Longitudinal study of PTSD and depression in a war-exposed sample – comorbidity increases distress and suicide risk

Traumatic experiences are frequent and widely spread in contemporary world, causing acute and chronic stress that might produce significant psychological sequelae and social consequences, especially to vulnerable people (Lecic-Tosevski et al., 2016, 2018).

Post-traumatic stress disorder (PTSD) and major depressive disorder (MDD) are among the two most common psychiatric sequelae of trauma that might develop independently. However, their co-occurrence could be considered a norm rather than an exception (O’Donnell et al., 2014). The comorbidity of PTSD and MDD was found in as many as 50% of cases following trauma (Charlson et al., 2012; Richardson et al., 2017) and it was also high across diverse epidemiological samples (Caramanica et al., 2014; Rytwinski et al., 2013). The co-occurrence of these two disorders is associated with greater symptom severity and psychological distress as well as higher levels of functional impairment and reduced life satisfaction and quality of life, compared to those in individuals with PTSD or MDD alone (Byllesby et al., 2017; Contractor et al., 2018; Goenjian et al., 2011; Post et al., 2011). This pattern of findings was previously reported in refugees (Momartin et al., 2004), various patient groups (Holtzheimer et al., 2005; Shalev et al., 1998) as well as in the studies of war veterans (Bowler et al., 2016; Ikin et al., 2010; Raab et al., 2015).

Individuals with comorbid PTSD and depression have more severe depressive symptoms, higher impulsivity and hostility and are more likely to make suicide attempt, compared to subjects with depression alone (Kimbrel et al., 2016; Sher, 2005). It was found that presence of both disorders was associated with prevalence rates of suicidal ideation nearly twice as high as having only one of these disorders (Cougle et al., 2009). The findings from a recent longitudinal study on Iraq and Afghanistan veterans suggested that comorbid PTSD–depression may be a significant risk factor for future suicidal behaviour (Kimbrel et al., 2016). Tural et al. (2012) found that comorbid PTSD and MDD patients had significantly lower rates of recovery from PTSD in comparison to those with PTSD but without MDD.

There is evidence that pre-existing major depression may render an individual more vulnerable to PTSD in the aftermath of trauma (Breslau et al., 2000) and vice versa (Horesh et al., 2017; Kessler et al., 1995). Furthermore, participants with a higher number of traumatic events were more likely to receive a PTSD–depression comorbid diagnosis (Kaltman et al., 2010).

The aim of our study was to investigate the extent to which PTSD and depression co-occurred in Serbian general population after a major trauma at baseline and 1 year after the follow-up, as well as how the co-occurrence of MDD and PTSD was associated with sociodemographic factors, personal distress, suicidality and quality of life.

Data on the participants’ age, gender, education level, income, employment and marital status were obtained through a brief structured sociodemographic questionnaire.

The history of potential traumatic experience was assessed using a specifically amended version of the Life Stressor Checklist-Revised (LSC-R) (Wolfe and Kimerling, 1997). The presence of PTSD, depression and suicidality was assessed by the MINI-5 (Sheehan et al., 1998), a structured diagnostic interview assessing the symptom criteria used in the Diagnostic and Statistical Manual of Mental Disorders (DSM)-IV. Inter-rater reliability was assessed in two mock interviews. Agreement on an item was reached when the interviewers gave the same answer for it. The Brief Symptom Inventory (BSI) (Derogatis and Melisaratos, 1983) was used as a measure of general psychological symptoms. The IES-Revised (IES-R) (Weiss and Marmar, 1997) was applied for the assessment of post-traumatic stress reactions, and the Manchester Short Assessment of Quality of Life (MANSA) (Priebe et al., 1999) for the assessment of quality of life. The interviews were conducted face-to-face by trained psychiatrists and psychologists.

The study was approved by the Ethics Committee of the Belgrade University School of Medicine. All participants provided written informed consent after being fully informed about the study content and procedures.

Although the starting sample consisted of 640 participants, in order for the findings from two phases (at baseline and follow-up) to be comparable, we included only 159 subjects who participated in both the phases in this analysis. The average age of this subsample was 43.61 (SD=10.74), ranging from 22 to 65 years. Females formed 67.9% of the sample. Also, 63.5% of the sample were married, 11.9% single, 13.8% divorced, 10.1% widowed and 0.6% were cohabitating. The minimum years of education was 4 and the maximum was 22, with a mean of 12.14 (SD=3.04). As for the employment status, 43.4% of the sample were currently employed, 32.7% unemployed and 17% were retired, while 5.7% of the participants were students. Also, 13.2% participants had previously participated in combat.

In spite of an increasing interest shown on post-traumatic conditions in war-affected populations, a comprehensive epidemiological profile of PTSD–depression comorbidity is lacking. Particularly scarce are studies with a longitudinal approach. Our study has shown that PTSD and depression were common in general Serbian population after major trauma, as well as their co-occurrence. The finding of a high overlap between PTSD and depression is partly consistent with some studies. The US National Comorbidity Survey reported the lifetime comorbidity between PTSD and depressive disorders to be higher than 50% in the general population (Kessler et al., 1995). Similar comorbidity results were found in several studies with war veterans – 50% (Bleich et al., 1997), 51% (Constans et al., 1997) and 52% (Ikin et al., 2010). A recent study conducted among Canadian veterans as well as currently serving Forces members has shown 61.5% of PTSD+MDD comorbidity in the sample (Richardson et al. 2017).

The rate of PTSD and depression varies highly in different post-conflict populations (Johnson and Thompson, 2008). A cross-sectional community study in South Sudan has shown the rate of PTSD to be 37.6% and MDD to be 15.9% in the sample, while their comorbidity was identified in 9.5% of the total sample (Ayazi et al., 2012), which are significantly lower than our findings. A recent study in South Lebanese civilian sample has shown PTSD+MDD comorbidity to be present in 23.4% of the sample (Farhood et al., 2016), which is also lower than the findings of our study.

The strong association between PTSD–depression comorbidity and high psychological distress, found in our study, is in accordance with other studies. A study on South Sudan war-affected population has shown a higher level of psychological distress in participants with PTSD–depression comorbidity than in participants with either condition alone (Ayazi et al., 2012). A recent study on combat veterans from the Ohio National Guards has shown that PTSD+MDD comorbidity is more related to underlying general distress or negative affectivity than the symptom categories of the PTSD diagnostic criteria (Byllesby et al., 2017). Furthermore, our findings provide support to a previous research indicating that PTSD and depression symptoms are negatively correlated with quality of life (Ikin et al., 2010; Pittman et al., 2012). Raab et al. (2015) found that combat veterans with comorbid PTSD+MDD reported significantly worse satisfaction-related quality of life than those with PTSD alone. A study on victims of civilian violence has shown that the presence of comorbid mental disorders as well as the severity of post-traumatic symptoms remained the strongest predictor for impaired health-related quality of life in PTSD outpatients (Pagotto et al., 2015).

In our study, individuals with PTSD+MDD comorbidity were more likely to report current suicide risk, which was also reported in other studies of post-traumatic conditions. The findings of a prospective study of Iraq/Afghanistan veterans suggested that PTSD+MDD comorbidity may be a significant risk factor for future suicidal behaviour (Kimbrel et al., 2016). Our findings have shown that presence of only one disorder (PTSD or MDD) was related to increased suicidality, in accordance with the study of Arenson et al. (2018) who found that 49% of war veterans with PTSD and depression comorbidity had suicidal ideations, unlike 11% of subjects with PTSD alone or 34% with MDD alone. Another cross-sectional and prospective analysis of a national probability sample of women confirmed a strong association between suicidality and PTSD/depression comorbidity (Cougle et al., 2009; Oquendo et al., 2003).

In the follow-up phase, we found a significant association between number of traumatic events and years of education with PTSD–depression comorbidity. This is in line with previous studies showing that participants with a higher number of traumatic events were more likely to develop PTSD–depression comorbidity (Ayazi et al., 2012; Shalev et al., 1998). In contrast with most previous studies (Ayazi et al., 2012; Roberts et al., 2009), but consistent with that of Tracy et al. (2011), our study did not show significant association between sociodemographic factors and PTSD+MDD comorbidity, with the educational level being an exception. This is in accordance with recent findings showing that the educational status significantly predicts co-occurrence of PTSD+MDD (Farhood et al., 2016). A possible explanation could be that persons with lower education may have lower cognitive capacity to cope with traumatic events (Priebe et al., 2009) and lower ability to use healthy coping strategies.

The limitation of our study might be related to the accuracy of data, since they are based on respondent’s memory. Exposure to traumatic events was assessed retrospectively and might have been influenced by a recall bias. This, however, is the problem of most studies (Mollica et al., 2007). Also, only part of the participants took part in both phases, since we deliberately selected participants who fulfilled the criteria for PTSD; hence, the sample shown in this paper might be biased and does not provide representative rates for the population. The major strengths of our study are its longitudinal design, the sample being representative of the general population as well as the sample size.

High prevalence of PTSD/MDD comorbidity has increased our focus on post-traumatic conditions, particularly in war-affected populations, during the last decade. Co-occurrence of these two disorders following traumatic experiences is associated with functional impairment, disability and prolonged suffering. Our findings may have implications for understanding complex psychopathology following trauma. Symptom severity, high personal distress and high suicidal risk are important both for individual and public mental health. Therefore, particular attention should be paid to screening both for PTSD and depression in traumatised populations. Comorbidity of PTSD and depression is important in terms of treatment efficacy. The consensus guidelines have suggested that the presence of MDD should change the treatment practice of PTSD (Flory and Yehuda, 2015; Zoellnar et al., 2014).