Long term course of chronic posttraumatic stress disorder in traffic accident victims: a three-year prospective follow-up study

Abstract

The purpose of the present study was to gather prospective longitudinal data on the long-term course and outcome of chronic posttraumatic stress disorder (PTSD). The target population for this study was 74 injured traffic accident victims who had been previously followed-up for one year after the trauma. Nineteen of the original 24 PTSD subjects (79%) and 39 of the original 50 Non-PTSD subjects (78%) were available for this study, which took place during the fourth year after the accident. Our results show that 10 (53%) of the 19 patients with PTSD at one-year still suffered from PTSD after another two-year follow-up interval, while 9 recovered from PTSD during this follow-up period. Only 2 of the 39 without PTSD at one year developed delayed onset PTSD. The best predictor of recovery from chronic PTSD was the initial level of posttraumatic reaction immediately after the accident. These results demonstrate that spontaneous recovery from PTSD can occur even among patients who are currently considered chronic. Severity of initial reaction to the trauma appears to be a major risk factor for non-remitting chronic PTSD.

Introduction

Over the past several years, prospective studies in trauma victims have made considerable strides towards increasing our understanding of the risk factors and natural course of PTSD development. For instance, they showed that early emotional and posttraumatic symptoms predict PTSD 6–12 months after the trauma (Blanchard et al., 1995, Koren, Arnon & Klein, 1999), that the rates of spontaneous recovery within the first year after the trauma are approximately 50% (Blanchard et al., 1997, Rothbaum, Foa, Riggs, Murdock & Walsh, 1992), and that about one third of PTSD patients have a delayed onset (Epstein, 1993, Mayou, Bryant & Duthrie, 1993, Mayou, Tyndel & Bryant, 1997). While these studies enhanced our understanding of the risk and progress of PTSD, the longer-term course of PTSD remains unclear, as most of these studies focused on the first year after the trauma (the only exception is Mayou et al. (1997), who followed their subjects for 5 years).

According to DSM-IV, posttraumatic symptoms that last longer than three months are considered chronic PTSD, while symptoms that start 3 months or more after the trauma are defined as delayed-onset PTSD. Reported prevalence of chronic PTSD varies widely from 1% (Malt, 1988) to 46% (Blanchard, Hickling, Taylor, Loos & Gerardi, 1994) in victims of traffic accidents followed 3–12 months after the accident. In our previous study (Koren et al., 1999) we found that about 30% of injured traffic accident victims suffered from PTSD one year after the accident. Moreover, our data also suggested that the first three months following the trauma, rather than one month as currently indicated in the DSM, is the critical time period for transition from acute stress response to PTSD. Yet, these findings provide only limited information about the long-term course of chronic PTSD, and above all who will suffer from a life long illness and who will eventually recover. This question is of particular importance in view of recent reports suggesting spontaneous remission of posttraumatic symptoms in a significant portion of chronic PTSD patients (Blanchard et al., 1995, Mayou, Tyndel & Bryant, 1997, Rothbaum, Foa, Riggs, Murdock & Walsh, 1992). Similarly, little is known about the prevalence and long term course of delayed PTSD. Finally, clinical observations suggest that in some patients chronic PTSD transforms over time to other Axis I disorders, particularly affective and anxiety spectrum disorders. However, these observations have received only sparse empirical attention.

Another area in which longitudinal studies have been instrumental was identification of early predictors of PTSD. Among the factors highlighted by these studies were pre-traumatic psychiatric problems, trauma severity, and litigation. A history of psychiatric problems was linked to PTSD in several studies (Breslau, Davis, Andreski & Peterson, 1991, Blanchard, Hickling, Barton et al., 1996). In our own study, we found a relationship between PTSD one year after the accident and presence of pre-traumatic Axis I morbidity, primarily affective and anxiety disorders. Yet, to date there is only limited data on the extent to which pre-traumatic psychiatric morbidity predicts chronicity of PTSD beyond one year.

Finally, the degree to which chronic PTSD is affected by litigation is still debated in the literature. Traditional views, dating back to Miller, 1961a, Miller, 1961b, suggest a motivating and perpetuating role for litigation in the formation of PTSD. Yet, recent studies that examined this hypothesis have yielded conflicting results. While some have found a relationship between litigation and PTSD (Blanchard, Hickling, Taylor et al., 1996, Bryant & Harvey, 1995, Culpan & Taylor, 1973), others did not find such a relationship (Mayou et al., 1997), or between settlement of compensation and recovery (Bryant, Mayou, & Lloyd-Bolstock, 1997).

The purpose of the current study was to extend our previous work by examining the original cohort of patients three to four years after the accident. More specifically, our aims were: (1) to assess the prevalence of PTSD (both chronic and delayed onset) three years after the accident, as well the rates of spontaneous recovery among those who suffered from PTSD at one year; (2) to examine the natural course of posttraumatic, as well as other psychiatric, symptoms among chronic and remitted PTSD subjects; and finally, (3) to investigate the predictive power for non-remitting, chronic PTSD of variables that were identified in previous research as predictors of PTSD, such as pre-traumatic psychiatric morbidity, injury severity, and litigation status.