Table 3

Cognitive domain scores at baseline and follow-up in controls and patients either with or without an interval stroke/TIA and interactions (each patient group vs controls×time)*

Domain	Controls		Patients with interval stroke/TIA				Patients without interval stroke/TIA
	Baseline	Follow-up	Baseline	Follow-up	Interaction		Baseline	Follow-up	Interaction
	Baseline	Follow-up	Baseline	Follow-up	Parameter estimate	p Value	Baseline	Follow-up	Parameter estimate	p Value
Attention	−0.11±0.13	−0.19±0.14	−0.51±0.19	−1.19±0.23	−0.60±0.25	0.01	−0.83±0.11	−0.73±0.13	0.17±0.17	0.32
Verbal memory	−0.10±0.10	0.16±0.13	−0.08±0.16	−0.27±0.21	−0.44±0.21	0.03	−0.12±0.09	−0.41±0.12	−0.54±0.14	<0.001
Visual memory	−0.07±0.11	0.31±0.12	−1.00±0.17	−1.04±0.20	−0.41±0.21	0.06	−0.86±0.10	−0.48±0.12	0.00±0.15	0.98
Processing speed	−0.10±0.16	−0.02±0.17	−1.05±0.25	−1.70±0.29	−0.72±0.23	0.002	−1.09±0.15	−1.14±0.17	−0.12±0.16	0.44
Reasoning	−0.11±0.12	0.06±0.13	−0.99±0.19	−0.86±0.21	−0.04±0.22	0.84	−1.11±0.11	−0.73±0.12	0.21±0.15	0.16
Working memory	−0.10±0.10	−0.17±0.11	−0.61±0.15	−1.01±0.18	−0.33±0.16	0.03	−0.66±0.09	−0.69±0.10	0.04±0.11	0.70
Language	−0.01±0.15	0.16±0.17	−0.33±0.22	−1.33±0.28	−1.16±0.27	<0.001	−0.82±0.13	−0.83±0.16	−0.17±0.18	0.36
Praxis-gnosis	−0.08±0.18	0.03±0.14	−0.92±0.28	−0.57±0.24	0.23±0.36	0.53	−0.92±0.16	−0.47±0.14	0.32±0.25	0.20
Visuoconstruction	−0.11±0.13	−0.14±0.14	−1.16±0.21	−1.50±0.24	−0.30±0.22	0.18	−0.76±0.11	−0.87±0.13	−0.6±0.15	0.66
Mental flexibility	−0.10±0.27	−0.19±0.22	−1.27±0.45	−2.14±0.42	−0.78±0.50	0.11	−2.14±0.25	−1.64±0.23	0.59±0.32	0.06
Verbal fluency	−0.10±0.10	−0.11±0.12	−0.85±0.15	−1.16±0.20	−0.29±0.20	0.14	−0.89±0.09	−0.85±0.12	0.04±0.14	0.74
Composite score	−0.10±0.09	−0.03±0.10	−0.75±0.14	−1.22±0.15	−0.54±0.10	<0.001	−0.96±0.08	−0.84±0.08	0.04±0.07	0.56

*Mean scores (±SE) are estimated mean values, adjusted for age, sex and education. Parameter estimates (±SE) are for the interaction terms (patient group vs controls×time) of the linear mixed models. A negative parameter estimate indicates a greater decline in patients than controls, and a positive parameter estimate indicates a smaller decline in patients than controls. p Values are for the statistical significance of the interaction parameter estimates. Analyses were repeated with outliers excluded: a significant decline in verbal fluency for patients with an interval stroke/TIA (p=0.016) was the only difference to the presented results.
TIA, transient ischaemic attack.