Purpose: To determine the long-term effects of a low fat intervention on sugar content in school lunches
Methods: We calculated contributions of total sugars, 6 specific sugars, and selected nutrients in National School Lunch Program meals served in 56 former intervention, 20 randomly selected control schools that had participated 3 years before in a low fat intervention, and 12 additional schools from neighboring school districts that had never been exposed to the intervention, whose goal was the lowering of total fat to under 30% of calories and saturated fat to under 10% calories. Analysis of variance adjusted for region was used to compare treatment groups. Pearson partial correlations controlling for the effects of region and treatment group were used to assess the strength of sugar and fat relationships.
Results: Three years after the low fat intervention, former intervention, control and unexposed schools lunches were similar in mean total sugars (25% of calories), and "added" sugars (e.g. sucrose, glucose, galactose, and maltose) at 15% of calories; and differed only in their lactose content. As percent of calories from fat or saturated fat in lunches decreased, that from sugars increased. Lunches that met reduced saturated fat goals were significantly higher than those not meeting goals in percent of calories from sugars both in meals as offered (27.6 +/- 0.3% vs. 26.2 +/- 0.3 SE p =.004) and as served (26.5 +/- 0.4 vs. 23.9 +/- 0.4 p =.009). Lunches meeting reduced total fat goals were significantly higher only in percent of calories from sugars as served. Seventy-five percent of total sugar in lunches offered was from chocolate milk, fruit/fruit juices, and white milk. More "added" sugar came from high micronutrient foods, such as fruit, fruit juices, and chocolate milk than from desserts and entrees.
Conclusions: The existence of a fat-sugar "see-saw" makes it important to emphasize substitutions of fat and saturated fat with starches and fiber in school lunches.