Assessment of the Dose–Response Relationship Between Folate Exposure and Cognitive Impairment: Synthesizing Data from Documented Studies

The dose–response relationship between folate levels and cognitive impairment among individuals with vitamin B12 deficiency is an essential component of a risk‐benefit analysis approach to regulatory and policy recommendations regarding folic acid fortification. Epidemiological studies provide data that are potentially useful for addressing this research question, but the lack of analysis and reporting of data in a manner suitable for dose–response purposes hinders the application of the traditional evidence synthesis process. This study aimed to estimate a quantitative dose–response relationship between folate exposure and the risk of cognitive impairment among older adults with vitamin B12 deficiency using “probabilistic meta‐analysis,” a novel approach for synthesizing data from observational studies. Second‐order multistage regression was identified as the best‐fit model for the association between the probability of cognitive impairment and serum folate levels based on data generated by randomly sampling probabilistic distributions with parameters estimated based on summarized information reported in relevant publications. The findings indicate a “J‐shape” effect of serum folate levels on the occurrence of cognitive impairment. In particular, an excessive level of folate exposure is predicted to be associated with a higher risk of cognitive impairment, albeit with greater uncertainty than the association between low folate exposure and cognitive impairment. This study directly contributes to the development of a practical solution to synthesize observational evidence for dose–response assessment purposes, which will help strengthen future nutritional risk assessments for the purpose of informing decisions on nutrient fortification in food.