Elsevier

Neuroscience & Biobehavioral Reviews

Volume 61, February 2016, Pages 132-155
Neuroscience & Biobehavioral Reviews

Review
Novel methods to help develop healthier eating habits for eating and weight disorders: A systematic review and meta-analysis

https://doi.org/10.1016/j.neubiorev.2015.12.008Get rights and content

Highlights

  • We examine the effectiveness of approaches used to change habitual eating patterns.

  • Implementation intentions have small effect sizes in altering eating.

  • Food-specific inhibition training has a medium effect size in reducing food intake.

  • Attention bias modification has a medium effect size in reducing food intake.

  • Proof of concept work seems warranted in eating and weight disorders.

Abstract

This paper systematically reviews novel interventions developed and tested in healthy controls that may be able to change the over or under controlled eating behaviours in eating and weight disorders. Electronic databases were searched for interventions targeting habits related to eating behaviours (implementation intentions; food-specific inhibition training and attention bias modification). These were assessed in accordance with the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines. In healthy controls the implementation intention approach produces a small increase in healthy food intake and reduction in unhealthy food intake post-intervention. The size of these effects decreases over time and no change in weight was found. Unhealthy food intake was moderately reduced by food-specific inhibition training and attention bias modification post-intervention. This work may have important implications for the treatment of populations with eating and weight disorders. However, these findings are preliminary as there is a moderate to high level of heterogeneity in implementation intention studies and to date there are few food-specific inhibition training and attention bias modification studies.

Introduction

The transdiagnostic term eating disorders covers syndromes with eating behaviours ranging from under to over controlled eating (Fairburn et al., 2003). The most common form of psychological treatment across the spectrum of illnesses is Cognitive Behavioural Therapy (CBT) and the main elements used to change eating behaviours are monitoring (eating and compensatory behaviour diaries) and setting goals for food plans, which include regularly spaced meals. However, even in the optimal conditions of a clinical trial these approaches are only moderately effective and 40–60% of patients remain symptomatic at the end of treatment. In order to help explain this resistance of eating disorder symptoms to change, explanatory models have begun to focus on the role of habit formation in the development and maintenance of psychopathology (O’Hara et al., 2015, Steinglass and Walsh, 2006, Treasure et al., 2014, Walsh, 2013).

A staging model of eating disorders (Treasure et al., 2014) has recently highlighted how eating disorder psychopathology may follow a projected trajectory across the lifespan with symptoms becoming more embedded and complex over time. In the severe and enduring stage of illness it is hypothesised that eating disorder habits are deeply entrenched resulting in neuroprogressive changes and decreased treatment responsivity. Excessive habit formation is proposed to be a mechanism that maintains the compulsive nature of dietary restriction in anorexia nervosa (AN) (Walsh, 2013) and the impulsive/compulsive nature of overeating behaviours in obesity, binge eating disorder (BED) and bulimia nervosa (BN) (Berner and Marsh, 2014, Smith and Robbins, 2013). Parallels have been drawn between the role of habit formation in eating and weight disorders and other impulsive/compulsive disorders (Robbins et al., 2012). Based upon this emerging area of research, the purpose of this review will be to examine the effectiveness of novel approaches for developing healthier eating habits that may be valuable in the treatment of eating and weight disorders.

Gardner (2015) defines habit as, “a process by which a stimulus automatically generates an impulse towards action, based on learned stimulus-response associations” (p. 280). Walsh (2013) proposed that excessive habit formation might be a maintenance mechanism in AN. This model states that dietary restriction is initially goal-directed with a variety of possible aims such as losing weight or managing emotions. Individuals develop fixed dietary patterns and exclude a wide variety of foods from their diet. They may also engage in over-exercising behaviours as a means of weight-loss. During this stage weight loss may be positively reinforced, with individuals often reporting that they received compliments or concern from their peers and family and an increased sense of self-esteem/mastery. However, when maintained over time, dietary restriction may develop into a deeply entrenched habit primarily driven by automatic (stimulus-response rather than goal-driven) processes that are initiated by cues that are both internal (e.g., negative affect, physiological effects) and external (e.g., interpersonal difficulties) to the individual. Over time excessive habit formation may underpin the shift from weight-loss being initially rewarding to becoming compulsive in its nature.

Compulsivity involves the repetitive performance of actions that often result in negative consequences. These actions are typically the result of rigid-rules and are performed as a means of avoiding the perceived negative consequences of not carrying out the action (e.g., strict dietary rules may be followed in AN due to a fear of becoming overweight if they are broken) (Dalley et al., 2011, Fineberg et al., 2014, Fontenelle et al., 2011). Robbins et al. (2012) have advocated for a transdiagnostic approach to compulsivity based upon commonalities in cognitive, behavioural and neural processes across disorders and co-morbidities. This approach suggests that research should focus on transdiagnostic constructs to help aid treatment development rather than on traditional diagnostic criteria.

Comparisons have been drawn between compulsivity in AN and other compulsive disorders such as: substance use disorder, obsessive-compulsive disorder (OCD), and obsessive-compulsive personality disorder (Godier and Park, 2014, Park et al., 2014). For example, the persistent nature of weight-loss in AN has been likened to compulsive drug-taking in substance use disorder as these behaviours both continue despite their detriment to health (Godier and Park, 2015). Theories of habit formation are well established within the fields of substance use disorder and OCD, suggesting that the excessive formation of stimulus-response behaviours leads to compulsive drug-taking behaviours in substance use disorder (Everitt and Robbins, 2015, Pierce and Vanderschuren, 2010) and stereotyped/ritualised behaviours in OCD (Gillan et al., 2014, Gillan and Robbins, 2014). It is possible that excessive habit formation might be a mechanism that underpins compulsivity across disorders (Robbins et al., 2012). Consequently, there is a need for interventions that could help to break the stimulus-response habits that maintain compulsive behaviours.

Impulsivity is multifaceted and predisposes individuals to act without forethought to potential negative consequences (Dalley et al., 2011, Fineberg et al., 2014). The construct is considered to comprise of numerous sub-domains including deficits in inhibiting responses, attention and decision-making (Reynolds et al., 2008). Impulsivity manifests in a range of illnesses such as: substance use disorder (Grant and Chamberlain, 2014), impulse-control disorders (Grant and Potenza, 2006), behavioural addictions (Blanco et al., 2009, Grant et al., 2010, Robbins and Clark, 2015) and Attention-Deficit Hyperactivity Disorder (ADHD) (Lopez et al., 2015). As a result, a transdiagnostic approach to impulsivity has been argued for alongside compulsivity (Robbins et al., 2012). It is thought that both constructs may co-exist within and across a range of disorders and that the balance of the two constructs might contribute to their specific psychopathology (Grant and Kim, 2014, Grant and Potenza, 2006).

The action-to-habit theory of substance use disorder suggests that drug use may initially begin as an impulsive action that becomes compulsive through excessive habit formation (Everitt and Robbins, 2005, Everitt and Robbins, 2015). This action-to-habit theory may also help to explain how compulsive overeating develops (Robbins et al., 2012, Smith and Robbins, 2013). For example, within western societies whereby palatable foods are widely available, impulsivity might predispose individuals towards overeating. Episodes of overeating might become linked with cues (e.g., advertisements/negative affective states) that trigger food cravings and further overconsumption (e.g., binge-eating). This may lead overeating to transition from being an impulsive action to compulsion. Evidence has been found to support the role of habit formation and impulsivity/compulsivity in obesity, BED and BN.

It is possible that impulsivity underlies obesity as individuals may find it harder to resist unhealthy palatable foods (Nederkoorn et al., 2006). Nederkoorn et al. (2006) found that treatment-seeking obese children had less inhibitory control and were more sensitive to rewards than age matched healthy controls. Notably, poorer inhibitory control was found to be associated with less successful weight-loss during treatment (Nederkoorn et al., 2006, Nederkoorn et al., 2007). Similar findings of increased impulsivity and deficits in inhibitory control have also been reported in obese adolescents (Batterink et al., 2010) and young adults (Chamberlain et al., 2015, Jasinska et al., 2012).

Excessive habit formation may lead to the maintenance of overeating behaviours. Hortsmann et al. (2015) recently found through a selective satiation task that a higher Body Mass Index (BMI) in adult males is linked with lower levels of behavioural sensitivity to changes in the motivational value of food. Habitual responding to food cues could be a mechanism leading to behaviours such as late meal cessation and eating in the absence of hunger (Hortsmann et al., 2015). Thus, food cues might prompt obese individuals to overeat paralleling substance use disorder whereby drug related cues can induce compulsive drug-taking (Everitt and Robbins, 2015).

Binge eating is defined by episodes of overeating objectively large amounts of food accompanied by a subjective sense of loss of control. BED involves recurrent binge eating episodes that are associated with feelings of distress and guilt (American Psychiatric Association, 2013). Obesity and BED are strongly connected with obese BED displaying high levels of impulsivity (Schag et al., 2013a, Schag et al., 2013b) and compulsivity (Davis, 2013). Nazar et al. (2014) reported an association between BED and ADHD in a cross-sectional study of treatment-seeking obese women. Inattention symptoms and impulsivity traits were found to be strong predictors of binge eating severity. It is thought that food cues might highly engage the attentional focus of BED patients and decrease awareness on other cognitive processes, thus leaving them vulnerable to impulsive triggers of binge-eating (Nazar et al., 2014, Schag et al., 2013b).

Impulsive binge eating episodes might become compulsive due to excessive habit formation. Voon et al. (2015) used a decision making task to demonstrate that obese individuals with BED show a greater tendency to favour habit-based rather than goal-based learning approaches relative to obese non-BED. Neuroimaging data showed lower orbitofrontal cortex (OFC) and caudate nucleus grey matter volume in obese BED in comparison to obese non-BED. Hence, a bias towards habit-formation and neural deficits might underlie the compulsive nature of obese BED. This is similar to substance use disorder; with research finding that lower grey matter volume in the OFC is associated with a longer duration of illness and greater levels of compulsivity (Ersche et al., 2011).

BN is characterised by binge eating episodes and compensatory behaviours to prevent weight-gain (purging, laxative abuse, over-exercise and dietary restriction) (American Psychiatric Association, 2013). Pearson et al. (2015) have outlined a risk to maintenance model of BN suggesting that binge eating and purging episodes might begin as emotion driven impulsive actions that can become maintained as maladaptive emotion regulation strategies. For example, binge-eating episodes might initially be experienced as rewarding as they help to distract from the experience of negative emotions. After episodes of binge eating, purging might reduce feelings of guilt and discomfort. Over time, these behaviours develop into a means of avoiding anticipated painful emotions altogether rather than distracting from them. This is thought to parallel substance use disorder whereby drug taking may also be a method of avoiding distressing emotions (Baker et al., 2004). Therefore, binge-purge behaviours in BN might transition from being impulsive to compulsive behaviours that serve an avoidant function. In support of this, Engel et al. (2005) have found that higher levels of self-reported impulsivity and compulsivity are associated with greater levels of eating disorder psychopathology, depression and drug/alcohol addictions in community and treatment seeking females with BN.

Taken together, findings from across, obesity, BED and BN seem to support the notion that impulsivity and excessive habit formation are mechanisms underlying the compulsive nature of their psychopathology (Robbins et al., 2012).

Based upon the role of excessive habit formation in eating and weight disorders interventions are needed to focus on creating new healthier habits alongside the disruption of the stimulus-response linkage that underpins maladaptive habits (Lally and Gardner, 2011, Wood and Rünger, 2015). Various novel interventions have recently been developed which interrupt this process by planning (e.g., implementation intentions) or acting through more automatic processes such as changing the attentional processes (e.g., attention bias modification training) or impulsive action tendencies (e.g., food-specific inhibition training) which determine eating behaviour (Quinn et al., 2010, Rothman et al., 2009, Wood and Neal, 2007). These approaches have been drawn from other fields such as substance use disorder (e.g., Cox et al., 2014, Wiers et al., 2013), and may be helpful as novel treatment enhancers for eating and weight disorders (Treasure et al., 2015).

An approach to optimise planning and goal setting for behaviour change is implementation intentions (Gollwitzer, 1999). The aim of this approach is to strengthen deliberate processes of behaviour (van Koningsbruggen et al., 2014) by building counter habits. It involves the creation of action plans that state when, where and which behaviours should be performed in order to achieve a desired goal (Gollwitzer, 1999). For example, ‘if I realise that I am calorie counting, then I will distract myself,’ or ‘if I need to buy a snack from a vending machine, I plan to get a whole-grain fruit bar’. These interventions are part of the motivational phase of behaviour change and are based upon the framework of the Theory of Planned Behaviour (TPB) (Ajzen, 1985). They can be used to either help develop a new healthier response to a situation or to increase self-control over maladaptive habits (Lally and Gardner, 2011).

Adriaanse et al. (2011b) performed a systematic review and meta-analysis of the literature relating to implementation intentions finding that they appear to be a helpful approach for increasing healthy food consumption (d = 0.51) and less so for reducing the consumption of highly palatable foods (d = 0.29). However, this study did not examine the long-term effectiveness of these changes or the impact of this approach on weight change. Furthermore, more studies in this area have recently been conducted meaning that it may be considered necessary to systematically review the effectiveness of this approach further.

Food-specific inhibition training is an approach that involves increasing inhibitory control specifically towards highly palatable foods (Veling et al., 2011a). It involves the use of computerised tasks such as the go/no go task and stop-signal task as a means of inhibiting automatic impulses towards highly palatable foods (Houben, 2011, Veling et al., 2011a). Go/no go training is based upon a choice reaction time paradigm whereby subjects are instructed to respond quickly and accurately to the presentation of a stimulus in the middle of a computer screen. This stimulus is presented alongside either a go or no/go cue such as the letters “P” or “Q” and only appears on screen for a brief period of time. Participants are instructed pre-task to respond to the presentation of a go cue (e.g., by pressing a computer key such as the space bar) and to withhold their response when the stimulus is presented alongside a no-go cue. Outcomes recorded for the go/no go task include reaction times to the stimuli and the accuracy of responses. When participants do not successfully withhold their response for the no-go cue it is indicative of a greater level of impulsivity (Band and van Boxtel, 1999). Through this approach pictures of highly palatable food stimuli can be consistently paired with no-go cues with the goal of increasing self-control towards these items of food.

Regarding the stop-signal paradigm, a similar procedure is followed to go/no go training with participants receiving instructions to respond rapidly to the onscreen presentation of stimuli whilst withholding their response when a stop-signal appears onscreen (e.g., a border around the target stimuli becoming bold). However, the procedure of stop-signal training differs from go/no go training in several ways: (1) participants must respond quickly to the presentation of both neutral and target stimuli onscreen; (2) participants are instructed to inhibit their response for only a proportion of the target stimuli; and (3) there is a variable delay between the presentation of the food stimulus onscreen and the presentation of the stop-signal (Verbruggen and Logan, 2008). The stop-signal paradigm can be used as an assessment of the capability to suppress an already initiated motor response with longer reaction times to the stop-signal suggestive of a higher level of impulsivity and poor inhibitory control (Logan et al., 1997). This approach may also be used to increase inhibitory control towards highly palatable foods and may be of value in the treatment of disorders of overeating such as obesity, BED and BN (Juarascio et al., 2015).

Research has suggested that biases in attention might underlie either under or over eating. For instance, patients with AN have been found to have biases in attention away from highly palatable foods (Veenstra and de Jong, 2012). In populations that overeat attentional biases towards highly palatable foods have been reported (Kemps et al., 2014a, Nijs et al., 2010, Nijs and Franken, 2012, Werthmann et al., 2015). The goal of attention bias modification is to remediate these cognitive biases in attention and to decrease the saliency of the environmental cues that may trigger eating habits.

The attention bias modification approach is computerised and is based upon a modified version of the dot-probe task and can be used to train early orientation styles in attention either towards or away from food or emotional stimuli (MacLeod et al., 1986). To do this, two stimuli appear onscreen either side of a fixation point; one food related, the other neutral. Following this a probe appears (e.g., the letter “E” or “I”) which subjects must respond to quickly by pressing a computer key. To train attention towards food the probe consistently appears in the position vacated by the food stimuli or the neutral stimuli to train avoidance. Attention bias modification training may have potential as a widely disseminable treatment enhancer for eating disorders (Renwick et al., 2013) in either helping to develop healthier food intake or diminishing unhealthy food consumption.

The aim of the present systematic review and meta-analysis of the literature is to examine and compare the effectiveness of methods that have been found to change eating behaviours (i.e., implementation intentions, food-specific inhibition training and attention bias modification training). This is with the overall aim of translating possible new methods into clinical practice.

Section snippets

Literature search

The electronic databases Embase, Medline, PsycINFO using Ovid and Science Citation Index Expanded (1900–present) and Scopus were searched for relevant articles written in English in peer reviewed journals during available years of publication to October 2014 following the PRISMA guidelines (Moher et al., 2009). The keywords used as search terms can be found in Table 1.

Inclusion/exclusion criteria

To be included in the systematic review and meta-analysis, studies were required to meet the following criteria: (1) measured

Overview of included studies

In total 44 studies were included examining the effect of implementation intentions on food intake and weight change. Studies were divided into different groups examining the effect of implementation intentions on; (1) increasing healthy food intake (see Table 2) and; (2) reducing unhealthy food intake (see Table 3). In the present meta-analysis a third group was identified examining the effect of implementation intentions on; (3) weight change (see Table 4). Results are discussed in these

Summary of the results

The aim of this review was to examine the effectiveness of approaches that may be beneficial for developing healthier eating habits in eating and weight disorders. A primary finding from this review is that no studies using these approaches in clinical populations with AN, BN and BED were found in the literature. This was surprising considering the recent formulations of these illnesses based around habit-theory and impulsivity and/or compulsivity (Robbins et al., 2012, Treasure et al., 2014,

Conflict of interest statement

The authors declare having no conflict of interests in the writing of this paper.

Acknowledgements

We would like to give a special thank you to Faisal Jamshaid and Danielle Wilcock for their help in reviewing the articles that were included in this paper. Robert Turton is part funded by the Institute of Psychiatry, Psychology & Neuroscience/Medical Research Council (MRC) excellence studentship and by the Psychiatry Research Trust (PRT) (grant reference number 29 Turton/ Treasure). The third and fifth authors receive salary support from the National Institute for Health Research (NIHR),

References (157)

  • L. Fleig et al.

    Intervention effects of exercise self-regulation on physical exercise and eating fruits and vegetables: a longitudinal study in orthopedic and cardiac rehabilitation

    Prev. Med.

    (2011)
  • C. Gillan et al.

    Enhanced avoidance habits in obsessive-compulsive disorder

    Biol. Psychiatry

    (2014)
  • J. Grant et al.

    Impulsive action and impulsive choice across substance and behavioral addictions: cause or consequence?

    Addict. Behav.

    (2014)
  • J. Grant et al.

    Compulsive aspects of impulse-control disorders

    Psychiatr. Clin. North Am.

    (2006)
  • R. Guerrieri et al.

    Disinhibition is easier learned than inhibition. The effects of (dis)inhibition training on food intake

    Appetite

    (2012)
  • R. Guerrieri et al.

    Inducing impulsivity leads high and low restrained eaters into overeating, whereas current dieters stick to their diet

    Appetite

    (2009)
  • C. Hardman et al.

    The effects of food-related attentional bias training on appetite and food intake

    Appetite

    (2013)
  • K. Houben

    Overcoming the urge to splurge: influencing eating behavior by manipulating inhibitory control

    J. Behav. Ther. Exp. Psychiatry

    (2011)
  • K. Houben et al.

    Training inhibitory control. A recipe for resisting sweet temptations

    Appetite

    (2011)
  • C. Jackson et al.

    Beyond intention: do specific plans increase health behaviours in patients in primary care? A study of fruit and vegetable consumption

    Soc. Sci. Med.

    (2005)
  • A. Jasinska et al.

    Impulsivity and inhibitory control deficits are associated with unhealthy eating in young adults

    Appetite

    (2012)
  • A. Juarascio et al.

    Could training executive function improve treatment outcomes for eating disorders?

    Appetite

    (2015)
  • N. Kakoschke et al.

    Attentional bias modification encourages healthy eating

    Eat. Behav.

    (2014)
  • B. Knäuper et al.

    Replacing craving imagery with alternative pleasant imagery reduces craving intensity

    Appetite

    (2011)
  • E.J. Kothe et al.

    Randomised controlled trial of a brief theory-based intervention promoting breakfast consumption

    Appetite

    (2011)
  • N. Lawrence et al.

    Stopping to food can reduce intake. Effects of stimulus-specificity and individual differences in dietary restraint

    Appetite

    (2015)
  • R. Lopez et al.

    A multidimensional approach to impulsivity in adult attention deficit hyperactivity disorder

    Psychiatry Res.

    (2015)
  • A. Luszczynska et al.

    Planning to change diet: a controlled trial of an implementation intentions training intervention to reduce saturated fat intake among patients after myocardial infarction

    J. Psychosom. Res.

    (2007)
  • O. Mobbs et al.

    Obesity and the four facets of impulsivity

    Patient Educ. Couns.

    (2010)
  • D. Moher et al.

    Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement

    J. Clin. Epidemiol.

    (2009)
  • B. Nazar et al.

    Influence of attention-deficit/hyperactivity disorder on binge eating behaviours and psychiatric comorbidity profile of obese women

    Compr. Psychiatry

    (2014)
  • C. Nederkoorn et al.

    Why obese children cannot resist food: the role of impulsivity

    Eat. Behav.

    (2006)
  • A. Achtziger et al.

    Implementation intentions and shielding goal striving from unwanted thoughts and feelings

    Pers. Soc. Psychol. Bull.

    (2008)
  • M. Adriaanse et al.

    Finding the critical cue: implementation intentions to change one's diet work best when tailored to personally relevant reasons for unhealthy eating

    Pers. Soc. Psychol. Bull.

    (2009)
  • M.A. Adriaanse et al.

    When planning is not enough: fighting unhealthy snacking habits by mental contrasting with implementation intentions (MCII)

    Eur. J. Soc. Psychol.

    (2010)
  • M. Adriaanse et al.

    Planning what not to eat: ironic effects of implementation intentions negating unhealthy habits

    Pers. Soc. Psychol. Bull.

    (2011)
  • I. Ajzen

    From Intentions to Actions: A Theory of Planned Behavior

    (1985)
  • American Psychiatric Association

    Diagnostic and Statistical Manual of Mental Disorders

    (2013)
  • C.J. Armitage

    Evidence that implementation intentions reduce dietary fat intake: a randomized trial

    Health Psychol.

    (2004)
  • C.J. Armitage

    Evidence that implementation intentions promote transitions between the stages of change

    J. Consult. Clin. Psychol.

    (2006)
  • C.J. Armitage

    Effects of an implementation intention-based intervention on fruit consumption

    Psychol. Health

    (2007)
  • C.J. Armitage

    Evidence that self-incentives increase fruit consumption: a randomized exploratory trial among high-risk Romanian adolescents

    Prev. Sci.

    (2014)
  • T. Baker et al.

    Addiction motivation reformulated: an affective processing model of negative reinforcement

    Psychol. Rev.

    (2004)
  • C. Begg et al.

    Operating characteristics of a rank correlation test for publication bias

    Biometrics

    (1994)
  • Y. Benyamini et al.

    A structured intentions and action-planning intervention improves weight loss outcomes in a group weight loss program

    Am. J. Health Promot.

    (2013)
  • L.A. Berner et al.

    Frontostriatal circuits and the development of bulimia nervosa

    Front. Behav. Neurosci.

    (2014)
  • M. Boffo et al.

    Combining cognitive bias modification training with motivational support in alcohol dependent outpatients: study protocol for a randomised controlled trial

    Trials

    (2015)
  • M. Bradburn et al.

    Metan—an alternative meta-analysis command

    Stata Tech. Bull.

    (1998)
  • S. Chamberlain et al.

    Obesity and dissociable forms of impulsivity in young adults

    CNS Spectr.

    (2015)
  • J. Chapman et al.

    Evidence that boosters augment the long-term impact of implementation intentions on fruit and vegetable intake

    Psychol. Health

    (2010)
  • Cited by (99)

    View all citing articles on Scopus
    1

    Joint last authors.

    View full text