Our cross-sectional retrospective study replicates and extends the findings of previous reports(4
) examining the prevalence and consequences of trauma exposure in urban, predominantly African-American and impoverished, populations. In this study, we found that African-American and subjects of other ethnicities with low socioeconomic status, ascertained while receiving primary care at an urban, public hospital, were at high risk of exposure to traumatic events and the development of stress-related psychiatric illness.
A number of studies have examined the prevalence of trauma exposure and risk for PTSD in both civilian primary care samples(4
) and veteran primary care samples(58
). Reported rates of current PTSD range from 2%(55
) to approximately 24%(4
) depending on sample socioeconomic characteristics and methodological variation in the determination of trauma exposure and psychiatric morbidity. Alim and colleagues(4
) conducted a study with a predominantly African-American primary care sample, similar to our own, and found a lifetime trauma exposure rate of 65% whereas 87.8% of our sample experienced some form of traumatic event. In our study, these events consisted of serious accidents and various forms of interpersonal violence. A greater percentage of male than female subjects reported exposure to non-interpersonal trauma types (natural disaster, serious accidents, life-threatening illness) as well as most forms of non-sexual interpersonal violence (attacks with and without weapons by non-intimate partner, attacks with weapon by intimate partner) as observed by other investigators(4
). Consistent with previous reports(4
), female subjects reported much higher rates of sexual assault during childhood and adulthood as well as attacks without a weapon by an intimate partner. Interestingly, the only category of trauma exposure that we were unable to identify a significant gender difference with respect to trauma exposure was that of physical abuse - being beaten as a child. It is not clear whether these differences represent gender-related differences with respect to actual trauma exposure, gender-related differences in reporting or recall bias, or an interaction of these two factors. These data are similar with respect to rates of trauma exposure (83%) to those we have previously published(39
) describing trauma exposure in a predominantly African-American sample of convenience recruited from an inner-city mental health clinic.
Compared to national averages, the lifetime prevalence of PTSD was substantially elevated in our sample (46.2%). In the National Comorbidity Survey, Kessler and colleagues (1
) reported that the lifetime prevalence of PTSD was 7.8% with female subjects being twice (10.4%) as likely as male subjects (5.0%) to develop PTSD. Similar to our findings, Alim and colleagues (4
) in their sample of urban African-American primary care patients, also found a high lifetime prevalence of PTSD (51%) with female subjects (60%) being nearly twice as likely as male subjects (33%) to develop PTSD consistent with gender differences in rates of PTSD previously reported by other investigators(62
). The lifetime prevalence of MDD was also very high (36.7%) in our sample and elevated relative to data from the National Comorbidity Survey(69
). As with PTSD, the lifetime prevalence of MDD within our sample was also comparable to the overall lifetime rate of MDD (35%) observed by Alim and colleagues(4
). Data from the National Comorbidity Survey indicate that PTSD and MDD are frequently comorbid conditions in trauma exposed individuals(1
In addition to our examination of the current and lifetime prevalence of PTSD and MDD, we also examined the effects of the type, extent, and developmental timing of trauma exposure on post-traumatic stress and depressive symptoms. Consistent with previous reports(70
), we found that trauma of an interpersonal nature such as sexual assault and physical child abuse followed by non-sexual assault had the largest impact on both post-traumatic stress and depressive symptoms. An important consideration in the interpretation of these data is the over-representation of female subjects in the sexual assault categories and over-representation of male subjects in the non-sexual assault categories. The increased vulnerability of women to the development of PTSD and MDD may be responsible for the large effects of sexual assault on post-traumatic stress and depressive symptoms. Conversely, the decreased vulnerability of men to develop PTSD and MDD (relative to women) may reduce the effects of non-sexual assault on post-traumatic stress and depressive symptoms. Of note, we did not identify a gender difference in the exposure rate of child physical abuse which had effects on post-traumatic stress and depressive symptoms comparable to those of sexual assault. Exposure to multiple traumatic events across the lifespan is relatively common, particularly in certain populations at high risk of trauma exposure(1
) and is associated with elevated risk for negative psychiatric outcomes(6
). Consistent with this, we found that exposure to either childhood trauma, adult trauma, or both produced graded effects on post-traumatic stress as well as depressive symptoms as described in previously reported data derived from patients with PTSD(39
) and MDD(78
There are several limitations to this study. We utilized a cross-sectional approach to retrospectively assess the prevalence of trauma exposure and psychopathology in subjects recruited from the primary care clinics of an urban county hospital. Although the demographic characteristics of our sample limit the generalizability of our findings to the general population as well as many primary care and African-American samples, we believe that our report highlights the extensive trauma exposure experienced by urban African Americans with limited financial means and poor access to health care. The use of retrospective data is also a significant limitation of our study. Retrospective bias in the recall of putative etiological events has been studied the most with subjects who have major depression(78
) and our use of retrospective data to assess trauma may have inflated our estimate of trauma exposure in subjects with PTSD as reports of trauma exposure are correlated with PTSD symptoms(86
). In general support of this, a recently reported prospective study of the effects of prior trauma exposure on risk for PTSD suggests that risk for PTSD is only increased by exposure to subsequent trauma in individuals who had previously developed PTSD in response to prior trauma(88
). Unfortunately, this report did not consider exposure to childhood trauma exposure in its analysis, which may limit interpretation of the effects of childhood trauma on adult risk for PTSD and MDD. Other additional limitations of this study include the possible induction of social desirability response bias caused by reading of items on screening instruments to subjects by study interviewers rather than having subjects submit written answers and the use of the BDI to assess depressive symptoms in our primary care-derived sample which may overestimate depressive symptoms due to overlap of medical and depressive symptoms on the BDI.
These data document the extraordinarily high levels of childhood and adult trauma exposure, principally in the form of interpersonal violence, in a large, sample of an urban, primary care population. Within this group of subjects, PTSD and depression are highly prevalent conditions. Civilian trauma-related disorders, especially among impoverished, urban populations, carry enormous societal burdens. Furthermore, these findings suggest that intergenerational cycles of violence and trauma may be endemic in America's urban environments.