To the Editor: The purportedly "simpler" explanation of our findings offered in the letter by Drs. Parvin and Swartz actually rests on an unsupported assumption. We found that the co-occurrence of panic disorder with bipolar disorder in families ascertained for a high prevalence of bipolar disorder was predicted to a significant degree by one’s membership in a family in which the bipolar proband had comorbid panic disorder. This, we concluded, implies the existence of an inherited risk factor for panic in a subset of families. Drs. Parvin and Swartz argue that the relationship between panic and what we called "bipolar disorder" was really between panic and anxiety disorder symptoms that overlap with manic symptoms. To accept this argument, one has to assume that our diagnosticians systematically overdiagnosed bipolar disorder. This is unlikely since we based our diagnoses on the work of trained interviewers and research psychiatrists using a reliable diagnostic instrument (1) and state-of-the-art diagnostic methods (2). Therefore, it would seem to be "simpler and more ordinary" to assume that we did not confuse anxiety with mania. It is also surprising that Drs. Parvin and Swartz prefer the explanation that panic disorder is associated with generalized anxiety disorder rather than bipolar disorder, since the validity of generalized anxiety disorder syndrome remains open to debate (3) and, at any rate, is apparently a syndrome distinct from panic disorder (4).

On the other hand, I agree with Drs. Parvin and Swartz that the boundaries between anxiety and mood disorders may not be as clear as the current diagnostic nomenclature makes them seem. This is the problem we hope further investigation can address.