Sex differences in trauma and posttraumatic stress disorder: a quantitative review of 25 years of research
- TOLIN David F., FOA Edna B.
- Journal article citation:
- Psychological Bulletin, 132(6), November 2006, pp.959-992.
- American Psychological Association
Meta-analyses of 290 studies yielding data on the sex-specific risk of potentially traumatic events and post traumatic stress disorder (PTSD) show that females are approximately twice as likely as males to meet the diagnostic criteria for PTSD despite the fact that they are at lower risk of experiencing a potentially traumatic event (PTE). Females were more likely than males to report experience of child sexual abuse or adult sexual assault, but no more likely to exhibit PTSD in response to these events. Males were more likely than females to report experience of accidents, non-sexual assault, witnessing death or injury, disaster or fire, and combat or war, but females exhibited greater PTSD in response to these events. Sex differences in the risk of exposure to particular types of PTE can only partially account for the differential PTSD risk in males and females, and further research is needed.
- Extended abstract:
TOLIN David F.; FOA Edna B.;
Sex differences in trauma and posttraumatic stress disorder: a quantitative review of 25 years of research.
Journal citation/publication details
Psychological Bulletin, 132(6), November 2006, pp.959-992.
This detailed quantitative analysis covers 290 studies, and reveals that females have an approximately two-fold higher risk of PTSD than males despite the fact that males report higher levels of exposure potentially traumatic events. This sex difference persists even after controlling for the type of event, and requires further investigation.
Post traumatic stress disorder (PTSD) was originally codified for the Diagnostic and Statistical Manual of Mental Disorders (DSM) as a result of the high numbers of Vietnam veterans seeking treatment. However, subsequent epidemiological studies suggest that it may be more prevalent among females than males. The aim of this review is to examine sex differences in vulnerability to PTSD, and to investigate whether methodological differences between studies affect results on sex difference. The authors use the term sex (representing biological characteristics) rather than gender, which represents a more complex set of social and psychological constructs beyond the scope of their review.
What sources were used?
Medline and PsycINFO were searched from 1980 (when the DSM definition of PTSD first appeared) onwards to July 2005. Relevant journals (some specified) were also hand searched, and published literature reviews (unspecified) were checked.
What search terms/strategies were used?
The search terms were: PTSD, posttraumatic stress disorder, post-traumatic stress disorder, trauma, disaster, accident, combat, war, abuse, assault, rape, crime.
What criteria were used to decide on which studies to include?
Eligible studies were quantitative, included both male and female participants, and 'were not treatment outcome studies'.
Who decided on their relevance and quality?
The searches yielded 2,477 papers, of which 2,187 were excluded. The primary reasons for exclusion are given by the authors, and they suggest that many of the papers were checked in full text. No details of who carried out relevance assessment are given.
How many studies were included and where were they from?
A total of 290 studies were coded for the review and are asterisked in the references to the paper. A list of the included and excluded studies is also available from the lead author.
How were the study findings combined?
Each of the studies was coded for 18 variables including: number of male and female participants; age range of participants; number of male and female participants who had experienced a range of types of potentially traumatic event (PTE); number of male and female participants diagnosed with PTSD; mean and standard deviation on a continuous measure of PTSD for male and female participants; whether the study sample was epidemiological or a convenience sample; whether an interview or questionnaire was used; etc. Full details of all the coded variables are available from the lead author. The data were analysed using Comprehensive Meta-Analysis Software (version 2.2).
Findings of the review
The findings are presented narratively and in a range of tables that require statistical knowledge to interpret, to answer the review's four main questions.
Are women and girls more likely than men and boys to meet diagnostic criteria for PTSD?
Fifty-two separate male-female comparisons from 40 non-overlapping studies indicate that the odds of meeting the diagnostic criteria for PTSD are approximately twice as high for females than males. This finding remains robust after testing for demographic and methodological variables.
Are women and girls more likely than men and boys to experience a traumatic event?
Twenty-two sex comparisons from 19 separate studies addressed this question. They show that adult males are significantly more likely to report experience of a PTE than adult females, although this result is not replicated in all studies. Thus the two-fold increased risk of PTSD among females occurs despite a significantly lower overall probability of PTE exposure. One explanation may be that females are more likely to experience certain types of event that are disproportionately likely to lead to PTSD.
Do male and female participants differ in terms of the type of traumatic experience?
Sixty-four papers provided 482 comparisons of the frequency of different types of PTE among males and females. These show that while males are more likely than females to report exposure to such events, this is only true for certain categories, namely: accidents, non-sexual assault, combat or war, disaster or fire, serious illness or unspecified injury, and witnessing death or injury. Females, however, are more likely to report sexual assault or abuse, whether in childhood or adulthood. In order to identify whether women and girls are at higher risk of PTSD because they are at higher risk of sexual assault and abuse, it is necessary to discover whether the PTSD prevalence difference between the sexes remains after controlling for the type of trauma.
Do sex differences in PTSD remain when controlling for type of trauma?
This analysis examined the frequency and severity of PTSD among male and female participants who reported the same category of PTE, and shows that for those events that males tend to report (accident, serious illness, disaster etc) females are more likely to meet the criteria for PTSD and to report greater severity of PTSD. However, for child sexual abuse and adult sexual assault no significant sex differences were apparent, except 'under certain methodological circumstances' such as when structured interviews were used. Thus the higher prevalence of PTSD among females cannot be attributed solely to a higher risk of sexual assault or abuse.
The authors then explore whether the sex differences in PTSD and in reported exposure to PTEs could be 'mere methodological artifacts', the result of methodological shortcomings in the primary research studies, or the product of other variables that are not captured by most studies. These could include differences within particular types of PTE; the possibility that the currently recognised symptomatic patterns of PTSD describe female behaviour more accurately than male behaviour; or sex differences in cognitive and behavioural responses to PTEs. In all these cases possibilities are raised that require further investigation.
'The issue of sex differences in trauma and PTSD is both complex and sensitive' and 'much more research is needed before strong conclusions can be reached regarding how sex acts as a vulnerability or resilience factors' in the face of potentially traumatic events.
Implications for policy or practice
None are discussed.
- Subject terms:
- men, post traumatic stress disorder, quantitative research, systematic reviews, women, boys, girls;
- Content type:
- systematic review
- Journal home page
- ISSN print: