Introduction

EDs are characterized by objective disturbances in eating patterns [1]. Both BN and BED include binge eating as a primary symptom, typically defined as the consumption of a large amount of food in a short period of time accompanied by a sense of lack of control over eating [2]. Binge eating episodes and behavior in both BN and BED (BE) often occur during the afternoon and evening hours [3]. People with binge eating report extreme difficulties in control and regulation of their eating; sometimes they even report dissociation symptoms during eating and lack of sense of what and how much they eat during a binge. The current study compares between people who have binges during the day and those who have binges at night.

Dietary patterns in patients with BN and BED

Dietary patterns, energy consumption, frequency of binge episodes and macronutrients were evaluated in patients with EDs. In patients with BN, objective food consumption in a laboratory setting ranged from 7101 to 9360 kcal per 24 h and from 3030 to 4479 kcal per binge episode [4]. However, subjective food records showed total energy consumption to range from 3117 to 4275 kcal per day [5] and from 1173 to 2415 kcal per binge episode [6]. The frequency of binge episodes among individuals with BN ranged from 5.7 to 10.9 episodes per week [6].

In terms of macronutrients, the fat content is higher in binge episodes, of patients with BN, than in non-binge meals, while carbohydrate and protein intakes do not differ. There is significant difference in the calories consumed by patients with BN during binge episodes compared to non-binge meals [5]. Likewise, no difference has been found between BN and controls in the percentage of daily energy consumed as carbohydrates, but the percentage of energy from fat has been found to be significantly higher, with percentage of energy from protein significantly lower among BN [7].

With regard to energy consumption, individuals with BED consumed more energy than did individuals without BED when both binge and non-binge days were combined [8]. Yet, no difference was identified between non-binge days and the average intake of controls [9].

The research indicates that the frequency of binge episodes among BED ranges from 10.7 to 17 episodes per 28 days (according to the Eating Disorders Examination Questionnaire—EDE-Q) [9, 10]. Regarding macronutrients, individuals with BED consume a higher proportion of energy from fat and a lower proportion of protein on binge days than on non-binge days and women with BED consumed less protein and more fat than controls with obesity [9, 11]. Given these findings, it is not entirely clear whether there are differences in macronutrient intake between women with obesity with and without BED.

Dietary patterns in individuals with NES

Night eating syndrome (NES) is conceptualized as a delay in the circadian pattern of food intake [12], manifested by two core criteria: evening hyperphagia (EH), defined as the consumption of large amounts of total caloric intake after the evening meal, and/or nocturnal ingestions (NI), defined as ingestion of food after sleep onset [13]. It is not entirely clear whether NES represents a distinct entity of eating disorder (ED) as proposed in the DSM-5 (2) or whether it is a variant of bulimia nervosa (BN) or binge eating disorder (BED) in which the binge episodes occur mainly during the evening [14].

Despite being part of the core criteria for NES, dietary patterns of individuals with NES have received relatively little research attention. In accordance with the core criteria for NES, studies revealed that the proportion of food intake during the evening and night differs significantly between individuals with and without NES.

Morning anorexia

Most of the studies reported that breakfast skipping pattern was significantly higher among individuals with NES than controls [15].

Consumption

Individuals with NES tend to consume significantly more energy in the evening (after evening meal) compared with controls [16, 17].

Moreover, some studies have found that individuals with NES consumed significantly more total calories than controls, while others found no differences between groups [17, 18]. In addition, it was found that individuals with NES tend to consume significantly more calories at night than controls [19].

Macronutrients

The carbohydrate percentage of total night eating energy intake among individuals with NES has been found to be significantly higher compared to the rest of the day [20, 21]. In contrast, another study indicated that patients with NES consumed fewer carbohydrates and more fat during the night, compared with controls [19]. Percentage intake of protein from daily consumption was shown to be lower among individuals with NES compared to healthy controls [15]. Other studies have found no significant differences in the percentage of calories from macronutrients between individuals with and without NES [18, 21].

Individuals with NES have been found to have more binge eating episodes over a 24-h period compared to controls [20]. Furthermore, individuals with NES reported significantly greater frequency of binge eating episodes during the past 28 days than those without NES [22]. In addition, individuals with NES reported more NI per week [23] and per night [16] than those without NES.

Thus, it remains unclear whether the timing and frequency of binge episodes, the amount of calories consumed and the macronutrient composition of those with binge eating behavior (both BED or BN) with NES are similar to those without NES. Moreover, to the best of our knowledge, studies to date have not yet examined whether NES (both EH and NI episodes) contributes to an overall elevated total caloric intake among patients with BN and BED.

The current study attempts to shed light on these questions. Thus, the aims of the current study were to compare the dietary patterns, number of calories consumed and macronutrient composition of patients with binge eating disorders (both BED or BN) occurring during the day or mainly occurring at night time.

Materials and methods

Participants

The sample consisted of 68 women aged 18–60 years referred to the Eating Disorders Institution at Rambam Medical Center, Israel, for ED assessment during the period of 2011–2013. The study sample was recruited upon admission, prior to treatment. Male patients and any female patients with sleep-related eating disorders (SRED), AN, or other severe psychiatric illnesses were excluded from the study. Nine patients were either excluded or dropped out of the study (non-responding). Thus, 59 women are included in the statistics (responding). No significant differences were found on any demographic and clinical characteristics between the responding and non-responding participants.

Participants were divided into two groups based on night binges and ED diagnoses assigned through a full clinical psychiatric evaluation conducted at intake to the ED institution (night binge and BE-only). The night binge group (n = 25; 44.4%) included patients with night eating that fulfilled the criteria for NES (NI and/or EH) [24] as well as an ED diagnosis that included BE disorders (both BN or BED) based on the DSM-IV (25). The BE group (n = 34; 56.6%) included patients diagnosed with BN (n = 12) or BED (n = 22) based on the DSM-IV but who were not diagnosed with NES using the new criteria.

Procedure

This was a cross-sectional study. The study was approved by the Ministry of Health and the hospital’s Helsinki Ethics Committee and all participants provided and signed written informed consent.

The study was conducted in three separate meetings. The first meeting comprised collection of demographic and clinical information. In addition, participants underwent full psychiatric evaluation and diagnosis as part of general admission to the institution. Diagnosis of EDs was obtained by the Institution’s psychiatrist using the Mini International Neuropsychiatric Interview (MINI-SCID) [25] for BN and BED (the DSM-IV criteria) [25] and the proposed diagnostic criteria for NES [24]. The diagnostic criteria were characterized by a significantly increased intake in the evening and/or nighttime, at least 25% of food intake, and/or at least two episodes of nocturnal eating per week; awareness and recall of evening and nocturnal eating episodes were present. In addition, the daily pattern manifested by at least three of the following features: morning anorexia, a strong urge to eat between dinner and sleep onset and/or during the night, insomnia, a belief that one must eat in order to sleep, depressed and/or mood worsenings in the evening. The disorder is associated with significant distress, maintained for at least 3 months, and not secondary to a medical or psychiatric disorder.

Individuals with night binges and/or BN or BED were recruited for the study. In the second meeting, those who agreed to participate were given an in-depth explanation about the study and procedures from the principal investigator, signed their informed consent and were instructed how to keep 7-day food diaries that assessed time and amount of food consumption including: type and amount of food eaten, food preparation details, brand name if prepackaged food, time of the eating episode, whether it was a binge and whether or not they purged (an in-depth explanation about the definition of binge and purge episodes according to DSM-IV criteria was given to the participants). In the third meeting, held after 1 week, participants met again with the principal investigator, returned the completed food diaries and completed the BDI questionnaire. After this meeting, the patients started the treatment process according to their needs.

Measures

The Mini International Neuropsychiatric Interview (MINI-SCID) is a short diagnostic structured interview developed to explore 17 disorders according to DSM diagnostic criteria. The results for the comparison of the MINI with the Structured Clinical Interview for DSM-IV for BN diagnosis revealed very good values for kappa coefficient, sensitivity and specificity [26].

Food diary

Food records are widely used in studies estimating eating patterns of populations with ED [27]. This method provides quantitative information on food consumption during the recording period [28]. Frequency of binge days and binge episodes, time of first binge episode, average of total caloric intake per day, percentage of energy consumed following evening meal, and macronutrient composition (percentage of calories consumed as carbohydrate, fat and protein—continuous variables) were calculated. Information was obtained from the 7-day food diaries completed by each participant and analyzed by “Tzameret”, food composition software system [29].

Depression symptoms for covariance were measured using the 21-item Beck Depression Inventory (BDI) questionnaire [30]. This questionnaire addresses cognitive and physical aspects of depression. Each item is rated on a four-point Likert scale ranging from 0 to 3. A score of 0–9 indicates none or minimal depression, 10–18 indicates mild to moderate depression, 19–29 indicates moderate to severe depression, and 30–63 indicates severe depression. Most of the researchers report average alpha-coefficients higher than 0.75 [31]. The instrument was translated into Hebrew [32].

Data analysis

All statistical analyses were conducted using SPSS version 21.0. The differences between groups in each variable were tested using independent sample t tests (for normally distributed variables) and for BMI as a covariate. The Mann–Whitney U test was used for non-parametric tests and the Pearson Chi-square or Fisher’s test for categorical variables.

Results

The mean age of the patients was significantly higher among the night binge group compared with the BE group. The mean BMI was 27.2 ± 7 and was also significantly higher among the night binge group as compared with the BE group (See Table 1). The Beck Depression Inventory (BDI) score differed significantly between the groups, with the BE group scoring significantly lower than the night binge group. No significant differences were found in formal education level (more than 12 years), occupational status and smoking. The mean length of formal education did not differ significantly between the groups. Half of the patients were diagnosed with BN and half with BED, but the NES–BE group contained significantly more patients with BED than the BE-only group.

Table 1 Demographic and clinical characteristics by groups using chi-square and T test (n = 59)
Full size table

Binge episodes: comparison between groups

Results according to Mann–Whitney independent samples testing indicated that based on food diary records, the night binge group reported a significant higher weekly frequency of binge days and binge episodes, as compared to the BE group.

The night binge group consumed significantly more total calories compared to the BE group during an average day.

t test results indicated that the night binge group had a greater percentage of daily caloric intake following the evening meal, compared with the BE group.

Table 2 presents the characteristics of binge episodes (number of binge days, binge episodes per week and daily caloric intake on binge and non-binge days and average caloric intake consumed daily per week) by groups (means ± SD, statistical test and p values, after covariate to BMI, age or BDI).

Table 2 Number of binge days, binge episodes, daily caloric intake during the week (Means ± SD, statistic test and p values) by groups
Full size table

Macronutrient comparison between groups

Average fat consumption per day was found to be significantly higher among the night binge group compared to the BE group. The average carbohydrate consumption per day was significantly lower among the night binge group compared with the BE-only group, and carbohydrate consumption was significantly lower on non-binge days. No significant differences were found in protein consumption between groups.

Table 3 presents means ± SD and p values for the percentage of daily caloric intake consumed as fat, carbohydrates and protein by groups, after covariate to BMI, age or BDI.

Table 3 Mean (sd) of macronutrients distribution comparison by groups (using T test)
Full size table

The relationships between macronutrients and binge episodes

The average amount of fat consumed per day was positively related to the number of binge days, the number of binge episodes per week, the kcals consumed per day and the percentage of calories consumed after the evening meal. The amount of fat consumed on binge days was positively related to the number of binge days, the kcals consumed per day and the percentage of calories consumed after the evening meal. The average carbohydrates consumed per day was positively related to the percentage of calories after the evening meal. The average protein consumed per day was negatively related to the number of binge days, the number of binge episodes per week, the kcal consumed per day and the percentage of calories consumed after the evening meal. The protein consumed on binge days was negatively related to the number of binge days, the number of binge episodes per week and the kcal consumed per day.

Table 4 presents the Pearson correlations between macronutrients and number of binges, kcal consumed per day and percentage of calories consumed after the evening meal as reported in the diaries.

Table 4 Pearson correlation between macronutrients and number of binges, energy intake per day and percentage of energy intake after evening meal
Full size table

Discussion

It is clear that people with night binges present with psychological distress and disturbed eating patterns that are more severe than among people without night binges as a main symptom. However, it is not entirely clear whether NES represents a distinct entity within EDs or whether it is a variant of binge eating spectrum disorders including BED or BN, in which the binge episodes occur mainly in the evening. Various studies have attempted to understand the relationship between NES, BED and BN.

The current study attempted to shed light on these questions from dietary patterns’ perspective. It is one of the first studies to use the new proposed diagnostic criteria for NES [24] examining patients who seek treatment for ED in an outpatient ED institution.

The aims of this study were to compare the dietary patterns of patients with binge eating behavior (BED or BN) with NES to those without NES in terms of frequency of binge episodes, number of calories consumed and macronutrient composition of their diet.

According to the aims of the study, the results indicated that patients diagnosed with binge eating behavior with NES reported a higher frequency of binge days and binge episodes, as well as a higher total caloric consumption, greater consumption of fat and lower consumption of carbohydrates compared with those diagnosed with binge eating behavior without NES, even when controlling for potential confounders such as BMI, age or depression (BDI).

To the best of our knowledge, only one other study has investigated dietary patterns among NES and found that patients with BED and NES did not differ from those without NES [33]. Their results, which are contrary to the results of the current study, may be related to the different diagnostic criteria used to define NES in both studies. Grilo & Masheb [33] defined NES as at least one NI episode over the last 28 days. In the current study, we used the new criteria outlined by Allison et al. [24] which include both EH as well as a minimum of two NI episodes per week. In addition, these differences may be related to the different methodology used in the current study to assess the frequency of binge days and binge episodes using food diaries as opposed to the EDE-Q.

Nevertheless, our findings are in line with communal studies indicating that based on the EDE-Q, patients with NES reported a significantly greater frequency of binge episodes in the past 28 days than patients without NES [22, 34].

The current study findings may indicate that patients who have NES in addition to BE may have higher levels of eating pathology than patients who have BED or BN alone. This explanation is supported by previous results indicating a higher level of psychiatric comorbidity among patients who demonstrate NES together with BE [35, 36].

To the best of our knowledge, this study was also the first to compare the number of calories consumed among individuals with BE with and without NES. The current results indicated that the BE–NES group consumed significantly more calories during an average day than did the BE-only group. This new finding indicates that NES (both EH and NI) episodes contribute to overall elevated total caloric intake among patients with BE. The current results are in line with previous studies conducted in a community setting among individuals seeking weight loss treatment who were not diagnosed with other EDs [15, 18]. Our findings may indicate that individuals with both BE and NES may represent a subgroup with higher levels of eating pathology than individuals with BE only, although further research is needed to better understand this finding.

Another finding of the current study is that the percentage of calories consumed after the evening meal was significantly higher among the BE–NES group, compared with the BE-only group. This difference may indicate that the research group in the current study was classified correctly and that patients diagnosed with NES indeed consume a greater number of calories after the evening meal. Yet, according to the new diagnostic criteria for NES [24], individuals with NES should consume at least 25% of their total caloric intake after the evening meal, and not 20% as found. This may be because dietary intake as reported in food diaries is seriously underestimated, as related to the level of motivation or to shame or embarrassment. It may also be related to a sense of relief and feeling of security during the first week of assessment that may be reflected in a temporary change in night eating patterns [28]. Another explanation may be due to cultural differences in terms of dinner time.

In addition, the results indicated that the percentage of calories consumed after the evening meal was positively related to general fat consumption and negatively related to carbohydrate consumption, so that patients who consumed a higher percentage of daily calories after their evening meal tended to eat more fat and less carbohydrates.

These new results are the first to show the macronutrient composition of patients with BE with and without NES. The results showing that the BE–NES group consumed more fat and less carbohydrates may be explained by the higher frequency of binge episodes found in this group. These results are in line with previous findings showing that binge episodes in subjects with BN are also characterized by a higher consumption of fat than in non-binge meals [5].

In the current study, the time of the first binge episode in both groups was in the evening. In the literature, binge eating behavior in both BN and BED often occurs during the afternoon and evening hours [3, 37].

Study limitations

Several potential limitations of this study should be considered. First, the small sample size reduces the statistical power for detecting group differences and for generalization. A larger sample size would also permit differentiation between patients with BN or BED and the identification of possible differences in nutrition consumption and binge episodes. Future research is also necessary to better clarify the definition of NES, which currently includes both night ingestion and evening hyperphagia sub-syndromes. As evening hyperphagia is very similar to BED, it is important to separate these two sub-syndromes patients with EDs to better determine if evening hyperphagia is part of BED or BN or a separate ED subgroup rather than constituting part of NES with regard to nutritional consumption and binge eating behavior. Future studies should examine dietary patterns for each disorder separately.

Second, our findings may not apply to individuals with NES without ED diagnosis in the community, since they may have different eating patterns than those seeking treatment for EDs.

Third, food diaries were used to assess dietary consumption. This tool is widely used in studies estimating eating patterns of ED populations. It provides subjective quantitative information on food consumption during the recording period, and is often regarded as the gold standard against which other dietary assessment methods are compared [38]. However, dietary intake as reported in food diaries may be seriously underestimated, in particular among individuals with overweight or obesity. Moreover, in the current study the diaries were completed during the first week of assessment, when the motivation for change in general and in eating behavior in particular may be elevated.

Future directions

The dietary patterns of individuals with binge eating behavior with and without NES represent an important area for further research. Comprehensive understanding of these patterns will help researchers and clinicians facilitate effective treatment efforts. Moreover, it will help clarify the important nosological question of whether NES represents a separate entity or a variant of BED or BN in which the binging occurs mainly in the evening.

Future studies should be conducted while trying to overcome the limitations discussed above. Furthermore, it would be interesting to analyze the caloric and macronutrient composition not only throughout the day but also during the binge episodes themselves. These studies should check whether binge episodes of patients with and without NES are different and whether nighttime binge episodes of patients with both binge eating behavior and NES differ from their daytime binge episodes.

Summary

To the best of our knowledge, the current study was one of the first to assess dietary consumption and binge episodes in patients who sought treatment for binge ED with and without NES. Our findings indicated that patients with NES in addition to BED or BN have higher levels of eating pathology than patients with BED or BN alone. They experience more binge eating episodes and consume more calories and more fat than patients with binge eating behavior alone. In light of this, we suggest that NES among patients with BED or BN does not represent just a variant of BED or BN, but rather a separate entity that leads to a more severe disorder. According to our findings, the presence of NES contributed to an overall elevated total caloric intake, which in the long term may lead to weight gain. Thus, from a clinical perspective this may highlight the importance of early assessment of the presence of NES in patients with BED and BN. We cautiously suggest that the treatment of patients who have NES in addition to BED or BN should be more intensive.