Descriptive Study

D - Click here for explanation of classification scheme.

NEUTRAL: See Quality Criteria Checklist below.

To collect descriptive data on participants who were selected using broad inclusion criteria for nighttime eating.

• Consumption of 50% or more of daily food intake in the evening (this could include supper)
• Not feeling hungry at all or only a little hungry in the morning
• Trouble getting to sleep or staying asleep at least once a month
• Nocturnal eating at least once a month (getting up in the middle of the night to eat).

• Less than 18 years old
• Pregnant or breast-feeding
• Working night or evening shifts
• Abusing alcohol or drugs
• Physical illness influencing metabolism and food intake (e.g., insulin-dependent diabetes mellitus, untreated thyroid disease, cancer, any condition requiring a special diet)
• Medication that might influence sleep or eating (e.g., diuretics, insulin, steroids; psychotropic medication was allowed if there was no clear temporal relation between onset of medication use and onset of nighttime eating).

Recruitment

Newspaper advertisements that sought individuals who "eat large amounts of food in the evenings" and "get up during the night to eat." 

Design

Descriptive study.

Intervention

Respondents to newspaper were screened via telephone call. Those meeting inclusion criteria were mailed a consent form; once signed consent received, time for telephone interview was scheduled. Each participant was administered several instruments via one- to 1.5-hour telephone interview:

• Night Eating Questionnaire (NEQ)
• Night Eating Syndrome History and Inventory (NESHI)
• Sleep Disorder Questionnaire (SDQ)
• Eating Disorder Questionnaire (EDQ)
• Structured Clinical Interview for diagnoses outlined in the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (SCID). 

Statistical Analysis

• Descriptive statistical approach used
• To compare groups, chi-square tests, T-tests, and analyses of variance (ANOVA) were used as appropriate
• Cluster and latent class analyses considered, but sample size judged too small to conduct such analyses.

Timing of Measurements

Between January and December, 2002.

Dependent Variables

•  Diagnosis of Night Eating Syndrome, using 1955 criteria
  
  • Eating 25% or more total daily calories consumed after the evening meal (81; 76.4%)
  • Morning anorexia with negligible food intake at breakfast (80; 75.5%)
  • Sleeplessness, at least until midnight more than half the time (70; 66%)
  • (Optional) Mood lower in the evening (17; 38.6%)
• Diagnosis of Night Eating Syndrome, using 1999 criteria
  
  • 50% or more total daily calories consumed after last evening meal (48; 45.3)
  • Morning anorexia, even if participant eats breakfast (80; 75.5%)
  • Awakenings at least once a night (67; 63.2%)
  • Consumption of snacks during awakenings (61; 57.5%)
  • Repetition of the provisional criteria for three or more months (106; 100%)
  • Current Binge Eating disorder, rules out NES by 1999 criteria (9; 8.5%)
  • Current bulimia nervosa, rules out NES by 1999 criteria (4; 3.8%).

Independent Variables

• Normal or obese BMI (mean BMI 30.8kg/m²)
  • BMI less than 25, normal (21; 19.8%)
  • BMI 40 or more, morbidly obese (12; 11.3%)                                            
• Lifetime history of eating disorder
  
  • BED (15; 14.2%)
  • BN (6; 5.7%)
  • AN (2; 1.9%)
• Psychiatric co-morbidity
  
  • Major Depressive Disorder (MDD) (59; 55.7%)
  • Bipolar I (3; 2.8%)
  • Dysthymia (current only) (12; 11.3%)
  • Anxiety disorders
    • Panic disorder (11; 10.4%)
    • Agoraphobia (1; 0.9%)
    • Social phobia (8; 7.5%)
    • Specific phobia (7; 6.6%)
    • OCD (6; 5.6%)
    • PTSD (19; 18.1%)
    • Generalized Anxiety Disorder (GAD; current only) (18; 17.5%)
• Sleep disorders (N=92)
  
  • Restless legs (46; 50%)
  • Periodic limb movement disorder (PLMD) (30; 32.6%)
  • Bruxism (32; 34.8%)
  • Nightmares (35; 38%)
  • Sleepwalking (11; 12%)
  • REM Sleep behavioral disorder (25; 27.2%)
  • Confusional arousal (14; 15.2%)
  • Night Terrors (6; 6.5%)
  • Apnea (9; 9.8%).

• Initial N: 286 participants screened via telephone, 56 excluded, 230 accepted, 106 individuals returned signed consent for and scheduled for telephone interview
• Attrition (final N): 106 individuals, 64.2% female
• Age: Average 49.2 years
• Ethnicity: 94% Caucasian
• Other relevant demographics:
  • Six (5.7%) diagnosed with type 2 diabetes
  • 30 (28.3%) taking antidepressants
  • 29 (27.4%) taking prescription sleeping pills at least once a week at time of interview
• Anthropometrics: Average BMI 30.8kg/m² (SD=7.2), range 15.8 to 54.3kg/m², 21 (19.8%) had BMI less than 25kg/m² and 12 (11.3%) had BMI 40kg/m² or more
• Location: Neuropsychiatric Research Institute, Fargo, North Dakota.

Key Findings

Association Between 1955 and 1999 Diagnoses of NES

Criteria for 1995 Diagnosis of NES Met	Criteria for 1999 Diagnosis of NES Met	Criteria for both 1955 and 1999 Diagnosis of NES Met	Statistical Significance of Agreement Between Old and New Criteria
31	14	8	χ2=6.067, P=0.014
29.2%	13.2%	7.5%

Association between Independent variables and NES

Variable	Association with 1955 Criteria for NES Significance	Association with 1999 Criteria for NES
Lifetime history of major depressive disorder 59, 55.7%	71% χ2 = 4.160, p-0.041	No difference Not significant
Other independent variables	No significant association with NES symptoms

Other Findings

• Participants with BMI less than 25kg/m² did not differ from participants with BMI more than 30kg/m² on any of the night eating symptoms, nor did they differ for mean age and gender
  • 60% of obese participants reported being overweight before night eating began, and there was no significant difference in age of onset of night eating between normal and overweight participants
• Participants with a lifetime history of eating disorder did not differ from participants without a lifetime history of eating disorders in any of the night eating symptoms. Participants with history of ED:
  • Had higher prevalence of major depressive disorder (80% vs. 50% χ2=7.244, P=0.27)
  • Were more likely to report current sleepwalking (20% vs. 8%, χ2=7.244, P=0.027)
  • Were more likely to report current confusional arousal (35% vs. 8.1%, χ2=12.616, P=0.002)
  • Women more likely to have history of ED (85% vs 59%, χ2=4.659, P=0.031)
  • Participants with eating disorder significantly younger (42.7 years vs. 50.7 years, T=2.514, P=0.013)
• Nocturnal eating (eating when waking during the night) occurred an average of twice a night among participants meeting full 1999 criteria, and almost all awakening was accompanied by eating
• Participants who met full criteria for either diagnostic construct at breakfast and lunch significantly less often compared with those who did not meet full criteria for either diagnostic construct
• Participants who met 1955 criteria reported significantly less control over evening eating than those who did not meet full criteria (42.9 vs. 62.7, T=2.685, df=95, P=0.009)
• Nocturnal eating episodes: 14 participants without any nocturnal eating episodes were excluded, final sample for this analysis of N=92 (31 men and 61 women)
  • On a scale of zero to 100, self-reported average amount of control over nocturnal eating episodes was 41.4 (range zero to 100)
    • 41 participants (44.5%) reported no or just a little control
    • 32 (34.8%) reported complete or very much control over nocturnal eating episodes
    • 80% with nocturnal eating reported consuming 25% or more and 46% reported consuming 50% or more of daily food intake after the evening meal
    • Average age of onset of nocturnal eating was 33.2 years with 20 (21.7%) reporting onset at less than 19 years
    • Nocturnal eating was not associated with purging behavior
    • 16 participants (17/5%) reported they were completely (N=1), mostly (N=5) or partially (N=10) unaware of nocturnal eating episode
    • Only 15 (16.3%) indicated that they woke up specifically to eat; 33 (35.9%) woke to use bathroom, 24 (26.1%) woke due to insomnia and 20 (21.7%) woke for other reasons
    • Nocturnal eating episodes were consistent throughout week for 76.1% participants; 17.4% reported improvement on weekends and days off; 3.3% reported worsening on weekends and days off
    • Most participants had never consulted a physician for nocturnal eating [N=80 (87%)].

• This descriptive study confirms some of the observations made by other researchers and has many faces and names [Night Eating Syndrome (NES), Nocturnal Eating/Drinking Syndrome (NEDS), Sleep Related Eating Disorders (SRED)]. It remains unclear if these are related or distinct syndromes, whether night eating is a disorder of arousal or circadian rhythm or whether it can be perceived as an eating disorder.
• The purpose of this study was not to offer a definitive method of classifying NES but, instead, using a community sample, to demonstrated that the criteria sets that are being used by different groups (obesity researchers, sleep researchers) overlap and do not overlap in noteworthy ways, and that the current systems are inadequate. The study points out areas of obvious weakness in the current systems that people are using. Most likely, the participants in our and other samples represent a heterogeneous group of participants, with different mechanisms being responsible for their night eating behavior.

University/Hospital:

Neuropsychiatric Research Institute, Fargo, North Dakota

• This study gave little information on what is associated with NES, and a lot of information about what was not associated with NES, based on a small sample, with no healthy control group. The sample was self-selected and less than one half of participants who qualified actually participated; it is unclear if results would generalize to all participants with nighttime eating problems.
• The only independent variable that was found to be statistically significantly related to NES was Major Depressive Disorder. The study showed a variety of associations between the different independent variables, and described variations among symptoms within the criteria for NES.
• The most important information from this study was that there is not agreement among researchers as to criteria (let alone names) for Night Eating Disorders. Before this night eating problems can be well researched, there needs to be agreement on diagnostic criteria. This is definitely an area that needs more research.