Ms. Donahue, Ms. Lanzara, and Mr. Felton are with the Center for Information Technology and Evaluation Research, New York State Office of Mental Health, Albany. Dr. Essock is with the Department of Psychiatry, Division of Health Services Research, Mount Sinai School of Medicine, New York City, and with the Mental Illness Research, Education, and Clinical Center, James J. Peters Veterans Affairs Medical Center, the Bronx. Dr. Carpinello is the commissioner of the New York State Office of Mental Health. Send correspondence to Dr. Essock, Department of Psychiatry, Mount Sinai School of Medicine, Box 1230, 1 Gustave L. Levy Place, New York, NY 10029-6574 (e-mail: email@example.com). This report from Project Liberty is part of a special issue of Psychiatric Services commemorating the five-year anniversary of the September 11, 2001, attacks. Susan M. Essock, Ph.D., served as guest editor.
Immediately after the September 11, 2001, terrorist attacks on the World Trade Center, President Bush declared a federal disaster area in New York City and the ten surrounding New York counties because they contained large populations that commute to Manhattan to work. The declaration made these areas eligible for a range of Federal Emergency Management Agency (FEMA) programs, including the Crisis Counseling Assistance and Training Program, which supports short-term interventions for individuals and groups experiencing psychological sequelae to large-scale disasters.
As the state mental health authority, the New York State Office of Mental Health (NYOMH) administered and oversaw the design, implementation, and evaluation of Project Liberty, New York's FEMA-funded crisis counseling program that was created to respond to the aftermath of the attacks of September 11, 2001. The agency sought and was awarded more than $155 million in federal funding for Project Liberty ($137 million was ultimately spent), making it the most highly funded crisis counseling program in FEMA's history. NYOMH created an infrastructure that made it possible for local mental health authorities and provider agencies to successfully meet the mental health needs in their communities that were associated with the attacks within the goals of FEMA's crisis counseling program. As a result Project Liberty was a successful collaboration between NYOMH, local governments, and nearly 200 local agencies that provided service delivery throughout the declared disaster area.
Project Liberty's overall goal was to alleviate the psychological distress that large numbers of New Yorkers experienced as a result of the attacks. The program did this by providing free and anonymous community-based disaster mental health services to help individuals recover from their psychological distress and regain their predisaster level of functioning. The assumptions underlying this broad-based response strategy were that most people's stress reactions, although personally disturbing, constitute normal responses to a traumatic event and will be short term in duration (1). Corresponding interventions therefore emphasized helping people to identify their trauma responses, understand them as normal reactions, and reconnect with preexisting social supports. Populations of special concern were those most highly affected, including victims' families, survivors and their families, displaced individuals, emergency and recovery workers, the elderly, children, certain cultural and ethnic groups, and people with limited financial and social support resources or mental illness.
Because the level of damage resulting from the attacks was unprecedented, possible consequences were unknown; past disasters provided little guidance. Thus the necessary infrastructure for an effective mental health response was built largely from scratch. This required adapting the crisis counseling program model, developed primarily for natural disasters, to respond successfully to the intentional, catastrophic disaster of September 11, 2001. To support effective program services, NYOMH quickly developed and implemented contracting mechanisms, data collection processes and procedures, public education materials, and curricula and training for thousands in the basics of disaster mental health counseling, all tailored to this event. These activities took place as Project Liberty simultaneously responded to the attack's aftermath, and they required the rapid development of new expertise within the public mental health system, drawing upon the assistance of a wide range of national experts.
Through December 31, 2003, Project Liberty provided face-to-face counseling and education and outreach services to an estimated 1.2 million individuals in the declared disaster area. Nearly 550,000 individuals were served with public education, and more than 565,000 received group or individual counseling. Approximately 87 percent of these services were provided in New York City. By comparison, approximately 600,000 individuals were served in the New York State public mental health system in 2003.
Service volume steadily increased throughout the first seven months of operation: 700 counseling and educational sessions were provided in September 2001, and the number of sessions almost doubled in each succeeding month through April 2002. In May 2002 monthly volume reached 41,000 sessions, and that level of service was sustained until August 2003, when the program phase-down began. Subsequently, from September 2003 through the close of the community-based crisis counseling program in December 2003, the number of sessions gradually declined from about 24,000 to 12,000 per month. In all, 753,015 counseling and education sessions were provided between September 2001 and the end of December 2003. Counseling sessions involving one person or two or more members of the same family were coded as being an individual session. Most sessions (95 percent) were coded as individual sessions. The remaining 5 percent were group counseling sessions involving unrelated individuals and public education sessions. Public education sessions were formal presentations to groups of participants on topics such as common reactions to trauma, recognition of distress among children, stress reduction techniques, and methods that can be used to build coping skills.
Although the community-based crisis counseling program administered by the New York City Department of Health and Mental Hygiene closed in December 2003, Project Liberty programs that were operated by the Fire Department of New York and the Department of Education continued to operate throughout 2004. During 2004 the fire department provided an additional 24,972 counseling and educational sessions to present and former firefighters and their families. In addition to Project Liberty's main community-based program, a school-based crisis counseling program was created that was operated by the Department of Education. This program, which began at the end of 2002 and continued until the end of December 2004, provided a total of 159,094 crisis counseling and educational sessions to an estimated 408,000 students, parents, teachers, and other staff in the city's school community. Because these programs were still in operation and data collection had not been completed when the analyses reported in this issue of Psychiatric Services were being conducted, all of the articles presented here exclude 2004 data from the fire department and all data from the part of Project Liberty that was operated by the Department of Education.
Project Liberty was the first FEMA-funded crisis counseling program to explicitly build in a program evaluation component. NYOMH conducted a multifaceted evaluation in collaboration with New York City, the participating counties, and academic partners, notably Mount Sinai School of Medicine's Division of Health Services Research and the New York Academy of Medicine. In addition to collecting data on service encounters, the evaluation gathered information from stakeholders who were involved in providing and receiving crisis counseling in order to document issues related to the program's implementation and operation and to the needs of the communities served. Feedback from crisis counseling recipients was sought through surveys and telephone interviews about their experiences, needs, and opinions on the helpfulness of Project Liberty and their satisfaction with the care received.
The articles in this issue of Psychiatric Services summarize evaluation findings related to a number of program dimensions. Their content reflects key policy decisions that were made while the program was being rapidly implemented in response to the aftermath of the September 11, 2001, attacks and highlights how Project Liberty responded to challenges presented by the disaster. It is hoped that this information will have utility for persons engaged in disaster preparedness activities and for providers of community-based mental health services.
Effectiveness of Project Liberty outreach
An immediate challenge was how to best organize to effectively serve a very diverse population spread over an extensive geographic area. The strategy adopted was for local mental health authorities to identify and contract with community-based providers capable of proactively engaging affected groups. In all, nearly 200 agencies participated in Project Liberty, including large and small mental health agencies, consumer run organizations, faith-based social service agencies, and agencies with experience serving particular ethnic, cultural, or racial groups. A second complementary strategy was to conduct a large-scale media campaign to inform the public of the availability of Project Liberty services and to educate them about common psychological reactions to traumatic events.
The articles "Demographic Characteristics of Individuals who Received Project Liberty Crisis Counseling Services" (2) (see page 1261) and "Service Utilization and Event Reaction Patterns Among Children Who Received Project Liberty Counseling Services" (3) (see page 1277) describe how effectively Project Liberty provided outreach to affected individuals across the declared disaster area. Their complementary analyses describe the demographic characteristics of service recipients, the extent to which service recipients represented the demographic characteristics of their communities, and patterns of service use during the 27 months after the attacks. Both articles provide evidence that the demographic characteristics of service users approximated closely those of their respective communities and that Project Liberty successfully provided community-based services that were accessible to individuals of diverse racial and ethnic backgrounds.
Elevated mental health needs of persons affected by disasters can extend over months and years, raising the question of whether meeting these needs may divert resources away from the supply of other mental health services. Whether the introduction of Project Liberty's services affected the ability of agencies to supply outpatient mental health treatment services reimbursed by Medicaid, the main public program for people with serious mental illnesses, is discussed in the article titled "Did Project Liberty Displace Community-Based Medicaid Services in New York?" (4) (see page 1309). The study's findings suggest that overall, the supply of Project Liberty services was not associated with a lower supply of Medicaid-reimbursed outpatient treatment services and that implementation of flexible "demand side" financing, such as that used in Project Liberty, can call forth a large volume of new services rapidly and without interfering with other community services.
From September 2001 to December 2002 Project Liberty spent $9.38 million on a large-scale media campaign to inform and educate the public about the availability of project services and common psychological reactions to traumatic events. "Impact of a Media Campaign for Disaster Mental Health Counseling in Post-September 11 New York" (5) (see page 1304) describes the campaign and how its patterns of spending were correlated with calls to LifeNet, the hotline that provided information about and referrals to Project Liberty. Call volumes increased during months when total monthly media expenditures peaked. Spending on radio and print advertisements was significantly associated with increased call volume, although television outreach had the largest positive influence on calls per dollar of media expense. The findings reported in this article suggest that advertising through electronic media, especially television, may be particularly effective in encouraging help-seeking behavior.
Responding to the needs of Project Liberty service recipients
To monitor service recipients' reactions to events related to September 11, 2001, and symptoms of service recipients, counselors completed service encounter logs that included checklists for event reactions. Analyses of event reactions were used to guide program planning and enhance Project Liberty's responsiveness to the needs of service recipients. "Service Utilization and Event Reaction Patterns Among Children Who Received Project Liberty Counseling Services" (3) (see page 1277) discusses how sadness, tearfulness, fear, anxiety, concentration difficulties, anger, irritability, intrusive thoughts and images, and difficulty sleeping were the most common event reactions reported for children and adults. Children's event reactions differed somewhat as a function of age. Counselors reported that compared with older children (12 to 17 years), elementary school-age children exhibited relatively more isolation and withdrawal, anxious and fearful reactions, and concentration difficulties. Event reactions listed on service logs suggested that older children were more similar to adults and more likely than younger children to exhibit a number of avoidance and "numbing" reactions across domains and to abuse substances.
The article "Clusters of Event Reactions Among Recipients of Project Liberty Mental Health Counseling" (6) (see page 1271) reports on how hierarchical cluster analysis was used to group the 31 event reactions reported on the service encounter logs in order to explore how well the event reaction checklist mapped onto traditional diagnostic criteria. The findings indicated that the event reactions experienced by respondents clustered in groups consistent with criteria for major depressive disorder and posttraumatic stress disorder. These results suggest that a simple checklist is a useful way to track event reactions, thereby making it possible to monitor the emergence and amelioration of event reactions suggestive of these disorders as people engage in crisis services.
"Use of Project Liberty Counseling Services Over Time by Individuals in Various Risk Categories" (7) (see page 1268) describes the timing of entry into service for individuals in various risk groups who accessed Project Liberty's individual counseling services for the first time over the 27 months after the attacks. Individuals who lost family members accounted for the largest percentage of first-time service visits in the early months after the attacks, whereas uniformed personnel used disproportionately larger percentages of services much later. Occupationally displaced and unemployed workers sought counseling at relatively steady rates. These patterns suggest that counseling services should be made available for extended periods after disasters and that the focus of counseling should be shifted to meet emerging needs of different high-risk groups.
"Characteristics of Project Liberty Clients That Predicted Referrals to Intensive Mental Health Services" (8) (see page 1313) describes risk factors that predicted whether service recipients received referrals from their Project Liberty crisis counselors for intensive, professional mental health services. The study found that individuals in need of intensive mental health treatment continued to enter the Project Liberty system for up to two years after the attacks. The study found that individuals with more pervasive distress and greater attack-related exposure were more likely to be referred for more intensive treatment, suggesting that counselors providing disaster services pay special attention to these risk factors. A limitation of the use of Project Liberty log data on service encounters for this analysis is that it cannot be used to determine the extent to which service recipients followed up on referrals to professional mental health services.
Enhancing the service delivery model
There is considerable interest in developing ways to optimize community response to disasters. During the first ten months of Project Liberty operations, NYOMH analyzed the accumulating evidence from the service delivery process and from researchers studying the mental health impact of September 11, 2001. The results of this analysis suggested that although crisis counseling services were sufficient to facilitate a rapid return to predisaster functioning and improved resiliency for a majority of individuals who used Project Liberty services, they were not sufficient for a sizeable minority of individuals who, because of the intensity of their exposure to the attacks or predispositional risk factors, experienced persistent traumatic symptoms at levels that resulted in substantial functional impairment. On the basis of this information, NYOMH sought approval from its federal partners to expand Project Liberty service offerings to include a broader set of free, evidence-informed trauma treatment interventions known as enhanced services and additional training in enhanced services, technical assistance, and ongoing supervision to a limited set of service providers. The approval in August 2002 of Project Liberty's service expansion by FEMA and the Substance Abuse and Mental Health Services Administration marked the first time that such an expansion of crisis counseling services had been authorized under a FEMA grant.
"Outcomes of Enhanced Counseling Services Provided to Adults Through Project Liberty" (9) (see page 1298) compares severity of impairment among recipients of crisis counseling and enhanced services. Compared with crisis counseling recipients, users of enhanced services were found more likely to have been more directly affected by the attacks. The study also examined whether enhanced services recipients reported improved functioning and fewer trauma-related symptoms. On the basis of follow-up interviews, receipt of a brief, manualized intervention from a trained mental health professional was associated with improvements in depression, complicated grief, and daily functioning at work, school, and home, suggesting that the provision of specialized services resulted in improvement for individuals beyond what was achieved by crisis counseling alone.
Enhanced services required a referral mechanism that could be quickly administered, easily scored by crisis counselors, and effectively employed systemwide. The mechanism would also need to be consistent with crisis counseling program principles and guidelines. "A Psychometric Analysis of Project Liberty's Adult Enhanced Services Referral Tool" (10) (see page 1328) presents a psychometric analysis of the referral tool, designed to be a brief measure of current (past-month) disaster-related distress, functional impairment, and felt need for professional help, regardless of the presence or absence of specific psychiatric conditions. The results suggest that the tool is an internally consistent measure of distress and dysfunction that provides an apparently successful, empirical basis for referral from crisis counseling to the more intensive enhanced service interventions. The concept of enhanced services and the development of an objective method for identifying adults most in need of them herald a maturation of the field of disaster mental health that is particularly appropriate in the aftermath of terrorism.
"Use of Clients' Self-Reports to Monitor Project Liberty Clinicians' Fidelity to a Cognitive-Behavioral Intervention" (11) (see page 1320) represents the first systematic effort to evaluate treatment fidelity in disaster mental health services. It describes a novel evaluation strategy to determine clinician adherence to enhanced services that utilizes brief questions to service recipients during telephone interviews. The method developed was informative with regard to enhanced services and as a possible new, cost-effective methodological approach to quality assurance monitoring.
"Screening for Complicated Grief Among Project Liberty Service Recipients 18 Months After September 11, 2001" (12) (see page 1291) describes the first survey to screen for disaster-related complicated grief, a recently identified condition marked by symptoms of continuing separation distress and accompanying bereavement-related traumatic distress. Nearly half (72 persons, or 48 percent) of 149 crisis counseling service recipients who participated in a telephone interview more than a year after the attacks indicated that they knew someone killed as a result of September 11. Among individuals who knew someone who died in the World Trade Center attacks, a substantial subgroup (44 percent) screened positive for complicated grief (23 percent), including subthreshold symptoms (21 percent). Individuals so identified had increased risk of other common disaster-related mental health disorders, as well as significantly greater functional impairment than those who screened negative for complicated grief. The study's results support the importance of complicated grief as a unique condition and indicate the need for clinicians and researchers working with victims of disaster to recognize and treat this condition.
Providing acceptable and effective services
Service recipient feedback was examined to determine how acceptable Project Liberty services were to recipients, who rated the program's effectiveness in helping them return to their predisaster level of functioning. "Effectiveness of Two Methods of Obtaining Feedback on Mental Health Services Provided to Anonymous Recipients" (13) (see page 1324) examined alternative methods for obtaining feedback from people receiving Project Liberty mental health services. Although responses to a brief questionnaire and telephone interview were similar, nearly two-thirds of participants chose to return their forms by mail, further underscoring the utility of having multiple response methods for obtaining feedback.
"The Road Back: Predictors of Regaining Preattack Functioning Among Project Liberty Clients" (14) (see page 1283) examines the likelihood and predictors of Project Liberty counseling recipients' reporting a return to satisfactory life functioning 16 to 26 months after the attacks across five life domains: functioning in job or school, maintaining relationships with family and friends, handling daily household activities, taking care of physical health, and staying involved in community activities. The study found that African Americans were two to four times more likely than respondents from all other races to report return to good or excellent functioning after the attack in four of the five domains (relationships with family or friends, daily household activities, physical health, and community activities). Also, respondents who lost their job as a result of the attacks were less likely to report a return to preattack functioning in three domains (job or school, daily household activities, and community activities). The authors concluded that responses to future terrorist attacks should consider demographic characteristics and the impact of the attack (for example, job loss), because these factors can enhance or inhibit return to preattack functioning.
"Clients' Satisfaction With Project Liberty Counseling Services" (15) (see page 1316) found that at least 89 percent of service recipients who responded to surveys from a sample of participating agencies rated Project Liberty as either "good" or "excellent" across 11 service quality dimensions and four domains of functioning; counselor respect and cultural sensitivity were rated particularly favorably. These findings provide support for the ability of crisis counselors to provide accessible and acceptable disaster-related mental health services even while working under extremely difficult conditions.
The immense size and scope of Project Liberty—apparent in the number of people served, the number of participating agencies and counselors, the demographic diversity of served populations, the scale of public awareness of the program, and as the program evolved, the range of services provided—demonstrate that the public sector is capable of mounting a large-scale disaster mental health response to a catastrophic terrorist attack. However, doing so required intense intergovernmental collaboration and flexibility—which fortunately were achieved in the aftermath of September 11, 2001—between the county, city, state, and federal entities responsible for emergency mental health response. Because large-scale terrorism is new to Americans, the necessary infrastructure for New York's mental health response to September 11, 2001, was built largely from scratch under intense pressure. Project Liberty drew on the expertise and experience of staff within the public mental health system but in atypical ways. It required an expansion of focus to the general population and large-scale provision of non-office based, psychoeducational services—it was, in essence, a public health model.
Project Liberty marks the first instance in the history of federally funded postdisaster crisis counseling programs that permission was given for funds to be used for the formal evaluation of service delivery. Thus, before Project Liberty, very few data were available concerning the operational challenges inherent in mounting and sustaining large-scale disaster mental health response programs, the impact of such programs on existing mental health service systems, the characteristics and traumatic reactions of disaster victims served by these programs, and the outcomes of disaster mental health services. Despite substantial methodological limitations imposed on the Project Liberty evaluation by the emergency context itself, which limits the generalizability and conclusiveness of evaluation findings, the data amassed during Project Liberty and the analyses presented in the wide-ranging set of articles included in this issue of Psychiatric Services collectively represent a significant advance in the knowledge base concerning postdisaster public mental health response efforts.
Project Liberty evaluation findings provide further evidence that the mental health impact of terrorism is substantial, varied, and often persistent and that an effective public mental health response to terrorism requires a population-based approach that anticipates a wide range of individual and collective reactions, ranging from emotional distress, to changes in behavior, to emerging psychiatric illness.
Findings from the evaluation of Project Liberty's innovative enhanced services components provide, for the first time, evidence that it is feasible, within the context of a large-scale emergency mental health response program, to identify disaster victims with severe and persistent trauma symptoms for whom specialized trauma treatment may be warranted. This identification can be accomplished by using screening tools and procedures that have a low administrative burden.
Including screening and specialized trauma treatment in Project Liberty as additional components to the crisis counseling and public educational services traditionally emphasized in federally funded disaster mental health programs can be considered the first trial of a new model of postdisaster emergency mental health response. This model incorporates the short-term supportive psychoeducational services appropriate for a majority of affected individuals and the more intensive trauma treatment services needed by a minority of severely impacted disaster victims. Potential additional costs for the broader continuum of care in this model may be controlled for by triaging access to the more expensive and scarce specialized services with a standardized screening process. Broadening the continuum of care made available with public funding to disaster victims may also help to enhance outcomes for recipients by avoiding or minimizing service system fragmentation, a systems-level barrier often cited as a major impediment to the effectiveness of public mental health services.
However, the data from Project Liberty alone are insufficient to determine the relative effectiveness of the broader continuum of care model tried in New York for the first time compared with the bifurcated model used in past disasters in which federal emergency funding is available only for short-term crisis counseling and services for persons with more intensive and persistent mental health problems resulting from the disaster are limited to referrals to external specialty mental health services, services that may or may not be readily available to victims in need. Further trials in other postdisaster settings and more rigorous investigation of outcomes are needed, but this can happen only if current federal guidelines governing the use of FEMA emergency mental health funds are amended to allow future disaster mental health response programs to incorporate the additional screening, enhanced services, and program evaluation components pioneered in Project Liberty.
Much remains to be learned about the impact of terrorism on affected populations. Enhanced funding is needed to support and increase the pace of clinical and services research concerning the mental health impact of terrorism and effective interventions (clinical and organizational), as scientific knowledge remains limited. Governmental entities responsible for public mental health also need to explore ways to more effectively disseminate scientific knowledge of mental health needs in disaster response with the overall goal of strengthening disaster response capabilities in our communities. We believe that much of the experience and knowledge gained in New York through Project Liberty has wide applicability to inform national-level planning and policy making concerning the role of mental health care and services research in emergency preparedness and overall homeland defense.
This evaluation was funded by grant FEMA-1391-DR-NY (titled "Project Liberty: Crisis Counseling Assistance and Training Program") to New York State from the Federal Emergency Management Agency. The Center for Mental Health Services of the Substance Abuse and Mental Health Services Administration administered the grant.