Human study · Nutrition · 12 Nov 2026
In four elementary schools, a free lunch program did not bring a consistent change in what students ate
Source: Nutrients
What the research found
A quasi-experimental study was published on 12 November 2026. Four elementary schools in Saskatoon, Saskatchewan, were compared. Two started a universal free lunch program and two did not. Lunch intake was weighed, with photographs, in May and June 2021 and again in May and June 2023. Foods were classed with Canada's Food Guide and the NOVA processing scale. The analysis included 980 student-period observations. Students are children, so the age tag is under 40. Sex is not in the abstract, so the gender tag is all. Total quantified lunch grams fell in both groups, with a smaller drop in the program schools (difference-in-differences 32.62 g). Vegetable and fruit grams fell less in the program schools (3.52 g), but the share of the tray did not favor those schools (minus 3.75 percentage points). Whole-grain foods fell more in the program schools, by 18.71 g and by 8.07 percentage points of the tray. Ultra-processed foods fell in both groups, with little difference in share (minus 0.17 percentage points). The authors say the program was not accompanied by a consistent improvement in what students ate. Four schools. DOI 10.3390/nu18223758. Not medical advice.
Evidence level: Human study. A study in people, but an association isn't proof of cause.
More nutrition news
- In 37,853 people with celiac disease, an eating-disorder diagnosis was more common than in matched peers
- In 61 adolescents with phenylketonuria, diet scores were more anti-inflammatory than in matched peers, and cytokine gaps did not hold after correction
- In 30 women aged 40 or older, 8 weeks of food-based resistant starch did not change metabolic syndrome measures
- In two adult cohorts, a gut microbe that makes isoCDCA was more common when cholesterol was high, and mice showed the same lipid pattern
- In obese mice, eight weeks of a 16:8 fasting schedule remodeled brown-fat mitochondria and raised energy expenditure