The present study aimed to (1) examine changes in diet quality throughout pregnancy and (2) identify maternal characteristics associated with trimester-specific diet quality. Pregnant women (n = 79) were recruited in their 1st trimester of pregnancy and completed, at each trimester, three web-based 24-hour dietary recalls, from which the Canadian Healthy Eating Index (HEI) was calculated. Physical activity, nutrition knowledge, and socio-demographic web-questionnaires were also completed. Although no variation in total HEI scores was observed across trimesters, we found an overall decrease in the following subscores: adequacy, total fruits and vegetables, unsaturated fats and saturated fats (p < 0.05). In the 1st trimester, overweight and obese pregnant women had a lower diet quality in comparison with normal-weight and underweight women (HEI scores: 63.1 ± 11.9 vs. 68.0 ± 9.3; p = 0.04). In the 3rd trimester, women younger than 28 years old, with no university degree, poorer nutrition knowledge and who reside in an urban setting, had a lower diet quality (p < 0.05). In conclusion, less educated, younger women who reside in an urban setting may be at a higher risk of poor diet quality in late pregnancy and could benefit from public health programs.

To assess how nutrition knowledge is associated with global diet quality and to investigate whether sociodemographic characteristics (ie, sex, age, education, income, marital status, and living with children or not) moderate this association. Cross-sectional web-based study. The PREDISE study aims at identifying correlates of adherence to healthy eating guidelines in French-speaking adults from the Province of Quebec, Canada. A probability sample of 1092 participants (50% female). The Nutrition Knowledge Questionnaire and 24-hour food recalls from which the Canadian Healthy Eating Index (C-HEI) was calculated. Multiple linear regressions performed to assess how nutrition knowledge is associated with the C-HEI. Interaction terms tested to evaluate whether sociodemographic characteristics moderate the association between nutrition knowledge and the C-HEI. Nutrition knowledge (B = 0.141 [95% CI: 0.075-0.208], P < .0001) was identified as a significant correlate of the C-HEI. Education significantly moderated the association between nutrition knowledge and the C-HEI (P interaction = .0038), with a significative association among participants with a lower education level (B = 0.295 [95% CI: 0.170-0.421], P < .0001) but not among participants with a higher education level (B = 0.077 [95% CI: −0.004 to 0.157], P = .06). Whether participants lived with or without children also significantly moderated the association (P interaction = 0.0043); nutrition knowledge was associated with the C-HEI only in participants who were not living with children (B = 0.261 [95% CI: 0.167 to 0.355], P < .0001). This study suggests that the association between nutrition knowledge and adherence to healthy eating guidelines is not the same in different subgroups of the population. Interventions aiming at increasing nutrition knowledge may be a promising approach to improve diet quality, especially among individuals with a lower education.

The purpose of this study was to develop and validate a questionnaire assessing food liking in a French-Canadian population. A questionnaire was developed, in which participants were asked to rate their degree of liking of 50 food items. An expert panel evaluated the content validity. For the validation study, 150 men and women completed the questionnaire twice. An Exploratory Factor Analysis (EFA) was performed to assess the number of subscales of the questionnaire. Internal consistency and test-retest reliability of the subscales were evaluated. Concurrent validity was assessed through correlations between liking scores and self-reported frequencies of consumption. Comments from the experts led to changes in the list of foods included in the questionnaire. The EFA revealed a two-factor structure for the questionnaire (i.e., savory and sweet foods) and led to the removal of nine items, resulting in a 32-item questionnaire. The two subscales revealed good internal consistency (Cronbach alphas: 0.85 and 0.89) and test-retest reliability (p = 0.84 and 0.86). The questionnaire demonstrated adequate concurrent validity, with moderate correlations between food liking and self-reported frequency of consumption (r = 0.19–0.39, ps < 0.05). This new Food Liking Questionnaire assessing liking of a variety of savory and sweet foods demonstrated good psychometric properties in every validation step. This questionnaire will be useful to explore the role of food liking and its interactions with other factors in predicting eating behaviors and energy intake.

The objectives were to assess whether social support for healthy eating and perceived food environment are associated with diet quality, and to investigate if sociodemographic characteristics moderate these associations. A probability sample of French-speaking adults from the Province of Québec, Canada, was recruited in the context of the PREDISE study. Participants reported their perceptions of supportive and non-supportive actions related to healthy eating from close others at home and outside of home (n = 952), and of the accessibility to healthy foods (n = 1035). The Canadian Healthy Eating Index (C-HEI) was calculated based on three Web-based 24 h food recalls. Multiple linear regression models showed that supportive (B = 1.50 (95% CI 0.46, 2.54)) and non-supportive (B = −3.06 (95% CI −4.94, −1.18)) actions related to healthy eating from close others at home were positively and negatively associated with C-HEI, respectively, whereas actions from close others outside of home were not. The negative association between non-supportive actions occurring at home and C-HEI was stronger among participants with lower (vs. higher) levels of education (p interaction = 0.03). Perceived accessibility to healthy foods was not associated with C-HEI (p > 0.05). These results suggest that the social environment may have a stronger influence on healthy eating than the perceived physical environment. This adds support for healthy eating promotion programs involving entire families, especially for more socioeconomically disadvantaged individuals, whose efforts to eat healthily may be more easily thwarted by non-supportive households.

Intuitive eating is a positive approach to eating involving high responsiveness to hunger and satiety cues. Although the association between intuitive eating and many psychological health indicators is well-documented, its associations with overall diet quality and physical health indicators are less explored. This study aimed to assess whether intuitive eating is associated with adherence to Canada's Food Guide recommendations and with physical health indicators. In the context of the PREDISE study, 1088 French-speaking Canadian adults from the Province of Québec completed a web-based survey including the Intuitive Eating Scale-2. Three web-based 24-h dietary recalls were completed and used to calculate the Canadian Healthy Eating Index (C-HEI) score, an indicator of adherence to Canada's Food Guide. Participants visited the research center to assess physical health indicators (body mass index, waist circumference, blood pressure, cholesterol, and blood glucose). Multiple linear regression analyses were performed to assess how intuitive eating is associated with diet quality and physical health indicators. The total intuitive eating score was not associated with C-HEI (B = 1.47, 95% CI: -0.10, 3.04). However, the "Body-Food Choice Congruence" subscale was positively associated with C-HEI (B = 5.60, 95% CI: 4.44, 6.76), while the "Unconditional Permission to Eat" subscale was negatively associated with C-HEI (B = -4.04, 95% CI: -5.28, -2.80). The total intuitive eating score was negatively associated with BMI in men (B = -4.83, 95% CI: -5.75, -3.91, p < 0.0001) and women (B = -5.01, 95% CI: -6.00, -4.02, p < 0.0001), with waist circumference in men (B = -3.60, 95% CI: -5.33, -1.87) and women (B = -2.20, 95% CI: -3.71, -0.70), and with HDL cholesterol only in women (B = -0.10, 95% CI: -0.17, -0.02). The associations of intuitive eating with diet quality are inconsistent, and only a few associations with physical health indicators were observed.