The Art of Therapy: Using of Arts-Based Therapies in Eating Disorder Treatment

Arts-based therapies are often used to supplement more “traditional” eating disorder treatment protocols in various different settings, ranging from individual therapy to inpatient units. However, as Frisch, Franko & Herzog (2006) note, no published research provides empirical support for the use of arts-based therapies for eating disorder treatment.

You might be wondering: if there is no empirical support, why are clinicians still using these therapeutic practices? You might also be wondering why I’ve chosen to dissect an article from 2006.

I’ll address the first question in this post (teaser: it’s really hard to say!). As for my delving back into the depths of academia, there is surprisingly little literature that touches on arts-based therapy, despite its continued use. This article provides an overview of why this might be, and where we can go from here.


Arts therapy is an umbrella term used to refer to the “medicinal use of creative arts,” including dance and movement, drama, music, and visual arts. The premise of arts-based therapy is that engaging with the arts will facilitate clients’ achievement of positive change and personal growth.

Arts therapies don’t often stand alone; even at their roots they are linked to a number of psychological or developmental theories, which inform the shape they take. Accordingly, these therapies vary widely in both their underlying assumptions and their implementation. I wrote about dance/movement therapy here; other types of arts therapies include music therapy, creative arts therapy, and drama therapy.


In short, not a whole lot. As the authors suggest, what literature does exist on this topic is largely in the form of case studies and theoretical discussions. Fewer articles take an outcome focus, particularly for eating disorders.

There is some support for the use of arts-based therapies for groups of trauma survivors, as well as for psychiatric patients. However, these studies have mixed results. Positive outcomes for psychiatric patients have included better relationships with oneself and with other. For trauma survivors, there has been some encouraging research suggesting improvements in symptoms.

These outcomes are unlikely to generalize to eating disorders. Resultantly, Frisch et al. sought to find out more about the use of arts-based therapies for eating disorders, in an effort to guide further research in this area.

Review of the Literature

The authors used 2 techniques to survey the landscape of arts-based therapies for eating disorders in North America. They sent out a survey of various types of therapy and their use for eating disorder treatment, assessing what treatment programs are currently using; they also conducted a thorough search of the literature.

The authors found 30 papers dealing with the use of arts therapies for AN, BN, or BED. Of these:

  • 17 discussed arts-therapies in general
  • 8 discussed music therapies
  • 5 discussed dance therapies


Music Therapy (MT)

Articles explored the use of music therapy for self-discovery or relaxation purposes. Some examples include:


  • Background music used to facilitate meditation exercises
  • Music played during mealtime to alleviate anxiety

I found this second point interesting; I had never considered music during a meal in the treatment context as “therapy,” but I suppose it could qualify.


  • Having patients listen and/or sing along with a song, and then examine or discuss the lyrics in relation to oneself

A unique example touched on the combination of CBT and music therapy. In this context, music is used to help clients to challenge cognitive distortions through creating music. For example, in one treatment setting, patients created and performed a “recovery rap” (Hillard, 2001).

Dance/Movement Therapy (DMT)

In dance therapy, movement is used to help patients become more connected to their bodies and (in theory) transcend gaps between body image distortions and reality. The authors suggest that DMT is promising because “the body is a central battleground in eating disorders” (p. 135). The links between body and mind are highlighted, such that work with the body will lead to changes in the mind.

Generally, DMT is linked to psychotherapy; dance therapists may work with patients considered to be in a “preverbal stage” to overcome body issues. To be honest, I didn’t love the way that this was framed in the article; the authors make a statement about eating disorder patients who are “emotionally stunted.”

I also wonder about the body-image focus, and the idea of real vs. ideal self; as I’ve written about before, there are some issues with the idea of body image as a real “thing” that all individuals have. However, it is possible that being in one’s body and exploring it’s potential through movement may have positive effects related or unrelated to body image.

Some examples of DMT exercises include:

  • Relaxation and centering
  • Mirroring movements of others
  • Facing mirrors
  • Having clients reflect on experiences of a movement session (often through drawing)
  • Videotaping body movement and then reflecting/critiquing

Blanche Evan (1991) created a form of DMT based on psychoanalytic theory that links a sense of the body to feelings, in an effort to establish “psychophysical unity,” or a better sense of linkage between body and mind.

Creative-Arts Therapy (CAT)

“Creative-arts therapy” encompasses a number of arts-based therapies including drama, role-playing, drawing, painting and sculpture. CAT is the most widely employed, but also the most varied, arts-based therapy. Commonly, CAT centers around generating insight through the use of symbolism.

A few examples of CAT include:

  • Diagnostic drawing, where patients examine and/or interpret drawing structure and content
  • Improvisation
  • Role playing

In some instances, CAT techniques are based in developmental psychology, and explore the root causes of eating disorders from early childhood using role-play and dramatization in order to help patients develop insight into the development of their disorder. Much like DMT, CAT is often theory-driven (developmental psychology in the case of CAT; body-image based in the case of DMT), which informs the approach it takes.

In another approach to CAT, those using phenomenal and nonphenomenal body image tasks (PNBIT; Rabin, 2003) look at the role played by self-esteem in the development of eating disorders. PNBIT looks very similar to body image interventions employing dance and movement, in that participants perform a number of arts-based tasks to narrow the gap between “real” body image and an internalized sense of one’s body.

Drama therapy has also been used in eating disorder treatment, for example through role-play. Some suggest that this form of therapy may lead to improvements in both communication and functioning (e.g. Jacobse, 1994).


Twenty-two programs met the authors’ criteria: residential treatment programs treating AN, BN or BED in North America. Of these, the authors obtained data from 19 (13 responded to their survey, and data from 6 was found on the Internet via publicly-available information). Surveys assessed the structure and implementation of the treatment program, with 4 of 30 questions asking specifically about arts therapies.

Survey Results

19 programs used arts-based therapy at least once per week

  • BED programs had a 90.55% client participation rate (CPR)
  • AN/BN programs has a 99.21% CPR

26% of programs had daily arts-therapy; these programs had 99.38% attendance

The amount of time per patient per week spent on arts-based therapy varied widely depending on the treatment program; 2 outliers spent over 10 hours per week on arts therapies, and thus were removed from the data set in order to calculate a representative average. Of the remaining 17 programs, an average of 2.8 hours per week per patient was reserved for arts therapy (not counting dance/music).

Programs endorsed a number of reasons for using arts therapies, including:

  • Self-discovery
  • Self-exploration
  • Self-expression
  • Addressing challenging issues e.g. self-esteem, body image, depression, isolation
  • Providing a “healthy outlet for expression of emotions and development of positive coping skills”
  • Giving access to non-threatening, alternative therapies
  • Facilitating change for those who have difficulty with more traditional forms of therapy


Reviewing these results, it becomes clear that despite lack of empirical studies, many programs still use arts-based therapies. The concept of “empirical validity” is not a simple one, when applied to arts therapies: there is no standard way of using these therapies, and arts-based therapies tend, by nature, to be individualized, and hard to accurately assess and measure.

The authors of this study suggest using several smaller randomized controlled studies assessing arts-based therapies, given the extent to which they are used in residential treatment programs. Given the difficulties associated with measuring these therapies, and the fact that they are almost always used in conjunction with other therapies, I have to wonder whether it would be possible to design a randomized controlled study that both isolates the effects of arts-based therapies and also holds real-life, clinical implications?

Arts-based therapies may themselves be at odds with the very idea of standardization. As the authors note, “standardization of arts-based therapies for the purpose of study may undermine the inherent therapeutic benefits” (p. 138). They raise the question of whether evidence-based and best practice are synonymous, a question I’ve often pondered myself.

While it is obviously not practical in many treatment settings to individualize treatment, it is also incredibly difficult (if not impossible…) to establish a program of treatment that will work for all individuals with eating disorders (see this post for more detail on this point).

So, what is the point of using arts-based therapies, if there is no evidence base to back them up? The service providers’ survey responses in this study speak to the potential value of providing an alternative means of treatment for those who do not respond to more “traditional” approaches.

I certainly think there is value in exploring new and creative ways to engage, but I am torn between an endorsement of the value of individualized treatment and the practicality of group-based, homogenous treatment strategies. The existence and widespread employment of arts-based therapies is in itself encouraging, in a way; providers are eager to reach as many patients as possible, through a multi-disciplinary approach to treating eating disorders.

We are still unsure how best to treat eating disorders, and we need to engage with complex ideas and complex solutions. Why not include the arts among these complex solutions? Nonetheless, there is still a need to better understand and be explicit about the theories and values underlying these approaches.


Frisch, M.J., Franko, D.L., & Herzog, D.B. (2006). Arts-based therapies in the treatment of eating disorders. Eating Disorders, 14 (2), 131-42 PMID: 16777810


Andrea is a PhD candidate focusing on individual, familial, and health care definitions and experiences of eating disorder recovery. She has an MSc in Family Relations and Human Development and a BA in Sociology. In her Masters research, she used qualitative and arts-based approaches (digital storytelling) to explore the experiences of young women in recovery from eating disorders. Andrea has recovered from EDNOS. She can be reached at andrea[at]scienceofeds[dot]org.


  1. As a published researcher, I am in favour of evidence-based practice – to a certain degree – and particularly for physical illnesses, which are much easier to delineate and measure biologically. Nevertheless, the fact that an established therapy is not evidence based, by way of published research/clinical trials, doesn’t mean that it is futile and inefficacious.

    People who have EDs form a widely heterogeneous group. Even if EDs are subdivided into categories, a group of people with (e.g.) BN, still form a heterogeneous group. IMHO, one mistake that is made in research (and thence potentially in treatment) is that a group of people with a particular ED is largely heterogeneous.

    One cannot take the person out of the illness. First and foremost, the person with an ED is a human, with a unique environmental background and unique interpersonal experiences – all of which have the potential to influence their presentation of an ED. I truly believe that the human mind is far too complex to be reduced to categorised patterns of neurological functioning.

    I know of some people with EDs who have responded remarkably well to art therapies. There is some anecdotal evidence to support the efficacy of such treatments. I know of people who have failed miserably at so-called evidence-based treatments for EDs, but recovered almost of their own accord, perhaps because they have met a wonderfully supportive person who has become a partner and made them feel that life is worth living. There are many examples of unorthodox treatments apparently working for EDs…

    Some will suggest that it is a waste of time; even harmful to use therapies that lack a published evidence base for people with EDs. I am inclined to say “suck it and see”.

    • Thanks for commenting. I totally agree re: “one cannot take the person out of the illness,” and also about the fact that just because something lacks empirical evidence does not mean it is futile- I hope that I haven’t presented art therapy in that way in this post. I do think it is interesting that treatment programs continue to use arts-based therapies despite a lack of empirical evidence. My main thought on that one is, well, why not. Even if it doesn’t help every patient, perhaps it would help some, particularly those who enjoy art. If nothing else, practicing art could become another outlet. Personally, I enjoyed art therapy simply because it WASN’T talk therapy. We spent a lot of time engaged in various forms of talk therapy (CBT, DBT, EFT, you name it) in my treatment program that honestly, it was a relief to just focus on an artistic task once per week.

      Something that I would like to see is some qualitative research in this area, both asking practitioners more about how and why they use art therapy (rather than just having a survey, as in this study) and asking patients whether they enjoyed it and why.

      Anyway, hoping it didn’t come off as if I were against arts-based therapies- far from it. As I mention, I think that these therapies might provide some good balance and innovation in the amalgam of complex approaches to incredibly complex disorders.

      • It didn’t come off as if you’re against arts-based therapies at all, Andrea 🙂

        I guess I was just emphasising that I have tended to inhabit a world where evidence-based practice is THE thing (Biomedical Sciences) – and a lot of people turn up their noses at anything that is not!

        During my recovery from AN (and I’m still in recovery), I have found some types of art therapy really helpful. They have not actually been delivered as ‘therapy’; I merely signed up for some ceramics classes to get me out and being more social. These provided a bi-weekly focus and made me feel so much more hopeful that life is worth living. And if you feel that life can be good then it’s all the more easier to eat. (Depression has played a big role in my AN).

        I agree that a qualitative study would be interesting!

  2. Interesting post, Andrea! It seems like this review highlights the vast amount of random “noise” that may obscure potential meaningful effects of arts therapy in a clinical setting. The populations examined, and the therapies practiced, are likely to contain substantial sources of variance themselves. I’m really curious to see if there is a study identifying individual characteristics that may be more or less benefited by arts-based therapies. Certainly some individuals are helped more than others, and some may see no benefits whatsoever, mitigating the enormous benefit for a select few. It just seems like there is so much unparsed variance to really figure out if these interventions are really beneficial or not.

    • Yeah, exactly. I’m sure it is not that really too complex either: people who like art might benefit from art-therapy, people who like drama might benefit from drama-therapy, etc.

      But I don’t see much use in pushing any kind of Tx on people, particularly things that are so individual-specific.

      General approaches to dealing with intense emotions and states, increasing mindfulness, being more aware of triggers, restoring normal-ish eating, etc.. those are general things that will probably help most (or many) people with EDs. Outside of that… meh.

      Treatment should have long-term benefits and provide you with skills/tools that will persist after you are done with treatment; it is not supposed to be a vacation.. and sometimes I feel the positive outcomes from these therapies are largely (or mostly, but not solely) due to non-specific effect.

      • I agree that treatment should have long-term benefits and provide people with the necessary skills and tools to navigate their post-treatment lives. Though it isn’t meant to be a vacation, when you’re in intensive treatment and you’ve been talking about your feelings all day sometimes it is nice to sit down and draw… however, maybe this isn’t everyone’s experience. I wonder if a lasting effect of arts-based therapy could be engaging in artistic pursuits following treatment as a coping mechanism? Not sure if that is the aim (haven’t studied arts-based therapies in any detail) but seems like a reasonable potential outcome…

        I also agree that pushing treatments on people is certainly not a good idea- as an option among many, however, arts-based therapies could be a good idea for *some*… The study was a bit unclear as to whether the arts components of the residential programs they studied were compulsory or not, but they did report attendance rates, leading me to believe that they weren’t.

    • Good point; I have a feeling that reactions to arts based therapies would be quite polarized- i.e. some people would probably love them and other would be more resistant. I would imagine some people would just see arts therapies as “airy fairy,” whereas other could find them very therapeutic, especially those who might not feel comfortable exploring issues verbally. I guess I don’t really know where I’m going with this comment… different things work for different people? Seems obvious but isn’t often acknowledged as deeply in therapeutic approaches as it could be (though of course it is hard to individualize treatment in a group setting, so trying out a bunch of different things and seeing what the patient/client prefers is likely best suited to individual treatment, evidently). Thanks for commenting, Liz!

Comments are closed.