28 December 2004 in Neurology
Objective: To evaluate the efficacy of a cognitive-motor program in patients with early Alzheimer disease (AD) who are treated with a cholinesterase inhibitor (ChEI). Methods: Patients with mild cognitive impairment (MCI) (12), mild AD (48), and moderate AD (24) (Global Deterioration Scale stages 3, 4, and 5) were randomized to receive psychosocial support plus cognitive-motor intervention (experimental group) or psychosocial support alone (control group). Cognitive-motor intervention (CMI) consisted of a 1-year structured program of 103 sessions of cognitive exercises, plus social and psychomotor activities. The primary efficacy measure was the cognitive subscale of the AD Assessment Scale (ADAS-cog). Secondary efficacy measures were the Mini-Mental State Examination, the Functional Activities Questionnaire, and the Geriatric Depression Scale. Evaluations were conducted at 1, 3, 6, and 12 months by blinded evaluators. Results: Patients in the CMI group maintained cognitive status at month 6, whereas patients in the control group had significantly declined at that time. Cognitive response was higher in the patients with fewer years of formal education. In addition, more patients in the experimental group maintained or improved their affective status at month 12 (experimental group, 75%; control group, 47%; p = 0.017). Conclusions: A long-term CMI in ChEI-treated early Alzheimer disease patients produced additional mood and cognitive benefits.
8 September 1999 in International Journal of Geriatric Psychiatry
CLConstantine G. Lyketsos
1 July 2011 in Archives of Gerontology and Geriatrics
20 July 2026 in SSRN Electronic Journal
ObjectiveTo evaluate the association between documented antipsychotic exposure and documented diagnoses of deep vein thrombosis (DVT) and pulmonary embolism (PE) using information extracted from unstructured EHRs through an AI-enabled clinical NLP platform, and to explore whether these associations varied according to antipsychotic agent, patient age, or psychiatric diagnosis.MethodsThis retrospective, non-interventional study included adults aged 18 years or older who received care at Hospital Universitario Puerta de Hierro Majadahonda (Madrid, Spain) between January 2017 and July 2023. Clinical information was extracted from unstructured EHR narratives using the SAVANA® EHRead® clinical NLP technology. Logistic regression models were used to estimate odds ratios (ORs) and 95% confidence intervals (CIs) for the association between antipsychotic exposure and documented DVT or PE. Model performance was assessed using Tjur's R².ResultsA total of 37,423 patients were included (mean age 62.9 years; 58% women). Among them, 1,489 (4.0%) had a documented diagnosis of DVT and 4,152 (11.1%) had a documented diagnosis of PE. Any documented antipsychotic exposure was associated with lower odds of documented DVT (OR 0.02, 95% CI 0.02–0.03, p < 0.001; Tjur's R² = 0.028) and PE (OR 0.01, p < 0.001; Tjur's R² = 0.028). Similar associations were observed for most individual antipsychotic agents. Clozapine and levomepromazine were not significantly associated with DVT. Across all models, explanatory power was low.ConclusionsIn this retrospective NLP-based analysis of unstructured EHR data, documented antipsychotic exposure was not associated with increased odds of documented DVT or PE. The observed inverse associations are unlikely to reflect a protective biological effect and should be interpreted in the context of residual confounding, temporal ambiguity, sparse-data bias, and limitations inherent to retrospective EHR-based studies. These findings illustrate both the opportunities and challenges of using AI-enabled clinical NLP for pharmacoepidemiological research based on real-world healthcare data.