By Jennifer Gaudiani, MD, CEDS-S, FAED
What’s in a name?
Let’s talk about “atypical anorexia nervosa.” (AAN) First of all, I hate the name and location within the DSM5. It’s embedded under “other specified feeding and eating disorders” (OSFED), which sounds like a technicality until you realize that many insurance plans specifically exclude OSFED diagnoses for any higher level of care coverage. Then it’s clear how important it is.
Additionally, the name itself is problematic on a number of levels. For those who don’t know, the DSM5 says AAN is diagnosed when “all of the criteria for anorexia nervosa are met, except that despite significant weight-loss, the individual’s weight is within or above the normal range.” As it turns out, almost no one uses the “significant weight loss” criteria to diagnose AAN. In a major meta-analysis, only 12.5% of high quality studies on the topic even mentioned weight or a classification of weight loss quantity when defining their patient population. (1) While I clinically prefer to use just “AN” for all with the psychological and behavioral manifestations, formally speaking I will diagnose AAN in anyone who is not formally emaciated and has the AN thoughts and behaviors, whether their weight has dropped, stayed stable, or risen.
The word “atypical” implies that it’s odd not to get underweight from these behaviors, and/or that these patients are less common than those with AN. Both premises are incorrect. My clinical experience and the broader experience of the millions of people who have tried diets only to promptly regain or gain more weight says that humans have generally evolved to protect their body weight in the context of insufficient food and to put on weight protectively whenever possible to prepare for the next “famine.” Therefore, I theorize (without any way to prove this) that there must be some polygenic reason why certain individuals become underweight and stay there when they underfuel consistently. In short: those with AN are actually the “atypical” ones in becoming underweight, defying biology.
Furthermore, studies for years have found the prevalence of AN to be about 0.9%. By contrast, more and more literature points to the lifetime prevalence of AAN being from 2-100 times higher than that of AN, depending on the diagnostic criteria used for AAN. (2) Therefore, once again AN is actually the atypical one. One recent study showed that among a sample of over 1000 United States military veterans, 13.6% of women and 4.9% of men met the criteria for probable AAN. (3)
Finally, I will argue in this blog that the arbitrary delineation of AN from AAN reflects primarily weight stigma rather than science. You will read below that those with AN and AAN have almost all the same medical complications and that those with AAN if anything have more severe psychopathology than those with AN. The only difference is whether someone becomes underweight, which is the so-called expected outcome. Nowhere else does such a distinction occur. We don’t say someone has “atypical substance use disorder” or “atypical cirrhosis” based on a single physical finding that separates two groups.
In sum: Atypical anorexia nervosa should just be called anorexia nervosa. There shouldn’t be a divide. Weight loss is not universal in AAN.
Medical and Psychological Complications and Mortality
Nearly all the medical complications of AN are found in equal severity and prevalence in AAN. (4) An important study showed that total weight loss and recent weight loss were better predictors than admission weight of severity of medical complications. (5) An important study showed that bradycardia (slow heart rate), orthostatic blood pressure changes, hypothermia (low body temperature), need for hospital admission, psychiatric comorbidities, and suicidal ideation were statistically the same between those with AN and AAN, while distress related to eating and body image were higher in those with AAN. (6) These findings have been borne out in multiple studies.
However, there are a few differences. Studies are mixed as to whether menstrual irregularity is different between the two groups, with some saying yes and some no. The data are fairly clear that those with AAN have better bone density than those with AN, although still worse compared with control groups. (7) I’m always struck by the absence of data on hypoglycemia (low blood sugar) in studies that discuss the medical complications of eating disorders. Many of you know that I believe based on my 15 years in the field that hypoglycemia is the leading cause of death in AN. When the heart and brain run out of fuel (glucose) and the body cannot use its own tissues to create more, the brain and heart will stop. There aren’t data I’m aware of about hypoglycemia rates in AAN. My clinical experience is those with AN at all body sizes can develop hypoglycemia. However, perhaps because muscle and fat stores persist in those with AAN, which is what we use to synethesize glucose when we aren’t getting enough food/carbs, the body is able to rescue itself better. I don’t have any other explanation for the fact that death rates from AAN are around double the age- matched population, while those from AN range from 6-20+ times age-matched peers.
In sum: Most of the medical complications seen in AN are also seen in AAN with the same severity and frequency, worse in the latter with rate and amount of recent weight loss. Psychological suffering as measured by questionnaires is the same or worse in those with AAN. It appears that bone density and deadly hypoglycemia are less common in those with AAN, as may be menstrual irregularity. Death rates are meaningfully lower in AAN than in AN, but still way higher than in the general population.
Treatment
This blog would get too long if I were to go into treatment extensively, but here are some key points, by no means comprehensive. Prevalence of those with AAN in treatment centers is rising; maybe this augurs well for insurance covering it as they should. Those with AAN get diagnosed later than those with AN due to provider weight bias and lack of awareness. I can’t tell you how many patients come to our clinic and tell us tales of being in need of residential care in the past, with their outpatient therapist and RD frantic about their safety, only to be told casually by a doctor, “Well, your weight is fine, so it can’t be that bad. Go home and keep trying.” (Or even worse, being congratulated on weight loss.) This of course contributes to medical trauma and reinforces the ED voice which is constantly whispering, “You’re not sick enough.” If this has been you, I’m so sorry. Keep advocating and keep honoring your healthy voice that has goals and dreams.
In the outpatient setting, our clinic treats AAN just as we treat AN, with certain adjustments. One question that commonly comes up in those with AAN is what the target weight should be. The literature has not resolved this question. I’m keenly aware of my potential for internalized weight bias as I assess this question with patients. If a patient or their outpatient team knows that there’s a minimum weight at which their brain works best and where they become medically stable, that, of course, becomes the proper minimum to aim for. And childhood growth charts are invaluable, especially for younger patients. Almost always, helping a child return to their most typical percentile for weight is necessary for recovery.
For many patients, especially adults, I have faith that monitoring medical signs and symptoms (measurable and narrative) will give us most of the information we need as we support recovery. For those who don’t have an absolute minimum weight we are aiming for, I ask the RD and patient to develop a satisfying, consistent, abundant meal plan that lands somewhere between a maintenance and a weight-gain plan given the hypermetabolism that so often occurs with refeeding. We’ll monitor symptoms of gastroparesis, cold hands, low energy, poor sleep, concentration, memory, heart rate, mood, fatigue, sex hormone activity, and intensity of ED thoughts along the way. (And almost no one has all those symptoms! Many have just one or two.) I’ve seen patients need meaningful weight restoration for recovery, and I’ve seen patients thrive with relatively little weight gain when they show complete fidelity to the nutrition their body needs every day. I don’t want anyone to take away from this, “Dr. G said I don’t have to gain weight if I have AAN.” This is a delicate, individualized process that has to be assessed over time. Persistence of intense eating disorder (ED) thoughts in addition to struggling to remain consistent on meal plan and resist ED behaviors almost always means the brain remains too starved, such that further weight restoration is needed. I often get asked whether patients with AAN who need weight restoration should be renourished accordingly “even if” they have medical complications that can be associated with higher weight, like diabetes, obstructive sleep apnea, etc. The answer is: absolutely. It is foundational to human physical and psychological health to nourish consistently and adequately. In sum: Treatment for AAN is still inadequate across levels of care. Weight restoration is an individualized process that currently lacks peer-reviewed evidence, except that in almost all cases, children should return to their typical growth chart percentile for weight. Weight is only one of many, many elements we need to pay attention to when it comes to eating disorder recovery, as medical and emotional recovery may or may not be measurable but are extremely important!
1 Walsh BT, Hagan KE, Lockwood C. A systematic review comparing atypical anorexia nervosa and anorexia nervosa. Int J Eat Disord. 2023 Apr;56(4):798-820. doi: 10.1002/eat.23856. Epub 2022 Dec 12. PMID: 36508318.
2 Harrop EN, Mensinger JL, Moore M, Lindhorst T. Restrictive eating disorders in higher weight persons: A systematic review of atypical anorexia nervosa prevalence and consecutive admission literature. Int J Eat Disord. 2021 Aug;54(8):1328-1357. doi: 10.1002/eat.23519. Epub 2021 Apr 17. PMID: 33864277; PMCID: PMC9035356.
3 Masheb RM, Ramsey CM, Marsh AG, Snow JL, Brandt CA, Haskell SG. Atypical Anorexia Nervosa, not so atypical after all: Prevalence, correlates, and clinical severity among United States military Veterans. Eat Behav. 2021 Apr;41:101496. doi: 10.1016/j.eatbeh.2021.101496. Epub 2021 Mar 2. PMID: 33711788.
4 Rastogi R, Rome Md ES. Restrictive eating disorders in previously overweight adolescents and young adults. Cleve Clin J Med. 2020;87(3):165-171. doi:10.3949/ccjm.87a.19034
5 Whitelaw M, Lee KJ, Gilbertson H, Sawyer SM. Predictors of Complications in Anorexia Nervosa and Atypical Anorexia Nervosa: Degree of Underweight or Extent and Recency of Weight Loss?. J Adolesc Health. 2018;63(6):717-723. doi:10.1016/j.jadohealth.2018.08.019
6 Sawyer SM, Whitelaw M, Le Grange D, Yeo M, Hughes EK. Physical and Psychological Morbidity in Adolescents With Atypical Anorexia Nervosa. Pediatrics. 2016;137(4):e20154080. doi:10.1542/peds.2015-4080
7 Nagata JM, Carlson JL, Golden NH, Long J, Murray SB, Peebles R. Comparisons of bone density and body composition among adolescents with anorexia nervosa and atypical anorexia nervosa. Int J Eat Disord. 2019;52(5):591-596. doi:10.1002/eat.23048