levels of nonspecific psychological distress. Family violence in the past year was assessed with the Standard Family Violence Measure, an eight-item subscale of the CTS, and an Alternate Family Violence Measure, which is the total number of violent acts committed or threatened in the past year; scores were categorized as low, medium-low, medium-high, or high violence. Both measures were completed by the veteran as a self-report and by the spouse/partner of the veteran. Family violence, perpetrated by the veteran or by the spouse or partner, was significantly more prevalent in families of veterans with PTSD, with about a third of the veterans with PTSD having engaged in some level of family violence in the past year compared with 15% of those without PTSD. The mean score on the Standard Family Violence Index for the 736 veterans with PTSD was 2.08 vs 0.54 for the 231 veterans without PTSD (p = 0.002), scores for the spouses or partners were 1.57 and 0.51, respectively (p = 0.001). In all, 372 spouses or partners were interviewed. The 122 spouses or partners of veterans with PTSD reported up to four times as much medium-high to very-high family violence perpetrated by the veteran as did the 252 spouses or partners of veterans without PTSD (7-12% vs 3%, standard error 1.4-3.9, p < 0.01) on both the Standard Family Violence Index and the Alternate Family Violence Index. The mean number of violent acts committed or threatened by veterans in the past year was 4.86 for those with PTSD vs 1.32 for those without PTSD; for spouses or partners of the veterans the mean number of violent acts was 3.03 and 0.96, respectively. Over 9% of the veterans with PTSD had committed 13 or more acts of violence in the past year. The analyses were weighted to compensate for differences in selection probabilities so that the data provide unbiased national estimates for all male theater veterans with a spouse or partner. Age, sex, race or ethnicity, and nonresponse were used to stratify the weights. Strengths of this study are the reporting of family violence perpetrated by both the veteran and the spouse or partner and the use of a nationally representative sample of veterans.
In a further analysis of the same 376 Vietnam veteran couples who had participated in NSVG, Orcutt et al. (2003) used structural-equation modeling to assess the influence of early-life stressors, war-zone stressors, and PTSD symptom severity on intimate partner severity. The modeling showed that there were four direct influences on intimate partner violence in male Vietnam veterans: a poor relationship with mother, combat exposure, perceived threat in the war zone, and PTSD symptom severity. All the influences resulted in more intimate partner violence except for combat exposure. Increasing combat exposure was related to less violence against a spouse or partner. The model also suggests that retrospective reports of a stressful early family-life and antisocial behavior during childhood acted indirectly on intimate partner violence via war-zone stressors and PTSD symptom severity.
Taft et al. (2005) also assessed the NSVG subsample of the 376 veterans and spouses or partners who had participated in the family interview. Veterans with a lifetime history of physical violence toward their spouse or partner but with none reported for the past year were excluded. In all, 40 male veterans were classified as both PTSD-positive and partner violent (one or more episodes of partner violence in the past year), 41 were PTSD-negative but partner violent, and 28 were PTSD-positive but nonviolent. Data analysis showed that PTSD severity did not differ significantly between the two PTSD groups, nor did violence severity between the two partner-violent groups. The PTSD-positive and partner-violent group scored higher for all the following factors than did the PTSD-positive/nonviolent or PTSD-negative/partner-violent groups: psychiatric disorders (antisocial personality disorder, major depressive disorder, and alcohol and drug-use/dependence), violence between the veteran’s parents, relationship variables (marital adjustment, family adaptability, family cohesion), and war-zone variables (combat