16 August 2018
Despite neuropsychological differences between individuals with Alzheimer’s disease (AD) and dementia with Lewy bodies (DLB), overlap in performance has been observed, including analogous scores in memory tests (Rey Auditory Verbal Learning, AVLT; e.g., ). All the while, differential diagnosis remains a costly exercise. However, a recent report  has shown that several cognitive scores, including some derived from the AVLT, discriminate between AD (n=32, age=72.2 ± 8.3) and DLB (n=29, age=74.7 ± 7.5), with effect sizes (Cohen’s d) ranging from 0.600 (verbal learning) to 0.870 (rotation).
One of the main findings of this report  was that recency performance (i.e., memory for items at the end of the list) at trial 1 of the AVLT was better in AD than in DLB (d=0.652), which is consistent with a vast literature on serial position effects in AD (e.g., ), including our own work with the recency ratio (Rr) score . The principle of Rr is that since immediate recency tends to be high in AD, only for delayed recency to be usually low, a ratio between these two measures will be higher in individuals with AD than controls, as well as higher in subjects at risk of decline. We therefore expected that Rr would be highly effective in discriminating between AD and DLB by exploiting this principle.
To test our hypothesis, we calculated Rr from the data reported in , using this formula: (trial 1 recency + 1) / (delayed trial recency + 1). Our results show that (natural log transformed) Rr is significantly higher in AD than in DLB (t(58)=4.114, p<0.001) and yields the largest effect size (d=1.060) compared to all other measures above. Similarly, the area under the ROC curve is higher for Rr (0.774) than for trial 1 recency alone (0.670).
Additionally, Rr was associated with the Mini-Mental State Exam score in individuals with AD (r=-0.511, p=0.003), but not in individuals with DLB (r=0.113, p=0.560). These results indicate that Rr scores were also sensitive to the degree of generalised cognitive decline in the AD group.
All in all, these findings support the use of Rr as a cognitive measure for the discrimination between AD and DLB. For reference, an Rr score of 3+ yields a positive predictive value for AD (against DLB) of 88%, and a negative predictive value of 61%, in this sample.
Davide Bruno1,*, Cinzia Busse2,3, Annachiara Cagnin2,4
1School of Natural Sciences and Psychology, Liverpool John Moores University, Liverpool, UK
2Department of Neurosciences, University of Padua, Padova, Italy
3Neurofarba, University of Florence, Firenze, Italy
4IRCCS San Camillo Hospital Foundation, Venezia, Italy
*Corresponding author: D.Bruno@ljmu.ac.uk
No conflicts of interest to disclose.
This research was presented at the International Neuropsychological Society meeting in Prague, July 2018.
 Perri R, Fadda L, Caltagirone C, Carlesimo GA (2013) Word list and story recall elicit different patterns of memory deficit in patients with Alzheimer's disease, frontotemporal dementia, subcortical ischemic vascular disease, and Lewy body dementia. J Alzheimers Dis 37, 99-107.
 Bussè C, Anselmi P, Pompanin S, Zorzi G, Fragiacomo F, Camporese G, Di Bernardo GA, Semenza C, Caffarra P, Cagnin A (2017) Specific verbal memory measures may distinguish Alzheimer’s disease from dementia with Lewy bodies. J Alzheimers Dis 59, 1009-1015.
 Carlesimo GA, Sabbadini M, Fadda L, Caltagirone C (1995) Different components in word-list forgetting of pure amnesics, degenerative demented and healthy subjects. Cortex 31, 735-745.
 Bruno D, Koscik RL, Woodard JL, Pomara N, Johnson SC (2018) The recency ratio as predictor of early MCI. Int Psychogeriatr, doi: 10.1017/S1041610218000467.