Objective: An initial finding of heightened risk of suicidal ideation
or behavior among individuals in a random community sample who met dia
gnostic criteria for panic disorder or panic attacks was not replicate
d in later studies of both general and specific groups of psychiatric
outpatients. The present study represented another effort to validate
the finding. Method: The participants included 209 outpatients who had
attempted suicide or were at high risk for continued suicidal behavio
r or eventual suicide. All subjects were evaluated with a structured c
linical interview for assigning DSM-III-R diagnoses, the Modified Scal
e for Suicidal Ideation, the suicidal ideation subscale of the Suicide
Probability Scale, and the Beck Hopelessness Scale. Results: The find
ings indicated the relative complexity and importance of diagnostic co
morbidity among these suicidal subjects. Mood disorders were the most
frequent primary diagnoses, followed by phobias, posttraumatic stress
disorder (PTSD), generalized anxiety disorder, and panic disorder. Pan
ic disorder was not present as an isolated, independent diagnosis; on
the contrary, all of the patients with panic disorder also received at
least one additional co-morbid diagnosis. Mean scores for suicidal id
eation and hopelessness were greatest for patients with current comorb
id primary mood disorder and panic disorder. However, a critical and e
qually important role was played by comorbid PTSD, generalized anxiety
disorder, and phobias. Conclusions: The findings represent another fa
ilure to validate, with a specific clinical group, panic disorder as a
n independent risk factor for suicidal ideation or behavior. However,
they highlight the possibility that panic disorder and other anxiety d
isorders are risk factors when they co-occur with a primary mood disor
der.