Premium
Beyond uncertainty: Negative findings for the association between the use of proton pump inhibitors and risk of dementia
Author(s) -
Park SunKyeong,
Nam Jin Hyun,
Lee Hyesung,
Chung Hyunsoo,
Lee EuiKyung,
Shin JuYoung
Publication year - 2019
Publication title -
journal of gastroenterology and hepatology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.214
H-Index - 130
eISSN - 1440-1746
pISSN - 0815-9319
DOI - 10.1111/jgh.14745
Subject(s) - medicine , propensity score matching , dementia , confidence interval , rate ratio , cohort study , incidence (geometry) , cohort , retrospective cohort study , relative risk , proton pump inhibitor , disease , physics , optics
Abstract Objectives We investigated the risk of dementia associated with the use of proton pump inhibitors (PPIs) as compared with the use of histamine‐2 receptor antagonist (H2RA). Methods We conducted retrospective propensity score‐matched cohort study using the National Health Insurance Service‐National Sample Cohort. Subjects were defined as the patients newly prescribed with PPI or H2RA between 2003 and 2013 without prior prescriptions of PPI/H2RA or diagnosis of dementia from their history within the past 1 year. We followed up participants until dementia occurrence, death, or the end of the study, whichever occurred first, with an intention‐to‐treat approach. A 1‐year lag time between exposure and outcome measure was used to reduce protopathic bias. The incidence rate per 1000 person‐years was estimated. The incidence rate of PPI was compared with that of H2RA, defined as incidence rate ratio (IRR), calculated with a 95% confidence interval. To control for potential confounds, propensity score matching at a 1:1 ratio was conducted, and the crude IRR was adjusted by risk factors. Results Our propensity score‐matched cohort included 87 562 patients on PPIs and 87 562 patients on H2RAs. The IRR was 1.01 (95% confidence interval 0.96–1.06) with 1‐year lag time. IRR showed the decreased trend as the longer lag time. Also, no treatment duration or dose–response relationship was observed. Conclusions Our finding demonstrated that PPIs did not associate with dementia more strongly than did H2RA. On this basis, we suggest that the previously reported risk for dementia associated with PPI may have been overestimated.