Journal of Clinical Psychology in Medical Settings

, Volume 20, Issue 1, pp 56–63

Does War Hurt? Effects of Media Exposure After Missile Attacks on Chronic Pain

Authors

    • Department of Psychology, The Self and Health (SEALTH) LaboratoryBen-Gurion University of the Negev
  • Zvia Rudich
    • Division of Anesthesia and Intensive CareSoroka University Medical Center
  • Golan Shahar
    • Department of Psychology, The Self and Health (SEALTH) LaboratoryBen-Gurion University of the Negev
    • Department of PsychiatryYale University School of Medicine
Article

DOI: 10.1007/s10880-012-9313-4

Cite this article as:
Lerman, S.F., Rudich, Z. & Shahar, G. J Clin Psychol Med Settings (2013) 20: 56. doi:10.1007/s10880-012-9313-4

Abstract

This study focused on the effects of exposure to terrorist missile attacks on the physical and mental well being of chronic pain patients. In this prospective and longitudinal design, 55 chronic pain patients treated at a specialty pain clinic completed self-report questionnaires regarding their pain, depression and anxiety pre- and post a three week missile attack on the southern region of Israel. In addition, levels of direct and indirect exposure to the attacks were measured. Results of regression analyses showed that exposure to the attacks through the media predicted an increase in pain intensity and in the sensory component of pain during the pre-post war period, but did not predict depression, anxiety or the affective component of pain. These findings contribute to the understanding of the effects of terrorism on physical and emotional distress and identify chronic pain patients as a vulnerable population requiring special attention during terrorism-related stress.

Keywords

Chronic-painTerrorismDepressionAnxietyMedia-exposure

Introduction

War and terrorism have become part of daily life for a growing number of individuals throughout the world. People exposed to terrorism, either directly or indirectly (e.g., through the media), have been found to experience elevated emotional distress, including symptoms of posttraumatic stress disorder (PTSD) and depression (Bleich, Gelkopf, Melamed, & Solomon, 2006; Bleich, Gelkopf, & Solomon, 2003; Gelkopf, Solomon, Berger, & Bleich, 2008; Laugharne, Janca, & Widiger, 2007; Marshall et al., 2007; Schlenger et al., 2002; Soskolne, Baras, Palti, & Epstein, 1996). A few studies have focused on the effects of exposure to terrorism on physical health. Moric et al. (2007), for example, found an increase in health care utilization associated with elevated depression in chronic pain patients following the September 11th terrorist attacks. Others studies assessed the effects of the 9/11 attacks on conditions such as cardiovascular ailments and found that the individual’s acute stress response after the attacks predicted a decrease in cardiovascular health 3 years later (Holman et al., 2008). On the other hand, in a large survey conducted one and 2 years after the 9/11 attacks, Adams, Boscarino, and Galea (2006) reported a complex pattern of results as to the adverse effects of exposure to the events, whereby exposure was associated with poorer physical well being but not with poorer physical health.

This study focuses on the effects of exposure to missile terrorist attacks on chronic pain. Patients with chronic pain were postulated to be particularly vulnerable to the effects of terrorism, in part due to a disruption in their stress regulation system due to the prolonged and adverse experience of pain (Blackburn-Munro & Blackburn-Munro, 2001; Moric et al., 2007; Polatin, Young, Mayer, & Gatchel, 2005). Consequently, it might be expected that chronic pain patients might react with elevated physical and emotional distress to all types of stress, including to exposure to terrorism. Accordingly, Young et al. (2002) found that chronic pain patients experienced an increase in pain levels after the events of 9/11. Raphael, Natelson, Janal, and Nayak (2002), on the other hand, found an increase in general pain symptoms 6 months after the 9/11 attack, but reported that no worsening of fibromyalgia-like (FM-L) symptoms was shown in women who displayed FM-L before the attack.

Building on this line of inquiry, we utilized data collected as part of an ongoing study focusing on the links between chronic pain, depression, anxiety, and related variables (Lerman, Rudich, & Shahar, 2010; Lerman, Shahar, & Rudich, 2012; Rudich, Lerman, Gurevich, & Shahar, 2010; Rudich, Lerman, Gurevich, Weksler, & Shahar, 2008). In the course of data collection at Soroka University Medical Center, located in the Negev (Southern) region of Israel, an extensive military operation in this same area took place. In this operation, known as “Cast Lead”, a large number of missiles were launched from the Gaza Strip into the areas in and around Beer Sheva. Sirens warning of imminent attacks were heard regularly and missiles fell into populated and unpopulated areas. Physical injuries and loss of property were common occurrences. The Israeli and foreign media broadcasted news relating to this military operation during the day and night. These attacks lasted for 3 weeks from December 27, 2008 until January 18, 2009. This repetitive exposure posed an imminent threat which elicited significant stress.

Because we began collecting data prior to this event, we were in a very unique position to assess changes in pain and related distress as a function of various types of exposure to these missile attacks. This is noteworthy, because most studies on the effects of war and terrorism are usually conducted after the occurrence of the traumatic event, hence they are silent on matters of symptomatic change (but see Henrich & Shahar, 2008; Shahar, Cohen, Grogan, Barile, & Henrich, 2009). We were able to gather data from 55 chronic pain patients living in the area. Prior to the attacks, these patients had been assessed as to their pain, depression, anxiety, as well as on other factors (i.e., personality and social support) not pertinent to the present report. Following the attacks, we assessed these patients again regarding their pain, depression and anxiety, as well as their level of exposure to the missile attacks. We examined whether physical, relational, and media exposure to the attacks were predictive of changes in pain, depression, and anxiety.

Materials and Methods

Participants and Procedure

The study was approved by the Hospital Helsinki Committee for Human Subjects. One hundred and forty chronic pain patients treated at an outpatient specialty pain clinic at Soroka University Medical Center completed pre-“Cast Lead” (Time 1) assessment as part of a longitudinal study focusing on vulnerability to emotional distress in chronic pain patients. As part of this study patients were approached at the pain clinic during a scheduled appointment and were asked to participate in a study on psychological adjustment to chronic pain. Those who agreed signed an informed consent, and completed a wide range of questionnaires. Upon completion, patients were thanked, debriefed, and were given a nominal gift (a heating pillow). Inclusion criterion for participation in this study was continuous pain for at least 3 months and sufficient understanding of the Hebrew language. Following the “Cast Lead” operation, participants were contacted by phone and were asked to complete questionnaires over the phone regarding their direct and indirect exposure to the missile attacks and their levels of pain, depression and anxiety.

We were able to collect this post-“Cast Lead” (Time 2) operation data from 55 participants in the time frame of the 2 months we designated for this data collection. Of the 85 not assessed at this time point from our database, 21 declined to participate, and the remaining could not be reached either due to changed or disconnected phone numbers (N = 22) or because they were unavailable to be reached despite multiple calls (N = 42). The chronic pain diagnoses of the patients in our cohort were representative of the variety of chronic pain syndromes and included low back and cervical radiculopathy (N = 38), idiopathic pain (musculoskeletal pain, headaches) (N = 7), and neuropathic pain (N = 10).

The Time 1 assessment, conducted approximately 1 year prior to the “Cast Lead” operation, included measures of participants’ pain, depression, anxiety, coping strategies, social support, and personality. Time 2 assessment was conducted 1–2 months subsequent to the “Cast Lead” operation.

Measures

Subjective pain ratings were measured at Time 1 by the visual analog scale (VAS) and at Time 2 by a numerical rating scale (NRS) since the questionnaire was administered by phone: patients rated their pain today on a scale between 0 and 100, zero meaning no pain and 100 meaning the worst pain imaginable. Multidimensional pain experience was measured by the Short-form of McGill Pain Questionnaire (SF-MPQ; Melzack, 1987). On this measure participants were asked to rate to what extent their pain corresponded to 15 pain adjectives on a 4-point Likert scale ranging from 1 = ‘none’ to 4 = ‘severe’. This scale has a sensory subscale composed of 11 items and an affective subscale composed of four items. Both scales were used separately. Internal consistency of the Sensory scale was: Time 1: Cronbach’s α = .78; Time 2: Cronbach’s α = .88, and for the Affective scale was: Time 1: Cronbach’s α = .67; Time 2: Cronbach’s α = .71.

Depressive symptoms were assessed using the Center for Epidemiological Studies-Depression Scale (CES-D; Radloff, 1977). This is a 20 item self report questionnaire that assesses symptoms of depression in non-psychiatric adults. Subjects were asked to indicate on a 4 point scale how often they have felt or behaved in a certain way in the past 2 weeks from 0 = ‘very rarely’ to 3 = ‘most of the time’. Geisser, Roth, and Robinson (1997) found that this questionnaire discriminated significantly between persons with and without major depression in chronic pain patients. Internal consistency of Time 1 CES-D was: Cronbach’s α = .92 and for Time 2 was: Cronbach’s α = .86.

Anxiety was measured by the State Subscale of the Spielberger State-Trait Anxiety Inventory (STAI); Spielberger, Gorusch, & Lushene, 1970). This is a 20 item questionnaire in which subjects rate the extent to which statements correspond to the way they have been feeling in the past week on a 4-point Likert-type scale. Internal consistency of Time 1 STAI was: Cronbach’s α = .94 and for Time 2 was: Cronbach’s α = .95.

Exposure to the missile attacks was measured by a 22 item questionnaire which assessed four types of exposure: physical exposure (how close and how loud the missiles were from their location; five items), relational exposure (how many people they knew who had been physically hurt, emotionally hurt, or suffered property damage as a result of the bombing; three items), media exposure (the amount of terror-related television they watched following the bombing and their emotional reactions to the programming; five items) and stress exposure (to what degree they felt the missile attacks were stressful; one item). This questionnaire has been used in a number of studies to measure exposure to terrorism (Shahar et al., 2009; Soffer-Dudek & Shahar, 2010).

Statistical Analysis

All analyses were performed using the data analysis software system STATISTICA (StatSoft, 2007). In order to assess the effect of exposure to missile attacks on chronic pain patients we conducted five multiple regression analyses, each time predicting a different outcome measure. We used a two tailed α of .05 significance.

Results

Descriptive Data

Our final sample was comprised of 55 participants (a 39 % retention rate). Average age was 55.3 years (14.24 SD). Thirty-four were female (62 %). Average pain duration for the sample was 80 months (89.3 SD) and average education was 10.93 years (3.31 SD). In Table 1 we present the means, standard deviations and range of each measure in the study at both time points. When comparing participants who completed Time 2 (post-“Cast Lead” operation) assessment to those who did not, no significant difference was found in the Time 1 study measures. Participants who did not complete both waves of measurement were significantly more educated (p < .05, t = 2.26) but there was no other difference in demographic variables between the groups.
Table 1

Means, standard deviations and range of the study variables (N = 55)

Variables

T1 Mean

T1 SD

T1 Range

T2 Mean

T2 SD

T2 Range

Depression

28.20

14.21

2–56

28.89

14.58

0–53

Anxiety

48.65

15.16

25–76

47.79

16.7

20–80

Pain intensity (T1VAS; T2NRS)

62.72

31.03

0–100

52.20

33.15

0–100

Sensory pain

22.03

7.48

7–33

21.67

9.12

0–33

Affective pain

7.71

3.52

0–12

7.15

3.59

0–12

Physical exposure

.16

1.22

0–9

Relational exposure

.38

.73

0–3

Media exposure

11.58

6.08

0–20

Stress exposure

4.90

2.19

1–7

VAS visual analog scale, NRS numerical rating scale

Intercorrelations between the study variables are presented in Table 2. All of the outcome measures exhibited strong test–retest correlations: depression (r = .63), anxiety (r = .65), pain intensity (r = .35), sensory pain (r = .46) and affective pain (r = .42).
Table 2

Intercorrelations between the study variables (N = 55)

Variables

T1 depression

T1 anxiety

T1 pain intensity

T1 sensory pain

T1 affective pain

T2 depression

T2 anxiety

T2 pain intensity

T2 sensory pain

T2 affective pain

Relational exposure

Stress exposure

Media exposure

Age

Gender

Pain duration

Education

T1 depression

1.00

                

T1 anxiety

.83**

1.00

               

T1 pain intensity

.20

.09

1.00

              

T1 sensory pain

.37**

.23

.46**

1.00

             

T1 affective pain

.57**

.46**

.23

.66**

1.00

            

T2 depression

.63**

.54**

.09

.45**

.49**

1.00

           

T2 anxiety

.63**

.65**

.06

.35**

.39**

.81**

1.00

          

T2 pain intensity

.36**

.29*

.35**

.31*

.17

.48**

.47**

1.00

         

T2 sensory pain

.27

.23

.14

.46**

.19

.55**

.53**.

.41**

1.00

        

T2 affective pain

.58**

.48**

.15

.36**

.42**

.69**

.65**

.49**

.63**

1.00

       

Relational exposure

.32*

.29*

.20

.36**

.22

.27

.37**

.29*

.33*

.28

1.00

      

Stress exposure

.43**

.36**

.37**

.40**

.21

.52**

.46**

.45**

.41**

.39**

.29*

1.00

     

Media exposure

.37**

.37**

.06

.42**

.35**

.62**

.52**

.36**

.49**

.50**

.11

.67**

1.00

    

Age

−.19

−.16

−.06

.07

.14

.01

−.08

−.27

−.07

−.06

.09

−.26

.01

1.00

   

Gender

.11

.10

.05

.32*

.10

.33*

.40**

.19

.46**

.19

.26

.36**

.35**

.−01

1.00

  

Pain duration

−.09

−.01

.09

.07

−.11

.12

−.02

.02

.17

.06

.03

−.01

.01

.22

.11

1.00

 

Education

−.38**

−.21

−.13

−.14

−.49**

−.36**

−.23

−.08

−.04

−.23

−.10

−.18

−.30*

−.13

.02

.12

1.00

* α < .05; ** α < .01

Regression Analysis

Five multiple regression analyses were conducted to assess the effect of exposure to the missile attacks on different outcome measures: depression, anxiety, pain intensity, sensory and affective pain ratings. These were predicted from the three missile attack exposure measures (physical exposure was not used due to a very small variance, only one patient was directly exposed to the attacks), the baseline levels of the outcome variables (measured prior to the attacks), age, gender and education. Results of these regressions are presented in Table 3.
Table 3

Results of the five multiple regression analyses (N = 55)

Variable

Β

P

t

Regression 1: predicting depression

Adjusted R2 = .55, F[7, 42] = 9.74, p < .001

T1 depression

.43

.001*

3.41

Relational exposure

.03

.77

.28

Media exposure

.23

.08

1.76

Stress exposure

.12

.42

.79

Age

.07

.50

.67

Gender

.12

.25

1.15

Education

−.16

.14

−1.47

Regression 2: predicting anxiety

Adjusted R2 = .54, F[7, 41] = 9.24, p < .001

T1 anxiety

.55

.001*

4.78

Relational exposure

.13

.25

1.15

Media exposure

.23

.10

1.67

Stress exposure

−.03

.83

−.20

Age

−.07

.50

−.67

Gender

.20

.07

1.83

Education

−.06

.52

−.64

Regression 3: predicting affective pain

Adjusted R2 = .24, F[7, 43] = 3.30, p < .01

T1 Affective pain

.27

.06

1.87

Relational exposure

.18

.19

1.30

Media exposure

.34

.06

1.98

Stress exposure

−.07

.69

−.39

Age

−.14

.29

−1.05

Gender

.08

.50

.66

Education

−.14

.33

−.98

Regression 4: predicting sensory pain

Adjusted R2 = .34, F[7, 44] = 4.75, p < .001

T1 sensory pain

.27

.03*

2.23

Relational exposure

.14

.27

1.10

Media exposure

.34

.03*

2.22

Stress exposure

−.10

.53

−.62

Age

−.11

.37

−.89

Gender

.28

.03*

2.23

Education

−.02

.81

−.23

Regression 5: predicting pain intensity (NRS)

Adjusted R2 = .30, F[7, 42] = 4.04, p < .001

T1 pain intensity (VAS)

.32

.01*

2.49

Relational exposure

.21

.12

1.58

Media exposure

.44

.01*

2.58

Stress exposure

−.13

.49

−.68

Age

−.28

.04*

−2.11

Gender

.01

.98

.01

Education

.03

.77

.28

* Significant

Regression 1, predicting depression, accounted for 55 % of the explained variance (Adjusted R2 = .55, F[7, 42] = 9.74, p < .001). Only Time 1 depression was a significant predictor (β = .43, p < .001). Regression 2, predicting anxiety, accounted for 54 % of the explained variance (Adjusted R2 = .54, F[7, 41] = 9.24, p < .001). Again, only Time 1 anxiety was a significant predictor (β = .55, p < .001). Regression 3, predicting the affective dimension of pain accounted for 24 % of the explained variance (Adjusted R2 = .24, F[7, 43] = 3.30, p < .01) but there were no significant predictors. Regression 4, predicting the sensory dimension of pain, accounted for 34 % of the explained variance (Adjusted R2 = .34, F[7, 44] = 4.75, p < .001). Significant predictors were Time 1 sensory pain (β = .27, p < .05), gender (β = .28, p < .05; men more than women), and media exposure (β = .34, p < .05). Finally, regression 5, predicting pain intensity (NRS), accounted for 30 % of the explained variance (Adjusted R2 = .30, F[7, 42] = 4.04, p < .001). Significant predictors were Time 1 pain intensity (VAS) (β = .32, p = .01), age (β = −.28, p < .05), and media exposure (β = .44, p = .01).

Discussion

Utilizing a unique opportunity to prospectively examine the effects of a three week missile attack on the physical and mental well being of chronic pain patients, we found that media exposure predicted an increase in pain intensity and in the sensory component of pain during the pre-post war period, but did not predict an increase in depression, anxiety or the affective component of pain.

Our findings are consistent with previous ones, attesting to the adverse effects of indirect exposure to terrorism through the media but not with the specific effects of this exposure on mental health. For instance, a number of studies assessed the effect of media exposure after the 9/11 attacks and found that those who watched more media coverage were at risk for elevated levels of PTSD symptoms and depression (Ahern et al., 2002; Ahern, Galea, Resnick, & Vlahov, 2004; Marshall et al., 2007; Schlenger et al., 2002). However, to the best of our knowledge, no study has assessed the effects of media exposure on pain or physical health.

Although there is evidence of the negative effect of stress on pain and physical health (Chapman, Tuckett, & Song, 2008; Epel, 2009) few studies have explored the specific effect of war on physical pain in individuals who were not directly wounded. Chronic pain has been studied in the context of war and terrorism only in a handful of studies with mixed results. For example, Williams, Brown, Clauw, and Gendreau (2003) did not find a change in pain levels after the 9/11 attacks in a small sample of FM patients, however Young et al. (2002) did find an increase in pain in their sample of chronic pain patients. Whereas, as a group, our chronic pain patients did not exhibit a mean level increase in pain, depression or anxiety after the missile attacks, we found that those patients intensively exposed to the media did experience an increase in sensory pain and pain intensity.

Pain is a multidimensional phenomenon with sensory-discriminative and affective-motivational components. The sensory dimension pertains to the physical elements of pain such as intensity and location whereas the affective dimension relates to the negative emotions and beliefs patients attribute to their pain (Merskey & Bogduk, 1994). Our results indicate that the exposure to the missile attacks through the media had a differential effect on patients’ sensory and affective pain, wherein only sensory pain was affected. This coincides with our finding that media exposure did not predict changes in other mental health dimensions such as depression and anxiety.

Although our findings are not in line with previous research in this field pointing to the negative effects of media exposure on mental health, a possible explanation for our finding can be found in the reaction of the media and the “psycho education” they provided as to the emotional effects of the missile attacks. Television and newspapers in Israel were filled at that time with advice on regulating distress in the face of the threat, and this was conceptualized as a national challenge, thereby increasing a sense of belonging and social support, important buffers against the effect of stress on emotional distress. However, there was minimal mention of the effects on physical health, which might have caused individuals with chronic health conditions to feel alone and isolated with no information on a normal physical reaction for their situation and advice on adaptive coping mechanisms.

Prominent stress and coping theories are united in viewing the coping process as located within person-environment exchanges (Hobfoll, 1989, 2001; Lazarus & Folkman, 1984). Chronic pain patients are, by definition, coping with chronic stress (i.e., pain), as well as the effect of this stress on their functioning (Thorn & Dixon, 2007). The missile attacks and the blasting sirens proceeding them were undoubtedly stress provoking as were other missile attacks on Israel [for example see Henrich & Shahar, 2008; Soskolne et al., 1996)] but we did not find that the exposure to the attacks (how many people they know who were affected by the attacks and how stressful they perceived the attacks to be) predicted changes in patients’ mental or physical distress. Arguably, the exposure to images on television is sufficient as a reminder to these patients of the imminent threat and may be especially stressful for them, tipping the balance of the demands to that with which they can no longer cope. From the point of view of the Conservation of Resources theory of stress (COR; Hobfoll, 1989, 2001), fewer resources may have been left to cope with pain and adherence to medical regime, in turn causing elevated levels of pain. This may be especially true when considering the notion that chronic pain patients possess a biological vulnerability to stress thought to be caused by a disruption in their stress regulation system (Blackburn-Munro & Blackburn-Munro, 2001; Polatin et al., 2005).

An alternative explanation for the elevated physical, but not emotional, distress experienced in our sample could be that these patients express their psychological distress somatically. Chronic pain has been found to be associated with somatization and hypochondriasis (Fishbain, Lewis, Gao, Cole, & Steele Rosomoff, 2009), suggesting that chronic pain patients might communicate emotional distress through physical complaints. These patients are faced with a medical condition causing them pain on a day to day basis. When they are confronted with an external stressor such as missile attacks, it might be easier for them to continue expressing their distress in a physical “language” rather than an emotional one (Brio, 2010).

Our findings show that T1 depression and anxiety were associated with levels of stress and media exposure during the missile attacks. This association suggests that patients’ previous levels of emotional distress may affect their ability to cope with stressful situations, making stressors more salient (vis-à-vis stress exposure) and perhaps making them more likely to seek out more information about the situation (as reflected in media exposure). Stress and media exposure were strongly correlated, suggesting that the amount of television viewing related to the missile attacks may have influenced how much stress the individuals experienced or perhaps vise versa. Despite this strong relationship between the two types of exposure, the regression analyses revealed that media exposure was a stronger predictor of distress.

The strength of this study includes the assessment of baseline levels of the variables of interest about a year prior to the attack. Limitations include a relatively modest sample size and a small retention rate between the two time points. This was largely due to the time constraints we designated for data collection after the missile attack in order to accurately measure participants levels of exposure. An additional limitation is that the sole use of self-report questionnaires may have caused an inflation of shared variance. As well, patients were recruited for this study during a visit to a pain specialty clinic, thus the results pertain to individuals enrolled in pain management treatment program and might not be generalized to all community-residing individuals with chronic pain or to patients treated at other types of clinics such as rheumatology or orthopedics.

To conclude, this prospective and longitudinal study adds to the body of research on the effects of war and terrorism on suffering, specifically, the effect on individuals with chronic pain. Our results show that indirect exposure to terrorism had an adverse effect on the physical pain of chronic pain patients but not on their emotional distress. This emphasizes the importance of interventions that target this population.

Acknowledgments

This research was supported by THE ISRAEL SCIENCE FOUNDATION (Grant No. 751/08) to Golan Shahar (Principal Investigator) and Zvia Rudich (Co-Investigator).

Copyright information

© Springer Science+Business Media, LLC 2012