Determinants of disability at 6 months after stroke: The GRECogVASC Study

Background and purpose: The aim of this study was to determine the contributions of background disorders responsible for participation restriction as indexed by a structured interview for the modified Rankin Scale (mRS-SI).

Methods: A subset of 256 patients was assessed at 6 months after stroke using the National Institutes of Health Stroke Scale (NIHSS), gait score, comprehensive cognitive battery (yielding a global cognitive Z-score), behavioral dysexecutive disorders (DDs), anxiety and depressive symptoms, epilepsy, and headache. Following bivariate analyses, determinants of participation restriction were selected using ordinal regression analysis with partial odds.

Results: Poststroke participation restriction (mRS-SI score > 1) was observed in 59% of the patients. In bivariate analyses, mRS-SI score was associated with prestroke mRS-SI score, 6-month NIHSS score, gait score, global cognitive Z-score, behavioral DDs, and presence of anxiety and depression (all: p = 0.0001; epilepsy: p =0.3; headache: p = 0.7). After logistic regression analysis, NIHSS score was associated with increasing mRS-SI score (p = 0.00001). Prestroke mRS-SI score (p = 0.00001), behavioral DDs (p = 0.0008) and global cognitive Z-score (p = 0.01) were associated with both mRS-SI score > 1 and mRS-SI score > 2. In addition, gait score was associated with mRS-SI score > 2 (p = 0.00001). This model classified 85% of mRS-SI scores correctly (p = 0.001). Structural equation modeling showed the contributions of gait limitation (standardized coefficient [SC]: 0.68; p = 0.01), prestroke mRS-SI (SC: 0.41; p = 0.01), severity of neurological impairment (SC: 0.16; p = 0.01), global cognitive Z-score (SC: -0.14; p = 0.05), and behavioral DDs (SC: 0.13; p = 0.01).

Conclusion: These results provide a statistical model of weights of determinants responsible for poststroke participation restriction and highlight a new independent determinant: behavioral DDs.