Journal of Rehabilitation Medicine 51-8 | Page 27

Psychosocial well-being after stroke No change was detected in the SOC after the inter- vention. However, the results indicated that a higher SOC level might have a protective effect against the potentially stressful situation imposed by a stroke. The SOC is considered to be strongly related to perceived health (16). Despite the adverse health experiences after a stroke, participants with higher SOC scores had higher odds of a normal mood and higher scores regarding health-related quality of life at T2. This result is consistent with Antonovsky’s (16) theory on saluto- genesis, which explains how some people manage well despite adverse health experiences. This finding is also supported by a recent study on how personal factors, such as proactive coping, influence stroke outcomes, seem to be stable over time (33) and demonstrated that in the period from 2 months to 2 years after stroke personal factors did not improve naturally. Our results differed from those reported in a com- parable study that evaluated the effect of motivational interviewing on mood (11). Watkins and colleagues reported a significant difference between the interven- tion and control groups at 3 months, with 49% of the participants in the intervention group and 39% of the participants in the control group having normal mood at 3 months. At 12 months, the difference between the groups remained significant, with 48% of participants in the intervention group and 37% of participants in the control group reporting normal mood. Watkins’ study was conducted in a hospital with a different and shorter intervention, and the data were primarily collected by mail. Comparably, a substantially higher percentage of participants in both groups (59.6%) had normal mood at 6 months in our study. Although the results failed to demonstrate effectiveness of this dialogue-based inter- vention for promoting psychosocial well-being after stroke, conducting an RCT in a complex setting with face-to-face data collection imposes several factors that may explain the neutral results obtained in this study. The Norwegian authorities have encouraged re- habilitation services delivered in the municipalities through the Coordination Reform (12) and through the clinical guidelines for treatment and rehabilitation after stroke (19). A considerable proportion of participants reported receiving rehabilitation services post-stroke (66% at T1 and 55% of T2). Therefore, the rehabilita- tion needs of some participants may have already been met through usual care. No standardized psychological support exists in the follow-up after stroke in Norway and we lack a complete overview of the content of the follow-up that the participants received. However, since the participants lived in more than 70 different municipalities, we must assume that the follow-up varied substantially. 563 The most frequently reported follow-up was physical therapy (59% at T1, 36% of T2). This finding was sup- ported by a recently conducted study comparing stroke rehabilitation in Norway and Denmark, suggesting that follow-up focuses more on physical rehabilitation in Norway, while psychological support is better organi- zed and implemented in Denmark (34). We experienced that some of the participants strugg- led to distinguish ordinary healthcare services from the intervention, and participants in the control group perceived the study participation and the structured as- sessment interviews as a form of intervention. Conduc- ting face-to-face assessments ensured high compliance and complete data during collection, but also had some disadvantages. Face-to-face interviewing is a form of social interaction and merely being involved in these interviews may have positively affected the patients’ psychosocial well-being. Having an interviewer visit their homes 1 and 6 months after discharge may have also led to the perception of having received an inter- vention, even by those in the control group. Social desirability bias (35) is another factor that potentially affected the results. If the respondents had acted in ways or held attitudes that they felt were undesirable, their answers might have been affected. Consequently, they might have underreported socially undesirable attitudes and behaviours while enhancing attitudes and behaviours they believed were expected in coping with their life after stroke. Stroke recovery is multifaceted. Finding the optimal instrument to capture the impact of a psychosocial intervention rather than the expected natural recovery after stroke was challenging. Recovery depends on dif- ferent mechanisms and treatments at different phases after the acute injury, which range from hours to many months. Although improvement varies among indivi- duals, most patients improve during the first period following the stroke because of acute care treatment and post-lesional plasticity (36). Another issue is a possible ceiling or floor effect of the outcome measures, which is a known risk with instruments addressing aspects of psychosocial well- being (37). Many participants had no or minor stroke symptoms and limited adverse effects on their daily living activities after their stroke. This outcome resul- ted in maximum scores on several questions, especially on the SAQOL-39g. Furthermore, well-being is a subjective phenome- non. Physical sequelae, side-effects of medication, and overall spontaneous improvement may affect how patients answer questions immediately after a stroke compared with 6 months post-stroke. This personal reporting of subjective appraisal introduces the possibi- J Rehabil Med 51, 2019