Jacques Jouberta; Lynette Joubertb; David Bartonc; Graham Meadowsd; Chris Reide; Toby Cummingf; Stephen Davisa
a Department of Neurology, Royal Melbourne Hospital
b School of Social Work, University of Melbourne
c Department of Psychiatry, Royal Melbourne Hospital
d Department of Psychiatry, Monash University
e Department of Epidemiology and Preventive Medicine, Monash University
f National Ageing Research Institute, Parkville
beyondblue Victorian Centre of Excellence
Project completion year
The occurrence of post-stroke depression (PSD) is receiving increasing research attention. The shared care process, in which the specialist and general practitioner take concurrent responsibility for the care of the patient, is an approach that has been successfully applied to a variety of chronic medical conditions. It has not however, been evaluated in the context of stroke. In this study, we incorporated regular telephone screening for depressive symptoms into a shared care model for stroke survivors. We evaluated the effects of this model of care on stroke survivors one year after discharge, with particular reference to depressive symptomatology.
A total of 186 stroke patients were included in the study: 92 were randomised to the treatment group and 94 to the control group. Treatment group patients were exposed to a model of shared care that emphasised vascular risk factor modification and the identification of PSD. Depression screening was done using the Patient Health Questionnaire (PHQ-9). The PHQ-9 is a nine item depression scale which assists primary care clinicians in diagnosing depression as well as selecting and monitoring treatment. Screening was done at three months, six months, nine months, and 12 months post-discharge, and these results were faxed back to the GPs. Control group patients were exposed to usual care from their GPs, and were evaluated for depressive symptoms at 12 months.
Patients exposed to shared care exhibited significantly fewer depressive symptoms than controls at 12 months post-stroke. At 12 months, 34% of the treatment group expressed depressive symptoms, compared to 54% of the control group (p < .01; see Figure 1). Patients in the treatment group exhibited a steady decline in depressive symptoms over 12 months.
Of the 31 treatment patients depressed at 12 months, 26% were taking antidepressants, compared to only 14% of the 51 controls who were depressed at this stage.
Cognitive performance at discharge was a significant predictor of depression at 12 months, but depression at discharge was not a significant predictor of cognitive performance at 12 months. Worsening of depression was associated with improving cognitive function, indicating that stroke patients who are more cognitively impaired are somehow emotionally insulated from their stroke-related deficits.
In line with previous findings PSD was associated with measures of stroke severity and functional disability.
Figure 1. Percentage of patients with depressive symptoms (PHQ-9 score of 5 or above indicating moderate to severe depression) at 12-month follow-up in treatment and control groups.
Depression was associated with: stroke severity; poorer functioning and activities of daily life; and greater physical handicap.
Implications for Policy and Practice
The shared care approach provides an effective framework for detecting and monitoring PSD. Telephone screening for depression is easy to administer, inexpensive, and acceptable to both patients and GPs.
The specific reasons for reduced depression in patients exposed to shared care are unclear and warrant further study. Regular contact with health care professionals with expertise in the area of stroke may have been important in ameliorating depression. With regular feedback to the GP of depression test results, there is increased awareness of the presence of depressive symptoms in stroke patients, resulting in greater rates of treatment.
The success of this preliminary study has resulted in the extension of the project both nationally and internationally. This includes research on the specific needs of stroke patients in rural areas, with a view to extending the model of care.
We are attempting to make a shift from case management to a model of patient empowerment. Future shared care interventions will place even greater emphasis on patient education and support, which should improve sustainability.
Williams, L. S., Brizendine, E. J., Plue, L., Bakas, T., Tu, W., Hendrie, H., & Kroenke, K. (2005). Performance of the PHQ9 as a screening tool for depression after stroke. Stroke, 36, 635-638.