Older adults' driving exposure during COVID-19 and its relationship with cognition
Restricted (Penn State Only)
- Author:
- Tian, Joanne
- Graduate Program:
- Human Development and Family Studies
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- August 19, 2024
- Committee Members:
- Heidemarie Laurent, Professor in Charge/Director of Graduate Studies
David Almeida, Major Field Member
Alyssa Gamaldo, Chair & Dissertation Advisor
Yiqi Zhang, Outside Unit & Field Member
Lesley Ross, Special Member
Zachary Fisher, Major Field Member - Keywords:
- Older adults
driving exposure
cognition
EMA
cognitive training
COVID-19 - Abstract:
- Age-related cognitive decline significantly influences daily activities for older adults, particularly in areas such as driving. Various theoretical frameworks have emphasized the importance of cognitive function, especially processing speed, in driving among older adults. Reduced driving exposure, usually defined as decreased time or distance an individual drives a vehicle, consistently correlates with lower processing speed performance. Previous research has observed that many older adults chose to limit their life space during the COVID-19 pandemic to reduce their risk of infection, which likely impacted their driving exposure. Moreover, older adults, especially those with cognitive impairments, experienced a sharp cognitive decline during this period. However, limited research has explored the association between cognition and driving exposure during the critical period of COVID-19. This dissertation aimed to address this research gap through two separate studies involving community-dwelling older adults. Study 1 examined the influence of COVID-19 on older adults’ driving exposure. Specifically, Study 1 utilized the secondary data from the Everyday Function Intervention Trail (EFIT) and the Elucidating the Necessary Active Components of Training (ENACT; N = 310, age range from 55 to 87) to test two aims. The first aim was to investigate the relationship between older adults’ perceived risks related to COVID-19 and driving exposure. Linear and ordinal logistic regression models revealed nonsignificant associations between driving exposure and COVID-19 perceived risks in older adults (driving mileage: b = 1.07; p > .05; driving frequency: b = 0.04; p > .05). However, a significant interaction effect of COVID-19 perceived risks and age was observed when predicting driving frequency (b = -0.01; p < .01). Decomposition of the interaction suggested that younger-aged older adults were more likely to increase their driving frequency when they perceived more COVID-19-related risks (OR = 1.10, 95% CI = 1.02, 1.18), particularly in relation to their family or friends within their direct environment. No significant interaction effects were observed when predicting driving mileage (COVID-19 and age: b = -0.36, p > .05; COVID-19 and gender: b = -2.29, p > .05). The second aim of Study 1 was to investigate the moderation effect of COVID-19 perceived risks on the existing relationship between processing speed and driving exposure. Results suggested that older adults with better performance on the Trail Making Test B (TMT B) were significantly less likely to drive frequently when they perceived less COVID-19-related risk (OR = 0.993, 95% CI = 0.986, 1.000). However, there were no significant interaction effects with other processing speed measurements or when predicting driving mileage (p > .05). Recognizing the challenges in detecting the relationship between processing speed and driving exposure among cognitively healthy older adults using traditional psychometric measurements, Study 2 utilized the Ecological Momentary Assessment (EMA) to collect more sensitive assessments of cognitive functioning across days within participants’ typical settings (e.g., home environment). Previous research has found that greater intraindividual variability (IIV) of cognition, as measured by EMA, is a sensitive indicator of subtle cognitive declines in older adults. In Study 2, mobile Symbol Search and UFOV (mUFOV) assessments were measured daily to examine the IIV of processing speed. Different from the original desktop version of UFOV measures, the distance between the selected and the correct location on the peripheral task was calculated to indicate mUFOV performance. mUFOV peripheral errors with a difficult central task and a medium central task were included in Study 2 based on their significant associations with the original UFOV scores. Study 2 used the secondary data from EFIT (N = 110, age range from 65 to 87) to test two aims. The first aim was to investigate the impact of driving exposure on IIV in processing speed performance. Results of heterogeneous variance multilevel models (MLM) suggest that higher reported driving mileage was significantly correlated with less IIV in mUFOV peripheral errors (bs = 0.998), but was not significantly correlated with IIV in Symbol Search performance (p > .05). The second aim of Study 2 was to examine the effectiveness of processing speed training in improving processing speed performance using EMA measurements, and whether the training effects differ depending on driving exposure. Results of heterogeneous variance MLM suggest that some of the within-person variances of processing speed performance were explained by the interaction between intervention groups and assessment phases (Symbol Search: χ^2 [5] = 247.70, p < .001; difficult mUFOV: χ^2 [5] = 236.23, p < .01; Medium χ^2 [5] = 236.23, p < .01). The IIV for both Symbol Search performance and mUFOV peripheral error (particularly with a medium central task) displayed a more stable pattern of change in the processing speed training group compared to the control group. However, minimal differences were observed in the IIV in mUFOV peripheral error with a difficult central task between the control and training groups. Additionally, older adults with less reported driving frequency had greater improvement on the average performance of mUFOV difficult task (b = -7.69, p = .03). Through these two studies, this dissertation provides new insights into the relationship between cognitive function (i.e., processing speed) and driving exposure, especially within the context of the societal impact of COVID-19. Study results also support the promising utilization of the EMA approach in strengthening the scientific understanding of cognitive changes in community-dwelling older adults. This dissertation also provides valuable implications on strategies to support and address the unique challenges older drivers are facing during the time of a critical public health crisis.