Eric F. Crawford

The association of posttraumatic stress disorder and metabolic syndrome: a study of increased health risk in veterans

BackgroundThere is accumulating evidence for a link between trauma exposure, posttraumatic stress disorder (PTSD) and diminished health status. To assess PTSD-related biological burden, we measured biological factors that comprise metabolic syndrome, an important established predictor of morbidity and mortality, as a correlate of long-term health risk in PTSD.MethodsWe analyzed clinical data from 253 male and female veterans, corresponding to five factors linked to metabolic syndrome (systolic and diastolic blood pressure, waist-to-hip ratio and fasting measures of high-density lipoprotein (HDL) cholesterol, serum triglycerides and plasma glucose concentration). Clinical cut-offs were defined for each biological parameter based on recommendations from the World Health Organization and the National Cholesterol Education Program. Controlling for relevant variables including sociodemographic variables, alcohol/substance/nicotine use and depression, we examined the impact of PTSD on metabolic syndrome using a logistic regression model.ResultsTwo-fifths (40%) of the sample met criteria for metabolic syndrome. Of those with PTSD (n = 139), 43% met criteria for metabolic syndrome. The model predicted metabolic syndrome well (-2 log likelihood = 316.650, chi-squared = 23.731, p = 0.005). Veterans with higher severity of PTSD were more likely to meet diagnostic criteria for metabolic syndrome (Wald = 4.76, p = 0.03).ConclusionThese findings provide preliminary evidence linking higher severity of PTSD with risk factors for diminished health and increased morbidity, as represented by metabolic syndrome.

Diagnostic efficiency of the AUDIT-C in U.S. veterans with military service since September 11, 2001.

BACKGROUND Alcohol screening with the 3-item alcohol use disorders identification test (AUDIT-C) has been implemented throughout the U.S. Veterans Health Administration. Validation of the AUDIT-C, however, has been conducted with samples of primarily older veterans. This study examined the diagnostic efficiency of the AUDIT-C in a younger cohort of veterans who served during Operation Enduring Freedom and/or Operation Iraqi Freedom (OEF/OIF). METHODS Veteran participants (N=1775) completed the alcohol use disorders identification test (AUDIT) and underwent the structured clinical interview for DSM-IV-TR for Axis I disorders (SCID) in research settings within four VA medical Centers. Areas under receiver operating characteristic curves (AUCs) measured the effiency of the full AUDIT and AUDIT-C in identifying SCID-based diagnoses of past year alcohol abuse or dependence. RESULTS Both measures performed well in detecting alcohol use disorders. In the full sample, the AUDIT had a better AUC (.908; .881-.935) than the AUDIT-C (.859; .826-.893; p<.0001). It is notable that this same result was found among men but not women, perhaps due to reduced power. Diagnostic efficiency statistics for the AUDIT and AUDIT-C were consistent with results from older veteran samples. The diagnostic efficiency of both measures did not vary with race or age. CONCLUSIONS Both the AUDIT and AUDIT-C appear to be valid instruments for identifying alcohol abuse or dependence among the most recent cohort of U.S. veterans with service during OEF/OIF within research settings.

Validity of the Personality Assessment Inventory Aggression Scales and Violence Potential Index in Veterans With PTSD

abstract In this study, we examined the validity of the Personality Assessment Inventory (Morey, 1991) Aggression (AGG) scales and Violence Potential index (VPI) in 399 male combat veterans presenting for formal evaluation of posttraumatic stress disorder (PTSD). The AGG scales exhibited convergence with other measures of hostility and violence and demonstrated discriminant validity with alternative constructs. When we examined reports of interpersonal violence in the past year, the AGG composite scale displayed substantial incremental validity over the effects of PTSD severity, demographics, a simple dichotomous question regarding violence in the past 30 days, and MMPI-2 (Butcher, Dahlstrom, Graham, Tellegen, & Kaemmer, 1989) scales measuring aggression. The VPI, however, added no unique explanatory power over the AGG composite scale.

Prolonged Exposure Therapy With Veterans and Active Duty Personnel Diagnosed With PTSD and Traumatic Brain Injury

The present study used archival clinical data to analyze the delivery and effectiveness of prolonged exposure (PE) and ancillary services for posttraumatic stress disorder (PTSD) among Operation Enduring Freedom, Operation Iraqi Freedom, and Operation New Dawn veterans (N = 69) with histories of mild to severe traumatic brain injury (TBI). Data from standard clinical assessments of veterans and active duty personnel treated in both inpatient and outpatient programs at 2 Department of Veteran Affairs medical centers were examined. Symptoms were assessed with self-report measures of PTSD (PTSD Checklist) and depression (Beck Depression Inventory-II) before and throughout therapy. Mixed linear models were utilized to determine the slope of reported symptoms throughout treatment, and the effects associated with fixed factors such as site, treatment setting (residential vs. outpatient), and TBI severity were examined. Results demonstrated significant decreases in PTSD, B = -3.00, 95% CI [-3.22, -2.78]; t(210) = -13.5; p < .001, and in depressive symptoms, B = -1.46, 95% CI [-1.64, -1.28]; t(192) = -8.32; p < .001. The effects of PE treatment did not differ by clinical setting and participants with moderate to severe injuries reported more rapid gains than those with a history of mild TBI. The results provide evidence that PE may well be effective for veterans with PTSD and TBI.

Surveying Treatment Preferences in U.S. Iraq–Afghanistan Veterans With PTSD Symptoms: A Step Toward Veteran‐Centered Care

This study examined health care barriers and preferences among a self-selected sample of returning U.S. veterans drawn from a representative, randomly selected frame surveyed about posttraumatic stress disorder (PTSD) symptomatology and mental health utilization in the prior year. Comparisons between treated (n = 160) and untreated (n = 119) veterans reporting PTSD symptoms were conducted for measures of barriers and preferences, along with logistic models regressing mental health utilization on clusters derived from these measures. Reported barriers corroborated prior research findings as negative beliefs about treatment and stigma were strongly endorsed, but only privacy concerns were associated with lower service utilization (B = -0.408, SE = 0.142; p = .004). The most endorsed preference (91.0%) was for assistance with benefits, trailed by help for physical problems, and particular PTSD symptoms. Help-seeking veterans reported stronger preferences for multiple interventions, and desire for services for families (B = 0.468, SE = 0.219; p = .033) and specific PTSD symptoms (B = 0.659, SE = 0.302; p = .029) were associated with increased utilization. Outcomes of the study suggested PTSD severity drove help-seeking in this cohort. Results also support the integration of medical and mental health services, as well as coordination of health and benefits services. Finally, the study suggested that outreach about privacy protections and treatment options could well improve engagement in treatment.

MMPI-2 assessment of malingered emotional distress related to a workplace injury : A mixed group validation

In this study, we examined the capacity of MMPI–2 (Butcher, Dahlstrom, Graham, Tellegen, & Kaemmer, 2001) validity indexes to identify malingered depression associated with a workplace injury. We compared 27 graduate students simulating depression with archival records of 33 inpatients diagnosed with major depressive disorder. We employed a mixed-group validation design to generate true positive rates (TPR) and false positive rates (FPR) for the various MMPI–2 validity scales [F, FB, F(p), FBS, F – K, Ds2] while we accounted for base rates of malingering in each sample. The Fake Bad scale (FBS) was the only validity measure that produced acceptable TPR and FPR or a significant correlation with malingering status.

Predicting mortality in veterans with posttraumatic stress disorder thirty years after Vietnam.

Previous research has demonstrated elevated mortality rates among Vietnam-era veterans with posttraumatic stress disorder, especially deaths resulting from nonmedical causes. However, information on the relative contribution of particular risk factors to increased mortality is limited. We used receiver operating characteristics methodology to identify patient-level characteristics that predicted 7-year mortality in 79,551 middle-aged, male, posttraumatic stress disorder–diagnosed outpatients seeking mental health treatment within the Veterans Affairs Health Care System between April 1, 1998 and September 30, 1998. Receiver operating characteristics models indicated that the strongest predictor of mortality was a recent history of medical hospitalization, followed by severity of medical diagnoses and presence of a substance disorder. Results highlight the importance of addressing comorbid medical illnesses and addictive disorders when caring for this population.

Diagnostic Efficiency of the Personality Assessment Inventory LOGIT Function for Posttraumatic Stress Disorder in Women

In this study, we examined the diagnostic efficiency of the Personality Assessment Inventory (PAI; Morey, 1991) for the assessment of posttraumatic stress disorder (PTSD) in a community-based sample of women (n = 128). Receiver operating characteristic curves (ROC) were generated to examine the efficiency of the PAI PTSD LOGIT function as a tool for diagnosing PTSD. Using the Clinician Administered PTSD Scale as the reference standard, the PTSD LOGIT function performed well (area under the curve [AUC] = .856, standard error [SE] = .034). This represents performance consistent with more commonly used self-report PTSD scales, the Davidson Trauma Scale (AUC = .863, SE = .033) and the PAI Anxiety subscale Anxiety-Related Disorders Traumatic Experiences (AUC = .861; S.E. = .033). Results of this study suggest that the PAI PTSD LOGIT function may be a useful tool in assessing PTSD.

Risk Factors for Concurrent Suicidal Ideation and Violent Impulses in Military Veterans.

Suicide and violence are significant problems in a subset of Iraq/Afghanistan-era veterans. This study investigates how posttraumatic stress disorder (PTSD) and resilience in veterans are associated with suicidal ideation and violent impulses while controlling for known covariates of both adverse outcomes. Structured clinical interviews were conducted of N = 2,543 Iraq/Afghanistan-era U.S. veterans. Compared with veterans denying suicidal ideation or violent impulses (n = 1,927), veterans endorsing both (n = 171) were more likely to meet diagnostic criteria for PTSD, report childhood abuse, combat exposure, physical pain symptoms, and drug misuse, and less likely to endorse self-direction/life purpose. Veterans reporting concurrent suicidal ideation and violent impulses had higher odds of misusing drugs and reporting pain symptoms relative to veterans reporting suicidal ideation only (n = 186) and had lower odds of endorsing self-direction/life purpose compared with veterans reporting violent impulses only (n = 259). The findings underscore the importance of examining drug abuse, physical pain symptoms, and self-direction/life purpose, as well as PTSD and history of trauma, in the context of clinical assessment and empirical research aimed at optimizing risk management of suicide and violence in military veterans.

Patient, therapist, and system factors influencing the effectiveness of prolonged exposure for veterans with comorbid posttraumatic stress disorder and traumatic brain injury

Abstract In contrast to concerns that cognitive limitations and neurobehavioral symptoms (NBS) associated with traumatic brain injury (TBI) may inhibit treatment effectiveness, a recent study found prolonged exposure (PE) led to large reductions in posttraumatic stress disorder (PTSD) symptoms among Iraq-Afghanistan veterans with a range of TBI severity (article by Wolf, Kretzmer, Crawford, Thors, Wagner, Strom, Eftekhari, Klenk, Hayward, and Vanderploeg [J Trauma Stress 28:339–347, 2015]). We further examined this sample of 69 veterans to determine whether system, veteran, and therapist factors predicted clinically significant responses. Results of hierarchical, logistic regressions revealed that therapist training in PE and lower service connection were associated with increased odds of large decreases in PTSD symptoms after adjusting for the robust effect of PE sessions completed. Other patient-level factors including age, time since injury, and baseline NBS were unrelated to significant improvements. Findings emphasized the impact of PE dosage, indicated greater mastery of the protocol was beneficial, and showed that service connection could impede self-reported, clinically significant change during PE in this important cohort.