Psychopathology and autobiographical memory in stroke and non-stroke hospitalized patients

Int J Geriatr Psychiatry. 2003 Jan;18(1):23-32. doi: 10.1002/gps.763.


Background: Psychopathology and autobiographical memory were investigated in a cohort of stroke and non-stroke hospitalized patients. Both these cohorts have been identified as having high levels of psychopathology (Katon and Sullivan 1990; Burvill et al., 1995). Difficulties recalling specific autobiographical memories (overgeneral memory) have been identified as important psychological variables in depression and predictors of outcome (Williams and Scott 1988; Brittlebank et al., 1993). Intrusive autobiographical memories have also been found to be associated with depression and overgeneral memory in depressed women (Kuyken and Brewin, 1995) and depressed cancer patients (Brewin et al., 1998a). This study looked at levels of psychopathology and autobiographical memories in stroke and non-stroke hospital patients.

Method: 417 patients were screened, of the 176 eligible 103 agreed to participate (54 stroke and 49 non-stroke). Participants were assessed for overgenerality using the Autobiographical Memory Test and intrusiveness of memories using the Impact of Events Scale. Also assessed were PTSD-like symptoms (PCL-S), mood (HADS, GHQ-28) and cognitive ability (MMSE, verbal fluency, digit span and estimated pre-morbid IQ).

Results: No significant differences were found between stroke and non-stroke patients on severity of depression, anxiety, severity of PTSD-like symptoms or autobiographical memories. Backward multivariate regression analyses for combined data (stroke and non-stroke) indicated that overgeneral memory recall, intrusive memories of past events and intrusive memories of illness were significant independent predictors of depression (HADS). Avoidance of intrusive memories and reported childhood distress were not predictors of overgeneral memory recall. Significant predictors of overgeneral memory recall were; Gender, antidepressant medication, and estimated IQ.

Conclusion: Significant levels of psychopathology were identified in this cohort. However, there were no significant differences in the levels of depression, anxiety, PTSD symptoms and autobiographical memory between stroke and non-stroke hospitalized patients. Of particular interest was the finding that PTSD-like symptoms did not appear to be influenced by the nature of the person's illness. In combined data (stroke and non-stroke) autobiographical memories (intrusive images of their illness, intrusive memories of other events and overgeneral memory recall variables) were significant predictors of depression in this cohort. This suggests that psychological intervention of memory processes may be a worthwhile target in psychological intervention for depression in these cohorts. Gender, cognitive impairment, antidepressant medication, and estimated IQ were significant predictors of overgeneral memory recall and further investigation into the validity of these findings are warranted. Suggestions for further research and limitations of the study are also discussed.

Publication types

  • Multicenter Study

MeSH terms

  • Aged
  • Aged, 80 and over
  • Anxiety Disorders / etiology
  • Cohort Studies
  • Depressive Disorder / etiology
  • Female
  • Hospitalization
  • Humans
  • Male
  • Mental Disorders / etiology*
  • Mental Recall*
  • Middle Aged
  • Psychiatric Status Rating Scales
  • Regression Analysis
  • Risk Factors
  • Stress Disorders, Post-Traumatic / etiology
  • Stroke / psychology*