SEARCH

SEARCH BY CITATION

Keywords:

  • Control;
  • night eating;
  • night-eating syndrome;
  • sleep

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Limitations
  8. Conclusion
  9. Conflict of interest statement
  10. Acknowledgements
  11. References
What is already known about this subject
  • Night-eating syndrome (NES) can be a feature of severe obesity.
  • NES is a dysfunction of circadian rhythm and is associated with impaired sleep.
What this study adds
  • Night eaters with severe obesity are more likely to be low in mood and unemployed compared with non-night eaters.
  • Night eaters with severe obesity describe compulsive and uncontrolled eating.

Research interest in night-eating syndrome (NES) has grown in recent years in line with increased rates of obesity. This study used a mixed-methods approach to investigate its characteristics in severe obesity. Eighty-one individuals (mean [standard deviation] age 44.6 [11.6] years, [body mass index] 50.0 [10.7] kg m−2; 43% men) from a hospital-based UK obesity clinic were interviewed for NES based on 2003 criteria. Full and partial NES were combined into one night-eating behaviour (NEB) group (n = 31). Demographic and clinical characteristics were compared with those of non-NEB individuals (n = 50). NEB characteristics were also identified through exploratory thematic analysis of interview data. NEB individuals had lower mood (P = 0.01) and were less likely to be employed (P = 0.03). Differences in mean age and reported sleep duration were not significant. Thematic analysis of patient perceptions of NEB highlighted the potential heterogeneity of NEB development: NEB developed in childhood, adolescence and adulthood. Individuals reported long-standing and current sleep difficulties, negative affect and conflictful relationships. Night eating was solitary, compulsive and uncontrolled, and daytime eating patterns were chaotic. Accounts of awareness of night eating were conflicting. Severely obese night eaters are characterized by low mood and lack of employment. Further studies are required to explore behavioural and cognitive influences on night eating in severe obesity.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Limitations
  8. Conclusion
  9. Conflict of interest statement
  10. Acknowledgements
  11. References

The increase in rates of obesity in the developed world in the past three decades has seen a resurgence of interest in night-eating syndrome (NES). First identified in 1955 by Stunkard [1], NES can be considered a circadian rhythm dysfunction with a disassociation between eating and sleeping [2]. Current diagnostic criteria require significant increased food intake in the evening and/or night-time, as manifested by one or both of the following: at least 25% of food intake is consumed after the evening meal, or at least two episodes of nocturnal eating per week together with significant distress and/or impairment in physical and psychosocial functioning [3]. Awareness of night eating must be present to distinguish from sleep-related eating disorder (SRED), which requires night-eating amnesia [4].

NES is known to feature in severe obesity [5], although studies identifying NES characteristics in severely obese groups are limited and it is unclear whether characteristics are different to those in mildly obese patients. NES was associated with higher body mass index (BMI), male gender and binge eating in Australian adults seeking weight loss surgery [6], although night eating and binge eating are considered separate pathologies [7]. Others have not found associations with NES and BMI and suggest this finding may be restricted to clinical populations [8]. Identifying and characterizing NES in severe obesity is important as these individuals are often refractory to standard treatments, including having less successful outcomes after bariatric surgery [9].

Depression and sleep debt are associated with severe obesity [10]. Depression is also associated with NES regardless of BMI [11, 12]. Yet, the impact of NES on the complex interplay between sleep, depression and obesity-related comorbidity in severe obesity is unclear [13]. NES females (mean [standard deviation {SD}] BMI 36.0 [7.0] kg m−2) report significantly worse sleep quality than non-NES females [14], but little is known about the sleep characteristics of severely obese individuals with co-existing NES.

The aims of this study were twofold: (i) to evaluate the demographic, clinical and behavioural characteristics of night eaters in comparison with non-night eaters in a severely obese UK outpatient clinic population (BMI > 40 kg m−2) and (ii) to identify participants' perceptions concerning the development of night-eating behaviours (NEBs) and living with NES.

Materials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Limitations
  8. Conclusion
  9. Conflict of interest statement
  10. Acknowledgements
  11. References

A mixed-methods design was used.

  1. NES was identified in participants using diagnostic interviews.
  2. Demographic and clinical characteristics of night eaters and non-night eaters were compared, and validated tools used to compare eating and psychological characteristics.
  3. Thematic analysis of night eaters' diagnostic interviews was undertaken and compared with non-night eater interviews

Recruitment

Following local research ethics committee approval (reference no. 04/Q1508/9), participants were recruited from a UK outpatient clinic over 18 months. All participants provided written informed consent. Participants were originally recruited to participate in a study to develop and test a screening tool for NES. As no screening tools existed for NES at the time the study was undertaken, participants with potential NES were recruited with a flyer. Clinicians alerted consecutive clinic attendees to the flyer, which asked three questions: ‘Do you find it difficult to eat breakfast?’ ‘Do you often eat during the night time?’ and ‘Do you have trouble sleeping?’ Those who responded positively to the flyer initially were given a patient information sheet (PIS). Additionally, other clinic attendees who did not report night eating and had not responded to the questionnaire were invited during clinic to consider participating as controls and given a PIS. All potential participants were then telephoned to confirm their interest. Inclusion criteria comprised new clinic attendees (individuals with BMI >40 or >35 kg m−2 with comorbidity) and follow-up patients (with a potentially lower BMI), who are willing to attend a diagnostic interview. Exclusion criteria comprised previous bariatric surgery, age <18 years and unable to speak English.

Of 400 clinic attendees, 91 responded positively to the flyer, of whom 54 consented to participate in the study. From consultations in normal clinical practice, 30 individuals deemed not to have NES were recruited as controls. Eighty-one individuals (mean [SD] age 44.6 [11.6] years, BMI 50.0 [10.7] kg m−2; 43% men) completed all study procedures; three others not completing questionnaire tools were excluded (Fig. 1).

figure

Figure 1. Recruitment flow chart.

Download figure to PowerPoint

Data collection

Diagnostic interview

To diagnose NES, participants underwent a standardized diagnostic interview based on 2003 criteria, which were current at the time the study was undertaken (Table 1) [15]. The interview comprised the Night Eating Symptom and History Inventory (NESHI), a 17-item semi-structured interview schedule, developed by Stunkard's team and used to confirm a diagnosis of NES. Although lacking in validity studies, it is widely used in NES research and is currently the only diagnostic tool for NES in an interview setting [16].

Table 1. 2003 diagnostic criteria for night-eating syndrome (NES)
NES 2003 criteria
  • Morning anorexia, even if subject eats breakfast.
  • Evening hyperphagia. At least 50% of the daily caloric intake is consumed in snacks after the last evening meal.
  • Awakenings at least once a night, at least three nights a week.
  • Consumption of high-calorie snacks during the awakenings on frequent occasions.
  • The pattern occurs for a period of at least 3 months.
  • Absence of other eating disorders.

To exclude the presence of other eating disorders (EDs), the interview included the Eating Disorder Examination (EDE) v12, a well-validated structured clinical interview containing diagnostic items for bulimia nervosa and anorexia nervosa [17]. The EDE assesses frequency of different forms of over-eating including objective bulimic episodes and yields four subscales of related psychopathology as variables of interest, namely, restraint, eating concern, shape concern and weight concern [18]. Supplementary questions for identifying binge ED (BED) were also included [19]. Audio-recorded interviews lasting 60–90 min were conducted in a dedicated research unit.

Additional tools

Participants completed the Night Eating Questionnaire (NEQ), a 14-item validated tool to characterize night-eating symptom severity [8]. Developed by the same research group, the NEQ reflects constructs in the NESHI interview schedule in a questionnaire format. Evaluation work on the validity and reliability of the NEQ was not available at the time this study was undertaken, therefore scoring the tool for diagnostic purposes was not appropriate, although items were useful in informing NES characterization. To assess mood, participants completed the Beck Depression Inventory (BDI-IA), a 21-item widely validated measure that assesses cognitive and behavioural depressive symptoms [20].

Demographic and clinical characteristics

Participants self-reported usual sleep and wake times, including differences on weekdays and weekend, work and non-work days. An average total sleep duration (hours) in the past week was calculated. Additional information on sleep quality was extracted from interviews when present. Participants also self-reported age, gender, work status and the presence of childhood obesity. Diagnoses of anxiety and depression, along with obesity-related comorbidity (type 2 diabetes and obstructive sleep apnoea [OSA]), were identified from medical correspondence. Height and weight were measured, and BMI was calculated prior to the diagnostic interview.

Data analysis

Classification of night-eating syndrome

Full NES was defined by participants testing positive for all disorder-specific 2003 criteria. During the course of this study, Stunkard's team acknowledged the restrictive nature of these criteria, suggesting a more inclusive approach and relaxation of the calorie requirement after the evening meal from >50 to >25% [21]. Thus, in keeping with this broader approach, a partial NES category was defined as >25% of daily calorie intake consumed in snacks after the last evening meal, waking up to eat and/or eating prior to going to bed very late, regardless of frequency and duration of behaviour, amount eaten, presence of morning anorexia or other ED. Full and partial NES individuals were combined to form one NEB group. All other participants were classed as non-NEB. To confirm reliability of diagnosis, the first 18 interviewees were re-interviewed by a second clinician blinded to the original results. Inter-rater reliability of NES diagnosis was high with a kappa coefficient of 0.89 (95% [confidence interval {CI}] 0.65, 1.0).

Comparison of night-eating behaviour and non-night-eating behaviour group characteristics

Data analysis was performed with Statistical Package for the Social Sciences (version 14; SPSS Inc., Chicago, IL, USA) and Intercooled Stata (version 9; StataCorp, College Station, Texas, USA). Independent samples t-tests were used for between-group analyses of means on approximately normally distributed data and Mann–Whitney U-tests on other ordinal data. Frequencies (%), with chi-square testing, were calculated for gender, work status, comorbidity and childhood obesity. A post hoc F-test of equality of group variances was performed on sleep duration. Ninety-five percent CIs for differences were presented where relevant. A Pearson correlation coefficient was calculated to examine the strength of relationship between the NEQ and BDI total scores, and NEQ scores and age. Nominal P-values are given without corrections for multiple testing because of the exploratory nature of the analysis.

Qualitative analysis of night-eating behaviour interview

NEB interviews were independently transcribed and quality checked by the author (JC). ‘Unclear’ items were clarified by replaying audio tapes or checking interview notes. Two transcriptions were examined line by line using open coding. Once all possible codes had been identified, both transcriptions were entered into the ‘NVivo’ computer software program (QSR International, Warrington, UK), and codes were organized into preliminary categories and subcategories [22]. All other interviews were subsequently entered into ‘NVivo’ and subjected to line-by-line coding. Four transcriptions were checked independently to ensure narrative interpretation was to an adequate depth and with an appropriate degree of theoretical sensitivity [23]. NEB-related themes emerging from the categories are presented in narrative form.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Limitations
  8. Conclusion
  9. Conflict of interest statement
  10. Acknowledgements
  11. References

Seven individuals with full NES and 24 with partial NES were identified from diagnostic interviews and combined into one NEB group (n = 31). All other individuals were classed as non-NEB.

Comparison between the night-eating behaviour and non-night-eating behaviour group (Table 2)

Table 2. Comparison of characteristics of the NEB and non-NEB groups
CharacteristicMean (SD)Mean difference (95% CI)aP-value
NEB (n = 31)Non-NEB (n = 50)
  1. Differences reported as mean (SD) and using independent sample t-tests unless otherwise stated.

  2. a

    NEB–non-NEB.

  3. b

    Chi-squared test.

  4. BDI, Beck Depression Inventory; BMI, body mass index; CI, confidence interval; NEB, night-eating behaviour; NEQ, Night Eating Questionnaire; SD, standard deviation.

Age (years)42.3 (14.1)46.0 (9.7)−3.7 (−9.5 to 2.1)0.21
Number (%) of menb14.0 (45)21.0 (42.0)3% (−20 to 30%)0.78
BMI (kg m−2)52.5 (11.7)48.4 (9.8)4.1 (−0.7 to 8.9)0.09
Weight (kg)155.3 (37.5)138.3 (34.7)17.0 (0.7–33.3)0.04
Number (%) of employedb9.0 (29.0)27.0 (54.0)−25% (−50 to −4%)0.03
Number (%) with type 2 diabetesb8.0 (26.0)10.0 (20.0)6% (−13 to 28%)0.54
Number (%) with obstructive sleep apnoeab6.0 (19.0)14.0 (28.0)9% (−12 to 28%)0.50
Number (%) with depressionb14.0 (45.0)17.0 (34.0)11% (−11 to 33%)0.32
Number (%) with childhood obesityb19.0 (61.0)23.0 (46.0)15% (−7 to 37%)0.25
Sleep duration (h)7.6 (2.0)7.8 (1.2)−0.2 (−1.0 to 0.6)0.57
BDI total score24.8 (10.9)17.7 (10.8)7.1 (2.1–12.0)0.01
NEQ total score28.3 (7.6)13.0 (5.3)15.3 (12.0–18.5)<0.001
Duration of night eating (years)9.3 (12.7)N/AN/A 

The NEB group was slightly younger, although no significant difference in mean age was noted. Relatively more NEB individuals were aged under 35 (13 [41.9%] vs. 5 [10%], P = 0.001, range 18–68 vs. 22–65 years). Just under half of both groups were men. The NEB participants had significantly higher weights, although the difference in BMI was not significant. NEB participants were less likely to be working. Depression and/or anxiety were common in both groups, and although not significantly greater in those with NEB, these participants scored significantly higher on the BDI-IA scale, indicating greater depressed mood.

The mean duration of NEB was 9.3 (12.7) years (range <3 months to 47 years). Both group means of self-reported average total sleep time were below 8 h (NEB: 7.6 [2.0], range 3.5–13.0 h vs. non-NEB: 7.8 [1.2], range 5–10 h). Although group means of individuals' self-reported average total sleep time were not significantly different (Table 2), a post hoc test showed significantly greater variability in individuals' average total sleep time among the NEB group (P < 0.01).

Night-Eating Questionnaire

As anticipated, the mean total score of the NEQ was significantly higher in the NEB group (Table 2). A moderate correlation between the NEQ and BDI-IA total scores was noted (r = 0.47, P = 0.01), along with a weak negative correlation between the NEQ total score and age (r = −0.28, P = 0.013). No significant overall differences between groups were seen for the median EDE global and subscale scores (data not shown).

Thematic analysis of night-eating behaviour interviews

Thirty transcripts were analysed as one participant declined to be audiotaped. Thirteen broad categories with 185 subcategories were identified (see coding example in Table 3). Eight NEB-related themes emerged and are summarized below.

  1. Perceptions of NEB development

    Six individuals reported childhood NEB, 6 teenage-onset NEB and 18 adult-onset NEB. Of those developing NEB as a child, four said it was in response to being forced to diet by parents as a result of childhood obesity.

    My mum and dad were always on at me, you know, ‘you're getting chubby, you know, you're a chubby child’ … so I would try and be good, … but I knew when my mum and dad was in bed, I can remember always sneaking down, a biscuit, a sandwich, or something out of the fridge (065).

    Reported influences on teenage-onset NEB were bullying, trauma and family conflict. Childhood loss (death/absence of a significant individual) featured prominently. Adult-onset NEB was related to milestone life events (e.g. divorce, bereavement). The influence of events on NEB development and obesity appeared mixed, with some obese before the event and others developing obesity subsequently.

  2. Effect of poor sleep on the onset of NEB

    Most NEB individuals (63%) reported long-standing sleep problems. Three described poor sleep caused by earlier work patterns, and others developed disrupted sleep and subsequent NEB after stopping work. One teenager formerly addicted to cannabis linked his subsequent NEB to smoking-related night-time hyperphagia. Another related disrupted sleep and the start of NES to the distress of bullying.

    I think honestly the problem (NEB) started when I was getting bullied in school … my sleeping problem was there first, I think, because I used to be worrying about going to school … there was like about 2 years where I'd wake every hour on the hour and I was so tired in school that the school actually said that I kept falling asleep and I didn't know I was doing that (009).

  3. Current sleep difficulties

    For all NEB individuals, sleep was a negative experience. Sleep-onset insomnia was reported by 86%, some described emotional influences, two reported safety concerns (e.g. threats of sibling violence) and one individual blamed his medical condition (chronic obstructive pulmonary disease).

    That's probably the most annoying thing in me at the moment is not getting to sleep. It just does my head in, why can't I just go to sleep (015).

    Most (87%) described being easily roused from debilitating short periods of fitful sleep. Reasons for wakening were either unknown, emotional or hunger related with physical pain and OSA featuring prominently.

  4. Night eating and control

    Many NEB participants (55%) described night eating as compulsive. Seventy percent of participants felt compelled to eat in order to return to sleep, although individuals who stayed up very late eating and then went to bed did not experience this.

    It's hard to say but I believe I was in control, … I knew what I was putting into my body and I knew what the results would be, but at the same time it was sort of like I knew I didn't want to do it but it was one of them things where you get something in your mind and you want to do it and you just go for it, … it's literally just eating and eating and eating, as if like a magnet, like that where I was getting pulled towards it (040).

    Three individuals had to get out of bed to eat before making an initial attempt to sleep. Those who tried to resist the urge to eat described failure as inevitable and of feeling relieved at finally being able to sleep after eating. Once awake, almost a third of the group described getting up to eat because of ‘hunger’. Others ate in response to negative affect, such as feeling empty or unhappy. Emotions during night eating appeared neutral, with participants struggling to clarify feelings during eating but acknowledging subsequent guilt and negative self-perceptions of themselves. Only two individuals reported visible upset.

  5. Awareness of night eating

    Accounts of awareness were conflicting. Fifty percent felt they were fully aware while night eating, although eight reported partial awareness, feeling ‘dazed’ or ‘half asleep’. Four individuals were aware at the time but could not remember in the morning, or ‘woke up’ while eating.

    How much awareness? Oh I know I've got the food there, I can see it. But again I look at what I'm looking at and I go ‘… have I ate them already? They've gone quick.’ I'm not always really aware that I've ate at all (024).

  6. Secret eating

    All night eaters reported eating alone at night. Six individuals brought food back to bed, although the majority ate downstairs; one individual ate in the bathroom. Popular food choices for all night eaters were high-fat/high-sugar snacks and sandwiches. Although some denied actively concealing night eating, others reported concealing both night-time and daytime eating from family members.

    I don't know why but I always do it in the dark … because the bedrooms were in the hallway and the pantry used to be in the hallway … I'd become so experienced at opening the door without making the loud click [to avoid] … ‘who's that … what are you doing out there’ you know, my dad, shouting from the bedroom. So I got to be an expert at opening the door without making a noise (007).

  7. Chaotic daytime eating

    NEB individuals described chaotic daytime eating patterns. Individual variability was noted, with irregular meal times interspersed with frequent snacking. The first daytime meal often occurred later in the day (typically at 1300 h or later), resulting in a time shift of all subsequent meals. Most (77%) described ‘all or nothing’ dieting attempts, resulting in cycles of restraint and subsequent over-eating. Emotional responses while eating ranged from enjoyment to guilt and obsession.

    Some days when you sort of think, ‘Oh, I'm sick of this’ and you give up. Then when I do stuff like that I feel worse, so I either just stuff my face even more, or I just stop it altogether and eat healthily and then I'm putting myself back in the same place, it's just … I'm going mad (051).

  8. Negative affect and conflictful relationships

    NEB individuals reported eating behaviour influenced by low mood and obesity-related shame. Stress resulting from work, social and family conflict was particularly apparent.

    Mood terrible … I'm like stressing because I can't have this and then I start thinking well you know, what it's all for? Why should I be like this, if I like it, eat it, it's a vicious circle – I'm depressed because I'm on a diet and then when I eat it I'm depressed because I've eaten it, so I can't win either which way (080).

Table 3. Example of categories and subcategories identified from a section of narrative from a study participant using open coding
Text of narrativeCategory/subcategory 1/subcategory 2

Interviewer: Are you snacking in between your meals on those days?

Participant: No, I'm not. I'm not as bad in the day. You know in the daytime, in the morning. No I'm not, because I've got my husband there with me, so no and I don't get that urge to keep going to the fridge. I'm satisfied in the day, yeah. I don't have the sugar craving, or sweetie craving, it's more later, you know in the night that I have the sugar craving. But in the day I'm not, not too bad, maybe because he is with me.

Activity/day to day

Control/compulsion

Eating at night/get up and eat/craving to eat

Eating at night/get up and eat/current trigger

Eating patterns/secret eating

Eating patterns/snacking/variability of snacking

Family and friends/living arrangements

Family and friends/relationship with family and friends/support

Food emotions/satisfied

Personality/self-judgement/misbehaving

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Limitations
  8. Conclusion
  9. Conflict of interest statement
  10. Acknowledgements
  11. References

Key findings

This study had two aims: (i) to evaluate the demographic, clinical and behavioural characteristics of night eaters in comparison with non-night eaters in a severely obese clinic population and (ii) to identify participants' perceptions concerning the development of NEBs and living with NES. Compared with non-night eaters, NEB individuals were shown to have lower mood and were less likely to be working. Thematic analysis of NEB interviews revealed perceived heterogeneity of NEB development. Individuals reported solitary, compulsive and uncontrolled night eating, and chaotic daytime eating. Accounts of awareness of night eating were conflicting. Sleep difficulties were both current and long-standing. Individuals also reported negative affect and conflictful relationships.

Diagnostic criteria

The 2003 NES criteria required at least 50% of the daily caloric intake to be consumed in snacks after the last evening meal. In this study, individuals fulfilling this criterion were rare, with more fulfilling the broader NEB criteria. Recently, updated diagnostic criteria for NES support this more inclusive approach. Adjustments to the timing and volume of food eaten reflect the proposal that NES is a dysfunction of circadian rhythm. Either ‘late’ eating (at least 25% of food intake consumed after the last evening meal), or ‘nocturnal’ eating (minimum of two episodes a week) are now classed as NES. All individuals with NEB in this study met this requirement.

Eating behaviour

Although not essential for diagnosis, two eating behaviour descriptors, namely, ‘the urge to eat before bed or during the night’ and ‘the need to eat to return to sleep’, are also now included in the updated criteria. Although reported in this study, night eaters also described compulsive and uncontrolled eating with similar behaviours and cognitions during daytime eating, too. This reflects constructs noted in BED and bulimia, which are also characterized by perceived loss of control [24]. Current NES criteria propose that if both NES and ED co-occur, NES should defer to the other ED and be considered symptom dimensions only [3]. Yet, in the context of severe obesity, this relationship between NES, other ED and perceived control appears less clear-cut and would benefit from further investigation. Control is identified as the main predictor of outcome in bariatric surgery patients, regardless of ED behaviour [25], with worsening control associated with higher degrees of depression [9]. It is possible, given the relatively higher BMI seen in the NEB group, that severely obese individuals with and without NES have different cognitions influencing loss of control, which make restraint at high-risk times easier for some than for others.

Current criteria also require awareness and recall of night eating to be present. Awareness of night eating in this study was variable, reflecting findings of others and has led to the recent proposal that SRED criteria be reviewed [4].

Sleep

The NEB group reported poor sleep quality. This reflects a previously noted correlation between poor sleep quality and NES symptoms [26], and the descriptor for sleep onset or sleep maintenance insomnia to be present at least four nights per week [27]. Yet, poor sleep quality is also associated with obesity [27]. Mean sleep durations in both NEB and non-NEB groups were shorter than those noted in non-obese groups, in keeping with other obese populations [10].

It is interesting that the presence of distress and/or functional impairment is now a core NES criterion [3], suggesting a potential broad range of physical and psychological factors influencing night-time awakenings. Although in this study, NEB individuals reported psychological factors such as shame, guilt and visible upset. In relation to night eating, physical factors were more commonly reported. This does not support the proposal that NES is a stress-related maladaptive coping mechanism resulting in emotional arousal from sleep [28]. In a severely obese population specifically, further study is required to examine the impact of obesity-related physical dysfunction on sleep disturbance in individuals with NES. It is possible that night eating in severe obesity may be a chronic, habitual behaviour with distress of influencing events long past. Work status was not subject to detailed investigation in this study, and the combined influence on work status of NEB and sleep difficulties due to work-related stress needs further exploration. Work status may also be a useful objective measure of functioning in an obese NES population.

Mood

‘Mood is frequently depressed and/or worsens in the evening’ has also been added to the diagnostic criteria. Yet, whether this would discriminate NES adequately in obese populations remains unclear. The high depression levels found in this study reflect those of other obese and night-eating populations [29] [12]. Given the known association between poor subjective sleep quality and depression, it is possible that this relationship influences further the negative affect reported by the NEB group [30]. It is also possible that low self-esteem and shame about obesity contributed to the degree of depression, and that the relationship between NES and depression is more complex in this context.

Perceptions of night-eating behaviour development

The broad range of reported duration of NEB reflects the reported heterogeneity of NEB development. Although individuals reported many life events as potential influences, recall of life events must be treated with caution as these are subjective and potentially inaccurate, and the cross-sectional nature of this study precludes inferences about causality.

The role of obesity on the development and maintenance of night eating in severely obese populations needs further exploration. Marshall et al. reported a similar NES duration (10.4 years) in individuals with a BMI of <25 kg m−2 but a longer duration in individuals with a BMI of >30 kg m−2 (17.4 years) [31]. Fifty-one percent of their obese group reported developing NES after obesity, but evidence as to whether NES results in obesity is conflicting and not all individuals with NES gain weight [32]. Levels of childhood obesity in our study were unsurprising, given its known association with adult obesity [33]. Its impact on night eating needs further exploration as parents may unknowingly encourage NES development with strict daytime dietary control.

Limitations

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Limitations
  8. Conclusion
  9. Conflict of interest statement
  10. Acknowledgements
  11. References

Study participants were white Caucasian, reflecting the limited ethnic mix of the clinic. This may not reflect other UK obesity clinic populations and differs from similar US NES populations [31]. Cultural differences may influence the timing of meals, thus findings may not apply to other ethnic groups. Participants were seeking treatment for severe obesity, and identified characteristics may not apply to other non-clinical NES populations.

Differences in sleep characteristics may have been reduced through the participant selection process, which is likely to have led the non-NEB group to be more similar to the NEB group than the clinic population (as more than 20 of the non-NEB group responded positively to the three NES screening questions); differences may therefore have been even more striking with a more representative sample.

Conclusion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Limitations
  8. Conclusion
  9. Conflict of interest statement
  10. Acknowledgements
  11. References

This study has examined NEB in a severely obese population and has identified priorities for future research. In particular, it is important to identify factors influencing perceived control over night-time eating and to understand the impact of obesity-related comorbidity on disturbed sleep in NES. Treating severe obesity in clinical practice is challenging and often unsuccessful. A better understanding of the contribution night eating and severe obesity make to the development and maintenance of each other may improve treatment outcomes for a significant number of individuals.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Limitations
  8. Conclusion
  9. Conflict of interest statement
  10. Acknowledgements
  11. References

All authors were involved in study design, writing the paper and approving the final submitted version. Dr. Cleator and Michaela James undertook study procedures. Professor Wilding received an educational grant from Roche Products Limited to support the study fees of Dr. Cleator during the study. Thanks are due to study participants and to Professor Nicky Cullum for her invaluable editorial help.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Limitations
  8. Conclusion
  9. Conflict of interest statement
  10. Acknowledgements
  11. References