1

Introduction

BACKGROUND

This report summarizes the presentations and discussions at the Workshop on Integrating New Measures of Trauma into the Substance Abuse and Mental Health Services Administration’s (SAMHSA’s) Data Collection Programs, which was held in Washington, D.C., in December 2015. The workshop was organized as part of an effort to assist SAMHSA and the Office of the Assistant Secretary for Planning and Evaluation (ASPE) of the U.S. Department of Health and Human Services in their responsibilities to expand the collection of behavioral health data in several areas. The workshop was structured to bring together experts in the measurement of exposure to traumatic events, the measurement of posttraumatic stress disorder (PTSD), and health survey methods to facilitate discussion of measures and mechanisms most promising for expanding SAMHSA’s data collections in this area.

The overall effort is being overseen by the Standing Committee on Integrating New Behavioral Health Measures into the Substance Abuse and Mental Health Services Administration’s Data Collection Programs.¹ In addition to the topics covered by this workshop, SAMHSA and ASPE are interested in expanding data collection on serious emotional disturbance in children, on specific mental illness diagnoses with functional impairment, and on recovery from substance use or mental disorder.

___________________

¹ For a description of the overall study, see http://sites.nationalacademies.org/DBASSE/CNSTAT/Behavioral_Health_Measures_Committee/index.htm [April 2016].

Workshops on all four topics were held as part of the overall effort, and reports from the first two workshops have been published.^2,3

WORKSHOP FOCUS

At the beginning of the workshop, Neil Russell of SAMHSA described the agency’s goals in exploring how to best measure and expand SAMHSA’s data collection programs to include measures of exposure to traumatic events and PTSD, which can have a profound impact on people’s lives. Studies have found associations between exposure to a traumatic event and a wide range of negative outcomes: substance use and dependence; depression, anxiety, and conduct problems; schizophrenia and personality disorders; PTSD and acute stress disorder; poorer psychological response to subsequent traumatic event exposure; and suicide. SAMHSA would like to better understand how these outcomes occur in order to connect people who are impacted with treatments that can facilitate recovery.

Russell said that the descriptions of traumatic event exposure first appeared in the third edition of the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders (DSM-III), when PTSD was added as a mental disorder (trauma-related disorders had previously been listed under other diagnoses). With the introduction of the DSM-IV, emphasis was placed on defining a traumatic event by the event itself rather than through people’s reactions to the event and on defining a traumatic event as an event that involved actual or threatened death or serious injury or a threat to the physical integrity of the person or others. A PTSD diagnosis required that the event result in feelings of intense fear, helplessness, or horror and that the person also meet criteria in several other symptom categories, such as re-experiencing the event, avoidance, arousal, duration of at least 1 month, and associated functional impairment.

The fifth edition of the DSM (DSM-5) brought about several further

___________________

² National Academies of Sciences, Engineering, and Medicine. (2016). Measuring Serious Emotional Disturbance in Children: Workshop Summary. K. Marton, Rapporteur, Committee on National Statistics and Board on Behavioral, Cognitive, and Sensory Sciences, Division of Behavioral and Social Sciences and Education. Board on Health Sciences Policy, Institute of Medicine. Washington, DC: The National Academies Press.

³ National Academies of Sciences, Engineering, and Medicine. (2016). Measuring Specific Mental Illness Diagnoses with Functional Impairment: Workshop Summary. J.C. Rivard and K. Marton, Rapporteurs. Committee on National Statistics and Board on Behavioral, Cognitive, and Sensory Sciences, Division of Behavioral and Social Sciences and Education. Board on Health Sciences Policy, Institute of Medicine. Washington, DC: The National Academies Press.

changes, Russell noted. In this edition, PTSD was moved from the category of anxiety disorders into a new category of trauma and stressor-related disorders. Symptoms were divided into four clusters: intrusion, avoidance, negative alterations in cognitions and mood, and alterations in arousal and reactivity. This change involved separating the DSM-IV avoidance and numbing criterion into two criteria: avoidance and negative alterations in cognitions and mood, as well as adding the requirement of at least one avoidance symptom for a PTSD diagnosis. Three new symptoms were also added in the DSM-5: persistent and distorted blame of self or others, persistent negative emotional state, and reckless or destructive behavior. One criterion was removed: fear, helplessness, or horror right after the trauma. The DSM-5 also revised several symptoms to clarify symptom expression.

Russell next turned to discussing the work SAMHSA has done in this area. One of the strategic initiatives on trauma and justice called for the development of a surveillance strategy for trauma and its association with mental and substance use disorders. As part of this initiative, SAMHSA began thinking about ways to obtain national estimates of exposure to trauma and posttraumatic stress symptoms, including subclinical and clinical PTSD, initially still based on the DSM-IV definition. Determining associations between trauma events and PTSD symptoms, as well as mental health and substance use problems, was also part of this initiative.

One of SAMHSA’s projects focused on the concept of trauma—including trauma exposure and trauma or posttraumatic stress response components—through three “Es”: event, experience of event, and effect. The event is the actual or extreme threat of physical or psychological harm or, for a child, severe, life-threatening neglect that imperils healthy development. The experience of the event is whether the event is traumatic, in other words, how the individual labels, assigns meaning to, and is physically and psychologically disrupted by an event. The effect is the adverse experience resulting from trauma exposure that may occur acutely, immediately after the event, or have a delayed onset.

Russell briefly described the National Survey on Drug Use and Health (NSDUH), SAMHSA’s existing survey on substance use and mental health. The survey has been conducted since 1990, and it involves approximately 67,000 interviews completed annually, with respondents aged 12 and over, sampled from the household (non-institutionalized) population in the United States. The interviews are conducted at respondents’ homes by trained lay interviewers and, for the sensitive portions of the survey, through an audio computer-assisted self-interview. In addition to questions about substance use and mental health, the NSDUH collects data on physical health conditions and demographic characteristics, including the respondent’s age, race, and veteran status. The questionnaire is available

in English and Spanish, but almost all of the interviews (around 96%) are completed in English. The survey is in the field practically every day of the year with approximately 600 interviewers, across all states. Russell underscored that the NSDUH is a very large undertaking.

Because legislation (Public Law No. 102-321) requires the estimation of serious mental illness by state, SAMHSA developed conceptual and operational definitions of serious mental illness and a methodology for producing estimates. The approach for producing the estimates relies on the Mental Health Surveillance Study (MHSS), which was a follow-on to the NSDUH conducted between 2008 and 2012. The MHSS involved a clinical interview administered to a nationally representative subset of the NSDUH respondents (500-700 people annually) a few weeks after they completed the NSDUH interview. This sample was limited to adults and conducted in English, by telephone. The primary goal was to produce overall model-based estimates of serious mental illness, but the survey also enabled SAMHSA to make estimates of specific mental disorders.

Russell said that the MHSS was not designed to measure exposure to potentially traumatic events and PTSD, but the survey contained measures that enabled SAMHSA to produce some estimates on these topics (which will be discussed later in the workshop). Despite its current limitations in terms of data on exposure to potentially traumatic events and PTSD, Russell said that the MHSS design is important to understand because it illustrates one possible approach for collecting data on a focused topic as a follow-on to an existing large-scale survey.

Russell noted that SAMHSA faces a number of challenges as it attempts to expand its data collection in this area. There is a need to differentiate between trauma exposure and the potential outcome of trauma exposure (e.g., PTSD). Responses to traumatic events may differ across the lifespan; by ethnicity, race, and culture; and by a person’s role in the event. In addition, individuals have different levels of resiliency, that is, ability to tolerate traumatic events.

Another challenge, said Russell, is that predicting the trajectory of symptom development and potential severity may not be possible on the basis of a simple “yes” or “no” question about traumatic exposure. The type of traumatic event, the intensity of the event, and the setting of the event can all influence the effects of the exposure, and all of them may need to be ascertained as part of the data collection.

Russell discussed several approaches and issues that SAMHSA will need to consider for collecting information on trauma exposure. One option might be to instruct respondents to self-report events in their own words and then analyze those responses after the data are collected. If the questionnaire is to be administered by an interviewer, a decision has to be made about whether to use lay interviewers or clinical interviewers.

If lay interviewers are used, a standardized set of question and answer choices could be administered, and the event would be classified as traumatic depending on the respondent’s understanding of what a traumatic event entails. By contrast, a clinical interviewer could ask a series of questions aimed at collecting detailed information and then make a judgment based on that information about whether the event should be classified as traumatic.

Russell said that less expensive, brief screening methods could also be considered. The existing screening instruments vary greatly in how they assess traumatic events and the resulting posttraumatic stress symptoms, and SAMHSA would be interested to know if any of them are suitable to meet the agency’s goals.

Another challenge associated with producing nationally representative estimates of exposure to trauma and its effects is that some of the populations most affected are not easily captured in typical household surveys, such as the NSDUH. These populations include active-duty military personnel (regardless of where they reside), people in jails or prisons, homeless people, youth living in foster care, and people in institutionalized settings.

SAMHSA would face additional challenges if the estimates are to include children. Some traumatic events are sensitive and difficult to assess in an interview, even with adult respondents. Asking children and adolescents about these issues would be particularly difficult, and special consent procedures might be required if including children is deemed important.

Russell said that SAMHSA is looking for input on the issues and challenges described. From the perspective of the Center for Behavioral Health Statistics and Quality, the center at SAMHSA that commissioned the study, the goal for expanding data collections on these topics is to understand the association between outcomes of exposure to trauma, mental health, and substance use. The agency would like to think more broadly than just PTSD and include other outcomes of trauma exposure. Other key covariates of interest for this research include language spoken, race and ethnicity, gender, age, education, income, medical conditions, and health insurance status. SAMHSA would like to be able to produce national estimates approximately every 3-5 years. This schedule means that it would be possible to consider a design similar to that of the MHSS, which involves pooling data from several years’ worth of interviews in order to produce some of the estimates.

Russell described several data collection strategies that SAMHSA has considered for producing estimates of trauma exposure and outcomes. One option would be to add questions directly to the NSDUH, but he emphasized that it would be important to maintain the average NSDUH

administration time at around 60 minutes, primarily because a longer survey could adversely affect response rates. In other words, if new questions are added to the NSDUH, some of the existing questions would likely have to be dropped. Another approach would be to reinstate the MHSS or develop a similar survey to collect data from a subsample of the NSDUH respondents in a follow-on interview.

A third option would be to develop a new, stand-alone data collection. This approach would be expensive, but it might be necessary if neither the NSDUH nor a follow-on to the NSDUH is deemed to be a suitable mechanism for collecting the data of interest. A fourth possibility would be to identify an existing source of national data that could be used to produce estimates of trauma. Russell said that SAMHSA has conducted some research to identify existing data sources, and none seems suitable for the agency’s current goal, but they are looking for further input on potential sources of data.

Russell concluded by saying that input from the workshop participants would be particularly useful on several key issues: how to measure exposure to potentially traumatic events and the outcomes of these events; survey and questionnaire design tradeoffs; mechanisms for collecting data; and the impact of potential changes to NSDUH. Since some of the possible approaches discussed could involve model-based estimation procedures, the agency would also appreciate guidance on these types of methods.

Larke Huang, who leads SAMHSA’s strategic initiative on trauma and justice, provided additional background on the agency’s interest in trauma. She said that trauma was one of the areas that the previous SAMHSA administrator wanted the agency to focus on from a programmatic and policy perspective. SAMHSA would like to have a solid foundation in this area, grounded in research and data. As a first step, the agency needs to crystalize its thinking about the concept of trauma outcomes, beyond just PTSD, and determine how to gain a better understanding of the connections between exposure to traumatic events and areas that SAMHSA is mandated to address, including mental health, substance use disorders, and other conditions, such as HIV. One of the challenges associated with measuring these concepts is that different studies show different prevalence rates, depending on the definitions used, and Huang noted that SAMHSA also wrestles internally with definitional issues. SAMHSA wants to focus on work that has the potential of translating research into policy. The agency would like guidance on how different research methodologies can be used to inform and advance critical programs.

WORKSHOP CHARGE

The specific statement of task for the workshop (shown in Box 1-1) was developed on the basis of the charge for the overall project, which was to expand data collections on several behavioral health topics. The main goals of the workshop were to discuss options for collecting data and producing estimates on exposure to traumatic events and PTSD, including available measures and associated possible data collection mechanisms.

ORGANIZATION OF THE REPORT

This summary describes the workshop presentations and the discussions that followed each topic: see the workshop agenda in Appendix A. Biographical sketches of the steering committee members and speakers are in Appendix B.

Chapter 2 covers existing studies and data, including the trauma module that was included in the MHSS (see above) and other national surveys that have collected data on this topic. Chapter 3 discusses the key concepts relevant in the context of measuring exposure to potentially traumatic events, PTSD, and other outcomes. The chapter also discusses the measures that are currently available. Chapter 4 focuses on issues specific to measuring trauma exposure and its effects in children and adolescents. The workshop participants’ discussions of the key themes and possible next steps for SAMHSA are summarized in Chapter 5.

This report has been prepared by the workshop rapporteur as a factual summary of what occurred at the workshop. The steering committee’s role was limited to planning and convening the workshop. The views

contained in the report are those of individual workshop participants and do not necessarily represent the views of all workshop participants, the steering committee, or the National Academies of Sciences, Engineering, and Medicine.