Jeffrey Sonis

Interventions to prevent post-traumatic stress disorder: a systematic review.

CONTEXT Traumatic events are prevalent worldwide; trauma victims seek help in numerous clinical and emergency settings. Using effective interventions to prevent post-traumatic stress disorder (PTSD) is increasingly important. This review assessed the efficacy, comparative effectiveness, and harms of psychological, pharmacologic, and emerging interventions to prevent PTSD. EVIDENCE ACQUISITION The following sources were searched for research on interventions to be included in the review: MEDLINE; Cochrane Library; CINAHL; EMBASE; PILOTS (Published International Literature on Traumatic Stress); International Pharmaceutical Abstracts; PsycINFO; Web of Science; reference lists of published literature; and unpublished literature (January 1, 1980 to July 30, 2012). Two reviewers independently selected studies, extracted data or checked accuracy, assessed study risk of bias, and graded strength of evidence. All data synthesis occurred between January and September 2012. EVIDENCE SYNTHESIS Nineteen studies covered various populations, traumas, and interventions. In meta-analyses of three trials (from the same team) for people with acute stress disorder, brief trauma-focused cognitive behavioral therapy was more effective than supportive counseling in reducing the severity of PTSD symptoms (moderate-strength); these two interventions had similar results for incidence of PTSD (low-strength); depression severity (low-strength); and anxiety severity (moderate-strength). PTSD symptom severity after injury decreased more with collaborative care than usual care (single study; low-strength). Debriefing did not reduce incidence or severity of PTSD or psychological symptoms in civilian traumas (low-strength). Evidence about relevant outcomes was unavailable for many interventions or was insufficient owing to methodologic shortcomings. CONCLUSIONS Evidence is very limited regarding best practices to treat trauma-exposed individuals. Brief cognitive behavioral therapy may reduce PTSD symptom severity in people with acute stress disorder; collaborative care may help decrease symptom severity post-injury.

Review and special articleInterventions to Prevent Post-Traumatic Stress Disorder: A Systematic Review

CONTEXT Traumatic events are prevalent worldwide; trauma victims seek help in numerous clinical and emergency settings. Using effective interventions to prevent post-traumatic stress disorder (PTSD) is increasingly important. This review assessed the efficacy, comparative effectiveness, and harms of psychological, pharmacologic, and emerging interventions to prevent PTSD. EVIDENCE ACQUISITION The following sources were searched for research on interventions to be included in the review: MEDLINE; Cochrane Library; CINAHL; EMBASE; PILOTS (Published International Literature on Traumatic Stress); International Pharmaceutical Abstracts; PsycINFO; Web of Science; reference lists of published literature; and unpublished literature (January 1, 1980 to July 30, 2012). Two reviewers independently selected studies, extracted data or checked accuracy, assessed study risk of bias, and graded strength of evidence. All data synthesis occurred between January and September 2012. EVIDENCE SYNTHESIS Nineteen studies covered various populations, traumas, and interventions. In meta-analyses of three trials (from the same team) for people with acute stress disorder, brief trauma-focused cognitive behavioral therapy was more effective than supportive counseling in reducing the severity of PTSD symptoms (moderate-strength); these two interventions had similar results for incidence of PTSD (low-strength); depression severity (low-strength); and anxiety severity (moderate-strength). PTSD symptom severity after injury decreased more with collaborative care than usual care (single study; low-strength). Debriefing did not reduce incidence or severity of PTSD or psychological symptoms in civilian traumas (low-strength). Evidence about relevant outcomes was unavailable for many interventions or was insufficient owing to methodologic shortcomings. CONCLUSIONS Evidence is very limited regarding best practices to treat trauma-exposed individuals. Brief cognitive behavioral therapy may reduce PTSD symptom severity in people with acute stress disorder; collaborative care may help decrease symptom severity post-injury.

Risk and Protective Factors for Recurrent Intimate Partner Violence in a Cohort of Low-Income Inner-City Women

The purpose of this study was to identify longitudinal predictors of any (versus no) episodes of recurrent intimate partner violence (IPV) and their severity among low-income inner-city women. A secondary analysis was conducted on data from an inception cohort of 321 previously abused women from the Chicago Women’s Health Risk Study. In a multivariable logistic regression model, pregnancy, frequency of IPV in the year prior to the baseline interview, and the partner’s use of power and control tactics increased the odds of recurrent IPV during the follow-up period and leaving an abusive partner reduced the odds. In a multivariate proportional odds logistic regression model, partner violence outside the home was associated with higher severity of recurrent IPV, but leaving an abusive partner was not. The results suggest that, for low-income women, leaving an abusive partner may reduce the risk of recurrent victimization without increasing severity of the recurrent attacks that do occur.

Teaching of human rights in US medical schools

Objective. —To determine the extent to which human rights issues are included in required bioethics curricula in US medical schools and to identify medical school characteristics associated with the extent of human rights issues covered. Design. —Cross-sectional survey. Participants. —Bioethics course directors and bioethics section directors of 125 US medical schools. Main Outcome Measure. —The extent of human rights teaching at each school was measured as the percentage of 16 human rights issues. Results. —Course directors at 113 (90%) of the 125 US medical schools responded to the survey. Medical schools included about half (45%; 95% confidence interval, 41%-49%) of 16 human rights issues in their required bioethics curricula. Domestic human rights issues, such as discrimination in the provision of health care to minorities (82% of medical schools), were covered much more frequently than international human rights issues, such as physician participation in torture (17% of schools). Public medical schools included substantially fewer human rights issues than private medical schools (F [1,112] =7.7; P Conclusions. —Required courses in medical education do not adequately address the medical aspects of human rights issues, especially international issues.

Association between duration of residence and access to ambulatory care among Caribbean immigrant adolescents.

OBJECTIVES The purpose of this study was to determine the association between duration of residence and access to ambulatory care among Caribbean immigrant adolescents. METHODS A cross-sectional survey of adolescents at a New York City high school was conducted. Multivariate modeling methods were used to adjust for demographic and socioeconomic covariates. RESULTS Duration of residence was strongly and directly associated with access to ambulatory care after adjustment for ethnicity. CONCLUSIONS Caribbean immigrant adolescents, particularly recent immigrants, have reduced access to ambulatory care.

PTSD in Primary Care—An Update on Evidence-based Management

Posttraumatic stress disorder (PTSD) is common in primary care but it is frequently not detected or treated adequately. There is insufficient evidence to recommend universal screening for PTSD in primary care, but clinicians should remain alert to PTSD among patients exposed to trauma, and among those with other psychiatric disorders, irritable bowel syndrome, multiple somatic symptoms and chronic pain. A two-stage process of screening (involving the PC-PTSD), and, for those with a positive screen, a diagnostic evaluation (using the PTSD-Checklist), can detect most patients with PTSD with few false positives. Evidence-based recommendations are provided for treatment in primary care or referral to mental health.

Posttraumatic Stress Disorder does not Increase Recurrent Intimate Partner Violence

ABSTRACT To determine whether posttraumatic stress disorder (PTSD) increases the risk of recurrent intimate partner violence, I conducted a secondary analysis of longitudinal data from 321 women in the Chicago Womens Health Risk Study. PTSD at baseline showed a moderate unadjusted association with recurrent IPV in the 2-year follow-up period. However, when important confounding factors, such as severity of previous abuse and use of control tactics by the partner, were controlled in a logistic regression model, there was no association between PTSD and recurrent victimization.

How to write an NIH R13 conference grant application.

ObjectiveThe purpose of this article is to provide recommendations for writing a successful R13 conference grant proposal for the National Institutes of Health (NIH).MethodsThe authors reviewed successful NIH conference grant proposal abstracts. They also reflect on their own experience in writing an NIH conference grant proposal and implementing a successful annual conference on research methods in the area of psychological trauma.ResultsThe key to a strong proposal is linkage among all of its sections, from the specific aims to the budget. The specific aims should be justified by the need for the conference and articulated in the background and significance section, and the aims, in turn, should drive the content and format of the conference.ConclusionConferences can be an important way to promote NIH scientific goals, by disseminating new findings, facilitating collaborations, and stimulating new lines of research.

Editorial: Innovations in trauma research methods, 2008

This issue of the Journal of Traumatic Stress contains a special section devoted to highlights from the fifth and final annual NIH-funded Conference on Innovations in Trauma Research Methods (CITRM), held in Evanston, Illinos, on November 16th and 17th, 2008. CITRM was designed to be a conference that focused on research methods, not content, in the area of psychological trauma. As at the four previous conferences, CITRM 2008 included presentations dealing with diverse aspects of the research process, including study design, sampling, measurement, ethics, and data analysis. The article by Newman and Kaloupek (this issue; pp. xxx-yyy) focuses on ethical aspects of designing and conducting trauma research studies. The authors concentrate on the basic ethical principles of autonomy (respecting individuals’ independence and protecting those individuals with diminished capacity) and beneficence (doing no harm and maximizing possible benefits and minimizing possible harms associated with research participation). It is noteworthy and comforting that their review of the literature indicates that most trauma research participants refer to their participation in positive terms and experience little, if any, distress from their participation. Equally noteworthy is that relatively few studies have empirically examined ethical aspects of trauma research. Thus, among the authors’ suggestions is for researchers to systematically collect data to enlarge the evidence base on which ethically informed research design decisions can be made. A specific recommendation is to study whether, and under what conditions, trauma-exposed individuals might have diminished capacity or be at heightened risk for emotional upset, and thus require special precautions that may affect recruitment procedures or study design. In 2008, the Institute of Medicine (IOM) of the National Academy of Sciences report on treatment of posttraumatic stress disorder concluded, to the surprise of many, that the research evidence was inadequate to determine the efficacy of any treatment for PTSD except for exposure therapies. The conclusion was based primarily on methodological limitations in the existing randomized clinical trials. Leon and Davis (this issue; pp. xxx-yyy) address some of the key shortcomings in PTSD treatment trials identified by the IOM panel and make thoughtful and useful suggestions for better practice. They discuss the problems created by research participant attrition, use of multiple outcomes, and inappropriate groups for comparison to active treatment. They recommend state-of-the-art methods to handle missing data (other than the problematic “last observation carried forward” method), adjust for multiple outcomes, and select a credible comparator. The article by Creswell and Zhang (this issue; pp. xxx-yyy) provides readers with a primer for the use of mixed methods designs in trauma research. Mixed methods is a methodological approach that involves the collection of qualitative and quantitative data within the context of a single study or well-designed series of studies. It is well-suited to either refine and better understand an established construct such as PTSD or begin the process of identifying factors for consideration in a newly studied construct. The authors first define mixed methods studies and provide an overview of four core features. They address strategies for combining or integrating the findings from qualitative and quantitative study components and go on to describe four mixed methods designs that systematically vary whether quantitative data are collected prior to, following, or concurrent with the collection of qualitative data. The authors identify an example of a mixed methods study in the trauma literature and provide the reader with five alternatives for potential follow-up studies. The article by Suvak, Walling, Iverson, Taft, and Resick (this issue; pp. xxx-yyy) is centered on the use of multilevel regression for the analysis of longitudinal data. Multilevel regression is ideal for such repeated-measures data because it enables the researcher to consider associations between variables within persons as between-persons or individual differences characteristics. These characteristics, in turn, can be associated with other between-persons variables. Using longitudinal data from a large sample of rape and robbery victims, Suvak et al. begin by presenting a simple model for the relation between symptoms of intrusion and avoidance. They next develop the logic for a growth curve variation, wherein separate models document the within-person associations between time (since trauma) and scores on intrusion and avoidance. Finally, the authors combine the separate growth curve representations into a composite model, thereby expanding the usual univariate growth curve approach to a multivariate approach and directly addressing individual differences in and the relation between change in intrusion and change in avoidance. Keane (this issue; pp. xxx – yyy) closes this special section with thoughts on the dual importance of both substantive conceptualization and methodological advancements to the enhancement of knowledge of psychological trauma and its consequences. He also provide a brief retrospective on the contributions of the CITRM conference series and a call for continuing development of new methods and technologies and their dissemination to the community of trauma researchers. Although enthusiasm for research methods remains strong among trauma researchers, CITRM will not be continuing. CITRM was funded for five years by the National Institute of Mental Health and we have used that funding to hold five successful conferences. (Please see our related article1 for suggestions on how to write a National Institutes of Health conference grant proposal). CITRM developed as an outgrowth of activities of the Research Methods Special Interest Group (RM-SIG) of ISTSS more than seven years ago. Now that CITRM has ended, the RM-SIG will again serve as a primary vehicle for advancing research methods among trauma researchers. The Co-Chairs of the RM-SIG, Dean Lauterbach and Patrick Palmieri, invite suggestions for methods-related programming at annual ISTSS meetings, and encourage membership and active participation in the RM-SIG to help develop novel ways to disseminate and make use of research methods to advance trauma research.

Access and Outcomes of Obstetric Care

To The Editor: The article by Larimore and Davis l is an important attempt to link access to maternity services (as measured by physician availability) to an important health outcome-infant mortality. There are several methodological and statistical problems in the study, however, that undermine the validity of the results and conclusions. The fundamental problem of their study is that the unique biases that can affect ecologic studies such as theirs are not even addressed. First, in contrast to control of confounding in individual-level studies, attempts to control confounding in ecologic studies rarely eliminate confounding.2•3 The inclusion of such variables as percentage of nonwhite study population, education, and income into the linear regression model does not mean that the association between INDEX (the indicator of physician availability) and infant mortality is not confounded by these variables. Second, even small errors in the measurement of covariates can result in a profound bias in an ecologic analysis, and this bias can produce effects vastly different from the effects introduced in individual-level studies.4 Income, for instance, is probably measured with some degree of error, and this measurement error might have a substantial effect on the regression coefficient for INDEX. (It is impossible to determine the magnitude and direction of this bias without analyzing individual-level data.) Further, the statistical considerations relevant to ecologic analyses are ignored. Correlation coefficients and, therefore R2, are spuriously inflated in ecologic analyses relative to individual-level studies.s The magnitude of this difference can be profound. Morganstern,S for instance, offers an example in which data that were analyzed at the individual level resulted in an R2 of 0.0 I, but when they were analyzed ecologically, the R2 was 1.00. S Despite the hazard of using correlation coefficients (and R2), the authors use the R2 for their linear regression model as their primary outcome measure. The reported R2 of 0.176 is almost certainly spuriously high. If the authors had used the regression coefficient for INDEX as their primary outcome measure (because regression coefficients are not falsely elevated in ecologic analyses), their conclusions would have been vastly different. INDEX, for instance, shows the weakest association with infant mortality rate of any variable studied, approximately 500 times weaker than the association between percentage of the nonwhite study population and infant mortality. These criticisms are not intended as a broadside against ecologic studies in general. \Vhile ecologic studies can provide valid estimates of individual-level effects under certain very limited conditions, the effect of unique biases must be evaluated and appropriate statistical methods must be used before any conclusions can be drawn. It is difficult to draw any valid conclusions from Larimore and Daviss study because they did neither. They must be congratulated for attempt-