A longitudinal study of psychological and social factors affecting recovery from psychiatric breakdowns.

Thirty-five people who had suffered a recent onset of symptoms of depression and/or anxiety were interviewed soon after being seen by a clinician, and again six months later. Symptom scores at the second interview were predicted by whether any plans subjects had formed by the first interview had gone wrong, by major non-health difficulties, and by internal, stable and global attributions made at the first interview about the event or difficulty that was most distressing before symptoms became severe.