A Pharmacoepidemiological Claims Data Analysis

Willy Gomm, PhD; Klaus von Holt, MD, PhD; Friederike Thomé, MSc; et al.

JAMA Neurol. 2016 April; 73(4):410-6

Abstract available at: https://www.ncbi.nlm.nih.gov/pubmed/26882076

This association was encountered by an epidemiological study. Findings of mouse models in which the use of PPIs increased the levels of β-amyloid in the brain are suggestive of a causal relation. Elevated prevalence of these amyloid plaques are reported in patients with Alzheimer’s dementia.

The researchers investigated the association between the use of PPIs (omeprazole, pantoprazole, lansoprazole, esomeprazole, or rabeprazole) and risk for dementia using insurance claims data (inpatient and outpatient diagnoses and drug prescriptions; 2004 to 2011) from 73,679 German patients aged ≥75 years. They used time-dependent Cox regression modeling, incorporating potential confounding factors of age, sex, comorbidities (depression, stroke, diabetes, ischemic cardiac disease), and polypharmacy.

Overall, patients receiving regular PPIs were significantly more likely to develop incident dementia compared with those not taking PPIs (hazard ratio, 1.44).

Comment: These findings need to be taken with a grain of salt, as the study did not adjust for well-recognized risk factors for dementia, particularly family history, heavy alcohol use, hypertension, and atherosclerosis. The proposed mechanism of harm for PPIs is elevation of β-amyloid, which has been independently associated with Alzheimer dementia, yet only 2.5% of patients had this diagnosis. I think it is premature to advise patients to stop PPI if they need it due to this reason, and we should reassure our patients that adequate interpretation of the study is necessary.