Disabilities and Eating Disorders: A Theoretical Model and Call for Research

Abstract

Objective:

There is an urgent need for research on eating disorders among individuals with disabilities. This paper highlights the lack of research on the relationships between disabilities and EDs, despite their common convergence.

Method:

In this paper, we aim to 1) highlight the need for further research investigating the relationships between disability status and EDs, 2) describe existing frameworks for conceptualizing disability, 3) utilize such frameworks to propose a novel theoretical model of ED/disability relationships and related sociocultural factors and 4) identify future directions for research in this area.

Results:

We propose a multidimensional theoretical model of the relationships between EDs and disabilities. Further, we describe how these relationships are likely influenced by a system of individual factors (e.g., disability (in)visibility, food access, and self-identification) and sociocultural factors (e.g., ableism/discrimination and weight stigma).

Discussion:

Scholars are encouraged to test our proposed model and further investigate experiences of disability and ED co-occurrence with participatory research and mixed-methods designs. ED prevention and screening programs, as well as treatment access and efficacy, need to be evaluated for disabled populations. Disability should also be routinely collected as a demographic across studies, and ED measures should be validated and/or developed for individuals with disabilities.

1. Introduction

Eating disorders (EDs) are severe psychiatric conditions characterized by dysfunctional or distressing eating behavior (e.g., restriction, binge eating, or compensatory behaviors; American Psychiatric Association, 2022). Examples include binge eating disorder, bulimia nervosa, anorexia nervosa, avoidant/restrictive food intake disorder, and other specified feeding or eating disorder. Contrary to colloquial beliefs and stereotypes about eating disorders (Pike et al., 2013; Sonneville & Lipson, 2018), individuals with marginalized identities experience EDs at higher rates compared to individuals with majority identities (Calzo et al., 2017; Gordon et al., 2021; Sonneville & Lipson, 2018). Furthermore, the presence of multiple marginalized identities further compounds ED risk (Beccia et al., 2021; Burke et al., 2022). Disability is also a marginalized identity, as people with a disability are frequently rendered “inferior” to non-disabled people, but also disabled by being denied social and economic mobilization (Dirth & Branscombe, 2018). However, to date, disability-related ED research has predominantly focused on the societal impact of EDs and/or the conceptualization of EDs as disabilities (e.g., Erskine et al., 2016; van Hoeken & Hoek, 2020). As such, little work has examined disability and its relationship with EDs, despite case study evidence of their co-occurrence (Cicmil & Eli, 2014). In effect, the actual mechanisms underlying the association between EDs and disability status are under-studied, representing a critical knowledge gap that must be filled to improve ED prevention and treatment and reduce health disparities for individuals with disabilities.

This paper outlines current models of disability to situate our proposed theoretical model which describes putative indirect and bi-directional relationships between EDs and disabilities. Thus, we propose that disability and EDs may result separately from shared influences, or contribute to the development or maintenance of each other. We also propose directions for future research including considerations to improve diagnostics, assessment, prevention, and treatment for individuals with EDs and disabilities.

2. Dominant models of disability

There are several conceptual models developed to define disability (see Table 1). The medical model conceptualizes disability as a pathology that must be identified and treated to provide ‘relief’ from impairment (Burton, 1983; Williams-Findlay, 2023). Given this model’s focus on deficits, it has been critiqued as “insufficient” and “oppressive” towards disabled populations (Hogan, 2019). Nevertheless, the medical model of disability remains commonly used in research and practice. Conversely, the social-environmental model of disability considers the ways that environments, rather than individual impairment, cause the challenges experienced by disabled individuals. This model conceptualizes disability as a social position and identity marginalized by ableism (Dirth & Branscombe, 2018).

Table 1.

Dominant models of disability.

	Medical Model	Social-Environmental Model	International Classification of Functioning, Disability, and Health
Disability definition	Disability results from impairment that should be identified and treated to accord with ‘normative functioning.’	Disability results from a mismatch between social/environment context that an individual is situated and the needs of the individual.	An individual’s disability is influenced by their body, their body’s functional abilities, their ability to execute tasks and participate in society, and their environmental context.
Example: Physical/visible/mobility-related disability	An individual is not able to enter a business primarily because they do not have legs that will carry them up the stairs. They cannot work at that location. Changing the accessibility of the location is beyond the scope of the medical model. For some individuals, physical therapy may seek to enable walking.	They are not able to enter a business building primarily because the entrance is only accessible by stairs (e.g., no ramp). They cannot work at that location. Changing the accessibility of the location would remove barriers.	They are not able to enter a business due to multidimensional relationships with leg function, difficulty moving and walking, and building design barriers. Their inability to work at that location interacts bidirectionally with these other factors. Individual functioning and disability are both measurable.
Example: Invisible disability (e.g., neurodivergence)	An individual is not able to take part in political events primarily because they communicate non-verbally. For example, speech therapy is used to ‘improve’ their capacity to communicate.	They are not able to take part in political events due to the event’s reliance on verbal communication methods. Since a person’s communicative needs are seen as unique differences, instead, their socio-environments should be adapted, such as including alternative forms of communication (e.g., closed captions).	They are not able to take part in political events due to a multidimensional relationship with their communication, others’ communication, and the event’s design. A person’s communicative needs are a symptom of a disability which is measurable and subject to therapy, but is targeted with environmental factors (e.g., education) and individual strengths (e.g., attention to detail) in mind.
Example: Eating Disorders	An individual is not able to perform at their normal capacity at work because they are unable to focus and preoccupied with thoughts of food. They often avoid business lunches because coworkers comment on food and engage in diet talk, which limits their networking opportunities. Managing the symptoms that cause them to avoid the lunches would be the key focus of treatment.	The individual is not able to take part in business lunches. Therefore, the individual cannot participate fully in work opportunities. Holding meetings that do not involve food, or focusing conversations on other topics, would enable them to participate equally with their colleagues.	The individual is not able to take part in business lunches due to multidimensional relationships with food-related triggers, difficulty handling stress and other psychological demands, and the social environment around food at work. Their lack of full participation in work opportunities interacts bidirectionally with these other factors. Individual functioning and disability are both measurable.
Example: AN/Autism Co-Occurrence	An individual is not able to celebrate holidays and special occasions with their family because they cannot eat the holiday meals. They avoid these meals because the foods trigger their sensory sensitivities. Due to social communication “impairments” and pathological autistic/ED thought patterns, they are anxious about responding to family members asking them why they are not eating. Treatment foci may include exposure to different foods, anti-anxiety medications, and social skills training.	An individual is not able to celebrate holidays and special occasions with their family because these events all focus on food and socializing. Planning celebratory events that do not involve food would enable them to join with other family members. Having a quiet space to retreat to during a large family event, or planning 1:1 activities, instead of large group activities, could give them more chances to engage in ways that are pleasant for everyone.	The individual is not able to celebrate holidays and special occasions with their family due to multidimensional relationships with sensory perception, food-related triggers, difficulty handling psychological stress/demands, and the social environment at celebrations. Their lack of full participation in special occasions interacts bidirectionally with these other factors; for example, impacting the level of family social support that can be enabling in all life domains. Individual functioning and disability are both measurable.

Efforts to integrate the medical and social-environmental models have resulted in the development of biopsychosocial, dimensional frameworks (Hogan, 2019; Van Der Veen et al., 2023; Wehmeyer, 2021). One such framework is the International Classification of Functioning, Disability, and Health (ICF; World Health Organization, 2001), a multidimensional interdisciplinary model that describes disability as resulting from the interaction(s) between presented health conditions (e.g., epilepsy), individual characteristics (e.g., cognition), and the environmental/social context (e.g., poverty). Nevertheless, prevailing research operationalizes disability through the medical model without integrating critical social-environmental factors that contribute to both the development and maintenance of disability. Thus, our working definition is built from a biopsychosocial, multidimensional framework wherein we propose that social-environmental influences, in addition to individual-level functioning, contribute to disability.

3. Theoretical model

Drawing on an ecocultural model (Weisner, 2002) and dominant models of disability, we highlight how intrapersonal, interpersonal, socio-cultural, and economic levels interact to construct individual experiences. We propose a multidimensional model of disabilities and EDs wherein EDs and disabilities may be both indirectly and bi-directionally related (Figure 1). Thus, disability status and EDs may result separately from these shared influences, or reciprocally, wherein one contributes to the development or maintenance of the other. Political/economic systems (e.g., capitalism) and systemic discrimination (e.g., ableism) perpetuate oppression of disabled individuals through segregation and exclusion from society and contribute to health disparities (e.g., increased ED rates among disabled people; Russell, 2019; Williams-Findlay, 2023). Intersectional structural factors like state-level structural sexism (e.g., gendered wage gaps, abortion bans) are also associated with increased risk of certain ED behaviors (Beccia et al., 2022) that may particularly affect disabled individuals who already experience barriers to healthcare and wage-earning. Systemic structures also affect individual/interpersonal influences: for instance, geography (e.g., urban versus rural environments; Douthit et al., 2015) and social support (Aitken et al., 2017) can influence access to healthcare, education, and employment (Sakellariou & Rotarou, 2017). Collectively, these influences are associated with an increased likelihood of ED development and maintenance through decreased treatment access (Cachelin et al., 2001; Hotte, 2018). Two individual or interpersonal factors – disability diagnosis identification and the visibility of a person’s disability – likely interact to affect the association between EDs and disability. Although individuals may meet disability criteria via eligibility for social security disability income, they may not identify as disabled (Tierney, 2001). Not identifying as disabled could be protective if one views their ED as treatable, or harmful if they negate the seriousness of their ED (Tierney, 2001). A disability’s visibility may impact that person’s relationship with their body and food. On a sociocultural level, one may experience more ableism when their disability is visible, which may result in more individual-level disordered eating behavior as a coping mechanism (Cicmil & Eli, 2014). The experience of ableism, like any discrimination, is highly contextual and intersectional. The minority stress model (Meyer, 2007) can be applied to explain how ableism, negatively impacts health and therefore may contribute to both disability and ED risk, across social identities (e.g., Botha & Gillespie-Lynch, 2022).

Figure 1.

Proposed Theoretical Model of the Bidirectional Relationship Between Disability and Eating Disorders

People with disabilities also may experience structural barriers to food access that may account for some disparities in ED prevalence (Schwartz et al., 2019). Disability is associated with increased risk of poverty, which limits the quantity and types of food available (Hughes & Avoke, 2010). Furthermore, the risk of food insecurity may be compounded by living in under-resourced areas with few or no grocery stores and limited access to fresh produce. Since food insecurity is an acute and long-term ED risk factor (Becker et al., 2017), restricted food access may increase the risk of EDs among disabled people.

In addition to sharing risk factors, EDs and disability also may be bi-directionally related. EDs may directly cause and/or contribute to disability-related impairment (e.g., cardiovascular or other medical complications; Mehler et al., 2010) and disability-related challenges may precipitate and/or exacerbate EDs (Cicmil & Eli, 2014). Certain diets are indicated for management of chronic illnesses like a gluten-free diet for celiac disease), or disabilities like eating sensory-congruent food; however, such ‘restrictive’ diets may inherently increase risk for ED onset and/or maintenance (Brede et al., 2020; Conviser et al., 2018; Nimbley et al., 2023; Quick et al., 2013). As noted, disability status may hamper treatment access, especially if one’s disability impacts transportation and/or employment, thus worsening ED pathology. Further, certain aspects of disability may influence one’s body image (e.g., body functionality; Bailey et al., 2015; Thomas et al., 2019) and/or eating concerns (e.g., sensory sensitivities in autistic individuals; Kinnaird et al., 2019). EDs may also arise as a way to cope with disability-related experiences – namely, minority stress arising from having a disability (Cicmil & Eli, 2014; Sim et al., 2017, 2021). Notably, in some instances, disability may serve as a protective factor against EDs: for instance, one study identified congenitally blind women reported lower thin-ideal internalization, body dissatisfaction, and disordered eating compared to women who became blind later in life or sighted women (Ashikali & Dittmar, 2010; Baker et al., 1998).

Finally, although EDs and disability are often connected, we should not assume this is always true. For example, someone who uses a wheelchair may develop an ED independent of their disability (Gross et al., 2000). Understanding factors that lead to reciprocal relationships between EDs and disabilities, especially compared to instances when EDs and disabilities are distinct, may clarify how, and under what circumstances, these phenomena share etiology to identify avenues for ED prevention for individuals with disabilities.

3.1. Sociocultural and economic factors

Above, we briefly highlighted the various levels of the proposed model (political/economic, sociocultural, and individual/interpersonal) and their interdependence. Previous ED literature has typically focused on individual factors (e.g., Barakat et al., 2023), therefore we aim to further describe the intersection of sociocultural and economic factors below. However, further exploration of the political/economic level of the model is beyond the scope of this paper since economic influences vary greatly across the world (e.g., depending on the healthcare systems and available social programs), although the role of capitalism on disability has been well-documented (Russell, 2019).

3.1.1. Ableism.

Understanding the association between disability and EDs requires attention to relevant systemic factors at this intersection – namely, ableism and discrimination (Hassouneh-Phillips et al., 2005; Iezzoni, 2022). Many disabled individuals report barriers to healthcare, including accessibility concerns (e.g., transportation to healthcare facility), feelings of ‘invisibility,’ and incompetence of healthcare providers (de Vries McClintock et al., 2016). Ableism in healthcare is associated with new or worsening disability later in life (Rogers et al., 2015), which may partially result from medical exclusion and inaccessibility (e.g., Reppermund et al., 2020). As such, limited understanding among practitioners on how to identify and treat EDs among those with disabilities results in poorer treatment outcomes and even iatrogenic harm (e.g., Nimbley et al., 2024). Prior case studies identified practitioners missing ED symptoms like food restriction, binge eating, or obsessive focus on calories) by erroneously attributing these to neurodevelopmental and intellectual disability (Hendrickson et al., 2015).

Health insurance, disability, and employment are also tightly intertwined under capitalism. Disability studies often trace the emergence of the medical model of disability to industrialization, whereby disabled people are seen as “unfit” workers (Davis, 1997; Ross & Taylor, 2017). In turn, health insurance in the United States is often dependent on having employment (i.e., employer-provided insurance) or a medically documented inability to work, which disproportionately burdens those who are unable to work, afford health insurance, or obtain the required documentation (e.g., Hashemi et al., 2022).

3.1.2. Weight stigma in healthcare and beyond.

Weight stigma is rampant in healthcare settings, with potentially dire consequences, with one, but non-exhaustive, example being withholding certain potentially life-saving interventions based on body mass index (BMI). Specifically, many surgical centers enforce strict BMI cutoffs for orthopedic and transplant surgeries, despite evidence reporting BMI as a non-reliable criterion for postsurgical complications and/or success (e.g., Rubenstein et al., 2019). Withholding surgery until weight loss is achieved could motivate or worsen ED symptoms, as individuals may resort to extreme behaviors to receive needed treatment.

To this end, people with disabilities may develop ED symptoms if they are told to lose weight as a “solution” to other disability-related challenges. In healthcare settings, weight-based “overshadowing” frequently involves blaming symptoms or experiences on weight, irrespective of the actual cause (Phelan et al., 2015). Given that weight loss is a gateway to intervention, or even accurate diagnosis, patients may engage in ED behaviors to have their healthcare needs addressed. Therefore, weight stigma may particularly affect certain disabilities that are associated with obesity (e.g., vascular or metabolic conditions; Udo et al., 2016). Because healthcare settings are fraught with weight stigma, increased healthcare utilization due to disability may also lead to increased exposure to weight discrimination, which could also increase ED risk. Finally, ED treatment access and outcomes may center weight status as a key indicator of ED illness and recovery (e.g., for anorexia nervosa), which can lead to insurance companies only paying for ED treatment if one has a low BMI or BMI being seen as an indicator of treatment “success” (e.g., Hill et al., 2023). For someone with a high BMI or binge eating disorder, weight loss may even be set as a treatment aim (Grilo et al., 2020). The centering of weight in ED treatment may therefore limit opportunities for long-term recovery and even contribute to disability (Lester, 2011).

4. Future research and practice

4.1. Defining EDs and Disability

In this paper, we focused on EDs and disability generally. However, when defining disability, we encourage researchers and practitioners to not view disability as monolithic, given the disabled community varies in background, experiences, and needs. As such, we recognize that the definitions of EDs and disability require careful consideration and with an intersectional lens (Moradi & Grzanka, 2017). Specifically, attending to the unique experiences that occur across a person’s intersecting identities (e.g., racialized ableism) is critical for understanding disability and EDs among people who hold multiple marginalized identities (Brinkman et al., 2023). For example, experiences of ableism may coincide with experiences of racism among individuals who are racially minoritized and disabled, leading to unique barriers to accessing effective and culturally-informed treatment. A disabled cisgender women may access ED treatment more easily than an autistic cisgender man, given EDs are under-diagnosed and under-recognized in men (Murray et al., 2017; Strother et al., 2012). Similarly, ED research must consider how intersectional identities may result in unique body- and appearance-ideals that differ across groups that may not be captured in current ED definitions focused on thinness (e.g., Mishra et al., 2023).

4.2. Measurement

Regular reporting of disability within ED research would clarify whether disabled populations are adequately represented within ED research, and how much extant findings generalize to disabled populations. We suggest researchers collect disability status in demographics questionnaires (see Cockburn et al., 2024) and use inclusive measurement tools to assess disability in ED samples – for example, multidimensional measures of disability like the World Health Organization Disability Assessment Schedule (WHODAS; Üstün, 2010) and the OCAP-18 (Hirai, 2021; Lorgelly et al., 2015). Moreover, adding ED measures into public health surveillance systems that track disability status would facilitate examination of disparities in ED prevalence and outcomes among individuals with disabilities. There have also been recent efforts to develop assessment and diagnostic frameworks for certain disabilities among patients receiving ED treatment. Most notably, the PEACE pathway focuses on identification of autism among those with restrictive eating disorders (Tchanturia et al., 2020). Expanding these efforts to other co-occurring disabilities may support disability identification within the ED community.

It is currently unclear whether existing ED measures are appropriate for use within specific disabled populations (Longhurst, Nimbley, et al., 2024). Hence, future research validating existing ED-related measures within specific disabled populations is critically important. For example, body image scales often depict non-disabled subjects which raises methodological challenges when employed among samples with physical and/or visible disabilities.

4.3. Testing ED/Disability Relationships

This paper proposes possible bidirectional associations between EDs and disability, and a putative theoretical model. However, these associations and model must be empirically evaluated. Such research necessitates the availability of validated psychometric measures for use in this population, as described above. Additionally, we encourage researchers to include psychiatric disability as a form of disability, given EDs themselves may be disabling and EDs almost always co-occur with other psychiatric conditions (Hambleton et al., 2022; Hudson et al., 2007).

4.4. Prevention

It cannot be assumed that ‘traditional’ ED risk and protective factors outlined in existing theories (e.g., body dissatisfaction/appreciation) necessarily apply to the disabled population, nor that the same risk and protective factors are shared across disabilities (Bailey et al., 2015; Messer et al., 2022; Stice & Shaw, 2002). For example, targeting body functionality appreciation may be a viable option for both disabled and non-disabled patients; however, thoughtful adaptations may be needed to avoid describing body functions in ableist ways that could exclude and harm those with certain disabilities (Rice et al., 2021). Finally, it is crucial to mitigate structural inequalities that may contribute to ED and disability risk, including via policy and legislation. For example, expanding nutritional support programs could reduce ED risk by improving food security, and reduce disability risk by improving health outcomes (Becker et al., 2017; Wang et al., 2021).

4.5. Treatment Access and Efficacy

Accumulating evidence indicates that healthcare professionals need specific training on both disability and EDs (Clemente et al., 2022; Field et al., 2023). Disabled individuals may also be less likely to see their own symptoms as indicative of an ED, particularly if they have experienced pressure to lose weight from healthcare providers (e.g., Grillot & Keel, 2018). Therefore, education for both healthcare providers and individuals about EDs and disabilities may be crucial for ensuring adequate treatment access. Future efforts should examine other ways to improve access to existing prevention/treatment programs (e.g., Field et al., 2023), but also the efficacy and feasibility of adapted interventions designed for use among disabled individuals (Roberts & Kwan, 2018). Scholars should, however, be mindful that barriers and adaptation needs will likely vary within and across disability statuses.

5. Conclusion

Previous work on EDs and disabilities focused on the socioeconomic outcomes of EDs as a disability (e.g., healthcare costs). Research on the complex multidimensional associations between EDs and disability and the prevalence and outcomes of EDs among disabled people is sorely lacking. We propose a theoretical model emphasizing the multidimensional association between disability status and EDs that encompasses individual and sociocultural influences. We identify directions for future research including examining ED screening, prevention, and treatment access and efficacy for disabled individuals. Foundational research is needed to understand whether existing ED- and disability-related constructs are applicable to disabled populations and are appropriate intervention targets. Finally, we encourage scholars to utilize qualitative analyses and a participatory approach (e.g., co-production, community-led; Nicolaidis & Raymaker, 2015), underpinned by a recognition of epistemic injustice (Okoroji et al., 2023).

Table 2.

Calls for Research.

Category	Calls for Research
Defining EDs and Disability	Gain consensus on how research conceptualizes/operationalizes disability Consider developmental differences and interactive processes linking EDs and disabilities Understand the heterogeneity of disabled experiences, both between (e.g., physical vs. sensory) and within (e.g., autism) differing disabilities Investigating how intersectional identities may contribute to unique experiences of EDs and/or disabilities (e.g., unique treatment barriers, body image ideals)
Measurement	Incorporate assessment of disability (e.g., WHODAS; Üstün, 2010) into routinely-assessed demographics forms in ED research studies Add ED measures in public health surveillance systems that track disability status Develop new or revise existing ED assessment tools (e.g., Eating Attitudes Test-26; Garner et al., 1982) for use in disabled populations Expand existing diagnostic pathways (e.g.., PEACE; Tchanturia et al., 2020) toward other disabilities
Testing ED/Disability Relationships	Empirically evaluate the proposed theoretical model and putative bidirectional and indirect relationships between EDs and disability Understand the extent to which secondary consequences of EDs may be disabling Investigate distinct functions of ED behaviors within disabled populations Examine psychiatric disability, given high rates of psychiatric disorder comorbidity with EDs Refer to mixed-method designs for elucidating factors which contribute to ED/disability relationships Prioritize co-production/participatory research to center lived experience perspectives
Prevention	Explore population-specific strengths that might protect against EDs in disabled groups Identify promoting factors (i.e., individual, socio-cultural) and/or processes necessary for positive outcomes (e.g., acceptance by others, community-level resistance) Develop new and/or revise existing ED prevention programs for disabled populations (e.g., Expand your Horizon; Alleva et al., 2017) Include community members in the development of prevention programs Work to mitigate inequalities that contribute to ED and disability risk (e.g., policy and legislation)
Treatment Access and Efficacy	Modify and validate existing ED treatments for use in disabled populations Examine disability-related barriers (e.g., accessibility) and facilitators to ED treatment Increase ED knowledge and training among providers who work with disabled populations Increase disability knowledge and competence among ED providers Evaluate the efficacy, acceptability, and feasibility of adapted interventions

Highlights.

• Disabilities commonly converge with eating disorders (EDs)

• Thus, we propose a novel multidimensional model of ED/disability relationships

• Additionally, we identify future directions for research in this area