Is the more bitter diet, the healthier?

Higher dietary consumption of bitter foods and beverages such as brassica vegetables, tea and coffee may have beneficial health properties and reduce the risk of chronic disease.^(1–3) However, little is known about the bitterness of diets and the characteristics of bitter diets. This study aimed to explore: (a) the bitterness of diets; (b) the nutrient and food group composition, and quality of bitter diets; and (c) the contribution of foods and beverages to the bitterness of diets, using data from a nationally representative population of Australia. This study was a secondary analysis of the 2011–2012 National Health and Physical Activity Survey. Dietary data from 9,118 Australian adults were obtained from 24-hour dietary recalls and were combined with an established sensory foods database to assess the bitterness (bitter score) of individuals’ diets (bitter score = bitter taste × amount of foods/beverages (g) consumed). Diets were then categorised into tertiles of bitterness (n = ~3,039/group) with 23,098 > and 44,166 < threshold bitter scores for the most and the least bitter diets. Data were analysed using ANOVA with the Bonferroni post hoc test. Across the whole sample (51% male, 46.6 ± 17.5 years old), the bitter score of individuals’ diets ranged from 377 to 270,403 (mean 37,719 ± 26,008). The bitter score of the most bitter diets was almost five times greater than that of the least bitter diets (66,966 ± 21,881 v. 13,193 ± 5,819; p < 0.001). The most bitter diets were higher in dietary fibre (7.1%) and total fat (4.6%), and lower in carbohydrate (3.8%) compared with the least bitter diets (p < 0.001 for all). The most bitter diets scored slightly higher on a diet quality index assessing compliance to the Australian dietary guidelines (44.3 ± 17.2 v. 41.4 ± 17.9, p < 0.001), consuming more serves of water (59.8%), vegetables (30.2%), fresh/canned fruits (33.9%), wholegrain/high fibre cereal (36.2%) and low-fat dairy (57.2%), and lower serves of refined/low fibre cereals (−18.0%) compared with the least bitter diets (p < 0.001 for all). Overall, beverages made a greater contribution to the total bitterness of diets than foods (87% v. 13%, p < 0.001), which was more pronounced in the most bitter diets compared with the least bitter diets (94.6% v. 79.7%, p < 0⋅001). Greater intakes of tea/coffee (285.4%) and milk/milk substitutes (5.6%), and a lower intake of soft drinks (−16.2%) and juice (−10.5%) were reported in the most bitter diets compared with the least bitter diets (p < 0.001 for all). The nutritional profile of the most bitter diets among Australian adults appeared to be healthier than the least bitter diets. A greater understanding of which specific foods are driving the bitterness of diets and how they are being consumed is warranted.