Is TV Traumatic for All Youths? The Role of Preexisting Posttraumatic-Stress Symptoms in the Link Between Disaster Coverage and Stress

Abstract

In youths, watching TV coverage of a disaster is associated with traumatic-stress symptoms. However, the role of predisaster symptoms in this link has not been addressed. In this study, urban-school youths who had experienced both Hurricanes Katrina and Gustav (N = 141; grades 4–8) were assessed 12 months and 6 months before Gustav and then 1 month after Gustav. The amount of TV viewing was associated with post-Gustav stress symptoms, controlling for pre-Gustav symptoms. However, pre-Gustav stress symptoms interacted with TV viewing in predicting post-Gustav symptoms such that for youths with higher preexisting symptoms, there was a stronger association between TV viewing and level of post-Gustav symptoms. The results advance the literature on the role of media coverage in stress reactions by showing that preexisting symptoms can be an important component of identifying which children are likely to be most negatively affected by TV coverage.

Keywords

mass media posttraumatic stress disorder

The role of the media in youths’ behavior continues to be a contentious issue in psychological science, with criticism often focusing on the need to better understand the conditions under which media influence behavior (Anderson et al., 2010; Ferguson & Kilburn, 2010). The effect of watching media coverage on youths’ stress reactions to disaster is one growing area of interest (Comer & Kendall, 2007; Pfefferbaum et al., 2001; Pfefferbaum et al., 2003). Research suggests that exposure to natural disasters is associated with a number of posttraumatic-stress reactions in youths (e.g., Eisenberg & Silver, 2011; La Greca, Silverman, Vernberg, & Roberts, 2002; Norris, Friedman, & Watson, 2002; Weems & Overstreet, 2008). One of the most common reactions involves the symptoms associated with posttraumatic stress disorder (PTSD), which include negative reexperiencing, avoidance, emotional numbing, and hyperarousal (American Psychiatric Association, 1994). A number of organizations have issued statements in light of the potential role of the media in disaster-related stress responses (see Hagan & American Academy of Pediatrics, Committee on Psychosocial Aspects of Child and Family Health, and Task Force on Terrorism, 2005; National Center for Child Traumatic Stress Network, 2006; Substance Abuse and Mental Health Services Administration, 2007). For example, the American Academy of Pediatrics recommends that pediatricians “help parents recognize the potential deleterious effects of indirect disaster exposure from media” (Hagan et al., 2005, p. 794).

Past research does suggest that watching more TV coverage of a disaster is associated with more negative stress responses, including PTSD symptoms (Comer & Kendall, 2007; Hagan et al., 2005; Pfefferbaum et al., 2001; Pfefferbaum et al., 2003). For example, following the terrorist attack in Oklahoma City, reports of exposure to broadcast media coverage of the bombing were correlated (r = .33) with PTSD symptoms in youths assessed 2 years after the attack (n = 88 sixth graders; analyses excluded children with direct exposure to the attack; Pfefferbaum et al., 2003). Similar results have been found regarding the September 11 attacks (e.g., Lengua, Long, Smith, & Meltzoff, 2005; Schuster et al., 2001). However, past studies have not both employed prospective designs and controlled for predisaster symptom levels. Consequently, the reported linkage between TV viewing and PTSD symptoms following disasters may simply be due to preexisting symptoms. This limitation in past research is due, in part, to the unpredictability of disasters, which makes the collection of predisaster data difficult. In one of the only pre- to postdisaster studies of media coverage, Propper, Stickgold, Keeley, and Christman (2007) analyzed the stress-related dream content of university students (N = 14) before and after the September 11 attacks. Stress-related dream content before the attacks was not related to amount of TV viewing, but postattack dream content was correlated with amount of TV viewing. However, Propper et al. did not control for dream content before the attacks in examining the association of TV viewing with postattack dream content. They concluded that their results provided “disturbing evidence that the media may have a deleterious impact” (p. 340).

Theoretically, watching TV coverage of disasters may leave youths with distressing images that foster a stress response, increase threat perceptions, and thus increase PTSD symptoms. Alternatively, preexisting PTSD symptoms might be responsible for the association between postdisaster TV viewing and PTSD symptoms (e.g., preexisting symptoms might increase viewing). In the latter case, controlling for predisaster symptoms might reveal that there was no direct postdisaster association between TV viewing and PTSD symptoms. Comer and Kendall (2007) suggested that preexisting conditions might potentiate distress experienced from watching media coverage. Therefore, TV viewing might interact with preexisting symptoms so that it primarily harms those individuals with existing stress-related problems. Similarly, TV viewing might serve to maintain or exacerbate existing stress. Indeed, trait anxiety was found to interact with the amount of TV news viewed to predict personal threat perceptions in a cross-sectional, non-trauma-exposed community sample of youths age 7 through 13 years (N = 90; Comer, Furr, Beidas, Babyar, & Kendall, 2008).

Drawing on these theoretical possibilities and past research, we predicted that the amount of TV disaster coverage viewed would be associated with PTSD symptoms following a disaster, even after we controlled for predisaster PTSD symptoms, perceptions of harm to self, and distress or fear during the disaster. We also tested whether watching media coverage interacts with preexisting PTSD symptoms, predicting that the relationship between watching TV coverage and postdisaster PTSD symptoms would be greatest for those individuals with elevated preexisting PTSD symptoms.

Method

This study utilized a prospective design. The participants were 141 fourth through eighth graders who were attending a single school in a New Orleans neighborhood that had experienced massive damage and flooding after Hurricane Katrina and who were part of an existing longitudinal study. The present sample was evaluated for PTSD symptoms 24 (Time 1) and 30 (Time 2) months after Katrina (1 year and 6 months before Hurricane Gustav, respectively); both PTSD symptoms and viewing of TV coverage of Gustav were evaluated 1 month (Time 3) after Hurricane Gustav. (Findings from Times 1 and 2 were reported by Weems et al., 2010; 41% of that study’s sample, which was the same as the present sample, reported moderate to severe Time 1 PTSD symptoms.) Gustav caused serious damage and casualties in the Caribbean and triggered one of the largest evacuations in U.S. history (more than 3 million people), making landfall (August 31, 2008) along the Louisiana coast with winds of 105 miles per hour. The sample at Time 3 was ages 9 through 16 years (median age = 12) and was 52% male and 48% female. The youths reported their ethnicities as follows: Ninety-six percent were African American, 3.5% were mixed African American or “other,” and 0.5% were Hispanic. The school served mainly low-income families (97% received free lunch, 2% gave reduced payment, and only 1% paid full price). Comparison of the youths who completed the Times 1 and 2 assessments only (n = 191) and those who also completed the Time 3 assessment (n = 141, or 74% retention) revealed no statistically significant differences in gender, ethnicity, level of distress or fear during Katrina, or Times 1 and 2 PTSD symptoms.

Symptoms of PTSD were measured using a modified version of the University of California at Los Angeles Posttraumatic Stress Disorder Reaction Index (UCLA-PTSD RI; Frederick, Pynoos, & Nadar, 1992; La Greca, Silverman, Vernberg, & Prinstein, 1996) in which participants were asked to rate their symptoms in response to hurricanes. To assess perceptions of self-harm, we asked participants, “Did you think you would get hurt during Gustav?” (yes = 1, no = 0). To measure overall distress or fear during Hurricane Gustav, we asked, “Overall how scared were you during Hurricane Gustav?” To assess viewing of TV coverage of Gustav, we drew from previous studies assessing TV exposure to disaster coverage (i.e., Lengua et al., 2005; Pfefferbaum et al., 2003) and asked participants, “Did you watch TV coverage of Hurricane Gustav?” Participants responded to both of these questions using the following 4-point scale: 0 = not at all, 1 = a little, 2 = a lot, and 3 = a whole lot.

Data collection was conducted as part of the school’s counseling curriculum. Each school year, parents provided written informed consent for the counseling curriculum and for use of the data in research (> 90% response rate). Children provided oral consent. Therefore, at all three waves, data were collected directly in school classrooms, and so the main reason for attrition at Time 3 was that students moved to a different school. The institutional review board of the University of New Orleans reviewed and approved the procedures for the study. Youths completed the measures in a group classroom setting and were assisted by trained staff. Staff members read the instructions and each item to the younger children, and trained staff helped individual children as necessary, as in previous studies (e.g., La Greca et al., 1996).

Results

Mean UCLA-PTSD RI scores were as follows (standard deviations in parentheses): 23.94 (17.6), 23.85 (15.2), and 17.08 (14.7) for Times 1, 2 and 3, respectively. Intraclass correlation coefficients (ICCs; absolute agreement) indicated a fairly high level of stability in PTSD symptoms (Time 1 to Time 2 ICC = .55; Time 1 to Time 3 ICC = .51; Time 2 to Time 3 ICC = .54). Regarding the amount of Gustav TV coverage watched, 11% of the sample indicated they had not watched it at all, 33% indicated they had watched “a little,” 25% indicated they had watched “a lot,” and 31% indicated they had watched “a whole lot.” The amount of Gustav TV coverage watched was significantly correlated with Time 3 PTSD symptoms (r = .19, p < .05), but not with Time 1 and Time 2 PTSD symptoms (both rs < .11). Regression analysis tested if TV viewing was associated with Time 3 PTSD symptoms (after Gustav), while controlling for both Time 1 and Time 2 PTSD symptoms as well as for perception of self-harm and distress. For these analyses, the sample was dichotomized into two groups on the basis of the amount of TV coverage viewed; the high-viewing group (coded as 1) consisted of children who reported viewing “a lot” or “a whole lot” of Gustav coverage, and the low-viewing group (coded as 0) consisted of children who reported viewing no or “a little” Gustav coverage.¹ Results of this analysis (Table 1) indicated that Time 3 PTSD symptoms were significantly predicted by preexisting (both Time 1 and Time 2) PTSD symptoms, perception of self-harm, and TV viewing.

Table 1.

Results of the Regression Analysis Predicting Time 3 Posttraumatic Stress Disorder (PTSD) Symptoms (After Gustav)

Step and predictor	Slope	t	Semipartial r
Step 1
Time 1 PTSD symptoms (before Gustav)	0.22 (0.06)	3.56*	.22
Time 2 PTSD symptoms (before Gustav)	0.40 (0.07)	5.57**	.34
Amount of TV coverage of Hurricane Gustav watched	4.18 (1.87)	2.23*	.14
Perceptions of self-harm	9.34 (3.46)	2.70*	.17
Fear during the disaster	1.76 (1.07)	1.66	.10
Step 2
TV Coverage × Time 1 PTSD Symptoms	0.32 (0.13)	2.56*	.15
TV Coverage × Time 2 PTSD Symptoms	−0.16 (0.15)	−1.10	−.06

Note: Standard errors are presented in parentheses. The measure of PTSD symptoms was score on the University of California at Los Angeles Posttraumatic Stress Disorder Reaction Index (Frederick, Pynoos, & Nadar, 1992; La Greca, Silverman, Vernberg, & Prinstein, 1996). The regression model explained a significant amount of variance, full-model R² = .53; adjusted R² = .50; F(7, 138) = 20.71, p < .001.

p < .05. **p < .001.

Interaction terms were added in a second step of the regression analysis to test the hypothesis that the amount of TV coverage of Gustav that participants viewed (high vs. low) interacted with preexisting PTSD symptoms (at Time 1 and Time 2) to predict postdisaster PTSD symptoms. Results indicated a significant interaction between PTSD symptoms at Time 1 and TV viewing (see Table 1). The nature of the interaction is depicted in Figure 1, which shows Time 3 PTSD symptoms as a function of the amount of TV coverage watched and Time 1 PTSD symptoms. Post hoc probing of the interaction (as suggested by Holmbeck, 2002) indicated that there was a stronger association between Time 1 and Time 3 PTSD symptoms among participants who had watched a large amount of TV coverage of Gustav (β = 0.69, p < .01) than among those who had watched a smaller amount (β = 0.40, p < .01). Moreover, as Figure 1 illustrates, there appeared to be no association between TV viewing and PTSD symptoms after Gustav among youths with low levels of PTSD symptoms before Gustav. Post hoc probing of the interaction with Time 1 PTSD symptoms as the moderator of the relation between TV viewing and Time 3 PTSD symptoms was consistent with this possibility. The association between TV viewing and Time 3 PTSD symptoms (after Gustav) was significant for those youths with high Time 1 PTSD symptoms (β = 0.27, p < .01) but not for those with low Time 1 PTSD symptoms (β = −0.02, p > .1).

Fig. 1.

Time 3 (after Gustav) posttraumatic stress disorder (PTSD) symp- toms (i.e., score on the University of California at Los Angeles Posttraumatic Stress Disorder Reaction Index; Frederick, Pynoos, & Nadar, 1992; La Greca, Silverman, Vernberg, & Prinstein, 1996) as a function of Time 1 (before Gustav) PTSD symptoms (low = 1 SD below the mean, high = 1 SD above the mean) and amount of TV coverage viewed.

Discussion

This study helps to empirically demonstrate that the link between TV viewing of disasters and PTSD symptoms in youths is not simply due to preexisting symptoms. However, preexisting symptoms may help researchers understand the role of watching media coverage in youths’ stress reactions. As Figure 1 shows, the present data suggest that among youths with elevated preexisting symptoms, those who watch more media coverage of a disaster are more likely to maintain their symptoms, and that the negative effect of watching a large amount of media coverage of a disaster is greatest for those youths who have preexisting difficulties. These results are consistent with previous cross-sectional research regarding youths who had not been exposed to disasters or trauma (Comer, Furr, Beidas, Babyar, & Kendall, 2008). As far as we are aware, no studies have established prospectively the link between TV viewing and preexisting PTSD symptoms in samples that have experienced a disaster.

The present findings identify a condition (having elevated preexisting symptoms) in which watching media coverage may be most detrimental for youths but also raise questions about the cognitive or emotional mechanisms (mediators) whereby the media exert their influence. Thus, a next step might be, for example, to examine cognitive biases prompted by media coverage and their role as possible mediators of the link between media viewing and stress reactions. The present findings also are relevant to postdisaster intervention efforts targeting the media literacy of parents of youths. Such programs have good potential for ameliorating and preventing stress reactions after disasters (Comer, Furr, Beidas, Weiner, & Kendall, 2008), and this study suggests that such efforts can be particularly effective if they target parents and families of youths with preexisting symptoms. This study also highlights the importance of preexisting susceptibility (e.g., trait anxiety, preexisting stress symptoms) in youths’ reactions to disaster (La Greca, Silverman, & Wasserstein, 1998; Weems et al., 2007). In the present sample, pre-Gustav PTSD symptoms predicted a large amount of variance (see the semipartial correlations in Table 1) in post-Gustav symptoms.

Although this study has made several methodological advances, including the use of prospectively collected data, it is not without limitations. The main one is that this study does not allow causal conclusions about the linkages observed. Moreover, all of the data came from youths’ reports. Although youths have consistently been found to be valid reporters of their own internalizing distress (see, e.g., Weems, Zakem, Costa, Cannon, & Watts, 2005), reports from other sources (such as parents) might have yielded different findings. Time 2 PTSD symptoms did not interact with TV viewing. It may be that these PTSD symptoms, which were assessed closer in time to the disaster, exerted only a main effect (e.g., the semipartial correlations for Time 1 and Time 2 were .22 and .34, respectively; Table 1). Moreover, although our strategy for assessing TV viewing was similar to the strategies used in other work on disasters and media coverage (Lengua et al., 2005; Pfefferbaum et al., 2003), the approach here may measure subjective appraisals of viewing more than actual disaster-related TV consumption. Note also that we did not measure exposure to other forms of media, such as videos on the Internet. Youths’ subjective appraisals of disaster-related media consumption may not correspond to quantitative measures of consumption, and “a lot” of viewing may convey different meanings to different children.

Another limitation is that there are likely a number of intermediary factors, not assessed in this study, that could influence the link between TV viewing and PTSD symptoms (e.g., parental monitoring and communication with the child). Yet no single study can capture all the potential influences, and given that our analyses controlled for both preexisting symptoms and level of fear or perceived potential for self-harm, it is clear that the link we observed was not due simply to media exposure level or preexisting symptoms. Finally, a limitation to generalizing our results to other populations is that the study sample consisted predominantly of minority youths. Although the nature of the sample is a positive feature of our study, in that it extends the literature to a population that has been understudied (Rabalais, Ruggiero, & Scotti, 2002), caution is necessary in generalizing the present findings to other samples of youths.

In summary, the present results clarify the role of media coverage in stress reactions to disasters: Although preexisting symptoms did not account for the linkages between TV viewing and post-disaster PTSD symptoms, they are a potentially important factor in identifying which children are likely to be most negatively affected by watching TV coverage of a disaster. Future research on the role of the media may benefit from examining potential moderators so as to contextualize findings of main effects. Finally, practitioners with young patients who have anxiety disorders such as PTSD may wish to emphasize to parents the potential effects of media.

Footnotes

Declaration of Conflicting Interests

The authors declared that they had no conflicts of interest with respect to their authorship or the publication of this article.

Supplemental Material

Additional supporting information may be found at

Notes

References

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author.

Anderson

C. A.

Shibuya

Ihori

Swing

E. L.

Bushman

B. J.

Sakamoto

. . . Saleem

(2010). Violent video game effects on aggression, empathy, and prosocial behavior in Eastern and Western countries. Psychological Bulletin, 136, 151–173.

Comer

J. S.

Furr

J. M.

Beidas

R. S.

Babyar

Kendall

P. C.

(2008). Media use and children’s perceptions of societal threat and personal vulnerability. Journal of Clinical Child & Adolescent Psychology, 37, 622–630.

Comer

J. S.

Furr

J. M.

Beidas

R. S.

Weiner

C. L.

Kendall

P. C.

(2008). Children and terrorism-related news: Training parents in coping and media literacy. Journal of Consulting and Clinical Psychology, 76, 568–578.

Comer

J. S.

Kendall

P. C.

(2007). Terrorism: The psychological impact on youth. Clinical Psychology: Science and Practice, 14, 179–212.

Eisenberg

Silver

R. C.

(2011). Growing up in the shadow of terrorism: Youth in America after 9/11. American Psychologist, 66, 468–481.

Ferguson

C. J.

Kilburn

(2010). Much ado about nothing: The misestimation and overinterpretation of violent video game effects in Eastern and Western nations: Comment on Anderson et al. (2010). Psychological Bulletin, 136, 174–178.

Frederick

C. J.

Pynoos

R. S.

Nadar

(1992). Reaction index to Psychic Trauma Form C (Child). Los Angeles: University of California.

Hagan

J. F.

Jr. , & American Academy of Pediatrics, Committee on Psychosocial Aspects of Child and Family Health, and Task Force on Terrorism. (2005). Psychosocial implications of disaster or terrorism on children: A guide for the pediatrician. Pediatrics, 116, 787–795.

10.

Holmbeck

G. N.

(2002). Post-hoc probing of significant moderational and meditational effects in studies of pediatric populations. Journal of Pediatric Psychology, 27, 87–96.

11.

La Greca

A. M.

Silverman

W. K.

Vernberg

E. M.

Prinstein

(1996). Symptoms of posttraumatic stress after Hurricane Andrew: A prospective study. Journal of Consulting and Clinical Psychology, 64, 712–723.

12.

La Greca

A. M.

Silverman

W. K.

Vernberg

E. M.

Roberts

M. C.

(Eds.). (2002). Helping children cope with disasters and terrorism. Washington, DC: American Psychological Association.

13.

La Greca

A. M.

Silverman

W. K.

Wasserstein

S. B.

(1998). Children’s predisaster functioning as a predictor of posttraumatic stress following Hurricane Andrew. Journal of Consulting and Clinical Psychology, 66, 883–892.

14.

Lengua

L. J.

Long

A. C.

Smith

K. I.

Meltzoff

A. N.

(2005). Pre-attack symptomatology and temperament as predictors of children’s responses to the September 11 terrorist attacks. Journal of Child Psychology and Psychiatry, 46, 631–645.

15.

National Center for Child Traumatic Stress Network, National Center for PTSD, U.S. Department of Veterans Affairs. (2006). Psychological first aid: Field operations guide (2nd ed.). Retrieved from http://www.nctsn.org/sites/default/files/pfa/english/1-psyfirstaid_final_complete_manual.pdf

16.

Norris

F. H.

Friedman

M. J.

Watson

P. J.

(2002). 60,000 disaster victims speak: Part II. Summary and implications of the disaster mental health research. Psychiatry, 65, 240–260.

17.

Pfefferbaum

Nixon

S. J.

Tivis

R. D.

Doughty

D. E.

Pynoos

R. S.

Gurwitch

R. H.

Foy

D. W.

(2001). Television exposure in children after a terrorist incident. Psychiatry, 64, 202–211.

18.

Pfefferbaum

Seale

Brandt

Pfefferbaum

Doughty

Rainwater

(2003). Media exposure in children one hundred miles from a terrorist bombing. Annals of Clinical Psychiatry, 15, 1–8.

19.

Propper

R. E.

Stickgold

Keeley

Christman

S. D.

(2007). Is television traumatic? Dreams, stress, and media exposure in the aftermath of September 11, 2001. Psychological Science, 18, 334–340.

20.

Rabalais

A. E.

Ruggiero

K. J.

Scotti

J. R.

(2002). Multicultural issues in the response of children to disasters. In La Greca

A. M.

Silverman

W. K.

Vernberg

E. M.

Roberts

M. C.

(Eds.), Helping children cope with disasters and terrorism (pp. 73–99). Washington, DC: American Psychological Association.

21.

Schuster

M. A.

Stein

B. D.

Jaycox

L. H.

Collins

R. L.

Marshall

G. N.

Elliott

M. N.

. . . Berry

S. H.

(2001). A national survey of stress reactions after the September 11, 2001, terrorist attacks. New England Journal of Medicine, 345, 1507–1512.

22.

Substance Abuse and Mental Health Services Administration. (2007). Building bridges: Mental health consumers and representatives of the disaster response community in dialogue (HHS Pub. No. 4250). Retrieved from Center for Mental Health Services Web site: http://store.samhsa.gov/shin/content/SMA07-4250/SMA07-4250.pdf

23.

Weems

C. F.

Overstreet

(2008). Child and adolescent mental health research in the context of Hurricane Katrina: An ecological needs-based perspective and introduction to the special section. Journal of Clinical Child & Adolescent Psychology, 37, 487–494.

24.

Weems

C. F.

Pina

A. A.

Costa

N. M.

Watts

S. E.

Taylor

L. K.

Cannon

M. F.

(2007). Pre-disaster trait anxiety and negative affect predict posttraumatic stress in youth after Hurricane Katrina. Journal of Consulting and Clinical Psychology, 75, 154–159.

25.

Weems

C. F.

Taylor

L. K.

Cannon

M. F.

Marino

Romano

D. M.

Scott

B. G.

. . . Triplett

(2010). Posttraumatic stress, context, and the lingering effects of the Hurricane Katrina disaster among ethnic minority youth. Journal of Abnormal Child Psychology, 38, 49–56.

26.

Weems

C. F.

Zakem

Costa

N. M.

Cannon

M. F.

Watts

S. E.

(2005). Physiological response and childhood anxiety: Association with symptoms of anxiety disorders and cognitive bias. Journal of Clinical Child & Adolescent Psychology, 34, 712–723.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.10 MB