Research design
The validation of the ESQ was carried out through a descriptive cross-sectional design study, where a purposive sampling was employed across the nursing and physiotherapy workforce in the four countries which participated in this study. Participants were recruited through intermediaries within one acute hospital in each partner country. These hospitals were selected based on their affiliation with the respective educational institutions. Participation in this research study was voluntary and anonymous. Data was collected from June 2023 to August 2023 via Webropol 3.0 in hospitals in Finland, Sweden, Latvia, and Malta.
Participants and research context
Study population
The study population consisted of a total of 275 nurses and physiotherapists at four major hospitals in Finland, Sweden, Latvia, and Malta.
Inclusion and exclusion criteria
Nurses and physiotherapists in acute healthcare environments with permanent employment, both part-time and full-time, were eligible to participate in the present study. Healthcare professionals were included regardless of minimum work experience. Students and agency hired staff or temporary workers were excluded.
Study contexts and recruitment procedure
Four major hospitals in Finland, Sweden, Latvia, and Malta were chosen for data collection. All hospitals were urban-based and nurses and physiotherapists working in acute care were invited to participate. Health professionals from non-acute settings were excluded. Purposive sampling was used to obtain the required sample across the four study settings. The hospitals involved in recruitment were all teaching hospitals affiliated with the educational entities conducting this research study. The respective management of the hospitals were informed about the study, and permissions were obtained to conduct the data collection. The researchers from each of the four countries, thereafter, introduced the study to the managers of the respective hospitals in the associated country and these managers acted as intermediaries by distributing the Webropol link to the questionnaire to all nurses/physiotherapists working at the hospital. Every participant received a link to the questionnaire with a cover letter informing participants about the aim and purpose of the study. The letter explained that participation entails the completion of a questionnaire (the ESQ) and it emphasised voluntary participation and assured anonymity in relation to participation. Completion of the questionnaire via Webropol implied a respondent’s consent to participate. A reminder to fill out the questionnaire was sent twice at two-week intervals.
Sample size
Purposive sampling was used to obtain the required sample over 200 participants completed the questionnaires. This sample size complies with validation studies standards ranging from 100 to 400 [12].
Data collection and instrument
Data collection tool
In this study, the ESQ was used [4]. Permission to translate and use the tool was granted by the developer, Tarja Poikkeus (also co-author of the present manuscript). The ESQ contains 11 items. Two items are general ethical safety statements and there are three subscales: ethical autonomy, ethical respect, and ethical confidence. The ESQ uses a 5-point Likert scale where 1 represents ‘strongly disagree’ up to 5 ‘strongly agree’. The three subscales all contain three items each. Higher scores on each ESQ subscale (Ethical Autonomy, Ethical Respect, and Ethical Trust) reflect stronger perceptions of ethical safety in the healthcare work environment.
Cross-cultural adaptation and face validity
The original ESQ in Finnish, was translated according to translation unit standards [13]. The forward translations (Finnish to English, Swedish, and Latvian) were completed by three translators who were fluent native speakers of the respective languages. This was followed by a reconciled translation based on the three translations; a third person, reviewed the translations and combined them into one reconciled ESQ. The reconciled ESQ-translated versions were back-translated into Finnish by two independent persons, for the respective language, who were fluent in English, Swedish and Latvian. The results of all steps (i.e., two forward translations, reconciliation, and two backward translations with comments) were proofread by an external proofreader who audited the final version of the ESQ in English, Swedish and Latvian.
Pilot testing
The translated versions in Swedish, English, Latvian were ready for linguistic validation through pilot testing. The pilot tests of the translated versions were conducted in each country [13]. The questionnaire was completed by 10 participants in each country consisting of nurses and physiotherapists for face validation of the ESQ. Only minor changes were made after the piloting, such as changing terminologies to incorporate both physiotherapists and nurses within the specific healthcare environments.
Psychometric evaluation
Statistical analysis
Items distributions are presented as frequency and percentage. Skewness and kurtosis statistics were computed to assess the distributional characteristics of the items. Values outside the range of −2 to + 2 for skewness and − 7 to + 7 for kurtosis were considered as potential deviations from normality [14].
The internal consistency of the ESQ was measured using Cronbach’s α and McDonald’s ω, with values higher than 0.7 indicating acceptable internal consistency. The item-total correlation (corrected for overlap) was further computed to measure the internal consistency with values higher than 0.4 indicating satisfactory consistency [15, 16].
The factor structure of the 9 items of the ESQ was assessed using exploratory factor analysis (EFA) with maximum likelihood extraction and varimax rotation. Sampling adequacy and the suitability of the correlation matrix were assessed using the Kaiser-Meyer-Olkin (KMO) test and Bartlett’s test of sphericity, respectively. A value higher than 0.6, along with a significant Bartlett’s test of sphericity, indicates the factorability of the data [17]. Significant factor loadings with a magnitude of > 0.40 were considered meaningful. The Scree plot and Kaiser rule were used to determine the number of factors to retain.
To investigate whether the items of the ESQ correspond with their underlying concept, a Rasch analysis using a partial credit model (PCM) was conducted [14]. Item fit was measured using indices including Outfit and infit mean square (MnSq) with values between 0.5 and 1.5 indicating good fit [18]. Local independence between the items was assessed by computing residual correlations between any two items of ESQ, with values higher than 0.2 indicating a violation of local independence. Differential item functioning (DIF) was computed to measure the measurement invariance of the ESQ items across gender and profession subgroups. A DIF contrast of 0.64 was set as a cut-off for a meaningful statistically significant DIF [19].
Finally, to evaluate the relationship between the ESQ items and general ethical safety, a network analysis was conducted. Network analysis is valuable for examining complex relationships and interactions between study variables and visualising these associations. In network analysis, each variable can be represented as a node, and the association between each node is referred to as an edge. The centrality of the nodes was determined by analyzing three metrics including degree centrality, betweenness centrality, and closeness centrality. The network analysis was computed using EBICglasso (Extended Bayesian Information Criterion Graphical Least Absolute Shrinkage Selection Operator) estimation, employing the non-parametric bootstrap procedure with 500 iterations to analyse the 95% confidence interval (CI) of edge weights [20]. All statistical analyses were performed using SPSS version 27, JASP version 0.18.3.0 and Winsteps version 4.3.0.
Ethical considerations
The study follows the guidelines on research from the Declaration of Helsinki, regarding research on human subjects [18]. The respective University Ethics Committees approved the present study, including the Research Ethics Committee of Turku University of Applied Sciences (1/2023), Riga Stradins University Research Ethics Committee (2-PEK-4/507/2023) and the University of Malta, Faculty of Health Sciences Research Ethics Committee (FHS-2023-00077), and ethical guidelines according Swedish law (SFS: 2003:460) was followed, and accordingly approval to conduct the study was gained by the head of the clinics of the involved hospital. Additional permissions were addressed depending on specific country’s requirements. Furthermore, permissions were obtained from the executive directors of the respective study sites and the directors of the departments or head nurses. Informed consent was obtained from individual participants through completion of the questionnaire and participation was voluntary and anonymous; participants could withdraw their participation if they do not submit the survey.