Introduction
Taiwan, an aging society that may be a superaged country as early as 2026 (National Development Council, Taiwan, ROC, 2020), faces a shortage of trained physicians and low resident coverage for healthcare. As the healthcare demands of the Taiwanese are increasingly challenging to meet, the nursing profession initiated changes and, in 2000, added a nurse practitioner (NP) title to the Nurses Act to better meet public healthcare needs. However, the NP care system took many years to develop, and the Taiwan Association of Nurse Practitioners (TANP) worked with the Department of Health and Welfare to set up regulations and standards for education, licensure, and practice (Tsay et al., 2019). The NP program was finally implemented in 2006. Since then, NPs have become important healthcare members in acute care practice. The TANP defines NP as an advanced practice nurse who provides medical and nursing care to populations with acute and chronic illness (TANP, 2022). It lists five dimensions or domains of NP competency in practice: providing healthcare, collaboration, education, care quality/research, and leadership/professionalism (TANP, 2022).
NPs originated in the United States during the 1960s mainly to increase access and quality of care for its underserviced population. The profession of NP was rapidly adopted throughout the United States, Canada, the United Kingdom, Australia, and other countries. Research has also supported the positive practice outcomes of NPs over the past 30 years. A recent systemic review further documented that NPs provide access to and effective care in a variety of settings, are equal in quality outcomes and safety, and are cost-effective compared with physicians (Geller & Swan, 2021). Despite these positive practice outcomes, there remains a lack of consensus on the definition of the NP role.
Many nursing organizations identify core competencies to ensure care quality and safe practice for NPs (American Association of Colleges of Nursing, 2015). For example, the National Organization of Nurse Practitioner Faculty (NONPF) of the United States released the "Nurse Practitioner Core Competencies," comprising the nine domains of scientific foundation, leadership, quality, practice inquiry, technology, information literacy, policy, health delivery system, and ethics (NONPF, 2022). Moreover, the American Association of Critical Care Nurses (AACN) announced standards of professional performance for NPs, including professional practice, education, collaboration, ethics, advocacy, and system/organizational thinking (AACN, 2021). According to the Nursing and Midwifery Board of Australia, the NP standards framework is based on four domains: clinical, education, research, and leadership (Nursing and Midwifery Board of Australia, 2021). Furthermore, the International Council of Nurses emphasized the NP's role in direct patient care as "integrating knowledge and skills related to nursing and medicine to assess, diagnose, and manage patients in healthcare settings" (International Council of Nurses, 2020). These organizations have collectively reported at least 354 specific competencies for all advanced practice registered nurses, including NPs, and stated them as core competencies (Chan et al., 2020; Crabtree et al., 2002). Chan and colleagues, based on NONPF's practice domain, used a Delphi method to reduce the redundancy of these competencies, resulting in 49 final major competencies. Many of these practice domains and competencies are redundant and overlapping in nature.
Clinical competency validation is essential for NP practice and patient safety. According to the AACN, competence is defined as "the array of abilities (knowledge, skills, and attitudes) across multiple domains or aspects of performance in a certain context. Competence is multi-dimensional and dynamic. It changes with time, experience, and settings" (AACN, 2017). Although there have been many discussions around what represents clinical competency and assessment, there remains a lack of clear and consistent definitions and measurements (D'Aoust et al., 2022). However, three scales are currently used to measure NP competencies: two for primary care NPs and one for physician assistants. However, none of these measures are specifically designed for acute care NPs (ACNPs).
First, Ackerman et al. (1996), based on the Strong Model of Advanced Practice, conducted a study that identified domains of practice and competencies under each domain, including direct comprehensive care, support of systems, education, research, publication, and professional leadership for primary care NPs. They developed an advanced practice role delineation (APRD) based on domains and activities, although they did not report the psychometric properties of the APRD. More recently, Sevilla Guerra et al. (2022) conducted a factor analysis of APRD (38 items) and obtained six main factors, including expert care planning, comprehensive care, interprofessional collaboration, training/education, evidence-based practice and research, and leadership, accounting for 63.72% of the total variance (Sevilla Guerra et al., 2022). However, APRD items are designed primarily for primary care NPs.
Second, Abdallah et al. (2005) developed the EverCare Nurse Practitioner Role and Activity Scale (ENPRAS) to measure the performance frequency of role activities by EverCare NPs, who provide primary care to nursing home residents. ENPRAS identified six main roles and 99 activities of NPs, including collaborator, clinician, care manager/coordinator, counselor, communicator/cheerleader, and coach/educator. Although ENPRAS exhibited psychometric properties, the tool is very long, and items are listed in excessive detail, which may limit applicability among ACNPs.
Third, Lin et al. (2016) published the Nurse Practitioners' Roles and Competencies Scale, suggesting it to be a valid and reliable instrument for assessing competencies. However, while evaluating the Nurse Practitioners' Roles and Competencies Scale items, it was found that most reflected the functions of physician assistant. Therefore, the scale is not deemed appropriate for the current NP role.
In summary, the focus of ACNPs is mainly on direct patient care provided to assess, diagnose, and manage patients in acute care settings. Although the competencies of ACNPs are important to providing quality and safe patient care, as existing tools focus on measuring the primary care role of NPs, no current tools specifically measure the competencies of ACNPs. Therefore, this study was designed to develop and investigate the psychometric properties of the Acute Care Nurse Practitioner Competencies Scale (ACNPCS). The findings are intended to provide a tool to measure the clinical competencies of NPs in acute care settings.
Conceptual Framework
The Strong Model of Advanced Practice was adopted for this study. According to this model, the five domains of NP practice include direct comprehensive care, support of systems, education, research, publication, and professional leadership. Direct comprehensive care includes patient-focused activities such as procedures, assessments, interpretation of data, and patient counseling. Support of systems includes NP activities that contribute to the highest expectations of nursing service. The domain of education includes activities that educate patients, caregivers, and students about health and illness. The domain of research includes providing better patient care through scientific inquiry. Finally, publication and professional leadership include promoting the dissemination of nursing and healthcare knowledge (Ackerman et al., 1996).
The conceptual elements of empowerment, collaboration, and scholarship that contribute to advanced practice domains, the approach to patient care, and the professional attitude in clinical practice were emphasized. In addition, the model recognized the progression continuum of NP experience from novice to expert within each domain that enhances the development of clinical knowledge and supports career development in practice (Ackerman et al., 1996; Mick & Ackerman, 2000).
Methods
Design
We adopted the Strong Model of Advanced Practice, and mixed-method research was conducted using a sample of experienced NPs. First, we used a focus group of seven experienced NP leaders currently working in a medical center or a regional or community hospital to identify the clinical competencies of their practice. After three rounds of discussions and revisions, 46 initial ACNPCS items were developed. Second, we implemented consensus validation using two Delphi rounds with 10 experienced NPs with at least 5 years of clinical practice, which finalized the ACNPCS as a 39-item tool. Third, these competencies were evaluated by an expert panel of nine NPs over three review rounds to assess the relevance, clarity, and measurability of each. The panelists were asked to rate each of the 46 competencies for relevancy on a Likert scale ranging from 1 to 4 (1 = strongly disagree to 4 = strongly agree). Next, the ACNPCS was revised to 36 items based on the expert panel's suggestions. The panel of experts on NP practice was recruited nationwide with assistance from the TANP, which represents over 90% of NPs in Taiwan. Finally, we conducted a nationwide quantitative survey on 390 NPs from 125 hospitals working in pediatric, obstetrics and gynecology, psychiatric, emergency room, intensive care unit, surgical, and medical departments to evaluate the reliability and validity of the ACNPCS. The hospital institutional review board approved this study (CMUH111-REC3-059).
Participants
This study recruited participants and collected data from an online survey sent to TANP members from March to May 2022. The inclusion criteria were NPs who (a) held national NP certification and (b) had worked in acute care settings for at least 1 year. We invited eligible participants to take part in this study via email, which included detailed information about the study's purposes and procedures. The eligible NPs who agreed to participate then filled out the online questionnaire using a specific uniform resource locator attached to the invitation mail. Three hundred ninety NPs completed the online survey. The effective sample size was 390, which was sufficient to conduct a six-factor model in exploratory factor analysis (EFA) with the minimum necessary sample size calculated as a variables-to-factors ratio (36/6 = 6, n = 300; Mundfrom et al., 2005).
Measurement
The ACNPCS is a 36-item scale that evaluates the extent of time an NP spends on each activity or competency listed. There are six subscales in the ACNPCS, including providing healthcare (11 items), evaluating care (three items), collaboration (four items), education (four items), care quality/research (six items), and leadership/professionalism (eight items). The ACNPCS is a 5-point Likert scale scored from 1 to 5, with 5 = to a very great extent, 4 = to a great extent, 3 = to some extent, 2 = a little, and 1 = not at all. The total possible score range is from 36 to 180, with higher scores indicating more time is spent on performing NP competencies in acute care practices.
Statistical Analysis
Demographic characteristics were analyzed using descriptive statistics such as number (n), percentage (%), mean, SD, and range. The skewness and kurtosis were utilized to examine the multivariate normality of data, which noted that the skewness should be less than 2 and that the kurtosis should be less than 4 (Tabachnick et al., 2007). Internal consistency reliability and the test-retest method were conducted to assess the reliability of the ACNPCS, and EFA, confirmatory factor analysis (CFA), and known-group comparison were conducted to assess construct validity. Both the Bartlett's test of sphericity and the Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy were performed to examine the suitability of the ACNPCS for EFA. The statistical significance of the Bartlett's test of sphericity and values of KMO above 0.50 indicate that the data were suitable for EFA (Hair et al., 2006). Factors were extracted using principal component analysis in EFA, and the rotation method was varimax with Kaiser normalization. The number of factors was determined using the eigenvalues, which should be above 1, and the variance was explained by a specific factor based on a factor loading above .40 (Costello & Osborne, 2005). For CFA, the chi-square test, the [chi]2/df ratio, and several model fit indices such as goodness-of-fit index (GFI), comparative fit index (CFI), Tucker-Lewis index (TLI), root mean square error of approximation (RMSEA), and standardized root mean residual (SRMR) were used to evaluate model fit. The model was identified as adequate based on the following criteria: statistical significance of the chi-square test, [chi]2/df ratio less than 3, GFI and CFI close to .90, TLI close to .95, and RMSEA and SRMR less than .05 (Sun, 2005). In addition, the known-group comparison between novice NPs and expert NPs in clinical competencies was evaluated to further support the construct validity of the ACNPCS.
The Cronbach's alpha of internal consistency and test-retest were conducted to assess the reliability of the ACNPCS. The statistically significant level ([alpha]) was .05. All statistical procedures were conducted using SPSS 24.0 and AMOS (IBM Inc., Armonk, NY, USA) software.
Results
Participants' Characteristics
Most of the participants were female (96.7%), with a mean age of 43.36 (SD = 6.07, range: 28-58) years. Most were educated at the college/university level (74.1%), with 25.9% holding a master's or PhD degree. The participants had an average of 8.52 years of practice experience (SD = 4.82, range: 1-15 years). Most held an internal medicine (n = 186, 47.7%) or surgical (n = 182, 46.7%) license and practiced in medical (n = 144, 36.9%) or surgical (n = 130, 33.3%) units in acute care settings. A majority held an NP advancement ladder (72.8%). In terms of experience level, the largest proportion was at the NP advanced beginner level (31.8%), followed by novice (19.0%), competent (9.7%), and expert (7.9%; see Table 1).
Descriptiveness of the Acute Care Nurse Practitioner Competencies Scale
The mean score of the scale was 144.21 (SD = 21.17, range; 65-180). Higher scores indicate that NPs spend more time on competencies in their clinical practice (Table 2). Item mean scores for the scale are also shown in Table 2. The skewness and kurtosis for all of the items met the conditions of multivariate normality, meaning that the skewness was less than 2 and kurtosis was less than 4 (Tabachnick et al., 2007).
Reliability of the Acute Care Nurse Practitioner Competencies Scale
Internal consistency was assessed for the ACNPCS. Cronbach's alpha value for the final scale was .92. The ACNPCS identified six factors, and the Cronbach's alpha value of these factors ranged from .71 to .89. The test-retest reliability was investigated by administering the 36-item scale to 20 NPs, with a gap of 14 days. The scores of the two tests were highly correlated (r = .85, p < .001). The correlation for items was also calculated, and all were positively correlated (p < .01).
Construct Validity of the Acute Care Nurse Practitioner Competencies Scale
Exploratory factor analysis
To investigate the underlying factors measured using the ACNPCS, an EFA was conducted using principal axis factor analysis and promax rotation. The correlation matrix analysis showed that all items were significantly correlated with each other (range: .24-.83). The result of the Bartlett's test of sphericity met statistical significance ([chi]2 = 12144.47, p < .001), indicating that adequate correlations existed among the items of the extracted factors. Moreover, the KMO measure of sampling adequacy was .96, suggesting that data for EFA exhibited good partial correlation. In other words, the data of this study were suitable for EFA.
According to the eigenvalues (> 1), scree plot, and the cumulative explained variance, six factors were determined in the principal factor analysis in EFA. The EFA results indicate that six factors correlated with factor items and that the variance explained by providing healthcare, evaluating care, collaboration, education, care quality/research, and leadership/professionalism were 47.97%, 3.19%, 3.30%, 2.87%, 10.41%, and 4.79%, respectively. The 36 scale items explained 72.53% of the total variance (Table 3). The factor loading of each item was moderate to high, ranging from .50 to .85, which indicates these items respond to the specific factor (Costello & Osborne, 2005; Table 4). The definition of each factor is stated as follows: (1) providing healthcare: NPs provide direct patient care as integrating knowledge and skills to assess, diagnose, and manage patients; (2) evaluating care: evaluating care outcomes; (3) collaboration: communicate and collaborate with multidisciplinary teams; (4) education: provide relevant healthcare education; (5) care quality/research: assess care outcomes and implement NP practice research; and (6) leadership/professionalism: demonstrate leadership skills and actively participate in professional organizations.
Confirmatory factor analysis
CFA was performed to confirm the six-factor structure of the ACNPCS based on the results of the EFA. CFA results revealed that the six-factor model showed good model fits ([chi]2 = 780.54, p < .01; [chi]2/df ratio = 1.6 < 3). All factor loadings were above .70, and the composite reliability of each factor was also above .60 (providing healthcare = .60, evaluating care = .66, collaboration = .73, education = .67, care quality/research = .72, and leadership/professionalism = .63). The average variance extracted of each factor was also above .50 (providing healthcare = .57, evaluating care = .66, collaboration = .73, education = .67, care quality/research = .72, and leadership/professionalism = .61). In addition, all model fit indices met the standards for satisfaction model fit (GFI = .90, CFI = .98, TLI = .97, RMSEA = .04, and SRMR = .04), which is consistent with the results of the EFA.
Known-Group Validity
Known-group validity was investigated by comparing the ACNPCS scores of novice and expert NPs who were certified under a professional advancement program given by TANP. Many studies confirm that NPs endorsed by an advanced higher clinical ladder show higher competencies or clinical expertise, leadership, research/evidence-based practice, and education (Arthur et al., 2020; Paplanus et al., 2014). The results showed that the total scores for the novice NPs differed significantly from those of the expert NPs (t = 3.26, p = .001). In addition, significant differences between the two groups were observed for each of the six factors (p < .01).
Discussion
This study was designed to develop the ACNPCS using the consensus of experts on NP practice and a quantitative study to test the reliability and validity of the scale. This goal was achieved by developing a final 36-item competencies scale for ACNPs. Furthermore, the results support the reliability and validity of the ACNPCS.
The study used the Strong Model of Advanced Practice to guide the process of developing the ACNPCS. During the qualitative process, we initially followed the five domains of NP practice, namely, direct comprehensive care, support of systems, research, publication, and professional leadership, as specified by the model (Ackerman et al., 1996). However, in the quantitative study, we identified the six domains of the scale, including providing healthcare, evaluating care, collaboration, education, care quality/research, and leadership/professionalism. These domains are comparable with those of the Strong Model of Advanced Practice. Similarly, the Advanced Practice Role Delineation Scale for primary care NPs recognizes six domains: expert care planning, comprehensive care, interprofessional collaboration, training/education, evidence-based practice and research, and leadership (Sevilla Guerra et al., 2022). The different names used for some domains may be attributed to the overlapping factors in the structure and to the complexity of the constructs (Chan et al., 2020). However, in the comparison domains of the Strong Model of Advanced Practice, the Advanced Practice Role Delineation Scale, and the newly developed ACNPCS, one can easily identify that NP practice focuses on providing comprehensive care, interprofessional collaboration, research, and leadership, independent of area of NP practice. Nevertheless, because of the complexity of NP competencies and the possibility of overlapping factors, further research is needed to clarify or support these NP practice domains in acute care settings.
The Cronbach's alpha value of the newly developed scale was .92, showing high internal consistency (a value above .70 is considered adequate; Schmitt, 1996). The six factors of the ACNPCS all earned Cronbach's alpha values between .71 and .89. The reliability of the developed scale was further enhanced and supported by test-retest reliability.
Internal consistency was assessed, and the Cronbach's alpha value for the final scale was .92. The six factors of the ACNPCS earned Cronbach's alpha values ranging from .71 to .89. Test-retest reliability was investigated by administering the 36-item scale to 20 NPs with a gap of 14 days. The scores of the two tests were highly correlated (r = .85, p < .001). Item correlations were also calculated, and all were shown to be positive (p < .01). The test-retest reliability results support the stability of the ACNPCS over time. However, Cronbach's alpha values over .90 and interitem correlation values over .70 may indicate redundancies among items (Taber, 2018). Thus, researchers may focus on refining ACNPCS items in future studies.
With regard to the validity of the ACNPCS, we confirmed construct validity using EFA, CFA, and known-group validity. The EFA showed that all items met the criterion of a factor loading of at least .5 (Meyers et al., 2017). As all 36 items loaded on the six factors with a factor loading above .5, they were all retained in the scale. Furthermore, the CFA results met the recommended standards for evaluating goodness of fit (Meyers et al., 2017). CFA is frequently used in the process of scale development to verify the number of underlying dimensions of an instrument. CFA results also provide evidence of convergent and discriminant validity after being adjusted for measurement error and an error theory of the ACNPCS (factor loading > .7, composite reliability > .6, and average variance extracted > .5).
Known-group validity was utilized to provide additional support for the construct validity of the scale. On the basis of the novice-to-expert model, NPs who are experts are more likely than novice NPs to report higher clinical competencies (Benner, 1982). Therefore, our results support the construct validity of the ACNPCS in terms of being able to distinguish clinical competencies between novice and expert groups.
In summary, the competencies of ACNPs are vital to providing quality and safe patient care. We developed the ACNPCS to measure the clinical competencies of NPs, with the overall results supporting the reliability and validity of the scale.
Strength and Limitations
A valid and reliable tool was developed in this study to measure the clinical competencies of ACNPs. The strength of the study was its use of both qualitative and quantitative methods to develop scale items and comprehensively test for psychometric properties. We applied EFA to conduct item reduction and factor identification and applied CFA to validate these factors. Two types of reliability and three types of validity were established for this scale. The scale's main limitation is that the overlapping factors in its structure may limit the variability of the sample. Another limitation is that an individual item is used as the scoring method. The total scale score is calculated as the sum of the scores of the individual items. This approach may ignore the specific contribution of individual items. Likewise, we used the same sample for a two-step analysis, which may have increased the risk of Type I error. However, the p values of EFA and CFA were satisfactory (p < .01) after the Bonferroni correction (p < .025), indicating the risk of Type I error to be minor. In addition, more data should be collected to verify the reliability and validity of the measure and to modify the specific competencies of the ACNP.
Conclusions
The developed ACNPCS, designed to measure NP competencies in acute care settings, comprises six factors and 36 items and is scored using a 5-point Likert scale. The ACNPCS showed respectable psychometric properties in two reliability tests and three construct validity analyses and is thus recommended for use in evaluating the competencies of NPs working in hospitals. On the basis of the competency profiles identified for current NPs, training curricula may be developed for graduate programs that better train future NPs to ensure their ability to provide quality care and ensure patient safety in practice.
Acknowledgments
The authors thank the Taiwan Association of Nurse Practitioners and the nurse practitioners who contributed to the ACNPCS development and testing process.
Author Contributions
Study conception and design: SLT, SSH, SCC
Data collection: KK, SSH, SLT
Data analysis and interpretation: SSH, SLT, SSC
Drafting of the article: All authors
Critical revision of the article: SLT, SCC
References