LAUSR

The onset of dementia is a gradual and insidious process, and many individuals who receive a diagnosis of dementia report experiencing changes in their memory in the years prior (Glodzik-Sobanska et al., 2007; Jessen et al., 2010). Evidence suggests that reports of such changes are linked to an increased likelihood of developing dementia, and specifically Alzheimer’s disease (AD) (Reid and MacLullich, 2006), but evidence for an association between subjective reports and demonstrable changes using objective testing is lacking (Fritsch et al., 2014). A key factor that may explain the association between subjective reports of memory changes and later development of dementia is depression, but this relationship is complicated (Hill et al., 2016). Depression can manifest as cognitive difficulties, and this is captured in measures such as the Geriatric Depression Scale (GDS) (Sheikh and Yesavage, 1986). On the other hand, associations may be driven by the impact of perceived memory problems on mood, whereby the subjective experience of memory problems can lead to worry and a sense of hopelessness about the future in line with the Hopelessness Theory of Depression (Abramson et al., 1989), and supported by prior research (Crane et al., 2007). Much of the existing research investigating this idea is crosssectional, and consequently it is difficult to understand the temporal relationship between the two variables. Longitudinal studies are needed to disentangle this relationship, with previous studies being few in number and conflicting in their findings. A further limitation to exploring the lack of relationship between subjective and objective memory changes is the variety of measures and items used to elicit data regarding subjective memory complaints (Rabin et al., 2015). Response modes for such questions can ask participants to rate their current memory, compare it to their peers, or make a comparison with their abilities in the past. This makes results of previous studies difficult to compare, as the concept is not uniformly measured nor clearly operationalized. It is likely that different ways of obtaining subjective reports of memory have an impact on the lack of association with objective memory performance, and the involvement of other explanatory variables such as depression (Hill et al., 2016; Opdebeeck et al., 2019). In this issue of International Psychogeriatrics, Hill et al. (2019) investigated longitudinal relationships between self-reported memory problems and depressive symptoms, seeking to examine whether self-reported memory problems occurred concurrently with, preceded or followed depressive symptoms over time, using three different measures of subjective memory complaints. Based on previous cross-sectional work (Chin et al., 2014), the authors hypothesized that at any given year in the study, reports of memory problems will be associated with increased depressive symptoms, and secondly, that an increase in memory problems at any given year will predict an increase in depressive symptoms reported the following year. The analyses conducted in this study included 1162 cognitively intact community dwellers aged over 70 years, who were participating in the Einstein Aging Study, completing up to 11 waves of data. Self-reported memory was measured using three items that considered the frequency of memory problems in the last year, perceived 1-year decline in memory, and perceived 10-year decline in memory. Depressive symptoms were measured using the 15-item GDS (Sheikh and Yesavage, 1986), but with the item “Do you feel you have more memory problems than most?” excluded from scoring, leaving 14 items. To analyze whether changes in selfreports of memory problems are related to changes in current and future depressive symptoms, separate multilevel linear models were created for each of the self-reported memory questions. Concurrent changes in self-reported memory problems and depressive symptoms were examined, and then changes in self-reported memory from the previous wave of data collection were added to the model as a lagged variable. This enabled the researchers to examine whether changes in self-reports of memory problems predicted future changes in depressive International Psychogeriatrics (2020), 32:6, 677–680 © International Psychogeriatric Association 2020