Introduction

Posttraumatic Stress Disorder (PTSD) is a recognized psychiatric disorder affecting millions of people worldwide [1–7]. Evidence shows connections between a range of traumatic event exposures and PTSD, including exposure to inter-personal traumas (e.g., violence) and large-scale natural disasters [8, 9]. However, fundamental questions remain regarding the nature of traumatic events that can lead to PTSD.

A firmly established connection exists between direct exposure to traumatic events, such as natural disasters, life-threatening accidents, physical and sexual assaults, military combat, war exposure, life-threatening illnesses, and the traumatic loss of a loved one, and the susceptibility to PTSD [1–4, 8].

Other work has suggested that indirect traumatic events might also lead to greater risk of PTSD [10–14]. The concept of indirect trauma was initially introduced in the literature through the terms "secondary traumatization" and "vicarious traumatization," and defined as the experience of trauma indirectly affecting a family member, a close friend, a therapy client [10] or the exposure to trauma indirectly while performing professional duties, such as those carried out by emergency rescue crews [11]. In the 2013 5^th version of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5), a change was introduced in the definition of PTSD, incorporating indirect exposure to trauma into diagnosis criterion A (Exposure Criterion), defined as the "Indirect exposure to aversive details of the trauma, usually in the course of professional duties (e.g., first responders, medics)" [15, 16]. The DSM-5 excluded other potential forms of indirect trauma, such as observing traumatic events through electronic media, television, video games, or pictures, from its definition [17–19].

It remains unclear, however, the role these indirect traumatic events play in shaping PTSD. Some researchers have raised questions about the validity of including indirect exposure in the definition of PTSD [20, 21], while others have supported the concept [3].

A second, and related question pertains to the role of additive traumatic experiences to the risk of PTSD [22, 23], and whether cumulative exposure to direct and indirect traumatic events similarly increase the risk of PTSD [24–26]. Theory and research suggest that prior encounters with traumatic events can heighten sensitivity to adverse consequences when facing subsequent potentially traumatic events [27]. Substantial evidence emerges on the cumulative exposure to childhood traumas influencing the subsequent development of PTSD [28]. Research also suggests a dose-response relationship between the number of childhood traumas and comorbid psychiatric problems, including PTSD, in certain populations [29]. Cumulative exposure models to traumatic events suggest a dose-response relationship [30–32]. To our knowledge, there has been no previous work that had specifically examined the cumulative effects of indirect traumatic exposures on PTSD.

Our study seeks to investigate the impact of exposure to both direct and indirect traumas, as well as cumulative trauma, on PTSD, within the context of Houston, Texas where residents experienced the large-scale natural disaster of Hurricane Harvey, followed by the COVID-19 pandemic.

In August 2017, Hurricane Harvey, classified as a Category 4 storm on the Saffir-Simpson hurricane scale, hit Texas and Louisiana upon reaching land, establishing itself as one of the most severe tropical cyclones in the history of the United States [33]. The hurricane led to 68 direct fatalities and 35 deaths attributed to indirect causes, causing flooding in over 300,000 structures and affecting up to 500,000 cars. Additionally, approximately 336,000 households lost power, and 40,000 flood victims were either displaced or sought refuge in shelters, resulting in an estimated total of $125 billion in damages [34].

Three years later, the COVID-19 pandemic emerged as a protracted natural disaster [35]. Alongside more than 770 million confirmed cases and nearly 7 million confirmed deaths [36], The pandemic introduced a variety of social and economic challenges that are strongly linked to PTSD within the general population [37], among those directly exposed and hospitalized [38], as well as among healthcare workers experiencing indirect exposures [39].

These two disasters presented an opportunity to study the impact of exposures to direct and indirect traumas, as well as the cumulative impact of these traumas on PTSD among individuals in Houston who faced both Hurricane Harvey and the subsequent COVID-19 pandemic.

Materials and methods

This study examined the exposure to both direct and indirect trauma during Hurricane Harvey and the COVID-19 pandemic and explored their relationship with the probable diagnosis of PTSD.

Analysis

The data was analyzed in 2023. First, using descriptive statistics and chi-squared tests, we assessed whether demographics and trauma exposures are related to DSM-5 PTSD (see Table 1).

Download:

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

Table 1. Demographics (n = 1167)^*.

https://doi.org/10.1371/journal.pone.0307593.t001

Next, we estimated multivariable logistic regressions with interaction tests to predict the adjusted associations between each category of trauma and PTSD, as shown in Tables 2 and 3. These analyses were adjusted for age, gender, race, education, marital status, and income.

Download:

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

Table 2. Associations between types of direct and indirect trauma and PTSD^*.

https://doi.org/10.1371/journal.pone.0307593.t002

Download:

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

Table 3. Associations between cumulative direct and indirect trauma and PTSD^*.

https://doi.org/10.1371/journal.pone.0307593.t003

All analyses were performed using SAS 9.4 program, incorporating complex survey weights, with statistical significance set at p<0.05.

The study received ethical approval from the institutional review board at Boston University Medical Campus, and all research methods adhered to the relevant guidelines and regulations. Consent was waived by the ethics committee.

Results

As shown in Table 1, the majority of study participants identified as either male (48.1%) or female (50.5%), with a small minority (1.4%) selecting another gender category. More than half of the participants (52.1%) had attained some college education or higher, and approximately half (50.2%) were married. The sample consisted of 43.2% Hispanic individuals, 36.6% White non-Hispanic individuals, 14.2% Black non-Hispanic individuals, and 6% individuals from other non-Hispanic racial backgrounds. The average age of participants was 46.1 years. A substantial proportion of participants (61.2%) indicated a current income of less than $75,000. The interview mode was evenly split between paper (i.e., mail) (48.1%) and web (51.9%). 12.6% of participants reported experiencing current PTSD.

Descriptive analysis of the data, shown in S1 Table, reveals that a substantial portion of participants experienced a variety of exposures. For Hurricane Harvey, this included direct traumas, such as worrying about being seriously injured or killed (34.6%), and indirect trauma indirect trauma involving concerns about the serious injury or death of family members or close friends (54.8%). Furthermore, a significant number of participants were exposed to COVID-related events, including direct trauma like the fear of contracting COVID (48.5%), and indirect trauma involving concerns that family members or close friends might die from COVID (64.6%), or the experience of having seriously ill family members or friends (39.6%) or losing family members and friends due to COVID (19.4%).

We noted an increased incidence of PTSD among individuals who experienced specific Hurricane Harvey exposures, including worry about serious injury or death during Harvey (direct trauma, 22.1%), and the death of family members or close friends (indirect trauma, 32.5%). Further, we noted a higher prevalence of PTSD among those who experienced certain COVID-related traumas, including hospitalization (direct trauma, 80.4%), and the death of family members or friends due to the pandemic (indirect trauma, 17.5%).

We found a greater likelihood of PTSD when individuals were exposed to a higher number of direct or indirect traumas, in accordance with the DSM-5 criteria. Prevalence and median splits are presented in S1 Table.

Table 2 shows the results of multiple logistic regression analyses examining the relationship between direct and indirect trauma exposure for each event and PTSD prevalence. These adjusted results reveal a statistically significant relationship between direct and indirect traumas related to Hurricane Harvey and COVID-19, and the presence of PTSD. These relationships were statistically significant in all categories except for cumulative COVID indirect trauma. Associations with PTSD were demonstrated for Harvey’s direct traumas (OR = 3.18, 95% CI 1.85, 5.46), and COVID direct traumas (OR = 2.13, 95% CI 1.20, 3.77), and to a lesser extent for Harvey’s indirect trauma (OR = 1.91, 95% CI 1.05, 3.46), as presented in Table 2. The point estimates are higher for direct compared to indirect trauma, although the confidence intervals overlap. The full models are presented in S2 Table. Estimates from the unadjusted models are presented in S3 Table.

Table 3 shows the OR for PTSD for cumulative exposure to the indirect traumas of Harvey and COVID (OR = 2.79, 95% CI 1.47, 5.28) and cumulative exposure to the direct traumas of these events (OR = 2.53, 95% CI 1.36, 4.70). The full models are presented in S4 Table. Data from the unadjusted model can be found in S5 Table.

Discussion

Our study demonstrates associations between exposure to both direct and indirect traumas resulting from large scale potentially traumatic events and the development of PTSD. Further, our data establishes associations between cumulative exposures to different large scale potentially traumatic events and the occurrence of PTSD.

In our study, approximately one out of eight participants (12.6%) fulfilled the criteria for current PTSD, three years after Hurricane Harvey and during the COVID pandemic. These data align with other studies examining PTSD among persons exposed to traumatic events using DSM criteria [4]. Additionally, our findings are consistent with other studies investigating the long-term consequences of natural disasters that occurred years earlier. For example, Raker et al. reported a 16.7% rate of post-traumatic symptoms in communities exposed to Hurricane Katrina 12 years after the event [46]. In a longitudinal study conducted by Goenjian et al., a PTSD rate of 11.6%, as per DSM-5 criteria, was identified among survivors of the Spitak earthquake in Armenia, 23 years after the incident [47]. Horesh et al. revealed a 16.5% rate of PTSD among veterans from Israel 20 years after experiencing the Lebanon war [48].

We showed an association between exposure to both direct and indirect traumas and PTSD. This finding supports the notion, consistent with other studies, that the inclusion of exposure to indirect trauma in the DSM-5 criteria was warranted. May and Wisco’s review suggests that indirect exposure to trauma among close associates, aligning with the definition used in our study, may lead to PTSD, albeit with a lower likelihood compared to direct exposure [49]. Solomon and Zerach’s analysis of indirect exposures to trauma among family members of former Israeli prisoners of war who were exposed to war trauma and subsequently developed PTSD, provides evidence of the development of secondary traumatization among spouses and offspring [50]. Common mechanisms for both direct and indirect trauma are also evident in the study by Ten Holt et al., which found associations between both direct and indirect exposure to trauma and PTSD among patients with substance use disorder [12].

Our observation that persons directly affected by traumatic events consistently exhibited a higher prevalence of PTSD compared to those indirectly affected aligns with current literature, underscoring the intensity of event exposure as a significant risk factor for PTSD. Galea et al. demonstrated the impact of indirect exposure to the September 11 terror attacks on New York residents in relation to the development of PTSD, albeit to a lesser extent compared to individuals directly affected by the attacks [13]. In contrast, Kar et al. found no significant difference in PTSD prevalence among people directly or indirectly exposed to the 2004 Asian Tsunami disaster. However, their study defined indirect exposure as being in close proximity to the sea and witnessing all the devastation and trauma around, without specifying indirect exposure through affected loved ones [14].

Findings regarding the influence of indirect exposure to trauma on the development of PTSD are not conclusive in the literature. Merdjanhoff et al. found that indirect exposure to Hurricane Katrina at the community level was notably linked to psychological distress but did not show a significant association with probable PTSD [51]. However, this study did not define indirect exposure in alignment with the DSM PTSD criteria. Instead, it included stressors in the indirect exposure category, such as economic loss and lost income. Further, some studies have identified distinct features associated with direct and indirect exposure to traumas. For example, Christou-Ergos et al. demonstrated in Australia that individuals who directly experienced significant human suffering were less inclined to accept the COVID vaccine during the pandemic, whereas those with an indirect exposure to severe suffering showed a higher willingness to receive the vaccine [52].

Our study demonstrates a cumulative effect of exposure to multiple traumas on the development of PTSD. These findings are consistent with other studies that have explored the impact of exposure to multiple traumatic events. Schock et al. observed such a cumulative effect among refugees in Germany who had been exposed to various traumatic events [23]. Lowe et al. identified a cumulative effect of exposure to two disasters in the United States, namely Hurricane Katrina and the Deepwater Horizon oil spill [53]. Harville et al. showed that exposure to multiple hurricanes among United States Gulf Coast residents increased the risk of mental health issues, including PTSD [54]. Yehuda et al. found that Holocaust survivors with PTSD had experienced significantly more traumas in life compared to survivors without PTSD [55]. These findings are also in line with other studies that have examined the combined effects of COVID-19 and natural disasters, such as the work of Agyapong et al., which revealed a higher likelihood of PTSD and other mental health disorders among people in Canada who had experienced both COVID-19 and traumas like floods or wildfires [56].

We demonstrated a cumulative effect of indirect traumas on the risk of PTSD. Such effect is supported by other studies that have investigated the cumulative mechanisms of indirect trauma exposures. Fernandez et al. discovered that patients exposed to various disaster-related experiences in the New York area in 2001, including the World Trade Center disaster and the Flight 587 crash two months later, had poorer health status compared to those exposed to only one disaster. This also involved indirect experiences related to the deaths of friends and family in these events [57]. Garfin et al. found that prior indirect exposure to three community traumas in the northern United States through the media increased the likelihood of citizens in the Boston and New York areas reacting more negatively to the Boston Marathon Bombing, which was a subsequent community trauma [58].

Conclusions

Our study provides compelling evidence supporting the links between exposure to both direct and indirect traumas resulting from disasters and the onset of PTSD. Additionally, it firmly echoes evidence supporting relationships between cumulative exposures to various disasters and the emergence of PTSD [27–32].

The findings concerning the influence of cumulative exposure to indirect traumas on PTSD underscore the vital role of indirect experiences in the development of PTSD. This highlights the importance of screening individuals who have been indirectly exposed to traumas involving family members or loved ones, and further emphasizes the necessity of creating intervention programs customized for these individuals.

Supporting information