Abstract
Large portions contribute to excess consumption. We tested whether a portion size labeling intervention influenced consumers to select smaller lunch portions. Two cafés introduced a reduced-sized main course (half the original size and ∼30% lower in price), January-May 2024. Cafés were randomized to a portion labeling intervention or control. During the baseline and follow-up periods, the reduced size was called "Small" and the original "Large" in both cafés. During the intervention period, labels remained unchanged in the control; the intervention café relabeled the reduced size "Standard" and retained "Large" for the original size. The primary outcome was average kcal per main course purchased. Secondary outcomes were: 1) average kcal per transaction; 2) number of main courses purchased; and 3) gross sales. We also examined whether introducing a reduced-sized main course irrespective of the labeling intervention was associated with 1) average kcal per transaction, and 2) sales during the study compared to 2023. Cafés sold 2940 reduced-sized and 8587 original-sized main courses. Average kcal per main course purchased (DID -7.5 kcal [95%CI: 25.8, 10.7]) and kcal per transaction (DID -21.6 kcal [95%CI: 46.6, 3.3]) did not significantly differ compared to the control during the intervention. Regardless of portion labels, transactions with a reduced-sized vs. original-sized main course had 405 fewer kcal (95%CI: 414.7, -396.1). Average weekly gross sales increased by $3131 (95%CI: 2,098, 4164) in 2024 vs. 2023. The labeling intervention did not reduce kcal purchased, but offering reduced sizes regardless of labeling lowered kcal per transaction and did not harm sales.