We’re not so different: Comparing American and immigrant trauma survivors - CONCEPT Professional Training

Sarah Filone, Drexel University
David DeMatteo, Drexel University

Recent immigration trends indicate that the United States is home to a remarkably diverse and rapidly growing population of displaced persons. Many of these individuals have survived exceptional trauma and are thus particularly vulnerable to trauma-related behavioral health disorders. Mental health professionals are commonly asked to assess immigrants within this population in the service of immigration court decision making. These assessments present a variety of challenges for clinicians, including the assessment and documentation of trauma-related symptoms across cultural bounds. The Trauma Symptom Inventory-2 (TSI-2) may be uniquely suited to the demands of immigration court assessments, but it has not been previously examined in a culturally diverse sample. The current study provided an examination of the TSI-2 within a sample of immigrants with histories of trauma. De-identified TSI-2 data were drawn from several clinicians’ existing immigration assessment files. Reliability and standardization sample comparison results indicated that the TSI-2 exhibits sufficient internal consistency within this population, and that immigrants with histories of trauma generally respond similarly to individuals in trauma-specific clinical samples (with several notable exceptions). Specific clinical implications are discussed.

TSI-2, immigration assessment, trauma assessment

“Each year, thousands of immigrants enter the United States seeking permanent residence. Numbers of refugees and asylum seekers from Central America have increased steeply in recent years, with “credible fear” applications (i.e., applications based on fear of persecution in one’s country of origin) increasing sevenfold from fewer than 5,000 applications in 2008 to more than 36,000 applications in 2013 (Chang & Linthicum, 2013). […] Many of these individuals have survived severe trauma related to experiences such as imprisonment, torture, interpersonal violence, war, famine, female genital mutilation, and other human rights violations (Dana, 2007). Accordingly, displaced persons are particularly vulnerable to trauma-related symptomatology including posttraumatic stress disorder (PTSD), depression, and anxiety.” (p. 701)
Immigrants’ rates of PTSD are much higher than those observed in American population. “Thus, immigrants with histories of trauma present a unique set of challenges for mental health professionals seeking to provide treatment and assessment services to this growing population.” (p. 701)
The Trauma Symptom Inventory-2 (TSI-2) is “a revised version of the original TSI (Briere, 1995), and includes 136 self-report items designed to assess a variety of trauma- and stress-related symptomatology (Briere, 2011). Respondents rate how often they experience each symptom on a scale of zero (“never”) to three (“all the time”). The measure yields scores for 12 clinical scales, 12 subscales, and 4 factors (composed of relevant scale and subscale scores). In addition, two validity scales assess whether a person is likely to underreport (response level or RL scale) or overreport (the atypical response or ATR scale) trauma-related symptoms. […] To date there have been very few published studies that provide real-world data related to the utility of the TSI-2 in culturally diverse or immigration court populations.” (p. 702).
“The present study examined a real-world sample of TSI-2 profiles that were administered during immigration court evaluations and provides preliminary data related to the test’s performance within an immigration court sample.” (p. 702).
“The immigrant sample was expected to exhibit elevations on many of the clinical scales and subscales when compared with the general (not trauma-specific) standardization sample. […] Analyses comparing the study sample to the three trauma-survivor groups within the TSI-2 CCV sample (combat veterans, domestic abuse survivors, and sexual abuse survivors; total N = 95) were largely exploratory. These analyses provided a crucial comparison of trauma-specific samples (immigration vs. American individuals with histories of trauma).” (p. 703).
“The TSI-2 sample included 97 immigrants with trauma histories (39 males; 58 females) who were assessed in relation to a variety of immigration relief proceedings. […] TSI-2 respondent ages ranged from 18 to 66 (M = 35.8; SD = 9.9). The study sample was culturally diverse, with 46 unique countries of origin represented.” (p. 704).
“The TSI-2 demonstrated excellent internal consistency on all but one subscale, suggesting that the scales and subscales are measuring cohesive underlying constructs within the study sample.” (p. 706).
“As expected, the immigration sample exhibited substantial elevations when compared with the standardization (non-trauma-specific) sample. […] The immigrant sample yielded significantly lower scores than the trauma-specific standardization sample on externalizing scales including the externalizing factor, the sexual disturbance clinical scale (including both subscales), the suicidality clinical scale (including both subscales), and the tension reducing behaviors clinical scale. These results raise the question of whether differences in externalization were related to cultural manifestations of symptomatology (i.e., perhaps individuals from nonwestern cultures are less likely to externalize), cultural norms related to reporting externalizing behaviors or to violence type (i.e., perhaps individuals from nonwestern cultures are less likely to report externalizing behaviors), or a general tendency to underreport symptoms (as suggested by RL elevations in the immigrant sample).” (p. 707)
“The immigrant sample also yielded significantly higher scores on the defensive avoidance clinical scale, raising the possibility that elevated levels of avoidance may contribute to reductions in instances of externalizing behaviors (or reductions in reports of externalizing behaviors). Interestingly, the hypothesis that the immigrant sample would yield significantly higher scores on the somatization scale and subscales than the trauma-specific standardization sample was not supported.” (p. 707).
“The immigrant sample exhibited ATR elevations when compared with the general standardization sample, but not when compared with the trauma-specific sample. […] The immigrant sample RL (underreporting) scale exhibited significant elevations when compared with the TSI-2’s general and trauma-specific standardization samples, indicating that the immigrant sample was more likely to deny common symptoms than all comparison groups.” (p.707-708).

“The present study provided an examination of the TSI-2 within a real-world sample of immigration court respondents. Further validation studies are needed to conclusively determine whether the available TSI-2 normative data are appropriate for use with immigrant and nonwestern populations, but the study findings suggest that TSI-2 profiles completed by trauma survivors in the context of immigration court proceedings yield results similar to those completed by American trauma survivors. In addition, the results of the present study can inform practitioners of immigration assessment. For example, consistent comparative elevations on the RL scales may suggest the presence of cross-cultural differences in the tendency to underreport or deny symptoms (particularly externalizing symptoms). Accordingly, clinicians conducting immigration court assessment should pay particular attention to the possibility of symptom denial, enhance efforts to increase evaluee comfort level, and normalize trauma-related symptomatology. While the RL elevations detected in the current study may inform clinical practice, they do not appear to raise concerns about cross-cultural utilization of the TSI-2 as the differences did not result in disproportionate invalidation of TSI-2 profiles.” (p. 708-709).

“The TSI-2 is uniquely suited to immigration evaluations for several reasons. First, it assesses for a broad range of symptomatology that may be associated with trauma (including somatization, a commonly observed trauma-response among immigrant and nonwestern samples, Rohlof, Knipscheer, & Kleber, 2014). Second, the TSI-2 contains validity scales to aid in the assessment of response style, which is often relevant in the immigration court context. The RL (underreporting) scale is helpful in the context of immigration assessment because it provides information related to an individual’s perception of his or her symptoms and his or her willingness to endorse or discuss symptomatology. The ATR (overreporting) scale is also highly relevant to immigration court assessments because mental health professionals are often asked to evaluate whether an individual is likely exaggerating his or her symptoms. Of course, an elevated ATR scale is not an indication that an individual is relaying erroneous factual data as part of his or her case, but it may help to identify a pattern of exaggerated responding or rare symptom endorsement. Additional advantages of the TSI-2 include efficient administration (roughly 20 min) and availability in both Spanish and English language forms. Although separate norms for the Spanish version of the TSI-2 are not available, the translation demonstrated satisfactory content validity and internal consistency in a Puerto Rican sample (Gutiérrez, Wang, Cosden, & Bernal, 2011).” (p. 702).
“Several limitations emphasize the importance of continued work in this area. One limitation of the current study was the method of convenience sampling from the available TSI-2 profiles of individuals who have completed immigration-related evaluations. This sampling method was used with consideration for population recruitment difficulties; however it limited the study’s ability to standardize administration procedures, collect data relevant to validity measurement, contact participants for repeated TSI-2 administration, and emphasize specific cultural groups. The nature of the convenience sample limits the ability to generalize the results of the study to immigration court participants broadly, given that the participating clinicians may have chosen to administer the TSI-2 only to specific subgroups within their evaluation pool (e.g., only administering to the most severe cases, or cases where feigning concerns might have been elevated). In addition, the sample comprised individuals from many diverse cultural backgrounds, and therefore advanced comparisons between cultural groups were not feasible. Future research should endeavor to compare specific cultural groups in order to assess for the presence of culturally influenced TSI-2 differences.” (p. 709).

As always, please join the discussion below if you have thoughts or comments to add!