Table 1 Summary of the studies investigating impulsivity focused interventions to reduce binge eating behaviour

 Ferrer-García et al. (2017) [3]

N = 64 adults with BED or BN after CBT fails

Virtual Reality & Cue Exposure Therapy (VR-CET) vs. Additional CBT (A-CBT)

6 individual sessions twice a week

1. EDE (binge eating episodes)

1. Both groups reduced binge eating, VR - CET was significantly superior compared to A-CBT regarding the reduction of binge eating frequency and achievement in abstinence from binge eating episodes (53% vs. 25%).

2. EDE (purging episodes)

2. Both groups reduced purging behaviour, VR-CET was also significantly superior to A-CBT for achievement in abstinence from purging episodes (75% vs. 31.5%).

3. EDI - Bulimia subscale (self-reported binge eating tendency)

3. Both groups improved self-reported binge eating tendency, but VR-CET was significantly superior to A-CBT.

4. FCQ (state and trait version)

4. Both groups reduced both state and trait food craving, but VR-CET was significantly superior to A-CBT.

 Preuss et al. (2017) [32]

N = 69 treatment seeking obesity patients (40.6% OSFED, 33.3% BED)

ImpulsE (psychotherapeutic treatment to increase inhibitory control and emotion regulation) + food-specific stop-signal inhibition training vs. TAU (CBT for obesity and BED)

10, 100-min sessions in group format

1. EDEQ (frequency of episodes of overeating and objective binge eating)

1. Frequency of disinhibited overeating decreased in both conditions. Significant binge eating reduction in patients with BED at post-treatment and 3-month FU in ImpulsE group, no change in TAU.

2. SST with food stimuli (inhibitory control)

2. Significantly greater reduction in inhibitory control in ImpulsE group compared with TAU.

3. UPPS Impulsive Behaviour Scale (urgency, lack of premeditation, lack of perseverance and sensation seeking)

3. Perceived lack of perseverance and urgency significantly decreased in both groups.

 Schag et al. (2019) [33]

N = 80 adults with BED

IMPULS (impulsivity focused group intervention) vs. Control group without intervention

8, 90-min sessions in group format

1. EDE (binge eating episodes)

1. Binge eating episodes in the past 4 weeks were significantly reduced in both groups at the end of treatment. Binge eating was reduced more in the IMPULS group vs. control group at 3 months follow up.

2. DEBQ (external eating subscale)

2. External eating was reduced more in IMPULS group at the end of treatment and follow up. Control group showed reduction only at follow up.

3. BIS-15 & BIS/BAS (trait impulsivity)

3. Trait impulsivity was not significantly reduced in any group.

 Chao et al. (2019) [34]

N = 150 obese adults with binge eating

IBT-alone vs. IBT-liraglutide vs. Multicomponent (IBT + liraglutide + portion-controlled diet)

21 sessions of IBT vs.

21 sessions of IBT + 3.0 mg/d as a once-daily vs.

21 sessions of IBT + 3.0 mg/d as a once-daily + 12-week, 1000- to 1200-kcal/d diet

1. EDEQ (binge eating episodes)

1. At week 24, the IBT-liraglutide and multicomponent groups showed a significant within-group mean decline. The multicomponent group had a greater decrease compared to the IBT-alone group at week 24.

All groups had significant within-group declines in binge eating at week 52, with a greater decline in the multicomponent group.

2. FCI (frequency of food cravings)

2. All groups had significant and similar declines in total food cravings at both 24 and 52 weeks.

3. EDI (dietary disinhibition)

3. All groups had a significant within-group decline at week 24 and 52. At week 24, the IBT-alone and IBT-liraglutide groups did not differ, but the decline was significantly less in the IBT-alone group compared to the multicomponent group. At week 52, there was no significant group difference.

 Da Porto et al. (2020) [35]

N = 60 type 2 diabetic outpatients with BED

dulaglutide vs. gliclazide modified release +metformin

dulaglutide 150 mg/week vs. gliclazide modified release 60 mg/day + metformin (dosage 2–3 g/day)

1. BES (binge eating)

1. Binge eating behaviour was only significantly reduced in the dulaglutide group.

 Quilty et al. (2019) [36]

N = 49 women with BED

Psychostimulant medication (Methylphenidate) vs. CBT TAU

12 weeks of medication usage (initial dosage: 18 mg; final dosage 72 mg) vs. 12 weekly individual sessions for CBT

1. EDE (objective binge eating)

1. Objective binge episodes decreased in both groups, with no treatment effect.

2. BES (subjective binge eating)

2. Subjective binge episodes decreased in both conditions, with no treatment effect.

3. The UPPS Impulsive Behaviour Scale (urgency, lack of premeditation, lack of perseverance and sensation seeking)

3. Perseverance and negative urgency scores decreased in both conditions over time with no treatment effect. Higher levels of UPPS perseverance and negative urgency scores were associated with a better treatment outcome in both conditions.

 Brockmeyer et al. 2019 [37]

N = 50 with BN or BED

Real ABM to avoid food cues vs. Sham ABM

10 sessions within 4 weeks

1. EDE (binge eating episodes)

1. Both groups had significantly fewer binge eating episodes after the training.

2. FCQ (trait food craving, food cue reactivity)

2. Both groups reported significantly lower trait food craving and reduced food cue reactivity after the training.

3. Bogus Taste Test (food intake)

3. There was no significant change in food intake in any group.

4. AAT (approach and attention bias towards food)

4. There was no significant change in approach and attention bias towards food in any group.

 Giel et al. (2017) [38]

N = 22 women with BED

Food specific inhibition training based on antisaccade paradigm vs. Control group with free vision instruction

3 individual sessions within 2 weeks

1. EDEQ (number of binge eating episodes in the last 4 weeks)

1. There were significantly lower numbers of binge eating episodes in both groups.

2. FCQ (state version)

2. Reduced error rates and increase in food related inhibitory control in both groups.

3. YFAS (food addiction total score)

3. No effect on food craving or food addiction were found in any group.

 Turton et al. (2018) [39]

n = 27 women with BN and n = 17 with BED

vs. lean and overweight controls

Food specific Go/No-Go training vs. General Go/No-Go training (within-subject-design)

1 individual session

1. Taste test for food consumption following the training

1. Small non-significant reductions in high-calorie food consumption in the food specific vs. the general training.

2. 24- h post food diary including a sense of ‘loss of control’ and purging episodes

2. No treatment effect on binge eating or purging symptoms in the 24-h post diary.

3. FCQ (food craving)

3. No treatment effect on food craving.

 Burgess et al. (2016) [40]

N = 30 adults with BED or sub-BED

Real tDCS on DLPFC (anode right, cathode left) vs. sham tDCS on DLPFC (within-subject-design)

2 individual sessions

1.FPCT (Food craving)

2. In-lab food intake test

1. & 2. Food craving and food intake were reduced after tDCS compared to sham stimulation.

3. 5-day at-home binge eating survey (urge to binge eat and binge eating frequency 5 days)

3. Urge to binge eat in men was reduced after tDCS vs. sham; no reduction concerning binge eating frequency in both conditions.

 Gay et al. (2016) [41]

N = 47 women with BN

High frequency rTMS on left DLPFC vs. sham rTMS on left DLPFC

10 individual sessions over 2 consecutive weeks

1. Number of binge episodes in the last 15 days after stimulation

1. No significant reduction was found in any groups, and there was no difference between groups.

2. Number of vomiting episodes in the last 15 days after stimulation

2. No significant reduction was found in any groups, and there was no difference between groups.

 Kekic et al. (2017) [42]

N = 39 adults with BN

tDCS on DLPFC (anode right/cathode left) vs. tDCS on DLPFC (anode left/cathode right) vs. sham tDCS on DLPFC (within-subject-design)

2 individual sessions

1. Urge to binge eat on visual analogue scale

1. Both active conditions vs. sham show significant reduction in urge to bingeeat.

2. FCT (food craving)

2. There was no group difference for food craving.

3. Temporal Discounting (general reward processing)

3.Increased discounting in both active conditions vs. sham condition

4. Self-reported binge eating and purging frequency 24 h after stimulation

4. No differences between conditions were found.

 Max et al. (2020) [43]

N = 27 with BED

anodal 1 mA tDCS on DLPFC vs. sham tDCS (within-subject) vs. anodal2 mA tDCS on DLPFC vs. sham tDCS (within-subject design)

2 individual sessions

1. food-related antisaccade task (latency, error rate)

1. Significant reduction of error rate over time in all conditions; Latencies were decreased in the 2 mA vs. sham and vs. 1 mA condition.

2. Frequency of binge eating episodes in the past seven days

2. Compared to sham stimulation, the frequency of binge eating episodes decreased at the 2 mA condition over time whereas it did not change significantly at the 1 mA condition.

 Schmidt & Martin (2016) [44]

N = 75 healthy women with subjective binge eating episodes

EEG-neurofeedback with cue exposure vs. mental imagery with cue exposure vs. waitlist

10 individual sessions

1. EDEQ (binge eating episodes)

1. EEG-neurofeedback and MI groups showed decreased binge eating frequency, but this decrease was significant only at EEG-neurofeedback group at post test and 3-months follow up.

2. FCQ (trait version)

2. Food craving was reduced in both EEG-neurofeedback (large effect) and MI groups (medium effect).

 Van den Eynde et al. (2010) [45]

N = 38 adults with BN or EDNOS-bulimic type

High frequency rTMS on the left DLPFC vs. sham rTMS on the left DLPFC

1 individual session

1. Urge to eat, urge to binge eat, hunger on visual analogue scale immediately after stimulation

1. Urge to eat was significantly reduced in real rTMS group vs. sham stimulation. Urge to binge eat and hunger were reduced in both real rTMS and sham conditions.

2. Binge eating frequency 24 h after stimulation

2. Significantly fewer binge-eating episodes over the 24 h following were reported in in real rTMS compared to sham.

3. FCQ (state version)

3. Both groups reduced food craving, and there was no group difference.