Mental health and resilience in the Irish defense forces during the COVID-19 global pandemic

ABSTRACT

The Irish Defense Forces (DF) responded to the COVID-19 pandemic and national public health crisis by deploying personnel to aid domestic civil authorities in medical and care settings, contact tracing, logistics, and operations. Current research on COVID-19 reveals increased psychological distress among frontline workers and the general public. Resilience has previously been associated with lower levels of psychological distress. This study sets out to test these associations, and to examine mental health differences between DF personnel deployed in Ireland on pandemic-related duties (DIPD) and non-DIPD. Participants were 231 DF members who completed the: Connor-Davidson Resilience Scale-10, Patient Health Questionnaire-9, Generalized Anxiety Disorder-7, Perceived Coronavirus Threat Questionnaire, Brief Trauma Questionnaire, Post-traumatic Stress Disorder Checklist-5, and Alcohol Use Disorder Identification Test. Independent t-tests revealed no differences between DIPD and non-DIPD on measures of psychological distress or on self-rated mental health prior to COVID-19 (PC19) and during COVID-19 (DC19). Results of multiple hierarchical regression analyses revealed that depression predicted lower levels of resilience, while multiple traumatic events predicted higher levels of resilience. The total adjusted variance explained by the model was 25%. Clinical and policy implications for improving access to psychological support within the DF and military populations are discussed.

KEYWORDS:

• Mental health
• military
• resilience
• COVID-19
• pandemic

What is the public significance of this article?—This study suggests that DF members deployed for pandemic-related duties did not experience more mental health difficulties than those DF members who were not deployed. Also, positive mood and exposure to multiple traumatic events are associated with higher levels of resilience. These findings may be helpful in designing psychological interventions to promote resilience and thereby reduce psychological distress and mental health difficulties.

The coronavirus (COVID-19) outbreak was declared a global pandemic on March 11, 2020, by the World Health Organization (WHO). Many countries initiated stringent public health directives including “stay at home” orders to reduce the reproduction rate of the disease and associated mortality. The Irish Defense Forces (DF) are tasked with defending the state against armed aggression, participating in United Nations peacekeeping missions, and aiding the civil power. The DF responded to the COVID-19 national emergency by deploying personnel from all three DF components; Army, Naval Service, and Air Corps to assist the civil authorities in dealing with the COVID-19 pandemic. The DF provided support that included, but is not limited to, the provision of military healthcare professionals to civilian medical and care facilities, contact tracing, logistics and transport, engineering support, and military medical assets. There is a growing body of evidence that the COVID-19 pandemic and the ensuing societal restrictions have had a significant impact on the psychological wellbeing of frontline medical staff (Kang et al., 2020; Rossi et al., 2020), the general public (Li et al., 2020; Xiong et al., 2020), and military health care workers (Pan et al., 2020). Research is also emerging on pandemic-specific stress, e.g., increased social isolation (Gonçalves et al., 2020), and the advent of new measures for pandemic-specific worries, such as the Perceived Coronavirus Threat Questionnaire (PCT), which was developed using two sample populations from the general public (Conway et al., 2020). Deployment in Ireland for pandemic-related duties (DIPD) is unique in being both a domestic deployment and because it poses a risk to health and life, not just to military personnel, but also to their families and loved ones. This study is interested in how this “dual stressor” might impact the mental health and psychological wellbeing of the DF almost 1 year after the COVID-19 outbreak was declared a global pandemic.

Considerable research resources have been devoted to exploring the mental health of military personnel after deployment (Bonde et al., 2016; Hoge et al., 2004), much of it emanating from the United States (Thomas et al., 2010) and the United Kingdom (Fear et al., 2010) in the wake of combat operations in Iraq and Afghanistan. Mental health outcomes associated with military deployment include stress and burnout (Adler et al., 2017), post-traumatic stress disorder (PTSD) (Forbes et al., 2016), depression (Krauss et al., 2019) and alcohol misuse (Fear et al., 2010; Jacobson et al., 2008). While studies have found an increase in mental health disorders including PTSD (Hoge et al., 2004) among soldiers after combat deployments, some studies of the UK military have found a low prevalence of “probable PTSD” (Fear et al., 2010; Jones et al., 2012). Factors contributing to this discrepancy may include deployment duration, differences in health care and social support (Rona et al., 2016) or differences in national attitudes toward non-combat deployments such as Peace Keeping Operations (Greenberg et al., 2008). Research on suicide among Irish active-duty DF personnel between 1970 and 2002 estimated the yearly suicide rate as 15.3 per 100,000 (Mahon et al., 2005), a rate higher than found in the general population (11.6 per 100,000) in 2005 (National Suicide Research Foundation, n.d.). In comparison, rates of suicide among Swedish peacekeeping personnel were reported to be lower than the general Swedish population (Michel et al., 2007). Military mental health professionals are also at-risk of psychological ill health after deployment (McCauley et al., 2012).

Risk factors associated with psychological distress in a study of the UK military included younger age, female gender, perceived threat to life and poorly perceived leadership (Mulligan et al., 2010). Additionally, older age, being male, as well as exposure to trauma, was associated with odds of experiencing posttraumatic stress (PTS) among New Zealand military personnel (Richardson et al., 2020). Furthermore, exposure to potentially traumatic events among Australian soldiers on peacekeeping missions demonstrated increased and significant rates of psychopathology compared to a civilian sample (Forbes et al., 2016). Ongoing stigma around mental health and underreporting of psychological distress (Sturgeon-Clegg & McCauley, 2019) remains a barrier to accessing mental health services. Fear of being treated differently, or the potential impact on military career are often cited as barriers to access, even across national samples (Gould et al., 2010).

Protective factors such as morale, perceived good leadership, and perceived unit cohesion were all associated with lower levels of common mental health disorders and PTSD (Jones et al., 2012). Resilience is a psychological trait that refers to the individual’s ability to positively adapt in the face of adversity and traumatic events (TE). As noted by Davydov et al. (2010) there is no unifying definition of the concept of resilience, therefore, this study uses the term resilience to include concepts of hardiness (Kobasa, 1979) and psychological flexibility (Kashdan & Rottenberg, 2010). Gulf War soldiers with low scores on hardiness were found to be at greater risk of PTSD (Bartone, 1999), and using more avoidance coping strategies, which in turn was related to increased depression (Bartone & Homish, 2020). High hardiness scores were linked to lower levels of depression, PTS, and aggression (Krauss et al., 2019), and higher scores on hardiness combined with secure attachment style predicted positive mood (Escolas et al., 2014). In their meta-analysis Hu et al. (2015) found resilience was associated with adulthood, and male gender. Previous research assessing the heritability of resilience found gender-based differences with environmental mastery accounting for some of the variance in men but not women (Boardman et al., 2008).

Aim of the present study

This study is interested in the mental health and psychological wellbeing of the DF, and how DF personnel have been impacted psychologically almost 1 year after the COVID-19 outbreak was declared a global pandemic. Mental health difficulties, including symptoms of depression, anxiety, posttraumatic stress, and alcohol misuse, may vary depending on individual demographics, such as gender, age rank, TEs, and resilience.

Hypotheses

This study anticipates that 1) DF personnel who are DIPD will report higher levels of psychological distress than their non-DIPD colleagues. 2) Higher levels of resilience will be negatively associated with mental health difficulties including depression, posttraumatic stress (PTS), generalized anxiety and alcohol misuse.

Method

Participants

Participants consisted of 196 male (85%) and 35 female (15%) members of the DF. Females were overrepresented in this study as they account for just under 7% of the total strength of 8,534 DF members in 2020 (Houses of the Oireachtas, 2020). The mean age was between 35 and 44 years (range = 18–over 50); see Table 1 for participant demographics. Seventy-two per cent reported being married or cohabiting. Participants were current serving members of the DF at the time of the study (January–February 2021) with most participants having served more than 15 years in the DF (71%, n = 164). Forty-seven per cent reported being DIPD, and of those: 23% were frontline medical, 54% were involved in operational and other duties, 17% in logistics, and 6% in contact tracing. A small proportion (8%) of DIPD also reported being deployed overseas at the time of the study. Due to the relatively small sample sizes across the three groups, DIPD, non-DIPD, and deployed overseas, it was not possible to analyze the groups separately.

Table 1. Mean and standard deviations of demographic variables for total sample and gender.

Demographic variables	N = 231%	M	SD	Male n = 196%	Female n = 35%
Gender Male Female	84.8 15.2	1.15	0.36
Age (range) 18–24 25–34 35–44 45–54 Over 50	2.2 24.2 48.5 20.8 4.3	3.01	0.84	2.6 23.5 47.4 21.9 4.6	0 28.6 54.3 14.3 2.9
Marital Single Married/Cohabiting Separated/Divorced Widowed Other	24.7 72.3 1.3 0 1.7	1.82	0.62	20.9 75.5 1.5 0 2.0	45.7 54.3 0 0 0
Children None 1–2 2–4 4 or more	34.6 40.3 21.6 3.5	0.94	0.84	31.6 43.4 21.9 3.1	51.4 22.9 20 5.7
Service Army Naval Service Air Corp	82.3 10.8 6.9	1.25	0.57	82.7 9.7 7.7	80 17.1 2.9
Rank Commissioned officer Noncommissioned officer Private Recruit	54.5 26.8 17.7 0.9	1.65	0.80	53.1 26.5 19.9 0.5	62.9 28.6 5.7 2.9
Military service (years) < 5 5–10 10–15 15–20 20 or more	6.1 13 10 26.4 44.6	3.90	1.27	6.1 12.8 8.7 25 47.4	5.7 14.3 17.1 34.3 28.6
Deployments overseas None 1 2 3 4 or more	8.7 19.5 17.7 14.7 39.4	2.57	1.40	6.6 19.9 17.3 11.7 44.4	20 17.1 20 31.4 11.4

Participants were recruited as a convenience sample from the Irish DF (N = 231) during a 4-week period between January and February 2021. DF personnel were contacted about the study via DF internal communication messaging, e.g., IKON, chain of command group messaging, and representative associations. All DF members were invited to take part in the study, including all ranks and those personnel deployed overseas to ensure a broad representation. As there is no singular method of communication within the DF, and much of the internal communication relies on group messaging, it is not possible to calculate how many DF members directly saw the study link and to calculate the response rate. However, a total of 304 possible participants opened the study link and of those 76% completed the study. Participants were directed to the study through a hyperlink or QR code where they were given detailed information about the nature of the study, anonymity, data protection, and consent. Reminders were sent approximately 3 weeks after the study was initiated. Participants self-selected and no incentives were offered to participate in the research study.

Ethics

This research study was granted approval by the School of Psychology Human Research Ethics Committee, Trinity College Dublin, and the Irish Permanent Defense Forces.

Variables

Background information was gathered on gender, age, relationship status, family, DF component, rank, years of military service, overseas deployment, DIPD, previous mental health difficulties, and concerns related to COVID-19 (Table 2). The following self-report standardized measures were used to examine psychological wellbeing and mental health.

Table 2. Impact of COVID-19 pandemic and restrictions, and previous mental health experiences.

	Mental Health Pre-COVID-19	Mental Health during COVID-19
	% (N = 231)	% (N = 231)
Excellent	29.9	11.7
Very good	39.8	31.2
Good	19.9	28.1
Fair	6.5	24.7
Poor	3.9	3.9
Not sure	–	0.4
Total	100	100
Items related to COVID-19 and mental health	Yes %	No %
Deployed overseas March 2020	19.5	80.5
Deployed overseas Jan – Feb 2021	15.6	84.4
Deployed Ireland COVID-19	47.2	52.8
Past mental health difficulties?	42.4	57.6
Attend GP or Mental Health professionals for mental health difficulties?	22.9	76
Ever attended counseling?	34.2	65.8
Did service during COVID 19 have a negative impact on psychological wellbeing?	41.6	58.4
Worried loved ones would contract COVID 19?	84.8	15.2
Watched a lot of news coverage on COVID 19?	76.6	23.4
Did Government shutdowns have a positive effect on work/life balance?	42.4	57.6
Did shutdowns have a negative effect on personal life?	68	32
Experienced a bereavement due to COVID 19 (family member, friend, neighbor or relative of friend)?	18.6	81.4
Were roles and duties during COVID 19 rewarding?	63.6	36.4
Were roles and duties during COVID 19 frustrating?	57.1	42.9

Coronavirus anxiety

The Perceived Coronavirus Threat Questionnaire (PCTQ; Conway et al., 2020) is a new measure developed to measure worry, fear, and threat associated with the COVID-19 pandemic. It is composed of six items concerning how threatened or worried people are about COVID-19 using a 7-point Likert scale from “Not true of me at all” to “Very true of me” and demonstrated Cronbach's Alpha = .88 (Conway et al., 2020). In this study, Cronbach’s Alpha = .832.

Resilience

The Connor-Davidson Resilience Scale (CD-RISC 10; Connor & Davidson, 2003) is a brief version of the original 25-item CD-RISC. It is a self-reported unidimensional measure of resilience, using a 5-point Likert Scale from “Not true at all” to “True nearly all the time.” Higher scores indicate a higher degree of resilience. The 10-item CD-RISC demonstrated a Cronbach’s Alpha value of .85 (Campbell-Sills & Stein, 2007). Reliability analysis for this study found Cronbach’s Alpha = .888.

Posttraumatic stress

The Brief Trauma Questionnaire (BTQ) is a self-reported questionnaire derived from the Brief Trauma Interview (Schnurr et al., 1999). It provides a complete assessment of Criterion A for PTSD in the Diagnostic and Statistical Manual of Mental Disorders 5^th Edition (DSM-5). Respondents are asked about their experience of Criterion A events with dichotomous responses of “Yes/No.”

The Posttraumatic Stress Disorder Checklist for DSM-5 (PCL-5; Weathers et al., 2013) was used to measure PTS. It is a 20-item self-report measure that corresponds to the DSM-5 symptoms of PTSD using a 5-point Likert Scale from “Not at all” to “Extremely.” Scores above 31 suggest “probable” PTSD diagnosis, which is recommended for screening due to higher sensitivity at this cut point (Wortmann et al., 2016). The PCL-5 previously demonstrated excellent internal consistency (α = .96) (Bovin et al., 2016). In this study, Cronbach’s Alpha = .959.

Alcohol use

Alcohol Use Disorders Identification Test (AUDIT; Babor et al., 2001) consists of 10 questions concerning recent alcohol use, alcohol dependence symptoms, and alcohol-related problems. The first eight questions use a five-item continuous scale (scored 0–4), while the last two questions use a three-item scale (scored 0, 2, or 4). Scores above 8 indicate hazardous drinking, while scores above 20 are indicative of a “possible dependence” on alcohol. The AUDIT has previously demonstrated a good internal consistency with Cronbach’s Alpha = .80 (Fleming et al., 2009). Reliability analysis for this study found Cronbach’s Alpha = .842

Depression

The Patient Health Questionnaire (PHQ-9; Kroenke et al., 2001) is a multipurpose instrument for screening and measuring the severity of depression. The PHQ-9 consists of nine items that correspond to the DSM-IV criteria for depression. Items are rated on a 5-point Likert Scale from “Not at all” to “Nearly every day.” Scores of 5, 10, 15, and 20 constitute cut-points for mild, moderate, moderately severe, and severe depression, respectively. Internal consistency of the PHQ-9 found Cronbach’s Alpha was between 0.86 and 0.89 (Kroenke et al., 2001). In this study, reliability analysis revealed Cronbach’s Alpha = .906.

Generalized anxiety

Generalized Anxiety Disorder (GAD-7; Spitzer et al., 2006) is a 7-item scale used to measure or assess the severity of generalized anxiety disorder. Items are rated on a 5-point Likert Scale from “Not at all” to “Nearly every day.” Scores of 5, 10, and 15 represent cut-points for mild, moderate, and severe anxiety, respectively. The internal consistency of the GAD-7 was reported as Cronbach's Alpha = .92 (Spitzer et al., 2006). Cronbach’s Alpha for this study = .947.

Data analysis strategy

The data were analyzed using IBM SPSS Statistics (Version 26) predictive analytics software. Data were found to be missing for the AUDIT questionnaire (n = 16, 6.9%). On close inspection of these cases, respondents that answered “Never” to Item 1 “How often do you have a drink containing alcohol?” subsequently skipped Item 2 “How many drinks containing alcohol do you have on a typical day when you are drinking?” Item 2 contains no option for “None.” Where the mean score for the other nine items in the AUDIT was zero, a score of zero was awarded for item 2. The missing values for the AUDIT were thus reduced to 3.5% and were excluded pairwise for subsequent analyses. Little’s MCAR was calculated for the other measures and confirmed the data were missing completely at random (p = .178).

Univariate analyses were performed and means and standard deviations for all standardized measures can be found in Table 3. Histograms, skewness, and kurtosis were inspected for normality of the distribution for each of the predictor variables. Some variables displayed slight positive skewness; however, this was expected in a military population where resilience and psychological flexibility are actively recruited, and none exceeded acceptable values of ± 2 (George & Mallery, 2010). The sample size was considered large enough to perform parametric analyses. Independent and paired samples t-tests, Levene’s Test for Equality of Variance and eta squared methods were used to determine differences between groups. Pearson product-moment correlation coefficients were used to examine the relationships between variables (see Table 4). Multivariate regression analyses were performed to assess the predictive ability of the scale scores on the variables of interest.

Table 3. Descriptive statistics for predictor variables and independent t-tests comparing predictor variables for DIPD and non-DIPD.

			t-test for Equality of Means
Variables	N = 231^a		DIPD N = 109	Non-DIPD N = 121	t	Sig. (2 – tailed)	95% Confidence Interval
	Mean	Std. Deviation	Mean (SD)			p =	Lower	Upper
PCTQ	26.47	7.66	26.64 (7.70)	26.31 (7.66)	0.33	.744	−1.66	2.32
CD-RISC 10	29.68	6.30	28.87 (6.96)	30.41 (5.57)	−1.86	.064	−3.16	0.09
BTQ	2.72	1.92	2.80 (1.90)	2.65 (1.94)	0.55	.581	−0.37	0.66
PCL-5	16.74	16.53	19.20 (19.34)	14.53 (13.20)	2.12	.036	0.32	9.03
AUDIT	6.74	5.70	6.63 (6.18)	6.85 (5.19)	−0.28	.779	−1.72	1.29
PHQ-9	5.67	5.61	6.37 (6.49)	5.06 (4.65)	1.74	.083	−0.17	2.81
GAD-7	4.53	5.17	5.26 (5.99)	3.88 (4.22)	2.00	.047	0.02	2.75
Mental Health PC19	2.15	1.05	2.16 (1.13)	2.14 (.96)	0.12	.904	−.26	.29
Mental Health DC19	2.79	1.09	2.88 (1.11)	2.71 (1.06)	1.17	.243	−.12	.45

^aDue to missing values and varied sample sizes, means were compared pairwise.

*p < .007 (Bonferroni adjusted)

BTQ = Brief Trauma Questionnaire, PCL-5 = Posttraumatic Stress Disorder Checklist (DSM-5), PHQ-9 = Patient Health Questionnaire 9, GAD-7 = Generalized Anxiety Disorder 7, CD-RISC 10 = Connor-Davidson Resilience Scale 10, AUDIT = Alcohol Use Disorder Identification Test, PCTQ = Perceived Coronavirus Threat Questionnaire, PC19 = Pre COVID-19, DC19 = During COVID-19.

Table 4. Results of correlation analyses between outcome variables.

Variables	1	2	3	4	5	6
1. Resilience	–
2. PTSD Criterion A (TE)	0.06	–
3. Posttraumatic stress	−.41**	.32**	–
4. Depression	−.46**	.28**	.83**	–
5. Generalized anxiety	−.44**	.23**	.83**	.88**	–
6. Coronavirus anxiety	−.15*	−0.03	.23**	.21**	.28**	–
7. Alcohol use	−.17*	.17*	.34**	.32**	.32**	−0.02

^aDue to missing values, correlation coefficients were calculated pairwise; variables 1 & 6, n = 231.

variable 3, n = 230; variables 4 & 5, n = 229; variable 12, n = 223; variable 2, n = 218.

Variable measures: 1.CD-RISC 10, 2. BTQ, 3, PCL, 4. PHQ-9, 5. GAD-7, 6. PCTQ, 7. AUDIT.

*p < .05. **p < .01.

Results

Psychological characteristics

As most participants were male (85%) and of Officer rank (81%), t-tests were used to explore mean differences in all variables based on gender and rank. To correct for multiple comparisons, a Bonferroni adjusted alpha p < .007 was used. None of the results reached statistical significance, indicating that neither gender nor rank was a confounding variable. Participants were asked about their previous mental health history and the impact the COVID-19 pandemic has had on them. Forty-two percent reported having previous mental health difficulties. Almost one-fifth reported moderate-to-severe depression symptoms, while 11.4% reported moderate-to-severe symptoms of anxiety. Scores from the PCTQ indicated that 31% of the sample reported “strong” or “very strong perceived threat” by the Coronavirus. Most participants reported directly experiencing or being exposed to several TEs on the BTQ (M = 2.72, SD = 1.918, 95% CI = 2.46, 2.98) with only 12.6% (N = 218, n = 29) reporting no TEs. A crosstabulation analysis was performed to determine the percentage of those with probable PTSD against TEs reported on the BTQ (see Table 5). Eleven percent reported suicidal ideation, including suicide attempts, while 6% reported PTSD. Almost 32% exceeded the cutoff (≥ 8) for harmful drinking, with 4% of those at risk for possible dependence. Most variables were correlated with each other in the expected direction, with correlation coefficients ranging from weak to strong (see Table 3).

Table 5. Crosstabs analysis of traumatic events (BTQ), probable diagnosis of PTSD^a, and gender.

		PTSD (n = 41**) %
Brief Trauma Questionnaire (BTQ)^b	Yes N = 231* %		Male (n = 35) %	Female (n = 6) %
1. Served in war zone or exposed to war-related casualties	48		54	7†
2. Serious accident (car or work)	25		29	5
3. Been in natural or technological disaster	12		22	0
4. Had a life threatening illness	7		5	0
5. Physically punished before age 18 by parent, carer or teacher	20		17	2
6. Attacked or beaten or mugged by family, friends or stranger (excluding reported in 5.)	49		66	10
7. Pressured into unwanted sexual contact	11		12	7
8. Other situation where you were seriously injured or feared being seriously injured or killed	36		51	7
9. Close family member died violently (car crash, mugging, or attack)	17		29	7
10. Ever witnessed someone else seriously injured or killed or feared they would be	44		39	10

^aScores > 31 are indicative of probable diagnosis of PTSD on Posttraumatic Stress Disorder Checklist DSM-5 (PCL-5).

^bBTQ items are summarized for ease of presentation.

† Percentages do not add up to 100 as not all participants endorsed each item.

* Item 3, n = 227; Item 4 & Item 9, n = 230; Item 6, n = 229; Item 7 & Item 10, n = 228.

** Items 3, 4, & 10, n = 40.

Hypothesis testing

Hypothesis 1

Independent sample t-tests were used to analyze differences in psychological distress (coronavirus anxiety, resilience, TEs, PTS, alcohol use, depression, generalized anxiety) between DIPD and non-DIPD. To reduce the probability of making a Type I error, a Bonferroni adjusted alpha of p < .007 was used. None of the results achieved statistical significance (see Table 3). These results indicate that DF personnel who were DIPD did not differ in scores for coronavirus anxiety, resilience, TEs, PTS, alcohol use, depression, and generalized anxiety compared to those who were non-DIPD. Additionally, no differences were found between DIPD and non-DIPD on self-rated mental health prior to COVID-19 (PC19) and during COVID-19 (DC19) (see Table 3). A paired sample t-test was used to analyze differences across the entire sample on mental health PC19 (M = 2.15, SD = 1.05) and mental health DC19 (M = 2.79, SD = 1.09) which revealed a statistically significant difference in mental health ratings, t(230) = −9.79, p < .001. This indicates that self-rated mental health had worsened over the course of the pandemic. The mean difference in scores was −.65, 95% CI [−.78, −5.15] with a large effect size, Cohen’s d = 1. A moderately negative correlation was found between mental health PC19 and resilience, r = −.4, p < .001, and mental health DC19 and resilience, r = −.47, p < .001. Also, a negative weak correlation was found between coronavirus anxiety and resilience, r = −.15, p < .05.

Hypothesis 2

To assess whether psychological distress predicted levels of resilience, a multiple hierarchical regression was performed with resilience as the dependent variable, while depression, posttraumatic stress, coronavirus anxiety, and alcohol use were entered into the model as predictor variables. Due to a strong correlation between the PHQ-9 and the GAD-7 (r = .88, p < .001) and violations of multicollinearity, the Variance Inflation Factor (VIF) = 5.23 and 5.25, respectively, the GAD-7 was excluded from this regression analysis. Gender, age, rank, and TEs (BTQ) were entered into Step 1, and these accounted for 6.7% of the variance in the model. The predictor variables were entered into Step 2 and the total adjusted variance explained by the model was 25%, F(8, 202) = 9.82, p < .001. Depression, posttraumatic stress, coronavirus threat, and alcohol use explained an additional 20% of the variance after controlling for covariates, R² change = .20, F change (4, 202) = 19.5, p < .001. Depression made the strongest unique contribution to the final model, β = – .34, p = .003, 95% CI [–0.62, – 0.13], as depression severity increased, resilience scores decreased. Exposure to TEs also made a statistically significant contribution, β = .17, p = .011, 95% CI [0.13, 0.99], whereby increased exposure to TEs predicted an increase in resilience scores. Exposure to TEs made a unique contribution of 2.3% to the variance in resilience, while depression uniquely contributed 3.3% of the variance. Cohen’s f² = .34 indicated a medium effect size for the model. There were no differences found between DIPD and non-DIPD on resilience scores (see Table 3).

Discussion

While the full extent of the psychological impact of the COVID-19 pandemic may not be fully understood for many years, emerging evidence points to increased psychological distress across the general population (Li et al., 2020; Xiong et al., 2020) as well as frontline medical workers (Kang et al., 2020). This study sets out to explore the mental health of the DF and the psychological impact of the pandemic. Additionally, extending on previous research establishing an association between resilience and mental health (Bartone & Homish, 2020; Krauss et al., 2019), we set out to test this association with an Irish military population.

Contrary to our expectations, we did not find a statistically significant difference between service members who were DIPD compared to their colleagues who were non-DIPD on measures of psychological distress (coronavirus anxiety, resilience, TEs, PTS, alcohol use, depression, generalized anxiety). This ran counter to current research that has shown increased rates of vicarious trauma and mental health difficulties in medical, frontline workers and the public, during the pandemic (Kang et al., 2020; Li et al., 2020; Rossi et al., 2020). It is possible that given the high rates of self-reported mental health difficulties (42%) and exposure to traumatic events (87.4%), that these may have masked any additional psychological impact of the pandemic. Over one-fifth of DIPD was frontline medical which may be considered a “high-exposure” role compared to colleagues in operational and logistics roles. A meta-analysis of clinical staff working in high-risk epidemic and pandemic health emergencies, which included COVID-19, Ebola, SARS, etc., found that high-exposure clinical work was associated with only a small additional burden of mental health symptomatology (Bell & Wade, 2021). Furthermore, there was some overlap between the two groups in this study as some DF personnel who had been DIPD were subsequently deployed overseas and this may have been a confounder. In using a Bonferroni adjustment to control for Type I errors, the possibility of Type II errors increases, thus there may in fact be differences between the two groups that were not detected.

Of additional note in this study were the concerns expressed about COVID-19 and the impact it has had on participants’ lives. Over 60% reported greater perceived threat from the Coronavirus. This was at a time when Ireland was experiencing record COVID-19 related deaths and the vaccination program had just commenced rollout (January/February 2021). While medical and frontline staff were the first cohorts being vaccinated at this time, many of the other DIPD and non-DIPD personnel were not eligible for vaccination until months later, which may have contributed to coronavirus anxiety. Although participants may be subject to recall bias, they perceived their mental health to be worse at the time of data collection than pre-COVID-19. While there were no differences between DIPD and non-DIPD on mental health PC19 and DC19, there was a significant difference between PC19 and DC19 ratings across the whole sample, indicating that their mental health had suffered during that timeframe (≈10 months). While 9% of our sample endorsed suicidal thoughts, and 2% endorsed suicide attempts that are comparable to a Veteran sample of Australian peacekeepers (Forbes et al., 2016), this is still a grave concern, particularly in a cohort who may have easier access to lethal means. In this study, alcohol use was weakly correlated with resilience, depression, anxiety, and PTS compared to previous research (Forbes et al., 2016), which cautiously suggests that avoidance coping (Bartone et al., 2017) was not a strong feature for this sample.

Based on empirical findings from other military samples, we expected psychological distress to predict lower levels of resilience (Hu et al., 2015). This was partially supported as only depression made a significant unique contribution to the final model after controlling for gender, age, rank, and TEs. There was no difference in resilience scores between DIPD and non-DIPD groups, which suggests the results were not confounded by deployment status. Somewhat contrary to our expectations, exposure to TEs also made a significant contribution to the overall model, suggesting that those who had been exposed to, or experienced more TEs had higher levels of resilience. This may be a result of posttraumatic growth, which has been found in ex-military personnel (Habib et al., 2018). The “healthy soldier” effect refers to military personnel being healthier and fitter at recruitment than the general population (McLaughlin et al., 2008). This effect may also apply to psychological health and those who are less resilient may select to leave service at the earliest opportunity.

Correspondingly, and of particular concern, is the high level of posttraumatic stress being experienced by DF personnel. Overseas deployments by the DF consist of peacekeeping or humanitarian missions, yet almost one-fifth of the sample scored above the cutoff for a “probable” diagnosis of PTSD, which is markedly higher than among other peacekeeping samples (Greenberg et al., 2008). Participants in this study accessed counseling for their mental health difficulties within the DF at a lower rate than those who accessed counseling externally (8% and 26% respectively). This may be related to a perceived stigma in accessing psychological support in general, as well as within the military (Sturgeon-Clegg & McCauley, 2019), and fears that their mental health status will adversely affect their military career and ambitions (Gould et al., 2010).

There are some limitations to the present study. More than 80% of the participants were commissioned and noncommissioned officers, while they account for just 50% of the overall strength of the DF (Houses of the Oireachtas, 2020). This may be due to the lack of a singular communication system within the DF, which may have disadvantaged privates or other junior enlisted personnel during the recruitment phase of this study. It is possible that officers received individual e-mails alerting them to the study, while private and junior enlisted ranks were reliant on command group messaging and communication bulletins from their representative associations. Therefore, caution should be exercised when seeking to generalize our findings across the DF population. However, rank was not a predictor of resilience in this sample and comparisons between ranks revealed no statistical differences in the variables measured. The response rate, which could not be calculated, was relatively low compared to the overall strength of the DF (8,534). This may partially be explained by the varied forms of communication within the DF, where the research invitation was “lost” within multiple group messages. There may also have been a self-selection bias whereby the participants chose to opt-in to the study due to an interest in mental health topics. Also, research on mental health within the DF is not well-established, and the response rate may reflect a reluctance to report subjective mental health experiences. The cross-sectional design of this study limits our interpretation of the results as it reflects a point in time during unprecedented global anguish. Restrictions on seeing family members and friends and participating in sports and recreational activities may be significant contributors to reduced psychological wellbeing. This effect may diminish as societal restrictions ease. Nonetheless, a notable strength of this study was the gender distribution. Women who make up just under 7% of the DF were overrepresented in this study, which allowed us make comparisons with their male colleagues. While the study did not set out to compare differences between genders, these comparisons allowed us to explore the possibility of gender as a confounding variable. In addition, this is the first study to directly investigate the mental health and psychological wellbeing of the Irish DF, and uniquely, amid a global public health emergency. These outcomes provide a baseline for future research with this population.

Clinical and empirical implications

Over 14% of the participants reported not sharing their worries with anyone, which may make these individuals particularly vulnerable to the cumulative effects of psychological distress as they are less likely to seek help. Furthermore, only a small proportion reported accessing psychological support from within the DF, which may be related to stigma and concerns about their military career. Consequently, there may be utility in reviewing the downgrading requirements for DF personnel accessing support services. This may encourage increased engagement with military mental healthcare systems and allay any implicit concerns about career implications. Additionally, an important future direction should include consideration of the composition and resourcing of the mental health services within the Medical Corps, and the Personnel Support Service (PSS), to ensure the optimization of access and provision, both nationally and during overseas deployments.

Previous research has shown that unit cohesion and supportive leadership are associated with lower levels of psychological distress and PTSD among the military population (Jones et al., 2012). Considering the reported high rates of PTS and exposure to TEs in this sample, these protective factors may be even more crucial for DF personnel during a public health emergency and are certainly critical for DF personnel deployed on peacekeeping and humanitarian missions where exposure to TEs are especially possible. Interventions that promote increased unit cohesion, staff morale, and support from DF military leadership should be explored and implemented.

Future research exploring the relationship between resilience, exposure to TEs, posttraumatic growth, and the healthy soldier effect would aid a better understanding of why in this study, exposure to multiple TEs was associated with greater resilience. Previous research has found little difference between males and females in mental health outcomes after deployment (Vogt et al., 2011), and while female military personnel were found to have lower levels of resilience (Adams et al., 2021), this association was no longer significant in a sample of female and male Veterans once trauma type was controlled for (Portnoy et al., 2018). The finding in this study that males and females demonstrated no significant differences in the mental health and psychological wellbeing variables measured is worth researching further to ascertain if this finding in an Irish population was novel.

Acknowledgments

The authors would like to thank the Irish Defence Forces for their support in conducting this research study, particularly to the DF personnel who kindly participated.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

These data, which are owned by Trinity College Dublin, are not publicly available.

Additional information

Funding

This research was conducted in part fulfillment of the Doctorate in Clinical Psychology at Trinity College Dublin, by the first author. This research was not funded by an external body.