Background
Traumatic grief in healthcare professionals, particularly in nurses, constitutesa complex emotional response after repeated exposure to traumatic situations in their work environment [1]. In the healthcare context, events related to grief are common. Despite death being considered an inevitable part of an individual’s life process, healthcare professionals may be exposed to various traumatic events, such as patient death, verbal or physical abuse from patients or family members, and making difficult decisions that may have serious consequences [2].
The bereavement experience among healthcare professionals is significantly influenced by sociodemographic factors such as age and experience, as well as by contextual elements such as workload, job insecurity, and type of care unit [3‐5]. These factors determine how individuals process grief, particularly in high-stress settings such as pediatric and intensive care units. Job insecurity can also lead to increased stress, which complicates the grieving process. In addition, the care unit where the profession is practicing also plays a crucial role; for example, pediatric ICU practitioners face unique challenges due to societal expectations around infant mortality [6]. Prior training of healthcare professionals is crucial to equipping them with psychological and emotional tools that foster effective coping strategies during bereavement [7‐8].
Anzeige
When this process becomes traumatic, can have a significant impact on nurses’ mental and physical health, which, in turn, may compromise their ability to provide quality patient care [9].
It can even be a devastating and prolonged experience, especially in specific situations such as the context of the Covid-19 pandemic, in which healthcare professionals have been on the front line of the fight against the disease and have experienced an unprecedented amount of loss and trauma in their work environments [10]. In particular, nurses have been exposed to a high risk of infection and have endured an overwhelming workload under stressful and dangerous conditions [11]. Traumatic grief in nurses during the Covid-19 pandemic is a significant problem that needs to be studied to better understand its impact and develop effective interventions to support the mental health and well-being of these critical professionals.
Prevalence rates of professional traumatic grief have increased significantly following the Covid 19 pandemic, intensifying the pre-existing complications of pathological grief and increasing the risk of complicated or prolonged psychological distress. Indeed, numerous investigations have highlighted the increase in the prevalence of professional bereavement, with nursing staff in particular showing the highest rates [9, 10, 12‐14]. The high mortality rate, ethically challenging work conditions, lack of support and disorganization by hospital institutions, job insecurity, along with the ongoing and substantial number of deaths, and the nearly non-existent support network and resources to cope with the situation, could be considered potential contributing factors to this increase [15‐19]. In addition, recent research seems to consider COVID-19 as a potential factor of increased complicated grief in health professionals as the consequences of the pandemic, widely addressed in numerous studies [19‐21], have accentuated severe and persistent grief in many nurses, probably influenced by personal characteristics and the rapidity with which the sudden deaths of their patients occurred.
For bereavement experts, sensitization and subsequent awareness of the specific risks of complicated bereavement associated with the pandemic represent urgent issues to be addressed [19, 22]. In addition, while the context of the pandemic is undeniably a significant and challenging experience for healthcare professionals, it is just one facet of the broad spectrum of grief encountered in the field of healthcare. Recent research has illuminated the versatility and applicability of the grief scale in various other healthcare contexts and situations, shedding light on the unique challenges faced by professionals in environments such as the Intensive Care Unit (ICU) and pediatric care [23‐24]. Beyond the COVID-19 pandemic, healthcare workers in ICU settings routinely face profound grief as patients succumb to severe illnesses or traumatic injuries. These professionals often carry the emotional burden of witnessing patients’ struggles for survival, making life-or-death decisions, and consoling grieving families.
Anzeige
The emotional burden of these experiences should not be underestimated, and the grief scale helps in understanding and addressing the complex nature of grief that may arise. Similarly, pediatric care presents its own set of challenging and painful circumstances.
Healthcare professionals working with children and their families may find themselves grappling with the emotional turmoil of losing young patients, which is an exceptionally challenging aspect of their profession. The applicability of the grief scale in this context underscores its importance in recognizing and supporting healthcare professionals who are often overlooked in discussions about grief and loss.
The bio-psycho-social consequences, the threat perceived by the nursing population, and their association with sociodemographic and contextual factors require the use of standardized language for the accurate identification of professional traumatic grief. A clearer understanding of its predictors and specific symptomatology is essential to facilitate prevention, follow-up, and treatment for affected individuals.
The recent proposal for a nursing diagnosis defines professional traumatic grief as “a state of physical, psychological and social distress experienced by health professionals and caused by prolonged exposure to multiple sudden deaths in the work context” [25]. The symptomatology includes anxiety, fear and other emotional states such as feelings of helplessness, insomnia, psychological distress, exhaustion, depressive symptoms, somatization and feelings of stigmatization and frustration, among others [9, 20, 26‐29]. Some manifestations of post-traumatic stress disorder have even been observed in these health professionals [30, 31]. These consequences can negatively impact the physical and emotional state of these workers, leading to decreased job performance and an increased risk of errors or omissions.
The use of measurement, identification and diagnostic instruments and/or scales is essential to mitigate the threat to personal identity, foster coping resources in professionals, help minimize negative self-evaluation of loss, and improve the quality of life and health care of this at-risk population. Although there are instruments designed to measure prolonged or pathological grief, suchas the Inventory of Complicated Grief [32], the Revised Texas Grief Inventory [33] or the Grief Evaluation Measure [34], these tools have important limitations. Most of these scales, widely used in the literature, are not adapted to the construct of professional traumatic grief, as their items typically refer to close personal relationships with the deceased, leaving out specific contexts such as that of health professionals. In addition, most of these instruments do not assess all the criteria proposed for the disorder according to the diagnostic classifications, and those that do are highly demanding. Most have been validated in populations exposed to a single traumatic event, limiting their generalizability to individuals who have experienced multiple traumas. Finally, many of them have not been developed or validated with a Spanish population, raising concerns about their cultural applicability.
In the specific case of health professionals, although there are scales that measure attitudes or competencies related to coping with death and assessing their ability to handle a patient’s death [35‐37], there is no specific tool to assesses the symptomatology of traumatic grief in those who experience multiple deaths of patients suddenly, unexpectedly and unfairly.
Therefore, considering that there are no quantitative instruments designed to measure professional grief in the healthcare context with psychometric guarantees of reliability and evidence of validity, further research on the validation of a specific scale seems necessary since the lack of specific and robust screening tools to specifically identify the symptomatology of professional traumatic grief and the use of traditional assessment tools may lead to under- or overdiagnosis.
In addition, the combined administration of different quantitative scales and/or qualitative interviews takes a long time. This can be addressed through the development and psychometric validation of a shortened version of a specific tool for assessing professional grief symptomatology. The first research on the construct to be measured was the study into the Elaboration and Content Validation of the Inventory of Symptoms of Professional Traumatic Grief (ISDUTYP) as a specific and adequate measuring instrument to identify possible risk factors and a symptomatology of professional traumatic grief in nursing professionals [1]. This preliminary research employed a mixed-methods approach to design and validate the content of a scale for identifying individuals with professional traumatic grief. The study was structured into two phases: Phase I, Instrument Design, and Phase II, Instrument Content Validation. Ultimately, the final version of the ISDUTYP was successfully developed, comprising a total of 25 items.
Anzeige
Following this approach, this article presents a psychometric analysis whose objective is to evaluate the psychometric properties, including reliability, construct validity, criterion validity, convergent and discriminant validity of the ISDUTYP in a sample of nursing professionals in order to provide a widely used instrument in the health sector for the early identification of individuals with professional traumatic grief.
The specific objectives of this study are:
(1)
To analyze the sociodemographic characteristics of a population undergoing professional traumatic grief. This objective aims to contextualize the impact of professional traumatic grief within different demographic and professional profiles, providing a basis for understanding differences in symptomatology.
(2)
To evaluate the psychometric properties and validity of the ISDUTYP in a sample of nurses experiencing professional grief. This includes an assessment of the scale’s reliability, construct validity, criterion validity, and convergent and discriminant validity, ensuring its robustness as a tool for early identification and intervention.
The hypotheses formulated to achieve these objectives are as follows:
Anzeige
-
Individuals whose professional traumatic grief symptomatology is more pronounced are hypothesized to employ active coping strategies. This hypothes is aligns with the second objective by examining the scale’scapacity to reflect the connection between grief and adaptive behaviors. Research indicates thatgreater training in the health field can reduce negative reactions to traumatic situations, such as patient death, by promoting the use of problem-focused strategies to manage grief, particularly in high-stress environments [38‐40].
-
Professional traumatic grief symptomatology will be more accentuated in individuals who have greater stress, anxiety and depression. This supports the assessment of the scale’s convergent and discriminant validity, given that these psychological states function as catalysts of emotional and cognitive responses associated with bereavement [19, 41]. Similar results were found in a recent study during the COVID-19 pandemic [42].
-
Symptomatology of professional traumatic grief will be more pronounced in individuals who score higher on the Impact on Event Scale (IES-R). This hypothesis fulfills the second objective by confirming the ability of ISDUTYP to align with recognized indicators of trauma, as elevated IES-R scores are associated with increased risks of post-traumatic stress and mental health problems [43]. In the context of professional traumatic grief, it is reasonable to assume that those individuals who show a more intense response to traumatic events, as reflected in high scores on the IES-R, would also experience more severe symptoms of grief.
-
The symptomatology of professional traumatic grief will be more pronounced in individuals with a high level of compassion fatigue and a low level of job satisfaction. Compassion fatigue, characterized by profound emotional exhaustion, is often related to symptoms of bereavement and traumatic stress. By relating bereavement to emotional exhaustion and job dissatisfaction, this hypothesis reinforces the scale’s ability to capture complex emotional constructs [11, 44, 45].
-
Individuals with higher scores on the social support scale will show less professional traumatic grief symptomatology. This hypothesis aligns with the second objective by highlighting the protective function of social support, particularly among peers in healthcare environments, in alleviating bereavement symptoms [46, 47].
-
Professional traumatic grief symptomatology will be more pronounced in individuals who score higher on the Posttraumatic Growth Inventory. This shows the complexity of reactions to bereavement and trauma, as growth, usually seen as adaptive, may reveal maladaptive coping strategies that aggravate bereavement symptoms [48]. Moreover, events such as the COVID-19 pandemic can exacerbate this phenomenon, leading to maladaptive responses and increasing the symptomatology of complicated grief [49].
-
The symptomatology of professional traumatic grief will be more pronounced in individuals who score higher on the PHQ9 scale. This hypothesis supports the second objective by confirming the ability of the scale to show the effect of depressive symptoms on occupational traumatic grief [42].
Method
Design
Descriptive cross-sectional scale validation.
Participants
The study population used to analyse the validation of the scale consisted of nurses who had worked during the Covid-19 pandemic in the Andalusian Health Service (SAS) in the provinces of Andalusia. The inclusion criteria were: (1) Nurses who worked and/or work in Andalusian Health Service hospitals during the Covid-19 pandemic; (2) Individuals between 18 and 65 years of age; (3) Nurses who agreed to participate in the study after signing the informed consent form. Exclusion criteria were: (1) Nurses diagnosed with a mood or anxiety disorder previously and during the COVID-19 pandemic; (2) Individuals refusing to participate in the study. The sample was selected according to the inclusion and exclusion criteria.
The sample was selected using a convenience sampling strategy combined with a proportional stratified approach to ensure representation across specific strata. In relation to the sample size, different calculations were made according to the objectives. For the validation of the scale, a 10:1 ratio (subjects per item) was adopted to ensure a sample large enough to perform the factor analysis. In this case, since the scale has 25 items, a sample size of at least 250 individuals was considered. On the other hand, for the calculation of the sample size for convergent and discriminant validity, the hypotheses proposed were taken into account and a calculation was made for each of them, leaving us with the largest sample size. Since there is no literature comparing grief with the other constructs to be assessed in healthcare personnel, the final sample size was calculated when the study began, based on a pilot sample of 50 individuals. For this, a confidence level of 95% was determined, and a power of 80% in all cases. The calculation was performed from the sample calculation for the correlation coefficient, in which the total score of the Grief scale is compared with the rest of the scales, or sample calculation for mean differences, when the scale is compared with a qualitative sociodemographic variable (Table 1). Taking into account the calculations made in Table 1, the sample size required for our study was at least 908 subjects. After calculating the total sample size, a representative sample was taken from each province, with the number of participants from each province being proportional to the total number of nurses working in hospitals in each one.
Anzeige
Table 1
Sample size calculations
ISDUTYP Global score | Sample size | ||
|---|---|---|---|
Mean (SD) / Rho | |||
Gender | Men | 77.50 (26.37) | 908 |
Women | 82.73 (29.76) | ||
Job experience | Years | -0.153 | 262 |
Unit working during the pandemic Covid | Covid plant | 89.67(30.92) | 72 |
No Covid plant | 84.67(32.8) | ||
Covid ICU | 88.27(31.4) | ||
ICU | 103.83(24.5) | ||
Emergency Room | 61.25(13.7) | ||
Operating room | 72.5(25.8) | ||
Frequency of exposure to deaths | Never | 86(31.5) | 44 |
Rarely (once a week or less) | 74.95(23.1) | ||
Quite a bit (2 to 4 times a week) | 97.23(19.5) | ||
A lot (5 or more times per week) | 117.25(40.6) | ||
Preparation towards death as a professional | Poor | 82.67(34.2) | 328 |
Moderate | 87.74(32.47) | ||
Satisfactory | 97.75(27.8) | ||
Good | 88.17(23.9) | ||
COPE-28 | FACET 1 | -0.388 | 39 |
DASS-21 | Depression | 0.759 | 9 |
Anxiety | 0.799 | 8 | |
Stress | 0.729 | 10 | |
IES-R | Intrusion | 0.732 | 10 |
Avoidance | 0.658 | 13 | |
Hyperarousal | 0.716 | 11 | |
PROQOL | Compassion fatigue | 0.780 | 9 |
Compassion satisfaction | -0.483 | 25 | |
Burnout | 0.487 | 25 | |
Social support | Laboral emotional support | -0.276 | 80 |
Non-Laboral emotional support | -0.090 | 762 | |
Laboral instrumental support | -0.125 | 394 | |
PTGI | Relationship with others | 0.166 | 223 |
New possibilities | 0.174 | 203 | |
Personal strength | 0.300 | 67 | |
Spiritual change | 0.152 | 266 | |
Life appreciation | 0.166 | 223 | |
Global post-traumatic growth index | 0.201 | 151 | |
PHQ9 | Global score (severity of the clinical picture) | 0.691 | 11 |
Instruments
A structured questionnaire was used to collect the following data: sociodemographic characteristics (sex, age, place of residence, province of origin, marital status, number of children, academic degree, job position, job experience, unit where the pandemic occurred, frequency of exposure to deaths, preparation for death as a professional, perception of being trained to cope with death, support systems of the institution, being under medication), the ISDUTYP and other additional measures, which were included to characterize the sample and for use in the validity analyses, described below.
ISDUTYP: assesses symptomatology of professional traumatic grief on a 25-item scale with a Likert-type response format of 1 to 7 points (1 = Never and 7 = Always), where 1 indicates better status and 7 indicates worse status. The overall score of the scale is the sum of the scores of all items, leading to an overall score of 25 (best condition) to 175 (worst condition).
COPE-28: evaluates the characteristics of the person when coping with responses to stressful situations [50]. It is composed of 28 items with a frequency scale response format, with scores between 0 and 3 (0 = at no time and 3 = all the time). The scale is composed of 14 subscales, each composed of two items: active coping, planning, emotional support, social support, religion, positive reinterpretation, acceptance, denial, humour, distraction, self-blame, behavioural disengagement, relief, and substance use. To obtain the total score for the different types of coping, the scores for the items that make up each type of coping are summed and divided by the number of items that make up each type. Thus, the scores for the coping types range from 0 to 3 points, with higher scores indicating greater use of the strategy.
DASS-21: A short version of the Depression, Anxiety and Stress Scale (DASS-21) [51]. The three-dimensional self-report scales assess the presence and intensity of the affective states of depression, anxiety, and stress. Each item is answered according to the presence and intensity of each symptom on a Likert-type response scale from 0 to 3 points. Each scale consists of seven items and its total score is calculated with the sum of the items belonging to that scale and varies between 0 and 21 points.
IES-R: This scale was created by Horowitz, Wilner and Alvarez [52], and validated in Chile by Caamaño et al. [53]. It measures the symptomatology of post-traumatic stress. It consists of 22 items divided into three subscales (8 for re-experiencing, 7 for avoidance and 7 for hyperarousal). This is a Likert-type instrument for the evaluation of symptomatology intensity (from 0 = Not at all to 4 = Extremely). With a maximum total scale score of 88 points, the instrument has been used in several studies analysing the impact of disasters and emergencies.
PROQOL-VIV [54]: The professional quality of life questionnaire consists of 30 items that are answered on a Likert-type scale with 6 response alternatives ranging from never (0) to always (5). The questionnaire is divided into three subscales: compassion satisfaction (the mean score on this scale is 37, a score higher than 42 considered high compassion satisfaction while low compassion satisfaction corresponds to scores below 33); burnout (the mean score is 22, and scores higher than 27 symbolize a high risk of suffering from it, while scores below 18 imply a low risk); and secondary post-traumatic stress or compassion fatigue (the mean of this score is 13).
Social support scale [55]: The scale distinguishes between emotional and instrumental social support from various sources. It is a questionnaire for individual application and identifies the main sources of occupational and non-occupational social support for the worker. Work-related emotional support is assessed by items a1, b1, a2, b2 and b4, non-occupational emotional support by items c1, d1, c2 and d2, and instrumental job support by items a3, b3 and a4. The score of each block is obtained by the sum of the values of its items. Thus, the maximum score is 15 for emotional job support, 12 for emotional non-job support and 9 for instrumental job support.
Posttraumatic Growth Inventory: It was developed by Tedeschi and Calhoun [56] and evaluates the degree of positive changes perceived by the individual after an adverse event. The instrument has a six-alternative Likert response mode where zero equals no change and five equals a very important change; the higher the score, the greater the perceived change. It consists of 21 items and has 5 factors: Relationship with others (items 6, 8, 9, 15, 16, 20 and 21), New possibilities (items 3, 7, 11, 14 and 17), Personal change (items 4, 10, 12 and 19), Spiritual change (items 5 and 18) and Appreciation of life (items 1, 2 and 13), where the total value (sum of all items) corresponds to the post-traumatic growth index.
PHQ-9 [57]: Consists of 9 items that are statements of the clinical criteria of the DSM-IV, which in addition to diagnosing depression helps to discover the severity of the clinical condition. The response format is given by a Likert scale with the following 4 response options and scores: not at all = 0, several days = 1, more than half of the days = 2, almost every day = 3, the sum of which indicates the severity of the depressive disorder, the total score ranging from 0 to 27.
Procedure
The study participants were contacted through the SAS nursing care coordinators. They were sent an e-mail with the invitation to collaborate in the study. The e-mail explained the objectives of the study and the self-administered questionnaire was attached using a link made with GoogleForms. In this link, the participants were informed that by answering the questions they would automatically give their informed consent. This study was conducted in accordance with ethical principles outlined in the Declaration of Helsinki and was approved by the appropriate ethics committee.
Psychometric testing and statistical analysis
A descriptive analysis was carried out, showing absolute and relative frequencies in the case of the qualitative variables. For the quantitative variables, central tendency measures (mean and median) and dispersion (standard deviation (SD) and interquartile range (IR)) are reported. The distribution of the latter was checked via the Kolmogorov-Smirnov test.
Validity analysis
In order to analyse the psychometric properties and validity of the instrument, the relevant analyses were conducted to study reliability and validity.
Construct validity
First, the construct validity of the scale was assessed to analyse its factorial internal structure, through the application of exploratory factor analysis (EFA), in order to identify the factors capable of explaining the maximum amount of information collected in the data. The variables included in this analysis were the 25 items of the scale. The Principal Component Method was used to extract the factors. The scree plot and the criterion of eigenvalues over 1 were used to select the number of factors. A varimax rotation was subsequently applied. The factor loadings were used to decide the items that comprised each factor. Finally, an appropriate name was assigned to each factor, within the clinical context of the analysed construct.
The global score of the scale was the sum of the scores of all the items. Each of the 25 items are scored from 1 (best status) to 7 (worst status), leading to a global score from 25 (best status) to 175 (worse status). Based on the results of the factorial analysis, a specific score is proposed for each factor, consisting of the sum of the scores of the items that comprise it, divided by the total number of items. This division responds to the need to be able to compare the scores of the subscales, regardless of the number of items that comprise them. These scores range from 1 (best status) to 7 (worst status).
Subsequently, the characteristics of each item in the questionnaire, the subscales, and the overall score were analysed, calculating the mean, standard deviation, median, minimum, and maximum. Furthermore, we analysed the floor and ceiling effect as the percentage of people with the minimum and maximum score, respectively. These analyses were performed for the 25 items, the global score, and the subscales scores.
Reliability
To assess the questionnaire’s reliability, its internal consistency was evaluated, which was checked with Cronbach’s Alpha test for the global score and each subscale. Values above 0.70 are generally considered indicative of good internal reliability and suggest that the instrument can be used for group comparisons. Additionally, the item-scale correlations were examined using Spearman’s Rank Correlation Coefficient (rho), considering that correlations of 0.35 or less indicate low correlations.
Criterion validity andconvergent and discriminant validity
Below is a description of the analyses conducted for criterion validity and convergent/divergent validity. We did not measure any gold standard for the criterion validity of the whole scale. Instead, we determined the main concept measured by each item and analysed its correlation with the scale that measured that construct, using Pearson’s or Spearman’s correlation coefficient. For the convergent and discriminant validity, we analysed the relationships of our scale (and subscales) with the remaining scales and variables collected, according to the hypotheses proposed and described previously. For this, the Mann-Whitney U test and the Spearman Rank Correlation Coefficient (rho) were used. A significance level α = 0.05 was established for all the analysis.
All analyses were performed using IBM’s SPSS© (version 26.0) software for Windows (IBM Corp., Armonk, NY, USA).
Results
Sociodemographic characteristics of participants
A total of 930 people took part in the study. Their main characteristics can be seen in Table 2. They were mostly women (78.4%), with a mean age of 37.1 years. They were mostly single (44.1%) or married (39.5%) and the majority had 2 children or less (86.4%), with 31% standing out as having no children. In addition, almost all of them were nurses (77.3%), and 76.1% were base nurses, with around 13 years of experience. Seville was the province providing the highest number of participants (22.1%). Most of the subjects worked on Covid (40.6%) or Covid ICU (33.8%) wards.
Preparation for and support for death
The majority of participants (43.1%) reported having a ‘moderate’ level of preparedness to cope with death in their professional environment. However, a significant percentage of the total respondents (53.5%) indicated that they did not feel capable of handling situations related to death in their professional context. On the other hand, slightly more than half of the participants (51.7%) acknowledged the presence of support systems, while a substantial portion (48.3%) stated that they did not perceive such support systems (Table 2).
Table 2
Characteristics of the sample
Variable | Category/Unit | n | % | |
|---|---|---|---|---|
Sociodemographic | ||||
Gender | Male | 201 | 21.6 | |
Female | 729 | 78.4 | ||
Age (N = 927) | Years Mean (SD) Median (IR) | 37.1 (11.9) 34 (22) | ||
Province of residence (N = 929) | Málaga | 176 | 18.9 | |
Sevilla | 205 | 22.1 | ||
Cádiz | 148 | 15.9 | ||
Almería | 72 | 7.7 | ||
Jaén | 79 | 8.5 | ||
Córdoba | 94 | 10.2 | ||
Huelva | 55 | 5.9 | ||
Granada | 101 | 10.8 | ||
Marital status (N = 929) | Single | 410 | 44.1 | |
Cohabitant | 101 | 10.9 | ||
Separated | 12 | 1.3 | ||
Married | 367 | 39.5 | ||
Widow(er) | 11 | 1.2 | ||
Divorced | 28 | 3 | ||
Number of children (N = 583) | 0 | 185 | 31.7 | |
1 | 94 | 16.1 | ||
2 | 225 | 38.6 | ||
3 | 62 | 10.6 | ||
4 | 12 | 2.1 | ||
5 | 5 | 0.9 | ||
Academic degree (N = 889) | Nurse | 687 | 77.3 | |
Specialist nurse | 131 | 14.7 | ||
Official master’s degree | 60 | 6.7 | ||
DEA/Investigative Adequacy | 3 | 0.3 | ||
PhD | 8 | 0.9 | ||
Work | ||||
Job position (N = 929) | Base nurse | 707 | 76.1 | |
Specialist nurse | 163 | 17.5 | ||
Case manager nurse | 9 | 1 | ||
Coordinator | 45 | 4.8 | ||
Professor | 5 | 0.5 | ||
Job experience (N = 928) | Years Mean (SD) Median (IR) | 13 (11.1) 9 (19) | ||
Unit where he worked during the pandemic | Covid plant | 378 | 40.6 | |
Plant no Covid | 78 | 8.4 | ||
UCI Covid | 314 | 33.8 | ||
UCI | 63 | 6.8 | ||
Emergencies | 76 | 8.2 | ||
Operating room | 21 | 2.3 | ||
Frequency of exposure to deaths | Never | 144 | 15.5 | |
Rarely (once a week or less) | 449 | 48.3 | ||
Quite a bit (2 to 4 times a week) | 252 | 27.1 | ||
A lot (5 or more times per week) | 85 | 9.1 | ||
Preparation towards death as a professional | Poor | 190 | 20.4 | |
Moderate | 401 | 43.1 | ||
Satisfactory | 231 | 24.8 | ||
Good | 108 | 11.6 | ||
Perception of being trained to cope with death | Yes | 432 | 46.5 | |
No | 492 | 53.5 | ||
Support systems of the institution | Yes | 481 | 51.7 | |
No | 449 | 48.3 | ||
Health | ||||
Under medication | Yes | 236 | 25.4 | |
No | 694 | 74.6 | ||
In the analysis of the PROQOL Scale, an average score of 28.5 (SD: 9.6) was observed in the compassion fatigue subscale, an average score of 48.6 (SD: 8.3) in the compassion satisfaction subscale, and an average score of 36.8 (SD: 6.9) in the Burnout dimension (Table 3).
Table 3
Characteristics of the sample scores of accessory scales
PROQOL scale scores. | |||
Compassion fatigue | Mean (SD) Median (IR) | 28.5 (9.6) 27 (14) | |
Compassion satisfaction | Mean (SD) Median (IR) | 48.6 (8.3) 50 (10) | |
Burnout | Mean (SD) Median (IR) | 36.8 (6.9) 37 (9) | |
Social support scale scores. | |||
Laboral emotional support | Mean (SD) Median (IR) | 16.6 (2.7) 17 (4) | |
Non-Laboral emotional support | Mean (SD) Median (IR) | 13.4 (2.4) 14 (3) | |
Laboral instrumental support | Mean (SD) Median (IR) | 10.1 (2.5) 11 (3) | |
PTGI scale scores. | |||
Relationship with others | Mean (SD) Median (IR) | 26 (8.6) 27 (12) | |
New possibilities | Mean (SD) Median (IR) | 20.1 (6.5) 21 (10) | |
Personal strength | Mean (SD) Median (IR) | 16.1 (5.2) 17 (8) | |
Spiritual change | Mean (SD) Median (IR) | 8 (2.7) 8 (4) | |
Life appreciation | Mean (SD) Median (IR) | 11.8 (4) 12 (6) | |
Global post-traumatic growth index | Mean (SD) Median (IR) | 82 (25.7) 86 (37) | |
DASS21 scale scores. | |||
Depression | Mean (SD) Median (IR) | 12.4 (4.4) 11 (6) | |
Anxiety | Mean (SD) Median (IR) | 13.7 (5) 13 (7.3) | |
Stress | Mean (SD) Median (IR) | 15 (4.7) 14 (6) | |
COPE28 scale scores. | |||
Style 1: Problem-Focused Coping | Mean (SD) Median (IR) | 2.5 (0.5) 2.5 (0.6) | |
Style 2: Emotion-Focused Coping | Mean (SD) Median (IR) | 2.1 (0.4) 2.1 (0.5) | |
Style 3: Avoidant Coping | Mean (SD) Median (IR) | 2.4 (0.4) 2.4 (0.5) | |
Facet 1: Active coping | Mean (SD) Median (IR) | 2.6 (0.6) 2.5 (1) | |
Facet 2: Use of informational support | Mean (SD) Median (IR) | 2.9 (0.7) 3 (1) | |
Facet 3: Positive reframing | Mean (SD) Median (IR) | 2.4 (0.7) 2.5 (1) | |
Facet 4: Planning | Mean (SD) Median (IR) | 2.1 (0.6) 2 (1) | |
Facet 5: Emotional support | Mean (SD) Median (IR) | 1.5 (0.6) 1.5 (1) | |
Facet 6: Venting | Mean (SD) Median (IR) | 2.9 (0.6) 3 (1) | |
Facet 7: Humour | Mean (SD) Median (IR) | 2.8 (0.7) 3 (1) | |
Facet 8: Acceptance | Mean (SD) Median (IR) | 1.7 (0.6) 1.5 (1) | |
Facet 9: Religion | Mean (SD) Median (IR) | 2.2 (0.6) 2 (0.5) | |
Facet 10: Self-blame | Mean (SD) Median (IR) | 2.2 (0.5) 2 (0.5) | |
Facet 11: Self-distraction | Mean (SD) Median (IR) | 2.2 (0.7) 2 (1) | |
Facet 12: Denial | Mean (SD) Median (IR) | 2.8 (0.6) 3 (0.5) | |
Facet 13: Substance use | Mean (SD) Median (IR) | 2.1 (0.6) 2 (1) | |
Facet 14: Behavioural disengagement | Mean (SD) Median (IR) | 2.4 (0.7) 2.5 (1) | |
PHQ9 scale scores. | |||
Global score (severity of the clinical picture) | Mean (SD) Median (IR) | 8.5 (6.1) 8 (8.3) | |
Classification | Minimum | 290 | 31.2 |
Mild | 285 | 30.6 | |
Moderate | 193 | 20.8 | |
Serious | 112 | 12 | |
Severe | 50 | 5.4 | |
IES-R scale scores. | |||
Intrusion | Mean (SD) Median (IR) | 18.2 (6.1) 18 (9) | |
Avoidance | Mean (SD) Median (IR) | 17.7 (5.8) 18 (9) | |
Hyperarousal | Mean (SD) Median (IR) | 13.3 (4.7) 13 (8) | |
In the analysis of scores on the social support scale, different levels were observed in the evaluated dimensions, being the laboral emotional support dimension with the higher mean score (16.6 (SD: 2.7)) and the laboral instrumental support dimension with the lowest score (10.1 (SD: 2.5)) (Table 3).
Regarding the scores on the Posttraumatic Growth Inventory (PTGI) for assessing personal growth after traumatic events, the following average scores were observed: 26 (SD: 8.6) for relationships with others, 20.1 (SD: 6.5) for openness to new opportunities, 16.1 (SD: 5.2) for personal strength, 8 (SD: 2.7) for spiritual change, and 11.8 (SD: 4) for appreciation for life. The Global Posttraumatic Growth Index, which summarizes all these dimensions, revealed an overall posttraumatic growth score of 82 (SD: 25.7) (Table 3).
In the analysis of the DASS21 Scale, the results revealed that participants obtained an average score of 12.4 (SD: 4.4) on the depression subscale, with a median of 11. Regarding the anxiety subscale, participants achieved an average score of 13.7 (SD: 5) with a median of 13. Finally, on the stress subscale, an average score of 15 (SD: 4.7) was observed, with a median of 14 (Table 3).
Furthermore, the COPE28 Scale was employed to assess the coping strategies employed by participants in response to stressful situations. The scores in various dimensions and facets of this scale provided valuable insights. Concerning general coping styles, participants exhibited scores for problem-focused coping (Style 1) with a mean of 2.5 (SD 0.5), emotion-focused coping (Style 2) with a mean of 2.1 (SD 0.4), and avoidance coping (Style 3) with a mean of 2.4 (SD 0.4). Additionally, these strategies were further delineated into specific coping facets. High scores were observed in facets such as ‘Use of informational support’ (Facet 2, mean: 2.9, SD 0.7) and ‘Venting’ (Facet 6, mean: 2.9, SD 0.6), while ‘Emotional support’ (Facet 5, mean: 1.5, SD 0.6) and ‘Acceptance’ (Facet 8, mean: 1.7, SD 0.6) exhibited lower scores (Table 3).
The results regarding the PHQ-9 Scale showed an average score of 8.5 (SD 6.1) (Table 3).
In the IES-R Scale results, an average score of 18.2 (SD: 6.1) was obtained in the Intrusion dimension, an average score of 17.7 (SD: 5.8) in the Avoidance dimension, and an average score of 13.3 (SD: 4.7) in the Hyperarousal dimension (Table 3).
None of the variables had a normal distribution, according to the results of the Kolmogorov-Smirnov tests (data not shown).
Differences of the professional traumatic grief in sociodemographic characteristics.
Women obtained higher scores on the ISDUTYP compared to men: median of 78 vs. 66 in the global score, 2.31 vs. 1.92 in behavioural symptoms and 4 vs. 3.33 in emotional symptoms. In addition, the scores decreased (better situation) as the years of experience increased (rho < 0 in all cases). On the other hand, higher scores were obtained for the symptomatology of professional traumatic grief in people who worked in the Operating Room (84.62) and ICU (83.52) compared to those who worked in other units. There was no statistically significant difference, however. On the other hand, the professionals who presented higher traumatic bereavement symptomatology mean scores were those who experienced 5 or more deaths per week (108.85) and were moderately (83.9) or poorly (82.19) prepared for death as a professional. The same occurs in the subscales.
Construct validity. Factor structure of the ISDUTYP scale
The exploratory factorial analysis carried out to examine the factorial validity of the scale, shown that according to the eigenvalues and the scree plot (Fig. 1), the optimal number of factors was 2. These factors were composed of 13 and 12 items, respectively, and they made clinical sense. Factor 1 can be named “Behavioural symptoms”, while factor 2 could be called “Emotional symptoms”.
Fig. 1
Scree plot
×
![]()
Table 4 shows the factor loadings after Varimax rotation, a colour code being used for belonging to each factor. Factor 1 includes the items 12–14, 16–25; and factor 2 includes the items 1–11 and 15.
Table 4
Factor loadings after varimax rotation
Item | Factor 1 loadings | Factor 2 loadings |
|---|---|---|
Behavioural symptoms | Emotional symptoms | |
1 Feelingsofsadness | 0.0893206 | 0.679516 |
2 Feelingsofhelplessness | 0.155079 | 0.787298 |
3 Feelingsoffear | 0.208207 | 0.74933 |
4 Emotional exhaustion related to the traumatic event | 0.225395 | 0.769674 |
5 Difficulty falling or staying asleep | 0.254905 | 0.616138 |
6 Anxiety | 0.381397 | 0.637518 |
7 Guilt about the causes or consequences of the traumatic event | 0.516644 | 0.563449 |
8 Continuous negative mood in the form of rage, anger and/or annoyance | 0.40558 | 0.654701 |
9 Desperation | 0.42629 | 0.746102 |
10 Memories, thoughts, or feelings of distress | 0.473093 | 0.750927 |
11 Feelingsof irritability | 0.456808 | 0.662021 |
12 Performing behaviours or experiencing sensations or emotions as if the event were happening again | 0.505309 | 0.614048 |
13 Experiencing unhealthy behaviours (smoking, drinking, other) after the traumatic event | 0.542071 | 0.197192 |
14 Depression | 0.628694 | 0.336955 |
15 Feelingsof dullness | 0.417341 | 0.51486 |
16 Experiencing unpleasant and repetitive memories or images of the event involuntarily | 0.578247 | 0.536143 |
17 Avoidance of places, objects or thoughts that recalled the event | 0.732567 | 0.411521 |
18 Difficulty breathing | 0.783897 | 0.315264 |
19 Intrusive thoughts or images related to the traumatic event | 0.77225 | 0.392673 |
20 Panicattacks | 0.812608 | 0.234219 |
21 Loss of ability to care about patients or feeling distant with patients you used to care about since the traumatic event | 0.722754 | 0.177847 |
22 Difficulty in accepting reality | 0.819859 | 0.275787 |
23 Experiencing confusions when remembering what happened | 0.83197 | 0.223685 |
24 Nervousness or tremors | 0.754436 | 0.297627 |
25 Problems adapting to the situation experienced | 0.800883 | 0.326889 |
According to the descriptive analysis of the scale and subscales, the mean score of the behavioural symptoms subscale was 2.56, and the mean score of the emotional symptoms subscale was 3.84, indicating that the sample had better behavioural symptoms compared to emotional ones. The global mean score was 79.44. None of the scores had a normal distribution (Table 5).
Table 5
Scores and distribution of the ISDUTYP subscales
Subscale | Mean (SD) | Median (IR) | Normal distribution |
|---|---|---|---|
Behavioural symptoms | 2.56 (1.25) | 2.23 (1.85) | No |
Emotional symptoms | 3.84 (1.13) | 3.92 (1.75) | No |
Global score | 79.44 (28.05) | 76 (42) | No |
No ceiling effect was detected in any item, in the subscales, or in the overall score. However, a certain floor effect was observed in some of the items, such as 13, 14, 20 or 21, among others. It should be noted that the interpretation of this scale is that the higher the score, the worse the situation. Therefore, the floor effect in some items could indicate a good general situation with respect to that aspect. In any case, the scores of the subscales and the global score did not show any floor effect (Table 6).
Table 6
Floor and ceiling effect of the ISDUTYP items and subscales
Item/Subscale | Worst score (floor effect) n (%) | Best score (ceiling effect) n (%) |
|---|---|---|
Items | ||
Item 1 | 2 (0.2) | 59 (6.3) |
Item 2 | 8 (0.9) | 71 (7.6) |
Item 3 | 26 (2.8) | 50 (5.4) |
Item 4 | 25 (2.7) | 57 (6.1) |
Item 5 | 70 (7.5) | 45 (4.8) |
Item 6 | 87 (9.4) | 32 (3.4) |
Item 7 | 207 (22.3) | 6 (0.6) |
Item 8 | 127 (13.7) | 19 (2) |
Item 9 | 157 (16.9) | 20 (2.2) |
Item 10 | 113 (12.2) | 12 (1.3) |
Item 11 | 101 (10.9) | 23 (2.5) |
Item 12 | 195 (21) | 11 (1.2) |
Item 13 | 430 (46.2) | 14 (1.5) |
Item 14 | 383 (41.2) | 14 (1.5) |
Item 15 | 109 (11.7) | 58 (6.2) |
Item 16 | 151 (16.2) | 40 (4.3) |
Item 17 | 311 (33.4) | 18 (1.9) |
Item 18 | 399 (42.9) | 6 (0.6) |
Item 19 | 282 (30.3) | 21 (2.3) |
Item 20 | 507 (54.5) | 11 (1.2) |
Item 21 | 402 (43.2) | 10 (1.1) |
Item 22 | 360 (38.7) | 12 (1.3) |
Item 23 | 361 (38.8) | 11 (1.2) |
Item 24 | 333 (35.8) | 29 (3.1) |
Item 25 | 291 (31.3) | 15 (1.6) |
Subescales | ||
Behavioural symptoms | 32 (3.4) | 1 (0.1) |
Emotional symptoms | 0 (0) | 1 (0.1) |
Global score | 0 (0) | 1 (0.1) |
Reliability
The Cronbach´salpha values showed excellent reliability in all the scores. Particularly, Cronbach´s alpha was 0.964for the global score, 0.950 for the behavioural symptoms subscale and 0.950 for the emotional symptoms subscale. Regarding the item-scales correlations, all of them were direct and statistically significant. In addition, the correlations were stronger between each item and the subscale to which it belonged. All of them correlated moderately or strongly with the overall score, correlations ranging between 0.517 and 0.871 (Table 7).
Table 7
Correlations item-scales. Spearman’s rank correlation coefficient (rho)
Item | Factor 1 (Behavioural symptoms) | Factor 2 (Emotional symptoms) | Global score | |||
|---|---|---|---|---|---|---|
rho | p | rho | p | rho | p | |
Item 1 | 0.416 | < 0.001 | 0.616 | < 0.001 | 0.540 | < 0.001 |
Item 2 | 0.500 | < 0.001 | 0.734 | < 0.001 | 0.643 | < 0.001 |
Item 3 | 0.506 | < 0.001 | 0.728 | < 0.001 | 0.645 | < 0.001 |
Item 4 | 0.536 | < 0.001 | 0.774 | < 0.001 | 0.687 | < 0.001 |
Item 5 | 0.488 | < 0.001 | 0.668 | < 0.001 | 0.608 | < 0.001 |
Item 6 | 0.614 | < 0.001 | 0.772 | < 0.001 | 0.731 | < 0.001 |
Item 7 | 0.678 | < 0.001 | 0.752 | < 0.001 | 0.755 | < 0.001 |
Item 8 | 0.650 | < 0.001 | 0.786 | < 0.001 | 0.752 | < 0.001 |
Item 9 | 0.706 | < 0.001 | 0.868 | < 0.001 | 0.826 | < 0.001 |
Item 10 | 0.764 | < 0.001 | 0.890 | < 0.001 | 0.871 | < 0.001 |
Item 11 | 0.694 | < 0.001 | 0.809 | < 0.001 | 0.790 | < 0.001 |
Item 12 | 0.763 | < 0.001 | 0.752 | < 0.001 | 0.793 | < 0.001 |
Item 13 | 0.577 | < 0.001 | 0.395 | < 0.001 | 0.517 | < 0.001 |
Item 14 | 0.714 | < 0.001 | 0.572 | < 0.001 | 0.679 | < 0.001 |
Item 15 | 0.599 | < 0.001 | 0.671 | < 0.001 | 0.668 | < 0.001 |
Item 16 | 0.793 | < 0.001 | 0.714 | < 0.001 | 0.792 | < 0.001 |
Item 17 | 0.847 | < 0.001 | 0.668 | < 0.001 | 0.804 | < 0.001 |
Item 18 | 0.822 | < 0.001 | 0.631 | < 0.001 | 0.774 | < 0.001 |
Item 19 | 0.867 | < 0.001 | 0.683 | < 0.001 | 0.820 | < 0.001 |
Item 20 | 0.774 | < 0.001 | 0.561 | < 0.001 | 0.712 | < 0.001 |
Item 21 | 0.670 | < 0.001 | 0.458 | < 0.001 | 0.602 | < 0.001 |
Item 22 | 0.802 | < 0.001 | 0.584 | < 0.001 | 0.737 | < 0.001 |
Item 23 | 0.795 | < 0.001 | 0.542 | < 0.001 | 0.710 | < 0.001 |
Item 24 | 0.794 | < 0.001 | 0.601 | < 0.001 | 0.739 | < 0.001 |
Item 25 | 0.840 | < 0.001 | 0.637 | < 0.001 | 0.782 | < 0.001 |
Criterion validity
There is no gold standard for the criterion validity of the whole scale. Alternatively, Table 8 shows the correlation of each item with the scale that measures the same construct. The empty spaces in the table correspond to correlations that are not required a priori by any theoretical framework. All the expected correlations were statisticallysignificant, being in the majority of the cases moderates correlations.
Table 8
Correlation of each item with the scale that measures the same construct
Item | COPE-28. Coping. | DASS-21. Depression, anxiety, and stress. | IES-R. Event impact | PROQOL-VIV. Burnout. | PTGI. Post-traumatic growth. | PHQ-9. Clinical picture severity. |
|---|---|---|---|---|---|---|
1 Feelings of sadness | Depression subscale Rho = 0.306 p < 0.001 | |||||
2 Feelings of helplessness | Emotion focused subscale Rho = 0.160 p < 0.001 | |||||
3 Feelings of fear | Intrusion subscale Rho = 0.410 p < 0.001 | |||||
4 Emotional exhaustion related to the traumatic event | Stress subscale Rho = 0.501 p < 0.001 | Compassion fatigue subscale Rho = 0.409 p < 0.001 | ||||
5 Difficulty falling or staying asleep | Hyperarousal subscale Rho = 0.455 p < 0.001 | Rho = 0.433 p < 0.001 | ||||
6 Anxiety | Anxiety subscale Rho = 0.542 p < 0.001 | Hyperarousal subscale Rho = 0.500 p < 0.001 | ||||
7 Guilt about the causes or consequences of the traumatic event | Emotion focused subscale Rho = 0.227 p < 0.001 | Intrusion subscale Rho = 0.544 p < 0.001 | Rho = 0.149 p < 0.001 | |||
8 Continuous negative mood in the form of rage, anger and/or annoyance | Stress subscale Rho = 0.509 p < 0.001 | |||||
9 Desperation | Anxiety subscale Rho = 0.606 p < 0.001 | |||||
10 Memories, thoughts, or feelings of distress | Anxiety subscale Rho = 0.674 p < 0.001 | |||||
11 Feelings of irritability | Stress subscale Rho = 0.586 p < 0.001 | Hyperarousal subscale Rho = 0.555 p < 0.001 | ||||
12 Performing behaviours or experiencing sensations or emotions as if the event were happening again | Hyperarousal subscale Rho = 0.512 p < 0.001 | Compassion fatigue subscale Rho = 0.526 p < 0.001 | Rho = 0.113 p = 0.001 | |||
13 Experiencing unhealthy behaviours (smoking, drinking, other) after the traumatic event | Rho = 0.301 p < 0.001 | |||||
14 Depression | Depression subscale Rho = 0.491 p < 0.001 | |||||
15 Feelings of dullness | Compassion fatigue subscale Rho = 0.408 p < 0.001 | |||||
16 Experiencing unpleasant and repetitive memories or images of the event involuntarily | Intrusion subscale Rho = 0.562 p < 0.001 | Compassion fatigue subscale Rho = 0.554 p < 0.001 | ||||
17 Avoidance of places, objects or thoughts that recalled the event | Avoidance subscale Rho = 0.109 p = 0.001 | Avoidance subscale Rho = 0.550 p < 0.001 | ||||
18 Difficulty breathing | Anxiety subscale Rho = 0.664 p < 0.001 | Hyperarousal subscale Rho = 0.561 p < 0.001 | Rho = 0.555 p < 0.001 | |||
19 Intrusive thoughts or images related to the traumatic event | Intrusion subscale Rho = 0.620 p < 0.001 | Compassion fatigue subscale Rho = 0.583 p < 0.001 | ||||
20 Panic attacks | Anxiety subscale Rho = 0.582 p < 0.001 | |||||
21 Loss of ability to care about patients or feeling distant with patients you used to care about since the traumatic event | Avoidance subscale Rho = 0.217 p < 0.001 | |||||
22 Difficulty in accepting reality | Avoidance subscale Rho = 0.145 p < 0.001 | Avoidance subscale Rho = 0.486 p < 0.001 | Compassion fatigue subscale Rho = 0.513 p < 0.001 | |||
23 Experiencing confusions when remembering what happened | Intrusion subscale Rho = 0.495 p < 0.001 | |||||
24 Nervousness or tremors | Anxiety subscale Rho = 0.645 p < 0.001 | |||||
25 Problems adapting to the situation experienced | Avoidance subscale Rho = 0.119 p < 0.001 | Compassion fatigue subscale Rho = 0.546 p < 0.001 | Rho = 0.112 p = 0.001 |
Convergent and discriminant validity
Finally, almost all the hypotheses defined for the convergent and discriminant validity were met (Table 9). First, we highlight that we posed the hypothesis that people with more accentuated symptoms of professional traumatic grief would have active coping strategies. However, this could only be demonstrated for the behavioural symptoms subscale (Rho=-0.083, p = 0.012), which makes perfect sense. Nevertheless, it would have been desirable to find the same relationship with, at least, the global score (Table 9).
Table 9
Convergent and discriminant validity
ISDUTYP Behavioural symptoms | ISDUTYP Emotional symptoms | ISDUTYP Global score | |||||
|---|---|---|---|---|---|---|---|
p | p | p | |||||
Gender | Men (Median (IR)) | 1.92 (1.62) | < 0.001a | 3.33 (1.63) | < 0.001a | 66 (35) | < 0.001a |
Women (Median (IR)) | 2.31 (1.85) | 4 (1.58) | 78 (42) | ||||
Job experience | Years | -0.175 | < 0.001 | -0.158 | < 0.001 | -0.176 | < 0.001 |
Unit working during the pandemic Covid | Covid Plant(Median (IR)) | 2.23(1.79) | 0.269b | 3.92(1.75) | 0.143b | 77.5(44) | 0.139b |
No Covid Plant (Median (IR)) | 2.35(2) | 4(1.69) | 81(41) | ||||
ICU Covid (Median (IR)) | 2.15(1.62) | 3.83(1.42) | 74(36) | ||||
ICU (Median (IR)) | 2.46(2.38) | 3.92(1.58) | 79(43) | ||||
Emergency Room (Median (IR)) | 2(1.44) | 3.67(1.98) | 74.5(40) | ||||
Operating Room (Median (IR)) | 2.38(3.35) | 3.75(3.08) | 75(76) | ||||
Frequency of exposure to deaths | Never (Median (IR)) | 1.92(1.29) | < 0.001b | 3.46(1.48) | < 0.001b | 66.5(35) | < 0.001b |
Rarely (once a week or less) (Median (IR)) | 2.15(1.46) | 3.67(1.58) | 72(35) | ||||
Quite a bit (2 to 4 times a week)(Median (IR)) | 2.54(2.06) | 4.25(1.25) | 83(41) | ||||
A lot (5 or more times per week) (Median (IR)) | 3.92(2.35) | 5.08(1.37) | 111(41) | ||||
Preparation towards death as a professional | Poor (Median (IR)) | 2.31(1.63) | < 0.001b | 4(1.67) | < 0.001b | 78(40) | < 0.001b |
Moderate (Median (IR)) | 2.46(2) | 4(1.5) | 79(41) | ||||
Satisfactory (Median (IR)) | 2.08(1.38) | 3.67(1.75) | 71(36) | ||||
Good (Median (IR)) | 1.69(1.88) | 3.42(2.17) | 65(50) | ||||
COPE-28 | FACET 1 Active coping | Rho=-0.083 | 0.012 | Rho=-0.032 | 0.336 | Rho=-0.060 | 0.069 |
DASS-21 | Depression | Rho = 0.659 | < 0.001 | Rho = 0.613 | < 0.001 | Rho = 0.674 | < 0.001 |
Anxiety | Rho = 0.719 | < 0.001 | Rho = 0.675 | < 0.001 | Rho = 0.736 | < 0.001 | |
Stress | Rho = 0.680 | < 0.001 | Rho = 0.678 | < 0.001 | Rho = 0.717 | < 0.001 | |
IES-R | Intrusion | Rho = 0.633 | < 0.001 | Rho = 0.614 | < 0.001 | Rho = 0.657 | < 0.001 |
Avoidance | Rho = 0.563 | < 0.001 | Rho = 0.520 | < 0.001 | Rho = 0.571 | < 0.001 | |
Hyperarousal | Rho = 0.630 | < 0.001 | Rho = 0.623 | < 0.001 | Rho = 0.660 | < 0.001 | |
PROQOL | Compassion fatigue | Rho = 0.673 | < 0.001 | Rho = 0.612 | < 0.001 | Rho = 0.682 | < 0.001 |
Compassion satisfaction | Rho=-0.239 | < 0.001 | Rho=-0.120 | < 0.001 | Rho=-0.190 | < 0.001 | |
Burnout | Rho = 0.408 | < 0.001 | Rho = 0.462 | < 0.001 | Rho = 0.461 | < 0.001 | |
Social support | Laboral emotional support | Rho=-0.168 | < 0.001 | Rho=-0.071 | 0.031 | Rho=-0.130 | < 0.001 |
Non-Laboral emotional support | Rho=-0.092 | 0.005 | Rho=-0.023 | 0.488 | Rho=-0.064 | 0.051 | |
Laboral instrumental support | Rho=-0.116 | < 0.001 | Rho=-0.028 | 0.386 | Rho=-0.073 | 0.025 | |
PTGI | Relationship with others | Rho = 0.124 | < 0.001 | Rho = 0.222 | < 0.001 | Rho = 0.181 | < 0.001 |
New possibilities | Rho = 0.066 | 0.045 | Rho = 0.182 | < 0.001 | Rho = 0.127 | < 0.001 | |
Personal strength | Rho = 0.034 | 0.297 | Rho = 0.146 | < 0.001 | Rho = 0.091 | 0.005 | |
Spiritual change | Rho = 0.046 | 0.158 | Rho = 0.152 | < 0.001 | Rho = 0.101 | 0.002 | |
Life appreciation | Rho = 0.121 | < 0.001 | Rho = 0.251 | < 0.001 | Rho = 0.195 | < 0.001 | |
Global post-traumatic growth index | Rho = 0.087 | 0.008 | Rho = 0.203 | < 0.001 | Rho = 0.151 | < 0.001 | |
PHQ9 | Global score (severity of the clinical picture) | Rho = 0.576 | < 0.001 | Rho = 0.553 | < 0.001 | Rho = 0.596 | < 0.001 |
The scores on the ISDUTYP were higher in individuals with higher stress, anxiety, and depression (rho > 0.61, p < 0.001 in all cases), and these subjects also obtained higherscoreson the scale measuring the impact of the event (rho > 0.52, p < 0.001 in all cases) (Table 9).
The ISDUTYP also correlated with the PROQOL, with inverse correlations with the subscale of compassion satisfaction and direct correlations with compassion fatigue and burnout, although the latter were weaker. That is, the symptoms of professional traumatic grief were more accentuated in individuals with higher levels of compassion fatigue (rho > 0.61, p < 0.001 in the subscales and the global scale) and burnout (rho > 0.41, p < 0.001) and lower levels of compassion satisfaction (rho<-0.12, p < 0.001 in the subscales and the global scale). Additionally, the ISDUTYP scores were lower among individuals with higher scores on the social support scale, as inverse relationships were found (rho < 0). However, it is worth mentioning that the correlations with factor 2 (emotional symptoms) were not statistically significant in non-laboral emotional support and laboral instrumental support (Table 9).
Furthermore, we hypothesised that the symptoms of professional traumatic grief would be more accentuated in individuals with higher scores on the post-traumatic growth scale. This direct relationship was observed in most of the subscales, but it was not statistically significant in some cases. In particular, no relationships were found between behavioural symptoms and either personal strength or spiritual change. There was a relationship between the subscales and the global score of the ISDUTYP and relationship with others, life appreciation and global post-traumatic growth index (Table 9).
Finally, the ISDUTYP also correlated positively with the PHQ9, i.e., the higher the bereavement score, the greater the depressive disorder (rho > 0.55, p < 0.001 in the subscales and the global scale) (Table 9).
Discussion
Present study provides new insights into the psychometric robustness and the relevance of the ISDUTYP scale for capturing the multifaceted nature of professional traumatic grief among nursing professionals. Notably, the scale’s internal consistency, as evidenced by the Cronbach’s alpha values, attests to its reliability for both overall and individual item responses. This highlights the scale’s utility across diverse healthcare settings, reflecting the complex emotional and behavioral responses toprofessional grief that have been echoed in prior qualitative research [58].
Furthermore, our findings extend the current understanding of professional traumatic grief by confirming a two-factor structure that encompasses both behavioral and emotional dimensions, aligning with previous literature on grief symptomatology [26, 59]. Interestingly, our study departs from earlier phases of the scale’s development by Gilart [1], where physical symptoms were initially considered but later excluded, aligning our results more closely with predominant grief research.
Our investigation into the scale’s convergent and discriminant validity reveals several intriguing findings. For instance, gender-based differences in grief intensity align with existing literature, which highlights the heightened vulnerability of women in health sciences to prolonged grief [39, 46].
Additionally, in line with previous research, the results indicated that professional traumatic grief scores were lower among individuals with more work experience [60]. However, variations in scores were observed depending on the specific services. In fact, it should be noted that operating roomprofessionals scored high on ISDUTYP symptomatology. At first glance, this data may seem surprising, however, the importance of the unprecedented pandemic situationshould not be underestimated. During this time a reorganization of infrastructures and human resources was necessary, leading to the reassignment of many operating room nurses to other units with greater needs, where they were suddenly exposed to multiple deaths. Moreoverthe chronicity and complexity of bereavement become more evident, as our results indicate in the professionals of the ICU service. It is in this context where nursing professionals routinely face cumulative deaths and have a greater exposure to deaths [61]. On the other hand, emergency professionals exhibited lower levels on such symptomatology. This result could be attributed, to the service environment and the lack of time to communicate that prevents individualized and sensitive care and attention being provided to patients.
It is important to note that although health professionals have to be trained to perform different skills and to face the death of patients, in some cases the preparation to deal with these complex situations is inadequate, as our results indicate. For this reason, they are more predisposed to suffer a prolonged type of bereavement. Training can make a difference, as evidenced by studies that have found that more training results in fewer negative reactions [38, 39].
Similarly, we found that individuals with higher levels of stress, anxiety, and depression scored higher on the ISDUTYP. This finding aligns with previous studies that report elevated rates of anxiety, stress and major depression [19, 41, 62], which may be considered risk factors for the development of bereavement syndrome. In fact, a recent study conducted in China during the COVID- 19 pandemic found that health care workers reported symptoms of depression (50%) and anxiety (45%), and more specifically that nurses who have provided direct care to COVID-19 patients have a higher risk of more severe psychological distress [42].
In the same vein, it is important to highlight thatthe symptomatology of professional traumatic grief is more pronounced in individuals who have significantly higher scores on the impact of the event scale, as reported in other studies [43]. The relationship between the symptomatology of professional traumatic grief and significantly higher scores on the Impact of Event Scale may be explained by the overlap of symptoms between post-traumatic stress and professional traumatic grief. Both conditions share common features, such as intrusive thoughts or distressing memories, avoidance of event-related reminders, and hyperarousal symptoms. Consequently, individuals with higher IES scores may experience symptom overlap with professional traumatic grief. On the other hand, traumatic events at work, such as exposure to multiple deaths due to COVID-19, can have a lasting impact on individuals’ mental health. Higher scores on the IES may reflect a greater severity of trauma exposure and, consequently, a higher intensity of symptoms of professional traumatic grief.
Regarding the correlation between the ISDUTYP and the PROQOL, our results indicated, in agreement with other research, that the symptoms of professional traumatic grief are more accentuated in individuals with higher levels of compassion fatigue and lower levels of compassion satisfaction [11, 44, 63]. These data are consistent with those obtained in other countries with a severe COVID-19 crisis situation, such as China [42], and with studies conducted before the pandemic showing that continuous contactwith patient suffering can increase compassion fatigue and burnout in health professionals [64, 65].
Another notable finding is that ISDUTYP scores were lower in individuals with higher scores on the social support scale, indicating an inverse relationship. This aligns with existing literature suggesting that nurses who receive peer support are better able to cope with patient deaths [31, 47].
Furthermore, the hypothesis that professional traumatic grief symptoms would be more accentuated in individuals with higher scores on the post-traumatic growth scale was corroborated. A possible explanation lies in the complex and highly variable nature of individual responses to trauma.
Individuals experiencing post-traumatic growth may exhibit a more complex overall response to trauma. While post-traumatic growth has traditionally been regarded as an adaptive phenomenon [66, 67], recent research has suggested that PTG may represent a maladaptive coping response, akin to cognitive avoidance. This could encompass an intensified manifestation of symptoms of traumatic grief as part of such a response [48].
Furthermore, in the context of “atypical” deaths, such as the COVID-19 pandemic, the circumstances associated with this crisis may have exacerbated suffering, elicited maladaptive behaviors, and consequently influenced the manifestation of complicated grief symptoms [49].
This direct relationship was observed in most subscales, although it was not statistically significant in some cases. Notably, no relationships were found between behavioural symptoms and personal strength or spiritual change. It is worth mentioning that the reported structures of the PTGI across different countries have been controversial, possibly because of the weight of the sociocultural differences that each item of the scale has, which may have a different valuation according to each culture. Similarly, Joseph [68] proposed that the items on religion and spirituality are confounding in measures of post-traumatic growth. Shaw et al. [69], argue thatsome trauma victims experience an increase in their religiosity and spirituality, while in other cases these experiences decrease. Therefore, Joseph [68] suggests that religiosity and spirituality should not be part of the definition of post-traumatic growth.
Finally, regarding the correlation between the ISDUTYP and the PHQ-9, our results indicated, in agreement with other studies, that symptoms of professional traumatic grief are more accentuated in individuals with depressive symptomatology [42].
In summary, the correlations between ISDUTYP and various diagnostic constructs provide clear evidenceof its convergent and discriminant validity, in relation to the scale’s overall scores, its two subscales, and the criterion variables.
Therefore, the high reliability and validity of this scale in this population is confirmedandconsequentlyit can be deduced that it is an effective measuring instrument to evaluate the symptomatology of professional traumatic grief. Although preliminary results suggest good validity and reliability, they should be interpreted with caution due to certain methodological limitations, such as the lack of test-retest reliability assessments and the absence of a gold standard for criterion validation.
Limitations
Finally, it is essential to highlight some of the limitations encountered throughout the research. One of the significant limitations of the study resides in the challenge of assessing criterion validity due to the absence of an appropriate gold standard. Given that the measurement instrument employed is of recent development and lacks an established reference point, correlation analyses were conducted among the aforementioned questionnaires. Nevertheless, efforts were made to identify the primary concept measured by each item and evaluate its correlation with the scale addressing that construct.
Another important methodological limitation is our decision to exclude nurses with pre-existing mood or anxiety disorders, to minimize potential biases in result interpretation. These conditions could have significantly influenced responses to the studied variables, making it difficult to isolate and evaluate the specific relationship between traumatic grief and the analyzed factors, which are the primary focus of this study.
Although this methodological choice aims to reduce bias, it may also limit the generalizability of the findings. Given the high prevalence of these conditions among healthcare professionals, they could significantly interact with traumatic grief symptoms. Therefore, future studies should include these groups to explore this interaction, providing a more comprehensive understanding of traumatic grief in this population.
Another potential limitation is the absence of sensitivity to change tests, which remain pending for subsequent studies.
Finally, there is a potential selection bias in the sample, which was obtained via the internet through an online survey. This could have hindered the participation of individuals lacking technological skills.
Furthermore, although the results suggest that the scale can be applied in healthcare settings, methodological limitations, such as the absence of test-retest reliability and a gold standard, indicate the need for further research. Indeed, one of the main challenges is its lack of longitudinal validation, which restricts theability to assess sensitivity to change across different temporal and clinical contexts. This means that, although the instrument presents initial evidence of reliability and validity, its ability to identify changes over time or in response to specific interventions has not yet been verified and needs further research.
Similarly, item information may not fully reflect the variety of experiences associated with traumatic grief in different health care settings. For example, although it was created for health professionals, its relevance to particular subgroups, such as informal caregivers, has not been thoroughly examined.
Conclusion
As a final conclusion, it can be stated that the present study provides evidence on the factorial structure, reliability and validity of the ISDUTYP, a new self-report instrument designed to assess symptoms of professional traumatic grief. It is hoped that the results obtained through the development and validation of this scale of the specific symptomatology of professional traumatic grief in health professionals can guide the objectives and interventions needed to treat this study population.
In conclusion, although the ISDUTYP presents significant potential as a tool for assessing professional traumatic grief, the methodological limitations identified require caution in interpreting its results. Its use should be considered preliminary and complemented by other assessment tools until these issues are addressed. Future research should focus on validating its factor structure, temporal reliability and applicability in cross-cultural contexts, with the aim of ensuring its effectiveness as a robust and generalizable clinical instrument.
This scale will serve as a valid tool that can facilitate future research endeavors, allowing for an assessment of its validity and reliability when applied to other healthcare professionals and informal caregivers. Collaterally, through such a measurement tool, the workflow of healthcare professionals will become more efficient, thus enabling them to focus on the comprehensive and specific care of the consequences of exposure to the death of patients. In relation to the costs associated with the provision of health services, we consider that this instrument can optimize and control of professionals who suffer a traumatic professional bereavement and help them to overcome it in less time, which would lead to lower direct and indirect social and health costs. Finally, it should be noted that this study will establish synergies between different health professions and could be a good starting point for future research, not only in the context of the pandemic but also for other complicated and/or traumatic situations, natural catastrophes, etc.
On a practical level, the implementation of the ISDUTYP in occupational wellness programs could facilitate the early detection of symptoms of professional traumatic grief. The results could be used to design personalized interventions, such as psychological support programs, grief management workshops or emotional resilience trainings. From a health policy perspective, the results suggest the need to include tools such as the ISDUTYP in regular workplace mental health assessments. This could be integrated into crisis management protocols in units with high mortality, such as intensive care or emergency units. Also, the use of the scale could be linked to personnel management systems, optimizing the distribution of human resources according to the emotional and psychological needs of the team.
In conclusion, according to preliminary results, the ISDUTYP has shown to be a promising tool for assessing professional traumatic grief in nursing. However, identified methodological shortcomings, such as the lack of test-retest reliability and of a gold standard for criterion validation, require further research to verify its factor structure, temporal reliability, and applicability in different contexts. The use of ISDUTYP should be considered preliminary and in combination with other assessment tools until such research is conducted.
Acknowledgements
We would like to express our gratitude to all individuals who actively participated in the present study and Own Plan for Research and Transfer 2025-2027 – University of Cadiz (UCA).
Declarations
Ethics approval and consent to participate
This study was conducted in accordance with ethical principles outlined in the Declaration of Helsinki and was approved by the appropriate ethics committee (PEIBA 0875-N-22).
Consent for publication
No applicable.
Competing interests
The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
