Skip to main content
NCBI home page
Journal of Research in Nursing logoLink to Journal of Research in Nursing
. 2023 Jul 20;28(5):338–351. doi: 10.1177/17449871231175788

Cultural adaptation of the revised Basel Instrument for Rationing of Care to the Turkish context: a study of validity and reliability

Havva Arslan Yürümezoğlu 1,✉, Maria Schubert 2, Emine Sarıoğlu 3, Gülseren Kocaman 4
PMCID: PMC10599310  PMID: 37885955

Abstract

Background:

Studies have shown that nurses do not complete one or more necessary nursing care elements in their last shift due to lack of time. The Basel Extent of Rationing Nursing Care (BERNCA) instrument is one of the most used scales to measure the rationing of nursing care.

Aim:

This study is aimed to culturally adapt the revised BERNCA (BERNCA-R) instrument to the Turkish language and to test its validity and reliability.

Methods:

A cross-sectional and methodological study was used. The instrument was adapted in three stages: translation and adaptation, content validity testing and validity and reliability. Data were obtained from 350 nurses working in two public acute care hospitals in Turkey, between September 2019 and January 2020. The descriptive statistics, content validity index, exploratory and confirmatory factor analyses, Cronbach’s α, Guttman split-half and inter-item reliability analyses were performed for the analysis of the data.

Results:

The Turkish version of the BERNCA-R instrument with a 27-item and three-subscales (monitoring, daily care and needs and psychosocial care) structure was found to have acceptable and good fit indices.

Conclusion:

The Turkish version of the BERNCA-R instrument is valid and reliable tool to measure rationing of nursing care.

Keywords: factor analysis, instrument adaptation, nursing, rationing care, reliability, validity

Introduction

Due to a global nursing shortage, healthcare institutions worldwide have an insufficient number of nurses on their staff. It is estimated that currently there are only half as many nurses available as are needed, and that this shortage will worsen in Eastern Mediterranean regions by the 2030s (World Health Organization, 2016). Studies have demonstrated that low nurse–patient ratios and a poor working environment for nurses affect the quality of patient care and outcomes (Aiken et al., 2012; Ausserhofer et al., 2013; Bruyneel et al., 2015; Recio-Saucedo et al., 2018). One of the most important negative consequences of this nursing shortage is rationing of nursing care, which, as the evidence also shows, has a negative effect on patient outcomes (Ausserhofer et al., 2013; Schubert et al., 2008, 2009). Studies have shown that almost all nurses surveyed were unable to complete one or more necessary nursing care tasks in their last shift due to time constraints (Aiken et al., 2018; Ausserhofer et al., 2013; Ball et al., 2012, 2014, 2018).

In the literature, researchers use various concepts to describe necessary nursing care that is not provided, its influencing factors and related outcomes, such as unfinished care, care left undone, missed nursing care and implicit rationing of nursing care. Although these concepts have different definitions, their common characteristic is that they all refer to nurses being unable to perform necessary patient care due to time and resource constraints (Bail and Grealish, 2016).

Implicit rationing of nursing care has been described as the lack or inability to take necessary precautions for patients due to insufficient nursing resources (staffing, skill mix and time) (Schubert et al., 2005). This definition refers to a set of nursing tasks or treatment measures based on clinical consensus that are important for achieving targeted outcomes for patients. The tasks identified may be influenced by the professional standards, education levels and cultural characteristics of the country (Schubert et al., 2007).

The Rationing Care in Switzerland (RICH Nursing Study) project was the first work in which the concept of implicit rationing of nursing care was used. The RICH Nursing project was based on the International Hospital Outcomes Study and the nursing care left undone concept (Schubert et al., 2005). Results of the RICH Nursing Study underlines that in healthcare units with insufficient resources, nurses have to minimise or skip certain tasks, thereby increasing the risk of negative patient outcomes (Schubert et al. 2008). Kalisch (2006) then coined the term ‘missed nursing care’, which is defined as an error of omission. The relevant antecedents of missed nursing care are described as labour resources and material resources, as well as relationship and communication factors associated with the provision of care. Unless nurses are specifically asked, they avoid communicating this shortcoming; making it a problem experienced every day in patient care but not openly discussed (Kalisch et al., 2009).

In the literature, mainly in developed countries, the effects of implicit rationing of nursing care on patient care quality and safety have been identified, monitored and used as a policy-making instrument for accurate nurse planning. In Turkey, negative working conditions, high patient-to-nurse ratios, and overtime are generally held to have led to the problem of implicit rationing of nursing care. In order to be able to measure implicit rationing of nursing care in a valid and reliable way, a suitable instrument was selected based on a literature review. Due to its content and ease of use, the Basel Extent of Rationing of Nursing Care–revised (BERNCA-R) was determined to be most appropriate for measuring this phenomenon in Turkey. The BERNCA-R measures implicit rationing of nursing care that occurs when nurses are unable to provide the necessary care to patients due to lack of time and resources. This measurement process also includes clinical decision-making and nurses’ prioritisation skills. The decision-making process of nurses is shaped by the organisational culture, characteristics of the nurse working environment and philosophy of care (Schubert et al., 2007). The BERNCA-R is an indicator and a validated instrument to measure the effects of nursing work environment on patient outcomes (Schubert et al., 2008). The BERNCA-R has been used in the past in Switzerland and in several other countries like Poland (Uchmanowicz et al., 2019). The adaptation of BERNCA-R to Turkish will create an opportunity to compare the results of different countries.

Aim

The aim of this study was to adapt the BERNCA-R instrument to the Turkish culture and to test the validity and reliability of this Turkish version of the BERNCA-R.

Hypotheses

To this end, the hypotheses of the study were determined as: (H1) there is a strong positive correlation among the items and the overall construct of implicit rationing of nursing care, and (H2) there is a negative and significant correlation between the BERNCA-R Turkish version and the NWI subscale (Staffing and Resource Adequacy).

Methods

Design, setting and sample

For the evaluation of the validity and reliability of the cultural-adapted finalised Turkish version of the BERNCA-R, a cross-sectional design was used. This portion of the study was conducted in two public hospitals located in the third largest city of Turkey, with 567 and 702 beds, respectively. The number of nurses working in the two public hospitals was 478 and 623, respectively. A total of 1101 nurses were recruited from two hospitals by convenience sampling. Of the 1101 nurses, 842 met the inclusion criteria and were included, and of this, 358 nurses participated in the study (response rate: 42.5%). The sample consisted of 358 nurses from all educational levels who worked in direct patient care for at least three months in the two study hospitals. Nurses who were in management positions, or were not directly involved in patient care, or had not worked in the same healthcare institution for at least three months, were excluded from the study. For factor analysis, it is suggested that at least five to ten participants be included per item (Pett et al., 2003). Therefore, it was necessary to include 330 nurses in this cross-sectional study, which was conducted to test the validity and reliability of the Turkish version of the BERNCA-R, which has 33 items. A researcher distributed the instrument to all nurses who could be reached, to ensure the needed sample size in case there should be missing or empty survey data. Eight nurses who did not meet the inclusion criteria and/or had missing survey data were excluded. The analyses were then conducted including 350 nurses and their data.

Data collection

The survey data were collected between September 2019 and January 2020. To ensure a high response rate, a member of the research team (ES) met with all of the nurses in person to explain the purpose of the study and to answer their questions regarding the instrument items. The surveys were then given to the nurses in sealed envelopes. After the nurses completed the surveys, they were placed in boxes in the clinics and were then collected by the researcher.

The instrument used includes the following: sociodemographic information, the Staffing and Resource Adequacy subscale (four items) of the Practice Environment Scale of the Nursing Work Index (PES-NWI) and the Turkish version of BERNCA-R. The sociodemographic portion comprises eight questions regarding age, gender, education level, tenure in the profession, tenure in the organisation, current worked unit, working shift and number of patients on the last shift. The PES-NWI, which measures quality of the nurse work environment, was used as a criterion validity in the first BERNCA development study (Schubert et al., 2007). In this study, the Staffing and Resource Adequacy subscale of PES-NWI was used to test the validity of the BERNCA-R Turkish version. The subscale refers to having adequate number of staff and support resources to provide safety and quality patient care and spend enough time for patients. The PES-NWI developed by Lake (2002) and adapted to Turkish by Turkmen et al. (2011) was used to determine nurses’ perceptions of their work environment. The Staffing and Resources Adequacy subscale includes four items and uses a 4-point Likert scale (4 = strongly disagree to 1 = strongly agree). In the study of Turkmen et al. (2011), the Cronbach’s α of the subscale was found to be 0.80.

The BERNCA instrument was developed by Schubert et al. (2007) and later revised by the same authors (Schubert et al. 2013). The original BERNCA, with 20 items, has a unidimensional structure (Schubert et al., 2007). The BERNCA-R instrument, revised by Schubert et al. (2013) and adapted for use in this study, contains 32 items. Using a 5-point Likert scale (0 = not required, 1 = never, 2 = rarely, 3 = sometimes and 4 = often), respondents are asked to rate how frequently they could not perform the necessary nursing tasks in their last seven working days due to inadequate time, staffing levels and/or skill mix. The validity and reliability of the BERNCA instrument were confirmed in several national and international research projects such as Rationing of Nursing Care in Switzerland (RICH Nursing Study) study and the Registered Nurses Forecasting study. There are no published results available regarding the dimensionality of the revised version of the BERNCA-R with 32 items. Depending on the version, the reported Cronbach’s α varied between 0.93 (Schubert et al., 2007) and 0.96 (Uchmanowicz et al., 2019). To calculate the levels of implicit rationing of nursing care, the scores were summed and divided by the number of respective items (32) to calculate the total score and mean (with the range of total scores being 0–128). It is interpreted that, as the score increases, the frequency of rationing of nursing care increases.

Cultural adaptation

The BERNCA-R was adapted to the Turkish culture in three stages: translation and adaptation, content validity testing and evaluation of validity and reliability.

Stage 1: translation and adaptation

The BERNCA-R was first forward translated from English to Turkish by three independent translators in order to reflect the original meaning of the items. The first translator had a good grasp of the instrument structure and the terminology used in the field of nursing. The second and third translators were nurses and fluent in Turkish and English. After the English to Turkish translation, the instrument was reviewed by researchers who were native speakers of Turkish. The Turkish version of the BERNCA-R was then back translated into English by two translators with nursing experience in both Turkish and English-speaking countries. The Turkish BERNCA-R instrument was evaluated in terms of conceptual consistency, similarity in meaning and content validity compared to the original version.

In addition to the translation, the following adaptations were made as part of the cultural adaptation of the BERNCA-R into the Turkish language. An item related to monitoring pain management was added to the Turkish version of the instrument, taking the total number of items to 33. Although pain is defined as the fifth vital sign, nurses less effective pain assessment and monitoring when compared other tasks. Therefore, it has been decided the pain management was necessary to measure rationing nursing care. In addition, the answer category ‘family member did’ was added. Turkey is a country with a predominantly collectivist culture, in comparison to Western countries. This characteristic also influences relationships between patients, family members and health professionals. Therefore, in Turkey, family members play an important role in providing patients with basic activities of daily living (ADLs) and daily needs in acute care hospitals. Patients prefer the support of their families in activities such as eating, basic hygiene and bathing and family members want to stay with the patient in the hospital to support these activities.

Stage 2: content validity

Two approaches were used to evaluate content validity. The first was an expert panel who reviewed the content. The second was by calculating the content validity index (CVI). The literature recommends having at least three experts perform this evaluation (Polit et al., 2007). However, considering the different educational levels of nurses and the varying characteristics of the environment in Turkish acute care hospitals, it was decided to include a larger number of experts in this study. Twenty experts working in various hospitals were invited to participate in the expert panel, with 17 experts accepting and subsequently attending the panel. The experts participating in the panels were experienced nurses with bachelor degrees or masters degrees and who provided direct patient care in clinics. For the evaluation of the content validity of the Turkish items of the instrument, three separate panels with experts were held.

The experts evaluated the Turkish version of the BERNCA-R with its 33 items in terms of the content to be measured. The 17 experts participating confirmed the content validity of the Turkish version of the BERNCA-R. However, some of the experts suggested removing some of the items (items 9, 13, 14, 20, 21 and 24) because they did not consider them included in the responsibility of nurses in Turkish acute care hospitals. The inclusion of these items in the nursing curriculum and nursing job descriptions was discussed by the expert panel, and a consensus was reached to keep the items in the instrument during the data collection.

The CVI was determined by means of the Davis technique (Polit et al., 2007). For acceptable content validity of an instrument, the ratio of experts giving the items 3 and 4 points to the total percentage of experts should be 80% or above (Polit et al., 2006). It is suggested that the CVI of each item (I-CVI) should be at least 0.78 when there are nine or more expert ratings and 0.83 in the case of six expert ratings (Polit et al., 2007). For the calculation of the CVI, the participating experts were asked to score each BERNCA-R item from 1 (not relevant) to 4 (highly relevant). The researchers reviewed the scores given to each item by the experts and revised those items rated with only 1 or 2 points.

Stage 3: evaluation of validity and reliability

Validity and reliability testing

Validity testing

For the Turkish version of BERNCA-R, the content validity, construct validity and criterion (concurrent) validity were established. Explanatory factor analysis (EFA) and confirmatory factor analysis (CFA) were conducted to evaluate the construct validity. The EFA was used to explore the dimensionality and number of factors. The CFA provided insight into how well the hypothesised factor structure fits with the data. Correlation analysis was carried out to explore the concurrent validity of the correlation between the Turkish version of the BERNCA-R and the PES-NWI subscale ‘adequacy of resources’. Reliability tests were conducted by means of Cronbach’s α coefficient, split-half method and inter-item correlation. The data analyses were conducted with SPSS software v. 24.0 (IBM Corp., Armonk, NY., USA), and the LISREL 8.7 (Computer Software, Lincolnwood) statistical package program was used for the CFA.

Construct validity

The internal structure of the Turkish BERNCA-R and the underlying dimension were explored using EFA. For the justification of the appropriateness of the 33 items of the Turkish BERNCA-R for factor analysis, the Kaiser–Meyer–Olkin (KMO) −0.50 and above – and the Bartlett’s test of sphericity – significant Bartlett’s test (p < 0.05) was used (Pett et al., 2003). In order to identify the dimensionality of the Turkish BERNCA-R version and the related items, the items loading on each component (factor), a Principal Component Analysis and an initial analysis with Varimax rotation were conducted, and several factor models were tested. To ensure that the items contribute adequately to a given factor, the factor loading categorisation 0.30 = minimal, ±0.40 = important and ±0.50 = practically significant developed by Hair et al. (2014) was employed, and items with a factor load of 0.40 and above were accepted.

CFA was carried out to test the structure of the Turkish BERNCA-R. Several criteria were used to test the fit of the model with the data. These were ratio of chi square to degrees of freedom (χ2/df), the root mean square error of approximation (RMSEA), the normed fit index (NFI), the comparative fit index (CFI) and the standardised root mean square residual (SRMR). Threshold values indicating model fit were <2 for χ2/df, <0.08 for RMSEA, >0.90 for NFI and CFI and <0.08 for SRMR.

Criterion (concurrent) validity

In order to evaluate the criterion validity of the Turkish version of the BERNCA-R and PES-NWI-R subscale, ‘Staffing and Resource Adequacy’ (subconcept adequacy of resources and skill mix of the nurse work environment) was used. This subscale refers to the availability of an adequate number of staff and support resources to ensure safe and adequate quality of patient care and to enable sufficient time for patients. The PES-NWI, developed by Lake (2002) and adapted to Turkish by Turkmen et al. (2011), measures nurses’ perceptions of the quality of their work environment. The Staffing and Resources Adequacy subscale includes four items and uses a four-point Likert scale (4 = strongly disagree to 1 = strongly agree). In both the original studies (Lake, 2002 and Turkmen et al. (2011), the Cronbach’s α of the subscale was found to be 0.80.

Reliability testing

To evaluate of internal consistency and homogeneity, Cronbach’s α, split-half method and inter-item correlation were calculated. Cronbach’s α is the most important and widely used measurement technic for the internal consistency. The reliability of the Turkish version of the BERNCA-R was examined by means of the Cronbach’s α for 27 items and three dimensions of the instrument. The Cronbach’s α reliability coefficient 0.70 or higher was considered acceptable. Following the split-half method, the items of the BERNCA-R were divided into two parts, and a Cronbach’s α coefficient was calculated to establish the correlation between the two halves. Inter-item correlation was calculated to evaluate whether there were items in the BERNCA-R instrument that measured the same construct.

Results

Table 1 provides descriptive data on the demographic and work characteristics of the nurses who participated in the study. The mean age of the nurses was 35.9 years old, the majority were female and two-thirds had a bachelor’s degree. On average, the participants had worked 14.5 (standard deviation (SD): 8.6) years in the nursing profession, and 7.7 (SD: 7.2) years in their current hospital. Most of the nurses worked in surgical units in both day and night shifts. It was reported that the nurses care for 7.5 patients on average during the day shift, and 17.5 patients during the night shift. Table 2 shows the descriptive results the BERNCA-R.

Table 1.

Demographic and work characteristics of the nurses (n = 350).

Characteristics Mean (SD), %
(n = 350)
Age, mean (SD) 35.9 (7.9)
Gender, n %
 Female 95.1
 Male 4.9
Education degree, n %
 Vocational school 8.6
 Associate 12.6
 Bachelor 65.4
 Postgraduate 13.4
Duration of professional experience (years), mean (SD), (min–max) 14.5 (8.6), (0.3–43)
Duration of experience in the current hospital (years), mean (SD), (min–max) 7.7 (7.2), (0.3–43)
Units at which nurses worked, n %
 Medical 44.3
 Surgical 55.7
Shift, n %
 Day 7.7
 Night 3.4
 Day–night 88.9
Number of patients cared for during the last day shift, mean (SD), (min–max) 7.5 (3), (3–16)
Number of patients cared for during the last night shift, mean (SD), (min–max) 17.5 (6.1), (5–40)
Open in a new tab

SD: standard deviation.

Table 2.

The descriptive results of Turkish BERNCA-R instrument (n = 350).

Items of Turkish BERNCA Not required
(%)		 Never
(%)		 Rarely
(%)		 Sometimes
(%)		 Often
(%)		 Family member did
(%)		 Missing value
(%)
1. Sponge bath 20.6 2.9 3.4 6.0 14.6 52.5 0
2. Partial sponge bath 18.0 2.3 5.7 7.4 10.9 55.7 0
3. Skin care 11.7 7.7 6.3 8.3 9.1 56.0 0.9
4. Oral hygiene 13.7 10.3 10 9.4 6.0 50.6 0
5. Dental hygiene 16.3 6.3 7.7 6.3 9.4 54.0 0
6. Assist food intake 13.1 13.7 8.6 8.9 5.7 49.7 0.3
7. Mobilisation 10.9 20.0 15.4 14.9 7.4 30.3 1.1
8. Change of the position 11 18.3 18.6 16.6 6.0 28.6 0.9
9. Change of the bed linen 25.5 27.6 14.3 10.3 1.4 20.3 0.6
10. Emotional and psychological support 6.3 20.3 30.0 28.0 15.4 0 0
11. Necessary conversation 3.1 23.1 28.9 32.6 9.7 2.3 0.3
12. Information about therapies 6.2 27.4 26.6 28.6 10.6 0 0.6
13. Continence training (diapers) 39.8 27.3 17.4 10.9 4.3 0 0.3
14. Continence training (insert catheter) 39.7 28.6 18.0 10.0 3.4 0 0.3
15. Activating or rehabilitating care 17.4 28.3 22.3 17.1 14.6 0 0.3
16. Education and training 6.9 43.3 24.9 18.9 5.4 0 0.6
17. Preparation for discharge 2.3 45.1 28.3 17.4 6.6 0 0.3
18. Monitoring patients as described by physician 1.4 45.7 30.3 18.6 3.7 0 0.3
19. Monitoring patients as The nurse felt necessary 3.7 36.9 29.4 26.3 3.7 0 0
20. Monitoring of confused patients and use of restrains 30.5 24.3 23.7 16.6 4.3 0 0.6
21. Monitoring of confused patients and use of sedatives 37.7 31.4 16.6 10.3 3.4 0 0.6
22. Delay in measure because of a physician delay 12.8 38.0 26.9 17.7 4.6 0 0
23. Administration of medication, infusions 2.2 55.4 28.6 12.9 0.9 0 0
24. Change of wound dressings 32.5 34.0 20.6 10.0 2.3 0 0.6
25. Preparation for test and therapies 8.9 42.0 24.3 17.7 7.1 0 0
26. Keep patient waiting who rung 2.0 22.8 24.0 30.9 20.0 0 0.3
27. Adequate hand hygiene 1.2 49.6 26.6 17.7 4.9 0 0
28. Necessary disinfection measures 1.2 53.1 29.4 12.9 3.1 0 0.3
29. Studying care plans 2.3 33.7 33.7 21.4 8.9 0 0
30. Assessment of newly admitted patient 2.3 42.3 31.7 20.3 3.4 0 0
31. Set up care plans 1.1 43.7 30.9 20.0 4.3 0 0
32. Documentation and evaluation of the care 0.6 32.8 33.1 26.6 6.9 0 0
33. Pain assessment and evaluation 1.2 42.0 32.0 19.4 5.1 0 0.3
Open in a new tab

BERNCA: Basel Extent of Rationing Nursing Care; BERNCA-R: Basel Extent of Rationing Nursing Care–revised.

Validity and reliability testing

Content validity

Based on 17 expert ratings, the I-CVI ratings range between 0.82 and 1.00 for the Turkish version of BERNCA-R.

Construct validity

For the construct validity, the KMO test scores at 0.89 and the statistically significant of the Bartlett test (χ2 = 5271.317; p < 0.001). The EFA results of the 33-item and the 27-item instruments are given in Table 3. In the 27-item instrument, those items with factor loadings below 0.40 were excluded (9, 13, 14, 20 and 21). These items were also those that had not been considered to be part of nurses’ roles during the expert discussions. EFA is defined as an analysis in which conceptual and statistical assumptions should be considered together, but often the former (conceptual assumptions) take precedence (Hair et al., 2014). Therefore, it was decided to remove the items, which both contained roles and tasks that were not expected of nurses in their practice in Turkey according to the expert panel discussions and had low factor loading in the EFA analysis. To further explore the construct validity, a comparison was made between the results of EFA conducted with the 33-item instrument and the 3, 4 and 5-factor models that were obtained and those conducted with the 27-item instrument and the 3-factor model. The experts suggested removing item 24, ‘Change of wound dressings’, from the instrument as this task is carried out by medical residents in Turkish hospitals. Therefore, in addition to the items described above, item 24 was removed from the adapted instrument. After testing different factor models, it was decided that the 27-item and 3-factor model best explained the variance. The three factors included in the final model were assessment, planning and monitoring (subscale 1), daily care and needs (subscale 2) and psychosocial care (subscale 3).

Table 3.

Exploratory factor analysis and factor loadings of two different factor models for the Turkish BERNCA-R instrument.

Model 1* Model 2**
Factor 1
Assessment, planning and monitoring		 Factor 2
Daily care and needs		 Factor 3
Psychosocial care		 Factor 1
Assessment, planning and monitoring		 Factor 2
Daily care and needs		 Factor 3
Psychosocial care
Cronbach’s α of the subinstruments 0.91 0.88 0.80 0.91 0.90 0.80
Items of Turkish BERNCA
1. Sponge bath 0.047 0.769 −0.090 0.075 0.795 −0.067
2. Partial sponge bath 0.079 0.842 −0.051 0.105 0.861 −0.028
3. Skin care 0.055 0.878 0.009 0.071 0.882 0.032
4. Oral hygiene 0.014 0.846 0.115 0.037 0.851 0.131
5. Dental hygiene −0.014 0.859 0.087 0.013 0.872 0.107
6. Assist food intake 0.088 0.664 0.135 0.106 0.662 0.151
7. Mobilisation 0.068 0.530 0.418 0.084 0.527 0.438
8. Change of the position 0.175 0.476 0.366 0.184 0.463 0.383
9. Change of the bed linen 0.242 0.341 0.117 – – –
10. Emotional and psychological support 0.176 0.059 0.727 0.183 0.052 0.737
11. Necessary conversation 0.189 0.029 0.779 0.192 0.017 0.784
12. Information about therapies 0.228 0.068 0.774 0.234 0.049 0.773
13. Continence training (diapers) 0.347 0.349 0.238 – – –
14. Continence training (insert catheter) 0.379 0.377 0.211 – – –
15. Activating or rehabilitating care 0.233 0.257 0.586 0.238 0.238 0.582
16. Education and training 0.498 0.097 0.415 0.501 0.085 0.423
17. Preparation for discharge 0.513 0.162 0.368 0.517 0.145 0.371
18. Monitoring patients as described by physician 0.655 0.164 0.238 0.657 0.145 0.240
19. Monitoring patients as the nurse felt necessary 0.637 0.120 0.263 0.642 0.105 0.257
20. Monitoring of confused patients and use of restrains 0.385 0.346 0.253 – – –
21. Monitoring of confused patients and use of sedatives 0.377 0.360 0.162 – – –
22. Delay in measure because of a physician delay 0.516 0.070 0.162 0.514 0.049 0.159
23. Administration of medication, infusions 0.579 0.015 0.071 0.591 0.027 0.079
24. Change of wound dressings 0.470 0.265 0.076 – – –
25. Preparation for test and therapies 0.565 0.152 0.221 0.567 0.140 0.232
26. Keep patient waiting who rung 0.545 –0.009 0.234 0.569 0.006 0.232
27. Adequate hand hygiene 0.703 0.071 0.079 0.709 0.066 0.087
28. Necessary disinfection measures 0.740 0.051 0.051 0.738 0.036 0.059
29. Studying care plans 0.706 0.018 0.151 0.718 0.010 0.147
30. Assessment of newly admitted patient 0.746 0.127 –0.005 0.755 0.118 –0.004
31. Set up care plans 0.720 0.117 0.075 0.730 0.108 0.075
32. Documentation and evaluation of the care 0.746 0.074 0.113 0.750 0.061 0.110
33. Pain assessment and evaluation 0.693 –0.011 0.170 0.704 –0.010 0.171
Open in a new tab
*

Model 1: Three factors with 33 items.

**

Model 2: Three factors with 27 items, items with factor loadings of >0.40 are shown in bold.

Confirmatory factor analysis was conducted with 27 items and the three-subscale structure (Figure 1). Some of the CFA fit indices (RMSEA, NFI, CFI and SRMR) remained below acceptable limits, which led to an examination of the modification suggestions of the LISREL 8.7 statistics program. Three modifications were made based on the results of the CAF analysis and considered theoretically suitable for the structure of the instrument. In order to improve the fit indices, three modifications were proposed for items 27 and 28, items 7 and 8 and items 31 and 32. In the 27-item CFA model for the Turkish version of BERNCA-R, acceptable and good fit indices were obtained after the proposed modifications were implemented (χ2/df = 1.83, RMSEA = 0.069, NFI = 0.91, CFI = 0.96 and SRMR = 0.07) (Kline, 2016).

Figure 1.

Figure 1.

Open in a new tab

The path diagram of the Turkish BERNCA-R instrument.

BERNCA-R: Basel Extent of Rationing Nursing Care–revised.

Criterion (concurrent) validity

There was a weak negative, statistically significant correlation between the Turkish BERNCA-R and the Staffing and Resource Adequacy subscale (r = −0.218, p < 0.001).

Reliability results

The Cronbach’s α value of the overall instrument with 27 items was 0.91, and the Cronbach’s α values of the three subscales were 0.91, 0.90 and 0.80, respectively. The Guttman split-half reliability coefficient for the three subscales was 0.85, 0.79 and 0.78. The inter-item correlation for the three subscales were 0.41 (0.26–0.76), 0.52 (0.25–0.83) and 0.50 (0.36–0.62).

Discussion

The aim of this study was to translate and culturally adapt the BERNCA-R questionnaire, developed by Schubert et al. (2013) to measure implicit rationing of nursing care, for use in Turkey and to test its validity and reliability. The results of this study indicate that the Turkish version of the BERNCA-R has good validity and reliability in measuring implicit rationing of nursing care in Turkey.

The content validity of the Turkish version of the BERNCA-R was confirmed by the experts. The experts suggested removing items 9, 13, 14, 20, 21 and 24. Although these items are included in nursing regulations and nursing education curriculum in Turkey as the responsibility of nurses, in daily routine in clinic, these tasks are often accepted as the responsibility of the assistant staff (items 9, 13 and 14) and the residents (item 24), or in the decision-making area of the physicians (items 20 and 21). In Turkey, the working environment of nurses in public hospitals is poor. In particular, centralised and physician-oriented decision-making mechanisms have a negative effect on the scope of responsibility and tasks expected to be carried out by nurses. As Table 2 shows items in subscale 1 assessment-planning-monitoring are most frequently rationed, whereas the items referring to the subscales 2 psychosocial care and subscales 3 daily care-needs were rationed less frequently.

The evaluation of the construct validity of the 33-item BERNCA-R version using EFA revealed that items 9, 13, 14, 20 and 21 had low factor loadings in the various factor models tested. Therefore, it was decided to remove these items low factor loading as well as item 24, as these refer to roles and tasks that, according to the discussion with the expert panel, nurses (in public hospitals) in Turkey are not expected to carry out. As opposed to the version BERNCA-R (Schubert et al., 2013) and the Polish version (Uchmanowicz et al., 2019) with one-factor, the Turkish version of BERNCA-R showed a three-factor structure: (1) planning-monitoring, (2) daily care and needs and (3) psychosocial care. Based on the EFA results, the 27-item and three-factor instrument structure showed the most compatible results with the data. This three-factor structure has facilitated evaluating the implicit rationing of nursing care in Turkey. In the version of the BERNCA that Zúñiga et al. (2016) revised for nursing homes, a 19-item and four-factor structure was obtained after the factor analysis. The results of CFA undertaken for the Turkish version of BERNCA-R confirmed the three-factor model structure of the adapted instrument. Accordingly, the fit indices of the Turkish version of BERNCA-R were shown to be acceptable and have fit index values in acceptable ranges for the instrument structure. However, as the gap between the roles and responsibilities of nurses in Turkey described in nursing curricula and job descriptions and the actual situation in practice is reduced, the adapted instrument should be reviewed again to determine whether it is necessary to reinsert the items removed from the instrument.

In this study, the weak negative but statistically significant correlation shown between this NWI subscale and BERNCA-R indicates a relationship between the perceived adequacy of staffing and resources and the levels of rationed care. From a clinical point of view, this low correlation is not relevant and H2 cannot be confirmed. It should be considered here that, in addition to nurses’ perception of the adequacy of resources and skill mix, changes in nurses’ roles and job descriptions may also be factors affecting rationed nursing care. These factors may have contributed to the low correlation between the NWI subscale and the adapted instrument.

In the Turkish version of the BERNCA-R instrument, the Cronbach’s α of the three factors varied between 0.80 and 0.91, which indicates a high internal consistency of the adapted instrument and confirms H1. The Guttman split-half reliability coefficient and the inter-item correlation values also provided favourable results. These results confirm H1 and indicate that the instrument is internally consistent and reliable. In the Polish BERNCA-R version, the Cronbach’s α coefficient of the one-dimensional scale was found to be 0.96 (Uchmanowicz et al., 2019). Although it is interpreted that the reliability of a scale increases as the Cronbach’s α coefficient approaches 1, in cases where this coefficient is above 0.90, it is recommended to review the scale again considering the possibility that different items may measure the same construct (Tavakol and Dennick, 2011). But each item in the BERNCA-R scale measures a different patient care activity. Furthermore, the Guttman split-half reliability coefficient and the inter-item correlation values were also provided favourable results and which was also evaluated to indicate that the scale had a good level of reliability.

Conclusion

This study provides evidence that the Turkish version of the BERNCA-R instrument is a valid and reliable tool for measuring rationing of nursing care. The Turkish version of the BERNCA-R instrument with a 27-item and three-factor (monitoring, daily care and needs and psychosocial care) structure was found to have acceptable and good fit indices. The BERNCA-R instrument will contribute to measures used in determining of rationing of nursing care in the different healthcare settings.

Key points for policy, practice and/or research.

  • The Turkish version of BERNCA-R scale is a valid and reliable tool for assessing rationing of nursing care.

  • In the collectivist cultures, in comparison to Western countries, family members play an important role in providing patients with basic ADLs and daily needs in acute care hospitals. Therefore, when using this scale in collectivist cultures, the answer option ‘family member did’ should be used to accurately assess the frequency of rationing nursing care.

  • By assessing the rationing of nursing care, nurse managers can evaluate their organisations.

  • This instrument will also offer the opportunity to compare data for the rationing of nursing care between developing countries and developed countries.

  • The rationing of nursing care is a complex concept, and further research is needed to assess the Turkish BERNCA-R scale for external validity with different samples and in different settings.

Biography

Havva Arslan Yürümezoğlu is an associate professor in the Faculty of Nursing at Dokuz Eylul University, Turkey. Her research interests include healthy work environment, rationing nursing care, intention to leave nursing, effects of structural empowerment, nursing care effects on nurse and patient outcomes.

Maria Schubert is a professor and co-head in the Institute of Nursing at ZHAW School of Health Professions, Switzerland. Her research interests include patient safety and quality of care in hospitals, rationing of nursing care, missed nursing care, quality of the nurse work/practice environment, staffing and skill mix, leadership and delirium management.

Emine Sarıoğlu is a nurse in Katip Celebi University Ataturk Training and Research Hospital, Turkey. She has a master’s degree in nursing management.

Gülseren Kocaman is an emeritus professor in the Faculty of Nursing at Dokuz Eylul University, Turkey. Her research interests include quality of nursing education, healthy work environment, intention to leave nursing, nursing care effects on nurse and patient outcomes.

Footnotes

Authors’ note: The manuscript was edited by a native English-speaking editor before the submission.

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding: The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethical approval: In order to adapt the BERNCA-R instrument to Turkish, permission was obtained via e-mail from Schubert et al. (2007) who develop the instrument. Prior to embarking on the study, the managers of the hospitals were informed about the study, and their written permission was obtained. The study protocol was also approved by the university ethics committee (2019/19–39). In addition, the participants were informed regarding the confidentiality of their personal information and the voluntary nature of participation.

ORCID iD: Havva Arslan Yürümezoğlu Inline graphic https://orcid.org/0000-0001-7180-9833

Contributor Information

Havva Arslan Yürümezoğlu, Associate Professor, Department of Nursing Management, Faculty of Nursing, Dokuz Eylul University, İzmir, Turkey.

Maria Schubert, Professor, Institute of Nursing, School of Health Professions, Zurich University of Applied Sciences, Winterthur, Switzerland.

Emine Sarıoğlu, Nurse, Department of Research and Education, Katip Celebi University Ataturk Training and Research Hospital, Izmir, Turkey.

Gülseren Kocaman, Emeritus Professor, Department of Nursing Management, Faculty of Nursing, Dokuz Eylul University, İzmir, Turkey.

References

  1. Aiken LH, Sermeus W, Van den Heede K, et al. (2012) Patient safety, satisfaction, and quality of hospital care: Cross sectional surveys of nurses and patients in 12 countries in Europe and the United States. BMJ 344: e1717. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Aiken LH, Sloane DM, Ball J, et al. (2018) Patient satisfaction with hospital care and nurses in England: An observational study. BMJ Open 8: e019189. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Ausserhofer D, Schubert M, Desmedt M, et al. (2013) The association of patient safety climate and nurse-related organizational factors with selected patient outcomes: A cross-sectional survey. International Journal of Nursing Studies 50: 240–252. [DOI] [PubMed] [Google Scholar]
  4. Ausserhofer D, Zander B, Busse R, et al. (2014) Prevalence, patterns and predictors of nursing care left undone in European hospitals: Results from the multicountry cross-sectional RN4CAST study. BMJ Quality Safety 23: 126–135. [DOI] [PubMed] [Google Scholar]
  5. Bail K, Grealish L. (2016) “Failure to Maintain”: A theoretical proposition for a new quality indicator of nurse care rationing for complex older people in hospital. International Journal of Nursing Studies 63: 146–161. [DOI] [PubMed] [Google Scholar]
  6. Ball JE, Pike G, Griffiths P, et al. (2012) “RN4Cast Nurse survey in England”. National Nursing Research Unit Report. King’s College London. Avaliable at: https://www.kcl.ac.uk/nmpc/research/nnru/publications/reports/rn4cast-nurse-survey-report-27-6-12-final.pdf (accessed 8 December 2021). [Google Scholar]
  7. Ball JE, Bruyneel L, Aiken LH, et al. (2018) Post-operative mortality, missed care and nurse staffing in nine countries: A cross-sectional study. International Journal of Nursing Studies 78: 10–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Ball JE, Murrells T, Rafferty AM, et al. (2014) “Care left undone” during nursing shifts: Associations with workload and perceived quality of care. BMJ Quality Safety 23: 116–125. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Bruyneel L, Li B, Ausserhofer D, et al. (2015) Organization of hospital nursing, provision of nursing care, and patient experiences with care in Europe. Medical Care Research and Review 72: 643–664. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Hair JF, Black B, Black WC, et al. (2014) Multivariate Data Analysis, 7th edn. Harlow: Pearson Education. [Google Scholar]
  11. Kalisch BJ. (2006) Missed nursing care: a qualitative study. Journal of Nursing Care Quality 21: 306–313. [DOI] [PubMed] [Google Scholar]
  12. Kalisch BJ, Landstrom GL, Hinshaw AS. (2009) Missed nursing care: A concept analysis. Journal of Advanced Nursing 65: 1509–1517. [DOI] [PubMed] [Google Scholar]
  13. Kline RB. (2016) Principle and Practice of Structural Equation Modelling, 4th edn. New York, NY: The Guilford Press. [Google Scholar]
  14. Lake ET. (2002) Development of the practice environment scale of the Nursing Work Index. Research in Nursing & Health 25: 176–188. [DOI] [PubMed] [Google Scholar]
  15. Pett MA, Lackey NR, Sullivan JJ. (2003) Making Sense of Factor Analysis: The Use of Factor Analysis for Instrument Development in Health Care Research. Thousand Oaks, CA: Sage Publications, Inc. [Google Scholar]
  16. Polit DF, Beck CT, Owen SV. (2006) The content validity index: Are you sure you know what’s being reported? Critique and recommendations. Research in Nursing & Health 29: 489–497. [DOI] [PubMed] [Google Scholar]
  17. Polit DF, Beck CT, Owen SV. (2007) Is the CVI an acceptable indicator of content validity? Appraisal and recommendation. Research Nursing & Health 30: 459–467. [DOI] [PubMed] [Google Scholar]
  18. Recio-Saucedo A, Dall'Ora C, Maruotti A, et al. (2018) What impact does nursing care left undone have on patient outcomes? Review of the literature. Journal of Clinical Nursing 27: 2248–2259. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Schubert M, Ausserhofer D, Desmedt M, et al. (2013) Levels and correlates of implicit rationing of nursing care in Swiss acute care hospitals-A cross sectional study. International Journal of Nursing Studies 50: 230–239. [DOI] [PubMed] [Google Scholar]
  20. Schubert M, Clarke SP, Glass TR, et al. (2009) Identifying thresholds for relationships between impacts of rationing of nursing care and nurse-and patient-reported outcomes in Swiss hospitals: A correlational study. International Journal of Nursing Studies 46: 884–893. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Schubert M, Glass TR, Clarke SP, et al. (2007) Validation of the Basel extent of rationing of nursing care instrument. Nursing Research 56: 416–424. [DOI] [PubMed] [Google Scholar]
  22. Schubert M, Glass TR, Clarke SP, et al. (2008) Rationing of nursing care and its relationship to patient outcomes: The Swiss extension of the International Hospital Outcomes Study. International Journal for Quality in Health Care 20: 227–237. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Schubert M, Schaffert-Witvliet B, De Geest S, et al. (2005) RICH-nursing study: Rationing of nursing care in Switzerland. Effects of rationing of nursing care in Switzerland on patient’ and nurses’ outcomes. Final Report, Basel, Switzerland: Institute of Nursing Science, University of Basel. [Google Scholar]
  24. Tavakol M, Dennick R. (2011) Making sense of Cronbach’s alpha. International Journal of Medical Education 2: 53. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Turkmen E, Badır A, Balcı S, et al. (2011) The adaptation of the practice environment scale of the Nursing Work Index into Turkish: Reliability and validity study. Journal of HEMAR-G 13: 5–20. [Google Scholar]
  26. Uchmanowicz I, Kirwan M, Riklikiene O, Wolfshaut-Wolak R, et al. (2019) Validation of Polish version of the Basel Extent of Rationing of Nursing Care revised questionnaire. PloS One 14: e0212918. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. World Health Organization (2016) Global Strategy on Human Resources for Health: Workforce 2030. Geneva, Switzerland: World Health Organization. Avaliable at: https://apps.who.int/iris/bitstream/handle/10665/250368/9789241511131-eng.pdf?sequence=1 (accessed 5 December 2021). [Google Scholar]
  28. Zúñiga F, Schubert M, Hamers JPH, et al. (2016) Evidence on the validity and reliability of the German, French and Italian nursing home version of the Basel Extent of Rationing of Nursing Care instrument. Journal of Advanced Nursing 72: 1948–1963. [DOI] [PubMed] [Google Scholar]

Articles from Journal of Research in Nursing are provided here courtesy of SAGE Publications

RESOURCES