Interactive relationship between alexithymia, psychological distress and posttraumatic stress disorder symptomology across time

ABSTRACT

Alexithymia, psychological distress, and posttraumatic stress disorder (PTSD) are highly related constructs. The ongoing debate about the nature and relationship between these constructs is perpetuated by an overreliance on cross-sectional research. We examined the longitudinal interactive relationship between alexithymia, psychological distress, and PTSD. We hypothesised that there is an interactive relationship between the three constructs. Military personnel (N = 1871) completed the Toronto Alexithymia Scale, the Kessler 10 and a PTSD Checklist (PCL-C) at pre-deployment, post-deployment, and at 3–4 years following the post-deployment assessment. We initially tested whether psychological distress is either a moderator or mediator in the relationship between alexithymia and PTSD across the time points. General psychological distress was a partial mediator of total PTSD severity and hyperarousal symptomology at all three time points. Psychological distress fully mediated re-experiencing and avoidance symptomology at all three time points. Our results suggest that those with alexithymia are at longitudinal risk of developing more severe PTSD symptomology and experiencing hyperarousal irrespective of temporal proximity to traumatic exposure. Further, vulnerability to the emergence of re-experiencing and avoidance symptomology for those with alexithymia is increased when one experiences greater distress. Our results show that alexithymia is a persistent risk factor for PTSD symptomology.

KEYWORDS:

• Alexithymia
• psychological distress
• posttraumatic
• PTSD

Posttraumatic stress disorder (PTSD) is a heterogeneous disorder that may develop after exposure to a traumatic event such as threatened death, injury, or sexual assault (American Psychiatric Association, 2013). The signs and symptoms of PTSD span four symptom clusters: (1) intrusive symptoms or re-experiencing of the traumatic event; (2) avoidance of reminders that are associated with the traumatic event; (3) hyperarousal; and (4) negative cognitions about oneself, the world, and others. Although several gold-standard, first-line treatments have been developed for PTSD, nearly 40% of patients will experience a chronic course of the condition (Santiago et al., 2013). One potential driver of this poor treatment response is that not all patients are able to adaptively engage with distressing emotions, which is necessary to process traumatic experiences (for example, see Putica et al., 2022b).

Alexithymia

Alexithymia is a dimensional, subclinical construct characterised by difficulties in identifying and communicating one’s emotional states. Alexithymia is most commonly assessed using the Toronto Alexithymia Scale (TAS-20), which has three subscales: (1) Difficulty in Identifying Feelings (DIF); (2) Difficulty in Describing Feelings to other people (DDF); and (3) Externally-Oriented Thinking (EOT; Bagby et al., 1994). The prevailing hypothesis in the literature is that alexithymia is associated with emotion dysregulation, which drives the development of psychopathology (Sheppes et al., 2015). The emotion dysregulation hypothesis is supported by findings that show that alexithymia hinders general affective processing capacities (Preece et al., 2022).

Existent literature demonstrates a robust relationship between alexithymia and PTSD (Declercq et al., 2010; Edwards, 2019). Although some postulate alexithymia to be a post-trauma coping mechanism, i.e. PTSD-related emotional numbing, which buffers one against further psychological distress (Brady et al., 2017; Eichhorn et al., 2014), emerging research has indicated that alexithymia is a distinct construct to PTSD-related emotional numbing (posttraumatic defence from further distress; Litz & Gray, 2002). Putica et al. (2021) found that alexithymia is associated with Default Mode Network (DMN) alterations, as opposed to PTSD-related emotion-numbing, which shows alterations in the Salience Network. Rather, alexithymia has been conceptualised as a stable personality trait that correlates positively with neuroticism (Gaggero et al., 2022), depression (Li et al., 2015), and anxiety (Palser et al., 2018). This is supported by studies showing that despite participants demonstrating positive treatment responses, their alexithymia levels tend to remain constant (Luminet et al., 2001; Martínez-Sánchez et al., 2003). This high association between pre-treatment and post-treatment alexithymia scores has led to the conceptualisation of alexithymia as a trait lacking absolute stability, but rather acting as a trait of relative stability like neuroticism and extraversion (Luminet et al., 2001).

The depiction of alexithymia as a trait of relative stability has also been presented in the posttraumatic field. Yehuda et al. (1997) found a positive association between PTSD and alexithymia in Holocaust survivors but no association between traumatic experiences and alexithymia, indicating that alexithymia seems more connected to all posttraumatic symptoms clusters (re-experiencing, avoidance and hyperarousal) than to trauma exposure itself. Hence, the researchers concluded that alexithymia might be a pre-existing condition that increases the likelihood of one developing PTSD (Yehuda et al., 1997). Further, the Default Mode Network has been found to be implicated in personality and self-referential processes (Li et al., 2022), with the connectivity of the Default Mode Network found to occur early in life (Bulgarelli et al., 2020). Similarly, alexithymia has demonstrated etiological trajectories characterised by childhood maltreatment/neglect (Khan & Jaffee, 2022; Sripada et al., 2014). Taken together with evidence showing the relative stability of alexithymia, this research supports the notion of alexithymia as a pre-existing trait with limited within-person changeability. However, the nature of this stability as relative rather than absolute suggests that there may be an interaction modulating the relationship between alexithymia and psychopathology. One such modulator may be one’s level of general psychological distress. Mikolajczak and Luminet (2006) found that, of this variance in follow-up alexithymia scores which indicated relative stability rather than absolute stability, a significant amount was predicted by psychological distress severity.

Psychological distress

Psychological distress is a broad term that refers to a range of unpleasant emotional experiences, including anxiety, depression, and stress. Alexithymia and PTSD have been shown to be closely associated with general psychological distress (Kerswell et al., 2020; McIntosh, 2017), with one’s level of pre-trauma general psychological distress also found to be associated with PTSD symptomology (Sareen, 2014). Given the associations between alexithymia and PTSD, psychological distress and PTSD, and alexithymia and psychological distress, it is important to investigate how the relationships between these constructs may pertain to PTSD presentations. However, this has been difficult to understand due to the cross-sectional nature of established works.

Given the high association between alexithymia and general psychological distress, some have also raised concerns about the discriminant validity of the TAS-20 (and alexithymia as a construct), suggesting that it might actually be a measure or by-product of one’s current levels of general baseline psychological distress or psychopathology symptoms rather than measuring or being a distinct construct (Brady et al., 2017; Eichhorn et al., 2014). Cross-sectional studies exploring the discriminant validity between the TAS-20 and other measures of general psychological distress have yielded equivocal results. Item-level analysis of alexithymia and other measures have found that TAS-20 items have generally loaded onto first-order factors separate from distress and psychopathology (Müller et al., 2003). Results become more mixed, though, when examining overlap among subscales (second-order factors). Preece et al. (2020) conducted a second-order factor analysis of the subscales from the TAS-20 and Depression, Anxiety, and Stress Scales (N = 300 across five samples, three of which included the TAS-20) and found that TAS-20 DIF scores showed mixed loadings across both alexithymia and a general distress factor. Based on these equivocal results, the question of how alexithymia and general psychological distress interact and how this interaction relates to the development of psychopathology remains. One way to test this is to explore the impact of alexithymia and general psychological distress on PTSD symptomology across time.

PTSD symptomology across time

Initially, it was thought that PTSD, in the vast majority of cases, would develop within the first few months after the potentially traumatic event. On the contrary, emerging evidence now indicates that PTSD presentations are heterogeneous, with many individuals experiencing delayed onset of symptomology years after trauma exposure (Bonde et al., 2022; Utzon-Frank et al., 2014). Systematic reviews based upon numerous large longitudinal studies across populations of civilian and military personnel consistently demonstrated that the onset of the disorder, defined as the time at which all diagnostic criteria are met, is delayed beyond six months (and up to five years) post-deployment in 20–30% of cases. This incidence in delayed-onset symptomology is found to be even more frequent in military personnel returning from deployment compared to civilian samples (Bonde et al., 2022; Utzon-Frank et al., 2014). These findings indicate that there are variables outside of trauma exposure that may predict delayed PTSD presentations.

The vast majority of research exploring the association between alexithymia and PTSD has largely been cross-sectional in nature (for example, Eichhorn et al., 2014; La Rosa et al., 2022). However, in one available prospective study by McCaslin et al. (2006), pre-9/11 alexithymia scores significantly predicted 9/11-related PTSD symptom severity, over and above pre-9/11 PTSD symptoms. McCaslin and colleagues suggest that alexithymia is not simply related to the level of trauma exposure but rather to symptomatic response(s) to the event. However, further work examining the association between alexithymia and PTSD symptomology across multiple time points following trauma exposure is needed to test this conclusion. Investigating these relationships over time may yield insights into how these variables interact to predict psychopathology.

Present study

To date, the work exploring the association between alexithymia and PTSD symptomology has utilised cross-sectional designs or only focused on symptomology immediately after trauma exposure. It is surprising, given that alexithymia has been found to be closely related to general psychological distress, that no work has explored whether an interactive relationship exists between alexithymia, general psychological distress, and PTSD across time. Our study is the first to explore whether psychological distress explains the relationship between alexithymia and PTSD over time. Exploring the longitudinal relationship of these constructs may help elucidate how alexithymia increases one’s vulnerability to the development of PTSD symptomology. Additionally, if the relationship between alexithymia and PTSD is explained by psychological distress, this would enhance identification of those at risk of developing PTSD psychopathology.

Given that alexithymia has been found to be linked to PTSD at a disorder and symptom cluster level among cross-sectional studies, the aim of the current study is to elucidate the longitudinal nature of the relationship between alexithymia, general psychological distress, and PTSD (at a disorder and symptom level). Specifically, we aim to examine how alexithymia and general psychological distress interact to predict the development of PTSD symptomology. If psychological distress does moderate or mediate the relationship between alexithymia and PTSD, this will further reinforce that alexithymia and PTSD are distinct constructs whose association is strengthened during periods of increased psychological distress. Firstly, we hypothesised that there would be an interactive relationship between alexithymia, psychological distress, and PTSD (at a disorder severity and symptom cluster level) across time, either via a moderation or mediation effect. Next, we tested which model (moderation or mediation) best explains the relationship between alexithymia and PTSD (at a disorder severity and symptom cluster level) across time.

Method

Participants and procedure

Members of the Australian Defence Force (ADF) who deployed to the Middle East Area of Operations after June 2010 and returned from that deployment by June 2012 were invited to participate in the study (N = 3074). Data was collected immediately prior to deployment (pre-deployment; no more than four months prior to deploying), again directly after deployment (post-deployment; on average 4.2 months after deployment), and at follow-up (3–4 years post-deployment). A total of 1871 ADF personnel (female = 172; M_age = 29.90; SD = 0.10) completed the measures at pre-deployment, post-deployment (N = 1324; female = 172; M_age = 28.90; SD = 0.10), and follow-up (N = 574; female = 98; M_age = 35.00; SD = 0.20). Response rates between measures varied (as outlined in the results). Participants completed a pre-deployment self-report questionnaire following an initial study briefing or by mail via prepaid envelope. Only participants who completed a pre-deployment questionnaire were invited to complete post-deployment and follow-up measures.

The study protocol was approved by the Department of Veterans’ Affairs Human Research Ethics Committee (E014/018) and was recognised under expedited review processes by the Australian Government Department of Defence and the University of Adelaide Human Research Ethics Committee. The study protocol was also submitted to the Australian Institute of Health and Welfare Ethics Committee, which granted approval (EO 2015/1/163). The study was conducted in accordance with the Australian Code for the Responsible Conduct of Research (https://www.nhmrc.gov.au/sites/default/files/documents/attachments/grant%20documents/The-australian-code-for-the-responsible-conduct-of-research-2018.pdf). Informed consent was obtained from all participants; participants were also made aware that they could withdraw from the study at post-deployment data collection time points.

Measures

Toronto alexithymia scale (TAS-20)

The TAS-20 (Bagby et al., 1994) was used to measure the presence and severity of alexithymia at pre-deployment. The TAS-20 is a self-report measure comprised of 20 items rated using a 5-point Likert scale (1 = strongly disagree to 5 = strongly agree). The total alexithymia score is the sum of responses to all 20 items. Scores equal to or greater than 51 indicate the presence of alexithymia. Good test-retest reliability and excellent internal consistency, with alpha coefficients ranging from .74 to .77 for the full scale, have been reported (Bagby et al., 1994). The Cronbach’s alpha for the total TAS-20 score was .80. Although the TAS-20 is a well-validated measure of alexithymia, some have demonstrated that at a subscale level, there are some concerns, such as the discriminant validity of the Difficulties Identifying Feelings (DIF) subscale and the reliability of the Externally-Oriented Thinking subscale (see Preece et al., 2020). For our sample, the DIF factor and psychological distress (K10) showed some shared variance (r = .187, p = .01), however, most of the variance for this factor was explained by the TAS-20 (r = .827, p = .01). Given this acceptable discriminant validity between the DIF and K10 factors, we ran our analyses at a TAS-20 total score level.

PTSD checklist – civilian version (PCL-C)

The PCL-C (Weathers et al., 1994) was used to capture PTSD symptomatology at pre-deployment, post-deployment and follow-up. The PCL-C is a summated rating scale that yields a continuous measure of PTSD symptom severity, with scores ranging from 17 to 85. The PCL-C also yields symptom cluster sub-scale scores assessing the severity of re-experiencing, avoidance and hyperarousal symptomatology. The civilian measure was preferred over the military iteration as it measures posttraumatic symptomology in response to all traumatic events, not just those linked to stressful military experiences/combat. Respondents were asked how much they have been bothered by each PTSD symptom over the past month on a 5-point severity scale (ranging from 1 =  not at all to 5 = extremely). The PCL-C can be scored by adding up all items for a total severity score, with scores of and above 33 indicative of a PTSD diagnosis (as per the current study; Weathers et al., 1994), or via symptom categories with at least one item endorsed on the re-experiencing subscale, three items endorsed on the avoidance subscale, and two items endorsed on the arousal subscale. In this sample, test-retest reliability was .96 and internal consistency ranged from .92 to .97 for the total scale and subscales. The Cronbach’s alpha for our sample was .94.

Kessler psychological distress scale (K10)

The K10 (Kessler et al., 2002) is a summated rating scale that measures global psychological distress. Respondents were asked how often they experienced anxiety and depressive symptoms (ranging from 1 = none of the time to 5 = all of the time) over the preceding 4-week period. Scores range from 10 to 50 with higher scores indicating a higher severity of psychological distress. The K10 has good demonstrated convergent validity with other measures of distress (Cornelius et al., 2013) and good validity (Bougie et al., 2016). The Cronbach’s alpha for our sample was .88.

Statistical analysis approach

Firstly, we used descriptive analyses to describe demographic and working characteristics. Secondly, bivariate correlations between all the study variables were examined. We then conducted a simultaneous regression analysis to estimate the association of alexithymia and general psychological distress with PTSD across each time point. Next, we used the PROCESS macro for SPSS (Model 1; Hayes, 2017) to test if psychological distress was a moderator on the relationship between alexithymia and PTSD over the three time points (pre-deployment, post-deployment, and follow-up). The PROCESS macro model has the advantage of eliminating multicollinearity by automatically providing the mean-centring function of the independent variable and moderating variable. Furthermore, the method can verify, in more detail, the significance of the simple slope, which is the effect of the independent variable on the outcome variable according to the moderating variable (Hayes, 2017). The PROCESS macro (Model 4; Hayes, 2017) was then used to examine if general psychological distress was a mediator of the relationship between alexithymia and PTSD (total disorder severity and each symptom cluster: re-experiencing, avoidance, and hyperarousal) across each time point. The bootstrapping sample size was set to 2,000 and a 95% confidence interval (CI) did not include 0, indicating a statistically significant mediation. Bonferroni corrections were applied to control for type 1 error within families of tests, with p < .004 indicating statistical significance. Missing data were missing completely at random, therefore listwise deletion was used to handle missing data in all analyses. We included gender in our mediation analyses to examine whether it would warrant further analysis and inclusion within our manuscript. Gender was not found to have a significant effect on our mediation models for total PTSD scores at any time point measured. Therefore, it was not included as a variable within our analyses.

Results

Demographics

Outcome measure descriptive statistics are presented in Table 1 below. The majority of the sample had prior deployment experience (57.7%) and 13.80% of the sample reported prior combat experience. However, previously published a priori analyses for the current sample found that there was no association between prior deployment experience and baseline psychological symptoms for the current sample (Davy et al., 2012). Further, for our sample, previous combat experience was not found to be associated with baseline alexithymia, F_{(1, 218)} = 0.06, p = .81, or baseline general psychological distress, F_{(1, 204)} = 2.87, p = .09. Deployment-related traumatic experiences for the sample are presented in the Supplementary Information (Table S1). The current sample was ethnically homogenous, with 87% of the sample identifying as Australian and being Australian-born. Thirty-five percent of the sample was university educated, with 30.9% having reported completing secondary school and a diploma. The sample was comprised of 39.4% non-commissioned officers, 15.2% commissioned officers, and 45.4% from other ranks. The average length of deployment was 6.72 months (SD = 2.13). Nine percent of the sample reported experiencing a mild traumatic brain injury and 2.3% reported experiencing a physical injury, of which the most prevalent were knee injuries. The most common post-deployment medical complaints reported for the sample were fatigue (49.2%), followed by lower back pain (38.7%), and muscle aches or pains (32.8%). Alexithymia scores, as measured by the TAS-20, and general psychological distress, as measured by the K10, were modestly correlated, r = .19, p < .001.

Table 1. Outcome measure descriptive statistics for overall sample.

	Pre-deployment		Post-deployment		Follow-up
	M (SD)	Range	M (SD)	Range	M (SD)	Range
TAS-20	44.09 (10.30)	20–82
K10	13.88 (5.00)	10–47	13.98 (5.20)	10–47	13.05 (5.17)	10–47
PCL-C	24.03 (5.43)	2–62	22.46 (8.30)	17–77	29.43 (12.66)	18–97
Re-exp.	6.94 (3.37)	5–25	6.35 (2.57)	5–24	7.1 (4.70)	5–25
Avoidance	9.13 (4.52)	6–30	8.77 (3.07)	6–25	3.93 (1.61)	2–10
Arousal	8.25 (3.88)	5–25	7.18 (3.23)	5–24	9.62 (4.00)	6–28

TAS-20 = Toronto Alexithymia Scale; K10 =  Kessler Psychological Distress Scale; PCL-C = Posttraumatic Stress Checklist – Civilian; Re-exp. = PTSD-related re-experiencing; Avoidance = PTSD-related avoidance; Arousal = PTSD-related arousal.

Associations among variables

Correlations among the study’s variables are presented in the Supplementary Information, Table S2. Alexithymia was correlated with psychological distress, total PTSD across the three time points, and all symptom clusters (except for pre-deployment re-experiencing and follow-up re-experiencing and avoidance), r = .139 to .346, with higher levels of alexithymia associated with greater psychological distress and PTSD symptomology. Psychological distress correlated with all variables, r = .189 to .757, with more psychological distress being associated with greater PTSD symptomology.

Regression analyses

A simultaneous regression was conducted to determine whether both alexithymia and psychological distress were uniquely associated with PTSD symptom severity across the three time points (see Table 2). Alexithymia and psychological distress were found to be unique predictors of PTSD across all three time points.

Table 2. Associations of alexithymia and psychological distress with PTSD.

Model		Unstandardised Coefficients		Standardised Coefficients	t	Sig.	95.0% Confidence Interval for B
		B	Std. Error	Beta			Lower Bound	Upper Bound
T1	TAS	.140	.012	.311	12.085	< .001	.117	.163
	K10	.313	.024	.338	13.122	< .001	.266	.360
T2	TAS	.172	.023	.172	5.95	<.001	.092	.182
	K10	1.227	.033	.739	36.683	< .001	1.161	1.292
T3	TAS	.153	.050	.140	3.086	.002	.056	.251
	K10	.756	.125	.296	6.056	< .001	.511	1.001

T1 = pre-deployment; T2 = post-deployment; T3 = follow-up.

Moderation

For all models, general psychological distress was not found to be a significant moderator. Therefore, only the mediation models were further explored.

Pre-deployment: model of alexithymia and PTSD mediated by distress

See Figure 1 below for a summary of the tested pre-deployment mediation models. Alexithymia was associated with greater PTSD severity at pre-deployment. General psychological distress was found to be a partial mediator of the relationship between alexithymia and PTSD severity at pre-deployment (see Figure 1(A)). The mediation model accounted for 25.05% of the variance in total PTSD, F_{(2, 1173)} = 196.02, p < .001, 95% CI [0.1172, 0.1626]. Examination of specific indirect effects revealed a significant path through general psychological distress b = .003, 95% CI [.0143, .1015], in that those with greater levels of alexithymia reported greater psychological distress, which in turn was associated with greater PTSD severity, at pre-deployment.

Figure 1. Pre-deployment mediation models for PTSD symptomology.

Note: Unstandardised path coefficients and standard errors are shown beside each line. Significant direct and indirect paths are boldfaced. ***p < .001.

Alexithymia was also associated with greater hyperarousal symptomology at pre-deployment. No main effect of alexithymia on re-experiencing or avoidance symptomology was observed (see Figure 1(B and C), respectively). However, a significant mediation model emerged for re-experiencing when psychological distress was included as a mediator for re-experiencing, F_{(2, 1173)} = 15.16, p = .001, 95% CI [0.1074, 0.2360], explaining 26.73% in variance in the relationship between alexithymia and re-experiencing symptoms at pre-deployment, b = .16, 95% CI [.0835, .3683],. A further significant mediation model also emerged for avoidance symptoms at pre-deployment, F_{(2, 1173)} = 16.53, p < .001, 95% CI [0.1490, 0.3288], explaining 28.02% of the variance in the relationship between alexithymia and avoidance symptoms at pre-deployment, b = .02, 95% CI [.1452, .6051],. General psychological distress was found to be a partial mediator of the relationship between alexithymia and hyperarousal at pre-deployment (see Figure 1(D)). The mediation model accounted for 31.19% of the variance in total hyperarousal symptoms, F_{(2, 1173)} = 20.90, p < .001, 95% CI [0.1443, 0.3009], in that those with greater levels of alexithymia reported greater psychological distress, which in turn was associated with greater hyperarousal symptomology, at pre-deployment, b = .02, 95% CI [.0560, .2887]. Our results demonstrate that general psychological distress fully mediated the relationship between alexithymia and re-experiencing and avoidance symptomology at pre-deployment, and partially mediated the relationship between alexithymia and total PTSD severity and hyperarousal symptomology at pre-deployment.

Post-Deployment: model of alexithymia and PTSD mediated by distress

See Figure 2 below for a summary of the tested post-deployment mediation models. Alexithymia was associated with greater PTSD severity at post-deployment, General psychological distress was found to be a partial mediator of the relationship between alexithymia and PTSD severity at post-deployment (see Figure 2(A)). The mediation model accounted for 55.00% of the variance in total PTSD, F_{(2, 1140)} = 70.51, p < .001, 95% CI [1.1611, 1.2924] at post-deployment, b = .28, 95% CI [.0409, .2406]. Examination of specific indirect effects revealed a significant path through general psychological distress, in that those with greater levels of alexithymia reported greater psychological distress, which in turn was associated with greater PTSD severity, at post-deployment.

Figure 2. Post-Deployment Mediation Models for PTSD Symptomology.

Note: Unstandardised path coefficients and standard errors are shown beside each line. Significant direct and indirect paths are boldfaced. ***p < .001.

Alexithymia was also associated with greater hyperarousal symptomology at post-deployment. No main effect of alexithymia on re-experiencing or avoidance symptomology was observed (see Figure 2(B and C), respectively). However, a significant mediation model emerged when psychological distress was included as a mediator for re-experiencing, F_{(2, 1140)} = 29.26, p = .001, 95% CI [0.2561, 0.3023], accounting for 58.28% of the variance in the relationship between alexithymia and re-experiencing symptoms at post-deployment, b = .03, 95% CI [.2154, .6335]. A further significant mediation model emerged for avoidance symptoms at post-deployment, F_{(2, 1140)} = 59.95, p < .001, 95% CI [0.4154, 0.4670], explaining 50.02% of the variance in the relationship between alexithymia and avoidance symptoms at post-deployment, b = .05, 95% CI [.1153, .5715]. General psychological distress was found to be a partial mediator of the relationship between alexithymia and hyperarousal at post-deployment (see Figure 2(D)). The mediation model accounted for 71.17% of the variance in total hyperarousal symptoms, F_{(2, 1140)} = 59.20, p < .001, 95% CI [0.4176, 0.4694], in that those with greater levels of alexithymia reported greater psychological distress, which in turn was associated with greater hyperarousal symptomology, at post-deployment, b = .05, 95% CI [.1452, .6051]. Our results demonstrate that general psychological distress fully mediated the relationship between alexithymia and re-experiencing and avoidance symptomology at post-deployment, and partially mediated the relationship between alexithymia and total PTSD severity and hyperarousal symptomology at post-deployment.

Follow-up: model of alexithymia and PTSD mediated by distress

See Figure 3 below for a summary of the tested follow-up mediation models. Alexithymia was associated with greater PTSD at follow-up. General psychological distress was found to be a partial mediator of the relationship between alexithymia and PTSD at follow-up (see Figure 3(A)). The mediation model accounted for 10.13% of the variance in total PTSD, F_{(1, 397)} = 21.00, p < .001, 95% CI [0.5105, 1.0013], b = .07, 95% CI [.0239, .1002]. Examination of specific indirect effects revealed a significant path through general psychological distress, in that those with greater levels of alexithymia reported greater psychological distress, which in turn was associated with greater PTSD severity, at follow-up.

Figure 3. Follow-up mediation models for PTSD symptomology.

Note: Unstandardised path coefficients and standard errors are shown beside each line. Significant direct and indirect paths are boldfaced. ***p < .001.

Alexithymia was also associated with greater hyperarousal symptomology at pre-deployment. No main effect of alexithymia on re-experiencing or avoidance symptomology was observed (see Figure 3(B and C), respectively). However, a significant mediation model emerged when psychological distress was included as a mediator for re-experiencing, F_{(2, 395)} = 15.15, p = .001, 95% CI [0.1074, 0.2360], b = .02, 95% CI [.0835, .3683] and avoidance symptomology, F_{(2, 395)} = 16.59, p < .001, 95% CI [0.0571, 0.1227], at follow-up, b = .01, 95% CI [.1788, .6969]. General psychological distress was found to be a partial mediator of the relationship between alexithymia and hyperarousal at follow-up (see Figure 3(D)). The mediation model accounted for 8.85% of the variance in total hyperarousal symptoms, F_{(2, 395)} = 19.04, p < .001, 95% CI [0.1332, 0.2925], in that those with greater levels of alexithymia reported greater psychological distress, which in turn was associated with greater hyperarousal symptomology, at follow-up, b = .002, 95% CI [.0523, .2873]Our results demonstrate that general psychological distress fully mediated the relationship between alexithymia and re-experiencing and avoidance symptomology at follow-up, and partially mediated the relationship between alexithymia and total PTSD severity and hyperarousal symptomology at follow-up.

Discussion

The present study examined the interactive relationship between alexithymia, general psychological distress, and PTSD severity for military personnel at pre-deployment, post-deployment, and follow-up. We found that psychological distress was not a moderator of the relationship between alexithymia and PTSD severity at any time point. This suggests that one’s level of alexithymia is relatively stable (Mikolajczak & Luminet, 2006; Luminet et al., 2001, 2007) and does not change significantly in the context of modulating psychological distress. Although these previous studies indicated that alexithymia scores did change with large changes in psychological distress or psychopathology, these scores remained associated with baseline levels. Our results show that while alexithymia is correlated with distress, the magnitude of psychological distress does not change the magnitude or direction of the relationship between alexithymia and PTSD. This finding may also be explained by the nature of our sample, given that the relationship between alexithymia and PTSD symptomology is most prominent for military samples (for a review, see Edwards, 2019), which is hypothesised to stem from cumulative and continued traumatic exposures. However, it is important to note that we only captured alexithymia at pre-deployment (as it was defined as a trait), therefore, any inferences about relative and absolute stability should be considered with caution.

Psychological distress was instead found to be a partial mediator of the relationship between alexithymia and total PTSD severity across all three time points. It may be that trait alexithymia generally interferes with adaptive emotion regulation processes, resulting in increased psychological distress, which in turn exacerbates PTSD symptomology – this is supported by McCaslin et al. (2006), who found that baseline alexithymia predicted post-9/11 PTSD in urban New York City police officers (a time of considerably increased psychological distress), over and above pre-9/11 symptoms. Further, trauma-exposed individuals with alexithymia, when under increased distress, show a de-coupling of subjective and psychophysiological affective components (Putica et al., 2022b). Overall, this suggests that alexithymia increases one’s vulnerability to the development of pathology when one experiences increased psychological distress.

At a PTSD symptom cluster level, general psychological distress was also found to partially mediate the relationship between alexithymia and hyperarousal symptomology. It may be that interoception (or, at least, one’s interoceptive sensibility – the extent to which individuals report being able to perceive their internal sensations) may be at play within this relationship. For example, it may be that a heightened sensibility for interoceptive signals, combined with difficulty in attributing these sensations to emotions, increased an individual’s vulnerability to psychopathology (Palser et al., 2018). In our sample, those with alexithymia may read that any bodily signals are catastrophic or dangerous, ultimately perpetuating distress and fear-based pathology (for example, see Putica et al., 2022a).

On the other hand, general psychological distress was found to fully mediate the relationship between alexithymia and re-experiencing and avoidance symptomology. The finding that the relationship between alexithymia and avoidance is fully mediated by psychological distress is somewhat surprising as alexithymia has been shown to be associated with a propensity for avoidance-based coping (Bilotta et al., 2016; Panayiotou et al., 2015). One would expect that an avoidance-based coping style would therefore be independently associated with alexithymia. However, it may be that, for our military sample, there is less opportunity for one to engage in avoidance-based coping, especially while maintaining deployment readiness. Taken together, our results suggest that those with greater levels of alexithymia are vulnerable to experiencing PTSD-related re-experiencing or avoidance symptomology during periods of increased psychological distress.

Overall, our results show that the relationship between alexithymia and psychopathology may not be linear but rather interactive in nature, such that moderators/mediators create the “perfect storm” for the development of psychopathology. Our findings that alexithymia seems to be a trait vulnerability that is amplified during certain conditions (e.g. increased psychological distress) may explain why alexithymia is a transdiagnostic risk factor for the development of various psychopathologies (Celikel et al., 2010; Edwards, 2019; Nowakowski et al., 2013).

Clinical implications

Our investigation into the interactive relationship between alexithymia, psychological distress, and psychopathology has several important clinical implications. Firstly, our findings show that those with alexithymia may be at increased risk of developing PTSD (and potentially other psychopathology) when experiencing an increased period of distress, which suggests that regular proactive screening for alexithymia may be indicated. For example, providing psychoeducation to a client about how and when they may be at greater risk of pathology may be pivotal in increasing awareness/insight, promoting help-seeking, and then engaging in treatment targeting emotion regulation mechanisms if indicated. Considering alexithymia in assessment and treatment planning is paramount as it has been found to account for 30% of PTSD treatment outcome variance (Kosten et al., 1992). Furthermore, given the robust relationship between general psychological distress and PTSD, regular, pro-active screening of psychological distress among military personnel may be indicated to screen those at risk of developing (further) psychopathology.

Limitations and future research

We note several limitations of this study. Firstly, females were underrepresented in the current study. While we acknowledged that females are underrepresented in defence forces, they only represented 10.12% of our sample – as such, results should be interpreted with some caution, and the replicability of our findings should be tested across genders. Similarly, the sample was ethnically homogenous, with approximately 87% of participants being Australian-born and identifying as Australian. Secondly, we only captured alexithymia at pre-deployment – assessing alexithymia at multiple time points may have afforded us more scope to investigate the stability of alexithymia, discuss possible aetiologies of alexithymia for our sample, and further elucidate the distinctions and nature of the interactions between alexithymia, general psychological distress, and PTSD across time. A further limitation of the current study was the decreasing sample size at post-deployment and follow-up. Similarly, due to the convenience sampling utilised, we do not know if our participants differed from those who did not volunteer for the study. A further potential limitation is the use of a mediation model at pre-deployment as it can be difficult to discern how variables are placed within the model. Given the established findings that alexithymia is a trait variable and robust predictor of PTSD, and psychological distress is more akin to a state phenomenon, we included ordered the model to include alexithymia as the predictor and general psychological distress as a mediator. However, due to the known limitations of such an approach our results for the pre-deployment model should be interpreted with some caution. Finally, the use of self-report measures to assess alexithymia is inherently limiting, as those with difficulties in recognising emotions in themselves and others would not have objective insight into their experience. However, there is considerable evidence for self-report measures of alexithymia as the nature of the construct is inherently about how people are experiencing their emotions, so, to some extent, it is difficult to target that without directly asking the individual. The TAS-20 is also limited when it comes to assessing facet levels of alexithymia – for those investigations, newer measures, such as the Perth Alexithymia Questionnaire (Preece et al., 2018), should be considered. Future studies would benefit from utilising clinician-administered interviews (or a multimodal approach), such as the Toronto Structured Interview for Alexithymia (Bagby et al., 2006), or psychophysiological measures that allow for multimethod measurement and decrease reliance on participants who may not have the most accurate insight into their emotional recognition ability.

Conclusion

Ongoing debate ensues regarding the interplay between alexithymia, psychological distress, and PTSD. However, research exploring the relationships between these constructs has relied on cross-sectional designs, limiting inferences regarding these relationships. We examined whether there is an interactive relationship between alexithymia and PTSD, with general psychological distress acting as a moderator or a mediator over time. Psychological distress was found to be a complementary partial mediator of the relationship between alexithymia and total PTSD severity and hyperarousal symptomology across all three time points. Psychological distress was a full mediator of the relationship between alexithymia and re-experiencing and avoidance symptomology across all three time points. The interactive relationship between these constructs suggests that those with higher alexithymia are at increased risk of developing PTSD and greater hyperarousal symptomology, irrespective of temporal proximity to trauma exposure. Further, our results suggest that alexithymia is a predisposing factor that increases one’s vulnerability to total PTSD severity and to re-experiencing, avoidance, and hyperarousal symptomology during periods of increased psychological distress, irrespective of temporal proximity to trauma exposure. Overall, general psychological distress had a differentiated mediating effect on PTSD symptomology, indicating that there may be complex interactions occurring at a cluster level that should be investigated to elucidate the interplay between alexithymia, psychological distress, and PTSD.

Disclosure statement

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

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.