The complexity of producing and interpreting dietary vitamin A statistics.

Producing, reporting, and interpreting vitamin A statistics present multiple challenges largely attributable to the systems of equivalence used to convert pro-vitamin A carotenoids into retinol equivalents, and to the criteria used by institutions to set recommendations. This study describes the information on total vitamin A, retinol and provitamin A carotenoids available in 90 food composition tables/databases (FCTs/FCDBs). It also evaluates the effect of the definition of vitamin A intake (Retinol Equivalents [RE] or Retinol Activity Equivalents [RAE]) and the source of requirements on the potential contribution of dietary intake to the population's requirements. We found that 43 percent of the FCTs/FCDBs reviewed, many of them from high-income countries, do not provide total vitamin A or sufficient information for computing it, or present inconsistencies between the metadata and the published values; 9 percent publish total vitamin A in RE and RAE; and 28 percent provide information on retinol and provitamin A carotenoids that enables calculating total vitamin A in both definitions. Vitamin A adequacy ratios are lowest when the consumption unit is RAE and the source of requirements is the US Health and Medicine Division. When the consumption definition is RE, adequacy ratios are higher using FAO/WHO than EFSA requirements. It is imperative to reach consensus on the system of conversion of provitamin A carotenoids into retinol equivalents.