Pre-deployment threat learning predicts increased risk for post-deployment insomnia: Evidence from the Marine Resiliency Study
Introduction
Clinically significant insomnia, defined broadly as difficulties falling or staying asleep that interfere with daily functioning, is a common condition that can have serious adverse effects on physical and mental health (Roth, 2007; Cunnington et al., 2013). Among other consequences, insomnia has been linked to increased risk for new mental health conditions (Pigeon et al., 2017), exacerbation and/or poorer management of chronic medical conditions (Fernandez-Mendoza & Vgontzas, 2013), reduced quality of life (Olfson et al., 2018; Katz & McHorney, 2002), and premature death (Dew et al., 2003; Kripke et al., 2002). The burden of insomnia is particularly concerning among Veterans, where prevalence rates are as high as 50% and 2–3 times higher than the general population (Mysliwiec et al., 2013; Martin et al., 2017; Colvonen et al., 2020; Byrne et al., 2021). Moreover, rates of Veteran insomnia have been growing at an alarming rate in recent decades (Campbell & Riggs, 2015; Mysliwiec et al., 2013), with Veterans deployed in post-9/11 conflicts appearing particularly prone to sleep problems (Colvonoen et al., 2020).
Not surprisingly, the vast majority of Veteran insomnia cases are thought to develop following military deployment (Bramoweth & Germain, 2013). While there has been a great deal of research into the consequences of post-deployment insomnia (e.g., Hermes & Rosenheck, 2014; Kartal et al., 2021; Pigeon et al., 2012; Pigeon et al., 2013), surprisingly little research has been devoted to identifying predictors of post-deployment insomnia (Hughes et al., 2018). Indeed, only one previous investigation to our knowledge has attempted to predict post-deployment insomnia from pre-deployment data (Miller et al., 2021). Identifying factors that predispose Veterans for post-deployment insomnia is the first step toward understanding its etiology and preventing its development, which is critical given the cascade of adverse mental and physical consequences that may arise from post-deployment sleep issues (Hermes & Rosenheck, 2014; Kartal et al., 2021; Pigeon et al., 2012, 2013).
One factor that may be particularly important for predicting post-deployment insomnia is threat learning. Broadly speaking, threat learning refers to the ability to flexibly form, maintain, and extinguish fear responses to a particular stimulus depending on whether the stimulus has, has not been, or is no longer paired with an intrinsically aversive event. Military deployment can involve numerous instances of new threat learning where previously neutral activities (e.g., driving) are paired with aversive consequences (e.g., an IED explosion). Soldiers with a propensity for acquiring excessive, inflexible, or unnecessary threat responses may be more chronically aroused following aversive deployment experiences and thus be at higher risk for developing the persistent sleep-related arousal that often contribute to sleep problems among Veterans (Capaldi et al., 2011; Wright et al., 2011; Wallace et al., 2011; Bramoweth & Germain, 2013). This proposition is also consistent with prevailing models of post-deployment insomnia, which suggest that individuals who react excessively to or cope poorly with deployment-related stressors are at elevated risk for developing insomnia (Bramoweth & Germain, 2013; Hughes et al., 2018).
Additionally, there is a well-established link between disruptions in sleep-related processes (e.g., less total sleep time, lower proportion of rapid eye movement [REM] sleep) and abnormal threat learning (Straus et al., 2017; Colvonen et al., 2019; Schenker et al., 2021). While most studies have examined the effect of sleep on subsequent threat learning, evidence of the opposite pattern has been found in the few laboratory studies that have examined it – namely, that differences in threat learning constructs (e.g., heightened fear conditioning) predict changes in sleep quality the following night (Marshall et al., 2014; Straus et al., 2018). Heightened stress reactivity more broadly has also been linked to the development of insomnia, as individuals who experience more frequent cognitive intrusions and utilize more maladaptive coping strategies following stressful events have been found to be at elevated risk for future sleep problems (Drake et al., 2014; Pillai et al., 2014). Together, these data offer preliminary evidence that individual differences in threat learning could bias individuals toward or away from disrupted sleep following the stressful, combat-related experiences that often occur during military deployment.
Given the potential involvement of threat learning in the development of deployment-related insomnia, we investigated whether pre-deployment levels of threat learning could predict the incidence of post-deployment insomnia in a large longitudinal dataset of active-duty military members (i.e., the Marine Resiliency Study; Baker et al., 2012). Our study had three aims. Our first aim was to identify cross-sectional differences between participants with versus without insomnia at post-deployment across three key areas of threat learning: Threat conditioning (i.e., higher responding to a cue associated with an aversive stimulus), safety learning (i.e., lower responding to a cue that is not associated with an aversive stimulus), and fear extinction (i.e., lower responding to a cue that was formerly associated with an aversive unconditioned stimulus [US]). Consistent with research linking elevated threat learning to insomnia, we predicted that participants with insomnia would show greater fear conditioning, less safety learning, and less fear extinction at post-deployment relative to participants without insomnia.
Our second aim was to investigate whether pre-deployment levels of each insomnia-related difference in threat learning could predict post-deployment insomnia. This was done to test whether threat learning variables that were linked to insomnia at post-deployment acted as risk factors for insomnia or were merely consequences of post-deployment insomnia. If post-deployment insomnia was linked to a threat learning variable both cross-sectionally (i.e., at post-deployment) and longitudinally (i.e., at pre-deployment), it would suggest that abnormalities in that threat learning variable preceded the development of insomnia, consistent with the abnormality acting as a risk factor for post-deployment insomnia. However, if a threat learning variable was only linked to post-deployment insomnia cross-sectionally, it would imply that abnormalities in that threat learning variable arose after the development of insomnia (i.e., post-deployment), consistent with the abnormality being a consequence of post-deployment insomnia. Given the theory that heightened stress reactivity confers increased vulnerability for deployment-related insomnia (Bramoweth & Germain, 2013; Hughes et al., 2018), we hypothesized that threat learning variables would act as risk factors for post-deployment insomnia, as observed by levels of threat responses (fear potentiated startle, self-reported US expectancy) at pre-deployment predicting greater risk for insomnia at post-deployment.
Finally, our third aim was to examine potential moderators and mediators of the longitudinal threat learning-insomnia relationship. The purpose of this aim was to better understand the mechanisms that contribute to post-deployment insomnia. We hypothesized that the effects of pre-deployment threat learning on post-deployment insomnia would be moderated by the degree of combat exposure (i.e., higher combat exposure would be associated with a stronger effect) and mediated by higher levels of three post-deployment arousal symptoms (hypervigilance, startle, and nightmares). These hypotheses are consistent with theoretical models of post-deployment insomnia, wherein the propensity for heightened stress reactivity interacts with combat-related stress to produce excessive arousal, which in turn interferes with sleep (Bramoweth & Germain, 2013; Hughes et al., 2018).
Section snippets
Participants
Three infantry battalions of Marines and Navy Corpsmen (N = 1387) stationed in Southern California were enrolled in a longitudinal study of mental and physical health changes related to military deployment. One battalion (n = 546) did not deploy during the course of the study and was not included in the present analyses. Additionally, 21 participants did not complete the fear conditioning task at post-deployment, three failed to provide expectancy rating data for the first phase of the task,
Demographics
Demographic and clinical differences between participants with versus without post-deployment insomnia are shown in Table 1. At post-deployment, 21.5% of participants met criteria for insomnia disorder while 15.3% met criteria for lifetime insomnia when assessed at pre-deployment. Regarding rates of PTSD, 5.9% met criteria for PTSD at pre-deployment while 20.3% met criteria for PTSD at post-deployment. Prior to deployment, 86.7% of the sample reported having experienced a criterion A trauma as
Discussion
The purpose of the present study was to examine pre-deployment threat learning as a predictor of post-deployment insomnia and to test potential moderators and mediators of this effect. Consistent with expectations, post-deployment insomnia was associated with increased self-reported threat expectancy and fear potentiated startle to the conditioned danger cue, both across acquisition and during early extinction. Unexpectedly, only self-reported threat expectancy during acquisition at
Conclusions
The purpose of the present study was to investigate the prospective value of pre-deployment threat learning in predicting post-deployment insomnia, as well as examine whether this effect was moderated by combat exposure and mediated by post-deployment arousal symptoms. Consistent with expectations, service members with versus without insomnia showed a number of differences in threat learning at post-deployment. Moreover, one of these threat learning differences – higher pre-deployment threat
CRediT authorship contribution statement
Christopher Hunt: Conceptualization, Formal analysis. Daniel M. Stout: Conceptualization, Formal analysis. Dean Acheson: Conceptualization, Formal analysis. Peter J. Colvonen: Conceptualization, Formal analysis. Caroline M.Nievergelt: Data collection and analysis. Kate A.Yurgil: Data collection and analysis. Dewleen G. Baker: Funding acquisition, Writing – review & editing. Victoria B. Risbrough: Conceptualization, Formal analysis.
Declaration of competing interest
None.
Acknowledgments
Support for this work includes NIMH P50MH096889 (DGB, VBR), DOD (VBR) VA Merit Awards and Research Scientist Award (VBR 5I01BX004312; 1I21BX005872; IK6BX006186), the Office of Academic Affiliations Advanced Fellowship Program in Mental Illness Research and Treatment (CH), and the Center of Excellence for Stress and Mental Health (CH, DA, DS, DGB, CMN, VBR). The Marine Resilience Study was originally funded by project No. SDR 09–0128 (Drs Baker and Risbrough) from the Veterans Administration
References (62)
- et al.
Conditioned fear and extinction learning performance and its association with psychiatric symptoms in active duty Marines
Psychoneuroendocrinology
(2015) - et al.
Rating data are underrated: Validity of US expectancy in human fear conditioning
Journal of Behavior Therapy and Experimental Psychiatry
(2013) - et al.
Influence of chronic corticosterone and glucocorticoid receptor antagonism in the amygdala on fear conditioning
Neurobiology of Learning and Memory
(2004) - et al.
Prior exposure to a single stress session facilitates subsequent contextual fear conditioning in rats: Evidence for a role of corticosterone
Hormones and Behavior
(2003) - et al.
Prevalence, pharmacotherapy and clinical correlates of diagnosed insomnia among Veterans Health Administration service users nationally
Sleep Medicine
(2014) - et al.
Insomnia in United States military veterans: An integrated theoretical model
Clinical Psychology Review
(2018) - et al.
Sex differences in fear conditioning in posttraumatic stress disorder
Journal of Psychiatric Research
(2013) - et al.
Stress differentially modulates fear conditioning in healthy men and women
Biological Psychiatry
(2006) - et al.
Cross-lagged relationships between insomnia and posttraumatic stress disorder in treatment-receiving veterans
Behavior Therapy
(2021) - et al.
Estimated prevalence of insomnia among women veterans: Results of a postal survey
Women's Health Issues
(2017)