Validation and psychometric properties of the Persian version of the MISSCARE survey

Abstract

Aim

To conduct, translate, and psychometric evaluation of the MISSCARE-Persian Survey.

Design

A cross-sectional study was conducted in Iran in February–June 2021.

Methods

The translation of the MISSCARE Survey was accomplished according to World Health Organization (WHO) guidelines. Construct validity was performed by (N = 300) exploratory factor analysis and confirmation. To assess the reliability, internal consistency was assessed using Cronbach's alpha coefficient, and relative stability was assessed using the interclass correlation coefficient (ICC). The study adhered to COSMIN guidelines.

Results

The exploratory factor analysis, which resulted in the identification of three factors in the second part of the tool, explained 79.6% of the total variance. Confirmatory factor analysis indicated the model's good fit of information. The reliability of the first and second parts of the tool was 0.912 and 0.901, respectively. Additionally, the ICC was found to be 0.687 for the first and 0.706 for the second part of the tool.

INTRODUCTION

The quality of nursing services is a major factor in patient safety (Kim et al., 2018). Missed care is a quality index identified for nursing care and patient safety recently (Bragadóttir et al., 2017). As a significant hidden problem, missed care could result in nursing dysfunction (Kalisch et al., 2009; Zeleníková et al., 2019). It could also lead to undesirable outcomes, such as morbidity, mortality, patients' dissatisfaction with healthcare providers (Gibbon & Crane, 2018), and increased financial burden on the healthcare system since it results in readmission and increases the length of hospital stay to manage complications and adverse outcomes (Janatolmakan & Khatony, 2022). Thus, using an instrument that has the capability to investigate the extent and nature of this phenomenon and help study staff perceptions of care omission and the reasons for that is necessary (Siqueira et al., 2017).

BACKGROUND

One of the objectives of patients' rights charter is assurance of the health system's responsiveness to patient needs. However, lack of time and personnel could lead to ignorance of key points in patient care (John et al., 2016). In other words, patients' numerous care demands in a healthcare environment with lack of resources may cause nurses to prioritize the demands (Kalisch et al., 2009). Under such circumstances, the nurse decides which responsibilities to carry out or ignore (John et al., 2016). Consequently, some nursing responsibilities are missed (Kalisch et al., 2009). The neglected responsibilities may be different depending on various reasons, such as pressure for prioritization, teamwork, and the nurse's internal value system (Bagnasco et al., 2017). However, what occurs during nursing care provision has not been well identified. In other words, the specific outcomes of what nurses do or do not have not been well determined. In this context, negative outcomes seem to be associated with what Kalisch has called missed care (Moreno-Monsiváis et al., 2015).

Missed nursing care is a global challenge threatening patient safety and health (Janatolmakan & Khatony, 2022). The concept of missed care encompasses all aspects of clinical, emotional and prescriptive nursing, which have been partially completed, delayed, or not completed (Recio-Saucedo et al., 2018). Missed care has also been known as implicitly rationed care, nursing care left undone, unmet patient needs, unfinished nursing care and delayed nursing care (Jones et al., 2015).

In developed countries, it has been estimated that one out of every 10 hospitalized patients is hurt due to care negligence (Moreno-Monsiváis et al., 2015). In this regard, a previous study demonstrated that missed care was usual in U.S. hospitals. In that study, nearly three-fourths of nurses reported at least one missed care during their last shift. This measure has been found to be considerably higher in developing countries as all nurses have reported the incidence of missed care in some hospitals (Lake et al., 2017). In another study conducted by Matin et al. on the reporting of nursing errors in Iran between 2000 and 2017, it was found that the prevalence of nursing errors was different in diverse regions of Iran and was reported between 17% and 88%, with an average of 53% of the total number of nursing errors (Matin et al., 2018). The aforementioned studies demonstrated that the rates of missed care could be different in various hospital wards and in different communities. Hence, in order to achieve an accurate picture of missed nursing care, a reliable and valid tool based on each community's culture and language is required (Bragadóttir et al., 2015).

Several conceptual frameworks have been published on rationed/missed/unfinished nursing care (Hessels et al., 2015; Lucero et al., 2009; Schubert et al., 2007). However, the conceptual framework of the present study was the missed care model presented by Kalisch et al. (2009), which consists of three elements: 1- structure (hospital, ward, and nursing personnel's characteristics), 2- process (missed nursing care), and 3- outcome (staff's outcomes including job satisfaction and patients' outcomes including falling and prevalence of pressure ulcers) (Kalisch & Lee, 2012). This framework represents the interaction between missed nursing care and staff's and patients' outcomes (Bragadóttir & Kalisch, 2018). The model also indicates the characteristics of the organizational work environment (Zeleníková et al., 2019). In fact, this model has been designed based on three antecedents: financial resources, human resources and relations, which enable nurses to carry out the care process. These three antecedents are under the influence of the nurses' internal processes, including team norms, decision-making priorities, habits and values (Bagnasco et al., 2018). According to this model, a valid tool is beneficial in identifying causes and outcomes of missed care and solutions to reduce this phenomenon (Siqueira et al., 2017).

To this end, Kalisch designed the missed care standard tool in 2006 to determine omitted or delayed care and assess the barriers against care provision using 25 focus groups with nurses and nursing assistants. This tool was validated for adult patients hospitalized in internal, surgical, and intensive care units (Kalisch, 2006).

Two generations of missed care tools have been psychometric properties evaluated to date. the first one was, the validity and reliability of the original version which has been reported in several languages and countries and the second one was developing the original tool for other groups of patients (Palese et al., 2021), such as neonatal intensive care units (Tubbs-Cooley et al., 2015), paediatrics care units (Bagnasco et al., 2018) and prenatal care units (Simpson et al., 2019). Both generations of tool development have recently been continued with new validations for languages and cultures which adds to its value and makes it preferable for measuring missed nursing care (Bragadóttir et al., 2015; Palese et al., 2021).

Although comparison of different countries can provide tremendous insights regarding nursing care provision methods and their outcomes for patients, the lack of a reliable measurement tool adapted to a language and culture is a significant obstacle to investigating nursing care quality (Kalisch et al., 2012).

Therefore, given that the MISSCARE survey is accepted worldwide, developing a tool in compliance with this instrument can be a preference to evaluate missed nursing care in adult patients. Moreover, it is essential to validate an international tool and provide appropriate information about its psychometric process. Therefore, the present study aims to investigate the reliability and validity of the Persian version of the MISSCARE Survey in several provinces to take into account differences in professional nursing practice in the Iranian context.

METHODS Study design

The study was methodological and cross-sectional. This study was designed according to the COnsensus-based Standards for selection of health Measurement Instruments (COSMIN) guideline.

Sampling

To establish diversity in hospital size (from 100 to 1000 beds) and type (educational, non-educational, general, and private), a purposive sample of 10 acute care hospitals was used in Tehran, Shiraz, and Semnan provinces, Iran. In this study, six hospital units (oncology, trauma ICU, general medicine, surgery, cardiac care unit (CCU), and COVID-19) were included.

Using the convenience sampling method, nurses providing direct patient care were invited to participate in an online survey. Inclusion criteria included: being nurses, nursing technicians, and nursing aides with an employment relationship with the selected hospital, assignment to various units, and working in the unit for at least a month. Excluded participants included professionals on vacation or medical leave. This gave a potential sample size of 500. In the present study, the response rate was approximately 60%.

Data collection

Considering the prevalence of the COVID-19 pandemic, data collection was performed using the MISSCARE – Persian Survey from February to June 2021 through an online survey.

Characteristics of the MISSCARE survey

The MISSCARE Survey consists of three sections with 41 items. The first section contains demographic and underlying information, including work conditions and satisfaction levels. Part a (24 items) involves nursing activities to assess missed care elements. In addition, part B (17 items) investigates the reasons for missed care in three dimensions, namely relations/teamwork (nine items), financial resources (3 items), and human resources (5 items).

In this study, the participants were requested to choose the answer with the best level of agreement with the item. For the first part, the respondents were asked to mark always = 5, often = 4, sometimes = 3, rarely = 2, or never = 1 for the care missed by all personnel in their wards. For the second part, they were required to choose a very important reason = 1, moderate reason = 2, weak reason = 3 or no reason = 4. The answers to items in parts A and B were reversely scored, with (1 = 5) (2 = 4) (3 = 3) (4 = 2) (5 = 1) for part A and (1 = 4) (2 = 3) (3 = 2) (4 = 1) for part B, as instructed by the author of the original survey. Higher scores represented higher levels of missed nursing care in part A and a stronger reason for its occurrence in part B.

Job satisfaction

The first part of the instrument contained three items related to job satisfaction: satisfaction with the job, satisfaction with teamwork in the ward, and satisfaction with the job position. The respondents were asked to evaluate the items using a five-point Likert scale (5: very dissatisfied, 4: dissatisfied, 3: not satisfied and not dissatisfied, 2: satisfied and 1: very satisfied).

Procedure Translation

In order to translate the missed care tool, the guidelines proposed by the World Health Organization (WHO) were used (WHO, 2016). Considering the defined standard in tool localization, permission was first obtained from Dr. Kalisch, the tool designer at the University of Michigan. Afterwards, the tool was translated into Persian by two proficient translators separately. The two translations were discussed in a specialized panel, and the two translators' opinions were matched. The approved Persian version was then given to a specialist in Persian Literature who was required to edit it with respect to grammar and wording. Afterwards, the Persian version was back-translated into English by a proficient translator. Finally, the translated version was sent to the designer, and the items were reviewed and confirmed accordingly.

Content and face validity Cognitive interview

The face validity was assessed via face-to-face interviews with 10 nurses with various clinical experiences and job tenures. These nurses were requested to evaluate the items concerning ambiguity, difficulty and appropriateness and suggest terms for substitution, if necessary. Accordingly, the necessary modifications were applied. After forward-backward translation, based on the results of the investigation of face validity, the understandable version of the tool appropriated to the Iranian culture was selected.

Content validity

The scale's content validity was assessed according to 10 experts' opinions based on Waltz and Bausell's Content Validity Index (CVI) (Waltz & Strickland, 2010). In order to determine CVI, ‘cultural relevancy’ was used for all items based on a four-point Likert scale. To this end, 15 nursing specialists were asked electronically or in person to determine the relationships between the tool items regarding the subscales using a four-point Likert scale (1: not related, 2: somehow related, 3: related but needs revision, and 4: completely related). Afterwards, the items receiving the highest scores (rating 3 or 4) were divided by the total number of experts.

Construct validity

In order to determine the construct validity of the tool and discover the hidden factors, exploratory factor analysis (EFA), the most common method for an item classification, was used. The sample size was determined by considering an adequate sample balance and consensus to estimate the minimum sample necessary to conduct a reliable factor analysis (Tabachnick et al., 2007). According to Munro (2005), 20–30 participants per factor were needed to conduct confirmatory factor analysis (CFA) (Munro, 2005). Given that three dimensions are considered in part B original version of the scale; 100–150 samples were required for CFA. Therefore, 150 nurses were included in the study for CFA. Indicators of model fit in CFA in three general categories are as follows: (1) absolute fit: the root mean square error of approximation (RMSEA), standardized root means square residual (SRMR), Goodness-of-Fit index (GFI) and chi-square; (2) comparative fit: index (CFI), incremental fit index (IFI), relative fit index (RFI), normed fit index (NFI) and Tucker-Lewis index (TLI); (3) affordable fit: parsimony comparative fit index (PCFI), parsimony normal fit index (PNFI), adjusted goodness-of-fit index (AGFI) and Akaike's information criterion (AIC) (Samitsch, 2014).

Reliability

The reliability of the tool was assessed by internal consistency and test–retest methods. Internal consistency was determined using Cronbach's alpha coefficient and temporal consistency using the test–retest method. For this purpose, the tool was completed by 30 nurses within a two-week interval.

Data analysis

Considering descriptive statistics, quantitative variables were presented through mean ± Standard Deviation (SD), and qualitative variables were reported as frequency (percentage). In order to determine the psychometric properties of the tool, factor analysis, the Kaiser-Meyer-Olkin (KMO) test, and Bartlett's test of sampling adequacy were used (Table 1). At first, EFA, using maximum likelihood, was used to analyse the related items. It should be noted that the normal distribution of data was the prerequisite for this method. In addition, skewness and kurtosis of the variables were within the range of ±3. Then, maximum likelihood with PROMAX rotation was used to promote the initial factor structure. The minimum cutoff value for factor loading was considered >0.3. All data analyses were carried out using the SPSS statistical software, version 25, and Lisrel 8.8. p < 0.05 was considered statistically significant.

TABLE 1 Final results of KMO and Bartlett's test of sphericity to determine the validity of the tool.

Bartlett test
KMO	Chi-Square	df	p-value
0.906	2160.809	136	p = 0.000*

^*Significant effect, p < 0.05.

Ethical considerations

This study was approved by the Ethics Committee of the University of Social Welfare and Rehabilitation Sciences, Tehran, Iran (code: IR.USWR.REC.1399.058) and performed in accordance with the Helsinki Declaration of 1975 and the ethical standards of the institution's Committee on Human Experimentation. The necessary permissions were obtained from the tool designer and the authorities of the healthcare centres. Moreover, the nature and objectives of the research were completely explained to the participants. They were also informed about the voluntary nature of the study and their freedom to voluntarily participate in or withdraw from the study. Confidentiality of the data was also taken into consideration, and the results were published anonymously.

RESULTS

This study was conducted on 300 nurses caring for adult patients in different hospital wards (except paediatric wards). Three hundred participants included 5.65% aides, 1.3% nursing technicians, 85.3% nurses, and 7.5 nurses with administrative roles. The mean age of participants was 24.15 (SD = 1.32) years. Most participants (80.9%) were female and 40–30 years old (40.6%). The majority of the participants (97.3%) had at least a bachelor's degree in nursing, 99.3% worked as staff nurses and 33.7% had more than 10 years of work experience (Table 2).

TABLE 2 The characteristics of participants (N: 300).

Variable		n	%
Age (years) Mean (SD): 24.15 (SD = 1.32)	20–30	120	39.6
	40–30	123	40.6
	50–40	52	17.2
	>50 years	5	1.7
Gender	Female	245	80.9
	Male	55	18.2
Professional experience	less than 1 year	16	3.5
	1–2 years	31	10.2
	2–5 years	63	20.8
	5–10 years	86	28.4
	more than 10 years	102	33.7
Experience in the current unit	less than 1 year	30	9.9
	1–2 years	54	17.8
	2–5 years	77	25.4
	5–10 years	72	23.8
	more than 10 years	64	21.1
Work hours	morning	57	18.8
	evenings	7	2.3
	evening and night	45	9.14
	rotating shifts	190	62.7
Adequacy of staffing	100% of the time	11	3.6
	75% of the time	38	5.12
	50% of the time	83	27.4
	25% of the time	95	31.4
	never	73	24.1

Content validity index

In this study, the CVI was found to be >0.79 for all items in the first part of the tool. Besides, Ave/scale-CVI was 0.9 in the first part. The mean CVI for all 41 items was equal to 0.86.

Reliability

ICC computed through the test–retest method was found to be 0.687 for the first section and 0.706 for the second. The Cronbach's alpha coefficient was calculated as 0.912 for the first section and 0.901 for the second. Cronbach's alpha was greater than 0.70, an acceptable level of internal consistency. Cronbach's alpha coefficients for all subscales are presented in Table 3.

TABLE 3 Correlation between test scores and retesting the final form of Missed Nursing Care Survey and the Reasons affecting it (N: 30).

[Table omitted. See PDF]

Abbreviations: ICC, Intra-class Correlation Coefficient; SD, Standard Deviation.

Note: Significant effect, p < 0.05.

Construct validity Exploratory factor analysis

Prior to the calculation of construct validity, a pilot study was conducted on 70 nurses working in various healthcare centres. According to the results, the reliability coefficient of the first and second parts of the tool was 0.94 and 0.92, respectively. Considering the acceptable results at this stage, the tool was entered into the next step. The results indicated that the three factors explained 79.6% of the variance. Since part one included the independent nursing care tool, factor analysis was only conducted for the second part. Extraction of factors using the maximum likelihood analysis with Eigenvalue>1 led to the identification of three factors, which explained 79.6% of the total variance. The matrix of the rotated factors is presented in Table 4. This matrix revealed the dimensions to which the items belonged. The final tool included 41 items and three factors in two sections. After the factor analysis and achievement of the dimensions, Cronbach's alpha was computed again.

TABLE 4 Exploratory factor loadings of items in Missed Nursing Care in Iran with three factors and Cronbach's Alpha.

Factors	Items	Variance	Eigen value	Cronbach's alpha	Factor loading
					1	2	3
Communication	Tension or communication breakdowns with the medical staff	29.4	5.33	81.6			0.866
	The nursing assistant did not communicate that care was not provided.					0.748
	Tension or communication breakdowns within the nursing team					0.716
	Caregiver off the unit or unavailable					0.552
	Tension or communication breakdowns with other ancillary/support departments					0.413
	Heavy admission and discharge activity					0.383
	Lack of backup support from team members					0.507
	Other departments did not provide the care needed.					0.393
	The inadequate hand-off from the previous shift					0.463
Material Resources	Supplies/equipment not available when needed	29.1	5.08	79.2		0.803
	Supplies/equipment not functioning properly					0.776
	Medications not available when needed					0.426
Labour resources	Unexpected rise in patient volume and/or acuity in the unit	21.1	3.6	80.7	0.756
	Inadequate number of staff				0.721
	The inadequate number of assistive and/or clerical personnel				0.597
	Urgent patient situations				0.537
	Unbalanced patient assignments				0.387

Confirmatory factor analysis

Another sample consisting of 150 nurses was selected for CFA. The results of the chi-squared test (x² = 214.19 and p = 0.00) and other fit incises showed that the three-factor model extracted from EFA had a good fit of the data (RMSEA: 0.075; NFI: 0.88; CFI: 0.94; IFI: 0.94; RFI: 0.86; AGFI: 0.81; NNFI: 0.93; PNFI: 0.75; standardized RMR: 0.074). Finally, the results indicated that CFA based on the three-factor model extracted from EFA with the obtained data had a good fit (Figure 1).

FIGURE 1. The final structure of the model.

Elements of missed care

Out of the 24 nursing activity elements in the first section of the tool, five were responsible for the highest percentage of missed nursing care. These five elements were assisting a patient to walk, position change, emotional support, oral care, prescription of medications in 30 min and documentation of essential information.

Reasons for missed care

Regarding relations, financial resources and human resources, relations and teamwork were the main reasons for missed care in males and females.

Mean job satisfaction (1.136 ± 2.68), satisfaction with teamwork in the ward (2.53 ± 0.967) and satisfaction with the current job position (3.13 ± 1.022) indicated a balanced level of job satisfaction, teamwork and job position.

DISCUSSION

Overall, the investigations on missed care have revealed that this concept is a global phenomenon that can affect patient safety and the quality of nursing care. The MISSCARE Survey has been widely used to evaluate this concept and analyse its causes worldwide (Zeleníková et al., 2019). Hence, this methodological study aimed to evaluate the psychometric properties of the MISSCARE Survey to be used by Iranian nursing professionals.

In the present study, the translation and testing of the MISSCARE Survey from US English to Persian were successfully carried out in six steps, including a back-translation process and psychometric testing. When the content validity assessment began, some experts in our study disagreed on the usefulness of some items in Part A in Iran. However, the revision following the cognitive interview led to some modifications to the tool interface, and no items from either part of the survey were removed. In the translation of the MISSCARE Survey- Icelandic version, the participants had difficulty understanding the meaning of ‘missed nursing care’, and the researchers had to provide more explanations to the participants in responding to the questionnaire (Bragadóttir et al., 2015).

The study findings revealed that female professionals with more than 10 years of experience and a bachelor's degree were predominant. This finding is consistent with prior research (Kalisch & Williams, 2009; Siqueira et al., 2017).

The psychometric testing demonstrated that this tool was reliable and valid for evaluating missed nursing care among adults in various therapeutic wards in Iran. The Cronbach's alpha coefficient of the first and second parts of the localized tool was 91.2 and 90.1, respectively, confirming the reliability and internal consistency of the tool. Similar results were also obtained in the studies conducted in the US (Kalisch & Williams, 2009), Czech and Slovakian (Zeleníková et al., 2019), Brazilian (Siqueira et al., 2017), Turkish (Kalisch et al., 2012), and Icelandic (Bragadóttir et al., 2015) versions of the tool.

In order to conduct construct validity, we divided part B of the survey into three factors: human resources, financial resources, and relations. Most studies, likewise, divided part B into three factors: communication/teamwork, labour resources and material resources (Bragadóttir et al., 2015; Kalisch & Williams, 2009; Zeleníková et al., 2019). Some studies have divided Part B into five factors to more accurately determine the reason for missed care (Castner et al., 2015; Siqueira et al., 2017). The Cronbach's alpha coefficients ranged from 79.2 to 81.6, indicating the internal consistency of the tool. Other studies evaluated the internal consistency of section B of the tool, results ranging from 0.699 to 0.936 (Kalisch & Williams, 2009; Papastavrou et al., 2016). For the construct validity of part B, the CFA showed a good model fit with the three factors of reasons for missed nursing care. The resulting factor loadings established the items belonging to each factor.

In section A of the tool, the construct validity has not been fully explored because it is simply a list of nursing care activities that could provide a better assessment of the extent of missed nursing care, reflecting a unidimensional construct. However, if section A requires users to rank the interventions missed according to their frequency from rarely to always, this section might require other assessments, such as the Mocken scale, which is a unidimensional scale that consists of hierarchically-ordered items that measures the same underlying, latent concept (Palese et al., 2021).

In the present study, the most common missed cares included assisting a patient to walk, changing the patient's position every 2 h, oral care, bathing the patient and training the patient. In the same vein, Zeleníková et al. (2019) reported that the mostly ignored nursing activities were assisting a patient to walk, position change, emotional support, oral care, medication administration in 30 min and documentation of essential information. It was also stated that in case the nurses could not provide patients with the necessary services, they had to prioritize them (Zeleníková et al., 2019). Dehghan-Nayeri et al. (2015) also disclosed that priorities were determined based on the system's expectations, the physicians' orders, and the nurses' roles in the system. These cases would cause the nurses to omit care and consider priorities in care provision (Dehghan-Nayeri et al., 2015). For instance, in the study by Sung-Hyun, nursing activities such as patient reassessments, timely medications and patient teaching were prioritized more than other nursing activities (Cho et al., 2020).

In the present study, relations were the most important reason for the incidence of missed care. In fact, nurses believed that this dimension was faced with more significant challenges. However, studies in Slovakia and Brazil reported that labour resources were the most critical reason in those countries (Siqueira et al., 2017; Zeleníková et al., 2019).

The present study findings showed moderate satisfaction with the occupation, teamwork, and position, indicating overall job satisfaction. When nurses could not fulfil the care they felt was necessary, they reported dissatisfaction with their jobs (Zeleníková et al., 2019). These results were in agreement with the research by Morshedi et al. (2020), which demonstrated that 77% of the nurses were satisfied with their status compared to the past (Morshedi et al., 2020).

In light of these findings, this tool can be used as an exploratory tool for missed nursing care resulting from active errors, such as carelessness, distraction and ignorance or hidden factors, such as high workload, inadequate supervision, deficient relationships, inappropriate care for facilities and reduction of standard processes (Siqueira et al., 2017).

One limitation of the present study was that the researcher was unable to be present in healthcare centres due to the COVID-19 pandemic, and the tools were electronically (online) completed by nurses in various health centres. Hence, the rate of missed care might have been underestimated by nurses. Furthermore, the nurses could not be observed during care provision to explore the neglected activities. As a result, some participants might have been unwilling to indicate missed care, despite the careful attention given to anonymity.

CONCLUSION

The present study results indicated that the MISSCARE- Persian Survey with 41 items enjoyed appropriate reliability and validity and could be used to evaluate the missed care during hospitalization among adult patients. Nonetheless, missed care might not necessarily represent the low quality of nursing care; rather, it may result from a nurse's clinical judgement and prioritization of services based on the patient's conditions. Further studies should be conducted to determine the causes and outcomes of missed care, and the applicability and strength of the MISSCARE- Persian survey.

Clinical applications

By identifying the facilitating and preventive factors of missed care, nursing managers and instructors will be able to design and run their management and educational activities based on the scientific findings that provide the necessary conditions for learning and proper implementation of nursing interventions.

ACKNOWLEDGEMENTS

The authors would like to thank Dr. Kalisch for providing permission to use the missed care tool. They would also like to appreciate the authorities of the healthcare centres and the nurses who contributed their time and effort to this research despite their high workload. The authors are most grateful to the participants who shared their experiences and contribution to the conversation which enabled this research.

FUNDING INFORMATION

This research was supported by the University of Social Welfare and Rehabilitation Sciences (USWR), Tehran, Iran.

CONFLICT OF INTEREST STATEMENT

All authors declare no financial, personal, or potential conflicts of interest.

DATA AVAILABILITY STATEMENT

Data available on request due to privacy/ethical restrictions.

ETHICS STATEMENT

The University of Social Welfare and Rehabilitation Sciences Ethics Committee approved the study. All participants signed an informed consent form.