CAN TIMED UP-AND-GO (TUG) DUAL-TASK PERFORMANCE AID DIAGNOSIS OF DEMENTIA?

Åberg A.C.^1,2, Cedervall Y.¹, Giedraitis V.¹, Berglund L.³, Lennhed B.⁴, Rosendahl E.⁵, Mc Kee K.², Kilander L.¹

¹Uppsala University, Department of Public Health and Caring Sciences, Geriatrics, Uppsala, Sweden, ²Dalarna University, School of Education, Health and Social Studies, Falun, Sweden, ³Uppsala University, Uppsala Clinical Research Center, Uppsala, Sweden, ⁴Falu Lasarett, Department of Geriatrics and Rehabilitation Medicine, Falun, Sweden, ⁵Umeå University, Department of Community Medicine and Rehabilitation, Physiotherapy, Umeå, Sweden

Background: Impaired cognition is associated with reduced performance in mobility tests, including the Timed Up-and-Go (TUG). TUG is a well-established test of one mobility sequence: time is recorded when an individual starts from a seated position, stands up, walks 3 meters at comfortable speed, turns around walks back and sits down again. Moreover, dual-tasking such as combining a mobility test with a verbal task; TUG dual-task (TUGdt), is significantly more challenging for people with dementia, which may indicate that TUGdt can be useful as an aid in dementia diagnosis. Here we present initial results from an ongoing longitudinal study.

Purpose: To investigate if TUGdt outcome can differentiate those who get a dementia disease diagnoses from others in a population of persons attending specialist outpatient clinics for memory assessment.

Methods: Data collection at baseline (before diagnosis) included demographic data, depression screening and a battery of established cognitive and motor function tests. Data on diagnosis were collected from medical records. TUG and two types of TUGdt i.e. enumerating animals (TUGdt animal) and months of the year in reverse order (TUGdt months) were performed. These test were recorded by two synchronised video cameras (frontal and sideways viewing). Number of correct words for TUGdt:s were calculated from the videos and related to the time i.e. correct words per 10 seconds (s). Dual-task cost was calculated according to the formula: (TUGdt-TUG) x 100 TUG Associations between the outcome dementia (yes/no) and the TUG variables were examined in logistic regression models adjusted for age, gender and education.

Results: One hundred and sixty-nine adults were consecutively recruited from two memory clinics. Nine with non-dementia related diagnoses were excluded. Thus, 160 adults; 88 males, 62 with university education and age M=72 years (range 39-94) were included. Forty-one were diagnosed with any dementia diagnosis and 119 with mild (n=74) or subjective (n=45) cognitive impairment. For dual-task cost months, subjects with/without dementia attained M=34.2 (range -18.7—293.9) and 18.2 (-21.6—114.3), respectively (p>0.05) while dual-task cost animals resulted in 19.4 (-7.2—148.7) and 12.1 (-18.7—55.4), respectively (p>0.05). Mean (range) for TUGdt months per 10 s were 2.8 (0.0—8.8) and 5.0 (0.0—14.0) for subjects with/without dementia, respectively, and adjusted odds ratio (OR) for outcome dementia was 2.33 (95 % confidence interval (CI) 1.41-3.70, p 0.001) per one standard deviation decrease. For TUGdt animals per 10 s, M (range) were 2.9 (0.4—11.1) and 4.8 (0.0—9.8) for subjects with/without dementia, respectively, and adjusted OR was 11.90 for tertile 1 versus. tertile 3 (95 % CI 3.12-45.38, p 0.0001).

Conclusion(s): Unexpectedly, dual-task cost did not differ significantly between persons with/without dementia diagnoses. In contrast, number of correct words per 10s during TUGdt animal as well as TUGdt months showed capacity to discriminate between persons diagnosed with dementia and those with mild or subjective cognitive impairment.

Implications: The results indicate that TUGdt may be useful in the early process of investigating a suspected cognitive impairment, as an easily administered aid, alongside pure tests of cognitive function.

Funding acknowledgements: The Uppsala-Örebro Regional Research Council, Alzheimer Foundation, Promobilia Foundation, Thuréus Foundation for Geriatric Research and the Swedish Society of Medicine

Topic: Disability & rehabilitation

Ethics approval: The Regional Ethical Review Board in Uppsala has approved the study

All authors, affiliations and abstracts have been published as submitted.