Distinguish eating disorders by behavior and nutrition, recognize medical instability, and connect safe nutritional recovery with effective psychological care.
Two patients describe binge eating and vomiting. One has severe weight loss with fear of gaining weight; the other maintains a stable weight. Their behaviors overlap, but the diagnoses may differ. First reconstruct restriction, loss of control, compensation, and weight trajectory. Then ask whether either patient needs medical stabilization today.
Build the eating history before naming the disorder
A binge involves both an objectively unusually large amount of food within a discrete period and a sense of lost control. Eating more than intended at a celebration is insufficient. Ask what happened, how long it lasted, whether stopping felt possible, and what followed. Compensation includes vomiting, laxative or diuretic misuse, fasting, and driven exercise. Ask directly without assuming that appearance reveals behavior.
Compare the pattern, then assess severity separately
Restriction with significantly low weight
Anorexia nervosa requires restriction producing significantly low weight, fear of gaining weight or persistent behavior preventing gain, and disturbed weight or shape experience, disproportionate influence on self-evaluation, or poor recognition of seriousness. Recurrent binge eating or purging during the last three months identifies the binge-eating/purging subtype.
Binges with compensation
Bulimia nervosa requires recurrent binges and inappropriate compensation, both averaging at least weekly for three months, with self-evaluation overly influenced by weight or shape. The episodes must not occur exclusively during anorexia nervosa.
Binges without regular compensation
Binge-eating disorder requires marked distress, at least weekly binges for three months, and at least three associated features such as rapid eating, uncomfortable fullness, eating without hunger, eating alone from embarrassment, or subsequent disgust, depression, or guilt.
Weight helps interpret anorexia nervosa, but a normal BMI does not automatically establish bulimia nervosa. A person with substantial weight loss and the psychological features of anorexia may have atypical anorexia nervosa within other specified feeding or eating disorder, even at a higher weight. OSFED also includes clinically significant presentations that miss a frequency or duration threshold. Obesity alone does not diagnose BED, and BED can occur without obesity. Amenorrhea is not required for anorexia nervosa. [1][19]
Avoidant/restrictive food intake disorder, or ARFID, involves restriction driven by low interest, sensory characteristics, or fear of consequences such as choking rather than weight or shape concerns. It causes weight or growth problems, nutritional deficiency, supplement dependence, or marked psychosocial interference. Exclude food unavailability, culturally sanctioned practices, and a medical explanation that fully accounts for the restriction. ARFID can be medically severe. [2]
Try it here · Checkpoint 1 of 3
Make your prediction before reading the choices. A first attempt is just a starting point.
Case 2
Show answer and explanations for case 2
A. Binge-eating disorder (Why this does not fit)
Compensatory fasting and vomiting are regular, which distinguishes this from BED.
B. Anorexia nervosa (Why this does not fit)
The required significantly low weight from restriction is absent.
C. Atypical anorexia nervosa (Why this does not fit)
There is no substantial restrictive weight-loss syndrome; normal weight alone does not establish atypical AN.
D. Bulimia nervosa (Best answer)
The stem supplies recurrent binges, compensation, the weekly three-month threshold and weight-shape overvaluation outside AN.
Takeaway: Ask about fasting and exercise as well as vomiting when assessing compensation.
Review the growth curve, recent weight loss, intake, exercise, purging, medications, fainting, chest symptoms, hydration, and suicidality. Obtain temperature, resting and orthostatic vital signs, and a focused examination. ECG and laboratory assessment are guided by presentation and commonly include electrolytes, bicarbonate, renal function, glucose, magnesium, phosphate, and blood count. Normal initial tests do not exclude malnutrition or future refeeding complications.
Starvation reduces energy expenditure. Bradycardia, hypotension, hypothermia, low triiodothyronine, leukopenia, and fine lanugo hair can accompany restriction. Bradycardia in an undernourished athlete is not automatically benign conditioning. Interpret it with trajectory, orthostasis, symptoms, and ECG. A low T3 starvation pattern does not by itself justify thyroid hormone, which may worsen cardiovascular and bone risk.
Syncope, severe dehydration, electrolyte disturbance, arrhythmia, acute food refusal, serious psychiatric risk, or failure of outpatient care may require hospitalization. For adolescents and young adults, SAHM lists severe bradycardia below 50/min by day or 45/min at night among factors supporting admission. These are assessment supports, not a rule to wait until a pulse reaches 40. Clinical circumstances and available support determine the setting. [2]
Medical stabilization, nutritional rehabilitation, and psychological treatment address different parts of the illness and usually proceed together. Restoration is individualized to age, growth, prior trajectory, and physiology. A universal BMI goal or a single discharge weight cannot replace a recovery plan. Discuss food access, family support, and practical barriers without turning the visit into a moral judgment about eating.
Read vomiting through physiology
Repeated gastric acid loss produces hypochloremic metabolic alkalosis. Volume depletion promotes renal sodium retention and potassium loss, adding hypokalemia. Diarrhea from laxatives more often causes bicarbonate loss and a non-anion-gap metabolic acidosis, although mixed behaviors and volume depletion can alter the final laboratory pattern. Determine the actual exposure instead of treating every low potassium value as proof of vomiting. [12]
Acid can erode dental enamel, especially the palatal surfaces of upper teeth. Repeated contact of fingers with incisors can produce dorsal knuckle calluses called Russell sign. Painless parotid enlargement and increased salivary amylase may occur. These findings support an eating history but are neither necessary nor sufficient for diagnosis. Increased total amylase without compatible abdominal pain or pancreatic enzyme findings does not establish pancreatitis. Studies support a salivary contribution but do not settle whether binge eating or vomiting is the sole driver. [13][14]
Hypokalemia and hypomagnesemia increase electrical instability and can prolong repolarization, particularly with other QT-prolonging medicines. Syncope with electrolyte depletion warrants urgent monitored assessment and correction of all relevant abnormalities. Potassium replacement alone may be ineffective when magnesium remains low. Avoid assigning every eating-disorder death to one electrolyte mechanism. [20][9]
Retching followed by hematemesis suggests a Mallory-Weiss mucosal tear near the gastroesophageal junction. Severe chest pain, systemic illness, or subcutaneous air after vomiting raises concern for full-thickness esophageal rupture, or Boerhaave syndrome. Both require assessment; the latter is an emergency. Do not reassure a patient about blood loss solely because a mucosal tear is common. [15][18]
Feed the patient while anticipating intracellular demand
Why a normal initial phosphate can become dangerous
Prolonged undernutrition
Total body phosphate, potassium, magnesium, and thiamine reserves may be depleted despite initially acceptable serum measurements.
Nutrition resumes
Insulin promotes intracellular uptake and new energy synthesis. Thiamine demand increases. Monitor fluid balance as nutrition resumes.
Refeeding risk depends on recent intake, weight loss, malnutrition, comorbid illness, and electrolyte history, not just current BMI. Someone with a BMI of 27 after prolonged negligible intake can be at risk. ASPEN's consensus definition considers decreases in phosphate, potassium, or magnesium within five days of calorie reintroduction, with severity related to the size of the decrease and organ dysfunction. Thiamine deficiency can also contribute. [9]
Use a supervised nutritional protocol with baseline and repeated electrolytes, fluid assessment, replacement, and thiamine according to risk. Do not withhold nutrition indefinitely while waiting for a perfect laboratory panel. Equally, do not introduce an aggressive unsupervised regimen in a severely depleted patient. The appropriate setting must provide timely monitoring and treatment. [21]
The old blanket instruction to start every patient extremely slowly is incomplete. A randomized trial at two specialized inpatient centers enrolled patients aged 12 to 24 with AN or atypical AN and at least 60% of median BMI. With close monitoring, starting at 2,000 kcal/day and increasing by 200 kcal/day restored stability sooner than starting at 1,400 kcal/day with increases every other day, without more electrolyte safety events.
These were study regimens, not universal starting prescriptions. This supports specialized protocols, not one calorie prescription for every patient, including patients below 60% of median BMI or with different diagnoses and medical circumstances. [10]
Try it here · Checkpoint 2 of 3
Make your prediction before reading the choices. A first attempt is just a starting point.
Case 8
Show answer and explanations for case 8
A. Renal phosphate wasting from proximal tubular dysfunction (Why this does not fit)
This can lower phosphate but would need supporting renal findings. The immediate relation to nutrition after prolonged depletion favors an intracellular shift.
B. Insulin-driven phosphate uptake into cells during refeeding (Best answer)
Prolonged inadequate intake depletes reserves; renewed energy synthesis can lower serum phosphate despite a higher BMI.
C. Intracellular phosphate shift from hyperventilation-induced respiratory alkalosis (Why this does not fit)
Respiratory alkalosis can shift phosphate intracellularly, but no hyperventilation or blood-gas evidence is supplied. The nutritional transition is the stronger explanation.
D. Renal phosphate wasting from increased PTH activity (Why this does not fit)
The close relationship to resumed nutrition favors refeeding physiology; calcium and PTH findings would be needed for that diagnosis.
Takeaway: Refeeding risk is about depletion and recent intake, not BMI alone.
Energy availability connects reproduction and bone
Low energy availability reduces hypothalamic reproductive signaling. Reduced pulsatile GnRH leads to low or inappropriately normal LH and FSH and reduced estradiol. Low leptin is one signal of depleted energy availability; it is not a stand-alone diagnostic test. Low gonadal hormone levels, nutritional deficits, and other endocrine adaptations impair bone accrual and strength. Adolescence matters because missing expected bone accrual can have lasting consequences.
Functional hypothalamic amenorrhea is a diagnosis of exclusion. Check pregnancy first when relevant, then assess thyroid function, prolactin, gonadotropins, estradiol, and other tests guided by findings. Absence of galactorrhea does not exclude hyperprolactinemia. A low body weight does not exclude coexisting PCOS. Amenorrhea during restriction supports the energy deficit story but cannot replace this evaluation. [6]
Restoring energy availability and an appropriate weight trajectory is central. Calcium and vitamin D adequacy support care but cannot substitute for nutrition. The Endocrine Society suggests a baseline DXA after at least six months of amenorrhea, or earlier when severe nutritional deficiency, another energy-deficit state, or skeletal fragility is suspected. Combined oral contraceptives should not be prescribed solely to regain menses or improve bone density in FHA; withdrawal bleeding can mask persistent hypothalamic suppression.
For adolescents and women whose menses have not returned after a reasonable trial of nutritional, psychological and exercise modification, the guideline conditionally suggests short-term transdermal estradiol with cyclic oral progestin. This requires individualized specialist assessment and does not replace energy restoration. [6]
In children and adolescents, interpret DXA using Z-scores, not adult T-score categories. A Z-score at or below -2 supports low bone mineral density for age. Pediatric osteoporosis generally requires fracture context rather than DXA alone; one or more vertebral compression fractures can establish osteoporosis in the absence of local disease or high-energy trauma. Without a vertebral fracture, ISCD requires both a Z-score at or below -2 and a clinically significant fracture history: at least two long-bone fractures by age 10 or at least three by age 19. Height and maturation affect interpretation. [7]
Choose treatment for the disorder and the person
For medically stable adolescents with anorexia nervosa, family-based treatment is a leading first-line psychological approach. Caregivers temporarily support reliable nutrition, with increasing age-appropriate autonomy as recovery progresses. This approach does not blame the family. Adults require eating-disorder-focused psychotherapy, nutritional rehabilitation, and medical monitoring adapted to individual needs. Admission is appropriate for instability, not a substitute for continuing therapy. [2]
For adults with bulimia nervosa, eating-disorder-focused CBT and, where appropriate, guided self-help address regular eating, rigid rules, shape-based self-evaluation, and the binge-compensation cycle. Fluoxetine has an FDA-labeled bulimia dose of 60 mg daily; a 20 mg depression starting dose should not be mistaken for the studied bulimia dose. Medication complements psychological and nutritional care. For children and young people, NICE recommends bulimia-focused family therapy; individual eating-disorder-focused CBT is an alternative when family therapy is unsuitable or ineffective. [3][8]
For BED, psychological treatment targets binge frequency, distress, and regular eating. Lisdexamfetamine is FDA-approved for moderate to severe BED in adults, not for obesity or weight loss. Consider cardiovascular status, blood pressure and pulse, sleep, psychiatric history, and misuse risk. A stimulant is neither required for every patient nor an approved pediatric BED treatment. [5][8]
No medicine replaces nutritional recovery in anorexia nervosa. An adult randomized trial found modest weight benefit from adjunctive olanzapine without a corresponding benefit on the primary obsession measure. It remains an individualized off-label option with adverse-effect monitoring. SSRIs do not reliably restore weight in AN, but the explanation is not a proven universal absence of brain tryptophan. Assess and treat comorbid depression on its own merits. Mirtazapine can increase appetite and weight in depression treatment; that does not establish it as the best treatment for AN. [11][17]
Bupropion is contraindicated in people with a current or prior diagnosis of anorexia nervosa or bulimia nervosa because of seizure risk. This applies even without current purging. The label does not make BED alone a categorical contraindication, although other seizure risks and contraindications still matter. [4]
Try it here · Checkpoint 3 of 3
Make your prediction before reading the choices. A first attempt is just a starting point.
Case 16
Show answer and explanations for case 16
A. 10 mg daily is the labeled recommended bulimia dose (Why this does not fit)
A low starting dose may help tolerability, but the labeled recommended BN dose is 60 mg daily.
B. 60 mg daily is the labeled recommended bulimia dose (Best answer)
The BN trials supporting the label demonstrated benefit at 60 mg; depression starting doses are not interchangeable with disorder-specific targets.
C. 20 mg daily is the labeled recommended bulimia dose (Why this does not fit)
The label found 60 mg, rather than 20 mg, effective on binge and vomiting frequency in the supporting trials.
D. 80 mg daily is the labeled recommended bulimia dose (Why this does not fit)
Doses above 60 mg have not been systematically studied for BN in the cited label. Other indications have different dosing limits.
Takeaway: Match the medication regimen to the diagnosis being treated.
A. Other specified feeding or eating disorder (Why this does not fit)
The supplied frequency, duration, loss of control, distress and three associated features meet full BED criteria rather than a subthreshold presentation.
B. Anorexia nervosa (Why this does not fit)
There is no restrictive low-weight syndrome or weight-gain avoidance pattern.
C. Binge-eating disorder (Best answer)
Weekly binges for three months, distress and at least three associated features are present without regular compensation.
D. Bulimia nervosa (Why this does not fit)
Regular compensatory behavior is missing.
Takeaway: BED is a behavioral diagnosis, not a synonym for obesity.
A. Primary ovarian insufficiency (Why this does not fit)
Ovarian failure usually produces high gonadotropins in response to low estradiol.
B. Prolactinoma (Why this does not fit)
The measured prolactin is normal; absence of galactorrhea alone would not have excluded it.
C. Pituitary gonadotropin failure from a structural lesion (Why this does not fit)
A structural lesion can produce low gonadotropins but is not established here. The clear energy-deficit context favors functional suppression after appropriate exclusion of other causes.
D. Suppressed pulsatile hypothalamic GnRH signaling (Best answer)
Low energy availability reduces hypothalamic drive, producing low estradiol without an appropriately high gonadotropin response.
Takeaway: Interpret gonadotropins relative to estradiol after excluding other causes.
A mucosal tear primarily explains bleeding, not this combination of chest pain and subcutaneous air.
B. Acute coronary syndrome (Why this does not fit)
ACS is an important chest-pain differential, but cervical subcutaneous air immediately after forceful vomiting particularly raises concern for esophageal rupture.
C. Acute pancreatitis (Why this does not fit)
Pancreatitis usually presents with abdominal pain and does not best explain the cervical air after vomiting.
D. Full-thickness esophageal rupture (Best answer)
Pain and extra-luminal air after vomiting suggest Boerhaave syndrome and require emergency evaluation.
Takeaway: Severe chest symptoms after vomiting warrant investigation beyond a mucosal tear.
A. Individual CBT as the preferred first approach despite available family support (Why this does not fit)
Individual therapies may have roles, especially when FBT is unsuitable, but family-based treatment has particularly strong support for this adolescent AN scenario.
B. Family-based treatment with medical and nutritional follow-up (Best answer)
FBT is a first-line approach for medically stable adolescents and uses caregivers as recovery supports.
C. Begin inpatient psychotherapy despite current medical and psychiatric stability (Why this does not fit)
Admission depends on medical and psychiatric needs, not the diagnostic name alone.
D. Begin fluoxetine without a disorder-focused psychotherapy (Why this does not fit)
Medication alone does not restore nutrition or address the full AN syndrome.
Takeaway: Match the setting to stability and use effective adolescent treatment promptly.
A. The same results justify home use without repeated laboratory testing (Why this does not fit)
The trial's setting and eligibility limit that extrapolation.
B. The trial directly establishes safety below 60% of median BMI (Why this does not fit)
Patients below 60% of median BMI were excluded; the finding cannot establish safety for that population.
C. A low-calorie start is required because higher-calorie trials did not include AN (Why this does not fit)
The trial specifically enrolled AN and atypical AN, making this reason for rejecting it incorrect.
D. The trial supports a monitored higher-calorie approach in this population (Best answer)
The randomized trial enrolled ages 12 to 24 with AN or atypical AN at or above 60% of median BMI in specialist inpatient units. Its findings support a supervised protocol here, with individual risk assessment.
Takeaway: Use evidence within its population and care setting.
