Davina Binng1, Moriah Splonskowski1, Claudia Jacova2. 1. School of Graduate Psychology, Pacific University, Hillsboro, Oregon, USA. 2. School of Graduate Psychology, Pacific University, Hillsboro, Oregon, USA, cjacova@pacificu.edu.
Abstract
INTRODUCTION: Distance or remote cognitive assessments, administered via phone or computer platforms, have emerged as possible alternatives to traditional assessments performed during office visits. Distance refers to any nontraditional assessment feature, not only or necessarily location. We conducted a systematic review to examine the psychometric soundness of these approaches. METHOD: We searched PubMed, PsycINFO, AgeLine, and Academic Search Premier for articles published between January 2008 and June 2020. Studies were included if participants were over the age of 50, a structured assessment of cognitive function in older adults was evaluated, the assessment method was deemed distant, and validity and/or reliability data were reported. Assessment distance was defined as having any of the following features: use of an electronic test interface, nonroutine test location (e.g., home), test self-administered, and test unsupervised. Distance was categorized as low, intermediate, or high. RESULTS/DISCUSSION: Twenty-six studies met inclusion criteria. Sample sizes ranged from n = 8 to 8,627, and the mean age ranged from 57 to 83. Assessments included screens, brief or full batteries, and were performed via videoconferencing, phone, smartphone, or tablet/computer. Ten studies reported on low distance, 11 on intermediate distance, and 5 studies for high distance assessments. Invalid performance data were observed with older age and cognitive impairment. Convergent validity data were reported consistently and suggested a decline with increasing distance: r = 0.52-0.80 for low, 0.49-0.75 for intermediate, and 0.41-0.53 for high distance. Diagnostic validity estimates presented a similar pattern. Reliability data were reported too inconsistently to allow evaluation. CONCLUSION: The validity of cognitive assessments with older adults appears supported at lower but not higher distance. Less is known about the reliability of such assessments. Future research should delineate the person and procedure boundaries for valid and reliable test results.
INTRODUCTION: Distance or remote cognitive assessments, administered via phone or computer platforms, have emerged as possible alternatives to traditional assessments performed during office visits. Distance refers to any nontraditional assessment feature, not only or necessarily location. We conducted a systematic review to examine the psychometric soundness of these approaches. METHOD: We searched PubMed, PsycINFO, AgeLine, and Academic Search Premier for articles published between January 2008 and June 2020. Studies were included if participants were over the age of 50, a structured assessment of cognitive function in older adults was evaluated, the assessment method was deemed distant, and validity and/or reliability data were reported. Assessment distance was defined as having any of the following features: use of an electronic test interface, nonroutine test location (e.g., home), test self-administered, and test unsupervised. Distance was categorized as low, intermediate, or high. RESULTS/DISCUSSION: Twenty-six studies met inclusion criteria. Sample sizes ranged from n = 8 to 8,627, and the mean age ranged from 57 to 83. Assessments included screens, brief or full batteries, and were performed via videoconferencing, phone, smartphone, or tablet/computer. Ten studies reported on low distance, 11 on intermediate distance, and 5 studies for high distance assessments. Invalid performance data were observed with older age and cognitive impairment. Convergent validity data were reported consistently and suggested a decline with increasing distance: r = 0.52-0.80 for low, 0.49-0.75 for intermediate, and 0.41-0.53 for high distance. Diagnostic validity estimates presented a similar pattern. Reliability data were reported too inconsistently to allow evaluation. CONCLUSION: The validity of cognitive assessments with older adults appears supported at lower but not higher distance. Less is known about the reliability of such assessments. Future research should delineate the person and procedure boundaries for valid and reliable test results.