Eligible participants were in the longitudinal Life After Stroke Phase 1 (LAS-1) study who participated in a 6-year follow-up. The LAS-1 was a prospective observational study of the rehabilitation process after stroke, which is described in detail elsewhere13. From an original sample of 349 patients diagnosed with stroke recruited consecutively in 2006-2007 to stroke units at Karolinska University Hospital in Stockholm, Sweden, 183 surviving individuals were mailed to be invited to participate. Informed signed consent was obtained from all participants. Ethical approval for the original study and 6-year follow-up was granted by the regional ethics committee in Stockholm (2011/1573-32, 2012/428-32) and the procedures were performed in accordance with the Declaration of Helsinki .
In the current study, all participants in the LAS-1 study who consented to participate in the 6-year follow-up and who completed the Hospital Anxiety and Depression Scale (HADS) were14 were included.
Data was collected through structured face-to-face interviews with experienced occupational therapists and physiotherapists. Most of the interviews were conducted in the participants’ homes. If necessary, a family member was present during the interviews.
Sociodemographic data were collected at baseline, within 5 days after stroke, and at the 6-year follow-up. The Barthel Index (BI) has shown good agreement with other measures of stroke severity15 and was used to rank stroke severity at baseline. A score 16. Cognitive function was assessed using the Mini-Mental State Examination (MMSE).17. One item from the Scandinavian Stroke Scale was used to assess the presence and severity of aphasia at baseline18. After 6 years the following data was collected.
Anxiety and depression scale in the hospital
Symptoms of depression were collected using the Hospital Anxiety and Depression Scale (HADS), a 14-item self-rated scale used to screen for anxiety and depression in individuals with physical illness14. The items cover non-physical symptoms of anxiety and depression, with seven items covering anxiety (HADS-A) and seven items covering depression (HADS-D). The respondent is asked to rate agreement with the statements for a period of 1 week on a four-point scale from 0 = no symptom to 3 = maximum symptom. The maximum score on HADS-D is 21 and a commonly used cut-off for depression is >819. However, it has been suggested that a lower score is more accurate for detecting depression in people who have had a stroke20 Therefore, a cut-off of 4 was used in this study.
Stroke Impact Scale
The perceived impact of a stroke was assessed using the Stroke Impact Scale (SIS) version 3.021, a 59-point scale that assesses eight domains: strength, hand function, ADL, mobility, communication, emotion, memory and thinking, and participation. Responses on each domain are converted into a score from 0-100, with 0 indicating maximum and 100 indicating no perceived impact of the stroke. The scale also includes a single item reflecting perceived recovery from stroke, which is rated on a visual analog scale with a range of 0–100, with 0 reflecting no and 100 reflecting maximum recovery.
The Barthel Index15 was used to assess ADL. The instrument consists of ten questions related to personal care and mobility activities. The overall score is 0-100, with higher scores indicating greater independence. Any score below 100 indicates some level of dependency.
Frenchay activity index
Participation in social and daily activities was assessed using the Frenchay Activities Index (FAI).22. The scale consists of 15 items covering housework, outdoor activities and leisure/work. Each item is scored from 0 to 3 depending on frequency of activity over the past 3 or 6 months. A higher score indicates more frequent participation in social and daily activities. An overall score of 23.
Mean, standard deviation (SD), median, interquartile range (IQR), minimum and maximum values were calculated for descriptive data. A cut-off of HADS-D ≥ 4 was used for depression symptoms in bivariate analyses. Group differences between participants with symptoms of depression versus no symptoms of depression were calculated using Chi2-Test for categorical variables and Mann Whitney u for ordinal scale data. The level of significance was set at p