Early attachment experience and recovery from psychosis

Two studies were carried out to investigate the relationship between coping styles and co-morbid depression in people adjusting to the onset of psychosis. Evaluative thinking and early attachment experiences were also examined.

In Study 1, a 39-item Recovery Style Questionnaire (RSQ) was developed to measure recovery style in people with psychosis, based on McGlashan, Levy & Carpenter's (1975) interview measure of recovery style. Fifty-six participants completed both the RSQ and McGlashan's interview-based measure. Study 2 explores the relationship between these styles of recovery, depression and early childhood attachment experiences. Thirty-six people participated.

The RSQ was both reliable and correlated highly with McGlashan's interview-based measure. We found that the RSQ, in keeping with the interview-based measure, was bimodally distributed, thus supporting McGlashan's contention that they define two distinct recovery styles termed 'integration' and 'sealing over'. As predicted, 88 per cent of the 'sealers' were moderately to severely depressed compared to 52 per cent of the 'integrators' who were mildly depressed with no members of the 'integration' group experiencing moderate to severe depression (p < .0003). Patients who employed the sealing over recovery style also made significantly more negative self-evaluations than did patients in the integration group and also perceived their parents to be significantly less caring than those in the integration group.

These findings are explained in terms of a multi-axial model incorporating personality structure and development as well as mental disorder. It is suggested that those individuals with a poorly developed sense of self defend against the threat of psychosis using denial. Clinical implications are discussed and more research is suggested to further investigate the links between evaluative and inferential thinking in co-morbid depression, and how such thinking relates to early childhood experience.