Resting Energy Expenditure in Adults With Williams Syndrome: Comparative Accuracy of Predictive Equations.
Standard resting-calorie formulas are too risky for adults with Williams syndrome—use Mifflin-FFM only as a rough draft.
01Research in Context
What this study did
The team compared real resting-energy use with seven common prediction equations in adults who have Williams syndrome.
They used a metabolic cart to measure actual calories burned at rest. Then they checked how close each formula came.
What they found
Adults with Williams syndrome burned fewer resting calories than same-age peers. Most of the gap came from lower muscle mass.
No equation hit the mark for clinical work. The Mifflin equation that includes fat-free mass was least wrong, but still off by enough to change a meal plan.
How this fits with other research
Hattier et al. (2011) saw the same story in kids with Down syndrome: only one skinfold equation (Slaughter) passed the accuracy test.
Beck et al. (2021) found standard body-size formulas also fail to predict fitness in adults with Down syndrome. Together these papers show that "one-size-fits-all" math breaks down in genetic syndromes.
González-Agüero et al. (2011) add that regular tools like BMI miss extra fat in youths with Down syndrome, echoing why calorie equations miss here: the body composition is simply different.
Why it matters
If you write meal plans or track weight for adults with Williams syndrome, skip the textbook equations. Use Mifflin-FFM if you must, but still budget for a 10–15 % error and monitor body weight and activity. A short indirect-calorimetry test beats any guess.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Swap in the Mifflin-FFM equation and flag any calorie target for weekly weight check instead of trusting the number alone.
02At a glance
03Original abstract
BACKGROUND: Williams syndrome (WS) is a rare neurodevelopmental disorder caused by a microdeletion on chromosome 7q. Previous research indicates that adults with WS are prone to having overweight and obesity and also have decreased fat-free mass (FFM) on body composition analysis. To date, no explorations of factors associated with measured resting energy expenditure (mREE) or with the accuracy of predictive resting energy expenditure (pREE) equations have been performed. This study aimed to (1) obtain mREE in adults with WS and examine contributing factors; (2) compare mREE between adults with WS and matched controls; and (3) assess the accuracy of widely used pREE equations. METHODS: A convenience sample of 41 adults with WS (mean age = 31.5 ± 10.2 years, mean body mass index [BMI] = 28.6 ± 7.7 kg/m2) completed in-person mREE assessment at a clinical research centre using indirect calorimetry. Anthropometric measurements, plus FFM and fat mass (FM) on dual energy X-ray absorptiometry (DXA), were obtained. Twenty-four of these adults with WS were matched to controls, concurrent or historical, on age, sex, race/ethnicity and BMI. Differences between the two cohorts were examined with and without adjusting for age and FFM (unadjusted analyses using independent samples t-tests and adjusted analyses using ANCOVA). pREE was computed for the WS cohort using equations derived from several distinct adult populations, including some with adults ages 60 and older, and others with a high proportion of adults with overweight or obesity. Prediction accuracy at an individual and group level was examined for each equation. RESULTS: Within the WS cohort, mREE was significantly lower in females (1183.1 ± 186.6 kcal/day) than in males (1366.4 ± 217.8 kcal/day) and in individuals with a BMI < 30 kg/m2 compared to a BMI ≥ 30 kg/m2; after adjusting for FFM, the sex-based difference in mREE persisted but with attenuated significance, while the difference between BMI categories no longer reached significance. WS participants had lower mREE than controls (males: 1310.0 ± 164.7 kcal/day vs. 1653.5 ± 406.7 kcal/day, respectively; females: 1163.1 ± 123.5 kcal/day vs. 1377.4 ± 401.5 kcal/day, respectively); after FFM adjustment, these differences were not statistically significant. Age was retained in all models even though it was not a significant predictor. Most predictive equations did not achieve ≥ 70% individual-level accuracy or a mean absolute percentage error ≤ 10%. The highest-performing predictive equation for individuals with WS was the Mifflin FFM equation, which yielded individual accuracy rates of 87.5% in males and 68.0% in females. The second top-performing equations differed by sex, with individual accuracy of 68.8% (Owen FFM) in males and 64% (Bernstein Height & Weight) in females. CONCLUSIONS: These findings indicate that (i) the amount of FFM explains most of the mREE reduction observed when comparing females and males with WS, and when comparing WS and matched controls, and (ii) few predictive equations provide clinically useful REE estimates in adults with WS. More accurate predictions tended to originate from equations developed in cohorts with a high prevalence of overweight and obesity, or with older adults. These findings highlight the need for cautious application of pREE equation values in the clinical care of adults with WS and underscore the importance of continued investigation into energy requirements and their determinants.
Journal of intellectual disability research : JIDR, 2026 · doi:10.1111/jir.70073