Purpose: The total cost of Alzheimer's disease (AD) is estimated to range from $50-$100 billion annually. This study examined the changes in overall healthcare utilization following treatment initiation with five AD regimens: donepezil, galantamine, rivastigmine, memantine, and memantine plus acetylcholinesterase inhibitor (AChEI) combination while controlling for potential confounding factors. Methods: Retrospective analysis of medical, pharmacy and eligibility data for members initiating AD therapy (index date) between 1/1/04 and 6/30/04 from a large managed care organization (MCO) in the United States. Members who were not continuously enrolled throughout the 6-month pre-index period and 6-month post-index period were excluded. Analysis of covariance (ANCOVA) was used to determine the change (post-index period minus pre-index period) in overall pharmacy utilization, outpatient physician visits, emergency room (ER) visits, acute hospitalizations, acute hospitalization days, short-term skilled nursing facility (SNF) stays, and SNF days while controlling for age, gender, comorbidity (Chronic Disease Score) and baseline total healthcare charges. Results: A total of 2,461 were newly started on AD therapy; 65% were female and mean age was 80 years (SD 7.2). Treatment was initiated with donepezil (62%), memantine (14%), galantamine (11%), rivastigmine (8%), and memantine+AChEI (6%). While adjusted pharmacy utilization (PMPM) increased for each regimen (increases ranged from 0.27 to 0.51 across regimens, p=N.S.), medical utilization decreased. The largest decreases were observed in physician outpatient visits (ranging from –0.13 to –0.10 across regimens, p=N.S.), acute hospitalization days (–0.27 to –0.11, p=N.S.), and SNF days (–0.49 to –0.04, p=0.013). Conclusions: AD treatment initiation was associated with increases in overall pharmacy utilization and decreases in medical utilization, most notably in physician office visits, acute hospitalization LOS, and SNF LOS. Short-term SNF LOS was the only type of utilization that differed significantly between AD regimens.