This is an archived article.

May 5, 2004

Collaborative care may prevent PTSD, alcohol abuse among trauma survivors

Approximately 2.5 million Americans are so severely injured each year that they require inpatient hospital admissions, and between 10 and 40 percent of these patients go on to develop symptoms of posttraumatic stress disorder (PTSD). Alcohol abuse, which has been associated with an increased risk of recurrent injury, is part of the diagnosis for 20-40 percent of trauma victims.

New research by investigators at the Harborview Injury Prevention and Research Center (HIPRC) indicates that a multi-faceted collaborative care intervention — one that combines medications and psychotherapy — can reduce alcohol abuse by trauma survivors as well as prevent the development of PTSD, compared to trauma care as usual. “A Randomized Effectiveness Trial of Stepped Collaborative Care for Acutely Injured Trauma Survivors” is published in the May issue of the Archives of General Psychiatry.

The research was conducted at Harborview Medical Center, the region’s only Level I trauma center, between March 2001 and January 2002. Patients in the study had severe injuries and showed symptoms of PTSD and/or depression. Those who met these criteria were also evaluated for alcohol abuse/dependence as well as for their levels of pre-injury symptoms, functioning and use of health services. A total of 2,610 patients were involved in the study, either as part of the collaborative intervention group or the control group, which received usual care for their injuries.

The collaborative intervention involved a team: a master’s level trauma support specialist provided case management and motivational interviewing intervention targeting alcohol, and a psychiatrist and psychologist delivered medications and cognitive behavioral therapy targeting PTSD.

The intervention was a stepped-up procedure, beginning with an intervention at the bedside in the surgical ward. The master’s level trauma support specialist would ask each intervention patient, “Of everything that happened to you since your injury, what concerns you the most?” The patient and case manager would problem solve around each concern the patient expressed. The case manager also followed patients through the surgical hospitalization to outpatient primary care visits and community rehabilitation. Case management was covered by team members around the clock to address any questions and needs of the injured patients.

Patients involved in the collaborative intervention who showed evidence of alcohol abuse/dependence participated in at least one 30-minute motivational interviewing session. Developed by Dr. Larry Gentilello, Dr. Fredrick Rivara, Chris Dunn, Ph.D., and other colleagues at HIPRC, the intervention can be delivered in a trauma surgical inpatient setting. The intervention explores the pros and cons of drinking, the importance of change, and specific drinking goals and action plans to bring about a change in behavior. Patients could receive optional booster sessions after the initial session.

Patients were reevaluated for PTSD three months after their injuries. Those with PTSD at this time received their preference of cognitive behavioral therapy, medications, or a combined treatment. The trauma support specialist remained in contact with study patients 6-12 months after their injuries to assess symptoms, function and rehabilitation. Patients who remained symptomatic of PTSD and/or showed evidence of alcohol abuse/dependence received a combination of ongoing trauma support and intensified care.

Over time patients in the intervention group were significantly less symptomatic for PTSD and alcohol abuse/dependence when compared to controls. On average, patients in the intervention group demonstrated no change in PTSD symptoms over the course of the year after the injury, while patients in the control group manifested a significant worsening of these symptoms. Patients in the intervention group markedly diminished their alcohol consumption over the course of the year as manifested by a 24.2 percent decrease in alcohol abuse/dependence, while patients in the control group averaged a significant 12.9 percent increase in their drinking.

“This investigation establishes the feasibility and effectiveness of delivering mental health interventions at trauma centers,” says Dr. Doug Zatzick, a University of Washington (UW) associate professor of psychiatry and behavioral sciences, and the study’s principal investigator.

“The significance of these findings is underlined by the Sept. 11, 2001 terrorist attack,” Zatzick explains. “The Centers for Disease Control reports that within 48 hours after the attack on the World Trade Center, 1,103 survivors were triaged through five Manhattan acute care centers. It’s critically important to find early interventions for PTSD that engage injured trauma survivors in settings, such as trauma centers, that form the front-line in our healthcare defense. Routine improvements in acute care mental health evaluation and treatment not only have the potential to improve the day-to-day quality of care for injury survivors, they also may inform mental health care delivery after a large-scale mass attack.”

In addition to Zatzick, the study was conducted by HIPRC collaborators Dr. Frederick Rivara, a UW professor pediatrics, and Dr. Gregory Jurkovich, a UW professor of surgery; as well as by Dr. Peter Roy-Byrne, a UW professor of psychiatry and behavioral sciences; Joan Russo, Ph.D., a UW research associate professor of psychiatry and behavioral sciences; and Dr. Wayne Katon, a UW professor of psychiatry and behavioral sciences. RoseAnne Droesch, M.S.W., formerly of the Harborview Injury Prevention & Research Center, served as trauma support specialist; Amy Wagner, Ph.D., a UW assistant professor of psychiatry and behavioral sciences, oversaw the development and implementation of the cognitive behavioral therapy component; and Chris Dunn, Ph.D., a UW assistant professor of psychiatry and behavioral sciences, oversaw the implementation of the motivational interviewing component.