A Machine Learning Approach to Predict Post-stroke Fatigue. The Nor-COAST study.

OBJECTIVE: This study aimed to predict fatigue 18 months post-stroke by utilizing comprehensive data from the acute and sub-acute phases after stroke in a machine-learning set-up. DESIGN: A prospective multicenter cohort-study with 18-month follow-up. SETTING: Outpatient clinics at 3 university hospitals and 2 local hospitals. PARTICIPANTS: 474 participants with the diagnosis of acute stroke (mean ± SD age; 70.5 (11.3), 59% male; N=474). INTERVENTIONS: Not applicable. MAIN OUTCOME MEASURES: The primary outcome, fatigue at 18 months, was assessed using the Fatigue Severity Scale (FSS-7). FSS-7≥5 was defined as fatigue. In total, 45 prediction variables were collected, at initial hospital-stay and 3-month post-stroke. RESULTS: The best performing model, random forest, predicted 69% of all subjects with fatigue correctly with a sensitivity of 0.69 (95% CI: 0.50, 0.86), a specificity of 0.74 (95% CI: 0.66, 0.83), and an Area under the Receiver Operator Characteristic curve of 0.79 (95% CI: 0.69, 0.87) in new unseen data. The proportion of subjects predicted to suffer from fatigue, who truly suffered from fatigue at 18-months was estimated to 0.41 (95% CI: 0.26, 0.57). The proportion of subjects predicted to be free from fatigue who truly did not have fatigue at 18-months was estimated to 0.90 (95% CI: 0.83, 0.96). CONCLUSIONS: Our findings indicate that the model has satisfactory ability to predict fatigue in the chronic phase post-stroke and may be applicable in clinical settings.

Physical Activity Behavior and Its Association With Global Cognitive Function Three Months After Stroke: The Nor-COAST Study†.

OBJECTIVE: The purposes of this study were to determine the association between physical activity (PA) behavior and global cognitive function 3 months after stroke and to explore the role of physical capacity as a mediating factor. METHODS: Participants with stroke were successively recruited at 5 different hospitals in Norway. PA was measured using accelerometers, with a follow-up period of 7 consecutive days, and global cognitive function was assessed using the Montreal Cognitive Assessment (MoCA). The general pattern of PA and the percentage of participants adhering to World Health Organization PA recommendations (at least 150 minutes of moderate-intensity aerobic PA per week) were investigated using descriptive statistics. Multiple regression and mediator analyses were used to examine the relationship between PA behavior and MoCA scores; physical capacity, measured with the Short Physical Performance Battery, served as the mediating variable. RESULTS: A total of 193 women (42.6%) and 260 men (57.4%) with a median age of 73.7 years (25th and 75th percentiles = 65.8 and 80.4, respectively) and a median MoCA score of 25 points (25th and 75th percentiles = 22 and 27, respectively) were included. Mean total time spent walking at moderate intensity was 251.7 (SD = 164.6) min/wk (mean bout length = 20.9 [SD = 7.3] seconds), which indicated 69.3% adherence to World Health Organization guidelines. With each point decrease in the MoCA score, there was an expected 8.6% increase in the odds of nonadherence to PA recommendations. Physical capacity was identified as an important mediating factor, explaining the strength of the association between cognition and PA behavior. CONCLUSIONS: In contrast to previous research, in the present study, most participants adhered to the updated global PA guidelines. However, people who had survived stroke and had reduced cognitive function were at higher risk of inactivity, an association mediated by physical capacity. IMPACT: A better understanding of the association between cognition and PA behavior after stroke might help for developing more targeted early-onset interventions.