Timed Up and Go test phases as predictors of future falls in community-dwelling older adults

Introduction: The Timed Up and Go (TUG) is a test widely used to assess the risk of falls in older adults. Although it is a complex task, only the total TUG time has been used for evaluation. The widespread use of smartphones has provided the development of applications for monitoring diagnostic procedures. Objective: To analyze the ability to predict future falls in older adults. Methods: A cohort study (1 year) of 42 participants using the sTUG Doctor. Fall events during 1-year follow-up were monitored by telephone. The number of days between assessment and first fall or last contact was calculated for survival analysis, assessed by unadjusted and adjusted Cox proportional hazards regression models. Tests with p <5 % were considered statistically significant and between 5% and 10% were indicative of significance (Epi-Info™ 7.2). Results: Falls were observed in 22 (52.38%) participants (fallers). The results indicated that cognitive impairment, depressive symptoms, women, and participants with fear of falling (FES-I) were more likely to fall. Fallers performed worse on all sTUG Doctor phases. Hazard ratios for predicting falls were significant for total TUG time (1.35; p = 0.029) and total number of steps (1.52; p = 0.057). Total TUG time remained significant when adjusted for sex, age group, FES-I, and depression level. Conclusion: The sTUG Doctor was an important tool to predict falls in community-dwelling older adults.


Introduction
The sensory system is responsible for starting the process of developing human body balance and is one of the first systems to undergo changes with the aging process. 1 With aging, the human sensory system is affected by reduced functional reserve in older people and by the diseases that frequently affect this age group.
Thus, several stages of postural control are altered, reducing the compensatory ability to maintain balance and, consequently, favoring postural disorders. 2 Along with aging, the musculoskeletal system also undergoes physiological changes that compromise its functioning. These changes are characterized by a decline in physical capacity related to decreased muscle strength, flexibility, agility, coordination, and joint mobility, which lead to increased postural instability and, consequently, to an increased risk of falls. [3][4][5] Given this context, several instruments are used to assess body balance in older people and, consequently, the risk of falls. Among them, the Timed Up and Go (TUG) test stands out because it is an easy-to-apply, low-cost test widely used in clinical and research settings. 6 Currently, the widespread use of smartphones, devices with sophisticated sensors, has provided the development of applications for monitoring diagnostic procedures. 7 Sensors such as accelerometers, gyroscopes, and magnetometers embedded in mobile devices are an inexpensive way to conduct studies of this magnitude. In addition, they have a high level of efficiency and have been used in several studies. [8][9][10] In this context, Milosevic et al. 11 implemented a smartphone application (Mobility Suite®) that includes the Smart Timed Up and Go (sTUG) Doctor. The sTUG Doctor application, in addition to the total test time, also evaluates body posture transitions during the TUG test. These quantifications allow us to better assess body kinematics and dynamics, obtaining parameters not yet explored in fall risk prediction. 11 The sTUG Doctor provides instantaneous feedback with the most relevant parameters to the user in the form of a report on the screen. 11 An advantage of the sTUG Doctor over other sensors is cost-effectiveness, as it only requires a smartphone, a device that is currently inexpensive and used daily by more than half of Brazilians, including older adults. 12 However, its study is recent, and it has been used basically in the field of research. Therefore, exploring this application becomes even more important to support the future introduction of this tool in other environmental contexts, such as in the home and hospital settings.  The TUG test was used as the main evaluation tool.
It is a functional mobility test widely used in research settings to assess fall risk. It evaluates gait performance, body posture transitions, and changes in direction during walking by measuring the time in seconds a person takes to complete the test. The test consists of the following tasks: rising from an armless chair with a backrest, walking a distance of 3 meters, turning around, walking back to the chair, and sitting down. 21 The time taken to complete the test generates a fall risk classification, as follows: low risk (< 10 seconds), moderate risk (10 -20 seconds), and high risk (> 20 seconds). 22 The TUG test was performed through the sTUG

Results
A total of 42 older adults participated in the study. Table 1 shows the distribution of the sociodemographic, clinical, and lifestyle characteristics of the participants according to the fall event during follow-up. Of those evaluated, 52.38% had a fall event. The rate of falls was higher in women (58.82%) than in men (25%). The mean age of fallers was higher. Regarding age group, the rate of oldest-old fallers was proportionally higher. Fallers obtained higher mean FES-I scores, lower mean MMSE scores, and higher mean GDS-15 scores. On average, fallers needed more time and took more steps to complete the total test. In addition, they had higher values for the sit-to-stand transition, stand-to-sit transition, and maximum angle change. Conversely, they had lower values for maximum angular velocity. In this study, no older adult was classified as being at high risk of falling. However, of those evaluated, 76.92% of those classified as being at moderate risk of falling fell during follow-up (Table 2). Table 3 shows the hazard ratios calculated by Cox regression to predict falls at each follow-up month, unadjusted and adjusted models for sTUG Doctor components. FES-I and MMSE were the variables that most influenced fall risk prediction by total TUG duration.
This finding demonstrates that fall risk prediction by total TUG duration is influenced but not dependent on the differences between FES-I levels. However, the presence of the MMSE in the model reduced this chance. This result indicates that fall risk prediction by total TUG duration is dependent on the MMSE level.
According to the variable 'number of steps,' GDS-15 and MMSE were the variables that most influenced fall risk prediction. These results demonstrate that fall risk prediction by reduced number of steps is dependent on the GDS-15 and MMSE levels. Results in bold represent significant tests or indicative of significance.
Comparisons between means were tested by unpaired Student's t test and between frequencies by the chi-square test.

Discussion
The results of the present study showed that the TUG test phases that best predicted falls in communitydwelling older adults were total TUG duration and the number of steps during the TUG test. According to Muir et al., 25 in general, the gait in older adults is associated with lower speed, shorter step length, and greater base of support compared with young adults. These gait changes may be a strategy to increase stability or a consequence of decreased muscle strength and poor physical performance. 25 In the present study, these changes were not sufficient to prevent falls, as they remained significant for fall risk prediction. In addition, other strategies used to reduce the risk of falling include decreased impulse or initial contact, knee extension, widening and narrowing the base of support, and decreased step length and height, with consequent reduction of speed. 26 The present study addressed falls as a postassessment event, where more than half of the followedup older adults fell at least once during the 1-year follow-up. The frequency of falls observed among the participants in this study was higher than that estimated by Brazilian longitudinal studies (25% to 35%). 3,28 It is worth noting that the present study started in  We highlight that, to date, no other study was found in the literature that used the assessment of the TUG test phases through the sTUG Doctor application to predict future falls in older adults. Given this scenario, the originality of the research is highlighted, as well as the methodological rigor with which it was conducted.
Another relevant point to be highlighted is that, based on the characteristics of the study sample, it is believed that the participants obtained relatively good results because they are healthy community-dwelling older adults, most of whom are physically active, and, possibly, also because of the mean age of the sample.
Finally, the sample size is pointed out as a limitation of the study, since, due to the COVID-19 pandemic, the expected sample size was not reached. Reaching the expected sample size means that it is large enough to provide a good approximation or an estimate for the behavior of the entire population. That is, we have a representative sample of the study population. Therefore, since the results are limited to the findings presented here, it is expected that, in a more favorable context, this research can be replicated with a more satisfactory sample size and including frailer older people, such as institutionalized older adults.

Conclusion
Fallers performed worse than non-fallers on all TUG test phases. The TUG test phases that best predicted falls were total TUG duration and number of steps, a finding that could be extracted through the use of a smartphone application. Older women, those with fear of falling, cognitive impairment, and depressive symptoms were more likely to fall.