It is broadly assumed that there is a strong causal association between exposure to childhood abuse and adult posttraumatic stress disorder (PTSD). It has been difficult to substantiate this causal link with the prospective longitudinal data that would arguably be necessary to confirm or disconfirm this relationship. However, the assumption remains viable based mostly on cross-sectional and/or retrospective studies of persons with psychopathology, and in particular PTSD, who report such early life trauma. This paper will review that assumption by exploring an alternative trajectory: that childhood abuse sets the stage for adult PTSD through a less direct route. For example, statistical modeling of such persons on whom extensive lifetime traumatic histories have been obtained has additionally identified revictimization as a more direct consequence of early exposure, and perhaps a more proximal cause of adult psychopathology such as PTSD (Nishith, Mechanic, & Resick, Reference Nishith, Mechanic and Resick2000).

In addition, we attempt to link recent endocrinological observations in survivors of early trauma with those in PTSD that may provide better models for understanding the association between early life events and subsequent and enduring biobehavioral consequences. It is critical to update this clinical literature with these biological models because previous models of the biology of stress have fallen short of being able to explain stress responses that are either enduring or that do not eventually abate in the absence of continuous provocation. Biological models of hypothalamic–pituitary–adrenal (HPA) axis dysfunction offer a paradigm for how early life events recalibrate stress response systems to facilitate altered responses to subsequent exposures. They may even explain some of the behavioral changes that increase risk of revictimization.

Childhood Maltreatment and PTSD

It is often assumed that childhood abuse leads to PTSD. However, most observations regarding the relationship between early trauma and adult PTSD have been correlative, retrospective, and have relied on examination of clinical populations. Despite the limitations associated with such studies (such as the inability to determine causal relationship, and both recall and sample bias), these observations have provided some support for the view that early maltreatment is related to adult PTSD: higher rates of childhood abuse and maltreatment have been found among individuals with PTSD than among those without (Bremner, Southwick, Johnson, Yehuda, & Charney, Reference Bremner, Southwick, Johnson, Yehuda and Charney1993; Cloitre, Scarvalone, & Difede, Reference Cloitre, Scarvalone and Difede1997; Follette, Polusny, & Milbeck, Reference Follette, Polusny and Milbeck1994; Lang et al., Reference Lang, Aarons, Gearity, Laffaye, Satz, Dresselhaus and Stein2008; Rodriguez, Ryan, Van de Kemp, & Foy, Reference Rodriguez, Ryan, Vande Kemp and Foy1997; Zlotnick, Reference Zlotnick1997).

The first indication that this link may not be direct and causal is that, following childhood sexual abuse, the prevalence of PTSD in children is lower than the prevalence of PTSD in adults. Although the prevalence of PTSD after sexual or physical abuse in children has been reported to be from 34% to 65% (Ackerman, Newton, McPherson, Jones, & Dykman, Reference Ackerman, Newton, McPherson, Jones and Dykman1998, McLeer et al., Reference McLeer, Dixon, Henry, Ruggiero, Escovitz and Niedda1998, Saigh & Bremner, Reference Saigh, Bremner, Saigh and Bremner1999), most studies of adults who report abuse in childhood show a much higher range of PTSD, ranging from 72% to 100% (Lindberg & Distad, Reference Lindberg and Distad1985; Rodriguez et al., Reference Rodriguez, Ryan, Vande Kemp and Foy1997). Methodologic differences may account for some of the discrepancy, but it is interesting to speculate about other explanations. The higher prevalence of PTSD in adults exposed to childhood abuse may suggest that the full impact of early childhood abuse is not realized until adulthood, particularly if no steps are taken to help the child overcome these experiences when he or she is young. Beitchman et al. (Reference Beitchman, Zucker, Hood, daCosta, Akman and Cassavia1992) referred to “sleeper” effects of abuse that require developmental maturity for their expression. Such effects are not evident as short-term consequences of a trauma but may emerge subsequently. An example of such an effect is sexual dysfunction, which has clearly been associated with childhood sexual abuse but may not be evident until adulthood. Similarly, Toth and Cicchetti (Reference Toth, Cicchetti, Russ and Ollendick1999) highlight the role of a dynamic model in which a child's response to the trauma influences the subsequent resolution of developmental tasks such that vulnerability to later pathology is increased. Another explanation is that the appearance of some psychological responses to early trauma may require more developed cognition and maturity. As one's perspective and cognitive capacities develop with age, perceptions and responses to an earlier experience can change as well. Thus, it is possible for adults to have delayed reactions to earlier events as they become more fully aware of the magnitude of betrayal implied by an experience. Certainly, it is not until adulthood that many victims of childhood abuse acknowledge that they actually did experience abuse. Furthermore, subsequent life circumstances can either confirm or negate cognitions associated with childhood trauma. That is, to the extent that children can be removed from a traumatic situation and placed in an environment in which they can thrive, they may be able to alter their perception about their personal safety. Alternatively, should early traumatic events be followed by subsequent traumas, both emotional and biological effects may be consolidated. These explanations suggest that perhaps the traumatic nature of abuse in childhood does not directly lead to the PTSD reported in adulthood; rather that it sets the stage for subsequent experiences that produce PTSD.

Maltreatment Versus Single Trauma: Differing Effects

Because PTSD is recognized as a consequence of single-episode traumas (DSM-IV-TR; American Psychiatric Association, 2000), it is important to examine the long-term impact of such events in childhood as well, such as random shootings in schools or other public places; terrorism, accidents, and natural disasters; single episodes of assault; and life-threatening illnesses. A greater understanding of the sequelae of these types of trauma will be critical in sorting out the consequences related to the nature of exposure to trauma from those related to other risk factors associated with the chronic traumatization that is inherent in abuse. The literature examining effects of childhood single-episode trauma show a different trajectory than that described for abuse and could be interpreted as suggesting that the immediate consequences of single-episode trauma are more severe than the immediate consequences of childhood abuse. For example, studies examining children immediately after natural disasters have found PTSD to occur in up to 87% of the children (Garrison et al., Reference Garrison, Bryant, Addy, Spurrier, Freedy and Kilpatrick1995; John, Russell, & Russell, Reference John, Russell and Russell2007; McDermott, Lee, Judd, & Gibbon, Reference McDermott, Lee, Judd and Gibbon2005; Scheeringa & Zeanah, Reference Scheeringa and Zeanah2008; Shaw et al., Reference Shaw, Applegate, Tanner, Perez, Rothe and Campo-Bowen1995). Generally speaking, rates of PTSD following extreme violence are quite high; for example, the rates of PTSD after being kidnapped, witnessing the murder of a parent, or experiencing violence have ranged from 95% to 100% (Horowitz, Weine, & Jekel, Reference Horowitz, Weine and Jekel1995; Schwarz & Kowalski, Reference Schwarz and Kowalski1991).

However, when considering these higher rates of PTSD for single-episode traumas, it is important to consider how the literature on abuse versus single-episode traumas may be impacted by difficulties in assessing the effects of abuse. Such comparisons do not account for the myriad other problems that can develop as a result of abuse. Childhood abuse has been associated with higher rates of anxiety (Mancini, Van Ameringen, & MacMillan, Reference Mancini, Van Ameringen and MacMillan1995; Molnar, Buka, & Kessler, Reference Molnar, Buka and Kessler2001), depression (Nelson et al., Reference Nelson, Heath, Madden, Cooper, Dinwiddie and Bucholz2002; van der Vegt et al., Reference van der Vegt, Tieman, van der Ende, Ferdinand, Verhulst and Tiemeier2009), suicidality (Cloitre et al., Reference Cloitre, Scarvalone and Difede1997; Klein et al., Reference Klein, Arnow, Barkin, Dowling, Kocsis and Leon2009), dissociation (Dancu, Riggs, Hearst-Ikeda, Shoyer, & Foa, Reference Dancu, Riggs, Hearst-Ikeda, Shoyer and Foa1996; Dutra, Bureau, Holmes, Lyubchik, & Lyons-Ruth, Reference Dutra, Bureau, Holmes, Lyubchik and Lyons-Ruth2009; Zlotnick, Reference Zlotnick1997), personality disorders (Johnson, Smailes, Cohen, Brown, & Bernstein, Reference Johnson, Smailes, Cohen, Brown and Bernstein2000; Luntz & Widom, Reference Luntz and Widom1994), substance abuse (Umeno et al., Reference Umeno, Morita, Ikeda, Koda, Abe and Endo2009; van der Vegt et al., Reference van der Vegt, Tieman, van der Ende, Ferdinand, Verhulst and Tiemeier2009), physical illness (Afifi et al., Reference Afifi, Enns, Cox, de Graaf, ten Have and Sareen2007; Greenfield & Marks, Reference Greenfield and Marks2009), and interpersonal problems (Callahan, Price, & Hilsenroth, Reference Callahan, Price and Hilsenroth2003; Liang, Williams, & Siegel, Reference Liang, Williams and Siegel2006).

The immediate developmental consequences of abuse and neglect can also be significant. Maltreatment has been found to detrimentally affect the attainment of important developmental goals: the processes necessary for emotion regulation (Fries & Pollak, Reference Fries and Pollak2004; Gaensbauer, Reference Gaensbauer1982; Gibb, Schofield, & Coles, Reference Gibb, Schofield and Coles2009), secure attachments (Barnett, Ganiban, & Cicchetti, Reference Barnett, Ganiban and Cicchetti1999; Egeland & Sroufe, Reference Egeland and Sroufe1981; van IJzendoorn, Schuengel, & Bakermans-Kranenburg, Reference van IJzendoorn, Schuengel and Bakermans-Kranenburg1999; Venet, Bureau, Gosselin, & Capuano, Reference Venet, Bureau, Gosselin and Capuano2007), positive self-concept (Alessandri & Lewis, Reference Alessandri and Lewis1996; Egeland, Sroufe, & Erickson, Reference Egeland, Sroufe and Erickson1983; Kim & Cicchetti, Reference Kim and Cicchetti2003), language abilities (Culp et al., Reference Culp, Watkins, Lawrence, Letts, Kelly and Rice1991; Noll et al., Reference Noll, Shenk, Yeh, Putnam and Trickett2010; Sylvestre & Merette, Reference Sylvestre and Merette2010), sociocommunicative competence as reflected through play (Alessandri, Reference Alessandri1991; Darwish, Esquivel, Houtz, & Alfonso, Reference Darwish, Esquivel, Houtz and Alfonso2001; Valentino, Cicchetti, Toth, & Rogosch, Reference Valentino, Cicchetti, Toth and Rogosch2006), internalized morality (Koenig, Cicchetti, & Rogosch, Reference Koenig, Cicchetti and Rogosch2000; Toth, Cicchetti, Macfie, Rogosch, & Maughan, Reference Toth, Cicchetti, Macfie, Rogosch and Maughan2000), social information processing (Pine et al., Reference Pine, Mogg, Bradley, Montgomery, Monk and McClure2005; Rieder & Cicchetti, Reference Rieder and Cicchetti1989), prosocial interactive behaviors (Haskett & Kistner, Reference Haskett and Kistner1991; Howe & Parke, Reference Howe and Parke2001), and school adaptation (Eckenrode, Laird, & Doris, Reference Eckenrode, Laird and Doris1993; Shonk & Cicchetti, Reference Shonk and Cicchetti2001; Trickett, McBride-Chang, & Putnam, Reference Trickett, McBride-Chang and Putnam1994). A recent study linked maltreatment to personality organization that was lower in agreeableness, conscientiousness, and openness to experiences and higher in neuroticism (Rogosch & Cicchetti, Reference Rogosch and Cicchetti2004); and children with a history of maltreatment have been found to exhibit higher levels of aggression (Anthonysamy & Zimmer-Gembeck, Reference Anthonysamy and Zimmer-Gembeck2007; Cullerton-Sen et al., Reference Cullerton-Sen, Cassidy, Murray-Close, Cicchetti, Crick and Rogosch2008). With such a range of potential negative consequences to consider, therefore, studies that evaluate the prevalence of one specific variable (such as PTSD) are necessarily taking a narrow perspective on outcome that fails to truly capture the complex consequences of chronic maltreatment.

Further compounding this difficulty is the reality that many of the problems that may develop as a result of chronic child abuse are likely to be difficult to operationalize for purposes of empirical study. Because trauma is cognitively mediated, it is worth considering the difference between a child who has experienced a single-episode trauma in the context of an otherwise stable childhood versus a child in an abusive environment where the associated trauma is the norm. Certainly it is feasible that the first child in this comparison will be more traumatized in the classic sense of experiencing an event that shatters core beliefs of safety and self, whereas the second child may develop much more complex, fundamental problems of development and mental health.

In addition to considering prevalence, comparisons of rates of childhood PTSD to single events and childhood abuse should also account for chronicity. A number of studies have evaluated retrospectively the impact of specific, single-episode childhood traumas including experiencing a disaster (Green et al., Reference Green, Grace, Vary, Kramer, Gleser and Leonard1994; Morgan, Scourfield, Williams, Jasper, & Lewis, Reference Morgan, Scourfield, Williams, Jasper and Lewis2003), a hostage situation (Desivilya, Gal, & Ayalon, Reference Desivilya, Gal and Ayalon1996), or a transportation accident (Tyano et al., Reference Tyano, Iancu, Solomon, Sever, Goldstein and Touviana1996). These studies indicate that when evaluated years after the accident, a small proportion of individuals do report continued symptoms of PTSD, but that the majority have experienced a decline in symptoms. Given the retrospective nature of these studies it is difficult to ascertain whether this represents a sustained course of PTSD symptoms, or whether subsequent life events have contributed to the development of PTSD. One study did observe that for those who reported sustained symptoms of PTSD, events that served as reminders of the trauma produced symptom exacerbations (Desivilya et al., Reference Desivilya, Gal and Ayalon1996).

Prospective studies evaluating the long-term impact of such events provide the strongest evidence, and data are beginning to accumulate regarding the long-term impact of single-episode events such as transportation accidents (Winje & Ulvik, Reference Winje and Ulvik1998; Yule et al., Reference Yule, Bolton, Udwin, Boyle, O'Ryan and Nurrish2000), exposure to combat zone (Laor et al., Reference Laor, Wolmer, Mayes, Gershon, Weizman and Cohen1997), or disasters (Green et al., Reference Green, Grace, Vary, Kramer, Gleser and Leonard1994) experienced in childhood. These studies suggest that the majority of those who develop PTSD in response to single-event childhood trauma recover from the disorder and do so within a relatively short period of time. A recent study of children and adolescents impacted by Hurricane Katrina indicated that, for most, a significant decline in PTSD symptoms can be observed within 2 to 3 years of the experience of single-episode trauma (Kronenberg et al., Reference Kronenberg, Hansel, Brennan, Osofsky, Osofsky and Lawrqason2010). Echoing observations by Desivilya and colleagues on the importance of environmental factors, continued symptoms were associated with PTSD maternal symptoms (Laor et al., Reference Laor, Wolmer, Mayes, Gershon, Weizman and Cohen1997; Winje & Ulvik, Reference Winje and Ulvik1998), suggesting familial and/or environmental factors mediate posttraumatic symptoms following childhood trauma.

The existing literature therefore indicates that, although some children develop PTSD in response to single-episode childhood trauma, the majority will either never develop PTSD, or will recover within a relatively short period of time. This would seem to further suggest that childhood maltreatment itself is not commonly related to adult PTSD and other factors need to be considered in understanding the association between childhood abuse and adult PTSD.

PTSD in Adults Who Survived the Holocaust as Children

One hypothesis for the apparent link between childhood maltreatment and adult PTSD is that the chronicity of childhood abuse differentiates it from single-episode traumas, producing different effects. However, creating a dichotomy between single and chronic exposure based on comparing single-episode traumas with child abuse may be an oversimplification. The importance of the interpersonal nature of abuse is ignored in such a comparison. There are multiple different types of chronic traumatic stressors that are not the result of repeated physical or sexual abuse, for example, children who are exposed to war, ethnic cleansing, political oppression, concomitant interpersonal violence, and flight from their homes. It is not currently clear whether the long-lasting impact of these events can be more likened to single-episode traumas (which are associated with a greater degree of recovery) or child maltreatment involving physical and sexual abuse (which are associated with the more complex, diverse, and enduring outcomes). Further evaluation of this issue is necessary to create a clear formulation of why childhood abuse produces such enduring outcomes.

In the context of this question, it is compelling to examine the long-term consequences of child survivors of war. Cross-sectional studies have indicated that among children who experience war trauma, between 25% and 62% suffer from PTSD (Catani et al., Reference Catani, Schauer, Elbert, Missmahl, Bette and Neuner2009; Catani, Jacob, Schauer, Mahendran, & Neuner, Reference Catani, Jacob, Schauer, Mahendran and Neuner2008; Elbert et al., Reference Elbert, Schauer, Schauer, Huschka, Hirth and Neuner2008; Klasen et al., Reference Klasen, Daniels, Oettingen, Post, Hoyer and Adam2010; Neugebauer et al., Reference Neugebauer, Fisher, Turner, Yamabe, Sarsfield and Stehling-Ariza2009). In recent years, investigators have begun to describe the adult mental health ramifications of having experienced the Holocaust in childhood. The Holocaust was inarguably an isolated experience in the lives of its survivors but one that lasted over a period of several years for most. The evaluation of Holocaust survivors affords an opportunity to examine the long-term effects of an event that is of comparable magnitude to childhood abuse.

In one sample of Holocaust survivors, nearly half continued to suffer from PTSD despite the passage of more than 50 years (Yehuda, Schmeidler, Siever, Binder-Byrnes, & Elkin, Reference Yehuda, Schmeidler, Siever, Binder-Byrnes and Elkin1997). Despite the differences in the types of traumas experienced by those in camps as opposed to those in hiding, these groups did not differ considerably in types of symptoms. However, there were considerable differences based on the age of trauma exposure. Particularly marked in those who were younger, rather than older, during exposure were hypervigilance, psychogenic amnesia, emotional detachment, and dissociation (Yehuda et al., Reference Yehuda, Schmeidler, Siever, Binder-Byrnes and Elkin1997). Thus, in a group of survivors who were exposed early in life to extreme trauma in the context of war and life threat (but not at the hands of family or other caretakers), there were long-lasting and severe manifestations of PTSD. Furthermore, the nature of the symptoms was closely related to developmental stage at the time of traumatization. These results demonstrate that both chronicity of trauma and age of traumatization may be relevant factors in the nature and persistence of posttraumatic symptoms. The above results were further in accord with a recent meta-analysis of survivors of the Holocaust that also found a significant effect size (1.36) on continued symptoms of PTSD among those who had experienced the Holocaust as children (Barel, van IJzendoorn, Sagi-Schwartz, & Bakermans-Kranenburg, Reference Barel, van IJzendoorn, Sagi-Schwartz and Bakermans-Kranenburg2010).

Although these studies suggest that PTSD symptoms can be enduring among those who experience chronic trauma in childhood, that factors that perpetuate the disruption may not be completely transparent. For example, in child survivors of the Holocaust, there were numerous challenges in addition to having to fight for one's very life. Even after achieving physical safety during the postwar period, many child survivors experienced disruptions in relationships (loss of parental figures, separation from family). These unmeasured factors are very likely to have had a significant impact on long-term symptoms and support the idea that disturbance in relationships with primary attachment figures may contribute with chronicity to result in persistent symptoms. It should also be noted that because this research was completed in adult survivors of the Holocaust, and the time of onset of their symptoms cannot clearly be determined, this does not resolve the issue of why more adults than children report PTSD following childhood abuse (whereas rates of PTSD to single-episode traumas in childhood declines over time), and points toward the need to identify intervening factors more specifically.

Early Life Trauma as a Risk Factor for PTSD

One such intervening factor is the occurrence of additional trauma. It has been repeatedly demonstrated that early life trauma is a risk factor for development of PTSD to subsequent traumas. Several retrospective studies have found that individuals who develop PTSD in response to events that occur in adulthood are more likely to have had earlier traumatic experiences, particularly childhood abuse (Bremner et al., Reference Bremner, Southwick, Johnson, Yehuda and Charney1993; Engel et al., Reference Engel, Engel, Campbell, McFall, Russo and Katon1993; Zlotnick, Reference Zlotnick1997). A meta-analysis of 77 studies of risk factors for PTSD identified a history of reported childhood trauma as a risk for PTSD following adult trauma across populations regardless of gender, trauma type, or population studied (civilian or military; Brewin, Andrews, & Valentine, Reference Brewin, Andrews and Valentine2000). However, the mean effect size was found to be small, and, to the extent that such subjects also have experienced trauma in adulthood, it is difficult to state with conviction that current PTSD symptoms are associated with early life trauma unless other important variables are taken into account. Rather, it appears that early trauma in combination with other individual characteristics and subsequent exposure variables may increase the risk for PTSD.

However, two recent studies of PTSD symptoms and lifetime trauma exposure in combat veterans raise questions regarding the assumptions of ubiquitously negative consequences of early life trauma in terms of susceptibility to adult PTSD (Owens et al., Reference Owens, Dashevsky, Chard, Mohamed, Haji and Heppner2009; Stein et al., Reference Stein, Tran, Lund, Haji, Dashevsky and Baker2005). Both studies found that among individuals exposed to low levels of combat, higher levels of childhood trauma were related to increased severity of PTSD symptoms. However, where levels of combat exposure were high, increased levels of childhood trauma actually served to lower the severity of PTSD symptoms. This led both sets of authors to suggest that perhaps the experience of coping with early life trauma resulted in coping skills that allowed the veterans in these two studies to deal more effectively with the trauma of severe combat, an idea that reflects Rutter's theories about the potential “steeling” effects of stress and adversity (Rutter, Reference Rutter1981).

The existing literature, therefore, clearly does not fully resolve the question regarding the effect that childhood trauma has on the subsequent development of PTSD following adult trauma. Further evaluation of this question is necessary before firm conclusions can be drawn. It is also worth examining in more detail the mechanisms by which early life trauma can present risk for adult PTSD in order to gain a more complete understanding of this process.

The Role of Revictimization

One particularly enlightening area of research indicates that the experience of childhood maltreatment, particularly sexual abuse, increases the likelihood of further victimization (Briere, Woo, McRae, Foltz, & Sitzman, Reference Briere, Woo, McRae, Foltz and Sitzman1997; Cloitre et al., Reference Cloitre, Scarvalone and Difede1997; Desai, Arias, Thompson, & Basile, Reference Desai, Arias, Thompson and Basile2002; Follette, Polusny, Bechtle, & Naugle, Reference Follette, Polusny, Bechtle and Naugle1996; Goodman et al., Reference Goodman, Salyers, Mueser, Rosenberg, Swartz and Essock2001; Humphrey & White, Reference Humphrey and White2000; Jankowski, Leitenberg, Henning, & Coffey, Reference Jankowski, Leitenberg, Henning and Coffey2002; Nishith et al., Reference Nishith, Mechanic and Resick2000; Widom, Reference Widom1999; also for review see Arata, Reference Arata2002; Classen, Palesh, & Aggarwal, Reference Classen, Palesh and Aggarwal2005; Roodman & Clum, Reference Roodman and Clum2001) and revictimization is associated with more PTSD symptoms (Arata, Reference Arata1999; Banyard, Williams, & Siegel Reference Banyard, Williams and Siegel2001; Gibson & Leitenberg, Reference Gibson and Leitenberg2001). This process of revictimization can be differentiated from retraumatization in terms of its interpersonal nature: it is not simply the experience of subsequent trauma such as accidents, disasters, combat, but involves the perpetration of traumatic acts on the victim by another person (such as rape or assault). In this way, it echoes a salient aspect of how childhood abuse differs from other types of single-episode trauma.

Studies of adult sexual assault victims have documented a high degree of revictimization (Briere & Runtz, Reference Briere and Runtz1987; Gidycz & Hanson, Reference Gidycz and Hanson1995; Sorenson, Siegel, Golding, & Stein, Reference Sorenson, Siegel, Golding and Stein1991). Retrospective studies demonstrate a high rate of childhood sexual assault in rape survivors (Briere & Runtz, Reference Briere and Runtz1988; Kendall-Tackett & Simon, Reference Kendall-Tackett and Simon1988; Wyatt & Newcomb, Reference Wyatt and Newcomb1990) and victims of intimate partner abuse (Alexander, Reference Alexander2009; Gomez & Speizer, Reference Gomez and Speizer2009; Halpern, Spriggs, Martin, & Kupper, Reference Halpern, Spriggs, Martin and Kupper2009; Whitfield, Anda, Dube, & Felitti, Reference Whitfield, Anda, Dube and Felitti2003), and the relationship between childhood sexual abuse and adult sexual revictimization remains even when controlling for other childhood and family factors (Fergusson, Horwood, & Lynskey, Reference Fergusson, Horwood and Lynskey1997). In light of previously discussed limitations associated with observations made from retrospective studies, it is even more persuasive that two recent prospective, longitudinal studies found that individuals with substantiated childhood abuse were significantly more likely to experience revictimization as compared to those with no such history (Barnes, Noll, Putnam, & Trickett, Reference Barnes, Noll, Putnam and Trickett2009; Widom, Czaja, & Dutton, Reference Widom, Czaja and Dutton2008). Early childhood victimization is the single best predictor of subsequent victimization (Briere et al., Reference Briere, Woo, McRae, Foltz and Sitzman1997; Elliott, Mok, & Briere, Reference Elliott, Mok and Briere2004). A meta-analysis of revictimization among individuals with childhood sexual abuse found such a strong overall effect size (0.59) that the authors concluded that this supported a causal link between childhood sexual abuse and adult sexual revictimization (Roodman & Clum, Reference Roodman and Clum2001). Perhaps most important of all, early childhood abuse is among the strongest of predictors of symptoms following subsequent victimization (Koopman et al., Reference Koopman, Palesh, Marten, Thompson, Ismailji, Holmes and Corales2005; Norris & Kaniasty, Reference Norris and Kaniasty1994). This literature points toward revictimization as playing an important role in the relationship between child abuse and adult PTSD.

An excellent model for how to best clarify this relationship comes from a series of studies performed by Widom. A cohort of over 650 victims of substantiated child abuse and neglect was identified between 1967 and 1971 and has since been followed into adulthood and evaluated on multiple different variables. Comparisons with control subjects matched for age, race, gender, and socioeconomic status has formed the basis of an increasing number of papers that can convincingly compare rates of multiple sequelae (including substance use, major depression, aggression, and relationship problems) between the abused group and those who did not suffer abuse. One such study provides support for the role of revictimization as a mediator of adult PTSD among individuals abused in early life (Widom, Reference Widom1999). Twenty years after the abuse occurred, diagnostic interviews were obtained on 676 abused and 520 comparison subjects. A prevalence rate of 20.4% of current or lifetime PTSD was found in the comparison subjects. Prevalence rates of PTSD were 37.5% among those who had experienced childhood sexual abuse, 32.7% among those who had experienced physical abuse, and 30.6% among those who had experienced neglect. The group that had been maltreated in childhood was also far more likely to experience subsequent traumatic events such as rape, physical assault, or other traumas. When correcting for family problems associated with the abuse (parents being arrested, drugs or alcohol in self and family, early behavioral problems prior to abuse experiences, low level of education subsequent to abuse) childhood sexual abuse was still associated with PTSD in adulthood but physical abuse and neglect were not. An analysis of the relationship between adult trauma and PTSD, controlling for childhood trauma, was not performed in this study. Koenen and Widom (Reference Koenen and Widom2009), when using this same longitudinal, prospective data, found that the effects of revictimization on PTSD were so significant they accounted for 39% of the gender differences in prevalence of PTSD in this sample.

Nishith, Mechanic, and Resick (Reference Nishith, Mechanic and Resick2000) provided further support for the pivotal role of increased revictimization in the relationship between childhood abuse and PTSD in adulthood. This important study used path analysis to demonstrate that a history of child sexual abuse directly increased the vulnerability to adult sexual and physical victimization, which in turn contributed substantially to the level of current PTSD symptoms. It is interesting that neither physical nor sexual abuse in childhood was directly associated with current PTSD symptoms except through the pathway of increased revictimization, although notably the abuse was assessed retrospectively and the analyses are cross-sectional. The results of this study, together with the extant literature on the effects of child abuse on revictimization and subsequent symptom development, clearly demonstrate a role for cumulative effects of childhood and subsequent traumatic events in the development of PTSD. Cumulative trauma appears to particularly account for the increased estimates of PTSD in adults with early childhood trauma relative to traumatized children.

Pathways to Revictimization

Multiple studies have examined the risk factors that contribute to revictimization among those with a history of childhood abuse. Identified risk factors include existing symptoms of PTSD (Acierno, Resnick, Kilpatrick, Saunders, & Best, Reference Acierno, Resnick, Kilpatrick, Saunders and Best1999; Dietrich, Reference Dietrich2007; Noll, Horowitz, Bonanno, Trickett, & Putnam, Reference Noll, Horowitz, Bonanno, Trickett and Putnam2003) particularly numbing symptoms (Ullman, Najdowski, & Filipas, Reference Ullman, Najdowksi and Filipas2009), emotion dysregulation (Dietrich, Reference Dietrich2007; Messman-Moore, Walsh, & Dilillo, Reference Messman-Moore, Walsh and Dilillo2010), deficits in social support or interpersonal relatedness and skills (Bender, Cook, & Kaslow, Reference Bender, Cook and Kaslow2003; Dietrich, Reference Dietrich2007; Kearns & Calhoun, Reference Kearns and Calhoun2010), risky sexual behavior (Fargo, Reference Fargo2009; Fergusson et al., Reference Fergusson, Horwood and Lynskey1997), and substance use (Lis-Turlejska & Polak, Reference Lis-Turlejska and Polak2002; Ullman et al., Reference Ullman, Najdowksi and Filipas2009). However, although all of these factors are also identified as common consequences of childhood maltreatment, cross-sectional studies do not completely clarify the relationship between these factors and the revictimization. Recent literature has begun the process of using structural equation modeling to address this question and examine the specific pathways to revictimization following childhood abuse (Yehuda et al., Reference Yehuda, Flory, Pratchett, Buxbaum, Ising and Holsboer2010). Various models that explain such pathways are now being identified that include behavioral markers such as adolescent risk-taking behavior, adolescent sexual victimization, problematic alcohol use, problematic alcohol use, frequent alcohol use before sexual activity, and engagement in risky sexual behavior (Fargo, Reference Fargo2009); increased avoidant coping style (Fortier et al., Reference Fortier, DiLillo, Messman-Moore, Peugh, DeNardi and Gaffey2009); emotion dysregulation, both directly and indirectly through risky sexual behavior (Messman-Moore et al., Reference Messman-Moore, Walsh and Dilillo2010); and shaming sexual beliefs and behaviors (Reid & Sullivan, Reference Reid and Sullivan2009).

Another recent paper specifically examined revictimization among the Widom cohort in efforts to obtain a better understanding of the phenomenon (Widom et al., Reference Widom, Czaja and Dutton2008). Comparisons with a matched set of control subjects allowed for analyses in a prospective study of issues including whether risk was increased for general retraumatization or specific types of revictimization, whether gender and ethnicity impacted risk for revictimization, whether risk differed depending on type of abuse or neglect, and whether risk was higher at particular ages. Neglected and abused children were not found to be any more likely to experience later trauma than controls in this sample. However, the number of traumas experienced was significantly higher among the maltreated group, and this difference appeared to be primarily accounted for by experiences of interpersonal revictimization specifically. Although maltreatment interacted with gender to produce differences in risk for specific types of revictimization, overall risk for revictimization associated with childhood abuse did not differ by gender. Maltreatment did interact with ethnicity in that risk for revictimization was higher in the maltreated group among White participants, but among ethnic minorities maltreatment was not associated with greater risk. The authors speculate that this is perhaps due to the comparatively higher levels of trauma experienced among ethnic minorities in this sample.

In general, this issue of high trauma rates is one that should be considered carefully with regard to this study. The rates of trauma in this sample (99% of the maltreated group, 98.7% of the controls) are higher than those reported in broad epidemiological studies that report ranges from 58% to 89% (Breslau et al., Reference Breslau, Kessler, Chilcoat, Schultz, Davis and Andreski1998; Kessler, Sonnega, Bromet, Hughes, & Nelson, Reference Kessler, Sonnega, Bromet, Hughes and Nelson1995), suggesting that the population studied is not representative of the population at large. The authors refer to socioeconomic factors that may contribute to this. Does this suggest that there are predetermined factors that influence the risk of both childhood maltreatment and later victimization? Our studies to date seem to focus on an assumption that the experience of childhood maltreatment is the key etiologic factor in triggering revictimization and initiating a progression of revictimization and symptoms. However, it is interesting to speculate about other explanations: perhaps an underlying factor, yet to be clearly defined, is responsible for each element in this model, including the childhood maltreatment. Given the impact of childhood maltreatment, it is imperative to further examine the issue and identify models that explain the cascade that culminates in PTSD among those who experience abuse and trauma early in life. A review of the animal literature on the biological effects of early stress can assist in this process.

Biological Models of the Effects of Stress: The Centrality of Sensitization as a Mechanism of Permanent Stress Responses

Considerable interest was generated by early research on the physiologic effects of stress (Cannon, Reference Cannon1914; Selye, Reference Selye1936, Reference Selye1956), leading to the idea that external events could have a profound impact on the biology and behavior of organisms. Both acute and chronic stressors could have effects. However, critical to stress theory was that the stressor had to be present to exert its effects; once the stressor had been removed, the organism would begin a recovery from the effects of stress. In the 1970s and 1980s, a series of observations in laboratory animals led to the suggestion that the biological and behavioral effects of stress could continue and strengthen even after the stressor had been removed (Antelman, Reference Antelman1988). Exposure to even brief single-episode stressors, such as needle sticks, was demonstrated to have profound effects on how an animal would respond to a subsequent stressor (Antelman et al., Reference Antelman, Caggiula, Kocan, Knopf, Meyer and Edwards1991). The model of time-dependent sensitization and observations of kindled seizures that followed (reviewed in Post, Weiss, Smith, Li, & McCann, Reference Post, Weiss, Smith, Li and McCann1997) laid the groundwork for current biobehavioral models of stress, which are grounded in the belief that some aspects of exposure can be permanent. These models provided a framework for understanding how traumatic events in humans could have biological and behavioral effects that persisted long after the termination of the event (Antelman et al., Reference Antelman, Caggiula, Kocan, Knopf, Meyer and Edwards1991; Post et al., Reference Post, Weiss, Smith, Li and McCann1997; Yehuda & Antelman, Reference Yehuda and Antelman1993).

Molecular neurobiologic studies have actually demonstrated changes in gene expression that respond within minutes to environmental perturbations, observations that represent a remarkable paradigm shift from earlier formulations. It is now clear that stress activates the formation of gene products, which in turn influence cellular processes that lead to gene expression, protein formation, and concomitant biological and behavioral change (Duman, Heninger, & Nestler, Reference Duman, Heninger and Nestler1994; Hyman & Nestler, Reference Hyman and Nestler1996). Thus, current stress theory is focused on the life-long implications of stressful experiences and interest has increased in examining biological cascades resulting from activation (Masten & Cicchetti, Reference Masten and Cicchetti2010). Such developmental cascades may begin with maltreatment impacting functional systems that then influence neurobiological development, stress reactivity, and subsequent psychopathology (Cicchetti & Cannon, Reference Cicchetti and Cannon1999; Gunnar & Quevedo, Reference Gunnar and Quevedo2007; Meaney, Reference Meaney2010). Furthermore, the possibility of stress-induced changes in gene expression lays the groundwork for intergenerational effects of traumatic stress (Yehuda, Bell, Bierer, & Schmeidler, Reference Yehuda, Bell, Bierer and Schmeidler2008; Yehuda et al., Reference Yehuda, Bierer, Schmeidler, Aferiat, Breslau and Dolan2000) and may eventually contribute to uncovering a genetic or epigenetic basis for individual differences in stress reactivity. It is important that a recent study of brain tissue in suicide victims showed that epigenetic regulation of the glucocorticoid receptor (GR) gene, which is a focus of substantial interest due to the role of glucocorticoids in stress response, is influenced by a history of child abuse (McGowan et al., Reference McGowan, Sasaki, D'Alessio, Dymov, Labonté and Szyf2009). Some support for a genetic basis is also offered by identification of interactions between child abuse and the stress-related gene FKBP5 on PTSD symptoms in adults (Binder et al., Reference Binder, Bradley, Liu, Epstein, Deveau and Mercer2008).

Against this backdrop, it becomes critically important to ask whether there may be differences in the long-term effects of stress depending on the developmental stage at exposure. There is abundant evidence suggesting that trauma occurring early in life may differentially alter neurobiologic and behavioral systems and influence subsequent responses to stress, as reviewed below.

Developmental Influences of Early Stress in Animals

Early experiences are believed to set the level of responsiveness of the HPA axis and autonomic nervous system in such a way that these systems either overreact or underreact to stressful events (McEwen & Magarinos, Reference McEwen and Magarinos1997). Different types of environmental experiences are associated with different neurobiologic outcomes. The elucidation of this biological heterogeneity may be critical to understanding individual differences in responses to early events.

Levine (Reference Levine1967) and Denenberg, Brumaghim, Haltmeyer, and Zarrow (Reference Denenberg, Brumaghim, Haltmeyer and Zarrow1967) were the first to show that exposure of infant animals to mild stress could permanently reduce the pituitary–adrenal responses to subsequent stressors. These observations were replicated and further developed by Meaney, Aitken, Viau, Sharma, and Sarrieau (Reference Meaney, Aitken, Viau, Sharma and Sarrieau1989), who showed that laboratory rats handled within the first 2 weeks of life (which creates stress by involving maternal separation for several minutes per day) showed significantly lower adrenocorticotropic hormone (ACTH) and corticosterone secretion at baseline and in response to stress in adulthood. Furthermore, early handled rats showed an enhanced cortisol suppression following dexamethasone, suggesting an enhanced negative feedback inhibition (Meaney et al., Reference Meaney, Aitken, Viau, Sharma and Sarrieau1989). Consistent with these observations, studies of GR concentrations revealed an increased number of receptors (Meaney, Aitken, van Berkel, Bhatnagar, & Sapolsky, Reference Meaney, Aitken, van Berkel, Bhatnagar and Sapolsky1988) and increased GR mRNA expression in the hippocampal tissue slices (Plotsky & Meaney, Reference Plotsky and Meaney1993). The biological and behavioral findings noted above are similar to those observed in adolescents and adults with chronic PTSD, that is, low basal cortisol levels (Mason, Giller, Kosten, Ostroff, & Podd, Reference Mason, Giller, Kosten, Ostroff and Podd1986; Yehuda et al., Reference Yehuda, Southwick, Nussbaum, Wahby, Giller and Mason1990; Yehuda, Boisoneau, Mason, & Giller, Reference Yehuda, Boisoneau, Mason and Giller1993; Yehuda, Kahana, et al., Reference Yehuda, Kahana, Binder-Brynes, Southwick, Mason and Giller1995), increased cortisol response to dexamethasone (Kellner, Baker, & Yehuda, Reference Kellner, Baker and Yehuda1997; Stein, Yehuda, Koverola, & Hanna, Reference Stein, Yehuda, Koverola and Hanna1997; Yehuda, Boisoneau, Lowy, & Giller, Reference Yehuda, Boisoneau, Lowy and Giller1995; Yehuda, Southwick, et al., Reference Yehuda, Southwick, Krystal, Bremner, Charney and Mason1993), increased GR concentration (Yehuda, Boisoneau, et al., Reference Yehuda, Boisoneau, Mason and Giller1993, Reference Yehuda, Boisoneau, Lowy and Giller1995), and evidence of increased GR activity in the hippocampus (Grossman, Yehuda, & Buchsbaum, Reference Grossman, Yehuda and Buchsbaum1999). However, there are some important differences between the biological consequences of early handling and those observed with PTSD in relation to corticotropin releasing factor (CRF) activity and subsequent neuroendocrine responses to stress. CRF appears to be elevated in PTSD (Baker et al., Reference Baker, West, Nicholson, Ekhator, Kasckow and Hill1999) but not in early handled rats. Trauma survivors appear to show a pituitary–adrenocortical hyperresponsivity to stress (Heim et al., Reference Heim, Newport, Heit, Graham, Wilcox and Bonsall2000), whereas early handled rats showed a lower response to stressors as well as a faster recovery to baseline.

Studies of laboratory rats exposed to early maternal deprivation show a different constellation of biological and behavioral changes than those receiving the manipulation of early handling. In response to subsequent restraint or novelty stress, rats exposed to maternal deprivation show an increased cortisol and ACTH response (Plotsky & Meaney, Reference Plotsky and Meaney1993). However, ACTH, but not cortisol, concentrations have been found to be elevated at baseline. Moreover, maternally deprived rats also showed a decreased cortisol response to dexamethasone and an increased expression of CRF mRNA expression in the hypothalamus (Ladd, Owens, & Nemeroff, Reference Ladd, Owens and Nemeroff1996). Infant rats deprived of their mothers for a 24-hr period show increased CRF receptors in many important brain regions, such as the frontal cortex, hippocampus, amygdala, and cerebellum, in addition to the hypothalamus. Therefore, the acute endocrine response to subsequent stress may result from biological sensitization due to prior alterations in CRF receptor regulation.

Studies of maternal deprivation in nonhuman primates have pointed to the importance of the early mother–infant relationship in mediating individual differences in subsequent coping with adversity (Levine & Wiener, Reference Levine and Wiener1988). Maternal deprivation in primates is stressful, but it is not clear whether it is a better model for abuse or for neglect. Nonetheless, the response to subsequent stress cannot be predicted solely on the basis of this deprivation. Rather, temperamental, genetic, and environmental effects appear to be associated both with the monkey's ability to cope with environmental challenge as an adult and with variability in basal and stress-related changes in hormones such as ACTH and cortisol (Levine, Wiener, & Coe, Reference Levine, Wiener and Coe1993; Lyons, Yang, Mobley, Nickerson, & Schatzberg, Reference Lyons, Yang, Mobley, Nickerson and Schatzberg2000).

Perhaps more relevant to studies of childhood stress are experiments of monkeys who were assigned to different rearing conditions in order to produce different types of foraging stress. In this paradigm, mothers and their infants were randomized into different rearing conditions requiring low (less stressful), higher (predictable stress), or variable (unpredictable, most stressful) foraging demands on the mother. The infant monkeys were subjected to these conditions until young adulthood (2–4 years), and then a series of biological tests were performed. Adult monkeys who had been reared in the variable foraging condition showed low cortisol but increased CRF levels (Coplan et al., Reference Coplan, Andrews, Rosenblum, Owens, Friedman and Gorman1996), similar to what has been observed in adults with chronic PTSD (Baker et al., Reference Baker, West, Nicholson, Ekhator, Kasckow and Hill1999; Bremner et al., Reference Bremner, Licinio, Darnell, Krystal, Owens and Southwick1997).

Additional recent studies in monkeys have highlighted HPA axis sensitization following repeated, brief, unpredictable mother–infant separations at 3–6 months of age (Sanchez et al., Reference Sanchez, Noble, Lyon, Plotsky, Davis and Nemeroff2005). Infants, particularly females, manifested higher cortisol levels at the end of the separation protocol, followed by persistently lowered morning cortisol, resulting in flattened diurnal cortisol rhythms. These HPA axis findings were accompanied by signs of increased anxiety in the form of increased acoustic startle response. Again, these findings echo those of flattened diurnal cortisol rhythms in children and adolescents with early maltreatment (Gunnar & Vazquez, Reference Gunnar and Vazquez2001).

These studies in animals underscore the important role of early adverse life events in behavioral sensitization of both the HPA and noradrenergic systems. Although extrapolation from animal studies of stress to childhood traumatization is problematic, it is interesting to observe that neither the effects of childhood trauma nor early stressful life events in animals result in a uniform series of consequences but rather are predicted on the basis of a complex set of variables, including genetics, social interaction, and other mediating environmental events. In particular, the role of adverse caregiving by the attachment figure (mother) has been shown to be relevant in this sensitization process, an issue that could arguably be expected to be more pivotal in child abuse than other types of trauma.

Biological Consequences in Adults Who Experienced Trauma in Childhood

Just as there are myriad behavioral and mental consequences associated with childhood traumatic experiences, there are likely to be a range of biological alterations associated with early trauma exposure. To the extent that biological differences have been associated with these diverse outcomes, it could reasonably be assumed that there is no specific biological consequence related to early trauma. It is more reasonable to focus on the type of alterations that might be expected (e.g., neuroendocrine, neurochemical). To date, many studies examine long-term biological correlates of early abuse in the context of PTSD and highlights from these studies are reviewed below.

Implications for Intervention

The therapeutic implications of this perspective are important. Generally, the field of mental health offers intervention to individuals (particularly children) who have been found to suffer from psychopathology. However, a perspective of mental health that involves a trajectory from childhood abuse to a plethora of developmental, behavioral, emotional, and biological consequences that may lay a foundation of risk for subsequent psychopathology (i.e., PTSD) provides a good justification for a preventive model of intervention.

Unquestionably, the prevention of childhood abuse would be the ultimate outcome, and therefore interventions targeted to this would be the most effective. To this end, treatments that focus on parent training have been developed, and there is some evidence for their effectiveness in reducing rates of childhood abuse in targeted families (Chaffin et al., Reference Chaffin, Silovsky, Funderburk, Valle, Brestan and Balachova2004; Peterson, Tremblay, Ewigman, & Saldana, Reference Peterson, Tremblay, Ewigman and Saldana2003). However, a recent review of such treatments concluded that the evidence base for primary prevention programs remains relatively weak (Reynolds, Mathieson, & Topitzes, Reference Reynolds, Mathieson and Topitzes2009); therefore, secondary prevention programs need to be considered.

Given the potential for problems that develop from childhood abuse to adult PTSD, early screening and identification of traumatized children is necessary to change this trajectory. If, as hypothesized, disruptions in the parent–child relationship are a pivotal issue in the development of adult PTSD subsequent to childhood abuse, interventions that target this relationship would be indicated. There is actually a growing body of evidence that supports interventions such as parent–child interaction therapy (Chaffin et al., Reference Chaffin, Silovsky, Funderburk, Valle, Brestan and Balachova2004), child–parent psychotherapy (Lieberman & Van Horn, Reference Lieberman and Van Horn2008) and trauma-focused cognitive behavioral therapy (Cohen, Mannarino, & Deblinger, Reference Cohen, Mannarino and Deblinger2006) that incorporate the parent–child relationship in the treatment of the consequences of child abuse. In light of the perspective that early responses to trauma may operate to increase risk for revictimization and subsequent PTSD, it is also important that any interventions for children who are abused do not simply focus on treating PTSD or other disorders as specifically defined by DSM-IV. Rather, in order to change this trajectory, it is imperative that treatments address the many subthreshold symptoms, and developmental, behavioral, and emotional sequelae already discussed. Recent developments in our understanding of the pathways to revictimization offer some theoretical guidance on targets of preventive intervention, and recent studies of psychotherapies for traumatized children (including trauma-focused cognitive behavioral therapy and child–parent psychotherapy) have indicated that they can positively impact a broad range of outcome variables (affective dysregulation, behavioral problems, shame and guilt-related cognitions, and interpersonal functioning) as well as improving symptoms of PTSD (Cohen, Deblinger, Mannarino, & Steer, Reference Cohen, Deblinger, Mannarino and Steer2004; Cohen & Mannarino, Reference Cohen and Mannarino2010). Further, there is some recent evidence that family-based interventions targeting behavioral change in children can alter HPA axis functioning (Brotman et al., Reference Brotman, Gouley, Huang, Kamboukos, Fratto and Pine2007; Fisher, Gunnar, Chamberlain, & Reid, Reference Fisher, Gunnar, Chamberlain and Reid2000), and a recent study linked the changes in HPA reactivity to treatment induced changes in behavior (O'Neal et al., Reference O'Neal, Brotman, Huang, Gouley, Kamboukos and Calzada2010). It is possible, therefore, that early intervention in a manner that addresses the developmental, behavioral, and emotional sequelae of childhood abuse could potentially alter a trajectory to adult PTSD by not only reducing the risk of revictimization, but also by regulating HPA axis stress responsivity that may have been negatively affected by the trauma. Further study is needed to determine whether this speculation is borne out. Longitudinal studies that assess the effectiveness of these treatments in preventing the development of adult PTSD are also still necessary.

Summary and Conclusions

The effects of early life trauma on adults are diverse, and although pathology is not the only outcome, mental disorders including PTSD can develop. However, this link is less strong among adults who experienced single-event traumas as children than it is among those who experienced chronic trauma such as childhood abuse. When considering this in the context of the findings that PTSD is more common among adults than children who experienced childhood abuse, assumptions of the direct link between childhood trauma and adult PTSD should be challenged. There is evidence that the experience of childhood trauma, and in particular abuse, may be a risk factor for subsequent development of PTSD following exposure to adult trauma, and a substantial literature exists supporting a pivotal role of the increased risk of subsequent victimization. The literature on the effects of early stress in animal models may provide some explanations for this. Specifically, it seems possible that permanent sensitization of the HPA axis and behavioral outcomes result from early life stress and then combine with the impact of retraumatization to sustain, perpetuate, and amplify symptomatology of those exposed to trauma in childhood.

In considering this model, the importance of disturbance in attachment relationships as a key traumatizing element should be noted. Animal models commonly use maternal deprivation as the stressor in these models, and it is possible that this actually reflects a key ingredient rather than simply a type of induced stress. This perspective allows us to view the different relationship that childhood abuse and single-episode trauma have with adult PTSD through a broader lens than one that perceives only chronicity, but rather emphasizes the importance of the disruption in parental or attachment relationships that are implicit in childhood abuse.

Several major gaps exist in our knowledge at this time. There is a need to better delineate the additional risk factors that mediate disease and recovery from trauma, the relationship between the biological and behavioral consequences of trauma, and the similarities and differences between the consequences of exposure to single and multiple traumas, between interpersonal trauma and other types of trauma and their cumulative effects. It is necessary to identify which environmental factors exacerbate and which reduce the effects of stress and likelihood of revictimization and to examine the impact of altering environments appropriately for children who have been exposed to trauma. As the mental health field questions our assumptions about the link between early life trauma and adult pathology, it will be possible to provide more focused interventions for both the child and the adult suffering from the effects of such exposures. These questions will only be answered unequivocally by longitudinal, prospective studies that use unbiased samples rather than relying on retrospective reporting by clinical patients.