Journal of Rehabilitation Medicine 51-3 | Page 18

164 M. Gustavsson et al. Table V. Multivariable logistic regression for the association between the independent variables and a positive outcome using Stroke Impact Scale 3.0, domain 8 (SIS), Frenchay Activities Index (FAI) and Occupational Gaps Questionnaire (OGQ), odds ratios (OR), 95% confidence intervals (95% CI) and p-values Positive outcome in SIS, OR (95% CI) p-value Positive outcome in FAI, Positive outcome in OR (95% CI) p-value OGQ, OR (95% CI) p-value Age Sex Women 0.98 (0.93–1.03) 1.02 (0.99–1.05) 0.898 Men FAI pre-stroke Stroke severity Moderate/severe 0.80 (0.31–2.05) 1.01 (0.95–1.08) Mild Intervention group, Rehabilitation type Inpatient 2.70 (0.97–7.69) Home Control group, Rehabilitation type Inpatient 0.45 (0.11–1.84) Independent variables 0.510 1 1 0.646 0.729 1 Home 4.17 (2–9.09) 1 0.73 (0.30–1.77) DISCUSSION The aim of this study was to explore the importance of client characteristics (age, sex, stroke severity and par- ticipation before stroke), context (inpatient or client’s home), and approach (enhanced client-centeredness or not) of rehabilitation on participation in everyday life one year after stroke. The main findings were that, among participants receiving usual rehabilitation, stro- ke severity and rehabilitation context were associated with a positive outcome in participation, whereas only stroke severity was associated with a positive outcome for participants receiving an enhanced client-centred intervention. Stroke severity, i.e. having had a mild stroke, was, in this study, associated with a positive outcome, 2.42 (1.00–5.87) 0.89 (0.49–1.62) 1.03 (0.99–1.07) 0.710 0.184 1 < 0.00 0.80 (0.37–1.72) 0.564 1 0.485 1 0.087 1.00 (0.97–1.03) 1 0.619 0.725 1 0.265 medical rehabilitation) (see Table IVb). No significant associations were found with OGQ (see Table IVc). Table V presents the results from the multivariable logistic regression analyses for both groups combined for the associations between the independent variab- les and positive outcome for the 3 outcome measures of participation. There were significant associations between having a mild stroke and a positive outcome using FAI (OR 1.00 vs 0.24 for moderate/severe stroke). Furthermore, a nearly significant association (p-value = 0.051) was found between home rehabilita- tion and a positive outcome using FAI for those in the control group (OR 1.00 vs 2.42 for in-patient rehabi- litation) (see Table V). For the participants receiving the enhanced client-centred intervention, there were no significant associations between rehabilitation context and a positive outcome using the FAI, SIS or OGQ. Neither age, sex nor participation in everyday life be- fore stroke were associated with a positive outcome in participation in everyday life after stroke. www.medicaljournals.se/jrm 0.86 (0.98–1.04) 1.02 (0.97–1.05) 1 0.058 1 3.25 (0.84–12.57) 0.309 2.70 (0.91–8.01) 0.072 1 0.051 1.13 (0.51–2.49) 0.759 using FAI for all participants, and SIS domain 8 for those in the control group . This result may reflect natural recovery in those with mild stroke, regardless of which rehabilitation they received. However, one explanation for the different results, depending on the instrument used, could be that the instruments assess different aspects of participation. The FAI measures the frequency of participation in daily activities, and the results of the current study show that a mild stroke is associated with an increased frequency of participation regardless of intervention received. The SIS, on the other hand, measures the perceived impact of stroke on participation in daily activities, and the results of this study imply that, although the actual frequency of participation did not change, using an enhanced client- centred approach may reduce the perceived impact of stroke on participation in everyday life. These results imply that stroke severity does impact the outcomes in participation in everyday life; however, stroke severity may be of less importance for outcome in perceived impact of stroke on participation when receiving an enhanced client-centred intervention. Thus, a client- centred approach may be particularly important for those with moderate to severe stroke. Results from previous studies on the association between stroke severity and participation are inconsistent. While some studies have shown that stroke severity, i.e. that a mild stroke could have a positive impact on participation both early on (at 1 month) (27) and a long time (at 6 years) after stroke (5), other studies have failed to confirm this association (2, 28). In addition to stroke severity, context of rehabilita- tion, i.e. receiving home rehabilitation, was associated with a positive outcome in participation according to the FAI for the control group. One could argue that participants receiving home rehabilitation have usually had a less severe stroke and therefore are more likely to