Serving as one of the five armed forces of the United States, the Coast Guard (CG) is unique in having domestic and international missions. The CG conducts 11 missions: aids in navigation, defense readiness, drug interdiction, ice operations, living marine resources, marine environmental protection, marine safety, migrant interdiction, port security, search and rescue, and other law enforcement. In Boat Stations, CG military service bears resemblance to civilian law enforcement: high risk missions, a high degree of unpredictability regarding when a mission is required and its degree of difficulty. Like firefighters, many CG personnel serving at Boat Stations have rotating shift schedules mixing on-duty and off-duty days on a weekly basis. Thus, active duty at Boat Stations represents a mixture of military and civilian life experiences.

Military stressors have long been associated with mental health problems, particularly posttraumatic stress disorder (PTSD) and major depressive disorder (MDD). Accordingly, the increased risk for PTSD and MDD among the larger services arising from deployment and attendant combat service are well-documented (Shen et al., 2010; Mayo et al., 2013; Hines et al., 2014; Stander et al., 2014; Mustillo et al., 2015). In parallel, studies of PTSD and MDD in civilian police and firefighters focus on critical incidents such as World Trade Center (Perrin et al., 2007; Biggs et al., 2010; Bowler et al., 2010) and Hurricane Katrina (Tak et al., 2007; West et al., 2008). These studies are in keeping with the view extreme stressors and traumas have a direct effect on PTSD expression.

CG personnel are an understudied population; prevalence rates for PTSD and MDD are largely unknown. A comparison between the different services may be made using hospitalizations for stress-related mental health disorders as a guide. Surveillance of hospitalizations for PTSD finds rates are highest for Army and Marines, lower among the Navy and Air Force, and lower still for CG (Armed Forces Health Surveillance Center, 2013). Similarly, rates of hospitalizations for MDD among CG personnel are about half that of Army, and similar to Air Force, Navy, and Marines (Armed Forces Health Surveillance Center, 2013). Active duty military are notoriously averse to seeking mental health support (Iversen et al., 2011), thus hospitalizations may severely underestimate actual prevalence.

In diathesis models of PTSD and MDD, trauma serves a necessary but insufficient role in etiology. Inherent vulnerabilities interact with risk factors in the development of stress-related mental health problems. Personality and dispositional factors serve as one source of vulnerability. For instance, behaviorally inhibited (BI) temperament, characterized by extreme withdrawal in the face of social and non-social challenges, potentiates aversive learning in civilian and veteran samples and is associated with expression of PTSD in veterans (Myers et al., 2012a,b). Further, distressed (Type D) personality, which is typified by high levels of negative affect (NA) and social inhibition (SI), is a vulnerability factor of poor physical and mental health outcomes, including PTSD and MDD.

Beyond the development and expression of mental health difficulties, there is growing concern that stress may degrade neurocognitive performance and adversely affect operational performance. Neurocognitive deficits have been reported after repetitive military environmental exposures (e.g., blasts; Luethcke et al., 2011), stress-related mental disorders (Marx et al., 2009; Luethcke et al., 2011; Haran et al., 2013), or mild traumatic brain injury (mTBI; Spira et al., 2014; Dretsch et al., 2015). The Defense Automated Neurobehavioral Assessment (DANA) battery, developed as a clinical decision support tool, consists of a battery of neurocognitive tests as well as psychological assessments for field screening of PTSD and MDD (Lathan et al., 2013; Spira et al., 2014). Using the DANA, multiple experiences of lifetime TBI were found to affect simple reaction time upon second testing, procedural reaction time and code substitution, but only in models unadjusted for PTSD and MDD (Spira et al., 2014). Although suggestive, the impact of PTSD and MDD on neurocognitive functioning was not separately evaluated.

The CG population in Boat Stations has a blend of military service and extensive non-military (but on-call) time, which offers the opportunity to examine the impact of different sources of trauma on PTSD. Therefore, the objectives of the current study were to: Objective (A) describe the demographic characteristics of the CG sample and responses to mental health and personality scales, Objective (B) determine prevalence of PTSD and MDD in CG personnel serving Boat Stations, Objective (C) examine vulnerability and risk factors for PTSD and MDD, and Objective (D) compare the neurocognitive performance of groups as a function of PTSD and MDD.

We expected rates of PTSD to be equivalent with respect to military or non-military trauma. Further, we expected personality scales to provide unique or distinguishable contributions to stress-related mental health difficulties, with BI temperament predicting PTSD and Type D personality, with its emphasis on negative affectivity, predicting general distress. We further expected MDD and its psychomotor slowing to be reflected in deficits in reaction time tasks and attentional processing (Thomas et al., 1999; Naismith et al., 2003), whereas PTSD would be associated with memory difficulties (Honzel et al., 2014).

A cross-sectional study was conducted in active duty CG personnel serving at Boat Stations. To gain insight into the source of traumatic stress symptoms, personnel were asked to complete the two version of the PTSD Checklist (PCL), with prompts related to military (PCL-M) and non-military (PCL-NM) traumatic experiences. Symptoms of MDD were assessed with the Personal Health Questionnaire (PHQ-8). In addition, personnel were administered scales assessing BI and Type D personality. Neurocognitive performance was assessed with the DANA.

Coast Guard personnel serving at Boat Stations experience shift duty with a mixture of operational and off-duty civilian stressors. Overall caseness for PTSD in this sample of CG personnel was similar to caseness using PCL-S (not specific to source) or PCL-C in large cross-sectional studies of Army (Hoge et al., 2008, 2014; Vasterling et al., 2010; Sundin et al., 2014), Marine (Hoge et al., 2014), and Navy (Harbertson et al., 2016) active duty personnel. Rates of PTSD were also consistent with cross-sectional studies of law enforcement (Maia et al., 2008; Bowler et al., 2010), civilian first responders (Skogstad et al., 2015) and firefighters in cross-sectional studies using similar instruments.

To capture the source of stress-related mental health difficulties, CG personnel were administered both the PCL-M and PCL-NM. Comparisons between the two scales suggest that CG personnel adhered to the difference in prompts and responded with different traumas in mind. The delivery of two PCLs is rarely reported, however, Hoge and colleagues recently reported on a comparison between the PCL-S which is based on DSM-IV criteria and the PCL-5, which is based on DSM-5 criteria, in a military sample. A high degree of agreement (κ = 0.67) was obtained with the PCL-S with PCL-5 using the same stressor prompt (Hoge et al., 2014). This contrasts to the modest agreement (κ = 0.37) obtained herein with two different prompts. Caseness of PTSD from non-military traumas was roughly double caseness from military trauma when accounting for personnel that met criteria on both versions. The bias toward non-military traumas may be attributable to: (a) less exposure to military traumas compared to non-military traumas, (b) greater proportion of time CG personnel spend engaged in off-base activities and non-military life events, or (c) an indirect effect of military traumas which depletes resources necessary to cope with non-military traumas. As to the former, there is no validated instrument to assess traumas encountered by CG personnel in performance of duty similar to the CES. Future work is necessary to more fully detail rates and sources of trauma in relation to PTSD symptoms. An open question arising from the cross sectional rates is whether enduring PTSD is more related to military, non-military or the combined burden. This question may be extended to hazardous civilian occupations.

Rates of MDD using symptom criteria were consistent with large surveys of active duty military (Wells et al., 2013), but half of what has been reported to post-deployed forces (Luxton et al., 2010). Further, the degree of co-occurrence of MDD and PTSD was similar to that observed in large survey studies (Wilk et al., 2012).

With the relatively large sample of females, gender as an influence on PTSD and MDD was evaluated. Rates of PTSD were equivalent in male and female CG personnel regardless of screening criteria or type of trauma experienced. Although, females are regarded as having higher risk of PTSD within the general population, studies of active duty military are mixed (Crum-Cianflone and Jacobson, 2014), or do not find differences (Jacobson et al., 2015). A lack of difference has also been reported in studies of police officers (Pole et al., 2001), whose duty demands are similar between males and females. Being female was a predictive factor for MDD when more liberal aggregate scoring criteria were imposed and attendant mild depression was included in models. The difference may be attributed to poor sleep quality which contributes to MDD symptoms, especially mild depression, and was more prevalent in females.

An important objective of the current study was to assess the ability of two personality factors previously identified as vulnerabilities—BI and Type D—to predict stress-related mental health problems. Consistent with research in veterans (Myers et al., 2012a,b), BI temperament was strongly associated with PTSD. BI temperament significantly classified subclinical and full PTSD emanating from either military or non-military traumas. However, BI temperament was not a significant predictor of MDD. In contrast, Type D personality strongly predicted MDD, but was not predictive of PTSD. Given the emphasis of Type D on negative affect it is not surprising that Type D is predictive of depressive mood. Although temperamental dispositions are considered stable personality characteristics, the cross-sectional nature of the study precludes stronger statements regarding either personality factor as a vulnerability to development PTSD or MDD.

There has been growing concern that mTBI may have long-term adverse mental and cognitive consequences. Although roughly half the sample screened positive for lifetime mTBI, lifetime mTBI or the number of mTBI events encountered were not predictors of PTSD, MDD or any deficits in neurocognitive performance.

Neurocognitive performance was, however, distinguishable based on caseness for mental health outcomes. Impaired neurocognitive performance was concentrated among those expressing caseness for both PTSD and MDD. The poor performance was mostly driven by deficits in simple reaction time and response inhibition. The DANA neuropsychological tests are subset of those comprising the Automated Neurobehavioral Assessment Metric (ANAM), but on a hand held form factor. Previous studies with active duty military assessing ANAM performance found deficits in simple reaction time (Vasterling et al., 2012; Wilson et al., 2016), code substitution learning and recall (Vasterling et al., 2012), code substitution learning, and match to sample (Neipert et al., 2014) in those meeting criteria for PTSD. Depressive symptoms were not evaluated in these studies. In these studies, the response inhibition module (“Go/No Go”) was not administered. Response inhibition is deficient in civilians (Falconer et al., 2013) and veterans with PTSD (Swick et al., 2012, 2013). Sensitivity to neurocognitive deficits in PTSD and MDD may depend on a number of factors, including form and venue of testing, and population size. To what degree the reduced throughput on these cognitive tasks adversely affects operational effectiveness of active military personnel, particularly decision making under stress, is unknown.

The main strength of the study is the examination of otherwise healthy high-functioning active duty CG personnel, an underrepresented military population. Not one participant was on reduced or modified duty. Another strength was the number of female volunteers, which allowed for assessments of the influence of gender in many of the outcomes. However, there are several weaknesses that need to be considered. For one, the sample was neither representative of all CG Boats Stations or of the general CG force. For another, the cross-sectional nature of assessment precludes sensitivity to fluctuations in emotional and cognitive performance. Although personality tendencies are assumed to be stable individual characteristics, the nature of the study does not preclude the possibility that experiences such as head injury or those leading to stress-related mental health symptoms can concomitantly change personality characteristics. Further, as a convenience sample of CG Boat Stations, many of the cofactors considered (e.g., sex, deployment, CES) were unbalanced and resulted in small numbers when trying to identify interactions. For example, females in the sample were generally younger and concentrated in lower ranks than males. Boat Stations are heterogeneous in military roles (boat crews, engineers, cooks, administrative, and support staff) and shifts. The inclusiveness of the study to all willing participants comes at the expense of focus on particular operational roles, which would require a much larger study to capture. Additionally, testing was conducted on an individualized basis, therefore time of testing—within a range of 700 to 1,700—was not controlled. Volunteers were allowed to schedule testing when testing would not interfere with duty. Therefore, control was also not exerted in terms of type of experiences immediately preceding testing (e.g., operations, exercise, meals, or training).

To date, this is the largest empirical study of stress and stress-related mental symptoms in CG personnel. The rates of PTSD related to military and non-military experiences are provocative, generating questions for future studies. Accounting for personality could provide opportunities for early intervention, promoting resiliency or targeted tracking to reduce expression of PTSD and/or MDD in the aftermath of trauma.

RS was responsible for study design, implementation, consenting participants, testing, study oversight, data analysis, and dissemination. JH was responsible for data analysis, write up, and dissemination. MD was responsible for study design, implementation and oversight. JT, CaEM, ChEM, RL, and WW were responsible for study design, development of tools and instruments for the conduct of the study, data analysis, write up and dissemination. PA and NK were responsible for consenting participants, testing and data entry. All authors contributed to and have approved the final manuscript.