Cholesterol-induced production of amyloid beta (Abeta) as a putative neurotoxin in Alzheimer's disease (AD), along with epidemiological evidence, suggests that statin drugs may provide benefit in treatment of the disorder. We tested the effect of once daily atorvastatin calcium (80 mg; two 40 mg tablets) on cognitive and/or behavioral decline in patients with mild-to-moderate AD. The study was designed as a pilot intention-to-treat, proof-of-concept, double-blind, placebo-controlled, randomized (1:1) trial with a 1-year exposure to study medication employing last-observation-carried-forward (LOCF) ANCOVA as the primary statistical method of assessment. Alternate statistical methods were employed to further explore the effect of atorvastatin treatment on progression of deterioration. Of the 98 individuals with mild-to-moderate AD (Mini-Mental State Examination score of 12-28) providing Informed Consent, 71 were eligible for randomization, 67 were randomized and 63 completed the 3-month visit and were statistically evaluable. The primary outcome measures were change in the Alzheimer Disease Assessment Scale-Cognitive (ADAS-cog) performance and the Clinical Global Impression of Change (CGIC). Secondary outcome measures included the MMSE, Geriatric Depression Scale (GDS), the Neuropsychiatric Inventory (NPI) and the ADCS Activities of Daily Living inventory (ADCS-ADL). Tertiary outcome measures included levels of total circulating cholesterol, LDL and VLDL, and circulating activity of the free radical scavenger enzymes superoxide dismutase (SOD) and glutathione peroxidase (GpX). Atorvastatin reduced circulating cholesterol levels and produced a positive signal on each of the clinical outcome measures compared to placebo, but did not elicit a difference in circulating SOD or GpX activities. The observed beneficial clinical effect reached significance for the GDS (p = 0.040) and the ADAS-cog at 6 months (p = 0.003), was all but significant for the ADAS-cog (p = 0.055) at 12 months, and was of marginal significance for the CGIC (p = 0.073) and NPI (p = 0.071) at 12 months when employing the primary statistical approach (ANCOVA with LOCF). Application of repeated measures ANCOVA statistics revealed the difference was significant for the CGIC and marginally significant for the ADAS-cog, but not significant for the other clinical indices. This evaluation indicated significant time-by-treatment interactions (altered progression) for the ADAS-cog and MMSE, but not the CGIC. Application of random intercept regression analysis revealed a significant difference for the CGIC, ADAS-cog and MMSE. Regression analysis also indicated that atorvastatin produced change in the slope of deterioration on the MMSE. Accordingly, atorvastatin therapy may be an effective treatment and may slow the progression of AD among mild-to-moderately affected patients.