Background/objective: Taste is of key importance in food choice and dietary patterns, but studies on taste profiles are limited. We previously assessed dietary taste patterns by 24 h recalls (24hR), but for epidemiological studies food frequency questionnaires (FFQ) may also be suitable. This study compared dietary taste patterns based on FFQ against 24hR and biomarkers of exposure.
Subjects/methods: A taste database including 467 foods' sweet, sour, bitter, salt, umami and fat sensation values was combined with food intake data to assess dietary taste patterns: the contribution to energy intake of 6 taste clusters. The FFQ's reliability was assessed against 3-d 24hR and urinary biomarkers for sodium (Na) and protein intake (N) in Dutch men (n = 449) and women (n = 397) from the NQplus validation study (mean age 53 ± 11 y, BMI 26 ± 4 kg/m2).
Results: Correlations of dietary taste patterns ranged from 0.39-0.68 between FFQ and 24hR (p < 0.05). Urinary Na levels, but not N levels, were positively associated with % energy intake from 'salt, umami & fat' tasting foods (Na; FFQ, r = 0.24, 24hR, r = 0.23, p < 0.001, N; FFQ, r = 0.08, p = 0.1394, 24hR, r = 0.05, p = 0.3427).
Conclusions: The FFQ's reliability against 24hR was acceptable to good for ranking of adults' dietary taste patterns. Associations between dietary taste patterns and urinary Na and N were similar for FFQ and 24hR. These findings suggests that both FFQ and 24hR can be used in combination with our taste database, to investigate potential relationships between dietary taste patterns and subgroups at risk of obesity and chronic diseases such as cardiovascular disease.
Trial registration: ClinicalTrials.gov NCT03233503.