Journal of Rehabilitation Medicine 51-9 | Page 39

DISCUSSION to the mRS correlates to higher self-perceived impact of strength, ADL and hand function. These correlations were moderate. Moderate correlation was seen between more severe stroke and health-related quality of life. Self-perceived impact of stroke From 1 to 5 years post-stroke, participants reported a worsening in the self-perceived impact of stroke. Most pronounced were changes in strength, emotion, and participation, as well as autonomy indoors and social life. There were moderate correlations between depen- dency at discharge from hospital (mRS) and strength, ADL and hand function (SIS) as well as with health- related quality of life (EQ5D) 5 years post stroke. The results that self-perceived impact of stroke could be more pronounced after several years are shown in a recent longitudinal study over 6 years (9). Similarly, the same trend can be seen in another study (18) where deterioration could be seen in functional outcome during the first year after stroke. An interview study (19) showed that persons with stroke are still facing challenges 5 years post-stroke and view it as an ongoing process. Persons with stroke have fewer social relations and participate less in society after 5 years. Furthermore, persons adjust their life after stroke by, for example, developing new skills and realizing their limitations (19). The results of that study can explain the results of the present study, which showed a lower score for the participation dimension after 5 years. In the present study, the emotion domain showed significant deterioration at 5 years compared with 1 year. Other studies (20, 21) have shown a relationship between mood state and participation, and this could not be ruled out in the present study. Poor community participation has been shown to be a predictive factor for depression in persons with stroke (20). Another study showed that mood disorders, as measured by Hospital Anxiety and Depression Scale (HADS), was a contributing factor of participation restriction (21). In the IPA, there was a significant deterioration in the autonomy indoors and social life subscales in 1–5 years. A recent study (22) indicated that between 1 and 6 years, the proportion who were satisfied with their family life were lower. This is in line with the present results, with more perceived problems regarding indoor autonomy, which will affect family life. Previous studies (23, 24) with shorter follow-up periods of 3 and 6 months, respectively, have shown that outdoor autonomy was most affected. It is pos- sible that, with time, people refrain from using mo- bility devices indoors, and therefore a deterioration in autonomy indoors occurs. It is also possible that the inability to continue doing the same activities as 663 before stroke contributes to the feeling of restrictions in social life (3). The present study showed moderate correlations bet- ween the strength, ADL and hand function dimensions of the SIS and dependency measured by mRS. Consi- dering that mRS measures functional dependency, it seems natural that these dimensions have the strongest correlations with mRS. The correlation in the present study between mRS and ADL was slightly weaker than was seen in another study (25) conducted approxima- tely 3 weeks into rehabilitation. In the present study, the analysis was performed 5 years post-stroke, when the rehabilitation period is completed and people have probably adjusted themselves to their new situation post-stroke. Limitations The present study has some limitations that should be taken into account. The study only includes patients who were treated at 1 hospital, where all the throm- bolysis and thrombectomy treatments in the region take place, resulting in, for instance, a selection bias towards younger people (since, at the time, there was an age limit on thrombolysis). Furthermore, participants included in the study initially all had impaired upper extremity function, which results in a selected group of participants and may limit the possibility to generalize the results. The method of gathering information, with interviews at 1 year and postal surveys at 5 years, may have influenced the results. However, it has been shown previously that there is a moderate agreement between postal surveys and interviews (26). There is a response bias, with persons with less severe stroke participating in the follow-up. This, as well as the small sample size, reduces the generalizability of the results. This indicates a need for longer, larger follow-up studies. The fact that there is no control population means that the normal ageing process, which could have an influence, is not addressed in the study design. However, since the mean age at stroke onset was 64 years, the healthy normal ageing process between 64 and 69 years will probably not have a large impact on participation. Conclusion It appears that the perceived impact of stroke becomes more prominent with time. This is the case even for persons with mild-to-moderate stroke. Some of the most affected areas are emotion and participation. The present study highlights the need for regular long- term follow up for persons with stroke. The content at follow-up should contain not only secondary preven- tion, but also an assessment of the whole situation for the person. The World Stroke Organization endorsed J Rehabil Med 51, 2019