**4. Results**

for 10 s to think). Later, the researchers gave the participant the mental reinstatement of the context instruction to encourage him/her to report all information that he/she could accurately

154 A Multidimensional Approach to Post-Traumatic Stress Disorder - from Theory to Practice

Specifically, the CI began with this recommendation: "I want that you tell me everything you can remember, every little detail you can remember, even if you think it is not important or if you are not sure about it." Four different instructions were then given. (1) The "free recall" technique [52] asked the participants to report as many as details as they could to accurately recreate the scene. This technique has been shown to be one of the most valuable components of the CI (e.g., [53]). (2) We also modified the "image" technique because we asked the participants to report as many as details as they could regarding the first and the most emotionally impactful images of the event. (3) A modified "changed perspective" technique was used, in which we asked the participants to try to observe the scene above from a bird'seye view and report as many details as they could, in addition to the emotion experienced.

All CI were audio-recorded and transcribed verbatim. Later, the CIs were scored to evaluate each unit of information recalled by the participants. The units of information or details were categorized according to whether they were reported in the free recall, image, or changed perspective instructions. Two different coders (psychology post-doc students) scored the interviews; only when an agreement between them was reached were the audiotape details

For all the three techniques, we extracted and analyzed the number of scenes recreated, negative emotions, vividness of mental images, and self-experiences. We also used ATLAS.ti

We performed a descriptive analysis concerning the demographic data for years of experience. We also reported the presence/absence of traumatic events experienced, the coping strategies adopted during the emergency, sleep disorders, intrusive memories of the event, and general

To evaluate the factors that predict post-traumatic stress and the post-traumatic growth, we performed separate stepwise regression analyses loading personality traits and coping

To examine the possible differences between the groups due to the exposure to the three different events, we performed an analysis of variance (ANOVA) with the group (Avellino, Viareggio, and L'Aquila) and the number of memories, emotions and experiences that emerged from the modified cognitive interview in the three different instructions sections (free recall,

remember.

**2.3. Coding and scoring**

**3. Statistical analyses**

health disorders (see **Table 2**).

strategies as independent variables.

mental image, and changed perspective).

for the analysis and coding of the interviews.

coded.

#### **4.1. Regression analyses**

To evaluate the personality traits and coping strategies that predict the traumatic effects in a critical event, we performed separate stepwise regression analyses considering the scores of the CISS and the BFQ as the independent variables and the scores in TSI and PTGI as the dependent variables. None of the sociodemographic variables correlated with any of the dependent variables, hence we did not have to control for any in the regression model.

### **4.2. CISS vs. TSI**

The stepwise regression analysis revealed a significant effect for the factor *emotion oriented* on TSI AA (*scale of dysphoric mood states; β* = 0.691, *t* = 3.582, *p* < 0.01), TSI AI (*anger/irritability; β* = 0.716, *t* = 3.836, *p* < 0.01), TSI DIS (*dissociation; β* = 0.786, *t* = 4.763, *p* < 0.001), and TSI TRB (*tension-reduction behavior; β* = 0.611, *t* = 2.889, *p* < 0.05).

#### **4.3. BFQ vs. TSI**

The stepwise regression revealed a significant effect only for the *emotional stability* factor of BFQ on TSI DIS (*dissociation; β* = −0.580; *t* = 2.568, *p* < 0.05), and the *conscientiousness* factor on TSI TRB (*tension-reduction behavior; β* = −0.658; *t* = 3.152, *p* < 0.01). No other factor was statistically significant.

#### **4.4. CISS vs. PTGI**

No CISS factors showed significant effects.

#### **4.5. BFQ vs. PTGI**

The stepwise regression performed on the PTGI scores showed a significant effect of *emotional stability* and *conscientiousness* as predictors of *new line possibilities* (respectively, *β* = −0.683; *t* = −3.803, *p* < 0.01, and *β* = 0.549; *t* = 3.056, *p* < 0.01), the factors of *conscientiousness and extraversion/vivaciousness on personal strength growth* (respectively, *β* = 889; *t* = 4.405, *p* < 0.001 and *β* = −530; *t* = −2.653, *p* < 0.05). Finally, the *emotional stability* factor also showed a significant effect on *spiritual change* (*β* = −0.533; *t* = 2.270, *p* < 0.05).

#### **4.6. ANOVAs**

To investigate the presence of significant differences between the three groups interviewed through CI, we performed separate ANOVAs for the three techniques (free recall, mental image, and change perspective) of the CI considering the group (L'Aquila earthquake, Viareggio derailment, and Avellino coach crash) as the independent variable and the verbal production in free recall/mental images and change perspective (number of scenes recreated, negative emotions, vividness of the mental image, and self-experience) as the dependent variables. The statistical power is 0.06. Hence, all nonsignificant results from the ANOVA cannot be interpreted to avoid making type II errors. We however report them for transparency reasons only. From the ANOVA performed on the free recall, only a main effect of the group with respect to the self-experience was seen (*F*(2,14) = 3.993; *p* = 0.04; see **Figure 1A**). All other differences were not significant most probably because of low power: the number of scenes recreated (*F*(2,14) = 0.985; *p* = 0.40), negative emotions (*F*(2,14) = 0.635; *p* = 0.54), and the vividness of mental images (*F*(2,14) = 1.495; *p* = 0.26). In the mental image section, the three groups differed in terms of the self-experience reported (*F*(2,14) = 3.993; *p* < 0.05) and of the number of scenes recreated (*F*(2,14) = 5.48; *p* = 0.02), but the other types of verbal production differences were not significant most probably because of low power (*F*s(2,14) = from 3.993 to 0.635; *p*s = n.s.; see **Figure 1B**). No significant differences emerged from the change perspective technique in the verbal production of the three groups (*F*s(2,14) = from 1.919 to 0.181; *p*s = n.s.; see **Figure 1C**).

**Figure 1.** (A) Means of verbal production (number of scenes recreated, negative emotions, vividness of mental image, and self-experience) of the three groups in the free recall of CI; (B) means of verbal production (number of scenes recreated, negative emotions, vividness of mental image, and self-experience) of the three groups in the mental image of CI; (C) means of verbal production (number of scenes recreated, negative emotions, vividness of mental image, and selfexperience) of the three groups in the change perspective of CI.

#### **5. Discussion**

Studies on traumatic reactions in first responders suggest that stress symptoms can continue over a significant period of time [54] and may include guilt, anxiety, depression, sleep disturvariables. The statistical power is 0.06. Hence, all nonsignificant results from the ANOVA cannot be interpreted to avoid making type II errors. We however report them for transparency reasons only. From the ANOVA performed on the free recall, only a main effect of the group with respect to the self-experience was seen (*F*(2,14) = 3.993; *p* = 0.04; see **Figure 1A**). All other differences were not significant most probably because of low power: the number of scenes recreated (*F*(2,14) = 0.985; *p* = 0.40), negative emotions (*F*(2,14) = 0.635; *p* = 0.54), and the vividness of mental images (*F*(2,14) = 1.495; *p* = 0.26). In the mental image section, the three groups differed in terms of the self-experience reported (*F*(2,14) = 3.993; *p* < 0.05) and of the number of scenes recreated (*F*(2,14) = 5.48; *p* = 0.02), but the other types of verbal production differences were not significant most probably because of low power (*F*s(2,14) = from 3.993 to 0.635; *p*s = n.s.; see **Figure 1B**). No significant differences emerged from the change perspective technique in the verbal production of the three groups (*F*s(2,14) = from 1.919 to 0.181; *p*s =

156 A Multidimensional Approach to Post-Traumatic Stress Disorder - from Theory to Practice

**Figure 1.** (A) Means of verbal production (number of scenes recreated, negative emotions, vividness of mental image, and self-experience) of the three groups in the free recall of CI; (B) means of verbal production (number of scenes recreated, negative emotions, vividness of mental image, and self-experience) of the three groups in the mental image of CI; (C) means of verbal production (number of scenes recreated, negative emotions, vividness of mental image, and self-

Studies on traumatic reactions in first responders suggest that stress symptoms can continue over a significant period of time [54] and may include guilt, anxiety, depression, sleep distur-

experience) of the three groups in the change perspective of CI.

**5. Discussion**

n.s.; see **Figure 1C**).

bances, flashbacks (e.g., intrusive thoughts), and excessive anger [55]. This is one of the reasons we decided to investigate this aspect 3 and 7 years from the three different catastrophic events that occurred in Italy in which the police officers interviewed had an important role in the protection and preservation of life, property, and the environment. They were on the front lines of these three events, and they also worked hard in the aftermath, which lasted several months. In particular, in one of these disasters (i.e., L'Aquila), the long-lasting effects of the traumatic experience are currently manifested mainly by feeling "on edge" in addition to with hyperarousal and sleeping problems. Indeed, we observed significant differences between the selfexperience and the number of scenes recreated by the police officers who responded. Both in the L'Aquila earthquake and in the Viareggio derailment, the first-responders were exposed to hazardous environmental conditions. In the L'Aquila event, the main shock (Richter magnitude 6.9) was followed by several thousands of aftershocks (30 of which had a Richter magnitude of greater than 3.5), which made the rescue difficult. Conversely, in the Viareggio train derailment, there was the tangible risk of further wagon explosions due to the high temperature emitted by the first wagon transporting liquefied petroleum gas (LPG) that had already exploded and destroyed houses alongside the railway line. Separately, in the Avellino coach crash, the first police officers arrived at the crash epicenter without knowing the severity of the situation because the initial information received from the operations room was not alarming. In some cases, the officers were able to rescue injured people in the upper part of the viaduct but were unable to reach the people trapped in the crashed coach approximately 100 ft under the viaduct.

Furthermore, only in L'Aquila earthquake scenario did the double condition of the victims (i.e., all lived in L'Aquila and in the earthquake-affected areas) and rescuers emerge. Indeed, the L'Aquila interviews are characterized by the reenactment of the rescue experiences that occurred later; many operators were at home with their families during the earthquake. As is well-known, police officers consider situations in which their own family is threatened especially stressful [10]. During the interview, their concerns for the safety of their families, friends, and peers who were not in the line of duty at the time of the earthquake emerged frequently. Moreover, all experienced the death of a fellow officer due to the collapse of the officer's house. This double experience as victims and rescuers in this group produced selfexperiences that were significantly different from those produced in the other groups. Although the sample observed is small and the result must be considered with caution, we maintain that it may provide some useful guidelines for emergency rescuers in long-term stress management when rescuers have the double role of being direct and indirect victims.

Concerning the absence of significant difference in negative emotions and the vividness of the mental images recalled, no conclusions or explanations can be drawn because of the lack of statistical power.

Although the three groups are different in their experiences, all participants have been exposed to a traumatic experience, and 12 out of 15 participants reported other traumatic experience related to their job activity. Among the most traumatic experiences on the job included fatal accidents involving children or adolescents and self-perceived errors in death notifications. According to Colwell et al. [56], these types of traumatic experiences are reported in 27.2 and 1.3% of their sample, respectively, and in some way can be considered part of the risk of the job.

We found interesting results when investigating personality traits and coping strategies with respect to TSI and post-traumatic growth.

Specifically, we found that some coping strategies, such as emotion-oriented ones, predict the result in some TSI scales. In particular, emotion-oriented strategies predict performance with respect to the anxious/arousal scale, a scale of dysphoric mood states; the anger/irritability scale, a scale strictly related to the experience of psychological trauma; dissociation, a scale concerns the reexperiencing of the traumatic event; and tension-reduction behavior, a scale that examines difficulties with oneself and affects regulation.

Generally speaking, emotion-oriented strategies are considered proactive, and studies on coping associate these strategies with better adjustment, higher self-rated coping effectiveness, and lower levels of depression (e.g., [57, 58]). Indeed, when an individual adopts this type of coping style, he/she contrasts negative emotional responses with the stressors. Considering the time from the traumatic event (3 or 7 years), this type of coping strategy is surely more functional than problem-focused strategies, which were used by all rescuers during the emergency, as reported from the CI. Folkman and Lazarus [59], in introducing the functions of coping strategies, highlight that individuals use these strategies to manage problems causing stress and to govern emotions relating to the stressors. In the light of this definition, several studies consider a situation stressful whenever the individual perceives a lower ability to cope with the situation. Furthermore, the adopted coping strategies are strictly related to the way in which a stressor is evaluated, which means that a stressor that is perceived as controllable tends to elicit more proactive coping mechanisms [60], while a stressor perceived as uncontrollable tends elicit more avoidance strategies (e.g., [2, 61]).

The few studies on coping strategies among police officers seem to suggest that police officers have the tendency to use maladaptive emotion-focused behaviors for immediate stress reduction (e.g., [62]), while the avoidance coping strategy been more frequently associated with psychological stress [63]. In contrast, Ortega and colleagues [64] found that only personality and tenure were significantly related to coping strategy and the management of the occupational stress. In a recent study, Grubb et al. [37] found that the UK police officers use fewer maladaptive cognitive strategies to regulate their emotions and cope with stress, suggesting that they are adept at avoiding strategies that are negative and dysfunctional. Here, we found that only proactive coping strategies allowed the respondents to cope with the traumatic experience. We also found that only emotional stability for personality traits is a predictor of dissociation as evaluated by the TSI scales; this is in line with the proactive emotional coping strategies used in this research. Moreover, the conscientiousness of personality traits predicts the tension-reduction behaviors of TSI.

Another interesting result was that personality traits predict post-traumatic growth. In particular, we found that emotional stability and conscientiousness predicts new line possibilities; the conscientiousness and extraversion/energetic factor predict personal strength growth (specifically, extraversion/vivaciousness predicts growth in the negative direction); and emotional stability predicts spiritual changes. These findings are of interest because the process of healing may consume mental and emotional energy, which could explain the negative prediction of the extraversion/vivaciousness trait. Indeed, people with a tendency toward being energetic may experience the reduction of the available resources more than others. Conversely, emotional stability may help in the building of new relationships and consolidating strong personal ties. In the same direction, individuals with conscientiousness traits are well organized, methodical, and thorough; this trait fosters the growth of personal strength as well as new life possibilities.

1.3% of their sample, respectively, and in some way can be considered part of the risk of the

We found interesting results when investigating personality traits and coping strategies with

Specifically, we found that some coping strategies, such as emotion-oriented ones, predict the result in some TSI scales. In particular, emotion-oriented strategies predict performance with respect to the anxious/arousal scale, a scale of dysphoric mood states; the anger/irritability scale, a scale strictly related to the experience of psychological trauma; dissociation, a scale concerns the reexperiencing of the traumatic event; and tension-reduction behavior, a scale

Generally speaking, emotion-oriented strategies are considered proactive, and studies on coping associate these strategies with better adjustment, higher self-rated coping effectiveness, and lower levels of depression (e.g., [57, 58]). Indeed, when an individual adopts this type of coping style, he/she contrasts negative emotional responses with the stressors. Considering the time from the traumatic event (3 or 7 years), this type of coping strategy is surely more functional than problem-focused strategies, which were used by all rescuers during the emergency, as reported from the CI. Folkman and Lazarus [59], in introducing the functions of coping strategies, highlight that individuals use these strategies to manage problems causing stress and to govern emotions relating to the stressors. In the light of this definition, several studies consider a situation stressful whenever the individual perceives a lower ability to cope with the situation. Furthermore, the adopted coping strategies are strictly related to the way in which a stressor is evaluated, which means that a stressor that is perceived as controllable tends to elicit more proactive coping mechanisms [60], while a stressor perceived as uncon-

The few studies on coping strategies among police officers seem to suggest that police officers have the tendency to use maladaptive emotion-focused behaviors for immediate stress reduction (e.g., [62]), while the avoidance coping strategy been more frequently associated with psychological stress [63]. In contrast, Ortega and colleagues [64] found that only personality and tenure were significantly related to coping strategy and the management of the occupational stress. In a recent study, Grubb et al. [37] found that the UK police officers use fewer maladaptive cognitive strategies to regulate their emotions and cope with stress, suggesting that they are adept at avoiding strategies that are negative and dysfunctional. Here, we found that only proactive coping strategies allowed the respondents to cope with the traumatic experience. We also found that only emotional stability for personality traits is a predictor of dissociation as evaluated by the TSI scales; this is in line with the proactive emotional coping strategies used in this research. Moreover, the conscientiousness of person-

Another interesting result was that personality traits predict post-traumatic growth. In particular, we found that emotional stability and conscientiousness predicts new line possibilities; the conscientiousness and extraversion/energetic factor predict personal strength growth (specifically, extraversion/vivaciousness predicts growth in the negative direction);

job.

respect to TSI and post-traumatic growth.

that examines difficulties with oneself and affects regulation.

158 A Multidimensional Approach to Post-Traumatic Stress Disorder - from Theory to Practice

trollable tends elicit more avoidance strategies (e.g., [2, 61]).

ality traits predicts the tension-reduction behaviors of TSI.

Our results seem to suggest that in the coping process, the traumatic event is important, both in terms of personality and methods of coping with the event. In some ways, coping strategies have an important function in protecting the individual and promoting his/her resilience. Personality has an important role in the meaning that the individual gives to his/her posttraumatic life and his/her capability to cope with the trauma. This an important result that takes into account the fact that police officers, as a consequence of their job, are not able to avoid future situations that are similar to the traumatic event or other potentially traumatic events. In this sense, they may be vulnerable to re-victimization; moreover, the exposure to new stressors may make spontaneous recovery more demanding because they would overburden an already overloaded system [11, 36]. For this reason, a goal of future studies should be to better investigate the role of all protective factors that may help rescuers in terms of resiliency and enhancing their capability for post-traumatic growth.

In conclusion, we believe that our results—above all, those related to the role of emotional coping strategies as a protective factor in the prevention of post-traumatic stress disorders may have important implications in terms of training new police officers. Indeed, preventive measure could be implemented to enhance the adaptive and proactive coping strategies in the management of stress. The evidence that personality traits may play an important role in not developing traumatic symptoms as well as in positive post-traumatic growth is information of interest that may have an impact in terms of recruitment. Furthermore, the use of structured cognitive interviews for investigating traumatic events could become a useful tool for supporting and reducing stress reactions. Moreover, one of the innovations of this study is related to the long amount of time that elapsed from the traumatic event exposure and the structured cognitive interviews. For example, a previous study by Marchand et al. [65] involved interviews of police officers at between 5 and 15 days and at 1 month, 3 months, and 12 months after the event. Our study is the first to investigate the psychological long-term effects of the exposure to a traumatic experience. From this perspective, the data in this study show that emotion-coping strategies and some positive emotional and mental personality traits appear to be beneficial for police officers. In this vein, our study suggests that through a modified cognitive interview, veteran police officers may share their experiences with cadets during academic training to help them avoid being overwhelmed from unexpected emotions and jobrelated experience.

It is also important to highlight that, even if our sample is small, which can be perceived as a limitation, interviewing police officers about a severe traumatic event may be very difficult because they are often prone to underestimating the magnitude of their distress. Moreover, the events investigated occurred several years ago, and the rescuers who intervened at that time are now at different police stations throughout Italy. However, we managed to interview all police officers who intervened as first rescuers who are now retired; all participants understood the implications of their contributions in terms of the academic training of cadets. This collective participation should be more appreciated in consideration of the "virility" culture [66], which prevents officers from expressing their emotions and feelings. This culture can force police officers into the role of superheroes, an idea that is supported by both the citizens and the police officers themselves. "Virility" can be considered a defensive mechanism strategy and may have an important role in protecting police officers from a high level of anxiety that could prevent them from doing their job. The embodiment of the "virility" culture does not allow displays of weakness and traumatic reactions. To offset risk, police officers can become risk takers. Police officers are generally perceived by citizens as superheroes and they, in line with this image, attempt to not fail such expectations by displaying "weaknesses" such as traumatic reactions or the long-term effects of trauma. For this reason, they generally never show any effects of the trauma exposure and tend to minimize their distress.

The participation of policemen in this study is a signal that knowledge and scientific results are changing the minds and culture in the direction of more effective attitudes toward the prevention and promotion of health.

### **6. Conclusion**

The present study shows that emotion-coping strategies and some positive emotional and mental personality traits appear to play an important role as a protective factor for the development of post-traumatic stress disorder, as well as for the post-traumatic growth.

Moreover, it explores the possibility to use a modified cognitive interview to allow veteran police officers to share their experiences with cadets during academic training to help them avoid being overwhelmed from unexpected emotions and job-related experience. In spite of its limitations (the small number of participants for traumatic event and lack of statistical power for some analyses), the results seem to suggest the importance to use systematic measurements of coping strategies adopted as well as debriefing in the immediate aftermath (for small and large emergencies) and over the course of the following years (for large emergencies).
