INTRODUCTION

The Night Eating Syndrome (NCS) was, in the 1950s, characterized by a triad of symptoms - absence of hunger in the morning, hyperphagia at night, insomnia - that occur associated with specific stressful events¹. Years later, anxiety, depressed mood that worsens at night, and waking up at night to eat were added to the syndrome’s components². In physiological terms, there is a mismatch in the circadian rhythm of secretion of several hormones involved in appetite and sleep control in NES³. This imbalance results in difficulty sleeping and an abnormal eating pattern, with excessive food intake after dinner hours - 25% or more of total daily calories. As a result of this inadequate diet, many scholars investigate whether there is an association between NES and increased Body Mass Index (BMI)⁴, but the results in the literature are still not a consensus⁵.

SCN carriers report that the appearance of the first symptoms of this syndrome occurred mainly between late adolescence and early adulthood^6,7. This study focuses on the critical period in the lives of college adults, whose lifestyle changes such as decreasing physical activity and adopting unhealthy eating habits can, in addition to result in increased body fat^8,9, persist throughout life¹⁰, with great harm to health.

In this context, Night Eating Syndrome stands out as an eating disorder that can affect university students. The proportion of students who meet the screening criteria for the syndrome varies between 1.2 and 15% in studies from different countries including Brazil^11-23, with a higher prevalence among obese individuals²⁴. It is known that obese individuals at 25 years of age are more likely to develop, ten years later, severe forms of obesity and metabolic alterations related to chronic diseases²⁵.

Most university students are in an age group and, therefore, in a stage of life that makes them more vulnerable to NES. In general, young adults (18 to 30 years old) have the highest food consumption after 11 pm compared to other ages²⁶. Psychological factors may also play a role in the outcome of this eating disorder. Students face psychological distress due to the difficulties of transitioning to adulthood and the trials of everyday academic life²⁷. It is known that emotional changes are important components of the set of symptoms of NES²⁸, and this association with the syndrome was confirmed among university students^17,18,29,30. Finally, it is common to find sleep disorders in this population³¹, another pillar of the syndrome. Thus, there is a concern to identify and characterize the unhealthy behaviors of university students in order to plan specific strategies for health promotion and the prevention of serious diseases associated with excess weight⁹.

Some additional factors that could interfere with eating, sleeping and the emotional state of students, and therefore with the probability of developing NES, have been less explored in the literature: class schedule, simultaneous employment and the arrangement housing for example. The activity of students at night, either for classes or for work; the double journey of study and work, with accumulation of responsibilities related to academic and professional performance; and moving away from the parents’ home, which could represent a loss of reference to the family pattern and a favorable situation for the disorganization of the daily routine, are variables that the authors intend to investigate in this research.

Thus, the present study aimed to analyze aspects of university students’ academic life, work and housing that could be associated with Night Eating Syndrome.

METHODS

RESULTS

Nine hundred university students from four courses offered by the analyzed higher education institution were evaluated: Architecture, Engineering, Medicine and Psychology (table 1). The percentage distribution of participating students among courses, academic semester of the course and academic year of the course approximately reflected the proportion of those enrolled at the beginning of data collection. The course with the highest adherence to the survey was Psychology with 75.7% (215 among 284 enrolled) and the course with the lowest proportion was Engineering with 53.2% (177 among 333 enrolled). In relation to the years and periods of the four courses, the percentages of adherence varied between 50 and 75% in relation to the total number of enrolled students.

Participants were mostly female (53.7%) and the median age was 22 years. As for the Body Mass Index (BMI), the median found was 23.5 kg/m². The median value of the NEQ score for all subjects was 19. Of the 900 participants, 151 had an NEQ score greater than or equal to 25, resulting in a prevalence of 16.8% of NES. Sixty students (6.67%) achieved a NEQ score equal to or greater than 30. As for the job variable, 31.7% of the students reported having a job. Among these students, the majority worked for 4 to 8 hours a day (54.4%) and in the morning and afternoon shift (44.6%). Regarding housing, 51.3% of university students reported that they did not live with their parents and of these, 46.3% said they lived with friends.

The NEQ score and the prevalence of NES (by the NEQ≥25 criterion) were analyzed according to the study variables (Table 2). Among students with a job, it was noted that the prevalence of NES was higher for those who work in the afternoon and lower for those who work at night (p=0.031). No statistically significant association was found between the prevalence of the syndrome for the other variables. Regarding the NEQ score of college students, there was no statistically significant association. It is noteworthy, however, that NEQ score comparisons for some variables were close to the limit of significance: women tended to score higher than men (p=0.053); there was a trend towards differences in scores between courses (p=0.078), working students tended towards lower scores (p=0.086). These trends motivated the search for associations between subgroups of the sample.

Comparisons of the prevalence of NES between men and women were performed according to some of the study variables (Table 3). In the Engineering course, the proportion of students reaching the SNC criterion was higher for females (p=0.007). In the Psychology course, the opposite occurred: the proportion of students reaching the SNC criterion was higher for males (p=0.035). It is noteworthy that in the Engineering course there was a predominance of enrolled men (76.3%) and in the Psychology course there was a predominance of women enrolled (79.1%). In the other two courses, the number of women enrolled was higher (50.8% in Architecture and 55.2% in Medicine), but the distribution of enrollments by gender was more balanced. No statistically significant association was found when comparing the prevalence of NES between men and women for the other variables surveyed (period of study, work, period of work and whether or not they live at their parents’ house).

The prevalence of the syndrome in students who live with their parents and those who do not live with their parents was also compared according to the study variables (table 4). It was possible to observe that the differences in the prevalence of NES for students who work in the afternoon and at night that were verified in the complete universe of the study are maintained for the subgroup that lives with their parents (p=0.026), but do not have statistical significance for the subgroup of students who do not live with their parents. No other significant differences were identified for this type of comparison.

DISCUSSION

In the higher education institution analyzed, the proportion of students reaching the SCN screening criteria was 16.8%, or approximately 1 in every 6 participants. This proportion stands out as the highest prevalence among findings in the scientific literature for university student populations in different countries, whose data were collected before 2020^11-23. Some of these studies exclude students who show evidence of binge eating disorders (eg¹²), reducing the prevalence rate. In the present study, other eating disorders were not investigated.

Data available in the literature suggest that the prevalence of NES in university students may be higher in some countries such as Turkey, Saudi Arabia, Malaysia, Brazil and China, where values of 9.5% to 15.0% were verified^15,17,20-22. The result found in this study is closer to this variation range. In the other countries, the prevalence of SCN among university students varied at a lower level, between 1.2 and 5.8%^11-14,20. It is possible to speculate that cultural, socioeconomic and educational system differences explain the disparities between countries in relation to the estimated prevalence of SCN in students. Another hypothesis to be considered would be some level of variation in the results obtained by the NEQ versions in different languages.

Among the participating students, there was a trend towards a higher NEQ score for women. When investigating this question in more detail, differences in the prevalence of SCN between genders were found that were related to the course in which the student was enrolled. In the Engineering course, the prevalence of SCN was higher among women, consistent with the trend towards higher NEQ scores for this gender. However, unexpectedly, in the Psychology course it was found that men had a higher prevalence of SCN. This point contrasts, in the two courses, the number of students enrolled in each of them, the Engineering course being mostly male, and the Psychology course mostly female. The higher prevalence of the syndrome in both cases could be associated with the discomfort of belonging to a minority. Despite society’s evolution towards reducing sexual segregation in the social and professional field, in higher education some areas still persist as predominantly female or male^40,41. The collision between the contemporary ideal of equality between the sexes and the realization of the existence of environments dominated by one of the sexes at the university could generate a particularly stressful conflict for the minority group.

Two hundred and eighty five working students were identified, corresponding to 31.7% of the participants. Worldwide, a relevant fraction of higher education students perform paid work during graduation^42-50. This university student’s work usually aims to pay for his personal expenses, his housing and his studies. The possible implications of working while attending classes are discussed at length. The daily journey extended by the sum of periods of study and work can interfere with health. Research shows that college students who work have worse sleep quality than their peers who don’t study, with an increased risk of sleep deprivation and sleep disturbances^45,46,48,49. Increased fatigue and reduced physical activity are also noted^46,48. In addition, several articles point to impaired academic performance of working students^42-44,47,50.

In this research, working students presented a distribution of NEQ scores that were close to statistically different from non-working students. However, the trend found was a lower NEQ score for working students. A similar result had been observed by our group in another sample of Brazilian students¹⁷. The eventual reduction of the NEQ score among working students, if confirmed, would contest the suspicion that working simultaneously with the study could favor the NES. Such preliminary evidence can be interpreted as indications that there is a more complex relationship between student work and signs of SCN. Aspects of the students’ work that reinforce the behaviors of the syndrome would coexist with aspects of opposite effect, attenuating the syndrome. The first step to verify the adequacy of this hypothesis would be to identify interfering cofactors.

It was found that the students’ living arrangements and working hours could be some of these cofactors. Among students who live with their parents, the prevalence of SCN was higher among those who worked in the afternoon and lower among students who worked at night. The implications of housing and work hours on SCN that were found are intriguing. Based on the related literature, it would be reasonable to expect the opposite: a higher prevalence of the syndrome among students who work at night and that this effect would be more pronounced among those who do not live at home. Regarding the workday, when it goes beyond bedtime, it can cause overweight and obesity due to changes in eating patterns^51,52. Regarding housing, when students leave the family home, their eating habits become less healthy^53,54. The results of this study point to relationships between work, housing and SCN among university students that apparently do not fit into a simple model. New studies will be needed to better clarify these relationships.

The present study presented as main limitations: the use of information obtained from self-administered questionnaires that may have intrinsic biases in the understanding of the questions and partiality in the answers of the interviewees; the lack of investigation of other eating disorders; the inclusion of participants from a single higher education institution; and the cross-sectional design that does not allow establishing causal relationships between the studied phenomena. Finally, the cultural and social context of Brazil, as well as its structure of higher education, present peculiar characteristics, and it is not possible to extrapolate a priori the interpretation of the results of this research to students from other countries.

In conclusion, the prevalence of NES found among Brazilian university students was high (16.8%), particularly in two situations: (1) being enrolled in an undergraduate course with a predominance of students of the other sex; and (2) live with parents and work in the afternoon. These observations may be helpful in identifying subpopulations of students at increased risk for eating disorders.