After 21 years of friendship, my dearest friend collapsed from a series of heart attacks. A week later, with no evidence of brain activity, her family made the heart-wrenching decision to turn off her life support. She was 39 years old. I gained some comfort from knowing that she had been released from a lifelong battle with food; it had been her comfort, her reward, her passion, her distraction, her enemy and the place she returnedto manage any and all her feelings.

Following a diagnosis of diabetes aged 14, my friend had rarely given her body the nutrition it needed. Instead, she would binge on sugar and refined carbohydrates most days, often abusing her insulin to rapidly lose weight through intentionally poor blood sugar control. By signing up to the latest diet plan, she repeatedly attempted to regain control, but consistently returned to the familiar, yet harmful, food behaviours. Our university years were marked with sudden but preventable distress, as she suffered regular, hypoglycaemic episodes, where her blood glucose level dropped dramatically, through poor management of her condition.

For much of her adult life, my friend’s weight fluctuated and her diabetic body suffered. Her health, so often sacrificed for her appearance, and her food, recruited to change her mood, brought grim consequences to every layer of her life. She developed diabetic retinopathy, cataracts, deep vein thrombosis and, eventually, kidney failure. She struggled with increasing anxiety, bouts of depression and a whole spectrum of eating disorder traits. Her final year of life was largely spent in and out of hospital. Neuropathy meant minor injuries quickly progressed to more serious medical problems, her wounds failing to heal due to poor blood circulation; one very distressing admission for diabetic foot ulcers had barely avoided foot amputation. She began dialysis and waited for a kidney and pancreas transplant. When the call finally came, the organs were unsuitable. It was devastating news.

The diabetic complications couldn’t be undone. I recall the moment shortly after discharge from yet another gloomy hospital visit, she repeated the doctor’s morbid warning about the increased probability of her early passing, most likely before we could celebrate our 50th birthdays. I have often wondered if, in that moment, she shared my fear her life would end far sooner; in fact, she died only a few months short of her 40th birthday.

About six months after my friend passed away, following a period of personal therapy and a significant midlife appraisal, I left a 20-year career in television and decided to retrain as an addiction therapist. It felt like the right time to embark on, for me, a more meaningful vocation, one which provoked uncomfortable self-exploration, multiple turning points and deeply rewarding work.

I find myself, 10 years later, identifying as an attachment-based therapist, specialising in addiction and dependency, resulting from an integration of two, distinct models of therapy: an abstinence-based addiction model and psychotherapy informed by attachment theory. Tragically, the nutritional-based interventions alone could not save my beloved friend. As my clinical work continues to provoke memories of her struggles, I wonder if awareness of her attachment insecurity could have provided an alternative insight into her dysfunctional eating that might have extended her life.

I would like to take a single case study (a female binge drinker and compulsive overeater) to describe my shift from an addiction model of treatment to an integrated, attachment-based framework for understanding and working with recovering addicts, whose dysfunctional food behaviour is maladaptive – and not dissimilar, in process or motivation, to substance misuse or alcoholism. I will share the challenges and small triumphs in helping my client surrender patterns of self-harm, while explaining why the mere avoidance of particular foods or cessation of damaging food behaviour were simply not enough to bring about lasting change. I highlight why, in my experience, the use of behavioural therapy within an addiction model of recovery has an enhanced capacity to heal and repair, when considered alongside the client’s attachment functioning.

In sharing a fragment of my work with this specific clinical population, I demonstrate the link between attachment insecurity and relapsing behaviour, and between a poor working model of self and a reluctance to pursue more constructive behavioural coping strategies. I also demonstrate why, in my experience, the client will not make the best use of therapy and a recovery community as a source of soothing and support, if they have little or no awareness of their attachment style. Additionally, I illustrate why an absence of behavioural skills training within addiction-focused, attachment-based work will limit clients’ learning, restrict their capacity for positive change and therefore impede their progress towards a more secure state.

I first met 28-year-old Ava when I was working as a therapist for a 12-step, abstinence-based, outpatient addiction programme. Ava stated that her drinking and overeating were out of control and she was ‘done with being fat’. She described her mood as ‘sad and anxious, for as long as I can remember’, with occasional panic attacks. She felt her ‘long-suffering boyfriend’ of eight years was a victim of ‘false advertising’, having met her when she was ‘a young and thin trainee doctor’, but who now found himself with a ‘failed, obese care assistant’.

Ava was under the care of a psychiatrist and had been taking medication for depression and anxiety for four years. She had attended weekly, psychodynamic psychotherapy for the past two years and was able to offer some understanding of the impact of her childhood on her adult choices. She listed the many weight-loss plans she had tried since early adolescence, only to regain the lost pounds – and more – with each attempt. Despite significant financial and emotional investment throughout her adulthood, she had not been able to overcome the destructive behaviour around food and alcohol that now dominated her life.

Ava easily recalled evidence to demonstrate her difficulty moderating alcohol. There were multiple stories of regrettable and risky behaviour resulting in black-out, which had been normalised in her friendship circle, yet which consistently resulted in heightened anxiety and shame. Ava was a binge drinker, both socially and alone at home, and she did not deny her compulsion to drink and struggle to moderate alcohol. Ava was nevertheless able to surrender alcohol from day one, with only a few wobbles, which was unusual in my experience. But while counting sober days gave her a sense of shared achievement with her programme peers, it also provoked disappointment that minimal change had occurred with her food behaviour. In fact, letting go of alcohol seemed to inflame her desire to overeat. She quickly identified that alcohol was not her primary prop to manage her feelings: that badge was unquestionably awarded to food.

It felt relevant to me to make sense of Ava’s historical dependency on food, but she was clear the primary goal for her treatment was behavioural change. She wanted both to lose weight and to acquire the skills to maintain a ‘right-size body’. But we agreed to explore what might provoke the cravings and why none of the slimming programmes had given her the long-term weight loss for which she yearned. If recovery was simply skills training, why had Ava not managed the change to which she desperately aspired?

Ava could track her compulsive eating back to her early childhood, when her father was largely absent from the home and she was subjected to her mother’s unpredictable mood swings and violent outbursts, which were later identified as symptoms of severe premenstrual tension. Her mother targeted Ava, as the ‘bad, fat daughter’. Ava would fail to dodge slaps and punches, while heeding a familiar and repeated message: ‘I hate you; you’ve ruined my life; I’m leaving.’ Ava was acutely aware of her childhood anxiety and desire to cling to her mother or grandparents to get her security needs met. She remembered feeling distressed by separations from her family members and her struggle to feel soothed when reunited. Ava had many examples of her mother’s misattunement, which often resulted in food-seeking behaviour to ease her anxiety and fear.

Chronic anxiety in childhood meant Ava had struggled and failed to develop an internal sense of security. She was starting to make sense of her relentless and unsatisfying search for safety in food. A mourning process needed to begin for the maternal attunement she had yearned for but never received. Each time Ava felt let down by another disappointing interaction, which led to a shameful return to overeating, I reminded her that unrealistic expectations of love or soothing from either her mother or food was an old fantasy, one that she might begin to surrender. It would be neither a quick nor a linear grieving process. Repeatedly,Ava would return to her mum and excessive food consumption in search of soothing.

In order to forgive her mother and surrender any expectations, it felt appropriate to make sense of her mother’s abusive and sometimes violent behaviour in Ava’s childhood. A mother’s own unresolved trauma might interfere with her ability sensitively to respond to her child, affecting the development of a secure attachment and potentially contributing to the intergenerational transmission of trauma. Ava knew that being slim and in control of one’s food was revered in her family; to be out of control was shameful. As an adult, Ava often reported feeling scrutinised by her mother, sensing her disapproval as she ate. Continuing to overeat was an act of rebellion, an expression of anger at her mother’s judgment and attempt to control Ava’s food. Overeating in her mother’s presence was an indirect communication of her autonomy.

We observed a similar pattern with her boyfriend, when Ava felt dismissed. How could Ava protect herself emotionally from their hurtful judgments, not reach for excess food and still manage openness, warmth and empathy in her relationships? It was an ongoing and deeply uncomfortable piece of work for Ava to assert and maintain a separateness by expressing her feelings and needs verbally, rather than with defiant, regressive behaviour, and by recruiting internal sources of comfort that were not an assault on her body.

The ability flexibly to regulate emotions in a context-appropriate manner is a distinct trait of secure attachment. Together, we noticed how Ava might recruit different emotional regulation strategies in different situations, such as self-blame, tolerance, criticism, rumination, obsession, shame, binge drinking or emotional eating. Of course, affect regulation and distress tolerance skills could rarely be neatly assigned to a particular point in a particular session. There was a need for flexibility and spontaneity from both myself and Ava when she began to dysregulate. To prevent her rapid slide into an unhelpful, emotionally negative place, which would inevitably result in overeating, I would appeal for a pause, to draw her towards a more cognitive state. We would frequently wonder if the detail she was so intent on sharing was serving her emotional stability or risked trapping her in unsteadying rumination. We both recognised that anything unsettling would precede overeating; hence, slowing down the session to create a break in that cycle was appealing. We recruited these opportunities to coach Ava into consciously activating her parasympathetic nervous system (the calming, restoring part), using specific skills and techniques to dissipate intense feelings, rather than automatically reaching for food to numb, escape or sedate herself until the feelings passed.

A mandatory aspect of the treatment programme was thrice weekly attendance at 12-step fellowship meetings, such as Alcoholics Anonymous and Overeaters Anonymous, where Ava was encouraged to seek support from others who were following an abstinence-based, addiction recovery programme. Evidence1 demonstrates that this community extends therapeutic support and provides a secure base to return to each week, to identify and share recovery experiences. When planning a binge and unable to access internal resources to manage the craving, Ava was encouraged to reach out to a fellowship peer who might empathise and help her to mentalise, or remind her to recruit the calming techniques. When Ava was infused with agitation or sinking in sadness, a peer might be hopeful. When Ava felt wobbly, a peer might steady her. But, inevitably, Ava experienced days when others were unavailable, unreliable or failed to provide the stability she needed in that moment, triggering historical responses to misattunement that resulted in ambivalence about attending these meetings.

Ava had a tendency to focus on what was missing and block any nourishing interventions. It was noticeably jarring for me when, in these moments, I was unable to penetrate her wall of disillusionment and defence. It often left me feeling frustrated and inadequate. Ava’s negative bias and mistrust of others threatened her connection with members of the 12-step community and therefore threatened another layer of support in her recovery journey.

In my experience, clients with traits of avoidant attachment found the 12-step meetings deeply uncomfortable, as they witnessed the emotional honesty of others. People with a more anxious style, such as Ava, demonstrated a greater desire to attend, but often reported feeling preoccupied with relationship dynamics within the community. Attachment theory2 suggests their withdrawal from the meetings or relational anxiety is an expression of their insecure attachment. A greater understanding of Ava’s attachment functioning – her fears of either intimacy or rejection – allowed us to make sense of, anticipate and counter her emotional responses provoked by the meetings.

While attendance at the meetings was indeed provocative, the interpersonal challenges provided an abundance of opportunity to practise mentalising skills, emotion regulation skills and social skills of honest communication, tolerance and reliability. Put simply, the recovery community presented an ongoing, experiential opportunity for Ava to try out new, ‘secure’ ways of thinking and behaving. Our sessions gave Ava an opportunity to unpack and make sense of her responses to the meetings, while taking risks in changing her old and unhelpful, insecure coping strategies.

Following completion of my attachment-based training, I felt an increasing dissonance between the pure addiction model of the 12-step treatment centre where I was employed and my attachment-based perspective. My experience highlighted the limitations in treating addictive symptoms as one disorder. If attachment insecurity was still active and untreated, it would interfere with the successful treatment of addiction and issues of dependency.

Without doubt, working solely within an addiction model, a task-focused, time-limited approach, had delivered results. It provided behavioural techniques to better manage affect, access to a support community and, for most, a significant shift in personal responsibility and a positive reworking of their narrative.

Yet in my first few years of working purely within an addiction model, I observed multiple episodes of substance and behavioural relapse. I witnessed many clients lose their sobriety or abstinence and return to the 12-step programme for a second, third or fourth treatment episode. It was, of course, distressing and disheartening for my clients. But I also felt frustrated, confused and questioning. Why was this happening, and how might I better support my clients in their bid to break their dependencies? Despite busting denial, developing motivation and teaching new skills, their reluctance to consistently recruit these skills when in need was baffling. Hearing an addicted client labelled as resistant, reluctant, defiant or a hopeless case in clinical discussions deeply frustrated me and did not make sense. Why would they seek admission to an addiction treatment programme, invest time, step away from their daily lives, then reject the step-by-step behavioural treatment being offered?

I began to question whether a therapeutic model of addiction alone could offer long-term rehabilitation, as well as short-term restoration. Attachment-based training provided the missing piece of the addiction treatment puzzle. Clients’ pervasive sense of inadequacy and their shame in finding themselves an ‘addict’ both had to be addressed in order to positively impact behavioural change. Attachment theory3 demonstrated how an insecurely attached person with low self-efficacy, due to their internal working model, is likely to avoid accepting challenges for fear of failure. Ava simply did not believe she was capable of succeeding, and consistently framed any perceived failures and adversities as personal shortcomings.

In my view, participating in an addiction treatment programme is only the beginning of recovery. If clients are curious and willing to engage in longer-term attachment work, they can begin to understand the root causes of their dysfunctional behaviours. Without awareness of their attachment history, a maladaptive aspect of themselves is left active, serving only to hinder their recovery process.

I yearned to offer my clients greater flexibility in the pace and length of the treatment in order to explore attachment history and attachment functioning in greater depth, with a focus on how this might influence their dependency issues and capacity to recover. So, after several years in addiction treatment centres, it felt time to step away and create a private practice in which I could integrate both models of training, specialising in addiction and dependency. I continued to work with Ava in weekly, face-to-face sessions.

Increasingly, I recruited attachment theory to assess clients, to frame their early developmental experiences, to help them make sense of their lifelong thinking and behaviour, particularly when under stress, and I used this theory to gain a fresh perspective on my relating style in the consulting room. Reviewing a client’s attachment history in an extended assessment helped me to foresee the level of care an individual might demand, depending on their attachment functioning. I could therefore consider if I had the availability and emotional capacity at that point to support their needs or whether I could instead make a referral, resulting in better care for the client and a conscious maintenance of my professional resilience. It was liberating and hugely enhanced my clinical work. Attachment theory had become a theoretical anchor, a base camp for clinical exploration.

Many of us can relate to Ava, scouting unnecessary food to alter mood, to ease the impact of unwanted news, to reward, to compensate, to soothe loneliness, to rebel, to indulge with friends when creating connection. Without doubt, food indulgence was a pleasurable activity shared with my diabetic friend. It’s culturally accepted to give and receive food to celebrate, to sympathise, as an expression of thoughtfulness. We all need to eat, and while for some there is a mild to moderate temptation to overindulge, for others – like my friend and Ava – a dependent relationship with food is shamefully destructive, a symptom of something far more complex than bad habits or gluttony.

Why did Ava have such a hard time following through her desire to change? I think it’s because the change required far more than breaking a habit. It demanded attachment-focused work to support Ava in shifting her thinking, building self-belief, nurturing self-compassion and learning to navigate her feelings without resorting to overeating. It also required the capacity to make sense of others’ behaviour (mentalisation), to challenge her expectations of others, to surrender an old coping strategy that no longer served her and – crucially – to forgive and grieve an unyielding mother, so she might no longer seek reassurance from someone who did not have the capacity to provide it. The change involved enlisting Ava’s developing internal parent. It required courage to drop her guard and expose her vulnerability, when she might turn to supportive relationships rather than salty and sugary foods, when she might seek connection rather than chocolate. It required reconfiguring her ‘internal working model’.2

For 28 years, Ava had recruited food to manage her anxiety, control her fury and soothe her sadness. Food helped minimise her social fears, temporarily escape body shame and numb her reality. Binge drinking and overeating were symptoms of Ava’s insecure attachment functioning, substitutes for human support and self-care gone wrong. When her attachment needs were activated, so too was a powerful urge to eat. And in a dysfunctional way, this had helped her survive for almost three decades. For Ava, like all her programme peers, change meant loss and lamenting the old way of living. It was unsettling and slow. The behavioural change was slippery, engaging with the recovery community was fraught with uncertainty, and acting ‘secure’ demanded huge discomfort and risk. The result, not uncommonly, was a vacillation between dependency and recovery, an expression of ambivalence towards change and the journey towards ‘earned security’.3

Importantly for Ava, her impulsive and compulsive eating began to lessen; her weight initially stabilised, then slowly began to diminish. Enlightened, she became more alert to self-harm, both behavioural and relational, to make better informed choices about what she needed and how she might meet those needs, both internally and externally. She began to flourish, with increasingly satisfying human connections rather than disappointing fast-food fixes.

Our work together has brought improvement in many aspects of Ava’s life, but I cannot fail to share the regressions over the past three years. Unsurprisingly, some of life’s stressful events (illness, job change, pregnancy, bereavement, COVID-19) activated her attachment system and old coping strategies triggered extended periods of relapse. Under difficult circumstances, Ava returned to overeating, demonstrating that change is not straightforward and feeling secure is pivotal in manoeuvring life’s inevitable curve balls.

For me, these past six years in private practice have brought an abundance of evidence supporting an integrated, attachment-based approach when working with addiction and dependency. Sensitivity to both my client’s and my own attachment functioning has allowed me to aim therapeutic interventions at key factors that maintain the addictive behaviour and complicate the therapeutic relationship. The dual focus on behavioural change and attachment functioning has significantly improved outcomes in my clinical work. For my clients, it has brought a shift from acute, short-term treatment to sustained recovery management. It has given me an evolving confidence and belief in my ability to help liberate my clients from the despair of dysfunctional eating.

From Attachment, Relationships and Food: from cradle to kitchen, 1st edition, edited by Linda Cundy and published by Routledge, 2021. ©2022. The edited extract was reproduced by permission of Taylor & Francis Group.


1 Greene D. Revisiting 12-step approaches: an evidence-based perspective. [Online.] Intechopen; 2021. (accessed February 2022).
2 Bretherton I, Munholland KA. Internal working models in attachment relationships: elaborating a central construct in attachment theory. In: Cassidy J, Shaver PR (eds). Handbook of attachment: theory, research, and clinical applications. Guilford: The Guilford Press; 2008 (pp102–127).
3 Odgers A. From broken attachments to earned-security: the role of empathy in therapeutic change. Abingdon: Routledge; 2018.