OBJECTIVES: Alzheimer’s disease (AD), a chronic and progressive neurodegenerative disease, accounts for approximately 60-80% of all dementia cases, and profoundly impacts patient and caregiver lives. This study characterizes the real-world clinical and humanistic burden of AD.

METHODS: Natural language processing was used to search and analyze the Amplity Insights database, composed of US-based transcribed physician notes, for patients diagnosed with AD (between January 2003 and December 2023). Patient-level characteristics, comorbidities, symptoms, quality of life (QoL), testing, and treatment use were summarized and described.

RESULTS: 144,687 patients were identified, with a mean (standard deviation) age of 81.1 (10.0) years. 61% of patients were female and 87% were white. 67,030 (46%) patients were seeing primary care physicians; 103,122 (83%), 65,001 (52%), and 35,187 (28%) had cardiovascular, metabolic, and psychiatric comorbidities. Among patients who reported symptoms, depression (37%), anxiety (26%), aggression (21%), cognitive impairment (21%), and confusion (20%) were the most frequently reported. Anxiety and depression were consistently reported in the 3 years leading to diagnosis whereas aggression, cognitive impairment, and confusion were reported in the year prior to diagnosis. Treatment mentions included donepezil (22%), memantine (17%), rivastigmine (4%), galantamine (1%), and lecanemab (n=2). 9% of patients had undergone testing with magnetic resonance imaging, 0.4% with lumbar puncture, and 0.9% with positron emission tomography scans. QoL impacts on patients who reported QoL issues included: falls (71%), inability to walk/sit (17%), and inability to cook (9%); 23% of these patients mentioned having “caregivers” and of these, 9% mentioned caregiver burden.

CONCLUSIONS: Analysis of this real-world database revealed that patients with AD are significantly impacted by their disease in terms of comorbidities, symptoms, and QoL burden, with QoL burdens that extend to their caregivers. Furthermore, these data suggest that AD diagnoses and treatment decisions may remain focused on clinical symptoms versus a biomarker-guided approach.