Characterising omission errors in everyday task completion and cognitive correlates in individuals with mild cognitive impairment and dementia.

The sample was highly educated, so the results may not apply to more diverse older adults. Tasks were completed in a campus apartment, not in participants' own homes, so everyday performance may differ from real home life. Some neuropsychological tests used to create cognitive composites were also used to assign diagnostic groups, which can blur the link between diagnosis and measured cognition. The cognitive composites depended on the tests that were administered, so other cognitive abilities were not examined. For the regression analyses, diagnostic groups were combined to increase statistical power, so the associations were estimated across the whole sample rather than within each group. One participant was excluded from the regression analyses because a processing speed measure was missing, reducing the regression sample to 129.

Do not read the memory association as proof that memory training will reduce everyday task omissions. The study observed existing diagnostic groups and did not test an intervention. It also showed that working memory and processing speed were associated with some omission categories, while executive functioning did not emerge as a significant predictor.