Incidence of type 2 diabetes among patients exposed to the combination of pravastatin and paroxetine.

PURPOSE: This study evaluated the effects of concomitant pravastatin and paroxetine use on the incidence of Type 2 Diabetes Mellitus (T2DM). METHODS: A new-user retrospective cohort design was employed using data selected from US health insurance claims databases (OptumInsight and MarketScan) between July 1, 2002, and December 31, 2009. Patients included were of age ≥18; newly prescribed pravastatin or paroxetine; and enrolled in the database for ≥180 days prior to the index date (i.e., first prescription of incident drug). Patients were assigned to either incident pravastatin or incident paroxetine user groups. Patients were followed until the study endpoint (T2DM), discontinuation of incident drug, second drug, or end of study/patient data. Cox proportional hazards models compared T2DM in users of pravastatin who were also taking paroxetine at index the date (combination users) versus pravastatinonly users. A similar analysis among users of paroxetine evaluated the use or non-use of pravastatin at index date. RESULTS: OptumInsight yielded 288,678 incident users of pravastatin or paroxetine; 443,137 were identified in MarketScan. The risk of T2DM among combination users compared to incident pravastatin only users was 1.05 (95% CI: 0.76, 1.44) and 0.94 (95% CI: 0.90, 0.97) in OptumInsight and MarketScan, respectively. The risk of T2DM among combination users compared to incident paroxetine only users was 1.03 (95% CI: 0.69, 1.54) in OptumInsight and 1.02 (95% CI: 0.97, 1.07) in MarketScan. CONCLUSION: The results indicate no increase in the risk of T2DM due to combined use of pravastatin and paroxetine compared to individual use of the two drugs; however, this study is limited by short mean follow-up.

Predictors of placebo group decline in the Alzheimer's disease Assessment Scale-cognitive subscale (ADAS-Cog) in 24 week clinical trials of Alzheimer's disease.

One limitation of several recent 24 week Alzheimer's disease (AD) clinical trials was the lack of cognitive decline detected by the AD Assessment Scale-cognitive subscale (ADAS-cog) in the placebo groups, possibly obscuring true medication effects. Data from 733 individuals in the placebo arms of six AD clinical trials performed 1996-1997 were pooled to examine the relationship of clinical, demographic, and genetic characteristics with the 24 week change in ADAS-cog. Baseline cognitive and functional status and the screening-to-baseline change in ADAS-cog were the strongest independent predictors of the 24 week change in ADAS-cog. The ADAS-cog did not detect progression in patients with mild dementia (screening Mini-Mental State Exam, MMSE, >or=20). The change in ADAS-cog from screening to baseline was inversely correlated with the 24 week change score; it was more difficult to detect cognitive decline at 24 weeks if individuals markedly worsened from screening to baseline. The effects of baseline MMSE and screening-to-baseline change in ADAS-cog generalized to the placebo group (N=106) of another AD study performed in 2004-2005. Overcoming lack of placebo decline in AD clinical trials will require scales more sensitive to cognitive decline in mild AD and strategies to reduce within-person variability in outcome measures.