Forty-nine families from the University of Rochester Child and Family Study were followed up 10 to 14 years after initial assessment. Two inclusion criteria were applied: at least one of the parents should have been hospitalized for a functional psychiatric disorder before initial assessment and the male index offspring should be 18 years or older at follow-up. Family communication was observed by the Consensus Rorschach procedure and coded with the Confirmation-Disconfirmation Coding System. Offspring childhood social competence was rated by peers, teachers and parents. Outcome measures included the Denver Community Mental Health Questionnaire, the Global Assessment Scale and hospitalization for a psychiatric disorder. Results showed that the family's level of confirmation and disconfirmation communication during Consensus Rorschach at initial testing predicted offspring interpersonal functioning and hospitalization for psychiatric disorders. These findings were not due to the initial social competence or IQ of the child, the level of functional impairment in parents or their social class.

Forty-nine families from the University of Rochester Child and Family Study were followed up 10 to 14 years after initial assessment. Two inclusion criteria were applied: at least one of the parents had been hospitalized for a functional psychiatric disorder before initial assessment and, second, the male index offspring should be 18 years or older at follow-up. Initial measures included observationally based coding of the family's level of disqualifying communication toward the index offspring, index child's scores on the Child Manifest Anxiety Scale, and ratings of the index child's social competence carried out by peers, teachers, and parents. Offspring outcome was measured by the Mental Health Inventory, Global Assessment Scale (GAS), and hospitalization for psychiatric disorder. The results showed that every measure of offspring outcome was predicted by the amount of disqualification directed to the offspring from the other family members. In addition, GAS score and mental health were predicted by the offspring's competence as a child. Family disqualification, childhood competence, and socioeconomic status accounted for 63% of the variance in adult GAS scores.