The committee reviewed the literature on PTSD treatment for definitions of the term recovery, finding that the term is used inconsistently and is not clinically meaningful in the same way that it is in other clinical domains (e.g., as in acute illnesses). PTSD can be chronic and can also remit and relapse over a patient’s lifetime (Wilson et al., 2001). No longer meeting PTSD diagnostic criteria is a common way to define recovery when inventory or questionnaire scores fall below an important threshold in the condition’s trajectory. However, the studies that constitute the evidence base on the efficacy of treatment modalities for PTSD use a variety of terms to denote a change for the better in PTSD status: improvement (significant improvement, reliable improvement, improvement in functioning), remission, therapeutic success, loss of PTSD diagnosis, symptom reduction or improvement, trauma recovery, good or high end-state functioning, treatment response, clinically meaningful change, and so on, while the term recovery is used in only three studies (Davidson et al., 1990; Gersons et al., 2000; McDonagh et al., 2005). In most cases, these terms simply describe the primary outcomes chosen in the individual study leading to a positive, negative, or neutral conclusion regarding efficacy. See Box 5-1 for three definitions of recovery (two pertain to mental health recovery in general, and one relates to PTSD specifically).
The studies the committee reviewed employed a range of specific definitions for “recovery” terms. These definitions may be divided into three categories: (1) absence or loss of PTSD diagnosis, (2) multiple domain measures used to determine good or high end-state functioning, and (3) a clinically meaningful threshold for “symptom improvement.” Not all studies seeking to show symptom improvement also reported PTSD diagnostic status, but almost all studies reporting loss of diagnosis did so by showing changes on PTSD symptom measures such as the Clinician administered PTSD Scale (CAPS), the Structured Interview for PTSD (SI-PTSD), and the Structured Clinical Interview for DSM-IV (SCID).
In the first category, studies that had as an outcome absence of or loss of PTSD diagnosis, defined recovery by a decrease in percentage or two standard deviations (SDs) improvement in CAPS score, by decrease or disappearance of a number of symptoms or an entire symptom cluster,4 by change in SI-PTSD scores, or by loss of the diagnosis using DSM criteria. The X percent or 2 SDs decrease are appropriate criteria for loss of diagnosis in many cases. However, a patient with a high score on a measure (>2 SDs above the mean for the PTSD population on which the measure