A computerized, self-administered test of verbal episodic memory in elderly patients with mild cognitive impairment and healthy participants: A randomized, crossover, validation study.

INTRODUCTION: Performance of "Revere", a novel iPad-administered word-list recall (WLR) test, in quantifying deficits in verbal episodic memory, was evaluated versus examiner-administered Rey Auditory Verbal Learning Test (RAVLT) in patients with mild cognitive impairment and cognitively normal participants. METHODS: Elderly patients with clinically diagnosed mild cognitive impairment (Montreal Cognitive Assessment score 24-27) and cognitively normal (Montreal Cognitive Assessment score ≥28) were administered RAVLT or Revere in a randomized crossover design. RESULTS: A total of 153/161 participants (Revere/RAVLT n = 75; RAVLT/Revere n = 78) were randomized; 148 (97%) completed study; 121 patients (mean [standard deviation] age: 70.4 [7.84] years) were included for analysis. Word-list recall scores (8 trials) were comparable between Revere and RAVLT (Pearson's correlation coefficients: 0.12-0.70; least square mean difference [Revere-RAVLT]: -0.84 [90% CI, -1.15; -0.54]). Model factor estimates indicated trial (P < .001), period (P < .001) and evaluation sequence (P = .038) as significant factors. Learning over trials index and serial position effects were comparable. DISCUSSION: Participants' verbal recall performance on Revere and RAVLT were equivalent.