3. Dietary assessment of flavonoids

The most common method of estimating flavonoid intake in epidemiologic studies is to use dietary questionnaires, such as food frequency questionnaires (FFQ), 24-h dietary recalls, and food diaries, to record all food consumption over a known period of time dietary surveys, especially FFQ, are prone to several limitations, particularly regarding the identification of specific foods (e.g., the large variety of products available on the market, different food processing and cooking techniques, distinct ingredients, and preparation of recipes) and accurate quantification of portion sizes [11]. Despite this, FFQs are the most common method used in large epidemiological studies, mainly because they are an easy, quick, and an economical way to record the participants' habitual diet. In order to improve both quantification and quality of the data gathered, the use of a previously validated FFQ on flavonoids or flavonoid-rich foods is essential. Moreover, the use of innovative technologies and methodologies for the dietary assessment of flavonoid intake such as the collection of multiple 24-HDRs and food records and interactive computer- and camera-based technologies will certainly improve this process [12].

In order to estimate the intake of flavonoids, a food composition table or database is also needed. There are currently two main databases. The first one is the US Department of Agriculture (USDA) database on flavonoids, isoflavones, and proanthocyanidins, which was created in 2003, 1999, and 2004, respectively [13–15]. Since then, several updates have been released. The USDA databases contain worldwide food composition data on the six main flavonoid classes (flavanols, flavonols, flavanones, flavones, isoflavones, and anthocyanidins) expressed as aglycones. Phenol-Explorer is the second most common database [16]. It was developed in 2009 and contains worldwide data, but in this case, on all nine flavonoid classes (including chalcones, dihydrochalcones, and dihydroflavonols) expressed as they are found in nature (mainly as glycosides) and analyzed with chromatography without hydrolysis. Phenol-Explorer also contains data on chromatography after hydrolysis and, in this case, flavonoids are expressed as aglycones, but these are not usually used. In the studies using Phenol-Explorer using flavonoid data expressed as aglycone equivalents, flavonoid glycoside contents are converted into aglycone contents using their respective molecular weights [17, 18]. The other important difference between databases is that Phenol-Explorer does not contain data on thearubigins because the composition data quality on thearubigins is very low [19]. The nonspecific spectrophotometric method used to date only provides a crude estimation of their concentrations in black teas, which are the only known food sources of thearubigins.

These differences between databases (classes of flavonoids, aglycones vs. glycosides, and inclusion of thearubigins) complicate the comparison among studies using different databases. In this book chapter, we have mostly compared studies using the USDA databases or Phenol-Explorer with data expressed as aglycones. In our comparisons, we have also made distinctions between the studies including proanthocyanidins and thearubigins, which are the most controversial flavonoid compounds.

sugar can be L-rhamnose, D-glucose, glucorhamnose, galactose, or arabinose [9]. This is very relevant because bioavailability differs among flavonoids, depending on the aglycone, the type

The most common method of estimating flavonoid intake in epidemiologic studies is to use dietary questionnaires, such as food frequency questionnaires (FFQ), 24-h dietary recalls, and food diaries, to record all food consumption over a known period of time dietary surveys, especially FFQ, are prone to several limitations, particularly regarding the identification of specific foods (e.g., the large variety of products available on the market, different food processing and cooking techniques, distinct ingredients, and preparation of recipes) and accurate quantification of portion sizes [11]. Despite this, FFQs are the most common method used in large epidemiological studies, mainly because they are an easy, quick, and an economical way to record the participants' habitual diet. In order to improve both quantification and quality of the data gathered, the use of a previously validated FFQ on flavonoids or flavonoid-rich foods is essential. Moreover, the use of innovative technologies and methodologies for the dietary assessment of flavonoid intake such as the collection of multiple 24-HDRs and food records and interactive computer- and camera-based technologies will certainly improve

In order to estimate the intake of flavonoids, a food composition table or database is also needed. There are currently two main databases. The first one is the US Department of Agriculture (USDA) database on flavonoids, isoflavones, and proanthocyanidins, which was created in 2003, 1999, and 2004, respectively [13–15]. Since then, several updates have been released. The USDA databases contain worldwide food composition data on the six main flavonoid classes (flavanols, flavonols, flavanones, flavones, isoflavones, and anthocyanidins) expressed as aglycones. Phenol-Explorer is the second most common database [16]. It was developed in 2009 and contains worldwide data, but in this case, on all nine flavonoid classes (including chalcones, dihydrochalcones, and dihydroflavonols) expressed as they are found in nature (mainly as glycosides) and analyzed with chromatography without hydrolysis. Phenol-Explorer also contains data on chromatography after hydrolysis and, in this case, flavonoids are expressed as aglycones, but these are not usually used. In the studies using Phenol-Explorer using flavonoid data expressed as aglycone equivalents, flavonoid glycoside contents are converted into aglycone contents using their respective molecular weights [17, 18]. The other important difference between databases is that Phenol-Explorer does not contain data on thearubigins because the composition data quality on thearubigins is very low [19]. The nonspecific spectrophotometric method used to date only provides a crude estimation of their

concentrations in black teas, which are the only known food sources of thearubigins.

These differences between databases (classes of flavonoids, aglycones vs. glycosides, and inclusion of thearubigins) complicate the comparison among studies using different databases. In this book chapter, we have mostly compared studies using the USDA databases or Phenol-Explorer with data expressed as aglycones. In our comparisons, we have also made distinctions between

of monosaccharide attached, and its position [10].

3. Dietary assessment of flavonoids

374 Flavonoids - From Biosynthesis to Human Health

this process [12].
