Date Presented Accepted for AOTA INSPIRE 2021 but unable to be presented due to online event limitations.
We examined cognitive and motor factors predicting adherence to smartphone-based ecological momentary assessment (EMA) for monitoring daily life participation in stroke survivors. Cognitive flexibility and dexterity were significant predictors of EMA adherence. We derived cutoff values to differentiate survivors with high and low adherence. OTs may use them to guide the selection of survivors who can use mobile health technology to monitor poststroke functioning.
Primary Author and Speaker: Shijia Li
Additional Authors and Speakers: Stephen C. L. Lau, Joanne N. Chin
Contributing Authors: Alex Wong
PURPOSE: As the age of onset of stroke is decreasing and the neurologic severity is increasingly mild, more stroke survivors are discharged home with limited to no rehabilitation services, but their long-term participation restrictions persist (Wolf, Baum, & Conner, 2009). Smartphone-based ecological momentary assessment (EMA) has been increasingly used in research to provide greater ecological sensitivity and temporal resolution for understanding moods, thoughts, and behaviors of individuals in naturalistic settings (Shiffman, Stone, & Hufford, 2008). Our prior research has demonstrated the initial feasibility of using EMA for measuring daily functioning in stroke survivors (Lau et al., 2019). However, specific post-stroke motor and cognitive factors contributing to EMA adherence are not well understood, limiting EMA as a clinical tool because therapists do not know what aspects of post-stroke function are needed to successfully utilize this mobile assessment approach (Lemke, Rodríguez Ramírez, Robinson, & Signal, 2019). Therefore, this study aimed to (1) identify key cognitive and motor predictors of EMA adherence in stroke survivors, and (2) provide recommended cutoff values of these predicted cognitive and motor functions.
DESIGN: A longitudinal study involving laboratory-based cognitive and motor assessments, as well as EMA surveys via a mobile application five times a day for 14 days.
METHODS: Participants were recruited through a stroke registry at Washington University. Participants completed a battery of laboratory-based cognitive and motor assessments, as well as EMA surveys of daily life participation in their own environments. Participants with the lowest 25% and the highest 25% of overall EMA completion rates were classified into low and high EMA adherence groups, respectively. We conducted binary logistic regression models to independently identify cognitive and motor predictors of these groupings. All models were adjusted for patient demographics and other clinical characteristics. We then conducted receiver operating characteristic (ROC) curve analyses to select optimal cutoffs for the tests of significant predictors.
RESULTS: Of the 176 participants (mean age = 60.5; 45.5% female) who completed the full study protocol, 44 and 52 were classified into low- (mean = 45.6% completion) and high-adherence (mean = 96.2% completion) groups, respectively. Cognitive flexibility, as measured by the Trail Making Test-B (TMT-B), was the only significant cognitive predictor (B = .08, SE = .03, p = .028). The ROC curve analysis indicated the criterion value for TMT-B as a fully adjusted T-score of 47 with adequate accuracy for predicting EMA adherence (area under the curve (AUC) [95% CI] = .65 [.54, .75]). Dexterity, as measured by the 9-Hole Pegboard Test-Dominant Hand, was the only significant motor predictor (B =.08, SE = .04, p = .029). The ROC analysis resulted in the criterion value for this motor test as a fully adjusted T-score of 36 with adequate accuracy in predicting EMA adherence (AUC [95% CI] = .625 [.520, .723]).
CONCLUSION: This study has identified cognitive and motor functions that can help predict EMA adherence among individuals with stroke residing in the community. Understanding the unique motor and cognitive profiles of each individual will allow practitioners to make appropriate EMA use recommendations. Researchers and clinicians may use these cutoff scores when recommending mobile health monitoring for stroke survivors. As the trend toward telehealth increases, this study may provide insights for selecting more appropriate participants with whom to implement mobile health technology for monitoring post-stroke functioning.
References
Lemke, M., Rodríguez Ramírez, E., Robinson, B., & Signal, N. (2019). Motivators and barriers to using information and communication technology in everyday life following stroke: a qualitative and video observation study. Disability and Rehabilitation, 1–9. https://doi.org/10.1080/09638288.2018.1543460
Lau, S. C. L., Macpherson, V., Lenze, E. J., Baum, C., Lee, J., Metts, C. L., Yingling, M. D., Fucetola, R. P., Fong, M. W. M., Depp, C. A., Heaton, R. K., Lu, C., Lai, A. M., & Wong, A. W. K. (2019). Feasibility and validity of ecological momentary assessment of daily function in people after stroke. Archives of Physical Medicine and Rehabilitation, 100(12), e179. https://doi.org/10.1016/j.apmr.2019.10.051
Shiffman, S., Stone, A. A., & Hufford, M. R. (2008). Ecological momentary assessment. Annual Review of Clinical Psychology, 4, 1–32. https://doi.org/10.1146/annurev.clinpsy.3.022806.091415
Wolf, T. J., Baum, C., & Conner, L. T. (2009). Changing face of stroke: Implications for occupational therapy practice. American Journal of Occupational Therapy, 63(5), 621–625. https://doi.org/10.5014/ajot.63.5.621
