P4-146: Talking about the past and the future: A characterization of responses to questions about personal past and future plans in treated and untreated people with Alzheimer's disease

Everyday correlates of Alzheimer's disease (AD) and its change are of interest. Our group has focused on identifying clinically observable changes in cognition and/or function which may be relevant for assessing treatment efficacy or have potential as 'early warning' signs of incipient AD. To discover whether responses to questions about personal past and future are sensitive to endogenous changes and/or mark treatment effects, we asked about personal past and future plans in mild/moderate AD patients. A subsample of patients (70/130) enrolled in a 4-month, double-blind placebo-controlled trial of galantamine (GAL) (CMAJ 2006;174:1099–1105; doi:10.1503/cmaj.051432) were video-recorded responding to a modified version of the Examination of Memory and Temporality (EMT) (Cortex. 1993;29:567–581). Responses were coded for content (e.g. informative, habitual, equivocal, confabulated, don't know), weighted and summed to yield scores (EMTL) for each patient (0–100). Mean scores and change scores were used to examine associations with the trial's other outcome measures. Forty patients (galantamine (GAL)=16, placebo (PLA)=24) had EMT interview data recorded at baseline and 4-months. Baseline: Patients in each arm were demographically and functionally similar. EMTL scores were similar in both groups for past (GAL mean=54.4 ±11.5, PLA mean=53.8±22.0) and future questions (GAL mean=50.3 ±17.9, PLA mean=48.6 ±20.1). Four months: The treatment arm significantly improved (Past GAL=7.4 ±16.1 PLA=-6.8 ±20.9, p=0.03; Future: GAL=18.4 ±15.1, PLA= -0.6 ±18.7, p=0.02). Even so, group changes on EMTL showed limited correlation with other outcomes. EMTL-Future correlated negatively with Orientation (r=-.561; p= .01) but not with ADAS-cog (p=.06). EMTL-Past correlated with worsening DAD (p=.02) scores, and showed trends on worsening ADLs (p=.08), IADLs (p=.09), Performance (p=.09), and CIBIC (p=.08) in the placebo group. Inspection of group mean changes showed non-significant improvement on some measures for GAL patients (Planning = 7.3±9.5; IADL 3.4±11.4; ADAS -3.9±5.9; Episodic Memory -2.8±3.9). Significant differences in group means appear only for Planning (p=.04). The trends - modest (non-significant) improvements in the GAL group on Planning, IADLs, ADAS-cog and episodic memory—suggest positive changes in EMTL-Future and EMTL-Past are not epiphenomena, but are capturing a real effect. This possibility needs to be explored in a larger sample.