Current evidence base

Studies conducted into restrictive eating have tended to focus mainly on Western Caucasian women; however, data are increasingly indicating that it impacts on all populations but may have a different aetiology and/or symptom profile. ^{Reference Halbeisen, Brandt and Paslakis3} Self-harm may be more prevalent among marginalised and under-researched groups, such as Lesbian, Gay, Bisexual, Transgender, Questioning, Intersex and Asexual+ (LGBTQIA+) youth and children in care. ^{Reference Varley, Fenton, Gargan, Taylor, Taylor and Kirby4} There is a risk that current diagnostic frameworks and treatment models disproportionately mischaracterise these groups, reinforcing existing health inequalities. Wang et al ^{Reference Wang, Fox, Boccagno, Hooley, Mair and Nock5} also expressed concern that the function of restricted intake needs to be considered, including in those with disordered eating who do not meet a diagnostic threshold. The empirical literature on restricted intake as a form of self-harm remains very limited. Only a small number of studies over the past 25 years have examined restrictive eating behaviours in relation to self-harm. ^{Reference Claes, Norré, Van Assche and Bijttebier6–Reference Washburn, Soto, Osorio and Slesinger10}

These studies consistently indicate that both eating disorder behaviours and self-harm can function as maladaptive strategies for emotion regulation, often involving intentional physical harm or distress. Fox et al ^{Reference Fox, Wang, Boccagno, Haynos, Kleiman and Hooley7} identified associations between restrictive eating and self-harm intent, suggesting that, for some individuals, restriction is enacted with the explicit purpose of causing harm.

Emerging quantitative evidence supports the clinical relevance of this phenomenon. Washburn et al ^{Reference Washburn, Soto, Osorio and Slesinger10} identified that 29.5% of adolescents and young adults in psychiatric care reported engaging in eating-related behaviours with self-injurious intent, conceptualised as self-injurious disordered eating symptoms. This group demonstrated greater psychopathology, poorer quality of life and increased functional impairment compared with those engaging in non-suicidal self-injury alone.

Similarly, Mahl et al ^{Reference Mahl, La Cour and Skandsen8} found that approximately 20% of high school students reported restrictive eating as a form of self-harm at some point. Together, these findings suggest that RISH behaviours could be relatively common among young people. However, the evidence base lacks qualitative exploration of lived experience and does not explicitly differentiate the function (self-harm versus weight/shape control) described in RISH.

Together, these studies ^{Reference Claes, Norré, Van Assche and Bijttebier6–Reference Washburn, Soto, Osorio and Slesinger10} and the PPI conducted by this author have framed restricted intake as serving different functions, including as a way of regulating overwhelming emotional experiences, as a form of self-punishment in response to negative thoughts or as a means of communicating distress, eliciting care or creating safety from those around them. In PPI sessions, Katy, an adult who identifies as having struggled with RISH episodes, stated: ‘My underlying need was for it to be understood I wasn’t OK, and restriction resulted in support and care.’ One young person with RISH discussed how they would acutely restrict their intake to avoid contact with a relative whom they described as emotionally abusive.

Conceptual clarification: RISH and anorexia nervosa

A key contribution of the RISH construct is the clarification of functional intent. Anorexia nervosa is primarily characterised by restricted intake due to weight and shape concerns, with ICD-11 stating that ‘body weight or shape is central to the person’s self-evaluation’. ¹¹ By contrast, RISH is not characterised by the primary aim of the restriction being weight or shape concern, but by the intent to self-harm through restriction of food and/or fluids. In ARFID, restriction is typically related to sensory sensitivities, fear of aversive consequences or limited interest in eating, rather than a desire to inflict harm. Importantly, RISH is not mutually exclusive with eating disorders and may co-occur. Likewise, RISH can occur in anxiety disorders, depression or psychosis, but these presentations may also involve other forms of reduced eating or drinking secondary to overwhelming affect, dissociation or other functions rather than as a form of self-harm. As with other forms of self-harm, careful assessment of intent, function and associated risk remains essential to differentiate these overlapping clinical presentations. In such cases, where there is a co-occurrence of RISH with a diagnosed mental health condition, clinicians should continue to follow the appropriate treatment guidelines for the diagnosed disorder while also recognising and addressing the self-harm function of restrictive intake, because this may require adapted risk management strategies and therapeutic approaches.

Misclassification of RISH could have iatrogenic consequences when treated solely within either an eating disorder or self-harm framework, as described in Simpson’s article on RISH in National Health Service (NHS) settings. ^{Reference Simpson and Gillings12} In an eating disorder framework, the strong focus on nutrition and weight restoration may inadvertently reinforce the behaviour, particularly if care becomes contingent on restriction. Conversely, a self-harm framework rarely involves input from dietetics and physical health monitoring, as described in the National Institute for Health and Care Excellence (NICE) eating disorder guidance, ¹³ which is required to ensure safety during a RISH episode.

Clinical and research implications

The RISH construct raises several critical clinical and research questions. First, there is a need for improved identification. Current terminology such as disordered eating lacks specificity and may obscure risk. ^{Reference Avent, Bryant-Waugh, Cardina, Chapman, Charalambous and Cini14} Young people involved in this work explicitly described such terms as invalidating. Second, treatment models require adaptation. Interventions developed for anorexia nervosa may be insufficient or even counterproductive when applied to RISH episodes without adaptation. NICE ¹⁵ recommends interventions for self-harming behaviours but does not describe a treatment approach for RISH specifically. Although following NICE guidance for self-harm would be important in RISH, this population also often requires input from dietetics and those with expertise in monitoring and managing restricted intake.