Depressed older adults with a history of suicidal behavior are at high-risk for progression to dementia. Whereas personality factors (i.e., neuroticism, conscientiousness) have been identified as key predictors of subjective cognitive complaints (SCC) in community-based older adults, it remains unknown how SCC relates to objective cognitive performance and personality factors in older suicide attempters. Using a case-control study design, we examined the relationship of SCC with objective cognitive performance and personality factors in depressed older adults with a history of suicidal behavior, depressed non-attempters, and older adults with no psychiatric history.

The sample (N=169) was aged 64.83 years (SD=9.21) and comprised 28 non-psychiatric comparison groups (CON), 89 depressed non-attempters (DNA), and 52 suicide attempters (ATT). Ordinary least squares linear regression models, including main effects and interaction terms with group were used to examine the extent to which the relationship between SCC (Memory Functioning Questionnaire; MFQ), objective memory performance (Rey Auditory Verbal Learning Test; AVLT), and personality factors (neuroticism and conscientiousness; NEO Five-Factor Inventory) differ based on suicide risk group status. This was followed by secondary regression analyses stratified by suicide risk group. All regression models were adjusted for age, sex, education, and race. Regression models in depressed groups additionally adjusted for depression severity.

Suicide attempters performed worse than the non-psychiatric comparison group on verbal learning (AVLT Immediate Recall CON: Beta=0.27 SE=2.11, p=.003) and performed worse than the depressed non-attempters on delayed memory recall (AVLT Delayed Recall DNA: Beta=0.19 SE=0.58, p=.038). Suicide attempters reported worse memory functioning than non-psychiatric comparisons but not depressed non-attempters (MFQ General CON: Beta=0.36 SE=7.41, p<.001; DNA: Beta=0.05 SE=6.04, p=.668). Conscientiousness partially explained suicide group differences in self-reported memory functioning (CON>ATT; CON Beta= 0.34 SE=8.86, p<.001, CON Beta after including conscientiousness in model=0.17 SE=10.04, p=.090). However, attempters reported using fewer compensatory strategies for memory difficulties than depressed non-attempters (MFQ Compensatory Strategies DNA Beta=-0.211 SE=1.85, p=.013).  The relationship between objective memory performance and SCC did not differ by suicide risk groups. Personality was independently associated with SCC such that higher conscientiousness was associated with better self-reported memory functioning (conscientiousness Beta=0.29 SE=0.33, p<.001) and higher neuroticism with self-reported serious functional difficulties attributable to memory failures (neuroticism Beta=-0.29 SE=0.24, p=.005). Group-stratified analyses indicated that neuroticism was only associated with serious functional difficulties attributable to memory failures in the depressed groups (MFQ serious-neuroticism Depressed Beta=-0.30 SE=0.24, p<0.001) but not in the non-psychiatric comparison group (MFQ serious-neuroticism CON Beta=0.12 SE=0.96, p=0.58). Conscientiousness was associated with better reported memory functioning in suicide attempters but not in depressed non-attempters or non-psychiatric comparisons (conscientiousness x DNA interaction Beta=-2.44 SE=1.82, p=.01; Group Stratified Analyses ATT Beta=0.50 SE=0.64, p<.001; DNA Beta=0.11 SE=0.50, p=0.32; CON Beta=0.34 SE=0.92, p=.14).

Suicidal depressed older adults reported using fewer compensatory strategies to cope with their memory difficulties despite having objectively worse memory performance relative to depressed non-attempters. Conscientiousness in suicidal depressed older adults may increase their sense of cognitive agency, whereas neuroticism may exacerbate their perception of cognitive decline.