Syrian crisis in Jordan

The Syrian crisis has been declared the worst refugee crisis since World War II, displacing over 6 million Syrians to neighboring countries and Europe [15]. According to a 2016 United Nations High Commissioner for Human Rights report, over 5.6 million registered Syrian refugees had fled from their homes and are living in surrounding countries such as Turkey, Lebanon, and Jordan; the number of refugees has grown additional 12.7% from the previous year [16].

Jordan is among the countries with the largest influx of displaced Syrians. As of August, 2020 there were, 658,756 Syrians refugees registered through the United Nations High Commissioner for Refugees (UNHCR), and the Jordanian government estimates approximately 1.4 Syrians remain unregistered in the country [8, 17, 18]. While many of the Syrians in Jordan have experienced the complex trauma of the war, Jordanians have been adversely impacted by the influx of hundreds of thousands of refugees [10, 19]. The inundation of refugees has put a strain on healthcare and social services, and has increased job competition, thereby increasing the potential for civil conflict and societal discord [10, 19, 20]. Given the fragility of Jordan’s overloaded system, many low-income Jordanians and displaced Syrians are without healthcare, jobs, and social services [19]. Moreover, systems in Jordan have been unable to provide adequate healthcare to individuals with chronic non-communicable diseases (NCDs) (e.g., hypertension, cardiovascular disease, diabetes, chronic respiratory disease) or mental health conditions (e.g., depression, post-traumatic stress, anxiety) [19, 21, 22].

Jordan has also experienced an epidemiological shift from the decreasing incidence of infectious diseases and fertility-related mortality to a rapidly increasing prevalence of chronic NCDs, such as diabetes, obesity, asthma, cardiovascular disease (CVD), and cancer [20, 23]. This transition is due in part to medical advances, which have resulted in reduced mortality and increased ability to prevent fertility-related mortality [24]. These advances have led to rapid population growth, which coupled with globalization, modernization, the refugee crisis, and urbanization, has initiated population level sociocultural changes [23].

In recent years, the global burden of NCDs has gained attention, particularly in low and middle-income countries. It is estimated that 80% of deaths globally are connected to NCDs, which are of particular concern to individuals from middle income countries, such as Syria and Jordan [25]. In Jordan, ischemic heart disease is the leading cause of death (18%), followed by cancer (15%), strokes (12%), and diabetes (7%) [26]. A study of NCDs among Syrians in Jordan found that participants reported at least one family member with an NCD, including hypertension (9.7%), arthritis (6.8%), diabetes (5.3%), chronic respiratory diseases (3.1%), and cardiovascular disease (3.7%) [21].

The global burden of mental health-related disorders has also emerged on the international stage, particularly in LMICs. Common mental health disorders such as anxiety and depression significantly contribute to the overall burden of disease in many LMICs [9]. The National Behavioral Risk Factor Surveillance Survey in Jordan found that 18.4 percent of low-income Jordanians reported frequent mental health distress [27]. Studies have suggested that, among displaced Syrians, clinical levels of post-traumatic stress are as high as 71% [28, 29]. Mental health difficulties in LMICs have been attributed to a myriad of factors, such as food and housing instability, forced displacement, exposure to potentially traumatic events, and the depletion of resources [9, 30].

Chronic diseases and mental health disorders often co-occur. The prevalence of depression in the global population ranges from 3–10%, while up to 33% of individuals with chronic diseases experience co-morbid depression [31, 32]. Al-Amer and colleagues [5] found 19.7 percent of Jordanians diagnosed with diabetes reported co-occurring depression. Studies of Syrian refugees in Jordan have also found that over half of those with chronic diseases also reported high levels of PTSD, and 35% of Syrians previously diagnosed with a chronic illness experienced clinical rates of depression [12, 33]. The co-morbidity of mental disorders and physical illness accounts for a 37% loss of healthy life years; smoking, alcohol consumption, lack of physical activity, unhealthy diet, and decreased access to healthcare have all been linked to this loss [34]. Co-morbid depression and diabetes, for example, has a slew of negative effects, such as decreases in self-care and self-management, a reduction in adherence to medication, and increases in medical expenditures [35–37].

The primary causes of mortality among individuals with mental health difficulties include preventable conditions such as cardiac illnesses, respiratory and infectious diseases, diabetes, and hypertension [38]. Furthermore, displaced populations often are unable to adequately treat chronic physical and mental health conditions due to a lack of medications and appropriate health care [1, 33]. Al-Rousan and colleagues [6] found that cost of living, lack of legal work opportunities, and inability to access healthcare were associated with psychological distress and health problems among displaced Syrians in Jordan. Given these startling rates of NCDs and mental health-related issues, scholars have called for continued research to explore those relationships between those social, biological, personal, and economic factors which impact physical and mental health in LMICs [9].

Social, physical, and mental health

Social health, which is defined as the combination of social networks and social support [39], involves the intersection of a feeling of belonging or acceptance and fulfillment in social roles, which in-turn can boost health and emotional well-being [14, 40, 41]. A growing body of evidence from the past several decades has illustrated causal relationships between role identity (e.g. social ties, self-esteem, social health), longevity, physical, and mental health [3, 42, 43]. A meta-analysis of 148 studies found a significant association between social health and reduced risk for overall mortality due to chronic diseases [44].

Scholars have suggested that social support and networks are comprised of two processes which can influence health and mental health outcomes. The first process involves the provision of informational, emotional, or instrumental resources to an individual to meet his or her physical and emotional needs. The second process, which we will examine in this paper, are the health benefits associated with participation and inclusion in a social group. Social interactions influence emotional, biological, or behavioral responses in an individual. In turn, these social interactions influence a person’s sense of personal control, self-concept, and self-worth [3, 40]. Social health may thereby reduce stress during adversity, which can protect individuals from physical health-related difficulties. Likewise, the social factors attributed to physical health outcomes are markedly similar to those that are associated with positive mental health outcomes and therefore should mediate the relationship between physical and mental health [3].

Research has examined the relationship between social, physical, and mental health, however, few scholars have examined the pathways which may influence these correlations [3]. The following study, therefore, sought to explore mediating and moderating role of social health on physical health and post-traumatic stress symptoms (PTSS) within a Jordanian community near the Syrian border. Our primary aim was to examine the mediating and moderating role of social health on the relationship between physical health and PTSS.

Measures

Statistical analyses

Data were cleaned and descriptive statistics calculated to assess demographic characteristics. A small percentage (1–3%) of the data were missing for our variables of interest (PTSS, social health, physical health), therefore, we excluded the missing data from the analyses. One-way ANOVA and independent samples t-tests examined mean scores of social health, PTSS, and physical health stratified by trauma exposure, age, gender, nationality, and education level. Bivariate correlations examined the relationship between PTSS, social health, and physical health.

To test the moderation we tested the relationships for: (i) the direct effect of the predictor (physical health) on PTSS, (ii) the direct effect of the moderator (social health) on PTSS, and (iii) the interaction effect of the moderator (social health) on physical heath and PTSS [51]. In order to further characterize the moderating effect that social health had on the relationship between physical health and PTSS we used the Johnson–Neymann (J–N) technique. The J-N technique was used to calculate regions of significance and confidence bands to assess the effect of the independent variable across all levels of the moderator variable [49]. The J-N technique identifies points in the range (0–100) of the moderator variable (social health) where the impact of the independent variable (physical health) on the dependent variable (PTSS) is significant and non-significant [52].

The mediation effects were tested in four steps examining the (i) direct effect of social health (mediator) on PTSS (ii) direct effect of physical health (predictor) on social health (mediator), (iii) total effect of physical health (predictor) on PTSS and (iv) significance of the indirect effect using bias-corrected bootstrapping with 95% confidence intervals (CI) [53]. Mediation is proven when the analysis found significant a- and b-paths, as well as a significant indirect effect with 95% confidence intervals that did not include zero [53]. We adjusted for the effect of sociodemographic correlates (age, gender, nationality, trauma exposure, and education level) in the moderation and mediation models. Income was not included as a control variable because most (91.2%) of the sample came from lower income households. All analyses were conducted in SPSS 27.0.

The moderation and mediational role of social health on the relationship between PTSS and physical health

Moderation analyses illustrated that higher physical health (B = -0.14, SE = 0.05, p<0.001) and social health (B = -0.13, SE = 0.03, p<0.001) were significantly associated with lower PTSS scores. The positive coefficient of the interaction term of physical health and social health indicated that social health was a significant moderator on the relationship between self-reported physical health and PTSS (B = 0.001, SE = 0.001, p<0.5). Table 5 presents outcomes of moderation analysis; Fig 1 visualizes the moderation effect).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Relationship between physical health and post-traumatic stress symptoms at low and high levels of social health.

https://doi.org/10.1371/journal.pone.0241036.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 5. Moderation analysis testing social health as a moderator between the relationships of physical health and post-traumatic stress symptoms.

https://doi.org/10.1371/journal.pone.0241036.t005

Results from the J-N technique indicated the conditional effect of physical health on PTSD transitioned in significance when the summed score of social health was 61; b = -0.04, SE = 0.02, t = -1.97, p = .05, 95% CIs [-0.07, .00]. The relation between physical health and PTSS scores were significant when social health scores were below a score of 61 and non-significant with higher values of social health (61–100).

Step one of the mediation analysis demonstrated a negative and significant direct effect of social health on PTSS, b = -0.07, s.e. = .015, p = < .001, indicating those with lower social health scored higher on PTSS. Step 2 found that the direct effect of physical health on social health was positive and significant, b = .10, s.e. = .03, p < .01, indicating higher physical health scores were associated with higher social health. Step three indicated the total effect of physical health on PTSS was negative and significant, b = -0.07, s.e. = .011, p = < .001, demonstrating those with higher self-reported physical health scored lower on PTSS. Step 4 demonstrated the indirect effect of physical health on PTSS via social health was statistically significant [β = -.0083, 95% C.I. (-.0147, -.0016)]. Results therefore indicated a negative predictive relationship between physical health and PTSS as mediated through social health. See Fig 2 for a visualization of the mediated model.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Mediation model of social health in relationship between physical health and post-traumatic stress symptoms.

Values presented are standardized regression coefficients. The values in parentheses represents the coefficient for direct (without mediator) path.

https://doi.org/10.1371/journal.pone.0241036.g002

Implications for policy and practice

Results from this study have implications for interventions designed to prevent mental and physical health related difficulties. Mental health support and NCD prevention in primary care is a growing field given the high rates of co-morbidity. The body of research in this area, specifically in LMICs and humanitarian settings, has significantly grown in the last decade, yet more research is needed to understand appropriate pathways to reduce mental health difficulties and manage or prevent physical health ailments. This is particularly relevant in LMICs with a large influx of displaced people, where much of the population is below the poverty level and without access to physical or mental health care [75]. The World Health Organization has recognized the global burden of NCDs and mental health and advocated for both in the international health agenda [76–78]. Therefore, our findings may inform NCD and mental health interventions in LMICs and humanitarian populations [79]. As Thoits [3] noted, it is crucial to understand intervening mechanisms such as social health on physical or mental health outcomes to design effective interventions.

Considering the high rate NCDs and mental health symptoms in this region, intervention approaches at the community level could be designed to increase individuals physical, social, and emotional well-being [30]. Such approaches may include integration of psychoeducation group work into the primary care setting informing individuals on common reactions to stress and trauma, the connection between post-traumatic stress and physical health, or on stress management. While this type of information does not require clinical mental health services, it may be a strategic way to reach those at risk for mental health-related symptoms. Activities which increase social interaction and inclusion could also be beneficial in health clinics and community centers. For example, activities for greater social contact could include religious events, participating in sporting events, or having community-based health and mental health education [80, 81].

Limitations

While this study provides novel findings on the moderating role of social health on physical health and PTSS, results should be considered in light of several limitations. First, our study is not generalizable to the wider population. The study was conducted in a region where many people (both Jordanian and Syrian) had high levels of PTSS. In the general global population, clinical rates of PTSS are much lower (approximately 3%) [82], therefore, additional studies in lower-risk populations would generalize the findings. Additionally, we used a self-reported measure of physical health. A more objective health measure, such as blood pressure, body mass index, or fasting blood glucose could further examine whether social health may influence prevention or management of specific NCDs such as hypertension or diabetes. It is also important to note that the reliability of the social health scale was acceptable (.65), however, future studies may want to employ a measure of social health designed specifically for an Arab population, which may yield higher reliability.

We also were unable to identify selection bias because we do not have information on how many people were approached by their doctor to participate in the study, who opted into the study from viewing the fliers that were posted in the health centers, and who fit the inclusion criteria, but opted out of the study. Finally, the data used for this study was cross-sectional, therefore longitudinal causal relationships cannot be determined.