Improving dementia prognostication in cognitively normal older adults: conventional versus novel approaches to modelling risk associated with neuropsychiatric symptoms

Background Studies in cognitively normal individuals on associations between psychiatric symptomatology and incident dementia have not reliably differentiated psychiatric syndromes from neuropsychiatric symptoms (NPS) that represent neurodegeneration. Conventional modelling often overlooks symptom natural history. Mild behavioural impairment (MBI) is a syndrome that leverages later-life emergent and persistent NPS to identify a high-risk group for incident dementia. Aim We aimed to explore associations of MBI, and conventionally-measured NPS (NPS-not-MBI), with incident dementia in cognitively normal individuals and the cognitively normal subset with subjective cognitive decline (SCD). Method Using National Alzheimer's Coordinating Center data, MBI was operationalised by the absence of past psychiatric disorders (symptom emergence) and the presence of symptoms at >2/3 of pre-dementia visits (symptom persistence). Kaplan-Meier survival curves and Cox proportional hazards regressions modelled dementia incidence across NPS groups and MBI domains, adjusted for age, gender, education, race, APOE-epsilon 4, and cognitive status. Results The sample comprised 1408 MBI (age 75.2 +/- 9.5; 54.3% female), 5625 NPS-not-MBI (age 71.6 +/- 8.8; 65.5% female) and 5078 No-NPS (age 71.2 +/- 8.9; 67.6% female) participants. Compared with No-NPS, MBI participants had lower dementia-free survival (P < 0.0001) and 2.76-fold greater adjusted dementia incidence rate (95% CI: 2.27-3.35, P < 0.001); incidence rate in NPS-not-MBI did not differ from No-NPS (hazard ratio 0.97, 95% CI: 0.82-1.14, P = 0.687). Of those with MBI who progressed to dementia, 76.0% developed Alzheimer's disease. Similarly, in the SCD subsample (n = 3485), persons with MBI had 1.99-fold greater dementia incidence versus No-NPS (95% CI: 1.46-2.71, P < 0.001) while NPS-not-MBI did not differ from No-NPS (hazard ratio 0.92, 95% CI: 0.70-1.19, P = 0.511). Conclusions Incorporating natural history into assessment of psychiatric symptoms in accordance with MBI criteria enhances dementia prognostication and modelling.