GPS helps detail outdoor activities in elderly with dementia

Activity patterns differ between cognitively healthy older adults and people with dementia (PwD), a recent study has found. The former tend to engage more in sports, while the latter undertake medically related activities.

“This study presents a framework for comparing outdoor mobility patterns of cognitively intact older adults and older adults with dementia based on a comprehensive set of global positioning system (GPS)-based indicators,” the researchers said.

“To provide an objective and inclusive view of the outdoor mobility of older adults with dementia, we focused on mobility indicators derivable from GPS data and domain knowledge,” they added.

A total of seven PwDs and eight cognitively intact controls (CTLs) were enrolled and made to carry a GPS device while travelling outside their homes. A framework incorporating 12 GPS indicators was applied, assessing spatial, temporal, and semantic dimensions of outside travel. The study lasted for 4 weeks; all enrolled participants were at least 65 years of age.

Participants contributed 27–28 days’ worth of GPS data, which showed that PwDs tended to spend more time at home (1,301.92±64.10 vs 1,221.08±83.71 minutes; p=0.06) and less time on outdoor activities (112.64±32.48 vs 149.21±66.48 minutes; p=0.2) than CTL controls, though the differences failed to reach statistical significance. [Gerontology 2021;doi:10.1159/000515391]

Average daily walking time, however, was significantly lower among PwDs (p=0.02).

In terms of timing, both groups tended to be most active during mornings and at noon. PwDs, in particular, preferred trips that started at noon. Although PwDs consistently took less trips during early mornings, mornings, and evenings, the differences relative to CTLs were not statistically significant.

Looking at the semantic dimensions of outdoor activities, the researchers then showed that PwDs and CTLs engaged in different types of outdoor activities.

For instance, PwDs attended a significantly higher proportion of medical- and health-related activities than CTLs (p=0.01) while simultaneously engaging in lower levels of sports (p=0.01). Other types of activities, such as those related to religion, shopping, food, and leisure, were not different between groups.

“The present study suggests that to overcome the apparent limitations of traditional life space measures and to objectively assess mobility behaviours of cognitively impaired populations, GPS technology can be used,” the researchers said.

“The proposed framework finds that in addition to determining the extent of the movements in the environment, which has been a primary focus of articles on older adult GPS-based life space assessments, GPS tracking devices can be used to automatically infer more in-depth information about the user’s mobility behaviour,” they added. This additional information can range from the timing of outdoor trips and the types of activities they typically engage in.

Future studies can develop machine learning algorithms that can predict movement patterns in this population, so assistance can be prepared in advance for those whose mobility diverges from what is expected, the researchers added. Such developments could help improve intelligence navigation systems and in turn improve support for outdoor mobility in older adults with impaired cognition.