Jeri E. Forster, PhD is the Statistical Core Director at the VA VISN 19 MIRECC. In concert with Dr. Brenner, she determines the statistical needs of the MIRECC and delegates projects. Along with her colleague, Dr. David Weitzenkamp, she assists with guidance on study design; sample size and power calculations; provision of written analytical plans describing methods and appropriate statistical tests; data analysis of completed studies; and assistance with manuscript preparation for publication.
Dr. Forster is also an Assistant Professor in the Department of Pediatrics, School of Medicine with a secondary appointment in the Department of Biostatistics and Informatics, Colorado School of Public Health. Within the MIRECC, Dr. Forster is actively involved in numerous research projects related to suicidality, TBI and PTSD and is also the PI (principal investigator) on a 3 year NIH National Institute of Drug Abuse R01 which aims to develop and evaluate new statistical methods to address the problem of nonignorable dropout in longitudinal clinical trials and cohort studies. These methods will additionally be applied to three existing databases to help answer clinical research questions.
Research Projects
Assessment of Cognitive Functioning as it Relates to Risk for Suicide in Veterans with HIV/AIDS
The purpose of this study is to evaluate the potential utility of neuropsychological testing as a key component of suicide risk assessment in Veterans with HIV/AIDS.
Medication overdoses account for substantial numbers of suicide-related behaviors. Non-adherence is a significant issue for those with psychiatric illness.
The purpose of this project is to explore the degree to which performance consistency on neuropsychological measures varies in a sample of Operation Iraqi Freedom (OIF)/Operation Enduring Freedom (OEF) Veterans with a history of mild traumatic brain injury (mTBI) with persistent self-reported symptoms.
Losing a close friend or family member to death can be an incredibly painful and stressful occurrence. Some bereaved persons may experience additional symptoms associated with a severe and unending type of grief now referred to as prolonged grief (PG). The connection between PG and suicide among Veterans has not been studied, even though Veterans already are at a higher risk for suicide than are other groups.
Through this research we hope to understand which values (e.g. power, achievement, security) and value dimensions (e.g., importance, success) have the strongest impact on suicidality in this population.
The proposed investigation is a pilot study intended to establish the safety and tolerability of lithium augmentation of standard psychopharmacological treatment of PTSD for combat veterans seeking treatment at the Denver VA Medical Center.
Military personnel are returning from current conflicts with traumatic brain injury (TBI). Brief and sensitive screening measures are needed to assess TBI in this population.
The purpose of this study is to examine the effect of PTSD on perceptions and beliefs about injury in Veterans with a history of combat mTBI or non-TBI injury.
This proposed project will be the first to replicate a groundbreaking psychological treatment for suicide prevention among those with moderate or severe traumatic brain injury, Window to Hope (WtoH), developed by PI Simpson and colleagues in Sydney, Australia. WtoH has been successfully evaluated in a Randomized Controlled Trial (RCT), in which treated Australian civilians (n=17) with severe TBI recorded significant decreases in hopelessness. The current project aims to (i) undertake the cross-cultural adaptation of the WtoH program; (ii) establish the acceptability and feasibility of WtoH within the VAMC context; and (iii) conduct a RCT within the VAMC to replicate the results from the original trial (efficacy). Deliverables are expected to include an intervention suitable for a larger trial and broader dissemination.
Bahraini NH, Gutierrez PM, Harwood JEF, Huggins JA, Hedegaard H, Chase M, Brenner LA. (2012). The Colorado Violent Death Reporting System (COVDRS): Validity and Utility of the Veteran Status Variable. Public Health Rep., May/Jun;127(3):304-309.
Objective. Using the Veterans Affairs Beneficiary Identification Record Locator Subsystem (BIRLS) as the criterion database, we evaluated the sensitivity and specificity of the death certificate information in the Colorado Violent Death Reporting System (COVDRS) to determine Veteran status for those who died by suicide. Methods. The study sample consisted of 3,820 individuals aged 18 years and older who died by suicide in Colorado from January 1, 2004, through December 31, 2008. To determine agreement on Veteran status, COVDRS data were submitted to the Veterans Benefits Administration for linkage to the BIRLS using Social Security numbers. Results. Sensitivity and specificity of the Veteran status information on the death certificate were 93.1% (95% confidence interval [CI] 90.7, 95.2) and 91.7% (95% CI 90.5, 92.8), respectively. The overall agreement between the death certificate and the BIRLS on Veteran status was very good (kappa 5 0.76; 95% CI 0.74, 0.79). Conclusions. This study of 3,820 suicide deaths in Colorado demonstrated a high level of agreement between the COVDRS Veteran status variable and the BIRLS. Such findings offer support for using the COVDRS in studying factors associated with suicide in the Veteran population.
Brenner, L. A., Betthauser, L.M., Homaifar, B. Y., Villarreal, E., Harwood, J. E., Staves, P. J., & Huggins, J. A. (2011). Posttraumatic stress disorder, traumatic brain injury, and suicide attempt history among Veterans receiving mental health services. Suicide & Life Threatening Behavior, 41, 416-423.
History of posttraumatic stress disorder (PTSD) or traumatic brain injury (TBI) has been found to increase risk of suicidal behavior. The association between suicide attempt history among veterans with PTSD and/or TBI was explored. Cases (N = 81) and 2:1 matched controls (N = 160) were randomly selected from a Veterans Affairs Medical Center clinical database. PTSD history was associated with an increased risk for a suicide attempt (OR = 2.8; 95% CI: 1.5, 5.1). This increased risk was present for those with and without a history of TBI. Results support incorporating PTSD history when assessing suicide risk among veterans with and without TBI.
Brenner, L.A., Breshears, R.E., Betthauser, L.M., Bellon, K. K., Holman, E., Harwood, J. E., Silverman, M. M., Huggins, J. & Nagamoto, H. T. (2011). Implementation of a suicide nomenclature within two VA healthcare settings. Journal of Clinical Psychology in Medical Settings.
Implementation of a Suicide Nomenclature within Two VA Healthcare Settings Those who work in the field of Veteran’s care, as well as educators, researchers, and professionals providing direct mental health services agree that learning more about and preventing suicide represents a highly critical goal. Yet, up to now, researchers and mental health professionals lacked a shared language for defining suicidal behavior. This study discusses implementation of the Center for Disease Controls’ Self-Directed Violence Classification System (SDVCS) and an accompanying Clinical Tool (CT) at two VA healthcare facilities (in Denver and Grand Junction, CO). Results of this study show that implementing a more unified language is possible, while at the same time highlights some of the challenges and barriers to adoption of this system. This study provides important information regarding implementation of the SDVCS throughout the VA system.
OBJECTIVES:: To assess the prevalence of traumatic brain injury (TBI) among Veterans seeking mental health services using a 4-item tool, the Traumatic Brain Injury-4 (TBI-4), and to establish the classification accuracy of the TBI-4 using the Ohio State University TBI-Identification Method as the criterion standard. STUDY DESIGN:: Archival and observational data collected from individuals seeking care at a Mountain State VA Medical Center. PARTICIPANTS:: The sample for the archival study was 1810. Three hundred sixteen Veterans completed observational study measures. MAIN MEASURES:: For the archival study, TBI-4 and demographic data extracted from electronic medical records. For the observational study, the Ohio State University TBI-Identification Method and a demographic questionnaire were used. TBI-4 data were also obtained from electronic medical records. RESULTS:: The prevalence of probable TBI among those seeking VA MH treatment was 45%. Sensitivity and specificity of the TBI-4 were 0.74 and 0.56, respectively. Veterans with all levels of TBI severity sought care within this VA mental health setting. CONCLUSIONS:: The prevalence of TBI in this VA mental health treatment population was higher than expected. Additional research is required to assess the clinical utility of screening for TBI among this population of Veterans.
This exploratory study was conducted to increase understanding of neuropsychological test performance in those with blast-related mild traumatic brain injury (mTBI). The two variables of interest for their impact on test performance were presence of mTBI symptoms and history of posttraumatic stress disorder (PTSD). Forty-five soldiers postblast mTBI, 27 with enduring mTBI symptoms and 18 without, completed a series of neuropsychological tests. Seventeen of the 45 met criteria for PTSD. The Paced Auditory Serial Addition Test (Frencham, Fox, & Mayberry, 2005; Spreen & Strauss, 1998) was the primary outcome measure. Two-sided, 2-sample t tests were used to compare scores between groups of interest. Presence of mTBI symptoms did not impact test performance. In addition, no significant differences between soldiers with and without PTSD were identified. Standard neuropsychological assessment may not increase understanding about impairment associated with mTBI symptoms. Further research in this area is indicated.
Breshears, R. E., Brenner, L. A., Harwood, J. E. F., & Gutierrez, P. M. (2010). Predicting suicidal behavior in Veterans with traumatic brain injury: The utility of the Personality Assessment Inventory. Journal of Personality Assessment, 92,349-355.
In this study, we investigated the Personality Assessment Inventory's (PAI; Morey, 1991, 2007) Suicide Potential Index (SPI) and Suicide Ideation scale (SUI) as predictors of suicidal behavior (SB) in military Veterans with traumatic brain injury (TBI; N = 154). We analyzed electronic medical records were searched for SB in the 2 years post-PAI administration and data via logistic regressions. We obtained statistical support for the SPI and SUI as predictors of SB. Analyses we performed using receiver operating characteristics suggested an optimal SPI cutoff of > or = 15 for this sample. Findings suggest that SPI and SUI scores may assist in assessing suicide risk in those with TBI, particularly when population-based cutoffs are considered.
Forster JE, MaWhinney S, Ball E, and Fairclough D. (2012) A varying-coefficient method for analyzing longitudinal clinical trials data with nonignorable dropout. Contemporary Clinical Trials, 33:378-85.
Dropout is common in longitudinal clinical trials and when the probability of dropout depends on unobserved outcomes even after conditioning on available data, it is considered missing not at random and therefore nonignorable. To address this problem, mixture models can be used to account for the relationship between a longitudinal outcome and dropout. We propose a Natural Spline Varying-coefficient mixture model (NSV), which is a straightforward extension of the parametric Conditional Linear Model (CLM). We assume that the outcome follows a varying-coefficient model conditional on a continuous dropout distribution. Natural cubic B-splines are used to allow the regression coefficients to semiparametrically depend on dropout and inference is therefore more robust. Additionally, this method is computationally stable and relatively simple to implement. We conduct simulation studies to evaluate performance and compare methodologies in settings where the longitudinal trajectories are linear and dropout time is observed for all individuals. Performance is assessed under conditions where model assumptions are both met and violated. In addition, we compare the NSV to the CLM and a standard random-effects model using an HIV/AIDS clinical trial with probable nonignorable dropout. The simulation studies suggest that the NSV is an improvement over the CLM when dropout has a nonlinear dependence on the outcome.
Homaifar BY, Brenner LA, Forster JE, Nagamoto H. Traumatic brain injury, executive functioning, and suicidal behavior: A brief report. Rehabil Psychol. 2012 Nov;57(4):337-41.
Objective: The aim of this pilot study was to explore the relationship between executive dysfunction and suicidal behavior in two groups of participants: (Group 1, n = 18) veterans with traumatic brain injury (TBI) and a history of at least one suicide attempt (SA), and (Group 2, n = 29) veterans with TBI and no history of SA. Controlling for the severity of TBI, it was hypothesized that participants in Group 1 would perform more poorly than those in Group 2 on measures of executive functioning. Design: The primary outcome variable was decision making as assessed by performance on the Iowa Gambling Task (IGT). Secondary outcome variables included laboratory-measured impulsivity as measured by the Immediate and Delayed Memory Test (IMT/DMT), abstract reasoning as measured by the Wisconsin Card Sorting Test (WCST), and aggression as measured by the Lifetime History of Aggression (LHA) scale. Results: Among those in Group 1, time between TBI and first suicide attempt postinjury varied widely (months to nearly 30 years). Only the WCST perseverative errors score differed significantly between individuals with and without histories of one or more suicide attempts (SAs). Conclusion: Suggestions for future study of SA among those with TBI are provided. When working with individuals with TBI, clinicians are encouraged to incorporate suicide risk assessment into their practice. Augmenting this process with a measure of perseveration may be beneficial. (PsycINFO Database Record (c) 2012 APA, all rights reserved).
Olson-Madden, J., Brenner, L.A., Harwood, J. E., Emrick, C. D., Corrigan, J. D., & Thompson, C. (2010). Traumatic brain injury and psychiatric diagnoses in Veterans seeking outpatient substance abuse treatment. Journal of Head Trauma and Rehabilitation, 25(6), 470-479.
OBJECTIVES: Explore the incidence of traumatic brain injury (TBI) in veterans seeking outpatient substance abuse treatment and the association between TBI and psychiatric diagnoses. MAIN MEASURE: The Ohio State University TBI identification method (OSU TBI-ID) was administered to veterans with positive TBI-4 screens; substance-related and psychiatric diagnoses were extracted from the medical record. PARTICIPANTS: : Over an 18-month period, 247 veterans completed the TBI-4. Of the 136 who screened positive, 70 were administered the OSU TBI-ID. RESULTS: On the basis of the TBI-4, 55% (95% CI: 49%-61%) of veterans screened positive for a history of TBI. The OSU TBI-ID was used to confirm screening results. Those who completed the OSU TBI-ID sustained an average of 3.4 lifetime TBIs. For each additional TBI sustained, after initial injury, there was an estimated 9% increase in the number of psychiatric diagnoses documented (99% CI: 1%-17%). For each additional documented psychiatric diagnosis, there was an estimated increase of 11% in the number of injuries sustained (99% CI: 1%-22%). Also, 54% (38/70) had a positive history of TBI prior to adulthood. CONCLUSION: These results emphasize the need for TBI screening in this vulnerable population, as well as the importance of increasing brain injury awareness among those abusing substances and their care providers. These findings also highlight the need for specialized services for those with TBI and co-occurring substance misuse aimed at decreased future TBIs or negative psychiatric outcomes or both. Further study is needed to clarify best practices.
Olson-Madden J, Forster JE, Huggins J, Schneider A. (2012). Psychiatric Diagnoses, Mental Health Utilization, High-Risk Behaviors, and Self-Directed Violence Among Veterans With Comorbid History of Traumatic Brain Injury and Substance Use Disorders. Journal of Head Trauma Rehabilitation. Volume 27(5), September/October 2012, p 370–378
Objectives: To describe various characteristics of veterans with co-occurring histories of traumatic brain injury (TBI) and substance use disorder (SUD) for purposes of hypothesis generation. Study Design: Archival data collected over a period of 4 years. Participants: Sixty-five veterans across eras of service with confirmed histories of TBI and SUD. Methods: Demographic and TBI information were obtained from an archival clinical database. Electronic medical records were reviewed for mental health utilization, psychiatric diagnoses, self-directed violence, and risk-taking behaviors. Results: In addition to a SUD, veterans were reported to have an average of 3 additional psychiatric diagnoses and a median of 3 TBIs per person. All utilized various mental health services in addition to substance use treatment. Individuals were found to have engaged in a variety of risky behaviors. There were significant associations between suicidal ideation and assaultive behaviors, as well as between suicide attempt and impulsivity. Conclusions: This study describes a sample of veterans with co-occurring histories of TBI, SUD, risk-taking behaviors, and self-directed violence. More research is needed to examine these complex interrelationships and to identify specific risk factors for intervention/prevention strategies.
Terrio, H. P., Nelson, L. A., Betthsuser, L. M., Harwood, J. E., & Brenner, L.A. (2011). Postdeployment traumatic brain injury screening questions: Sensitivity, specificity, and predictive values in returning soldiers. Rehabilitation Psychology, 56(1), 26-31.
OBJECTIVE: To evaluate the sensitivity, specificity, and predictive values of Post-Deployment Health Assessment traumatic brain injury (TBI) screening questions employed by the Department of Defense (DOD). Participants: Complete data was obtained from 3,072 soldiers upon return from a 15-month deployment to Iraq. METHOD: Comparisons were made between responses to the DOD four-item screener and a brief structured clinical interview for likely deployment-related TBI history. The interview process was facilitated using responses to the Warrior Administered Retrospective Casualty Assessment Tool (WARCAT). RESULTS: The sensitivity and specificity of the DOD screening tool (positive response to all four items) in comparison to the clinician-confirmed diagnosis was 60% and 96%, respectively. The sensitivity increased to 80%, with a slight decrease in specificity to 93%, for positive TBI screening when affirmative responses to questions 1 and 2 only were included. CONCLUSIONS: Affirmative responses to questions 1 and 2 of the DOD TBI screening tool demonstrated higher sensitivity for clinician-diagnosed deployment-related TBI. These two items perform better than positive responses to all four questions; the criteria presently being used for documentation and referral of a deployment-related TBI. These findings support further exploration of TBI screening and assessment procedures.
Surrogate endpoints frequently substitute for rare outcomes in research. The ability to learn about completed suicides by investigating more readily available and proximate outcomes, such as suicide attempts, has obvious appeal. However, concerns with surrogates from the statistical science perspective exist, and mounting evidence from psychometric, neurochemical, genetic, and neuroimaging studies suggests that surrogates may be particularly problematic in suicide research. The need for greater phenotypic refinement of suicide-related behaviors, development of and adherence to a shared suicide nomenclature, and conservative interpretation of investigational results that are limited to the precise population and suicide-related behavior under examination are discussed.
