Observed emotional involvement and overinvolvement in families of patients with bipolar disorder.

Many studies have examined the construct validity of the criticism component of expressed emotion, but little work has been done on clarifying the emotional overinvolvement (EOI) construct. In a sample of 115 recently episodic patients with bipolar disorder, the authors of the present study examined the construct validity of an observational coding system for both appropriate and inappropriate emotional involvement that permitted separate ratings for relatives' intrusiveness, self-sacrificing behaviors, and distress related to the patient's well-being. Findings support the measure's reliability and convergent validity and are moderately supportive of the measure's discriminant validity. Results also suggest that Camberwell Family Interview (C. E. Vaughn & J. P. Leff, 1976) EOI ratings do not discriminate among the different dimensions of the emotional involvement construct (or their appropriateness or inappropriateness) as revealed in laboratory-based interactions. The findings suggest that clinicians working with such families might consider differentiating among the various ways in which family members are involved with the patient and helping them learn to judge under what circumstances such involvement is appropriate and inappropriate.