IQCODE-Brief Version in Stroke Patients.

INTRODUCTION

Post-stroke cognitive impairment is frequently seen after stroke. We aimed to examine the psychometric properties, factorial structure, and screening utility of the IQCODE brief version in stroke patients using an informant-based approach to cognitive impairment screening.

METHODS

Patients aged older than 18 years with ischemic or hemorrhagic stroke were included. Within one week of completing the MoCA test, each patient's informant (spouse, caregiver, etc.) was contacted by phone, and the IQCODE brief version was administered. Cronbach's alpha, test-retest analysis, exploratory factor analysis (EFA), and confirmatory factor analysis (CFA) were performed to assess the reliability and construct validity of the IQCODE brief version.

RESULTS

A total of 179 patients with stroke (mean age: 58.6 ± 13.7 years) were included, 18 (10%) of them had hemorrhagic stroke. The IQCODE brief version demonstrated excellent internal consistency (Cronbach's alpha = 0.94). Factor analysis suggested a preliminary four-factor structure, explaining 76.5% of the total variance. CFA confirmed a good model fit (RMSEA = 0.07, CFI = 0.973, TLI = 0.964). Review of the factor loadings resulted in the exclusion of three items from the initial model. ROC analysis for the 16-item version yielded an area under the curve (AUC) of 0.771 (95% CI, 0.70 - 0.84), with an optimal cut-off score of 48.5 (sensitivity: 72.4%; specificity: 71.1%). A refined 13-item version demonstrated improved specificity (76%) without compromising sensitivity.

CONCLUSIONS

Both the 16-item and the proposed 13-item IQCODE brief versions demonstrated promising preliminary psychometric properties, and they may serve as effective screening tools. Their feasibility for telephone administration and strong psychometric properties support their use in routine clinical practice and remote screening.