Frequent smartphone- and tablet-based remote memory assessments may help detect subtle cognitive decline associated with prodromal Alzheimer’s disease over short periods, particularly among those with evidence of underlying Alzheimer’s pathology, while conventional in-person cognitive testing may not detect differences in short-term decline.
Researchers analyzed data from 202 patients aged 52 to 85 years participating in longitudinal aging studies in Germany and the United States. The cohort included 152 cognitively unimpaired patients and 50 patients with mild cognitive impairment. The participants completed memory tests remotely on their own smartphones or tablets weekly, biweekly, or in periodic testing bursts for at least 30 weeks and up to 1 year. The researchers examined changes in performance on tests of memory precision regarding objects and scenes, associative memory, and familiarity-dependent memory.
The researchers found that patients with mild cognitive impairment experienced greater decline compared with cognitively unimpaired patients on a remote test of familiarity-dependent memory during the 30-week study period. When the mild cognitive impairment group was stratified by beta-amyloid status, the decline was most pronounced among patients who were beta-amyloid–positive, a biomarker profile consistent with prodromal Alzheimer’s disease. In this subgroup, the researchers observed greater decline in both familiarity-dependent memory and memory precision for objects.
Not all remote memory measures showed evidence of short-term decline. The researchers did not identify statistically significant differences in rates of change for associative memory or memory precision for scenes. They suggested that these cognitive processes may deteriorate earlier in the disease course, resulting in measurable impairment at baseline but less detectable decline during the mild cognitive impairment stage studied here.
The researchers also examined whether changes on the remote tests reflected changes captured by established neuropsychological assessments. Decline on the familiarity-dependent memory task correlated with cognitive decline measured over multiple years using a harmonized cognitive composite derived from in-person testing. According to the researchers, the findings suggested that the remote assessment may be measuring a cognitive construct relevant to Alzheimer’s disease–related progression.
In contrast, conventional neuropsychological testing was not sensitive to differences in short-term decline. In a subgroup of patients with two in-person assessments conducted within approximately 1 year, the researchers observed baseline cognitive differences between groups but did not detect differences in rates of cognitive change over that shorter interval.
The researchers also demonstrated the feasibility of high-frequency remote assessment in older adult patients. Participants completed most assigned testing sessions, with average adherence of 77% and compliance of 89%. Neither cognitive status nor beta-amyloid–positivity was associated with a greater likelihood of dropping out prior to 30 weeks.
Several limitations should be considered when interpreting the findings. The analysis combined participants from multiple cohorts that used somewhat different protocols and assessment schedules. The subgroup of beta-amyloid–positive patients with mild cognitive impairment was relatively small, limiting the precision of subgroup analyses. In addition, the researchers did not include tau imaging data, preventing direct assessment of how tau pathology may have influenced cognitive trajectories. They also noted that longer follow-up periods may reveal changes in memory domains that were not detectable within 30 weeks.
“Frequent remote cognitive testing is a promising tool to feasibly capture and monitor subtle and short-term cognitive decline,” wrote lead study author Sarah E. Polk, of the German Center for Neurodegenerative Diseases in Germany, and colleagues.
The researchers concluded that smartphone- and tablet-based cognitive assessments may offer a practical approach for monitoring cognitive change in patients with mild cognitive impairment and prodromal Alzheimer’s disease, although larger studies will be needed to confirm the findings and establish clinical benchmarks.
Full disclosures of the study authors can be found in the study.
Source: npj Digital Medicine