Media use and insomnia after terror attacks in France

https://doi.org/10.1016/j.jpsychires.2017.12.006Get rights and content

Abstract

Direct exposure to traumatic events often precipitates sleep disorders. Sleep disturbance has also been observed amongst those indirectly exposed to trauma, via mass media. However, previous work has focused on traditional media use, rather than contemporary social media. We tested associations between both traditional and social media consumption and insomnia symptoms following 2015 terror attacks in Paris France, controlling for location and post-traumatic symptomology. 1878 respondents, selected to represent the national French population, completed an internet survey a month after the Bataclan attacks (response rate 72%). Respondents indicated different media use, post-traumatic stress and insomnia. Controlling for demographics, location and PTSD, insomnia was associated with both traditional (β 0.10, P = .001) and social media use (β 0.12, P = .001). Associations between social media and insomnia were independent of traditional media use. Interventions targeted at social media may be particularly important following mass trauma.

Introduction

A large body of evidence suggest that stress, and in particular, stress from direct exposure to trauma, is a precipitating factor for sleep disturbance (Bui et al., 2012, Harvey et al., 2003, Sinha, 2016). Studies with war veterans (Lewis et al., 2009) and individuals who experienced missile (Lavie, 2001), terror attacks (Galea et al., 2002, North et al., 1999) and natural disasters (Bui et al., 2012, McMillen et al., 2000) found increased difficulty in initiating sleep, returning to sleep after awakening, nightmares, and a general hyperarousal following traumatic events (Germain, 2013, Lavie, 2001, Sinha, 2016). A smaller literature has considered the impact of trauma on sleep amongst those indirectly exposed, primarily through the media. This includes associations between watching TV and dreams (following the Challenger disaster (Terr et al., 1999), and after 9/11 (Propper et al., 2007)), terror-related TV consumption and general sleep difficulties following missile file in Israel (Soffer-Dudek and Shahar, 2010), and reading internet news and disruptive nocturnal behaviour after the 2011 Great Japan earthquake (Bui et al., 2012). Most of this work has focused on engagement with traditional media; however, research following natural disaster suggests the immediacy and personalization offered by social media may have a greater psychological toll (Goodwin et al., 2013, Goodwin et al., 2017). As yet, the impact of these social media on sleep disturbance following trauma has been underexplored.

In November 2015 gunmen claiming allegiance to the Islamic State group attacked the Bataclan concert hall and restaurants in St. Denis, Paris, killing 130. Using data from a national survey (Goodwin et al., 2017), we consider the associations between traditional and social media use following the attack and subsequent insomnia. Because location (physical proximity to attack) has been positively associated with distress (Goodwin et al., 2017) we control for location within vs. outside of Paris. While the relationship between mass trauma and PTSD is complex, insomnia has been associated with acute psychological distress (Sinha, 2016, Germain, 2013, Krakow et al., 2015, Schoenfeld et al., 2012) primarily via the hyperarousal component of PTSD. We therefore report both associations between media use and insomnia, and media use and insomnia, controlling for the impact of post-traumatic symptomology on insomnia. This allowed us to assess the unique association of media use on insomnia, beyond the impact of distress.

Section snippets

Method

A major survey organization (https://www.surveygoo.com/) operates with a network of international panel surveys. We drew participants from their large French panel (approx. 500 000 individuals). Potential participants were sent a web link four weeks after the attacks (week of December 13, 2015) using validated addresses. Participants were selected using random stratified sampling, using weights for key demographic elements (age, gender), to create a nationally approximate representative sample

Measures

Insomnia was assessed using the Bergen insomnia scale (Pallesen et al., 2008), a six-item, seven-point scale assessing how many days per week, over the previous month, participants had problems with sleep onset, maintenance, early morning waking, non-restorative sleep, daytime impairment and dissatisfaction with current sleep (α = 0.89). PTSD was measured using the recently proposed ICD-11 scale (Cloitre et al., 2013). This scale comprises six items on a five-point scale (α = 0.91). Questions

Analyses

Preliminary statistical analyses report frequencies of media types used and associations of demographic background variables with different types of media use, PTSD symptoms, and insomnia symptoms including Pearson's Correlations and t-test. A priori multi-collinearity tests checked multi-collinearity amongst independent variables (Supplementary Table 2). Linear regressions tested associations between media use and insomnia, controlling for age, gender, location (Paris or elsewhere) and PTSD

Results

In the weekend after the attacks TV was the most widely used media, followed by radio and Facebook (Table 1 and Table S2). There were small positive correlations between distress and all forms of media use (Supplementary Table 2). Female respondents reported higher levels of insomnia (Ms 19.08 vs. 14.46, t (1872) = 8.38, P = .001), but there was no significant correlation between age and insomnia (r (1877) = −0.04, P = .054).

Discussion

Research on insomnia following traumatic events has focused on direct exposure to such events (Sinha, 2016). Media consumption, however, may prove significant for the national dispersal of stress beyond those directly impacted. In our data, media use in the immediate aftermath of attacks was significantly associated with insomnia a month after the event. Further, particular types of social media were more strongly associated with insomnia than traditional media formats. This association

Authors’ contributions

RG and MBE designed the study, obtained funding and organised data collection in France. RG and SL analysed the data; all authors drafted the initial paper and commented on and approved the final version. We would like to thank Dr. Anne-Cécile Hoyez for her help in translating the inventories, and for helpful comments from two anonymous reviewers.

Conflict of interest

None.

References (30)

  • M. Cloitre et al.

    Evidence for proposed ICD-11 PTSD and complex PTSD: a latent profile analysis

    Eur. J. Psychotraumatol.

    (2013)
  • L. Coviello et al.

    Detecting emotional contagion in massive social networks

    PLos One

    (2014)
  • S. Friedman

    Three Mile Island, Chernobyl and Fukushima: an analysis of traditional and new media coverage of nuclear accidents and radiation

    Bull At. Sci.

    (2011)
  • S. Galea et al.

    Posttraumatic stress disorder in Manhattan, New York City, after the September 11th terrorist attacks

    J. Urban Health

    (2002)
  • A. Germain

    Sleep disturbances as the hallmark of PTSD: where are we now?

    Am J Psychiatry

    (2013)
  • Cited by (0)

    View full text