Psychometric properties of the Farsi version of the Disaster Nursing Readiness Evaluation Index (F-DNREI)

Abstract

Background

Considering the vital role of nurses in responding to disasters, it is essential to measure their readiness with a valid and reliable tool. The present study aimed to assess the psychometric properties of the Farsi version of the Disaster Nursing Readiness Evaluation Index (F-DNREI).

Methods

This cross-sectional study was conducted between 2023 and 2024 among 200 nurses working at educational hospitals affiliated with one of the medical universities in Tehran, Iran. The Disaster Nursing Readiness Evaluation Index was translated into Persian. The face, content, and construct validity, as well as internal consistency, were analyzed.

Results

In the exploratory factor analysis, five factors were extracted: practical skills for disaster response, adaptability to stressful situations at the disaster site, communication and cooperation skills for teamwork, emergency nursing skills, and effective coping with daily stress. Together, these factors accounted for 39.7% of the total variance. The results of the confirmatory factor analysis indicated that the extracted model fit well: CMIN/DF = 1.519, CFI = 0.889, RMSEA = 0.051. The Cronbach’s alpha and McDonald’s omega coefficients for the entire questionnaire were 0.890 and 0.891, respectively.

Conclusions

Given that Iran is frequently exposed to disasters, it becomes essential to assess the preparedness of Iranian nurses using a valid and reliable scale. The availability of the Farsi version of the Disaster Nursing Readiness Evaluation Index (F-DNREI), which has undergone validation and reliability testing, facilitates accurate measurement of this concept.

Background

A disaster refers to a serious disruption in the functioning of a community or society, impairing the provision of essential services and resulting in the loss of human lives, financial assets, and economic and environmental resources. Such disruptions render the affected society unable to meet its resource needs independently and necessitate foreign assistance [1]. On a global scale, disasters occur approximately once a week, inflicting devastating effects on the health and well-being of individuals, families, and communities. The occurrence of natural hazards worldwide leads to a significant loss of life each year [2]. In 2023, 399 disasters were reported globally, resulting in the tragic loss of 86,473 lives. Notably, the Asian continent accounted for 163 of these disasters, leading to the deaths of 63,445 people [3].

Iran, with its unique geographical conditions, has consistently faced a broad spectrum of natural and man-made hazards. Over the years, a significant number of lives have been lost due to these calamities. Notably, the devastating Bam earthquake in 2003 claimed the lives of more than 30,000 people in southeastern Iran. Other earthquakes, such as those in Zarand (2005), Silakhor (2006), Eastern Azarbaijan (2002), and Bushehr and Southern Khorasan (both in 2003), also resulted in substantial casualties [4]. Given that accidents and disasters often cause widespread damage and injuries, the demand for specialized medical care significantly increases in affected areas. To meet this critical need, a robust workforce is essential on the front lines. As a result, mobilizing hospital nurses has become a central strategy in disaster response [5]. Since Florence Nightingale cared for the wounded and sick people during the Crimean War, nurses have played a crucial role in disaster response [5]. Prior to disasters, nurses identify potential risks and help prevent and mitigate them by creating strategies and plans. They actively contribute to the development of policies and guidelines related to disaster management [6]. During disasters, nurses provide holistic care to those affected and injured, both within and outside the hospital. Additionally, after a disaster, nurses collaborate with other medical teams to ensure that hospitals recover effectively [6, 7]. Despite their prior experience in trauma, wound care, infection control, or presurgical care, hospital nurses may not always be adequately prepared or available to provide care during disaster situations [5].

Currently, Iranian nurses face several challenges, including a shortage of nurses and nursing instructors, an aging nursing workforce, and insufficient resources for hiring nurses [8]. Additionally, disaster nurses in Iran are available only at the doctoral level, and there are no bedside disaster nurses. Moreover, the lack of standardized tools specifically focused on assessing the individual preparedness of nurses during disasters has hindered the ability to recognize and measure the current state of preparedness among Iranian nurses. In some studies, researcher-made tools have been utilized to investigate and evaluate nurses’ preparedness for disasters. For instance, the Readiness Estimate and Deploy Ability Index (READI), which has been validated in Iran [9], emphasizes nurses’ capabilities and attitudes in dealing with disasters rather than solely assessing preparedness. The disaster preparedness questionnaire was designed by Ghanbari et al. (2011). It assesses nurses’ knowledge, attitudes, and performance, serving as a measure of their preparedness [10]. In contrast, the Japanese Disaster Nursing Readiness Evaluation Index (JENREI) is a valid and reliable instrument that specifically focuses on individual nurse preparedness [11]. Therefore, the development of a Farsi version of the Disaster Nursing Readiness Evaluation Index (F-ENREI) becomes essential for evaluating the readiness of Iranian nurses to cope with both natural and unnatural disasters. Consequently, our present study aimed to assess the psychometric properties of the Farsi version of this index.

Methods

Sample and setting

This cross-sectional study was conducted between 2023 and 2024 among nurses working in teaching hospitals affiliated with one of the medical universities in Tehran. The inclusion criteria were willingness to participate in the study and having more than six months of work experience. To estimate the sample size for exploratory factor analysis, two approaches were considered: the lowest total sample size and the sample-to-variable ratio [12]. Comfrey and Lee (1992) suggested that a sample size of 50 people is very poor, 100 poor, 200 moderate, 300 good, and 500 very good [13]. The next recommendation is that the minimum sample-to-item ratio in exploratory factor analysis should be 5:1 [14]. In this study, we used the sample-to-variable ratio.

Translation

The goal of translation is to achieve equivalence between the tool in the source language (SL) and the tool in the target language (TL) [15]. After obtaining permission from the main designer of the scale, Dr. Maeda, the translation was conducted in a forward-backward manner. Initially, the questionnaire was translated from English to Farsi by two independent translators, and the two Farsi versions were subsequently reviewed by the primary researcher and the author responsible for the revisions. Next, the Farsi version was translated back from Farsi to English by two other translators, and the reverse version was emailed to the main designer. The designer’s feedback was incorporated into the backward version, resulting in the final English version.

Measures

Data were collected using a demographic questionnaire and the Farsi version of the Disaster Nursing Readiness Evaluation Index which was originally designed by Maeda et al. (2018) in Japan [11]. The demographic questionnaire included information on age, gender, marital status, educational qualification, and work experience.

Maeda et al. (2018) assessed construct validity, through exploratory factor analysis, revealed six factors that accounted for 54.41% of the total variance. CFA confirmed the goodness of fit of the extracted model. Additionally, the internal consistency of these six dimensions was deemed acceptable based on the Cronbach’s alpha coefficient [11]. To mitigate social desirability bias, we distributed questionnaires online to nurses. To accomplish this, we inputted the questionnaire questions into Poresline (similar to Google Forms) and then shared the link with nursing managers at hospitals through the corresponding author. The managers then shared the link through Telegram and WhatsApp groups. At the outset of the online form, we clearly stated the study objectives, and respondents completed it if they agreed. Online data collection occurred in December 2023 and January 2024.

Face validity

Formal validity, also known as “validity in terms of respondents,” refers to the extent to which an instrument accurately measures what it intends to measure. To assess the clarity of instructions, item content, and answer format, we conducted a pilot test of the Farsi pre-final version of the instrument with 5 nurses. During this cognitive debriefing stage, the nurses read the questionnaire aloud and identified any ambiguous or unclear wording. They also provided suggestions for improving each item [16].

Content validity

Content validity means “validity in terms of researchers” [15]. Content validity refers to the extent to which an assessment instrument accurately represents the intended construct or content domain [17]. This ensures that the items included in the instrument adequately cover the relevant aspects of the concept being measured. In your study, the Farsi version of the questionnaire was administered to five experts (three with PhDs in nursing and two with PhDs in disaster health). These experts reviewed the content of the items and provided suggestions for improving clarity by revising the wording.

Construct validity

Analyses were performed using Jamovi software version 2.4.14 and Amos 26. Prior to the analysis, item analysis was conducted to determine item-to-total correlations and inter-item correlations [16]. The Kaiser–Meier–Olkin (KMO) index and Bartlett’s test were also performed. To assess construct validity, exploratory factor analysis (EFA) was carried out using the maximum likelihood method. This method yields more generalizable and repeatable results [14]. Factors were then rotated for better interpretation, as unrotated factors can be ambiguous [14]. The purpose of rotation is to achieve an optimal simple structure in which each variable loads on the fewest possible number of factors while maximizing the number of high loadings on each variable [18]. Latent factors were extracted based on parallel analysis, and items with a factor load of less than 0.32 were excluded from any factor. To assess the reliability of the questionnaire, internal consistency was examined. Internal consistency evaluates whether all parts of an instrument measure the same characteristic [19].

CFA was conducted to validate and verify the factor structure derived from EFA. The following indices were used to assess the model fit: minimum discrepancy function by degrees of freedom divided (CMIN/DF), comparative fit index (CFI), incremental fit index (IFI), goodness-of-fit index (GFI), Tucker‒Lewis index (TLI), and root mean square error of approximation (RMSEA) [20]. Cronbach’s alpha and McDonald’s omega coefficients were utilized to evaluate internal consistency. The acceptable values for reliability are typically above 0.70 [16].

Ethical considerations

The study protocol was approved by the Ethics Committee of the Faculty of Nursing and Midwifery at Shahid Beheshti University of Medical Sciences (IR.SBMU.PHARMACY.REC.1402.085). Before collecting the data, the research objectives were thoroughly explained to the nurses. It was emphasized that their participation in the study was entirely voluntary, the questionnaires are anonymous, and the results are used solely for research purposes. Informed consent to participate was obtained from all of the participants in the study.

Results

A total of 200 nurses (62 men and 138 women) participated in the study. The mean age of the nurses was 33.4 (SD = 7.88) years, and they were aged between 20 and 64 years. The mean work experience of the nurses was 9.51 (SD = 7.54) years. Most of the participating nurses were married (53.8%), held a bachelor’s degree (72.7%), and were formally employed (39.7%).

Validity results

In terms of face validity and content validity, slight changes were made to the wording of the two items. Prior to conducting exploratory factor analysis, an item analysis was performed, which led to the removal of items 17 and 27 due to low inter-item correlation. During the exploratory factor analysis, five factors were extracted using the maximum likelihood method and quartimax rotation, which collectively explained 39.7% of the total variance. Items 16 and 29 were not assigned to any specific factor. However, Item 14 loaded onto both factors 1 (with a factor loading of 0.378) and 5 (with a factor loading of 0.323), indicating cross-loading. Consequently, item 14 was removed. To achieve an optimal simple structure, various rotations were employed in the analysis. The most interpretable factor structure, without item cross-loading and factor loadings above 0.30, was associated with quartimax rotation.

The first factor included seven items (items 7 to 13), as in the original version. This factor accounted for 10.63% of the total variance. Items 12 and 8 had the highest (0.879) and lowest (0.319) factor loadings, respectively. The second factor consisted of items 30 to 37 and was also included in the original version. The lowest and highest factor loadings for the items of this factor were related to item 32 (with a factor loading of 0.771) and item 31 (with a factor loading of 0.444), respectively. This factor accounts for 10.61% of the total variance. The third factor included items 15, 18, 19, 20, 21, 22, 25, 26, and 28. In the original version, items 15 to 22 were related to one factor, while items 25 to 29 were related to another factor. This factor was a combination of two factors from the original version. Items 16 and 29 were not assigned to any factor, while items 17 and 27 were excluded from the item analysis.

The factor loadings for items 25 (0.387) and 20 (0.695) were the lowest and highest, respectively. This factor explained 7.87% of the variance. The fourth factor included items 1 to 6, consistent with the original version. The factor loadings of these items ranged from 0.506 to 0.580. This factor explained 6.26% of the total variance. The fifth factor consists of two items, 23 and 24, which are also in line with the original version. The factor loadings for these two items were 0.615 and 0.717, respectively. This factor explained 4.33% of the total variance. (Table 1).

Table 1 The results of exploratory factor analysis

Reliability results

The Cronbach’s alpha and McDonald’s omega coefficients for the whole questionnaire were 0.890 and 0.891, respectively. The internal homogeneity was related to the second factor, whose MacDonald Omega and Cronbach’s alpha coefficients were 0.845 and 0.849, respectively. Additionally, the Cronbach’s alpha and McDonald’s omega coefficients for the entire questionnaire were 0.896 and 0.896, respectively. The lowest and highest scores were related to the first factor (2.88 ± 0.98) and the third factor (4.75 ± 0.40), respectively (Table 2).

Table 2 Characteristics of extracted factors in exploratory factor analysis

The mean score of the questionnaire was 125.04 ± 18.20. The correlation between the extracted factors varied between 0.11 and 0.53. Factors 4 and 3 had the lowest correlation (0.11), and factors 1 and 4 had the highest correlation (0.53). More details are provided in Fig. 1.

Fig. 1

Correlation heatmap

The results of the confirmatory factor analysis showed that the extracted model had a good fit: CMIN-680.402, DF = 448, CMIN/DF = 1.519, CFI = 0.889, GFI = 0.834, TLI = 0.877, IFI = 0.892, RMSEA = 0.051 (90% CI: 0.043–0.059) (Fig. 2).

Fig. 2

The final model

There was a significant positive correlation between the average score of the questionnaire and both average age (r = 0.292, p < 0.001) and work experience (r = 0.306, p < 0.001). Additionally, factors one, two, and four were correlated with the average age and work experience of nurses, while factors three and five were not correlated with these two variables. Furthermore, the mean total disaster preparedness score of male nurses was significantly greater than that of female nurses (131.70 ± 16.30 vs. 122.12 ± 18.27; p < 0.001). The results of the analysis with independent t tests showed that the average scores of factors two (34.59 ± 5.78 vs. 30.03 ± 8.07, p < 0.001), four (3.84 ± 26.01 vs. 4.52 ± 24.35, p = 0.014) and five (± 7.49 2.19 vs. 6.69 ± 2.33, p = 0.025) were significantly greater for male nurses than for female nurses. There was no relationship between the readiness score and marital status, type of employment or educational qualification.

Discussion

This study aimed to investigate the psychometric characteristics of the Farsi version of the Disaster Nursing Readiness Evaluation Index (F-DNREI). To evaluate construct validity in the current study, exploratory factor analysis was employed, resulting in the extraction of five factors, whereas the original study identified six factors [11]. Items 17 (I can use humor in my words), 27 (I can admit my mistakes), 14 (I can work with communication equipment), 16 (I can often talk to my colleagues) and 29 (I have good listening skills) are absent from the Farsi version. These differences may be attributed to demographic variations and the differing perceptions of Iranian and Japanese nurses regarding these issues. Additionally, the behaviors and skills described in these items might not align with Iranian cultural norms. Despite the identical constituent items across the two versions (Farsi and Japanese), the factors were named based on the original version.

The first factor, practical skills for disaster response, explained the highest percentage of total variance. This factor encompasses decision-making based on priorities, responding to mass casualties, triage experience in mass casualties, and care for radiological, chemical, and nuclear injuries, as well as evacuation and disinfection procedures. In the Farsi version, the highest factor loading was related to item 12, which pertains to deciding on priorities and tasks. Conversely, in the original version, the highest factor loading was related to item 7, which addresses appropriate responses to collective casualties. Triage aims to save lives by directing resources to injured individuals with greater medical needs [21]. Therefore, prioritizing medical requirements can reduce mortality among injured individuals [22]. In general, triage is effective when prioritizing patients with a higher probability of survival, while allocating scarce resources to patients with a low chance of survival is suboptimal in disaster situations [23]. Nurses in disaster scenarios often face ethical challenges because they must make decisions about priorities when forced to choose between different options [24]. In the Farsi version, item 14 was removed due to cross-loading, resulting in one fewer item compared to the original version. This difference can be attributed to the lack of skill and proper training in the use of communication equipment among Iranian nurses. This deficiency is rooted in international sanctions and the absence of up-to-date and efficient equipment in the country’s emergency rooms. Interestingly, the average score for this factor did not differ significantly between male and female nurses. Regardless of gender, nurses appear to exhibit similar clinical skills during disasters.

The second factor, adaptability to stressful situations, consists of 8 items that assess nurses’ ability to adapt and prepare for stressful situations during disasters. These challenges include coping with the death of victims, their own potential mortality, working in harsh weather conditions, dealing with limited privacy, and managing long-term work demands. During disasters, nurses encounter several stressors, including fear of the unknown, lack of experience in handling unfamiliar situations, mistakes due to incomplete information, and insufficient professional skills [25]. A chaotic and unpredictable environment, long working hours, limited resources, and unfamiliar surroundings further contribute to emotional and psychological stress among nurses during disaster response, making the provision of care more complex [26, 27]. The constituent items of this factor were consistent between the Persian and English versions of the questionnaire, with only differences in item arrangement. Notably, in the Persian version, item 32 (readiness to work in harsh weather conditions) had the highest factor loading, while in the English version, item 30 (readiness for the death of victims) showed the strongest association. Interestingly, male nurses scored higher on this factor than female nurses. This finding may be attributed to women’s potentially lower physical endurance for extended hours in harsh weather conditions and their emotional sensitivity when facing patient deaths and privacy concerns.

The third factor, termed communication and cooperation skills for teamwork, originally consisted of two separate factors: communication skills and teamwork. However, in the Farsi version, these two factors were merged, resulting in the combined name for this factor. In addition to providing clinical and technical care, nurses require diverse skill sets during disasters. These skills include team leadership, problem solving, resource management, and effective teamwork [24, 28, 29]. Continuous development of risk management competencies is essential for nurses to deliver high-quality nursing care before, during, and after disasters [30]. Teamwork plays a crucial role in disaster response for nurses. Without effective teamwork, patient safety can be compromised for various reasons, such as inadequate information exchange, misinterpretation of instructions, unclear phone orders, and oversight of critical situational changes [31]. Notably, the Farsi version of the questionnaire lacked four items (16, 17, 27, and 29). Items 16 and 29 were not associated with any specific factor, while items 17 and 27 were excluded from the item analysis.

The fourth factor focuses on emergency nursing skills. The items within this factor pertain to critical abilities such as managing fatal injuries, handling shock, following ABC protocols, practicing infection control, and ensuring proper airway management. During disasters, head injuries, chest injuries, and major vascular trauma are the primary causes of death [32]. Most fatalities occur either at the accident scene or within the first 4 h after a patient arrives at the trauma center [33]. To optimize outcomes and minimize the risk of undiagnosed injuries, systematic evaluation of injured patients is essential [34]. Trauma care should be organized around key concepts, including rapid assessment, triage, resuscitation, diagnosis, and therapeutic intervention [35]. While epidemics are not directly linked to natural disasters, the interplay of factors such as large population movements, environmental changes, victim conditions, and vulnerability to pathogens can increase the risk of an epidemic [36]. Diarrheal diseases have emerged as a significant cause of death in disaster settings and camps. Contaminated water sources, water pollution during transportation or storage, and shared cooking utensils contribute to this health challenge [37]. In the original version of the questionnaire, the highest factor loading was associated with item 1 (familiarity with shock), while in the Farsi version, item 2 (managing fatal injuries) had the strongest correlation. Interestingly, compared with female nurses, male nurses scored higher on the total score of items constituting this factor.

The fifth factor, effective coping with daily stress, included two items that referred to “coping with daily stress related to the family and the nurse’s finances”. In a qualitative study by Moghaddam et al. (2014), nurses stated that they needed to ensure that their families were supported and cared for during the nurses’ deployment [38]. In a study by French et al. (2002), nurses’ needs and the conflict between family and work commitments were highlighted. During Hurricane Floyd, nurses expressed a strong preference for being with their families and experienced heightened anxiety when separated from their loved ones. In fact, some nurses even made the difficult decision to stay with their families, even if it meant losing their jobs [39]. O’Boyle et al. (2006) suggested that incorporating support for nurses’ families could be an effective way to enhance nurses’ resilience during disasters [40]. The mean score for this factor was greater for male nurses than for female nurses.

The fit indices in the confirmatory factor analysis for the Farsi version were similar to those of the original version: the CFI was 0.889 in Farsi versus 0.897 in the original version; the RMSEA was 0.051 in Farsi versus 0.058 in the original version; and the GFI was 0.834 in Farsi versus 0.858 in the original version. Additionally, the reliability of the entire tool and all subscales exceeded 0.70, as assessed by both Cronbach’s alpha and McDonald’s omega coefficients. In the original version, the overall reliability of the tool was 0.93, while the reliability of the subscales ranged from 0.83 to 0.93 [11].

Limitations

The authors acknowledge that it is unconventional to perform both exploratory and confirmatory factor analyses on a methodological sample. However, due to limited access to nurses, confirmatory factor analysis was conducted using the same sample. The high patient load in Tehran hospitals (the capital of Iran) has resulted in nurses having heavy workloads and reluctance to complete research questionnaires. Although the online questionnaire link was provided to nurses for two months during this study, only a small number of them completed it. Future research should consider conducting studies in different medical centers with larger sample sizes to address these limitations and enhance the generalizability of findings obtained from this scale.

Conclusions

Given that Iran frequently faces both natural and unnatural crises and disasters, it is essential for Iranian nurses, who are among the first responders, to be equipped to handle disasters. Assessing nurses’ preparedness will inform future educational programs, necessitating valid and reliable measurement instruments. The availability of a validated and reliable Farsi version of the disaster nursing readiness evaluation index (F-DNREI) allows for effective evaluation of nurses’ readiness and the implementation of appropriate measures based on the findings.