Published in:
01-05-2016
Vitamin B12 deficiency induced by the use of gastric acid inhibitors: Calcium supplements as a potential effect modifier
Published in: The journal of nutrition, health & aging | Issue 5/2016
Login to get accessAbstract
Objective
Use of gastric acid inhibitors has emerged as a risk factor of vitamin B12 deficiency, especially in older adults. Calcium supplements could be an effect modifier of this relationship by its role in the absorption process of vitamin B12. The aim of this study is to examine whether the use of calcium supplements could be an effect modifier of the association between gastric acid inhibitors and vitamin B12 deficiency.
Design
Cross-sectional study based on medical chart reviews.
Setting
Geriatric Assessment Unit (GAU) of a university-affiliated hospital.
Participants
The study included 172 patients discharged from the GAU between 2008 and 2012.
Measurements
Cases of vitamin B12 deficiency were identified as those who had received a diagnosis of vitamin B12 deficiency, and/or were receiving a treatment for vitamin B12 deficiency. Use of gastric acid inhibitors and calcium supplements at admission was determined from the pharmacist report. Associations between medications and vitamin B12 status were investigated using logistic regression models.
Results
Seventy-one patients (41%) had vitamin B12 deficiency. At admission, 42% were taking gastric acid inhibitors and 45% calcium supplements. After adjustment for covariates, analyses revealed that vitamin B12 deficiency was more likely among users of gastric acid inhibitors who did not concomitantly received calcium supplements [OR=3.12; P=0.01]. Conversely, no significant association was observed in patients using both, gastric acid inhibitors and calcium supplements [OR=1.30; P=0.59].
Conclusions
The present study provides the very first evidence that the use of calcium supplements could be an effect modifier of the association between gastric acid inhibitors and vitamin B12 deficiency. Failure to consider calcium supplements as an effect modifier could have led to biased risk estimates in previous published studies.