Feasibility of the PAS was assessed by calculating response time and the number of missing values [27]. A missing rate of 15% to 20% is expected in educational and psychological studies [28].

Data analysis

Data were analyzed using SPSS software, version 21 for quantitative analyses and MAXQDA software, version 10 for qualitative data analysis.

Results

Item generation

An item pool containing 51 initial items was created by the research team during the first discussion session, based on the findings from the concept explanation stage. In the second discussion session, overlapping or similar items were merged or removed, resulting in a final instrument with 44 items for the psychometric phase.

Validity

Face validity

In the qualitative phase of face validity, the writing style of four ambiguous items was revised according to feedback from participants. In the quantitative phase of face validity, the item impact score (IIS) of all items was above 1.5; therefore, no item was deleted.

Content validity

In the qualitative phase of content validity, three unclear items were revised based on expert feedback, and seven items were merged with others due to overlapping, resulting in a reduction to 34 items. In the quantitative phase of content validity, with the decision of the research team, three items with a CVI score of less than 0.78 and two items with a CVR score of less than 0.62 were excluded. The S-CVI was calculated as 0.93. At the end of this phase, a scale with 29 items was prepared for the item analysis phase.

Item analysis

The internal consistency of this PAS was confirmed with a cronbach’s α value of 0.862. The total item correlation of all items was above 0.3. This indicates that each item correlated well with the total score.

Construct validity

The PAS with 29 items entered the construct validity phase. The detailed profiles of participants in the EFA phase are presented in Table 1. According to the results of the KMO and Bartlett tests, the data from 214 samples were suitable for factor analysis (Table 2). Using the 0.3 cut-off level, two items failed to load onto any factor and were removed. The scree plot is shown in Figure 1.

According to the plot in Figure 1 and the decision of the research team, a four-factor model with eigenvalues greater than 1 and factor loadings of 0.3 or higher was confirmed. The factors were named “protecting,” “valuing,” “informing,” and “mediating,” and they accounted for 49% of the total variance of the scale variables (Table 3).

Reliability

The PAS had a cronbach’s α coefficient of 0.862 and an ICC of 0.946 (95% CI, 0.865%, 0.979%), indicating good internal and external reliability. The SEM of the total scale was also desirable. The minimal important change percentage (MAC%) for each factor was less than 30%. Ceiling and floor effects were less than 15% for the total scale and its factors, which is an acceptable level of interpretability..

Feasibility

The average response time for the designed scale was 20 minutes, with a response rate of 99.6% for the entire scale.

Scoring

The PAS was a self-report measure consisting of 27 items across four subscales: Safeguarding (8 items), valuing (10 items), informing (5 items) and mediating (4 items). Participants rated each item on a five-point Likert scale ranging from 0 (never) to 4 (often). To aid understanding and facilitate score comparisons, all scores were converted to a scale of 0-100 using a linear conversion scoring method for each factor. Specifically, the factor score was calculated as follows:

Discussion

As moral representatives of patients, nurses have a responsibility to be prepared and able to advocate for them [2]. Patient advocacy is essential for ensuring comprehensive patient care, quality assurance, and promoting patients’ independent decision-making, as well as helping them understand their rights [29]. Despite its importance, few scales exist to measure this critical function [30]. This study was conducted to design a valid and reliable scale to measure the performance of nurses in patient advocacy. The resulting PAS consisted of 27 items distributed across four subscales: Safeguarding (8 items), valuing (10 items), informing (5 items), and mediating (4 items). These subscales are consistent with the nursing ethical codes provided by the ICN, emphasizing the importance of respecting patients’ rights, providing health information, addressing health-related needs, and ensuring social justice in access to healthcare resources [8].

Compared to similar instruments, such as the scale developed by Bu and Wu [15], which included 43 items across three dimensions (cognitive, affective, and behavioral), and the Protective Nursing Advocacy Scale developed by Hanks [16] with 43 items focusing mainly on protective aspects, the PAS has fewer items and a more comprehensive conceptual structure. While Bu and Wu’s [15] instrument emphasized nurses’ attitudes and Hanks’ tool focused on protective advocacy, the PAS covers broader domains, including valuing and informing, which are essential for holistic patient advocacy. Furthermore, in contrast to Garvey et al.’s scale [18], which targeted supportive care needs of cancer patients without specifically addressing advocacy behaviors, the PAS directly measures nurses’ actions related to advocacy.

Regarding psychometric properties, previous tools often reported limited content validity or focused mainly on internal consistency. For instance, Hanks reported a cronbach’s α of 0.94 but provided limited evidence of content validity [15, 16]. In contrast, the PAS demonstrated strong content validity through expert evaluations (CVI and CVR), face validity confirmed by nurses, and robust reliability (cronbach’s α=0.92). Moreover, EFA in this study supported a clear four-factor structure, explaining 49% of the total variance, which is comparable to or higher than similar scales.

Protecting is an important subscale of patient advocacy, as hospitalized patients are at risk of various hazards, including medical errors, misconduct, negligence, or inappropriate treatments by healthcare providers [31]. The results of this study emphasize to nurses that they should protect their patients from medical errors and other healthcare professionals’ misconduct. Valuing is another critical subscale of patient advocacy. This involves adherence to fairness and non-discrimination in the provision of care, as well as respect for patient privacy and their beliefs, culture, values, and preferences. Baldwin [32], Negarandeh et al. [33], and Mortell et al. [34] highlighted the importance of valuing in patient advocacy, with Baldwin identifying it as the cornerstone of her philosophy of patient advocacy and Negarandeh et al. emphasizing respect for the individuality and dignity of the patient. However, in cases where patients’ decisions may harm their health, their preferences should be overridden [33,.

Informing and interceding are also important subscales of patient advocacy. Nurses should not only answer patients’ questions but also inform them about the treatment process and legal issues related to their care. Informing is considered an essential component of patient advocacy by Sundqvist et al. [36] and Davoodvand et al. [5]. In addition, vulnerable patients may require someone to speak and intercede on their behalf [36, 37]. Therefore, nurses should act as mediators between patients and healthcare providers, advocating for their patients’ needs and concerns [36, 38].

This study has several limitations. First, the sample was restricted to nurses working in intensive care units of selected hospitals, which may limit the generalizability of the findings to other clinical settings or healthcare professionals. Second, the research was conducted in a single city (Qom, Iran), potentially affecting the applicability of the scale across different cultural and healthcare contexts. Third, since the PAS is a self-report instrument, responses may be subject to social desirability and response biases, which could influence the accuracy of the results.

Conclusion

The PAS is designed to evaluate nurses’ performance in patient advocacy and has demonstrated good validity and reliability. In this scale, the key components of patient advocacy included safeguarding, valuing, informing, and mediating. It is recommended that the validity and reliability of this scale be evaluated in future studies.

Ethical Considerations

Compliance with ethical guidelines

The study was approved by the Ethics Committee of Qom University of Medical Sciences, Qom, Iran (Code: IR.MUQ.REC. 1399.174). Prior to the commencement of the study, the objectives and procedures were explained to all participants, and written informed consent was obtained for their participation. Additionally, the participants were informed that their information would remain confidential and that they could withdraw from the study at any stage at their own discretion.

Funding

This research was financially supported by the Deputy of Research and Technology of Qom University of Medical Sciences.

Authors' contributions

Data analysis: Mohammad Abbasinia and Atye Babaii; Conceptualization, methodology, validation, writing, editing, and finalization of manuscript: All Authors.

Conflict of interest

The authors declared no conflicts of interest.

Acknowledgments

The authors would like to thank all participants in this research for their cooperation.