Introduction

Breakfast is the first meal that is consumed in the day. It is also considered the most important meal of the day and provides essential nutrients for the day’s activities for children, adolescents, and adults [1,2,3]. Despite the benefits of eating breakfast, skipping breakfast has become prevalent among school-age children, adolescents, and working adults [4,5,6]. Eating breakfast every day is recommended in various health strategies including the United States “Healthy people 2030” (United States Department of Health and Human Services) and Japan’s “Health Japan 21 (the second term)” (Ministry of Welfare and Labor). However, to increase willingness to eat breakfast, it is important to understand factors that affect eating breakfast, rather than simply publicizing the beneficial effects of eating breakfast.

Previous studies identified several factors related to breakfast intake. These include chronotype-related lifestyle issues such as wake time [7,8,9,10], appetite for breakfast [11,12,13], personality traits [14, 15], and self-esteem [16]. Associations between attitude toward breakfast and the frequency of eating breakfast have also been reported. For example, preadolescent attitudes toward breakfast were inversely correlated with preadolescents’ skipping breakfast [17]. Adolescents’ perception of the importance of breakfast was a significant predictor of adolescents’ skipping breakfast [18]. In practice, despite most children perceiving breakfast as “very important,” 23% reported skipping breakfast [17]. In addition, male students skipped breakfast more than female students, and students who placed a low value on breakfast consumption were likely to skip breakfast [17]. Attitudes about eating are developed early and may remain fairly fixed over time [19, 20]. The above studies suggested that attitude toward breakfast is associated with eating breakfast. If attitudes about eating are developed early and last a lifetime, associations developed in childhood will be observed in young adults. However, attitudes toward eating breakfast in young adults have not been investigated.

Attitude is defined as “a mental and neural state of readiness, organized through experience, exerting a directive or dynamic influence upon an individual’s response to all objects and situations with which it is related” [21]. It is, therefore, possible that the formation of attitudes toward breakfast may be related to daily habits such as wake time or appetite for breakfast, and psychological aspects such as personality traits and self-esteem.

Previous studies have reported an association between appetite and breakfast intake. An endogenous circadian rhythm is characterized by low appetite in the morning and a peak in appetite in the evening [11]. Furthermore, fasting appetite in the morning was higher with early wake time compared with delayed bedtime or control conditions [12]. People whose energy intake was higher at breakfast were significantly more likely to take in less energy (kcal) from their afternoon and evening meals, and therefore across the whole day compared with those with a lower energy intake from breakfast [13]. These findings suggest that wake time has some influence on appetite for breakfast.

Overall, attitude toward breakfast is associated with frequency of eating breakfast, and there may be relationships between attitude toward breakfast and wake time or appetite for breakfast, and between wake time and appetite for breakfast. Therefore, we hypothesized that there is an indirect association between wake time or appetite for breakfast and the frequency of eating breakfast, which is mediated by attitude toward breakfast, in addition to the direct association between these factors. To confirm this hypothesis, it is necessary to develop and test a model of frequency of eating breakfast with multiple possible factors.

In addition, gender-based differences in skipping breakfast have been reported over 15 years in Japan [6] and outside of Japan [22, 23]. Personality traits and self-esteem also show gender-based differences [24, 25]. Therefore, we also hypothesized that a gender difference may exist in the model of frequency of eating breakfast with multiple possible factors.

In this study, we investigated the associations among attitude toward breakfast, appetite for breakfast, wake time, personality traits, self-esteem, and frequency of eating breakfast among university students. We also aimed to develop a model of frequency of eating breakfast with multiple possible factors, many of which may be correlated.

Methods

Participants

We conducted a survey during university classes in 2017. Data were collected using an anonymous, self-administered questionnaire. Participants were 661 university students who attended classes in the liberal arts. Participants did not receive any remuneration for completing the survey. The questionnaires were delivered to all students in attendance (661 students), and completed questionnaires were collected. Of the 661 questionnaires, 555 contained valid responses. The response rate, calculated by dividing the number of valid responses by the number of questionnaires delivered, was 84.0% (n = 555, 177 men and 378 women, mean age 19.2 ± 1.1 years).

All participants provided informed consent to participate, and the study was approved by the Human Ethics Committee of the Graduate School of Human Development and Environment, Kobe University.

Measures

We used a questionnaire to determine participants’ anthropometry, living with family, frequency of eating breakfast, appetite for breakfast, attitude toward breakfast, wake time, personality traits, and self-esteem. The anthropometric items included self-reported height and weight [24, 26], consistent with previous studies [27,28,29]. Each participant’s body mass index (BMI) was calculated by dividing their weight in kilograms by their height in meters squared (kg/m2). To evaluate living with family, we asked whether participants lived with their family.

To assess the frequency of eating breakfast, we asked participants how many times per week they ate breakfast. Appetite for breakfast was evaluated with the question, “Do you usually have an appetite when you eat breakfast?” Participants were asked to rate their appetite on a scale from 1 (“never”) to 4 (“always”). Attitudes toward eating breakfast were assessed using a 13-item questionnaire developed and validated by Tapper et al. [30]. This questionnaire covers several domains, including concentration and behavior, energy, and the general importance placed on breakfast. Participants were asked to rate each item from 1 (“disagree a lot”) to 5 (“agree a lot”). Responses to the items were summed, and then divided by the number of items to produce a score between 1 and 5. The Cronbach’s α coefficient for this questionnaire was 0.86 in this sample.

Personality traits were evaluated using the Ten Item Personality Inventory (TIPI) developed by Gosling et al. [31]. This 10-item measurement tool was developed as an improved version of the five-factor model of personality [32]. The TIPI measures five important personality traits: extraversion, agreeableness, conscientiousness, neuroticism, and openness. It has adequate convergent and discriminant validity, test-retest reliability, and patterns of external correlates [31].

Self-esteem was assessed using the Japanese version of the Self-Esteem Scale [33], which was translated from the original Rosenberg Self-Esteem Scale [34]. The Japanese version is a 10-item measure, like the original, but responses are on a five-point Likert-type scale from 1 (“disagree”) to 5 (“agree”). Responses to the items were summed to give the total score. The Cronbach’s α coefficient for this scale was 0.87 in our sample.

Statistical analyses

Student’s t-tests were used to evaluate differences between men and women. Pearson’s correlation coefficients or Spearman’s rank correlation coefficients were calculated for physical status, breakfast-related items, self-esteem, and personality traits. We used the method described by Baron and Kenny [35] for the multiple linear regression analysis to verify if attitude toward breakfast was intermediate variable. In the first step, this was performed with frequency of eating breakfast as the dependent variable and other indicators (except attitude toward breakfast) as the independent variables. In the second step, attitude toward breakfast was the dependent variable and other indicators were independent variables. In the final step, we used frequency of eating breakfast as the dependent variable and other indicators (including attitude toward breakfast) as independent variables. In addition, multiple linear regression analysis was used to investigate associations between appetite for breakfast or wake time and psychological factors. Effect sizes for the multiple regression analysis were reported as Cohen’s f2 and interpreted according to Cohen’s recommendation of 0.02 for a small effect, 0.15 for a medium effect, and 0.35 for a large effect [36]. Structural equation modeling (SEM) was used to explore the association between frequency of eating breakfast and related factors. The fit indices used to evaluate the model were the ratio of the chi-square value to the degrees of freedom (χ2/df) < 2.0 [37], standardized root mean square residual (SRMR) < 0.08, root mean square error of approximation (RMSEA) including the 90% confidence interval < 0.05 (values between 0.05 and 0.08 were considered acceptable), and the comparative fit index (CFI) and Tucker-Lewis index (TLI) ≥ 0.90 [38, 39]. We also calculated Akaike’s information criterion (AIC) to estimate the relative quality of model. The level of statistical significance was set at 0.05. All statistical analyses were performed with SPSS® 27.0 for Windows (International Business Machines Corp, Armonk, NY) and Amos 27.0 for Windows (International Business Machines Corp, Armonk, NY).

Results

Frequency of eating breakfast, attitude toward breakfast, and appetite for breakfast were significantly higher in women than men. Compared with men, women had higher extraversion and neuroticism, and lower openness to experience. Self-esteem was higher in men than women (Table 1).

Table 1 Anthropometry, frequency, appetite, and attitude of breakfast, and psychometric indicators by gender
Full size table

Pearson’s correlation coefficients or Spearman’s rank correlation coefficients for frequency of eating breakfast, attitude toward breakfast, appetite for breakfast, wake time, BMI, personality traits, and self-esteem are shown in Table 2. For both men and women, frequency of eating breakfast, attitude toward breakfast, and appetite for breakfast were mutually significantly positively correlated with each other. Wake time was significantly inversely correlated with frequency of eating breakfast and attitude toward breakfast in both men and women, and with appetite for breakfast in men. In women, attitude toward breakfast was significantly positively correlated with extraversion, agreeableness, and conscientiousness.

Table 2 Pearson’s correlation coefficients between frequency of eating breakfast and attitude toward breakfast, appetite for breakfast, wake time, BMI, personality traits, and self-esteem
Full size table

The results of simple regression analysis is shown in Table 3, and those for multiple regression analyses in Tables 4 and 5. Multiple linear regression analysis was conducted in three steps as described by Baron and Kenny [35] to verify if attitude toward breakfast was considered as an intermediate variable (Table 4). The multiple linear regression results for Steps 1, 2, and 3 are shown in Table 4.

Table 3 Simple regression analysis with frequency of eating breakfast or attitude toward breakfast as the dependent variable
Full size table
Table 4 Multiple regression analysis with frequency of eating breakfast or attitude toward breakfast as the dependent variable
Full size table
Table 5 Multiple regression analysis with appetite or wake-up time as the dependent variable
Full size table

In Step 1, the multiple regression for men showed that appetite for breakfast was significantly positively associated and wake time and self-esteem were significantly inversely associated, with frequency of eating breakfast. For women, appetite for breakfast was significantly positively associated, and wake time was significantly inversely associated with frequency of eating breakfast.

In Step 2, the multiple regression among men showed that appetite for breakfast was significantly positively associated, and wake time was significantly inversely associated with attitude toward breakfast. Among women, appetite for breakfast, conscientiousness, and self-esteem were significantly positively associated, and wake time was significantly inversely associated with attitude toward breakfast.

In Step 3, appetite for breakfast and attitude toward breakfast were significantly positively associated, and wake time and self-esteem were significantly inversely associated with frequency of eating breakfast among men. Among women, appetite for breakfast and attitude toward breakfast were significantly positively associated, and wake time was significantly inversely associated with frequency of eating breakfast.

In addition, the multiple regression showed that wake time was significantly inversely associated with appetite for breakfast in men. Conscientiousness and neuroticism were significantly inversely associated with wake time among women (Table 5).

Figure 1 shows the results of the SEM and the relationships among frequency of eating breakfast, attitude toward breakfast, appetite for breakfast, wake time, self-esteem, and conscientiousness. The model was finalized after inspection of the modification indices and allowing the unique variance of self-esteem and conscientiousness to correlate. In men, appetite for breakfast was significantly positively associated (p < 0.001) and wake time was significantly inversely associated (p = 0.004) with attitude toward breakfast. Appetite for breakfast and attitude toward breakfast were both significantly positively associated with frequency of eating breakfast (appetite for breakfast, p < 0.001; attitude toward breakfast, p < 0.001). Wake time and self-esteem were both significantly inversely associated with frequency of eating breakfast (wake time, p = 0.002; self-esteem, p < 0.001). The model showed a good fit: χ2 = 1.658, df = 4, χ2/df = 0.415; SRMR = 0.028; GFI = 0.997, AGFI = 0.984, TLI = 1.032, CFI = 1.000, and RMSEA = 0.000. The AIC was 35.658. In women, self-esteem, conscientiousness, and appetite for breakfast were significantly positively associated with attitude toward breakfast (self-esteem, p = 0.011; conscientiousness, p = 0.020; appetite for breakfast, p < 0.001). Wake time was significantly inversely associated with attitude toward breakfast (p < 0.001). In addition, conscientiousness was significantly inversely associated with wake time (p = 0.023). Appetite for breakfast and attitude toward breakfast were both significantly positively associated with frequency of eating breakfast (appetite for breakfast, p < 0.001; attitude toward breakfast, p < 0.001). Finally, wake time was significantly inversely associated with frequency of eating breakfast (p = 0.003). The model showed a good fit: χ2 = 6.339, df = 4, χ2/df = 1.585; SRMR = 0.036; GFI = 0.994, AGFI = 0.971, TLI = 0.980, CFI = 0.995, and RMSEA = 0.039. The AIC was 40.339.

Fig. 1
figure 1

Associations between frequency of eating breakfast and wake time, appetite for breakfast, psychological conditions and attitude toward breakfast. Upper number is standardized coefficient for males and lower number is for females, and e1–e6 stand for errors in each item. Frequency: frequency of eating breakfast, Appetite: appetite for breakfast, Attitude: attitude toward breakfast. *p < 0.05

Full size image

Next, we removed the insignificant associations for each gender and determined the saturated models. The model for men (Fig. 2a) showed a good fit: χ2 = 3.287, df = 3, χ2/df = 1.096; SRMR = 0.025; GFI = 0.993, AGFI = 0.963, TLI = 0.996, CFI = 0.999, and RMSEA = 0.023. The AIC was 22.287. The model for women (Fig. 2b) also had a good fit: χ2 = 9.058, df = 6, χ2/df = 1.510; SRMR = 0.043; GFI = 0.992, AGFI = 0.973, TLI = 0.982, CFI = 0.993, and RMSEA = 0.037. The AIC was 39.058.

Fig. 2
figure 2

Associations between frequency of eating breakfast and wake time, appetite for breakfast, attitude toward breakfast, and self-esteem in men (a), and in women (b). The number is standardized coefficient and e1–e6 stand for errors in each item. Frequency: frequency of eating breakfast, Appetite: appetite for breakfast, Attitude: attitude toward breakfast. *p < 0.05

Full size image

Discussion

In this study, we attempted to clarify the associations between frequency of eating breakfast, attitude toward breakfast, appetite for breakfast, wake time, and psychometric factors in young Japanese adults. We also developed a model including different factors affecting frequency of eating breakfast.

All anthropometric values varied between men and women. Frequency of eating breakfast, appetite for breakfast, attitude toward breakfast, some personality traits, and self-esteem also varied significantly between men and women. These results showed differences between men and women in anthropometric indicators, breakfast-related indictors, and psychological indicators. Differences in anthropometric factors have previously been reported [24, 26], and the differences observed in this study were consistent with the previous results. The differences between men and women in neuroticism and openness as personality traits were also similar to previous results [24, 25].

We found that compared with men, women were more likely to eat breakfast more frequently. Similarly, previous results from Japan for 2003–2017 showed that fewer women skipped breakfast than men in each year [6]. Drewnowski et al. also found that women over age 18 years ate breakfast more frequently than men of the same age [22]. Keski-Rahkonen et al. reported similar findings in the United States [23]. These results were consistent with our findings.

We verified the relationships among frequency of eating breakfast, attitude toward breakfast, appetite for breakfast, lifestyle factors, and psychological indicators. Many of the indicators showed significant gender-based differences, so we separated the subsequent analyses by gender. Next, we confirmed the relationships between frequency of eating breakfast, attitude toward breakfast, appetite for breakfast, BMI, wake time, personality traits, and self-esteem by gender using Pearson’s correlation coefficients or Spearman’s rank correlation coefficients, followed by multiple linear regression analysis. In the multiple linear regression, the dependent variable was frequency of eating breakfast, and the other indicators were independent variables. Attitude toward breakfast was also used as an independent variable for the dependent variable “frequency of eating breakfast.” As attitude may be influenced by self-esteem, personality traits, wake time, and appetite for breakfast, we considered attitude toward breakfast as an intermediate variable. Therefore, we used the method described by Baron and Kenny [35] for the multiple linear regression analysis. In the first step, wake time, self-esteem, and appetite for breakfast were significant independent variables when frequency of eating breakfast was the dependent variable. In the second step, wake time, conscientiousness, self-esteem, and appetite for breakfast were significant independent variables when attitude toward breakfast was the independent variable. In the final step, wake time, self-esteem, appetite for breakfast, and attitude toward breakfast were significant independent variables when frequency of eating breakfast was the dependent variable. The standard regression coefficients for wake time and appetite for breakfast were much lower in the final step than the first. This suggested that wake time and appetite for breakfast were influenced by attitude toward breakfast as an intermediate variable. In addition, the multiple regression analysis in these steps showed large effect sizes, indicating there were firm associations among the frequency of eating breakfast, attitude toward breakfast, and appetite for breakfast. We also verified the associations between appetite for breakfast and wake time, personality traits, and self-esteem, and between wake time and personality traits or self-esteem using multiple regression analysis. There were definite associations between appetite for breakfast and wake time in men, and between wake time and conscientiousness in women.

Based on the results of the three-step multiple linear regression analysis, we proposed a model of frequency of eating breakfast, attitude toward breakfast, and related items. The diagram fit both men and women, but the model showed some gender-based differences. For example, the standardized estimate from attitude toward breakfast to frequency of eating breakfast was higher in women than men, and self-esteem and conscientiousness were only associated with attitude toward breakfast in women. In contrast, attitude toward breakfast was not influenced by self-esteem or conscientiousness in men. Personality is defined as dimensions of individual differences in tendencies that show consistent patterns of thoughts, feelings, and actions [40, 41]. Conner and Abraham [42] reported that conscientiousness was an independent predictor of behavior, and was also positively associated with attitude and intention. O’Connor et al. [43] also reported that conscientiousness and its facets may influence health status directly via changes in health behaviors and indirectly by influencing stress-health behavior relations. Frequency of eating breakfast has been positively correlated with conscientiousness [15]. However, intentions related to weight maintenance and weight loss or dieting in later adolescence and young adulthood were associated with low self-esteem [44,45,46]. Another study found that low self-esteem was associated with the consumption of an unhealthy breakfast, but not a healthy breakfast [16]. These studies suggested that both conscientiousness and self-esteem were associated with intention and actual health behaviors, which supported our findings of the relationship between self-esteem, conscientiousness, and attitude toward breakfast.

The reason for differences between men and women in the relationship between conscientiousness or self-esteem and attitude toward breakfast remain unclear. Among fifth and sixth graders in the United Kingdom, boys had a more positive attitude than girls toward breakfast [30]. However, no studies have examined gender-based differences in attitudes toward breakfast in young adults. A study among Chinese university students showed knowledge was positively associated with attitude, and attitude was positively associated with behavior [47]. However, the relationship between knowledge and attitude was stronger among women in that group [47]. This may be a factor underlying the differences between men and women in our study.

Our study had several limitations. First, the sample was collected from a restricted geographic area, and all participants were Japanese students. Height and body weight were self-reported and not measured objectively, which could have led to a misclassification of BMI because of information bias. However, the National Health and Nutrition Survey found mean BMIs of 21.1 kg/m2 in men and 20.7 kg/m2 in women aged 15–19 years, and 23.1 kg/m2 and 20.6 kg/m2 in those aged 20–29 years [6]. The BMIs for men and women in our study were similar (men, 21.3 kg/m2; women, 20.5 kg/m2). Second, the present study used a cross-sectional design. Therefore, we did not consider temporal relationships between factors, so it is not possible to directly prove causal relationships. However, intermediate factors presuppose a causal relationship. Therefore, a limitation of cross-sectional research is that whether a factor is an intermediate factor must be theoretically estimated from previous literature. Third, we did not collect data on the role of family, which may have an influence on breakfast intake in young adults. Despite these limitations, this is the first study to show the associations among attitude toward breakfast, appetite for breakfast, wake time, and psychometric factors in young adults, and to develop a model for breakfast intake with multiple possible factors.

Conclusions

This study clarified the relationships among factors related to eating breakfast and psychometric factors in Japanese university students. Our results suggest that attitude toward breakfast is directly related to the frequency of eating breakfast, and also plays a role as an intermediate factor in the frequency of eating breakfast. In women, self-esteem and conscientiousness are associated with the frequency of eating breakfast through attitude toward breakfast. In men, increased appetite for breakfast is important for attitude toward breakfast and increased frequency of eating breakfast, and early wake-up time is important for an increased appetite for breakfast. This indicates that men and women have different approaches to increasing the frequency of eating breakfast. In men, appetite for breakfast or wake-up time should be a focus to increase the frequency of eating breakfast; whereas for women, it is important to focus on the formation of the attitude toward breakfast and contributing psychological aspects.

What is already known on this subject?

Previous studies reported that attitude toward breakfast is associated with frequency of eating breakfast and there may be a relationship between attitude toward breakfast and wake time or appetite for breakfast, and between wake time and appetite for breakfast. These studies suggest that attitude toward breakfast is associated with eating breakfast, but no studies have investigated this factor in young adults. If attitudes about eating are developed early and last a lifetime, associations developed in childhood will be observed in young adults.

What does this study add?

This study shows that the frequency of eating breakfast is associated with attitude toward breakfast and mediated by appetite for breakfast and wake time. Some gender differences in factors influencing attitude toward breakfast were identified.