Journal of Womens Health, Issues and Care ISSN: 2325-9795

All submissions of the EM system will be redirected to Online Manuscript Submission System. Authors are requested to submit articles directly to Online Manuscript Submission System of respective journal.

Research Article, J Womens Health Issues Care Vol: 5 Issue: 2

Healthy Eating Attitude: A mediator of Nutrition Knowledge and Diet Quality Using the Healthy Eating Index-2010 in Young Women

Tamara Tabbakh and Jeanne Freeland-Graves*
University of Texas at Arlington, School of Social Work, 211 S. Cooper Street, Arlington, Texas, USA
Corresponding author : Jeanne H Freeland-Graves, PhD, RD, FACN
Bess Heflin Centennial Professor, Department of Nutritional Sciences, 103 W. 24th St., PAI 4.44, The University of Texas at Austin, Austin, Texas 78712, USA,
Tel: 512 471 0657;
Fax: 512 471 5844;
E-mail: [email protected]
Received: September 11, 2015 Accepted: March 11, 2016 Published: March 16, 2016
Citation: Tabbakh T, Freeland-Graves J (2016) Healthy Eating Attitude: A mediator of Nutrition Knowledge and Diet Quality Using the Healthy Eating Index-2010 in Young Women. J Womens Health, Issues Care 5:2. doi:10.4172/2324-8793.1000116


Objective: To investigate the impact of nutrition knowledge and psychological determinants of eating behavior (healthy eating attitude, self-efficacy, emotional eating, fat habits, and mindless eating) on diet quality using the Healthy Eating Index-2010 in young women.
Methods: In this cross-sectional design, 114 college women completed a 3-day food record, nutrition knowledge scale, and questionnaire measuring psychological determinants of eating behavior (healthy eating attitude, self-efficacy, emotional eating, fat habits, and mindless eating). Regression analysis was utilized to discern the influence of nutrition knowledge and psychological determinants of eating behavior on diet quality. A mediation model was conducted to further explore relationships between variables.
Results: Mean percent maximum score for diet quality was 59.6%. Nutrition knowledge (p<0.05), healthy eating attitude (p<0.01), and favorable fat habits (p<0.01) were positively associated with diet quality. Healthy eating attitude emerged as a partial mediator of nutrition knowledge on diet quality (β=0.72, CI: 0.3-1.4), resulting in a 34% reduction in the model.
Conclusions and Implications: Diet quality in this sample is less than ideal. Nutrition knowledge was a key predictor of this outcome, via enhancement of attitudes toward healthy eating.

Keywords: Diet quality; Nutrition knowledge; Healthy eating attitude; Healthy eating index-2010; Women


Diet quality; Nutrition knowledge; Healthy eating attitude; Healthy eating index-2010; Women


The prevalence of obesity among the adult population in the United States has reached alarming rates, with 68.5% classified as overweight or obese [1]. College is a critical time for expanding obesity trajectory, as young adults, ages 18 to 29 years exhibit the greatest increase in weight gain as compared to other ages. In addition, those who attend college are more likely to gain weight than their non-college counterparts [2]. When college students relocate to new environments and gain independence in their daily lives [3], weight may increase due to conventional barriers to healthy eating such as high cost, need for convenience, and time constraints [4]. A concomitant elevation in responsibilities, more frugal lifestyle, and academic and personal stressors [5] also may impact the ability to secure a healthful and balanced diet. Women students, in particular, may encounter psychological challenges such as body weight and shape concerns, raising their likelihood to engage in unhealthful behaviors [6,7].
Nutrition awareness may be a critical component for achieving favorable diet quality. Wardle et al. observed that adults who were in the highest quintile of nutrition knowledge were more likely to consume a healthy diet, defined as meeting recommendations for fruits, vegetables, and fat [8]. Similarly, de Vriendt et al. reported that nutrition knowledge was related to significantly higher intake of fruits and vegetables [9]. These studies suggest that nutrition knowledge is associated with a healthy diet. However, the impact of nutrition knowledge on diet quality in the literature is equivocal [10,11]. Brown et al. found that fruits, vegetables, protein, and dairy intake did not vary between college students who received nutrition education using text messages and controls [10]. Bu et al. observed similar results, with no significant differences in fat, calcium, sodium, or iron intake following a 13-week nutrition education intervention [11].
The lack of agreement with regards to the importance of nutrition knowledge may be in part attributable to the fact that nutrition awareness alone may be insufficient for effecting positive dietary behaviors. For example, the degree to which knowledge can translate into desirable dietary choices can be modulated by psychological factors, as behavior is subject to the ability of an individual to exercise control over their food environment [12]. In an intervention tailored for low-income mothers, the authors concluded that nutrition knowledge, social support and attitudes regarding healthful eating were significant predictors of weight loss [12]. Therefore, one component of achieving healthy weight status is regulation of psychological aspects of eating such as healthy eating attitude, selfefficacy, and emotional eating. Thus, both nutrition knowledge and psychological factors related to eating are potential mediators of optimal diet quality.
The goal of the current research was to examine diet quality of college women using the Healthy Eating Index-2010 and the extent to which nutrition knowledge and psychological characteristics of eating impact this outcome. This index reflects compliance with key recommendations of the Dietary Guidelines for Americans (DGA) [13,14]. A mediation effect was hypothesized to explain possible relationships between nutrition knowledge, psychological determinants of eating behavior (healthy eating attitude, self-efficacy, emotional eating, fat habits, and mindless eating), and diet quality. It is proposed that the relationship between nutrition knowledge and diet quality is not direct, but rather, mediated by psychological determinants of eating behavior.

Materials and Methods

Study design and subjects
A sample of 114 women was recruited from a traditional 4-year university. All participants completed a 3-day food record and online surveys using Qualtrics© including demographics, nutrition knowledge scale, and psychological determinants of eating behavior questionnaire (healthy eating attitude, self-efficacy, emotional eating, fat habits, and mindless eating) [15]. Qualtrics© is a data collection tool that provides extensive security and privacy (Qualtrics, Provo, UT 2011) [16]. The risks and benefits of the study were explained and informed consent was obtained. This study was deemed exempt by The University of Texas at Austin Institutional Review Board.
Demographic Survey
A revised version of a 29-item demographics questionnaire was administered to all participants [12]. Demographics include items such as age, academic major, grade level, and race.
Dietary intake
Participants completed estimated food records for 3 consecutive days; 2 weekdays and 1 weekend day, avoiding days of illness. Nutrient levels obtained from weighted food records are often regarded as the gold standard of dietary methods [17] and have been validated against urinary and blood metabolites, showing adequate mean correlations (r=0.38-0.54) [18,19]. Although estimated food records were used in this research, participants were provided with detailed instructions and training on how to record all foods and beverages consumed prior to completing the study. Additionally, food models and measuring guides were provided to enhance accuracy and validity of food records. Subjects specified type of food (brand, fresh or processed), preparation method and type of fat, cut of meat, fat content, size of item and amount consumed in a food diary. Each form was divided into 3 days, with sections for breakfast, lunch, dinner, and snacks. Completed food records were obtained and food entries were converted to appropriate measurement units. Mixed foods were deconstructed into their constituents and assigned respectively. Dietary intake data was analyzed using My Diet Analysis (Version 8.2.5, ESHA Research, Inc, Salem, OR). Diet quality was evaluated using the Healthy Eating Index-2010, a composite measure of key food groups and nutrients, derived from Dietary Guidelines for Americans [14]. A total of 12 components were calculated on a scale of 0-100 points for adequacy and moderation, with 100 points reflecting perfect adherence to recommendations. Density values were computed by conversion of the intake of each component to resemble criterion provided by the HEI-2010 (intake per 1,000 calories). Individual HEI components were computed by linear equations [14], using the standard for minimum and maximum scores. Component scores were summed to produce a total score. Predicted estimates were constrained to the highest possible value for each component.
Nutrition knowledge scale
The 20-item questionnaire of nutrition knowledge was modified from a previously validated scale [20]. It is composed of multiplechoice questions concerning nutrients, fast food nutrition, MyPlate recommendations, physical activity, and dietary guidelines. An expert panel in nutrition reviewed the items to establish content validity. Psychometrics of the current scale were evaluated in a sample of adult women (n=114) [15]. Internal consistency reliability and testretest reliability were α=0.74 and r=0.86 (p<0.001), respectively. The final scale was condensed to 19 items by eliminating questions that had high endorsement on one response (>97% sample answered correctly) because they offered low discriminatory power [21].
Psychological determinants of eating behavior
A 12-item scale developed and validated previously in a sample of adult women (n=114) was used to assess psychological determinants of eating behavior [15]. All sub-scales exhibited high content validity, as determined by a nutrition expert panel (n=10) composed of Ph.D. nutritionists and Registered Dietitians. Internal consistency reliability and test-retest reliability for subscales were: healthy eating attitude (α= 0.89, r=0.84, p<0.01), self-efficacy (α= 0.83, r=0.90, p<0.01), emotional eating (α =0.87, r=0.87, p<0.01), fat habits (α= 0.68, r=0.81, p<0.01), and mindless eating (α= 0.73, r=0.89, p<0.01). Subscales were operationally defined as degree of positive perspective regarding healthful eating, confidence in ones’ ability to ensure a healthy diet, consumption of food in response to emotional stimuli, use of fat for flavor enhancement, and engagement in screen time while eating, respectively. Questions were in a Likert scale format, ranging from 1=strongly disagree to 5=strongly agree. Fat habits were reversecoded such that a higher score reflected more desirable fat habits
Statistical analysis
All analyses were conducted using Statistical Product and Service Solutions (Version 20.0, SPSS, Armonk, NY, 2011). Analysis of variance was used to determine differences in HEI-2010 scores, according to Body Mass Index (BMI) categories. Linear regression with nutrition knowledge, healthy eating attitude, self-efficacy, emotional eating, dietary restraint, and fat habits as independent variables and total HEI score as the dependent variable was conducted. A mediation effect was tested; it was proposed that the relationship between nutrition knowledge (independent variable) and total HEI score (dependent variable) was mediated via psychological factors related to eating. The model was tested by the Preacher and Hayes Multiple Mediation Model, which utilizes bootstrapping. This technique involves sampling, with replacement, from the original data to obtain estimates of a sampling distribution curve and provides more robust estimates of confidence intervals, particularly for smaller sample sizes [22,23].


Figure 1 illustrates total and component HEI 2010 scores, according to BMI categories (underweight/healthy weight and overweight/obese). Mean score for total HEI was 59.6%, ranging from 30.5-89.5%. Linear trends were observed for non-overweight/ obese participants, with consistently higher scores on all categories, except dairy and refined grains. Scores for all components were <75% of the maximum possible score, with the exception of total protein (90.2%) and empty calories (78.5%). Thus, these college women were consuming below the recommended amounts for total fruits (65.6%), whole fruits (70.2%), total vegetables (71.5%), greens and beans (53.3%), whole grains (40.2%), dairy (53.3%), seafood and plant protein (72.6%), and fatty acids (15.3%). In contrast, their diets were greater than recommended for refined grains (68.9%), sodium (49.8%), and empty calories (78.5%).
Figure 1: Representative signals of AIMS sum signal and humidity of the exhaled breath sample measured with ChemPro®100 and multicapillary column used. The AIMS spectrum from the maximum point of the AIMS sum signal was analyzed further.
The contribution of nutrition knowledge and psychological determinants of eating behavior on diet quality was the second focus of this research. Using the enter method, the final regression model [F (3,110) = 19.7, p<.001] demonstrated that nutrition knowledge (p<0.05), healthy eating attitude (p<0.001), and fat habits (p<0.001) were positive predictors of diet quality (Table 1). Self-efficacy, emotional eating, and mindless eating were excluded in the final model, as they were not significantly associated with diet quality. The R2 for the total model was 34.9%, indicating that nutrition knowledge, healthy eating attitude, and fat habits explained one-third of the variability in diet quality.
Table 1: Psychological Characteristics of Eating Behavior according to Diet Quality, using the Healthy Eating index-2010 in College Women (n=114).
Figure 2 illustrates the effect of nutrition knowledge (independent variable) on diet quality (dependent variable), including significant mediators only. The effect of the independent variable on the mediator variable (pathway 2) and the mediator variable on the dependent variable (pathway 3) were both significant. Thus, steps of mediation were continued, and bootstrapped, regression coefficient for the indirect effect were computed. Lower and upper bias-corrected, bootstrapped confidence intervals for the indirect effect did not contain zero (healthy eating attitude: β=0.72, CI: 0.3-1.4), suggesting partial mediation, with a 34% reduction in the model. Estimates of the indirect effect derived from bootstrapping reflect the magnitude of mediation, i.e., how much of the effect of nutrition knowledge on diet quality was influenced by a more positive healthy eating attitude. In this analysis, proficient knowledge in nutrition appeared to be linked to more positive attitudes toward healthy eating, which was associated with enhanced diet quality.
Figure 2: Representative signals of AIMS sum signal and humidity of the exhaled breath sample measured with ChemPro®100 and multicapillary column used. The AIMS spectrum from the maximum point of the AIMS sum signal was analyzed further.


The present findings are congruent with national reports on diet quality of age-comparable females in the United States [24]. For example, mean total fruit and total vegetable intakes were 1.1 cup and 1.4 cup equivalents among NHANES females, as compared to 1.3 cup equivalent and 1.7 cup equivalents, respectively, in the present analysis. The overconsumption of refined grains and underconsumption of whole grains have been noted in both NHANES women and this study. These findings are supported by other research, in which college students consumed diets high in sodium [25], saturated fat [26], and calories [27], as well as low levels of fruits and vegetables [28] and whole grains [29]. In a recent study, Roy et al. showed similar patterns, with low adherence to fruit, vegetable, sodium, and discretionary calories recommendations among Australian young adults [30]. Another report revealed poorer diet quality in a sample of overweight and obese Australian young adults than in the present sample, with percent of maximum scores of 45.4% and 59.6%, respectively [31].
Nutrition knowledge was found to be a significant predictor of diet quality. These results are consistent with previous findings that revealed a positive association between nutrition knowledge and diet quality [32]. Additionally, a 3-week educational intervention focusing on nutrition knowledge in low-income women yielded substantial reductions in fast-food consumption, added salt, sugar, or butter to foods, and consumption of sodas, cookies and sweets and simultaneous increases in vegetable intake, use of nutrition labels, and physical activity [33]. The above investigations corroborate results from the present research, in which greater nutrition knowledge was positively related to diet quality. Although evidence of a positive association between self-efficacy and nutrition label use with diet quality has been documented by others [34], self-efficacy was not significantly related to diet quality.
In the present investigation, greater knowledge in nutrition was associated with desirable attitudes toward healthy eating, ultimately contributing to greater compliance with dietary guidelines. Thus, it appears that proficiency in nutrition fosters favorable attitudes in college women, which likely impacts their ability to make healthy food choices. A positive relationship between healthy attitudes and dietary behaviors has been reported in the literature [32]. This study extends on previous findings by providing evidence for a pathway by which nutrition knowledge and attitudes interact to effect positive dietary change.
This study contributes to the literature by providing a potential mechanism by which nutrition knowledge impacts overall diet quality of college women. Rather than a direct association, it was observed that greater nutrition knowledge was associated with more positive attitudes toward eating and consequently more optimal diet quality in college women. This finding offers a likely explanation for the discrepancies in the literature regarding the significance of nutrition knowledge as a determinant of diet quality. To our knowledge this study is the first to explore the impact of nutrition knowledge and psychological determinants of eating behavior on diet quality using the HEI-2010 in college women. This instrument has great utility, as it had demonstrated high content and construct validity [35] in various populations. Improvement of diet quality of college women is a public health priority, as this transition phase introduces and reinforces dietary habits that are associated with disease risk [36] and may persist throughout life. For example, dietary patterns of college attendees have been found to be significant predictors of levels of triglyceride, glucose, and high-density lipoprotein cholesterol, as well as waist circumference [37].
One limitation of the present study is that the cross-sectional nature restricts its ability to capture changes over time. Although the study spanned several seasons (Spring 2013-2014), major diet quality differences were not anticipated, as diets of young adults are low in fruits and vegetables [24] and are thus unlikely to be affected by seasonal variations. The rule of thumb for adequate power recommends a minimum of 10 participants per predictor variable [38]. The present analysis included 23 participants per predictor variable. Additionally, use of bootstrapped replications in the mediation model minimizes this effect through provision of more precise parameter estimates [22]. Thus, the relatively small sample size does not pose any threat to the validity of the study.
Social ecological models suggest incorporation of social and environmental factors for a more comprehensive evaluation of diet quality [39]. However, this was beyond the scope of this research, as this study focused on psychological determinants of eating behavior. The factors studied were chosen based on an extensive literature review highlighting the impact of healthy eating attitude [40], selfefficacy [34], emotional eating [41], mindless eating [42], and fat habits [43] on diet quality in young adults. Other psychological stimuli such as mindful eating, intuitive eating, and disordered eating were not part of the instrument yet reports on their associations with nutrition knowledge and impact on diet quality are warranted. For example, recent research suggests that intuitive eating, which emphasizes eating in response to internal cues, has a positive impact on psychosocial outcomes such as positive attitudes toward healthy eating in adults [44]. Women were chosen as the sample for this study, as they are more likely to face psychological challenges such as body weight and shape concerns, which may impact their likelihood to engage in unhealthful behaviors [6,7]. However, exploration of the impact of nutrition knowledge on diet quality via psychological determinants in young men and comparisons between the sexes is recommended for further investigation. Thus, future endeavors would benefit from a multi-level approach and exploration of a range of psychological, social, and environmental factors and their potential to act as mediators between nutrition knowledge and diet quality.


The current study observed that college women enrolled in introductory nutrition classes were not meeting recommendations according to the HEI-2010. Percent of maximum component scores were lower than 75% of maximum for all HEI-2010 categories, except total protein and empty calories. Adherence to HEI-2010 recommendations was lowest for fatty acids, whole grains, sodium, and greens and beans suggesting the need for improvements in these areas. Nutrition knowledge, healthy eating attitude, and desirable fat habits emerged as significant predictors of dietary outcomes. More specifically, it was observed that elevated nutrition knowledge exerted positive influences on attitudes towards eating, which are related to healthier dietary practices. Therefore, interventions targeting young women should incorporate these components in nutrition education programs.


Bess Hefflin Centennial Professorship


Track Your Manuscript

Media Partners