What Is Healthy Eating? Exploring Profiles of Intuitive Eating and Nutritionally Healthy Eating in College Women

Abstract

Purpose

Research suggests that food’s nutritional content and the feelings and behaviors associated with eating contribute to overall health, yet these constructs are rarely considered simultaneously. The current cross-sectional study investigated healthy eating profiles in college women that included both nutritional quality and intuitive eating (IE), with IE being an eating style that prioritizes physiological hunger/satiety cues and minimizes dieting and emotional eating.

Design/Setting/Subjects

Undergraduate women (n = 352) completed an online assessment and daily diaries (80% retention).

Measures

Nutritionally healthy eating, IE, dieting, body dissatisfaction, disordered eating, and psychological health were examined.

Analysis

Latent profile analysis explored patterns of nutritionally healthy eating and IE, and determined how these patterns related to psychological and disordered eating outcomes.

Results

A four-profile solution emerged: 1) Dieting (high nutritionally healthy eating + intermediate IE), 2) Unhealthy Eating (low nutritionally healthy eating + low IE), 3) Intuitive Eating (moderately high nutritionally healthy eating + high IE), and 4) Non-Dieting (low nutritionally healthy eating + intermediate IE). These profiles significantly differed from one another, such that the Intuitive Eating and Non-Dieting profiles reported lower psychological distress, body mass index, and disordered eating compared to other profiles, while the Dieting and Unhealthy Eating profiles showed the healthiest and poorest nutritional eating, respectively.

Conclusion

These findings suggest that IE and nutritionally healthy eating are distinct factors in conceptualizing the eating patterns of college women, and both should be considered when developing interventions. Future research should replicate these findings in larger/more diverse samples and examine eating profiles longitudinally.

Keywords

healthy eating nutrition intuitive eating body dissatisfaction obesity

Introduction

Background

The content of one’s diet is an important factor in overall health. Poor dietary quality is associated with abdominal obesity¹ and various long-term health problems.² Unfortunately, many college students do not meet the recommendations for dietary quality,^3,4 despite possessing reasonable knowledge about healthy nutritional requirements.⁵ Lifestyle interventions that target college students’ eating have traditionally focused on dietary content and nutrition literacy, but may also include instruction on mindfulness of hunger and satiety cues.⁶ Increasing awareness of physiological hunger and eating for both health and enjoyment rather than following external eating rules (such as a daily caloric limit) has been shown to improve eating habits, metabolic fitness, and psychological well-being.⁷ This suggests that healthy eating is determined not only by the content of one’s diet, but also the behaviors and feelings surrounding eating.

Increased emotional eating during college may play a significant role in weight gain.⁸ One common type of emotional eating occurs when an individual attempts to dampen negative feelings with positive eating sensations.⁹ The drive to eat for pleasure rather than in response to physiological hunger is known as “hedonic hunger” and is positively associated with poor diet quality and emotional eating.^10,11 Overeating in response to negative emotions (e.g., stress, anxiety, and depression) is more often reported by women than men.¹²

Intuitive eating (IE) is an eating style that includes giving oneself unconditional permission to eat until satiated, eating for physiological reasons rather than emotional ones, and choosing nutritious foods that match the needs of one’s body.¹³ IE is associated with better psychological health in college students and lower body shame, body mass index (BMI), and disordered eating.^14,15 IE interventions for college students have been shown to improve mindful eating and the recognition of physical hunger cues and satiety, further indicating that IE is a teachable construct that decreases propensity towards dieting and emotional eating.^16,17

Both the nutritional quality of foods and the motivations underlying their selection contribute to an individual’s overall well-being. Previous studies have often failed to incorporate these constructs simultaneously when interpreting the eating habits of college students. This is an important consideration given that lifestyle interventions traditionally have had minimal effects as far as changing college students’ eating behaviors.⁶ More work is needed to understand precisely what healthy eating is, and to determine how to help college students find a comfortable approach to eating that is also nutritious. To this aim, the current paper explores a two-factor definition of healthy eating; one that includes both nutritionally healthy eating and IE.

A viable technique for exploring a two-factor definition of healthy eating is to examine profiles of eating behaviors within a college sample. Latent profile analysis (LPA) is able to identify subgroups within a sample based on scores on observed variables.¹⁸ Research has used this general approach to examine subgroups of weight-related health behaviors in college student samples. For example, VanKim and colleagues found four classes of eating habits in both college men and women: 1) healthier eating habits, 2) moderate eating habits, 3) unhealthy weight control, and 4) healthier eating habits, more physically active.^19,20 Although previous research suggests that subgroups of college students can be categorized based on diet quality and exercise, studies have not included measures of healthy eating behaviors. Evaluating these two elements in conjunction may be important for capturing the eating profiles of college women, and a broadened conceptualization of what it means to “eat healthy” could help to guide lifestyle interventions within college populations.

Purpose

The current study used LPA to investigate healthy eating profiles in college women. We hypothesized that four profiles would emerge: 1) high nutritionally healthy eating + high IE, 2) high nutritionally healthy eating + low IE, 3) low nutritionally healthy eating + high IE, and 4) low nutritionally healthy eating + low IE. Given previous research showing that IE is associated with improved self-esteem and life satisfaction in college women, we hypothesized that profiles characterized by high IE would report lower body dissatisfaction and disordered eating, lower BMI and use of dieting strategies, and higher psychological well-being.¹⁵ We also hypothesized that profiles high on nutritionally healthy eating would report more dieting and lower BMI, as women who endorse a sustained effort towards monitoring food intake tend to eat healthier and have lower BMIs.²¹ An examination of differences at the intersection of both nutritionally healthy and IE was largely exploratory.

Materials and Methods

Design and Sample

A total of 620 women and men who were at least 18 years of age were recruited from a large university in the Southwestern United States. Recruitment occurred via announcements in psychology classes and through a web-based system. Participants received either extra credit or course credit. As part of the study, participants completed four daily nutrition recalls. Participants who did not complete these recalls (n = 88) were not invited to participate in the online survey. There were 532 participants who were invited to complete the online survey; 27 participants never completed the survey. The final sample included a total of 496 male and female college students, and the overall retention rate was 80.0%. Males and one individual identifying as transgender (n = 131) and participants over the age of 40 (n = 7) were excluded due to research findings about differences in eating and body image across biological sex and age.^22,23 The men in this sample were of interest for separate studies examining eating behaviors of college students overall. Four individuals were eliminated because they indicated in the online survey that they were not fluent in English, and two were eliminated because they endorsed being pregnant. The final sample was 352 college women. See Table 1 for demographic details.

Table 1.

Demographics Characteristics of the Overall Sample (N = 352).

Demographic Characteristics	M (SD)	Minimum - Maximum
Age	20 (3.2)	18–38
BMI	23.8 (4.9)	16.1–47.2
	Category	Percentage ( n )
Ethnicity	Not Hispanic, Latino, or Spanish origin	44% (n = 155)
	Another Hispanic, Latino, or Spanish origin	36% (n = 125)
	Mexican, Mexican American, Chicana	20% (n = 72)
^aRace	White	64% (n = 224)
	Other	23% (n = 81)
	American Indian/Alaskan Native	8% (n = 28)
	Black/African American	4% (n = 15)
	Unavailable/Unknown	4% (n = 13)
	Asian	3% (n = 12)
Marital Status	Never Married	84% (n = 295)
	Cohabitating with a Partner	10% (n = 36)
	Married	5% (n = 17)
	Divorced	1% (n = 4)
Living Situation	Live with Roommates	40% (n = 140)
	Live with Parents	37% (n = 132)
	Live with Significant Other	13% (n = 46)
	Live Alone	10% (n = 34)
BMI Category	Underweight	6% (n = 21)
	Normal Weight	67% (n = 236)
	Overweight	16% (n = 56)
	Obese	11% (n = 39)
Food Allergy	Yes	18% (n = 63)
Food Allergy	No	82% (n = 289)
Special Diet	Yes	7% (n = 26)
Special Diet	No	93% (n = 326)
Medical Condition Affecting Eating	Yes	7% (n = 26)
Medical Condition Affecting Eating	No	93% (n = 326)

^aNote: Individuals could select more than one race.

All study procedures were approved by the university’s Institutional Review Board. Informed consent was completed online. Participants completed four (3 weekday and 1 weekend) 24-hour food recalls using the online Automated Self-Administered 24-hour Dietary Assessment Tool (ASA-24). E-mail prompts (including a weblink and unique username/password) were delivered in the morning and participants had until midnight that day to complete the recall. Only those participants who completed all four recalls were invited to participate in a two-hour online survey. Participants had approximately one week to complete the online survey.

Measures

Automated Self-Administered 24-hour Dietary Assessment Tool (ASA-24)²⁴

Data from this online dietary assessment tool are converted into the Healthy Eating Index (HEI) using a macro available via the National Cancer Institute’s ASA-24 website.²⁵ The macro produced HEI scores for each of the four daily diary entries completed by each participant, and these scores were averaged to produce average HEI scores. The HEI is based on the 2010 dietary guidelines by the U.S. Department of Agriculture and appears to be generally valid for measuring the extent to which a population’s diet conforms to the standards of the U.S. Dietary Guidelines^25,26

Body Shape Questionnaire (BSQ)²⁷

This measure assesses the extent to which one has been dissatisfied with their body over the past four weeks. Participants respond to 34 items using a six-point Likert-type scale ranging from 1 (never) to 6 (always). The instrument shows good reliability and validity in women with body image distress, those with overweight/obesity, and undergraduate women.²⁸ In this study, internal consistency was excellent, α= .98.

Demographics

A demographics form was used to collect data on age, sex, race/ethnicity, education status, living situation, marital status, dietary restrictions, and height and weight (used to calculate BMI).

Dieting and Weight Control Behavior Checklist (DWCBC)²⁹

The Dieting and Weight Control Behavior Checklist (DWCBC) begins with a question that asks whether participants have engaged in any dieting behavior in order to lose weight in the past year. Participants are then asked to check “Yes” or “No” regarding if they have used any of the strategies (e.g., skipped meals) listed on the checklist in the past year.

Eating Disorder Examination-Questionnaire (EDE-Q)³⁰

The Eating Disorder Examination-Questionnaire (EDE-Q) assesses overall eating pathology, the presence of which suggests unhealthy eating practices. The first 21 questions ask participants to report on how many of the last 28 days they have engaged in different thoughts/behaviors related to disordered eating. The next 7 questions ask participants to rate from 0 (not at all) to 6 (markedly) the degree to which certain eating disorder symptoms affected them over the past 28 days. Scores are averaged across items to provide a global score. The EDE-Q has good reliability and validity in various samples,³¹ and the overall Cronbach's α in this study was .95.

Intuitive Eating Scale (IES-2)¹⁵

IES-2 items are on a five-point Likert-type scale ranging from 1 (strongly disagree) to 5 (strongly agree). The four subscales correspond to the different tenets of intuitive eating: unconditional permission to eat, eating for physical rather than emotional reasons, reliance on hunger and satiety cues, and body-food choice congruence. The IES-2 had good reliability and construct validity in a college sample.¹⁵ In the current study, internal consistency was good, α=.86.

Mental Health Inventory (MHI)³²

This 38-item measure of psychological health provides two global indexes: psychological well-being and psychological distress. The items are on a six-point scale that varies by question. The factor structure was cross-validated in several samples and demonstrated satisfactory internal reliability.³² In this study, internal consistency of the psychological well-being (α=.92) and psychological distress (α=.94) indexes was excellent.

Power of Food Scale (PFS)³³

This 15-item measure assesses hedonic hunger; motivation to eat beyond physiological hunger. The items are on a five-point Likert-type scale ranging from 1 (don’t agree at all) to 5 (strongly agree). The PFS has demonstrated good reliability and validity.³³ In this sample, Cronbach's α was excellent, α=.94.

Analysis

Data were analyzed using Mplus version 8 and SPSS version 26.0.^34,35 Data were examined for normality, univariate and multivariate outliers, and collinearity, and these assumptions were met except where specifically stated. Nine (3%) multivariate outliers based on the IES-2 were eliminated from analyses, for a resulting sample size of 343.

Latent profile analysis (LPA) is a type of mixture model that assumes the presence of latent profiles of individuals within the data represented by latent variables.¹⁸LPA maximizes homogeneity within profiles and maximizes heterogeneity between profiles. Measures of IE (latent factor scores on the four IES-2 subscales) and measures of nutritionally healthy eating (the HEI’s adequacy and moderation scales) were entered into the model. All models were run using 1000 random starts and 500 final stage optimizations. In each model, the loglikelihood value was inspected to ensure that replication was achieved and local maxima were not reached. Models were estimated using maximum likelihood estimation. Although four classes were expected, an exploratory framework was utilized due to the lack of prior research in this area; therefore, models with 2, 3, 4, 5, and 6 classes were estimated. The model with the best fit and most interpretable solution was retained using Bayesian Information Criterion (BIC), sample-size adjusted BIC (aBIC), bootstrap likelihood ratio test (BLRT), and entropy, which is an index of classification precision, with higher values indicating better separation across the profiles. Adequate power to enumerate the correct number of profiles in LPA is based on many factors, such as the number of indicators, number of profiles, and degree of separation between the profiles.³⁶ There is no clear recommendation for calculating power in LPA, and doing so requires information about the population parameters.³⁷ Previous research suggests a minimum sample size ranging from 300 to 1000.³⁸ Our sample size meets this minimum threshold; however, is below the threshold recommended by some studies.³⁹

To investigate the differences among profiles identified in the LPA, participants were categorized into groups according to their most likely latent profile. LPA is probabilistic and classification of individuals occurs based on most likely group membership. Ten separate one-way ANOVAS with follow-up comparisons were used to examine differences among the latent profiles. A Bonferroni correction was applied (.05/10 = .005) to account for multiple comparisons.

Results

Descriptive Statistics

Overall, scores on the IES-2 and HEI and their subscales were similar to means reported in previous research in college samples.^15,26 See Table 2.

Table 2.

Descriptive Statistics for the Overall Sample (N = 352).

Measure	M (SD)	Minimum - Maximum
IES-2 Unconditional Permission to Eat Subscale	3.3 (0.9)	1.8–5
IES-2 Eating for Physical Rather than Emotional Reasons Subscale	3.4 (0.8)	1–5
IES-2 Reliance on Hunger and Satiety Cues Subscale	3.6 (0.7)	1–5
IES-2 Body-Food Choice Congruence Subscale	3.5 (0.9)	1.3–5
IES-2 Total	3.4 (0.5)	1.6–4.8
MHI – Psychological Distress	61.3 (18.2)	24–119
MHI – Psychological Well-Being	54.0 (12.5)	21–80
BSQ	96.1 (39.4)	34–200
PFS	2.3 (0.9)	1–5
DWCBC Healthy Strategies	3.0 (1.4)	0–4
DWCBC Unhealthy Strategies	1.3 (1.7)	0–9
EDE-Q	1.7 (1.3)	0–5.7
HEI Adequacy Subscale Score	31.0 (8.6)	12.6–58
HEI Moderation Subscale Score	20.4 (6.1)	6.8–37.8
HEI Total Score	51.4 (13.4)	21.4–88

Note: IES = Intuitive Eating Scale, MHI = Mental Health Inventory, BSQ = Body Shape Questionnaire, PFS = Power of Food Scale, DWCBC = Dieting and Weight Control Behavior Checklist, EDE-Q = Eating Disorder Examination Questionnaire, HEI = Healthy Eating Index.

Factor Analysis of the IES-2

A 4-factor, 23-item model of the IES-2¹⁵ was an acceptable fit in the data [χ2 (224) = 1057.70, p < .001; CFI = .923; TLI = .913; RMSEA = .101 (90% CI = .098-.111) p <.001; SRMR = .078]. See Table 3.

Table 3.

Standardized Factor Loadings of the Intuitive Eating Scale (IES-2) in College Women.

Factors/Items	Standardized Factor Loadings
Unconditional Permission to Eat (UPE)
IES01*	.58
IES03	.63
IES04*	.61
IES09*	.61
IES16	.76
IES17	.73
Eating for Physical Rather than Emotional Reasons (EPR)
IES02*	.86
IES05*	.80
IES10 *	.80
IES11*	.87
IES12	.72
IES13	.67
IES14	.80
IES15	.80
Reliance on Hunger and Satiety Cues (RHSC)
IES06	.78
IES07	.71
IES08	.83
IES21	.77
IES22	.87
IES23	.88
Body-Food Choice Congruence (BFCC)
IES18	.76
IES19	.97
IES20	.95

Note: * = item is reverse coded; All factor loadings significantly loaded onto the latent factors (p <.001).

Latent Profile Analysis

Results of the LPA analyses showed improvements in model fit (as indicated by a drop in BIC/aBIC and a significant BLRT) until the 6-profile solution (see Table 4), however, the degree of improvement slowed around the 4- and 5-profile solutions. Upon closer examination of these solutions, the 4-profile solution was selected because the model displayed reductions in BIC/aBIC, highest relative entropy, and ease of interpretation. The 4 profiles were interpreted as: 1) Dieting (prevalence = 28%) characterized by intermediate levels of IE and the highest nutritionally healthy eating scores, 2) Unhealthy Eating (prevalence = 13%) characterized by low IE and lower nutritionally healthy eating scores compared to all but one profile, 3) Intuitive Eating (prevalence = 13%) characterized by the highest IE scores and higher healthy eating scores relative to all but one profile, and 4) Non-Dieting (prevalence = 46%) characterized by intermediate IE scores and the lowest nutritionally healthy eating.

Table 4.

Fit Indices for Latent Profile Analysis.

Latent Classes	Log-likelihood	Parameters	BIC	Difference in BIC	aBIC	Difference in aBIC	Entropy	BLRT
1	−3879.27	12	7828.59	—	7790.52	—	1.00	—
2	−3753.53	19	7617.98	210.61	7557.71	232.81	.77	p < .001
3	−3707.96	26	7567.70	50.28	7485.22	72.49	.75	p < .001
4	−3675.90	33	7544.44	23.26	7439.76	45.46	.78	p < .001
5	−3651.72	40	7536.94	7.5	7410.05	29.71	.77	p < .001
6	−3631.42	47	7537.21	−.27	7388.11	21.94	.79	p < .001

Note: BIC = Bayesian Information Criteria, aBIC = adjusted Bayesian Information Criteria, BLRT = Bootstrap Likelihood Ratio Test.

Investigation of Differences Among the Four Profiles

A series of one-way ANOVAS with follow-up comparisons were conducted to investigate differences among the four profiles identified in the LPA (see summary of profile differences in Figures 1-2 and Table 5).

Figure 1.

Standardized differences among the four profiles emerging from latent profile analysis.

Figure 2.

Summary of differences across four profiles emerging from latent profile analysis.

Table 5.

One-Way ANOVAs Comparing the Four Profiles from the Latent Profile Analysis on Study Variables.

		Means
Scale	F	Non-Dieting (n = 161)	Unhealthy Eating (n = 41)	Intuitive Eating (n = 37)	Dieting (n = 104)
HEI – Total	189.29	41.6^a,b,c	47.3^b,c,d	58.9^a,c,d	65.4^a,b,d
IES-2 – Total	108.16	3.5^a,b,c	2.7^b,c,d	4.1^a,c,d	3.3^a,b,d
MHI – Psychological Distress	9.24	59.2^a	74.5^b,c,d	57.0^a	61.0^a
MHI – Psychological Well-Being	15.10	54.4^a	43.3^b,c,d	59.9^a	55.6^a
BMI	16.22	23.1^a	28.5^b,c,d	22.5^a	23.6^a
BSQ	19.51	87.7^a,c	128.5^b,c,d	75.2^a,c	104.2^a,b,d
PFS	6.30	2.2^a	2.7^b,c,d	2.0^a	2.3^a
DWCBC – Healthy Strategies	14.37	2.7^c	2.9^c	2.9^c	3.7^a,b,d
DWCBC – Unhealthy Strategies	5.59	1.1^a	2.1^b,d	0.8^a	1.6
EDE-Q	24.90	1.3^a,c	2.8^b,c,d	1.0^a,c	2.2^a,b,d

Note: HEI = Healthy Eating Index, IES = Intuitive Eating Scale, MHI = Mental Health Inventory, BMI = Body Mass Index, BSQ = Body Shape Questionnaire, PFS = Power of Food Scale, DWCBC = Dieting and Weight Control Behavior Checklist, EDE-Q = Eating Disorder Examination Questionnaire. For one-way ANOVAs, all ps <.001 and df=3. Post hoc tests were conducted using pairwise comparisons.

^aDiffers significantly from the Unhealthy Eating profile

^bDiffers significantly from the Intuitive Eating profile

^cDiffers significantly from the Dieting profile

^dDiffers significantly from the Non-Dieting profile

Healthy Eating Index (HEI)

There were significant differences on healthy eating between each profile, F (3, 339) = 189.29, p <.001. Pairwise comparisons revealed that the Dieting profile reported the highest level of nutritionally healthy eating, the Intuitive Eating profile was second highest, the Unhealthy Eating profile was third, and the Non-Dieting profile was lowest.

Intuitive Eating Scale (IES-2)

Similar to the HEI, there were significant differences on this measure between each profile, F (3, 339) = 108.16, p <.001. Pairwise comparisons showed that the Intuitive Eating profile had the highest IE scores, the Non-Dieting and Dieting profiles had intermediate IE scores, and the Unhealthy Eating profile had the lowest IE scores. Although statistically significantly different, the Non-Dieting and Dieting profiles had negligible mean score differences on the IES-2.

Body Mass Index (BMI)

There were significant overall BMI differences, F (3, 339) = 16.22, p <.001, with the Unhealthy Eating profile self-reporting a significantly higher BMI relative to the other profiles. The mean BMI for the Unhealthy Eating profile was in the overweight range. There were no significant differences on BMI between the other profiles.

Mental Health Inventory (MHI)

The overall ANOVA was significant for psychological well-being, F (3, 339) = 15.09, p <.001, and psychological distress, F (3, 339) = 9.24, p <.001, respectively. According to pairwise comparisons, the Unhealthy Eating profile reported significantly lower levels of psychological well-being and significantly higher levels of psychological distress relative to the other profiles. There were no significant differences on the MHI between the other profiles.

Body Shape Questionnaire (BSQ)

A one-way ANOVA revealed significant differences on the BSQ, F (3, 339) = 19.51, p <.001, Pairwise comparisons revealed that the Unhealthy Eating profile reported significantly higher body dissatisfaction relative to the other profiles. The Dieting profile reported the second highest level of body dissatisfaction, which was significantly different from the other profiles. The Non-Dieting and Intuitive Eating profiles had the lowest body dissatisfaction and did not significantly differ from one another.

Power of Food Scale (PFS)

Overall significant differences on the PFS, F (3, 339) = 6.30, p <.001, were driven by the Unhealthy Eating profile, which reported significantly higher hedonic hunger relative to the other profiles. There were no significant differences on hedonic hunger among the other profiles.

Dieting and Weight Control Behavior Checklist (DWCBC)

There were overall differences on the use of healthy, F (3, 339) = 14.37, p <.001, and unhealthy dieting strategies, F (3, 339) = 5.60, p =.001, respectively. Pairwise comparisons revealed that the Dieting profile reported more use of healthy weight control strategies relative to the other three profiles. The Unhealthy Eating profile reported significantly higher use of unhealthy weight control strategies relative to the Non-Dieting and Intuitive Eating profiles. The Dieting profile did not significantly differ from any profile on use of unhealthy weight control strategies.

Eating Disorder Examination Questionnaire (EDE-Q)

There were significant differences on a measure of eating pathology, F (3, 339) = 24.90, p =.001. Pairwise comparisons showed that the Non-Dieting and Intuitive Eating profiles scored significantly lower on the EDE-Q relative to the Unhealthy Eating and Dieting profiles. The Unhealthy Eating profile reported the highest level of disordered eating, followed by the Dieting profile.

Discussion

The current study used LPA to explore eating patterns in college women and found that both IE and nutritionally healthy eating were informative in capturing the eating patterns of young women. Four distinct profiles emerged: A Dieting profile, an Unhealthy Eating profile, an Intuitive Eating profile, and a Non-Dieting profile. These profiles differed from each other on measures of eating behaviors and psychological health.

The original four-factor structure of the IES-2 was an acceptable fit in our sample of diverse college women and mean scores on the IES-2 were similar to those reported in other college female samples.¹⁵ Consistent with past research in young adult samples, the level of nutritionally healthy eating reported by participants was low.^3,26 On average, participants reported eating less than the daily recommended values for fruit, vegetables, whole grains, and dairy, and reported eating unhealthy levels of fat, sodium, refined grains, and empty calories.

Synthesis of LPA Findings

The four distinct profiles that emerged from the LPA differed in their level of IE and nutritionally healthy eating, suggesting that both components may be important when considering the eating patterns of young women. These profiles also differed from one another on outcome variables, such as disordered eating, body dissatisfaction, BMI, psychological health, and dieting, thereby suggesting that these profiles have potential clinical utility. It is important to remember, however, that LPA produces profiles that are probabilistic; thus, each profile consists of individuals who most likely belong to that profile, but there is error in categorization of individuals.

Intuitive Eating and Non-Dieting Profiles

The Intuitive Eating and Non-Dieting profiles displayed the highest levels of IE and reported lower psychological distress, BMI, disordered eating, and use of fewer weight control strategies. This is consistent with literature showing that IE is inversely associated with negative body image, eating pathology, and low self-esteem.¹⁵ Interestingly, the Intuitive Eating profile reported nutritionally healthier eating practices compared to the Non-Dieting profile, which may forecast better long-term health outcomes and weight management.¹ Furthermore, these findings support research suggesting that IE approaches to weight management have beneficial outcomes across physical and mental health domains.⁴⁰

Although the Non-Dieting profile had high IE, they reported the lowest HEI scores, which is consistent with research indicating that college-age students engage in a variety of unhealthy dietary behaviors.⁴¹ Due to its poor nutritional health, this profile may be at risk for future overweight, although the average current BMI for this profile fell in the normal weight range.⁴² Thus, this profile is a potential target for healthy eating interventions given that dietary habits are believed to change as emerging adults enter college and may become relatively entrenched over time.⁴

Unhealthy Eating Profile

The Unhealthy Eating profile was categorized by low nutritionally healthy eating and low IE. Importantly, this profile scored low on all measures related to overall health and highest on all measures related to body dissatisfaction and disordered eating. The Unhealthy Eating profile also had the highest BMI, which was in the overweight range. This finding is consistent with research suggesting that individuals with higher BMIs report higher levels of body dissatisfaction and eating disorder symptoms.^29,43 Although the Non-Dieting and Unhealthy Eating profiles both reported low nutritional quality, higher levels of IE in the Non-Dieting profile may have been protective against physical and mental health concerns. Thus, focusing exclusively on nutritionally healthy eating in college samples may be insufficient; one’s food choices and one’s feelings/behaviors associated with eating explain variance in body dissatisfaction, disordered eating, and psychological health.

Dieting Profile

Individuals who most probably belonged to the Dieting profile reported using more weight loss strategies in the past year and reported higher body dissatisfaction and disordered eating. The fact that the Dieting profile was high on variables such as body dissatisfaction and disordered eating is consistent with the dual pathway model, which identifies body dissatisfaction and dieting as two important factors leading to disordered eating.⁴⁴

Implications

Together these findings suggest that both nutritionally healthy eating and IE are important in conceptualizing eating patterns in college women, and interventions might best be tailored based on eating patterns. For example, a recent study of an 8-week IE intervention found significant reductions in body dissatisfaction and disordered eating behavior.⁴⁵ Individuals in the Non-Dieting profile, who have poor diet quality, may benefit from healthy eating interventions that focus on nutrition, whereas individuals in the Unhealthy Eating or Dieting profiles, who scored low on IE, may benefit from mindfulness-based or acceptance-based interventions.^40,46 The Intuitive Eating profile, which scored high on nutrition and high on IE, had scores consistent with low disordered eating, low body dissatisfaction, and high psychological well-being. These findings may suggest that incorporating both IE and nutritionally healthy eating could be important for increasing psychological well-being in college students; future research should specifically test this notion.

Strengths, Limitations, and Future Directions

Strengths of the current study were the size and ethnic diversity of the sample, the high quality of dietary data collected, and the high retention rate (80%). Limitations included: 1) the cross-sectional nature of the study, 2) the probabilistic nature of LPA and the fact that the measure of entropy for the LPA presented here was slightly below the recommended threshold,³⁷ 3) the restriction of the sample to female college students, which may limit generalizability, 4) the potentially reactive nature of participants’ eating during the study, and 5) the self-report nature of the measures. In addition, there were several factors that may have impacted statistical power, such as the degree of separation between classes, the unbalanced group sizes, and number of latent profiles.³⁶ Although the sample size was above the minimum recommendations based on some studies, the current study was likely underpowered.³⁷ This is not uncommon in applied research using LPA; however, the findings from the current study should be interpreted within this context.⁴⁷ Furthermore, although the current study found no group differences in the probability of profile membership based on ethnicity (χ² (6) = 4.20, p = .65), we were unable to compare the LPA analyses in non-Hispanic and Hispanic women given the small sample size (n <200) of each group. Future research should explore both ethnic/racial group and gender differences in eating profiles and should conduct longitudinal and experimental studies to examine causal attributions.

Summary and Conclusions

Our findings suggest that it is possible to categorize college women according to patterns of IE and nutritionally healthy eating, and that both factors are important in conceptualizing eating patterns in emerging adults. Four distinct profiles emerged (Non-Dieting, Intuitive Eating, Dieting, and Unhealthy Eating); individuals falling in the Intuitive Eating profile displayed the greatest overall physical and psychological health. If these findings are replicated, teaching the principles of IE should be considered when developing healthy eating interventions for college women.

So What?

What is already known on this topic?

Research suggests that food’s nutritional content and the feelings and behaviors associated with eating contribute to overall health, yet these constructs are rarely considered simultaneously.

What does this article add?

The current study investigated healthy eating profiles in college women that included both nutritional quality and intuitive eating,(IE) with IE being an eating style that prioritizes physiological hunger/satiety cues and minimizes dieting and emotional eating. Four patterns of nutritionally healthy eating and IE emerged that were related to psychological and disordered eating outcomes: 1) Dieting (high nutritionally healthy eating + intermediate IE), 2) Unhealthy Eating (low nutritionally healthy eating + low IE), 3) Intuitive Eating (moderately high nutritionally healthy eating + high IE), and 4) Non-Dieting (low nutritionally healthy eating + intermediate IE). These profiles significantly differed from one another on measures of psychological distress, body mass index, and dieting strategies, with individuals within the Intuitive Eating profile displaying the greatest overall physical and psychological health.

What are the implications for health promotion or practice?

Intuitive eating and nutritionally healthy eating are distinct factors in conceptualizing the eating patterns of college women, and both should be considered when developing interventions.

Footnotes

Author Contributions

Belon designed the study, ran participants, analyzed data, and wrote a draft of the manuscript.

Serier ran analyses and assisted with drafts of the paper.

VanderJagt assisted with reviewing the literature and writing drafts.

Smith assisting with designing the study and writing drafts of the manuscript.

All authors have approved the final article.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the University of New Mexico’s Student Research Allocations Committee.

Ethical Approval

The study was approved by the University of New Mexico’s Institutional Review Board (Study # 649 855-2). Informed consent was obtained online.

ORCID iD

Jane Ellen Smith

References

Asghari

Mirmiran

Yuzbashian

Azizi

. A systematic review of diet quality indices in relation to obesity. Br J Nutr. 2017;117(8):1055-1065. doi:10.1017/S0007114517000915.

Schwingshackl

Bogensberger

Hoffmann

. Diet quality as assessed by the healthy eating index, alternate healthy eating index, dietary approaches to stop hypertension score, and health outcomes: An updated systematic review and meta-analysis of cohort studies. J Acad Nutr Diet. 2018;118(1):74-100. doi:10.1016/j.jand.2017.08.024.

Davy

Benes

Driskell

. Sex differences in dieting trends, eating habits, and nutrition beliefs of a group of midwestern college students. J Am Diet Assoc. 2006;106(10):1673-1677. doi:10.1016/j.jada.2006.07.017.

Nelson

Story

Larson

Neumark-Sztainer

Lytle

. Emerging adulthood and college-aged youth: an overlooked age for weight-related behavior change. Obesity. 2008;16(10):2205-2211. doi:10.1038/oby.2008.365.

Abraham

R Noriega

Shin

. College students eating habits and knowledge of nutritional requirements. J Hum Nutr Diet 2018;2(1):13-17. doi:10.35841/nutrition-human-health.2.1.13-17.

Kelly

Mazzeo

Bean

. Systematic review of dietary interventions with college students: Directions for future research and practice. J Nutr Educ Behav 2013;45(4):304-313.

Schaefer

. A review of interventions that promote eating by internal cues. J Acad Nutr Diet. 2014;114(5):734-760. doi:10.1016/j.jand.2013.12.024.

Lazarevich

Irigoyen Camacho

Velázquez-Alva

Zepeda

. Relationship among obesity, depression, and emotional eating in young adults. Appetite 2016;107:639-644. doi:10.1016/j.appet.2016.09.011.

Macht

Simons

. Emotional eating. In: Nyklíček

Vingerhoets

Zeelenberg

, eds. Emotional Regulation and Well-Being. New York, NY: Springer; 2011:281-295.

10.

Mason

Dunton

Gearhardt

Leventhal

. Emotional disorder symptoms, anhedonia, and negative urgency as predictors of hedonic hunger in adolescents. Eat Behav. 2020;36:101343. doi:10.1016/j.eatbeh.2019.101343.

11.

Mason

Smith

Lavender

Leventhal

. Longitudinal prospective association between hedonic hunger and unhealthy food and drink intake in adolescents. Int J Environ Res Publ Health. 2020;17(24):9375. doi:10.3390/ijerph17249375.

12.

Thompson

Romeo

. Gender and racial differences in emotional eating, food addiction symptoms, and body weight satisfaction among undergraduates. J Diabetes Obes 2015;2(2):93-98. doi:10.154326/2376-0494.15.035.

13.

Tylka

Wilcox

. Are intuitive eating and eating disorder symptomatology opposite poles of the same construct? J Counsel Psychol. 2006;53(4):474-485. doi:10.1037/0022-0167.53.4.474.

14.

Tylka

. Development and psychometric evaluation of a measure of intuitive eating. J Counsel Psychol. 2006;53(2):226-240. doi:10.1037/0022-0167.53.2.226.

15.

Tylka

Kroon Van Diest

. The intuitive eating scale-2: Item refinement and psychometric evaluation with college women and men. J Counsel Psychol. 2013;60(1):137-153. doi:10.1037/a0030893.

16.

Humphrey

Clifford

Morris

. Health at every size college course reduces dieting behaviors and improves intuitive eating, body esteem, and anti-fat attitudes. J Nutr Educ Behav 2015;47(4):354-360. doi:10.1016/j.jneb.2015.01.008.

17.

Lattimore

. Mindfulness-based emotional eating awareness training: taking the emotional out of eating. Eat Weight Disord. 2020;25(3):649-657. doi:10.1007/s40519-019-00667-y.

18.

Gibson

. Three multivariate models: Factor analysis, latent structure analysis, and latent profile analysis. Psychometrika. 1959;24:229-252. doi:10.1007/BF02289845.

19.

VanKim

Erickson

Eisenberg

Lust

Rosser

BRS

Laska

. College women's weight-related behavior profiles differ by sexual identity. Am J Health Behav. 2015;39(4):461-470. doi:10.5993/AJHB.39.4.2.

20.

VanKim

Erickson

Eisenberg

Lust

Rosser

BRS

Laska

. Differences in weight-related behavioral profiles by sexual orientation among college men. Am J Health Promot. 2016;30(8):623-633. doi:10.4278/ajhp.140714-QUAN-331.

21.

Johnson

Pratt

Wardle

. Dietary restraint and self-regulation in eating behavior. Int J Obes. 2012;36(5):665-674. doi:10.1038/ijo.2011.156.

22.

Peat

Peyerl

Muehlenkamp

. Body image and eating disorders in older adults: A review. J Gen Psychol. 2008;135(4):343-358. doi:10.3200/GENP.135.4.343-358.

23.

Striegel-Moore

Rosselli

Perrin

, et al. Gender difference in the prevalence of eating disorder symptoms. Int J Eat Disord. 2009;42(5):471-474. doi:10.1002/eat.20625.

24.

Subar

KirkpatrickMittl

SIB

Mittl

, et al. The Automated Self-administered 24-hour Dietary Recall (ASA24): A resource for researchers, clinicians, and educators from the national cancer institute. J Acad Nutr Diet 2012;112(8):1134-1137. doi:10.1016/j.jand.2012.04.016.

25.

Kennedy

Ohls

Carlson

Fleming

. The Healthy Eating Index: Design and applications. J Am Diet Assoc 2005;95(10):1103-1108. doi:10.1016/S0002-8223(95)00300-2.

26.

Guenther

Kirkpatrick

Reedy

, et al. The Healthy Eating Index-2010 is a valid and reliable measure of diet quality according to the 2010 dietary guidelines for Americans. J Nutr 2014;144(3):399-407. doi:10.3945/jn.113.183079.

27.

Cooper

Taylor

Cooper

Fairbum

. The development and validation of the Body Shape Questionnaire. Int J Eat Disord 1987;6(4):485-494. doi:10.1002/1098-108X(198707)6:4<485::AID-EAT2260060405>3.0.CO;2-.

28.

Rosen

Jones

Ramirez

Waxman

. Body Shape Questionnaire: Studies of validity and reliability. Int J Eat Disord 1996;20(3):315-319. doi:10.1002/(SICI)1098-108X(199611)20:3<315::AID-EAT11>3.0.CO;2-ZEAT2260060405>3.0.

29.

Neumark-Sztainer

Wall

Guo

Story

Haines

Eisenberg

. Obesity, disordered eating, and eating disorders in a longitudinal study of adolescents: How do dieters fare 5 years later? J Am Diet Assoc. 2006;106(4):559-568. doi:10.1016/j.jada.2006.01.003.

30.

Fairburn

. Cognitive Behavior Therapy and Eating Disorders. New York, NY: Guilford Press; 2008.

31.

Mond

Myers

Crosby

, , et al.et al Screening for eating disorders in primary care: EDE-Q versus SCOFF. Behav Res Ther. 2008;46(5):612-622. doi:10.1016/j.brat.2008.02.003.

32.

Veit

Ware

. The structure of psychological distress and well-being in general populations. J Consult Clin Psychol. 1983;51(5):730-742. doi:10.1037/0022-006X.51.5.730.

33.

Lowe

Butryn

Didie

, et al. The Power of Food Scale. A new measure of the psychological influence of the food environment. Appetite. 2009;53(1):114-118. doi:10.1016/j.appet.2009.05.016.

34.

Muthén

. Mplus User’s Guide. 8th ed. Los Angeles, CA: Muthén & Muthén; 2017.

35.

IBM Corp . IBM SPSS Statistics for Windows, Version 26.0. Armonk, NY: IBM Corp; 2019.

36.

Ferguson

G Moore

Hull

. Finding latent groups in observed data: A primer on latent profile analysis in Mplus for applied researchers. Int J Behav Dev. 2019;44:458-468. doi:10.1177/0165025419881721.

37.

Tein

J-Y

Coxe

Cham

. Statistical power to detect the correct number of classes in latent profile analysis. Struct Equ Model 2013;20(4):640-657. doi:10.1080/10705511.2013.824781.

38.

Nylund-Gibson

Choi

. Ten frequently asked questions about latent class analysis. Trans Issues Psychol Sci 2018;4:440-461. doi:10.1037/tps0000176.

39.

Nylund

Asparouhov

Muthén

. Deciding on the number of classes in latent class analysis and growth mixture modeling: A monte carlo simulation study. Struct Equ Model 2007;14:535-569. doi:10.1080/10705510701575396.

40.

Bacon

Stern

Van Loan

Keim

. Size acceptance and intuitive eating improve health for obese, female chronic dieters. J Am Diet Assoc. 2005;105(6):929-936. doi:10.1016/j.jada.2005.03.011.

41.

Niemeier

Raynor

Lloyd-Richardson

Rogers

Wing

. Fast food consumption and breakfast skipping: Predictors of weight gain from adolescence to adulthood in a nationally representative sample. J Adolesc Health 2006;39(6):842-849. doi:10.1016/j.jadohealth.2006.07.001.

42.

Aljadani

Patterson

Sibbritt

Collins

. Diet quality and weight change in adults over time: A systematic review of cohort studies. Curr Nutr Rep. 2015;4(1):88-101. doi:10.1007/s13668-014-0115-1.

43.

Sira

White

. Individual and familial correlates of body satisfaction in male and female college students. J Am Coll Health. 2010;58(6):507-514. doi:10.1080/07448481003621742.

44.

Stice

Agras

. Predicting onset and cessation of bulimic behaviors during adolescence: A longitudinal grouping analysis. Behav Ther. 1998;29(2):257-276. doi:10.1016/50005-7894(98)800006-3.

45.

Kristeller

Wolever

Sheets

. Mindfulness-based eating awareness training (MB-EAT) for binge eating: A randomized clinical trial. Mindfulness. 2014;5(3):282-297. doi:10.1007/s12671-012-0179-1.

46.

Burnette

Mazzeo

. An uncontrolled pilot feasibility trial of an intuitive eating intervention for college women with disordered eating delivered through group and guided self‐help modalities. Int J Eat Disord 2020;53(9):1405-1417. doi:10.1002/eat.23319.

47.

Spurk

Hirschi

Wang

Valero

Kauffeld

. Latent profile analysis: A review and “how to” guide of its application within vocational behavior research. J Vocat Behav 2020;120:103445. doi:10.1016/j.jvb.2020.103445.

What Is Healthy Eating? Exploring Profiles of Intuitive Eating and Nutritionally Healthy Eating in College Women

Abstract

Purpose

Design/Setting/Subjects

Measures

Analysis

Results

Conclusion

Keywords

Introduction

Background

Purpose

Materials and Methods

Design and Sample

Measures

Automated Self-Administered 24-hour Dietary Assessment Tool (ASA-24) 24

Body Shape Questionnaire (BSQ) 27

Demographics

Dieting and Weight Control Behavior Checklist (DWCBC) 29

Eating Disorder Examination-Questionnaire (EDE-Q) 30

Intuitive Eating Scale (IES-2) 15

Mental Health Inventory (MHI) 32

Power of Food Scale (PFS) 33

Analysis

Results

Descriptive Statistics

Factor Analysis of the IES-2

Latent Profile Analysis

Investigation of Differences Among the Four Profiles

Healthy Eating Index (HEI)

Intuitive Eating Scale (IES-2)

Body Mass Index (BMI)

Mental Health Inventory (MHI)

Body Shape Questionnaire (BSQ)

Power of Food Scale (PFS)

Dieting and Weight Control Behavior Checklist (DWCBC)

Eating Disorder Examination Questionnaire (EDE-Q)

Discussion

Synthesis of LPA Findings

Intuitive Eating and Non-Dieting Profiles

Unhealthy Eating Profile

Dieting Profile

Implications

Strengths, Limitations, and Future Directions

Summary and Conclusions

So What?

What is already known on this topic?

What does this article add?

What are the implications for health promotion or practice?

Footnotes

Author Contributions

Declaration of Conflicting Interests

Funding

Ethical Approval

ORCID iD

References

Automated Self-Administered 24-hour Dietary Assessment Tool (ASA-24)²⁴

Body Shape Questionnaire (BSQ)²⁷

Dieting and Weight Control Behavior Checklist (DWCBC)²⁹

Eating Disorder Examination-Questionnaire (EDE-Q)³⁰

Intuitive Eating Scale (IES-2)¹⁵

Mental Health Inventory (MHI)³²

Power of Food Scale (PFS)³³