P1‐169: Characteristics of the self‐administered gerocognitive examination (SAGE) used as a screening tool

sure the frequency of behaviors clinically and theoretically linked to three frontal-behavioral domains: Apathy/Akinesia, Disinhibition/Emotional Dysregulation, and Executive Dysfunction. Behavior during the past two weeks is rated on a 5-point Likert scale (11⁄4almost never; 51⁄4almost always) by an informant who has knowledge of the subject’s daily behavior. The scale provides a total score and subscores for the three behavioral domains. FrSBe total and subscale scores were converted to z-scores using published normative data that corrected for age and gender. Results: FrSBe z-scores were subjected to a series of regression analyses that used diagnostic group, MMSE scores (entered as a continuous variable), and the interaction between diagnosis and MMSE to predict FrSBe total and subseaIe scores. The regression model for the total FrSBe score was significant, F (3,59)1⁄4 4.04, p1⁄4.OII, and IvIMSE score, 61⁄4.605, p1⁄4.004, and the Group X MMSE interaction, 61⁄4-1.315, p1⁄4.017, were significant predictors. The nature of the Group X MMSE interaction is illustrated by comparing patients with mildly (MMSE > 20) or moderately(MMSE :5 20) impairedMMSE scores (based on a median split). Mildly demented DLB patients were not impaired on the FrSBe (mean z-score 1⁄4 0.206 1.1), but moderately demented DLB patients were quite impaired (z-score 1⁄4 -2.9361.1). In contrast, mildly and moderately demented AD patients were impaired to a similar degree (zscores1⁄4 -1.9461.8 and -2.1661.5, respectively). Similar regression analyses were performed for each of the three FrSBe subscale scores. The regression model for the Apathy subscale score was significant, F(3, 59)-3.55, p1⁄4.02, and MMSE score was a significant predictor, 61⁄4.616, p1⁄4.003. The GroupXMMSE interaction approached significance, 61⁄4-.985, p1⁄4.075. Moderately demented (z-score -3.9061.4) were more impaired than mildly demented DLB patients (zscore 1⁄4 -1.3062.0), whereas mildly and moderately demented AD patients were similarly impaired (z-scores 1⁄4 -2.5061.9 and -2.8662.2, respectively). The regression model for the Executive Dysfunction subscale was significant, F(3, 59)1⁄43.71, p1⁄4.017, and MMSE score, 61⁄4.589, p1⁄4.005, and the Group X MMSE interaction, B1⁄41.198, p1⁄4.03, were significant predictors. Mildly demented OLB patients were not impaired on the executive dysfunction subscale (z-score 1⁄4 -0.5461.3), but moderately demented DLB patients were very impaired (zscore1⁄4 -3.6061.5). In contrast, mildly and moderately demented AD patients were similarly impaired (z-scores 1⁄4 -2.4961.9 and -2.7161.5, respectively). The regression model for the Disinhibition subscale score was significant, F(3, 59)1⁄44.73, p1⁄4.005, and the Group X MMSE interaction was a significant predictor, 61⁄4-1.231, p1⁄4.023; however, neither group was impaired on this subscale (Mild DLB: z-score 1⁄4 1.136O.3, Moderate DLB z-score1⁄4 O.lO60.8; Mild AD zscore -0.4261.4, Moderate AD zscore-0.13 . 1.4). Conclusions: Contrary to our hypothesis, behavioral symptoms associated with frontal-subcortical dysfunction were associated with milder dementia in AD than in DLB. Although the groups’ total FrSBe scores were similar, patients with AD exhibited abnormal behavior early in the course of disease, but behavior did not worsen significantly as dementia progressed. In contrast, themildly demented DLB patients scored in the normal range and did not develop significant behavioral symptoms until well into the course ofthe disease. These results suggest that fronto-subcortical circuits associated with the behaviors assessed by the FrSBe are affected in early AD, possibly due to the accumulation of AD pathology in frontal lobe cortex. The lack ofsignificant behavioral symptomology early in the course ofOLB suggests that frontal lobe cortex and associated fronto-subcortical circuits underlying these behaviors are not affected until later stages of disease. This possibility is consistent with the early cognitive manifestations ofDLB that include disproportionately severe visuospatial processing and attentional deficits that may predominantly arise from posterior cortical dysfunction. As DLB progresses, behavioral symptoms become more common (and may be more common than in more advanced AD), and other features ofthe DLB syndrome (e.g., extrapyramidal motor features) become more obvious as the disease affects fronto-subcortical circuits Even when the circuits are affected, it appears that dysfunction may be limited to those associated with apathy and executive dysfunction (i.e., anterior cingulate and dorsolateral prefrontal, respectively). The lack ofabnonnal disinhibition/emotional dysregulation in OLB andAD patients is consistent with previous results and suggests that dysfunction oforbitofrontal circuits thought to mediate this behavior may occur only in more severe stages of these diseases, if at all. From a clinical perspective, the present results suggest that the presence ofbehavioral symptoms related to frontal systems dysfunction in a mildly demented patient may aid in differentiating between AD and DLB. However, this possibility becomes less likely as the severity of dementia increases.