are a bmi under 16, under 75% of expected weight, and rapid weight loss refeeding risks?
“Less than 75% of expected body weight or BMI under 16 in adults, rapid or prolonged weight loss, even from a higher starting point.”
a host, a psychiatry resident said on PsychEd: Educational Psychiatry Podcast, 2026-10-01, at 11:38. hear it in context →
the episode: PsychEd Shorts 17: Medical Considerations in Eating Disorders (at 11:38). we link the show's own page; we do not re-host audio.
first, the part that holds
The risk factors the resident listed are the ones the main refeeding guidance uses. A BMI under 16 puts an adult in the top risk tier under both the UK NICE criteria and the 2020 ASPEN consensus. Recent weight loss counts on its own, whatever the person's starting weight.
what we found when we went and looked
We read secondary reproductions of the NICE and ASPEN 2020 criteria, because the primary NICE and Wiley pages blocked our fetches. NICE marks an adult as high risk with any one of these: a BMI under 16, unintentional weight loss of more than 15% over 3–6 months, little or no intake for more than 10 days, or low potassium, phosphate or magnesium. ASPEN's "significant risk" tier also starts at a BMI under 16, and its weight-loss threshold is lower: 7.5% in 3 months or more than 10% in 6 months. Neither adult guideline uses the "75% of expected body weight" cutoff. That measure comes from eating-disorder care. In one adolescent inpatient cohort, the odds of low phosphate during refeeding rose 6% for every 1% drop in percent of expected body weight. In the subgroup below 75% of expected body weight, 60–69% developed low phosphate.
assertion by assertion
| assertion | verdict | what the record says |
|---|---|---|
| a BMI under 16 in adults is a refeeding syndrome risk factor | confirmed | NICE: BMI under 16 alone is enough for high risk (under 14 is 'extremely high'). ASPEN 2020: BMI under 16 is 'significant risk', and 16–18.5 is moderate risk. We read both through reproductions in peer-reviewed reviews, not the primary documents. [1][2] |
| less than 75% of expected body weight is a refeeding syndrome risk factor | confirmed | Lower percent of expected body weight tracked with low phosphate during refeeding: odds rose 6% for every 1% drop. In the under-75% subgroup, 60–69% developed low phosphate. That cohort was adolescents, though, and the adult NICE and ASPEN criteria use BMI rather than percent of expected body weight. An adolescent position statement cited in an editorial treats under 70% of median BMI as severe malnutrition. So 75% is one cutoff in use, not a single agreed standard. [3][4] |
| rapid or prolonged weight loss is a refeeding syndrome risk factor | confirmed | NICE: unintentional loss of more than 15% over 3–6 months is high risk on its own, and more than 10% counts toward high risk when combined with another factor. ASPEN significant risk: 7.5% in 3 months or more than 10% in 6 months. [1][2] |
| weight loss counts even from a higher starting point | confirmed | In both NICE and ASPEN, the weight-loss criteria stand separately from the BMI criteria, so someone with a normal or high BMI can qualify through weight loss alone. The adolescent criteria also count more than 10% loss over six months whatever the current weight. [1][2][4] |
| these criteria reliably predict who will develop refeeding syndrome | overstated | The resident did not claim this; we checked it as context. In 327 critically ill COVID-19 patients, 82% met the ASPEN at-risk criteria and 36% developed refeeding syndrome. The criteria flag far more people than go on to develop it. [5] |
both directions
for the speaker: Both major adult guidelines name a BMI under 16 as a top-tier risk factor and count recent unintentional weight loss separately from BMI. That matches the resident's point that weight loss matters "even from a higher starting point." In eating-disorder care, lower percent of expected body weight predicted low phosphate during refeeding in a graded way, and patients below 75% of expected body weight had high rates of it.
against: The 75% figure is not in NICE or ASPEN, which use BMI. The eating-disorder evidence we read for it comes from adolescents, and the adolescent position statement uses 70% of median BMI rather than 75%. "Rapid or prolonged" is vaguer than the guidelines, which set specific thresholds (NICE more than 15% over 3–6 months; ASPEN 7.5% in 3 months or more than 10% in 6 months). The list also leaves out factors the guidelines weight just as heavily: days of little or no intake, low electrolytes before feeding, alcohol misuse, and some medications.
what this cannot say
We could not fetch the primary NICE CG32 page or the ASPEN 2020 paper (both returned 403) and read their criteria as reproduced in peer-reviewed reviews. One of those reviews gives NICE's weight-loss window as 6 months and another as 3–6 months. The data on percent of expected body weight comes from a single adolescent inpatient cohort of 87 patients, so it may not carry over to adults. The 2023 APA eating-disorder guideline was also blocked, so we could not check whether it uses 75%. Risk criteria are screening tools, not diagnoses, and how well they predict refeeding syndrome varies by population.
sources, each one fetched and read
- McClain CJ, Rios CD, Condon S, Marsano LS. Malnutrition and alcohol-associated hepatitis. Clin Liver Dis. 2021;25(3):557–570 (reproduces NICE refeeding risk criteria) What we read there: NICE high risk, one or more of: BMI <16 kg/m2; unintentional weight loss >15% over the last 6 months; little or no nutritional intake for >10 days; low K, PO4 or Mg before feeding. Or two or more of: BMI <18.5, weight loss >10%, little or no intake >5 days, alcohol abuse or certain drugs. Extremely high risk: BMI <14 or negligible intake >15 days. Fetched 2026-10-20 (HTTP 200). https://pmc.ncbi.nlm.nih.gov/articles/PMC8672300/
- Ponzo V, et al. The refeeding syndrome: a neglected but potentially serious condition for inpatients. A narrative review. Intern Emerg Med. 2021;16(1):49–60, Table 2 What we read there: ASPEN 2020 significant risk: BMI <16 kg/m2; weight loss '7.5% in 3 months or > 10% in 6 months'. NICE high risk: weight loss '> 15% within the last 3–6 months'. Neither table nor text uses percent of ideal or expected body weight. Fetched 2026-10-20 (HTTP 200). https://pmc.ncbi.nlm.nih.gov/articles/PMC7843537/
- Maginot TR, Kumar MM, Shiels J, Kaye W, Rhee KE. Outcomes of an inpatient refeeding protocol in youth with anorexia nervosa. J Eat Disord. 2017 What we read there: 87 patients, mean age 14.4. Percent body weight on admission was more strongly associated with low phosphate than the starting calorie prescription; each 1% decrease in %EBW raised the odds by 6%. Severely malnourished subgroup (<75% EBW, n=26): low phosphate in 69% vs 60% on higher- vs lower-calorie starts. Fetched 2026-10-20 (HTTP 200). https://pmc.ncbi.nlm.nih.gov/articles/PMC5209953/
- Kohn MR, Golden NH. Management of the malnourished patient: it's now time to revise the guidelines. J Eat Disord. 2022;10:56 What we read there: SAHM: BMI <70% of median BMI indicates severe malnutrition in adolescents, as does >20% body mass loss over one year or >10% over six months. A BMI <15 is recognized as severe malnutrition in adults. Cites evidence that low BMI predicted refeeding hypophosphatemia. Fetched 2026-10-20 (HTTP 200). https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9019959/
- Vahdat Shariatpanahi Z, et al. Refeeding syndrome and its related factors in critically ill coronavirus disease 2019 patients. Front Nutr. 2022;9:830457 What we read there: ASPEN criteria added loss of subcutaneous fat and muscle on exam to the earlier NICE criteria. Of 327 critically ill patients, 82% met ASPEN at-risk criteria and 36% developed refeeding syndrome. Fetched 2026-10-20 (HTTP 200). https://www.frontiersin.org/journals/nutrition/articles/10.3389/fnut.2022.830457/full
speakers are described by role, not named. one sentence is quoted under fair use, with attribution and a link to the original at the timestamp; we do not re-host audio or video. nothing on this page is a reason to start, stop or change a medication. if you have the file behind anything marked unverifiable, the research board is where people can answer. adam