← All sources
Ultra-processed diets cause excess calorie intake and weight gain: inpatient randomized crossover trial
NIH metabolic-ward crossover in 20 adults: ad lib ultra-processed vs minimally processed diets (2 weeks each) matched for presented calories, sugar, fibre, sodium, and macronutrients—ultra-processed arm increased energy intake by ~508 kcal/day and produced ~0.9 kg weight gain versus ~0.9 kg loss on unprocessed.
Design
- Setting: NIH Clinical Center inpatient metabolic unit
- Participants: 20 weight-stable adults (BMI ~27)
- Design: randomized crossover — 2 weeks ultra-processed diet then 2 weeks minimally processed (or reverse order)
- Meal matching: diets presented equal available calories, energy density, macronutrients, sugar, sodium, and fibre per protocol paper
Key results (ad libitum intake)
- Energy intake: +508 ± 106 kcal/day on ultra-processed vs unprocessed (P = 0.0001)
- Macronutrient excess: mostly carbohydrate (+280 ± 54 kcal/day) and fat (+230 ± 53 kcal/day); protein not significantly different
- Body weight: +0.9 ± 0.3 kg (P = 0.009) on ultra-processed vs −0.9 ± 0.3 kg (P = 0.007) on unprocessed; changes correlated with energy intake (r = 0.8)
Evidence hygiene
- Highly controlled feeding study—strong internal validity for ad lib energy intake mechanisms; external validity to free-living grocery patterns still requires pragmatic trials.
- Indexed under Cardiometabolic dietary patterns as a processed-food matrix anchor distinct from Mediterranean/DASH pattern trials.
Publication
Hall KD, Ayuketah A, Brychta R, et al. Cell Metab. 2019 Jul 2;30(1):67-77.e3. PMID 31105044.
Outcomes
- Ad libitum energy intake +508 ± 106 kcal/day higher during ultra-processed vs minimally processed diet (P=0.0001) despite matched presented energy density and macronutrient targets.
- Crossover weight change +0.9 ± 0.3 kg on ultra-processed vs −0.9 ± 0.3 kg on unprocessed (each P<0.01), tightly tracking energy intake (r≈0.8).