Conflict of interest: None.
Development and testing of a patient-centred care competency scale for hospital nurses
Article first published online: 13 NOV 2013
© 2013 Wiley Publishing Asia Pty Ltd
International Journal of Nursing Practice
Volume 21, Issue 1, pages 43–51, February 2015
How to Cite
Hwang, J.-I. (2015), Development and testing of a patient-centred care competency scale for hospital nurses. International Journal of Nursing Practice, 21: 43–51. doi: 10.1111/ijn.12220
- Issue published online: 13 JAN 2015
- Article first published online: 13 NOV 2013
- Manuscript Accepted: APR 2013
- National Research Foundation of Korea
- Korean Government. Grant Number: NRF-2009-327-H00039
- patient-centred care;
This study aimed to develop and test the psychometric soundness of a patient-centred care competency (PCC) scale for hospital nurses. A cross-sectional questionnaire survey was conducted among 594 nurses in two teaching hospitals (response rate 99.5%). Reliability and validity analyses were performed. The PCC scale consisted of 17 items divided into four subscales: respecting patients' perspectives (6 items), promoting patient involvement in care processes (5 items), providing for patient comfort (3 items) and advocating for patients (3 items). The Cronbach's alpha coefficient of the entire scale was 0.92, and those for the subscales were 0.85, 0.81, 0.84 and 0.80, respectively. Multitrait scaling analysis indicated that the four subscales had satisfactory convergent and discriminant validity. Significant correlations were found between total PCC scores and overall self-ratings of patient-centred care performance (r = 0.60, P < 0.001). The PCC scale was therefore determined to be a highly valid and reliable tool.
There has been a continued emphasis on quality improvement and patient safety in health care. Patient-centred care is widely acknowledged as a key characteristic of quality health care. The U.S. Institute of Medicine (IOM) has suggested that health-care professionals be prepared to deliver patient-centred care in addition to emphasizing teamwork, evidence-based practice, quality improvement and informatics competencies. Patient-centred approaches to care are being increasingly incorporated into training for health-care providers. Patient-centred care has been identified as a core competency in the education of nurses. The Quality and Safety Education for Nurses (QSEN) faculty has proposed knowledge, skills and attitudes for patient-centred care competencies that need to be developed in nursing education.
Patient-centred care is a global concept in which components interact. Patient-centred care encourages shared decisions about interventions or management of health problems with the patient and focuses on the patient as a whole person with individual preferences.[2-5] Its components include knowing and respecting patients' values, preferences, and needs; promoting partnership between providers and patients in care decisions; providing for patients' physical and emotional comfort; and advocating for patients.[4-12]
In order to achieve high-quality care, previous studies based on such a conceptualization of patient-centred care have suggested tools for measuring the patient-centredness of care in various health-care settings. Patient-centred care has been measured using both self-report instruments[12-15] and external observation methods focusing on the consultation or interview process with patients.[6, 16] Although some of these tools have been designed for use in hospital care settings, they have focused on the psychosocial environment of health-care settings or have measured patient experiences of care.[13, 17, 18] In addition, several tools have been proposed to measure nurses' perception of the patient-centredness of care.[11, 12] However, these tools have been designed for uses in specific care settings, such as nursing home or long-term aged care settings.[11, 12]
There is a need for appropriate instruments to measure patient-centred care competency (PCC) among hospital nurses in a reliable and economical way. A study in an academic setting measured the PCC of nursing students using six items that tested knowledge related to patient-centred care. However, to the authors' knowledge, no instruments developed for use in hospital settings that address nurses' PCC have been validated. Assessment of such competency will form an important basis with which to support the enhancement of patient-centred clinical practice among nurses.
The aim of this study was to provide an overview of the development of a PCC scale and its psychometric properties. The findings will help to determine which elements of nurses' competency need to be emphasized in order to enhance the transition towards patient-centred care in nursing education and clinical practice.
A cross-sectional survey design was employed for instrument development. The data reported in this paper were collected as part of a larger study of evidence-based knowledge translation among hospital nurses. The overall study was reviewed and approved by the Kyung Hee University Institutional Review Board (KHU IRB 2009-003).
Study setting and participants
This study was carried out in two teaching hospitals in Seoul, Korea. Hospital A (813 beds) had 34 556 inpatients and 866 071 outpatients as of 2010, and Hospital B (405 beds) had 23 410 inpatients and 517 705 outpatients during the same time frame. The study hospitals were chosen because they are all university-affiliated large teaching hospitals and they are considered leading hospitals in Korea. In addition, they willingly permitted the conduct of the research in their facilities. The number of nurses in inpatient departments was 578 in Hospital A and 412 in Hospital B. The subjects of this study were nurses working at adult general medical–surgical inpatient care units in the study hospitals. Thus, nurses working in psychiatric, paediatric, obstetric/gynaecologic, and special inpatient units; post-anaesthesia rooms; neonatal and paediatric intensive care units; emergency departments; and nursing administration were excluded. The sample consisted of 394 nurses who worked in 23 inpatient care units in hospital A and 200 nurses working in corresponding care units in hospital B. The response rate was 99.5% (n = 591). After cases with missing values were excluded, data from 577 participants were included in the final dataset. This sample size was considered sufficient according to the recommendation of at least 10 cases per item. In addition, we examined item communalities and variable-to-factor ratios to ensure the adequacy of the sample size for factor analysis.[21-23] Item communalities of 0.40–0.70 were considered as low to moderate.[21-23] Each factor contained three or more variables; this indicated that the factors were moderately overdetermined.[21, 22] When communalities are low and factors are not highly overdetermined, a sample size much larger than typically recommended (e.g. a 20:1 subject-to-variable ratio) is required to produce stabler factor solutions and better approximate population parameters. The present sample size was also set so as to meet these recommendations and thereby ensure stability of the factor solution.
Data collection procedure
We collected data by means of self-rating questionnaires from 23 March to 12 April 2010. The questionnaire was distributed to nurses with a return envelope. The covering letter included explanations about the purpose of this study, the voluntary nature of participation in the study and data confidentiality. The anonymity of the respondents was also ensured. Returning the questionnaire was considered as consent to participate in the study. To encourage voluntary participation of nurses, we provided a US$5 gift for each participant as a token of our appreciation.
The study was conducted in two phases: development and evaluation of the scale. The development phase included preparation and review of measurement items by a panel of content experts, and the evaluation phase consisted of psychometric testing of the scale.
PCC was defined as knowledge, skills and attitudes related to patient-centred care. After a review of previous research, we adopted the list of knowledge, skills and attitudes regarding PCC developed by the QSEN faculty for the purpose of this study. The list was adopted because the definitions of and statements about knowledge, skills and attitudes related to PCC developed by the QSEN faculty were judged to reflect all relevant aspects of patient-centred care suggested by previous studies.[4-12] Originally, the QSEN faculty suggested six core competencies that apply to all registered nurses in order to facilitate changes in quality and safety education: patient-centred care, teamwork and collaboration, evidence-based practice, quality improvement, safety and informatics. The faculty also specified the knowledge, skills and attitudes required for each competency on the bases of previous research findings and reviews by experts in nursing education, with the goal of clarifying the meanings of the competencies. In addition, the knowledge, skills and attitudes were originally developed for general use among all registered nurses; therefore, they were applicable to this study population.
An expert panel of content experts composed of three nursing professors and eight members of the board of directors of the Korean Quality Improvement Nurses Society rated the relevance of the 41 initial draft items using a four-point Likert scale (1 = not relevant, 4 = highly relevant). We computed the content validity index (CVI) of each item. The CVI was calculated as the proportion of experts assigning an item a rating of either 3 (quite relevant) or 4 (very relevant). The items with CVI values of 0.70 or greater (n = 25) were retained. A questionnaire including the resulting 25 items was pretested. The pretest was conducted with two head nurses and two registered nurses, who served as the sample of respondents to verify the clarity and comprehensibility of the questions. On the basis of the feedback received, minor modifications were made to the wording of the questionnaire.
The resulting 25-item questionnaire was distributed to nurses in the study hospitals. They were asked to rate their competencies for each item on a five-point Likert scale (1 = minimal, 2 = below average, 3 = average, 4 = good, 5 = excellent). For validation purposes, a question that asked nurses to rate their overall performance in patient-centred care from 0 to 100, with 100 being the highest, was added.
Additional variables collected were years of nursing experience, job position, workplace, and demographic characteristics such as age, gender, marital status and educational level.
The general characteristics of participants were summarized using descriptive statistics. The overall competency score was calculated by averaging the values of the competency-related items. Internal consistency reliability was determined using Cronbach's alpha coefficients. The item–total correlation matrix was also examined. Items with average correlation coefficients of less than 0.3 were excluded.
Before conducting factor analysis, we examined the sampling adequacy. Bartlett's test of sphericity returned a statistically significant result (P < 0.001), and the Kaiser–Meyer–Olkin measure was 0.945. Therefore, the data were appropriate for factor analysis. Exploratory factor analysis (EFA) was performed using principal components analysis and the varimax rotation method to obtain a distinct and maximally interpretable solution. We used a cut-off value of 0.5 for factor loadings to improve interpretability. Items cross-loaded on more than two factors were excluded.
Next, we performed multitrait scaling analysis to confirm the factor structure on the basis of the EFA results.[27-29] The extent to which the items could be combined into the multi-item scales determined via factor analysis was examined using Pearson's correlation coefficients between items and scales. Evidence of item convergent validity was defined as an item–scale correlation of above 0.40 (corrected for overlap). Item discriminant validity was confirmed if the magnitude of the correlation of an item with its own scale was higher than that with other scales. Scaling success rate was defined as the proportion of cases within each scale that met the criterion of a significantly higher item correlation with its own scale than with any other scale (by more than two standard errors) in the comparison of item–own scale correlations with item–other scale correlations.[27, 28]
To examine whether the scales derived from the EFA measure distinct dimensions of PCC, the correlation coefficients between scales were compared with each scale's internal consistency. If the Cronbach's alpha value of a scale is higher than the scale's correlation with the other scales, it indicates that the scale scores represent distinct dimensions of the construct.
In addition, we measured concurrent (intermethod) validity to evaluate the extent to which different tests of the same construct are correlated with each other, on the basis of the suggestion in a previous study that a single-item global rating of performance is acceptable. Concurrent validity, which indicates the extent to which different tests of the same construct are correlated, was tested by calculating the Pearson correlation coefficient between total PCC scores on the one hand and the global ratings of overall performance in patient-centred care on the other.[30, 31] Data were analysed using the SAS statistical software package, version 9.2 (SAS Institute, Cary, NC, USA). The level of statistical significance was set at P < 0.05.
General characteristics of participants and PCC scores
The general characteristics of participants are shown in Table 1. Nurses were aged 32.7 (SD 7.3) years on average, and 83.9% (n = 484) were 40 years old or younger. Most (98.8%, n = 570) were female, and 65.9% (n = 380) were unmarried. Approximately half held a bachelor's degree (47.5%, n = 274). Nurses had an average of 8.3 (SD 7.3) years of experience in nursing, and 87.7% (n = 506) were staff nurses. Participants worked in various departments: 26.7% (n = 154) in medical care units, 30.0% (n = 173) in surgical care units, 10.4% (n = 60) in Oriental medicine units, 19.4% (n = 112) in intensive care units and 13.5% (n = 78) in operating rooms.
|Age (years)||≤ 30||234||40.6||3.47||0.43||3.41||3.52||28.5||< 0.0001|
|Marital status||Unmarried||380||65.9||3.51||0.45||3.46||3.55||−5.27||< 0.0001|
|Master's or higher||116||20.1||3.75||0.50||3.66||3.84|
|Years of experience in nursing||Less than 3||173||30.0||3.46||0.45||3.39||3.53||12.66||< 0.0001|
|3 to less than 5||84||14.6||3.51||0.40||3.43||3.60|
|5 to less than 10||162||28.1||3.57||0.46||3.50||3.65|
|10 or more||158||27.4||3.75||0.45||3.68||3.82|
|Job position||Manager||71||12.3||3.94||0.46||3.83||4.04||7.24||< 0.0001|
Nurses rated their competencies as above average, with a mean score of 3.58 (SD 0.46) on a five-point Likert scale. There were significant differences in PCC scores according to nurses' age, marital status, educational level, years of nursing experience, job position and hospital type in which they worked (P < 0.05) (Table 1).
Internal consistency reliability
The overall Cronbach's alpha coefficient for the 25 draft items was 0.94. Item–total correlations ranged from 0.54 to 0.68. Item analysis was conducted in order to delete items that were not highly correlated with other items without decreasing internal consistency, but no items were deleted because of weak correlations.
Initial EFA revealed a four-factor solution with eigenvalues of 1.0 or greater, which explained 61.8% of the variance. Items that had factor loadings of less than 0.5 and that cross-loaded onto more than two factors were deleted.[25, 26] Thus, the final PCC scale consisted of 17 items.
Principal components analysis using varimax rotation was reperformed on the final 17-item PCC scale. This analysis resulted in four factors with eigenvalues exceeding 1, which explained 64.6% of total variance. These four factors accounted for 43.7%, 8.2%, 6.7% and 5.9% of the variance, respectively. Factor 1 was composed of six items with factor loadings of 0.52–0.77. This factor was named ‘respecting patients’ perspectives'. Factor 2 consisted of five items with factor loadings of 0.66–0.75. It was named ‘promoting patient involvement in care processes’. Factor 3 was composed of three items with factor loadings of 0.67–0.81. This factor was named ‘providing for patient comfort’. Factor 4 consisted of three items with factor loadings of 0.69–0.83. It was named ‘advocating for patients’. The overall Cronbach's alpha coefficient of the 17-item PCC scale was 0.92, and those of the four subscales were 0.85, 0.81, 0.84 and 0.80, respectively (Table 2).
|No.||Item contents||Factor 1||Factor 2||Factor 3||Factor 4||Cronbach's α|
|Respecting patients' perspectives||0.85|
|1||Value seeing health-care situations through patients' eyes||0.77a||0.09||0.25||0.11|
|2.||Elicit patient values, preferences and needs as part of clinical interview, implementation of care plan, and evaluation of care||0.74a||0.19||0.19||0.18|
|3.||Integrate understanding of multiple dimensions of patient-centred care such as patient and family preferences||0.72a||0.25||0.08||0.17|
|4||Communicate patient values, preferences and needs to other health-care team members||0.67a||0.22||0.14||0.23|
|5||Provide patient-centred care with sensitivity and respect for the diversity of human experience||0.67a||0.26||0.26||0.11|
|6||Support patient-centred care for individuals and groups whose values differ from own||0.52a||0.31||0.15||0.10|
|Promoting patient involvement in care processes||0.81|
|7||Examine barriers to active involvement of patients in their care processes||0.18||0.75a||0.05||0.17|
|8||Assess level of patient's decisional conflict and provide access to resources||0.23||0.75a||0.12||0.17|
|9||Describe strategies to empower patients or families in all aspects of the care process||0.18||0.71a||0.19||0.15|
|10||Engage patients or designated surrogates in active partnerships that promote health, safety and well-being, and self-care management||0.21||0.70a||0.14||0.18|
|11||Respect patient preferences for degree of active engagement in care process||0.30||0.66a||0.28||0.11|
|Providing for patient comfort||0.84|
|12||Assess presence and extent of pain and suffering||0.27||0.06||0.81a||0.21|
|13||Assess levels of physical and emotional comfort||0.22||0.24||0.78a||0.18|
|14||Elicit expectations of patient and family for relief of pain, discomfort or suffering||0.28||0.38||0.67a||0.12|
|Advocating for patients||0.80|
|15||Facilitate informed patient consent for care||0.14||0.09||0.19||0.83a|
|16||Communicate care provided and needed at each transition in care||0.25||0.31||0.19||0.71a|
|17||Participate in building consensus or resolving conflict in the context of patient care||0.29||0.40||0.12||0.69a|
|Variance explained by each factor||3.44a||3.34a||2.16a||2.04a|
Convergent and discriminant validity
The results of EFA with regard to the factor structure were then examined using multitrait scaling analysis. Pearson correlation coefficients, corrected for overlap between each item and its own scale, were above 0.40 for all items. The correlation of each item with its own scale was higher than that with the other scales for all items. Of the 51 comparisons between item–own scale correlations and item–other scale correlations, 49 met the criterion of being correlated with their own scales more strongly than with other scales by more than two standard errors. Thus, the overall scaling success rate was 96.1% (Table 3).
|Item convergent validity||Item discriminant validity||Scaling success rate, % (n)|
|Scale 1||Scale 2||Scale 3||Scale 4|
|Scale 1||6/6||94 (17/18)|
|Scale 2||5/5||100 (15/15)|
|Scale 3||3/3||89 (8/9)|
|Scale 4||3/3||100 (9/9)|
The correlations between the four PCC subscales ranged from 0.52 to 0.61. Each subscale's correlation coefficients with the other subscales were lower than the scale's internal consistency value. This result indicated that the four subscales represented distinct aspects of PCC.
The mean self-rating score of overall performance in patient-centred care was 76.1 (SD 11.9). There was a significant correlation between total PCC scores and the overall self-ratings of patient-centred care performance (r = 0.60, P < 0.001).
Patient-centred care is a core component of high-quality care and patient safety. Understanding nurses' competency for patient-centred care is necessary in order to facilitate the transition toward patient-centred care in clinical practice. The findings of this study demonstrated that the PCC scale has good psychometric properties to assess nurses' competency for patient-centred care. Multidimensional properties reflected a conceptual framework for the scope of patient-centred care. The four dimensions that emerged—respecting patients' perspectives, promoting patient involvement in care processes, providing for patient comfort and advocating for patients—were consistent with common attributes of patient-centred care suggested by previous research.[4-12]
The homogeneity of items is a major issue in assessing the psychometric properties of an instrument. There were no items with correlation coefficients of less than 0.3 with the total in the item–total correlation matrix. Internal consistency reliability estimates were calculated for all subscales as well as the overall scale. They all met the desirable criterion of Cronbach's alpha being ≥0.80. These results indicated that the PCC scale has excellent internal consistency.
Maximizing the explained variance and the parsimony of the scale was considered in factor extraction. We examined factors with eigenvalues greater than 1.0 using principal components factor extraction and the varimax rotation method in order to achieve the highest possible level of purity in the scale's factors. Another criterion for factor extraction was that the proportion of overall variance explained by a factor should be at least 5%. The factors that emerged in this study met both of these criteria. This supports the adequacy of the factors extracted. Subsequently, we used multitrait scaling analysis to confirm the factor structure. Convergent and discriminant validity were satisfactory for all four subscales. All subscales demonstrated high scaling success rates.[28, 29] Two items failed to reach the optimal criterion of having an item–own scale correlation coefficient more than two standard errors higher than its item–other scale correlation coefficients; nevertheless, these items had stronger correlations with their own scales than with others. Thus, multitrait scaling analysis supported the fit of the four-factor model identified by EFA. In addition, the magnitude of the interscale correlations was moderate, thus showing that the subscales—though related—represented distinct dimensions of PCC. Our results provide stronger support for the construct validity of the PCC scale, which contributes to the stability of the tool. Furthermore, nurses' PCC scores were significantly correlated with their global self-rating scores regarding overall patient-centred care performance.
The findings of this study provide evidence that the PCC scale is a reliable, valid instrument. To our knowledge, this is the first study to provide a survey instrument that measures competency with respect to patient-centred care in hospital nurses. Patient-centred care is one of the core values in health care, and it requires actions through a substantial partnership between patients and care providers. Nurses need to actively incorporate patient-centred care into practice by involving patients in care decisions on the basis of respect for patients' values, preferences and needs. Nurses also need to provide care in ways that result in better patient outcomes and advocate for patients in health-care systems. Competency assessments based on the PCC scale will provide a fundamental basis for future studies intended to improve nurses' competency with respect to patient-centred care. Furthermore, understanding the current status of patient-centred care will help to identify important elements to be enhanced in nursing education with the goal of providing patient-centred, high-quality care in clinical nursing practice.
The PCC scale is different from existing tools to measure patient safety and quality of care. For instance, several studies have suggested that patient-centredness is a key subdimension of health-care quality.[32, 33] In addition, patient-centredness has been considered as one aspect of hospital quality and patient safety competencies. Moreover, patient safety activities focus on the prevention of errors and adverse effects to patients associated with health care. A systematic review on tools to assess patient safety competencies suggested that the core competency domains were contribution to patient safety culture, teamwork, effective communication, risk management, optimization of human and environmental factors, and adverse event management. Therefore, although patient-centred care is closely linked to quality of care and patient safety activities, there are differences in the scope and focus areas of the tools that measure those constructs.
On the other hand, this study had several limitations. First, this study was conducted in only two teaching hospitals and thus requires replication in other clinical settings to ensure that the structure of the scale is stable across settings. Second, test–retest reliability was not calculated in this study because we measured the participants at only one time point. We suggest further validation studies in which the instrument is used and refined. We also recommend that confirmatory factor analysis be undertaken with samples composed of different nurses in hospital-care settings in order to support the generalizability of the PCC scale. In addition, this study did not consider the perspectives of patients who might have different viewpoints on patient-centred care. Thus, considering the perspectives of both care providers and patients on patient-centred care, a multitrait–multimethod matrix approach might be useful for confirmation of the validity of the scale and its subdimensions.
Nonetheless, this study has the following strengths: We selected nursing units for inpatient care with consideration of the representativeness of the sample, and then we surveyed all the nurses working in the care units. Moreover, we achieved a high response rate. These strengths contributed to minimizing the risk of selection bias in our results. In addition, the sample size was sufficient to achieve stability in the parameter estimates, exceeding the minimum recommended ratio of 10:1. Furthermore, there were few items with missing values. This seems to indicate that the tool is easily understood and the items are relevant to care practice.
In conclusion, the 17-item PCC scale demonstrated excellent reliability and validity, making it feasible for clinical use. The PCC scale developed in this study has several potential uses. For example, it can be used to assess nurses' competency for patient-centred care in hospital settings. In addition, the preliminary results in this study showed variations in total PCC scores according to individual and organizational characteristics. The application of this instrument will help to determine the factors related to patient-centred care. Furthermore, it can be used for evaluations of educational interventions intended to improve PCC. We also suggest future research using this instrument in order to better understand the relationship between PCC and patient outcomes such as patient satisfaction.
This research was supported by the National Research Foundation of Korea; grant funded by the Korean Government (NRF-2009-327-H00039).
- 1Institute of Medicine. Health Professions Education: A Bride to Quality. Washington, DC, USA: National Academies Press, 2003.
- 10Patient-centered care model demands better physician–patient communication. JAMA: The Journal of the American Medical Association 2012; 307: 441–442..
- 20Statistics and Data Analysis for Nursing Research, 2nd edn. Upper Saddle River, NJ, USA: Pearson Education, 2010..
- 23Best practices in exploratory factor analysis: four recommendations for getting the most from your analysis. Practical Assessment Research & Evaluation 2005; 10: 1–9., .
- 28Psychometric evaluation of the European Organization for Research and Treatment of Cancer In-Patient Satisfaction with Care Questionnaire (‘Sinhala’ version) for use in a South-Asian setting. International Journal for Quality in Health Care 2008; 20: 221–226., , .