Making Connections: Peer Support and Eating Disorder Recovery

I feel like a broken record when I say that we continue to lack an evidence base for most “alternative” forms of support for eating disorders. As I’ve noted in prior posts, just because something is not evidence based does not mean it does not work for anyone; often, an evidence base is established when researchers can secure enough funding to run a randomized-controlled trial (RCT) that would act as evidence.

Even when an RCT has been run, it is hard to say that one form of treatment is best for all. People with eating disorders, like people in general, respond to different things, based on personal preference, history, culture, age, gender, and so many other factors. It feels a bit simplistic to write that, but I sometimes think we need a reminder of that fact!

Ultimately, and unsatisfyingly, it can be hard to predict what will work best for someone to support their recovery. This can also change over time, as people’s life circumstances and desires change. Recently, I’ve heard rumblings around the potential to use peer support and self-help groups to support recovery. In my post about the EDAC conference I noted how impressed I was by the presentation by We Bite Back, highlighting an online peer-support program for people in recovery. Then, Tetyana pointed me in the direction of a recent study by Perez, Van Diest & Cutts (2014, Open Access) in the Journal of Eating Disorders exploring MENTORConnect, another peer-based program.


As the authors explain, peer support is not exactly a new phenomenon in the field of mental health more generally. Interestingly, they also cite many studies from the 1980s when introducing the idea that mentorship and support might be particularly beneficial for eating disorders. This made me wonder, if peer support and mentorship were being investigated in the 80s, what took away the attention in the intervening years until more recently?

There is more research about mentorship and support for other mental health issues, particularly in the addictions literature. When one thinks about peer support, what likely comes to mind is Alcoholics Anonymous, which is explicitly founded on the peer-support model.

Peer mentorship models for individuals in recovery from eating disorders have less often been the focus of research. This may partly be because self-disclosure of recovery when working in a therapeutic capacity is more contested in the eating disorder field (though it happens, and can be seen as a good thing or a bad thing- or a non-issue, for that matter…). Concerns about emotional entanglement and comparisons seem to be especially strong where eating disorders are concerned.

Further, while defining recovery from any mental health issue is no easy task, defining eating disorder recovery has been a particularly difficult nut to crack. People’s definitions of recovery vary enormously, and differences in recovery definitions could make peer support more challenging. For example, what if your mentor’s definition differs from your own? How can we establish benchmarks through which your mentor can guide you if we aren’t exactly sure what we’re aiming for?

There are a few other reasons that peer support for eating disorders is not wholeheartedly embraced. Perez et al. note some concerns raised in the literature, including:

  • Emotional entanglement between mentor and mentee
  • Mentor dominating the mentee’s process of recovery
  • Risk of relapse for the mentor

Despite these concerns, however, the authors suggest that peer-support might help to fill a gap in people’s experiences of treatment and recovery. They note that some of the main roles for social support might include:

  • Reducing financial burden
  • Providing “emotional support, information, and feedback” (p. 2)
  • Role modeling
  • Navigating triggers

One of the authors, Shannon Cutts, founded a program called MENTORConnect, which forms the basis of this preliminary, evaluative study.


MENTORConnect matches people who are actively engaged in eating disorder but working toward recovery and people who have been recovered from eating disorders for at least a year. The program is free and is designed to facilitate peer support. In order to be matched, volunteer mentors are screened to ensure that they have been recovered (which is defined here as “largely free from eating disordered thoughts and coping behaviours,” p. 4) for 12 months or longer. Each mentor is able to have up to 3 mentees.

Once they are matched, pairs plan for how their mentoring relationship will play out. The amount, type and method of contact varies depending on the particular pair, as does the length of the mentorship.


Perez et al. wanted to examine whether this free mentorship support program could help in recovery processes. They were quite explicit throughout the article that the aim was not for the program to replace medical/clinical support; in fact, treatment compliance was also one thing that they hoped the program would increase. Additionally, they were looking to determine whether mentorship could increase mentees’:

  • Motivation to recover
  • Energy and time spent on recovery
  • Quality of life

They also looked at the content and frequency of mentor-mentee interactions and whether there were also benefits for mentors’ recovery processes.

This study was one part of a larger evaluation of the content of and satisfaction with MENTORConnect. There were 141 participants in total, mostly (91%) from the US, mostly (81%) White, and aged 15-63 (mean age 31).

34 mentors participated; average length of recovery was 8 years. Importantly (in my opinion) 4 of the mentors had no history of an eating disorder; they were mental health professionals. The others had recovered from eating disorders as follows (note that participants were asked which eating disorders they had experienced, so they could select more than one):

  • 56% AN
  • 38% BN
  • 15% BED
  • 38% EDNOS

Mentees were divided into 2 groups (based on ability to find a match, not randomization): matched and unmatched. Average age was 31, and most were women (though I could not find any indication of exactly how many were men or about the gender of the mentors). Matched and unmatched mentees were quite similar in terms of demographics.


The authors looked at:

  • Quality of Life (using the Eating Disorder Quality of Life Scale)
  • Motivation, energy and confidence toward recovery (using 3 questions)
  • Treatment compliance (using 1 item about how many appointments the person had missed over the past 30 days)
  • Time and frequency of communication
  • The mentoring process (including what pairs talked about, based on a 27 item checklist)
  • Unmet needs (using an open-ended question)
  • Impact on mentors (using an open-ended, 250-word response question)

Over all, mentees who had been matched noted a better quality of life than those who had not been matched. They also missed fewer appointments (an average of 0.98, ranging from 0-4) than unmatched mentees, who missed an average of 2.14 (ranging from 0-5). However, there were no major differences between the groups in terms of motivation, energy, or confidence about recovery.

As expected, there were wide variations in how often mentees and mentors connected, and how long they had been in their mentorships. It seems that over time, certain areas of quality of life improved for mentees, including:

  • Education/vocation
  • Family and close relationships
  • Future outlook
  • Psychological
  • Emotional
  • Values and beliefs
  • Physical

At least some participants talked about each of the 27 items on the list of focus areas, though some were more popular than others. Most discussed were how to (adapted from a table on p. 8):

  • Manage/reduce thoughts and behaviours related to the eating disorder
  • Stay motivated and committed in recovery
  • Manage relapses
  • Choose recovery
  • Set, work toward, recovery goals
  • Deal with triggers (e.g. family, environment)

Those who weren’t matched shared that they wanted support in dealing with emotions and with comorbidities, including depression, anxiety, and self harm. Some noted wanting help with practical needs like grocery shopping and other food-related parts of life including sticking to a meal plan, and managing body changes that often come along with recovery.

Interestingly, 91% of the mentors noted that they found that the program was helpful to them, saying that mentoring was a positive force in their recovery and reminded them of the strides they had made in their own journeys.


Perez et al. suggest that their study provides some preliminary support for the idea that mentorship programs like MENTORConnect can be a positive force in the lives of both mentees and mentors in recovery. Again, they are not suggesting that the program can replace medical or clinical support in recovery, but that it may be a helpful addition. This might be particularly relevant given the lengthy wait lists and difficulty securing insurance for eating disorder treatment.

So, what do I think? I’m on the fence. If you’ve read my posts in the past, that might not surprise you. I continually toe the line between openness to new and innovative approaches to providing support to people with eating disorders and concerns about how to make sure people are getting the best kind of care possible to help them in recovery.

I firmly stand by my opening paragraph noting that the absence of a randomized controlled trial does not mean that this mode of support is not helpful. I think it is wonderful that people are trying new things to support the difficult process of recovery. With this in mind, a potential role for this program not explored in depth in the article is actually helping to navigate the process of finding the kind of clinical/therapeutic treatment that works best for the person (e.g. is culturally, age, etc. appropriate), as opposed to just encouraging treatment compliance.

When you are struggling, it can be incredibly difficult to figure out who you need to ask and for what. Things like locating treatment programs that fit with your personal needs and/or securing insurance coverage can be extremely time consuming and difficult- could it be helpful to have help from some who has been there to help you in this process?

One question I had about the design of the program (and, subsequently, the study) was that a few of the mentors did not have a history of an eating disorder. The kind of support and mentorship provided by these people might be quite different, and it might be a different kind of mentor-mentee relationship. I found it a bit odd that the authors did not comment more on this.

I agree with the authors that peer support and mentorship could be beneficial for both mentors and mentees, something that least warrants more research and I do find this study a great start. I think it would be interesting to ask an open-ended question or interview mentors and mentees about what they talked about instead of providing a checklist of potential topics–the authors note that all of the 27 items they listed were chosen by at least some of the pairs, which is unsurprising (i.e., when something is listed, you’re likely to think about it).

I would be quite curious to look at similarities and differences between this and other similar programs. I am also curious about how different definitions of recovery might change the process and results. I continue to go back to the question of who defines recovery, and what the impact of individual differences between mentors, mentees, and clinicians might be on these relationships. Could this be a place for one of the authors’ suggestions, that new mentors be mentored themselves by mentors with more experience providing support? At the same time, this does not alleviate my concerns about whose version of recovery dominates.

Tetyana also mentioned the idea of informal mentorship and peer support being helpful. In order to quit my rambling before you all fall asleep, I’d be curious to discuss in the comments. What do you think?


Perez, M., Van Diest, A.K., & Cutts, S. (2014). Preliminary examination of a mentor-based program for eating disorders. Journal of Eating Disorders, 2 (1) PMID: 25426293


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. Thanks for doing a write-up about this article Andrea!

    When it comes to something that’s meant to be an adjuncts to treatment and/or recovery, I hesitate to be too critical. After all, if people sign up voluntarily, and find it helpful, well, who am I to judge? When it comes to recovery, I generally feel like if people find something helpful, and it is not prohibitively expensive, then go for it. It is not like it is a tumor where we can show whether a certain treatment is or isn’t leading to a reduction in tumor size.

    Besides the four mentors not having had eating disorders (uh?? what), I was surprised that the authors did not do any qualitative interviews. I feel that preliminary studies are the time to do this kind of thing, to make sure that the researchers are not missing anything crucial. Would you agree?

    I generally found the paper to be missing a lot of information that I feel wouldn’t be too difficult to list. Like, for example, the age range of the mentors. It is mentioned for the mentees, but not for the mentors. On the face of it, it is not a big deal. But, omission of data for one group when it is listed for another group (or for a certain time point) always strikes me as incredibly fishy.

    I was surprised by the participants’ income: ~65% made <$100,000. That means means ~35% are making $100,000+. Affluent bunch!!! I wonder what's up with that. I can see the benefits of a more formal mentor-mentee relationship for some, particularly those who may dislike forums or online groups. Personally, I favour informal relationships that form on recovery forums. For one, the inherent power dynamic/hierarchical relationship structure of the mentor-mentee relationship as it is proposed in this study turns me off from it; but again, that's just my personal view. I can completely see how that would be something people would like and would benefit from much more than from the informal, less structured relationships that form organically. I can also buy the argument that those organic relationships might have the potential to be more damaging to either or both parties (if both individuals are in the process of recovery, versus one being clearly ahead of the other).

    • Yeah, I am on the same page re: adjuncts to treatment. I think the free aspect of the program is definitely one of it’s benefits; so yes, like you say- why not? I was happy to see that the authors were clear that they didn’t intend for the program to take the place of treatment.

      I was surprised not to see interviews and definitely think they would have been helpful. It seems like the rationale was that the limited prior work on peer mentorship had been qualitative/small scale so I think they were trying to make a more convincing case with a larger sample size. Like I said in the post though, I think this would have been especially helpful for exploring the things that mentors & menthes talk about- just asking in an open-ended way, instead of giving a list and having people choose the items they talk about… I feel like that would have been a more interesting way to get at the actual content in a non-primed way.

      Whoa, I can’t believe I missed the income thing.. that’s interesting. I also wonder why. I wanted more demographic info, but I wonder whether the authors had all the demographics or maybe not? Not sure how much info they collect as a matter of course.

      Can you explain more what you mean about the hierarchical relationship in the program? I think I get what you mean but curious how you think it impacts the relationship. Like, thinking about one person as “a role model” in an explicit sense or some kind of guru, as opposed to being in peer support with someone more on a level playing field with you like a more organic kind of friendship?

      • I feel like if you are just doing a preliminary study, then that’s the time to do open-ended questionnaires to see what you might be missing or maybe haven’t thought about as a researcher/author/creator of the program.

        More demographics would’ve been interesting, but like I said, I’m more surprised by the lack of some data for some groups but not others. Surely it was collected for both? To get an average age, you need to know all the ages and thus you have an age range… I feel like this kind of criticism people outside of research feel is miniscule, but I feel like it is exactly the kind of thing peer-reviewers should pick up on. Also, did they mention the average age difference between mentor and mentee pairs? I don’t see why not just do it and put it in the appendix or something.

        Yeah, the thing about the hierarchical relationship that turns me off is purely my personal view. It is not really a criticism of the program because obviously others would probably really like it. I just find the notion that just because someone recovered or is further along in recovery that means they know more or can give advice kind of meh. Or maybe it is just the title roles that turn me off, I am not sure.

        I wonder, do mentors think they know more because of their title as mentors? Do mentees value the advice of the mentor more than of a peer (on, say, a forum) because of the titles “mentor” and “peer”? What effect do those titles/roles have on the relationship? I much prefer interacting with people who may be, in some sense, further ahead than I am in ED recovery, or having experienced and gone through similar things that I have, but who have a “Yeah, I’ve been in a similar place, I think I know what it feels like, maybe this would be helpful.” The mentors might be saying the exact same thing, but does the fact that they are the “mentor” affect the way their advice is perceived by the mentee? Does it affect the relationship in someone sense? It may not. It may not even for me, but I am not sure.

        Surely there are sociological studies on this (probably not in field of EDs, but in other fields). I just edited an essay that touched upon something similar.

        • Good point on demographics maybe could have been in appendix & surprise about that not getting pick up in peer review…. I agree, it would have been interesting to know about average age difference (I didn’t spot it in the article). I think that could also play into the hierarchical-ness (yeah, that’s a word now…) of the relationship. Then again, maybe not. I’m not sure if I were in the position of either mentor or mentee whether I’d prefer a wider age gap or a smaller one… same thing for difference in length of time in recovery. Would it be harder or easier to relate to someone with more years of recovery experience? I’m not sure. I think there are so many intangibles about any kind of relationship, be it mentorship or therapeutic alliance or friendship that dictate who “clicks” with who, so maybe it’s hard to figure out a matching “formula” that will work. It did seem to me, at least, that there was room to decide pretty early on in the mentor-mentee match whether both people wanted to continue on.

          Also good point about the language of mentor/mentee vs. peer. I’ve definitely read articles about how labels and roles play into experiences; more so in terms of diagnosis if we’re talking health & mental health, but also in terms of status, so I don’t think it’s a stretch to say it could impact the relationship. For better or for worse, really… I guess the benefit is that having the title of “mentor” could reinforce the positive benefits of the relationship for the mentor (i.e. they’re calling themselves a mentor, so they should probably act like one… not that that doesn’t carry certain pressure, too, particularly around the need to be invulnerable… but I wonder if that could be mitigated by work with their own therapists/friends/etc.). On the downside, it does reinforce the expertise of someone who doesn’t necessarily have clinical training; then again, as we’ve talked about, just because someone has clinical training doesn’t mean they’re automatically empathetic or a good support person. So back to the idea of relationality and who clicks with who, I see this all as feeding into the complexity of any kind of supportive and/or therapeutic relationship. And I sit again on my wishy-washy fence wondering whether my rambles ever make any sense…

          • To me, it implies expertise about the recovery experience/process. I don’t think clinical training has much to do with this (add or take away). I guess I just don’t like the language. Again, completely personal.

            I would not want to be seen as a mentor to others even though I am routinely asked for advice. I would also be turned off by the idea of someone “further” in recovery mentoring me, but perhaps it is because I’ve met too many people who self-identify as recovered and offer advice that has a tinge of “I know more than you because I am recovered” kind-of thing. I also see so much patronizing, obnoxious, and self-righteous “advice” from self-identified recovered people online. So, my experience with respect to that would turn me off from something like that — participating as a mentor or as a mentee. I would never want to be see as knowing more than someone else with regard to recovery or the eating disorder process just because I might be less symptomatic or have struggled for longer or something else just because I don’t think there’s necessarily a linear relationship there.

            But I really, really do want to stress to people reading that this is my *personal* opinion. I think programs like MENTORconnect are fine. I don’t feel strongly either way; indeed, probably more positive because I am happy the option exists for people who are interested and who feel they would benefit from a one-on-one relationship like that.

            I think I am becoming very sensitive to language as a result of the type of content I edit and read for work.

          • Right, I get that for sure. I’m also uncomfortable with the linearity ascribed to recovery- it seems overly simplistic to reduce recovery to a trajectory from “unwell” to “well” when people’s lives are complex & diverse. But yeah, definitely wonder if some people might like that aspect of it… maybe in a role-modeling sense, if that’s their cup of tea (and I think a lot of people like role models for things like this, or at least that’s the sense I get. Kind of a hopeful thing?)

            I dig the sensitivity to language. It’s very social science-y of you.

  2. Great review. I’d love to learn more about Mentor Connect, but the are not sharing any of their info.

    We do peer support in two ways – group based, and one on one. Our groups run for 12 weeks, are closed, and are led by two facilitators who have both had several years of recovery behind them. These facilitators have experience running groups, and have additional training with us.

    We are evaluating the groups – ED Quality of Life and a MSCARED (Motivation stages of change for adolescents recovering from an ED- it’s also used with adults). Measured are administered pre and post, and then with a 6 and 12 month follow up. Still collecting data but have a qualitative piece too. Hearing a lot about reduction in isolation, shame,increase in hope, better relationships, increase in help seeking behaviour. Can’t wait to look at data in a year.

    There is good data on the impact of peer support for many other mental illnesses, but not much at all for eating disorders.

    Love the conversation.


    • Thanks for commenting, Shaleen! I thought you might like this post; your peer support groups sound really interesting and I’ll also be excited to hear about what your results are when the research is complete.

      Do the people in your groups tend to seek other forms of support (e.g. treatment) while in group, or are they on wait lists, or have they found other kinds of support to be less helpful? Maybe you don’t know yet without the data. I’m curious to learn more about it; I think the MSCARED results could be especially interesting if people in the groups are thinking about more intensive treatment. Also pleased to hear you have a qualitative component… in these kinds of preliminary studies (and as you say the peer support literature is scarce for eating disorders) I think qualitative data is especially helpful. It is great to hear directly from people experiencing something what they find helpful/unhelpful etc.

      Best of luck with the research, and let me know when it is shareable!

  3. I stumbled upon this while researching peer support for EDs. I am hoping to start an actual peer support program in my area as an adjunct to (or prior to for people who are waiting for) treatment. It’s nice to know there is some open-mindedness about this. I personally think, with the right training for peer supporters, it’s the missing link in ED treatment. As a longtime member of 12 step recovery programs, I have experienced how this can work. Careful training and selection of peers is necessary though. Thanks!

Comments are closed.