TUGA

: a nonverbal screening test for cognitive impairment and dementia

Abstract

In a longitudinal study that lasted 2 years, we assessed 150 subjects (91 control group / 59 dementia group) and compared their results in an experimental nonverbal test (TUGA), for cognitive impairment and dementia, with the results obtained by the same groups in two major screening tests, the Montreal Cognitive Assessment (MOCA) and the Addenbrooke Cognitive Examination Revised (ACE-R). To validate our test and to confirm the frontal assessment characteristics and the non-verbal abstract reasoning nature of TUGA, we also correlated the scores obtained, with the results in two other tests, the Frontal Assessment Battery (FAB) and the Raven Progressive Matrices Standard (RPM-std). Finally, to have a characterization of the level of autonomy of our dementia group and follow any changes on this dimension throughout the study, we have also applied the Barthel Activities of Daily Living Index (BI). The result showed that:(1) TUGA total scores have a strong correlation with MOCA (r=,796, p ≤ ,001) and ACER (r=,761, p ≤ ,001) total scores and a moderate correlation with FAB(r=,551, p ≤ ,001). (2) For an optimal cutoff score of 7.5, TUGA had a specificity of 80% and a sensitivity of 78%, with statistically significant differences with MOCA and ACE-R. (3) The evidences show that in both moments of evaluation, TUGA (78,0% -96,6%), is not only more sensitive detecting cognitive impairment related dementia, but detects it earlier than ACE-R (6,8% - 66,1%) and MOCA (3,4% - 22,4%). (4) TUGA is as sensitive to dementia patients with frontal lobe deficits and/or with psychomotor slowing. (5) The individual Deck observation of TUGA results, gives useful qualitative and quantitative information about the possible etiology of the scores. (6) TUGA total scores have a moderate correlation with RPM-std (r=,526, p ≤ ,001) and as expected, a very strong correlation with TUGA Deck D RPM-std (r=,914, p ≤ ,001), opening a wide range of clinical possibilities and application areas. These results become even more relevant if we consider the simplicity of TUGA tasks.

Date of Award	4 Apr 2017
Original language	English
Awarding Institution	Universidade Católica Portuguesa
Supervisor	Filipa Ribeiro (Supervisor)