Validity of the Intake of Sugars, Amino Acids, and Fatty Acids Estimated Using a Self-administered Food Frequency Questionnaire in Middle-aged and Elderly Japanese: The Japan Public Health Center-based Prospective Study for the Next Generation (JPHC-NEXT) Protocol Area

Background The Japanese database of food composition was revised in 2020, during which both the number of food items and the number of food items measured for sugars, amino acids, and fatty acids were increased. We evaluated the validity of estimated intakes of sugars, amino acids and fatty acids using a long food frequency questionnaire (long-FFQ) among middle-aged and elderly Japanese. Methods From 2012 to 2013, 240 men and women aged 40–74 years from five areas in the JPHC-NEXT protocol were asked to respond to the long-FFQ and provide a 12-day weighed food record (WFR) as reference. The long-FFQ, which included 172 food and beverage items and 11 seasonings, was compared with a 3-day WFR, completed during each distinct season, and validity was assessed using Spearman’s correlation coefficients. Results Percentage differences based on the long-FFQ with the 12-day WFR in men and women varied from −84.4% to 419.6%, and from −75.8% to 623.1% for sugars, −17.5% to 3.8% and −5.8% to 19.6% for amino acids, and −58.5% to 78.8% and −43.4% to 129.3% for fatty acids, respectively. Median values of correlation coefficients for the long-FFQ in men and women were 0.52 and 0.42 for sugars, 0.38 and 0.37 for amino acids, and 0.42 and 0.42 for fatty acids, respectively. Conclusion The long-FFQ provided reasonable validity in estimating the intakes of sugars, amino acids, and fatty acids in middle-aged and elderly Japanese. Although caution is warranted for some nutrients, these results may be used in future epidemiological studies.

eTable 1A.Daily intakes of sugars according to the short-FFQ, and percentage differences between intakes by the short-FFQ and 12d-WFR and their correlations in men and women aged 40-74 years Men (n=92) Women eTable 1B.Daily intakes of amino acids according to the short-FFQ, and percentage differences between intakes by the short-FFQ and 12d-WFR and their correlations in men and women aged 40-74 years eTable 1C.Daily intakes of fatty acids according to the short-FFQ, and percentage differences between intakes by the short-FFQ and 12d-WFR and their correlations in men and women aged 40-74 years  d Deattenuated CCx = energy-adjusted CCx × SQRT(1 + λx/n), where λx is the ratio of withinto between-individual variance for nutrient x and n is the number of dietary records (12 days).
eTable 2. Daily intakes of energy and major nutrients according to the long-FFQ, and percentage differences between intakes by the long-FFQ and 12d-WFR and their correlations in men and women aged 40-74 years Men (n=98)

Yokoyama et al. a
Women (n=142) d Deattenuated CCx = energy-adjusted CCx × SQRT(1 + λx/n), where λx is the ratio of withinto between-individual variance for nutrient x, and n is the number of dietary records (12 days).e 411 food items were substituted with the value of protein because there were unmeasured values for the sum of amino acid residues.f 423 food items were substituted with the value of fat because there were unmeasured values for triacylglycerol equivalents.g Total dietary fiber was measured by the Prosky or modified Prosky method.h Water-soluble dietary fiber measured by the Prosky or modified Prosky method.i Water-insoluble dietary fiber measured by the Prosky or modified Prosky method.
eTable 3. Daily intakes of energy and major nutrients according to the short-FFQ, and percentage differences between intakes by the short-FFQ and 12d-WFR and their correlations in men and women aged 40-74 years Men (n=92)  b The long-FFQ2 was conducted from 1 year after the long-FFQ1 (November 2013).c ICC was performed using the SAS MIXED procedure.d Total dietary fiber was derived by combination of the AOAC.2011.25 method with either the Prosky or modified Prosky method.e Total dietary fiber (AOAC.2011.25)measured using only the AOAC.2011.25 method.f Low-molecular-weight water-soluble dietary fiber that remains soluble in 78% aqueous ethanol.g High-molecular-weight water-soluble dietary fiber that precipitates from 78% aqueous ethanol.h Water-insoluble dietary fiber measured using only the AOAC.2011.25 method.i 95% CI not calculated due to the small number of measured food items and low intake.j Total dietary fiber measured using the Prosky or modified Prosky method.k Water-soluble dietary fiber measured using the Prosky or modified Prosky method.l Water-insoluble dietary fiber measured using the Prosky or modified Prosky method.
Yokoyama et al.Validity of Short and Long Self-Administered Food Frequency Questionnaires in Ranking Dietary Intake in Middle-Aged and Elderly Japanese in the Japan Public Health Center-Based Prospective Study for the Next Generation (JPHC-NEXT) Protocol Area.J Epidemiol.2016;26:420-32.12d-WFR,12-day weighed food record; CC, correlation coefficient; FFQ, food frequency questionnaire; SD, standard deviation; NA, not applicable.Percentage differences: (FFQ-12-day WFR)/12-day WFR × 100 (%).P-values refer to the paired t-test between intakes by the long-FFQ and those by 12-day WFR for each; * P<0.05, ** P<0.01.c Spearman's rank correlation coefficients based on energy-adjusted values (other than energy intake and total fat in % energy) and expressed as deattenuated CC. * P<0.05, ** P<0.01.
a b Yokoyama et al.Validity of Short and Long Self-Administered Food Frequency Questionnaires in Ranking Dietary Intake in Middle-Aged and Elderly Japanese in the Japan Public Health Center-Based Prospective Study for the Next Generation (JPHC-NEXT) Protocol Area.J Epidemiol.2016; 26:420-32.12d-WFR, 12-day weighed food record; CC, correlation coefficient; FFQ, food frequency questionnaire; SD, standard deviation; NA, not applicable.b Percentage differences: (FFQ-12-day WFR)/12-day WFR × 100 (%).P-values refer to the paired t-test between intakes by the short-FFQ and those by 12-day WFR for each; * P<0.05, ** P<0.01.c Spearman's rank correlation coefficients based on energy-adjusted values (other than energy intake and total fat in % energy) and expressed as deattenuated CC. * P<0.05, ** P<0.01.d Deattenuated CCx = energy-adjusted CCx × SQRT(1 + λx/n), where λx is the ratio of withinto between-individual variance for nutrient x, and n is the number of dietary records (12 days).e 411 food items were substituted with the value of protein because there were unmeasured values for sum of amino acid residues.f 423 food items were substituted withthe value of fat because there were unmeasured values for triacylglycerol equivalents.g Total dietary fiber was measured by the Prosky or modified prosky method.h Water-soluble dietary fiber measured by the Prosky or modified Prosky method.i Water-insoluble dietary fiber measured by the Prosky or modified Prosky method.Daily intakes of energy and nutrients according to the FFQs and their intra-correlations among men and women aged 40-74 years ICC, intraclass correlation coefficient; CI, confidence interval; FFQ, food frequency questionnaire; SD, standard deviation.a The long-FFQ1 was conducted in November 2012.