Abstract
OBJECTIVE: Dairy intake has been reported to be both pro-inflammatory and anti-inflammatory; this inconsistency has led to uncertainty in the field. We aimed to address this using genetic data to perform a causal analysis of the link between lactose intolerance, rheumatoid arthritis (RhA) and body mass index (BMI). Lactose intolerance necessarily leads to reduced dairy intake. RhA is an autoimmune disease, which has been linked to both increased and reduced dairy intake. Dairy intake has also been associated with elevated BMI, which is itself pro-inflammatory and is associated with increased risk of RhA. METHODS: We separately considered congenital lactose intolerance, and adult-onset lactose intolerance where a prolonged period of dairy intake has already occurred. We combined Mendelian randomisation (MR) and rare genetic variant association testing to determine the relationship between genetic liability to lactose intolerance, risk of RhA and BMI. As a positive control to optimise instruments for measurement of lactose intolerance, we used the causal link between lactose intolerance and osteoporosis. Rare variant analysis was performed under a recessive model. Conditional analysis of the effect of lactose intolerance on the risk of RhA via BMI used multivariable MR. RESULTS: We observed an opposite effect of congenital and adult-onset lactose intolerance on RhA risk. Congenital lactose intolerance increases the risk of RhA, but adult-onset lactose intolerance reduces the risk of RhA. The protective effect of adult-onset lactose intolerance on RhA is conditional on reduced BMI. DISCUSSION: We conclude that dairy intake is anti-inflammatory, which explains why congenital lactose intolerance and lifetime avoidance of dairy foods increases the risk of RhA. However, adult-onset lactose intolerance can reduce the risk of RhA because of an association with reduced BMI, which is also anti-inflammatory.