Are Women at Greater Risk for PTSD than Men?

July 1, 1998
Glenn Craig Davis, MD

,
Naomi Breslau, PhD

Volume 15, Issue 7

Differences between the sexes regarding the prevalence, psychopathology and natural history of psychiatric disorders have become the focus of an increasingly large number of epidemiological, biological and psychological studies. A fundamental understanding of sex differences may lead to a better understanding of the underlying mechanisms of diseases, as well as their expression and risks.

Differences between the sexes regarding the prevalence, psychopathology and natural history of psychiatric disorders have become the focus of an increasingly large number of epidemiological, biological and psychological studies. A fundamental understanding of sex differences may lead to a better understanding of the underlying mechanisms of diseases, as well as their expression and risks.

Community studies have consistently demonstrated a higher prevalence of posttraumatic stress disorder (PTSD) in females than in males. Recent epidemiologic studies conducted by Davis and Breslau and summarized in this article have begun to elucidate the causes of this higher prevalence of PTSD in women.

Davis and Breslau's studies addressing this issue include Health and Adjustment in Young Adults (HAYA) (Breslau et al., 1991; 1997b; in press) and the Detroit Area Survey of Trauma (DAST) (Breslau et al., 1996).

In the HAYA study, in-home interviews were conducted in 1989 with a cohort of 1,007 randomly selected young adult members, between the ages of 21 and 30, of a 400,000-member HMO in Detroit and surrounding suburban areas. Subjects were reevaluated at three and five years post-baseline interview. The DAST is a random digit dialing telephone survey of 2,181 subjects between the ages of 18 and 45, conducted in the Detroit urban and suburban areas in 1986. Several national epidemiologic studies that report sex differences in PTSD include the NIMH-Epidemiologic Catchment Area survey (Davidson et al., 1991; Helzer et al., 1987) and the National Comorbidity Study (Bromet et al.; Kessler et al., 1995).

Epidemiologic studies, particularly those focusing on the evaluation of risk factors for illness, have a long and distinguished history in medicine. However, it is important to understand that the proposition that there are factors predisposing individuals to the risk for PTSD was controversial in the early phase of characterizing this diagnosis. Many clinicians believed that a highly traumatic stressor was sufficient for the development of PTSD and that the stressor alone "caused" the disorder. But even early studies demonstrated that not all, and often a small number of, individuals exposed to even highly traumatic events develop PTSD.

Why do some individuals develop PTSD while others do not? Clearly, factors other than exposure to adverse events must play a role in the development of the disorder. In the late 1980s, a number of investigators began to examine risk factors that might lead not only to the development of PTSD, recognizing that the identification of risk factors should lead to a better understanding of the pathogenesis of the disorder, but also to a better understanding of the commonly comorbid anxiety and depression in PTSD and, most importantly, to the development of improved treatment and prevention strategies.

Since the diagnosis of PTSD is dependent upon the presence of an adverse (traumatic) event, it is necessary to study both the risk for the occurrence of adverse events and the risk for developing the characteristic symptom profile of PTSD among exposed individuals. One fundamental question addressed by the analysis of both types of risk is whether differential rates of PTSD could be due to differential exposure to events and not necessarily to differences in the development of PTSD.

Early epidemiologic studies identified risk factors for exposure to traumatic events and subsequent risk for the development of PTSD in such exposed populations (Breslau et al., 1991). For example, alcohol and drug dependence was found to be a risk factor for exposure to adverse events (such as automobile accidents), but was not a risk factor for the development of PTSD in exposed populations. However, a prior history of depression was not a risk factor for exposure to adverse events but was a risk factor for PTSD in an exposed population.

In an initial report (Breslau et al., 1991), the evaluation of risk of exposure and risk of PTSD in exposed individuals demonstrated important sex differences. Females did have higher prevalence of PTSD than males. Females were somewhat less likely to be exposed to adverse traumatic events but were more likely to develop PTSD if exposed. Thus, an overall increased prevalence of PTSD in females must be accounted for by a significantly greater vulnerability to develop PTSD after exposure. Why is this?

Before we attempt to answer this question, it is important to examine the overall pattern of a lower burden of trauma in females than in males. The fact that women are exposed to fewer traumatic events obscures an important variation across "types of traumatic events." In the DAST (Breslau et al., in press), adverse events are classified into various categories: assaultive violence, other injury or shocking event, learning of traumas of others, and sudden unexpected death of relative or friend. The category with the highest rates of PTSD is assaultive violence.

Do females experience proportionately more assaultive events than males? The answer is no. Actually, males experience assaultive violence more frequently then females. Assaultive violence as a category is composed of rape, sexual assault other than rape, military combat, being held captive, being tortured or kidnapped, being shot or stabbed, being mugged, held-up, or threatened with weapons, and being badly beaten up. While females experience fewer assaultive events than males, they do experience significantly higher rates of one type of assaultive violence, namely rape and sexual assault.

Does a differential rate of rape and sexual assault between males and females account for the rates of PTSD? No. Females actually have higher rates of PTSD across all types of events in the assaultive violence category, both for events to which they are more exposed (rape) and for events to which they have less exposure (mugged, held-up, threatened with a weapon).

To provide a more quantitative picture from one study (Breslau et al., in press), the conditional risk of PTSD associated with exposure to any trauma was 13% in females and 6.2% in males. The sex difference in conditional risk of PTSD was due primarily to females' greater risk of PTSD following exposure to assaultive violence (36% versus 6%). Sex differences in three other categories of traumatic events (injury or shocking experience, sudden unexpected death, learning about traumas of a close friend or relative) were not significant.

Within the assaultive violence category, women had a higher risk of PTSD for virtually every type of event such as rape (49% versus 0%); sexual assault other than rape (24% versus 16%); mugging (17% versus 2%); held captive, tortured or kidnapped (78% versus