The impact of forced displacement: trauma, increased levels of inflammation and early presentation of diabetes in women Syrian refugees

Abstract Background Forced displacement and war trauma cause high rates of post-traumatic stress, anxiety disorders and depression in refugee populations. We investigated the impact of forced displacement on mental health status, gender, presentation of type 2 diabetes (T2D) and associated inflammatory markers among Syrian refugees in Lebanon. Methods Mental health status was assessed using the Harvard Trauma Questionnaire (HTQ) and the Hopkins Symptom Checklist-25 (HSCL-25). Additional metabolic and inflammatory markers were analyzed. Results Although symptomatic stress scores were observed in both men and women, women consistently displayed higher symptomatic anxiety/depression scores with the HSCL-25 (2.13 ± 0.58 versus 1.95 ± 0.63). With the HTQ, however, only women aged 35–55 years displayed symptomatic post-traumatic stress disorder (PTSD) scores (2.18 ± 0.43). Furthermore, a significantly higher prevalence of obesity, prediabetes and undiagnosed T2D were observed in women participants (23.43, 14.91 and 15.18%, respectively). Significantly high levels of the inflammatory marker serum amyloid A were observed in women (11.90 ± 11.27 versus 9.28 ± 6.93, P = 0.036). Conclusions Symptomatic PTSD, anxiety/depression coupled with higher levels of inflammatory marker and T2D were found in refugee women aged between 35 and 55 years favoring the strong need for psychosocial therapeutic interventions in moderating stress-related immune dysfunction and development of diabetes in this subset of female Syrian refugees.


Introduction
Refugees experience high rates of post-traumatic stress disorder (PTSD), anxiety and depression from forced displacement and civilian war trauma.As of January 2020, it is estimated that ∼5.5 million Syrian refugees have fled conflict and violence and settled predominantly in the Middle East and North Africa region, mainly in Turkey, Lebanon, Jordan and Iraq, outside of camp settings in both urban and semi-urban areas. 1 Lebanon has the highest per capita concentration of Syrian refugees in the world with an estimated 183 refugees per 1000 inhabitants (UNHCR 2015).
Studies have shown that refugees endure high levels of acute stress and adverse lifestyle changes due to their forced displacement. 2Changes in diet, social isolation and exclusion, limited access to healthcare, separation from family, stress, anxiety and trauma associated with violence have been associated with the occurrence of chronic non-communicable diseases such as type 2 diabetes (T2D). 3In addition, clinically meaningful reductions in PTSD symptoms are associated with a lower risk of T2D. 4 Furthermore, forced displacement contributes to psychosocial stress and vulnerability; both of which are directly associated with increases in serum blood glucose levels and T2D.Moreover, PTSD has been shown to alter the production and secretion of pro-inflammatory cytokines and acute phase proteins [e.g.Creactive protein (CRP) and serum amyloid A (SAA)], which stimulate an immune response through the induction of NF-κB. 5,6Chronic activation of these pathways over periods of time can result in physiological responses becoming sensitized, dysfunctional and maladaptive.Previous research on the Syrian refugee populations in Lebanon and neighboring countries has demonstrated a high prevalence of T2D among the adult refugees. 7ro-inflammatory cytokines have also been shown to be elevated in individuals exposed to trauma and different forms of abuse compared with non-exposed healthy individuals. 8oupled to the pro-inflammatory cytokine response, levels of acute phase proteins such as CRP and SAA can also change during periods of trauma and infection. 9,10Increased levels of CRP have been reported in individuals with generalized anxiety disorder.Although conflicting studies report no relationship between PTSD and CRP, they have highlighted the potential role of demographic factors such as race and sex in determining the response to trauma. 114][15] The comprehensive data collected for this study from the Syrian refugee population in Lebanon present a unique opportunity to explore factors that may be involved in the development of T2D, due to the trauma of forced displacement.In this study, we investigate the relationship between stress, inflammation, psychopathology, development of T2D and obesity in trauma-exposed Syrian refugees.

Methods
The findings presented here are derived from the first set of data collected from a pilot cohort of 303 Syrian refugees, recruited from primary healthcare centers (PHCs), Lebanon between June 2019 and March 2020.Details on ethical approvals, study population, data collection and clinical assessment, mental health assessment, variables definition and statistical analysis are provided in the supplementary materials.The mental health of the recruited subjects was assessed using the Harvard Trauma Questionnaire (HTQ) and the Hopkins Symptom Check List-25 (HSCL-25) questionnaires.The HTQ and HSCL are the diagnostic tools for PTSD and anxiety/depression, respectively.

Demographic structure of the study population
This study included participants above 18 years old attending a PHC in Lebanon.The average age of the study participant was 47 years, and the study population was composed of 32.3% men and 67.6% of women.A large number or participants were from Aleppo (48.84%),Idlib (8.25%) and Raqqa (6.6%) where most of the fighting occurred (Fig. S1).The average age of men and women participants was 50.57± 12.52 and 44.28 ± 12.89, respectively.The average body mass index (BMI) of the study participants was 29.38 kg.m 2 with women displaying a substantially higher BMI (30.81 ± 6.69 kg.m 2 ) than their men counterparts (27.94 ± 4. 69 kg.m 2 ) with P value < 0.0001.About 55.57% study participants fell in the category of overweight and obese, whereas 4.95% of women participants were extremely obese.
Patient demographic and familial history of metabolic diseases are summarized in Table 1.The majority of the study participants (73.93%) were < 55 years old and 90.33% were married.Most participants self-reported familial metabolic disease such as T2D (59.74%), hyperlipidemia (26.40%) and hypertension (58.08%).Approximately half of the study participants over 55 years old, irrespective of gender, had a family history of T2D or hypertension (59.74 and 58.08%).
No significant difference in the prevalence of T2D and prediabetes was observed between men and women participants (Table S1).Although no difference was seen in the fasting blood sugar (FBG) and HbA1c levels, it is observed to be increased with age in both genders.Significantly high levels of uric acid and creatinine were observed in men (5.33 ± 1.75 and 0.89 ± 0.62 mg/dl) with P values of 0.0033 and < 0.000001 (Table 2).Chi-square test was performed for the categorical variables and the t-test was performed for the mean values.Categorical variables were displayed as counts and percentages.

Predictive model of SAA and CRP with stress scores for T2D in refugee population
To identify predictive trends for the onset of T2D progression in the refugee population, we carried out receiver operating characteristic analysis.Among the inflammatory markers tested, SAA levels exhibited a significantly high predictive value of 63.7% (P < 0.0001) followed by the CRP levels of 60.8% (P = 0.0040) (Fig. S2).Similarly, the predictive trend employed for the anxiety/depression and HSCL-25 predicted T2D onset with 59.7% (P = 0.012) followed by HTQ (57.8%) with P value of 0.0040.The levels of the inflammatory markers were investigated across all study groups: ND, PDD and NPDD (Fig. 1).The CRP levels were significantly higher in the PDD and NPDD groups, whereas SAA levels were significantly high in the NPDD group when compared with the ND.The SAA levels did not vary between the ND controls and PDD groups.The Hopkins anxiety/depression scores were significantly high in the PDD and NPDD groups when compared with the ND control group.The HTQ however, was significantly higher in the NPDD group compared with the ND.We sought to test whether the effects of trauma (PTSD), anxiety and depression, measured by the Harvard and Hopkins scores, impacted the onset of diabetes and obesity.
The regressions predicted T2D according to the anxiety/depression scores without (Table S4) and with (Table S5) adjustment by age, hypertension and hyperlipidemia.The results of these regressions are also presented in Fig. S3.Within the estimated sampling variation, the two anxiety/depression scores did not yield significant associations.Regressions analysis predicting T2D according to the anxiety/depression scores with adjustment by sex are shown in Table S6 and Fig. S3c and d.The results were suggestive, but not significant, given the estimated sampling variation ranges.The P values of multivariate logistic regression model showed variations within expected ranges for sampling variations.

Main finding of this study
We report a link between the inflammatory profile, gender and the development of T2D in a refugee population.Young Syrian refugee women, especially between the ages of 35 and 55 years exhibit more chronic PTSD trajectories and higher levels of inflammatory and anxiety/depression markers than men.The elevated levels of acute phase proteins of CRP and SAA are suggested to be the potential indicators of trauma-related pathophysiology and systemic inflammation.These markers have been linked to an increased risk, and later development, of diabetes among women experiencing war trauma.This is one of the unique studies that explicitly investigates the link between the inflammatory profile, gender and the development of T2D in a traumatized population.

The gender effects
The gender is an important risk factor for the development of trauma-related disorders among subjects exposed to specific forms of potentially traumatic events. 168][19][20][21] Although both men and women in our study had symptomatic anxiety/depression scores with HSCL-25, only women had considerably higher mean anxiety/depression scores and this was most apparent in the age group of 35-55.Gender differences in the proinflammatory cytokine response after severe trauma may account for the development of trauma-related psychopathologies 22 and suggest that gender is a biological variable that moderates the relationship between inflammation and psychological symptoms. 12Subclinical inflammation mediated by body fat has been shown to affect CRP levels more strongly in women than in men. 23Accordingly, women participants showed higher prevalence of overweight than men, which accounts for the lower CRP levels seen in men.Also, the differences in the number of study participants may have, in part, influenced the gender effects observed.The number of male subjects in the study population was considerably less than the female subjects and this difference may have had an impact on gender effect observed.Anxiety and depression are powerful players in the regulation and dysregulation of the hypothalamic-pituitary-adrenal (HPA) axis and its links to other cerebral circuits.The HPA axis, the autonomic nervous system and the immune system work together in harmonizing the hormonal inflammatory stress response. 24,25Sex-specific epigenetic and neuroendocrine influences and immune responses contribute to differential responses to trauma.The females show increased glucocorticoid production in response to various acute stressors supported by the findings of sex differences at all levels of the HPA axis.The impact of anxiety/depression on women can be magnified by at least two factors: leaving their primary home under duress and caring for their children while displaced.Women have been reported to have a larger cortisol response to stress and lower proinflammatory cytokines, which impact the stress response chain linked with the HPA axis and are associated with greater PTSD risk.

Obesity
Trauma is high among Syrian refugees 26,27 and several studies have shown that physiological stress that arises from the occurrence of traumatic life changing events also increases the risk of obesity and have adverse effects on the body's physiological functions. 280][31] In line with previous reports, we found a higher prevalence of obesity in women and higher prevalence of overweight in men.Several traumatic factors have been identified as a risk factor for increased obesity in refugee women, but the mechanism by which obesity regulated stress remains unclear. 32,33

Inflammation
Numerous findings indicate a link between PTSD, depression and anxiety with inflammation. 34The psychological response to stress causes various endocrine perturbations, which lead to obesity, inflammation and insulin resistance. 35,36Inflammation appears to be higher in adults exposed to early life adverse events. 35,36In particular, elevated levels of acute phase proteins such as CRP and SAA have been associated with traumarelated pathophysiology as levels of these proteins have been found to increase during infection and trauma. 11In this study, refugee women had significantly higher levels of inflammatory marker SAA.Although CRP levels in women with PTSD are not always correlated, it has been reported that SAA levels correlate with PTSD symptoms in adolescent girls and women. 37,38Our results confirm that the effect of stress on inflammation is greater in traumatized women than men.As with the HTQ and HSCL-25 scores, the levels of both CRP and SAA are most elevated in women aged 35-55 years.

Inflammation and T2D
Large-scale genetic and epidemiological studies have underlined the prominent role of environmental factors in T2D etiology. 39High prevalence of T2D was observed in the refugee population studied.This prevalence is significantly higher than that in the general Syrian population 40 as well as in the Lebanese population 41 where the Syrian refugee population under study is currently living.Likewise, a nearly 2.5-fold increase in the prevalence of pre-diabetes was observed Syrian refugee population when compared with the general Syrian population and the host Lebanese population. 42Interestingly, women displayed significantly higher prevalence of early diabetes at a very young age than men.These results suggest that women are more prone to physiological stress mediated by obesity and inflammation, which is likely to increase the risk of developing T2D at an early age.
The inflammatory markers of SAA and CRP analyzed in our study, while generally on the lower levels considering the relatively young age (mean age 47.4) of the study population, are found to be increased in patients with T2D. 43,448][49] Furthermore, our findings of higher SAA and CRP levels in NPDD and PD women are consistent with the results from prospective studies that have reported associations between depressive symptoms and T2D in women. 50Although we cannot rule out the role of poor nutrition and other socioeconomics factors, increased prevalence of T2D in refugees in our study, measurements of inflammatory markers (SAA and CRP) and the results from the trauma and anxiety and/depression questionnaires provide confirmatory evidence of relationships between the psychological stress and T2D risk 51 and suggest their potential impact in the development of T2D in the study population.

Limitations of this study
Psychopathology was determined in our study from selfreported questionnaires rather than a clinical interview, which would have given more details on the trauma experienced post displacement.However, two independent self-reported trauma questionnaires were administered in an adapted Arabic form to suit the study subjects.A longitudinal study would have been measured for the levels of inflammatory markers and glycemic control over time and would have allowed for repeated tests of recent T2D and pre-diabetes in the study sample, which together with sample sizes limits the study's ability to isolate emergent and prodromal T2D cases using blood markers.Furthermore, determining the undiagnosed diabetes before displacement was challenging, and hence they were not considered as cases.Our study is further limited with the lack of control groups attending the same PHCs.Notwithstanding the limitations of the study, the uniqueness of our findings in this small population provides a significant contribution to the understanding of the relationship between trauma-related disorders, gender and inflammation.Our findings are critical and show that there is an urgent need not only to alleviate the symptoms of traumatic stress disorder but also to prevent further deterioration in the health of the refugees.If trauma leads to inflammation and subsequent development of early diabetes and obesity in young women, then early-stage therapeutic interventions for PTSD, anxiety and depression may moderate immune dysfunction through improving stress-related immune damage.

What this study adds
This study is the first to document association between anxiety/depression (or trauma-like), inflammation, obesity and T2D in young Syrian refugee women especially between 35 and 55 years.Anxiety/depression-induced obesity and inflammation pose a significant T2D risk in Syrian refugees in Lebanon.Although more research is needed to identify the pathways that may explain the link between anxiety/depression, inflammation and T2D, psychosocial interventions aimed at mitigating the effects of stressful life changes may aid in the reduction of obesity and the prevention of T2D at an early age.

Table 1
Demographic and familial history of metabolic diseases among Syrian refugees

Table 2
Clinical characteristics of the study population aged <35 was observed in women aged 35-55 years (74/102, 72.54%) followed by women aged <35 years (37/61, 60.65%).With HSCL-25, both men and women had elevated anxiety/depression scores, regardless of age and gender (Table2).However, women displayed remarkably higher symptomatic anxiety/depression scores (2.13 ± 0.50, P = 0.00623), and this effect was consistently observed in the three age groups.Remarkably, highest score was observed in the age group of 35-55 (2.26 ± 0.53).On the other hand, highest anxiety/depression score in men was observed in the age group of > 55 years (2.06 ± 0.77).Similar trend was observed with HTQ, where only women participants aged between 35 and 55 years had symptomatic anxiety/depression score higher than 2.06 (2.18 ± 0.43).Even though, women displayed higher symptomatic anxiety/depression score than men (2.04 ± 0.43 versus 1.88 ± 0.52, P = 0.0048), no symptomatic PTSD scores with HTQ were seen in men participants across all the age groups.
*Multiple 't'-test was performed between men and women groups.P < 0.05 was considered significant.score

Table 3
Clinical characteristics of the diabetes subgroups in the study population *Multiple 't'-test was performed between total men and women groups.P < 0.05 was considered significant.( * ) indicate statistical significance.