Development and validation of the Online Rivermead Behavioral Memory Test (OL-RBMT) for people with stroke.

The purpose of this paper was to examine the reliability and validity (concurrent and construct) of a newly developed online Chinese version of the Rivermead Behavioral Memory Test (OL-RBMT) and its equivalence with the face-to-face version (FTF-RBMT). The OL-RBMT and FTF-RBMT were administered to 30 subjects with stroke in a two-week interval to establish their test-retest reliability, as well as to compare the two tests' equivalence. The OL-RBMT was further compared with another 30 age- and gender-matched, non-stroke patients to establish its construct validity. Its concurrent validity was established by computing scores with that of the Chinese version of Neurobehavioral Cognitive Status Examination (NCSE or Cognistat). The intra-class correlation for test-retest reliability of the OL-RBMT was 0.94 (P < 0.01). Pearson's correlation coefficients between the subscores of OL-RBMT and FTF-RBMT ranged from 0.84 to 0.93 (P < 0.01). Statistically significant correlation was found between OL-RBMT and NCSE scores (R = 0.797, P < 0.001), as well as in the differences in OL-RBMT scores between the stroke and non-stroke groups (Z = -4.041, P < 0.001). We consider that the reliability and validity of this newly developed online version of RBMT was established. Usability of the OL-RBMT was also discussed.