Liquid versus solid carbohydrate: effects on food intake and body weight DP DiMeglio 1 and RD Mattes 1 * 1 Purdue University, Department of Foods and Nutrition, West Lafayette, IN 47907-1264, USA BACKGROUND: Beverages are contributing an increased proportion of energy to the diet. Because they elicit a weak compensatory dietary response, they may increase risk of positive energy balance. OBJECTIVES: This study aimed to document the differential effects of matched liquid and solid carbohydrate loads on diet and body weight. DESIGN: In a cross-over design, seven males and eight females consumed dietary carbohydrate loads of 1880 kJ=day as a liquid (soda) or solid (jelly beans) during two 4 week periods separated by a 4 week washout. Subjects were permitted to consume the loads however they chose. In addition to baseline measurements, diet records were obtained on random days throughout the study, body composition was measured weekly, physical activity was assessed before and after treatments and hunger was assessed during washout and midway through each treatment. RESULTS: Free-feeding energy intake during the solid period was signi®cantly lower than intake prior to this period. Dietary energy compensation was precise (118%). No decrease in free-feeding energy intake occurred during the liquid period. Total daily energy intake increased by an amount equal to the load resulting in dietary compensation of 717%. Consequently, body weight and BMI increased signi®cantly only during the liquid period. Physical activity and hunger were unchanged. CONCLUSIONS: This study indicates that liquid carbohydrate promotes positive energy balance, whereas a compar- able solid carbohydrate elicits precise dietary compensation. Increased consumption of energy-yielding ¯uids may promote positive energy balance. International Journal of Obesity (2000) 24, 794±800 Keywords: liquid; solid; rheology; food intake; body weight; dietary compensation; obesity Introduction Recent increases in the prevalence of overweight and obesity 1 are likely to have behavioral etiologies that permit expression of inherent physiological processes for energy storage. Attention has focused on both decreased physical activity (ie energy expenditure) and increased energy intake. Survey data of changes in lifestyle and physical activity implicate the former. 2±5 However, trends reported by the CDC's Behavioral Risk Factor Surveillance System 6 indicate that, while activity levels are low in the population, they have been stable over the past decade. Thus, the high level of inactivity may be playing a permissive role in weight gain, but does not explain recent trends. Total daily energy intake has increased from the late 1970s by approximately 800 kJ=day, 7 due principally to increments in protein and carbohydrate consump- tion. While the US diet is high in fat and energy density, a condition that facilitates positive energy balance, 8 levels have been stable over the past two decades. 7 Thus, this practice also fails to account for recent body weight trends. Protein and carbohydrate are reported to elicit strong oxidative and behavioral responses that should dampen their potential to con- tribute to positive energy balance, 9 ± 12 but there is increasing reason to believe this can be modi®ed by the form of food consumed. 13 A trend in the US diet, which has only recently been recognized, is a marked increase in energy consumption through beverages. Since 1978, the onset of the rapid rise in overweight in the US, ingestion of soda has increased by about 40%. 14 In 1997, Americans consumed an estimated 204 liters per capita and only 24% were sweetened with high-intensity sweeteners. Data from the NHANES II survey 15 revealed soft drinks were the seventh largest contributor of energy to the diet, accounting for 4.72% of the total. Based on the 1996 CSFII survey (assuming an average energy content of 711 kJ=336 g serving), 16 the proportion has risen to 6.73%. Juice consumption has increased by about 22% over the same time period. In 1997, estimated per capita consumption was 33 liters. 14 Sports drinks were a trivial market in the late 1970s, but growth has averaged 10.5% throughout the 1990s and this category now represents the sixth largest among beverages. 17 Estimated 1997 per capita con- sumption was 6.1 liters. Coffee and tea consumption provided 1.35% of daily energy intake according to NHANES II data, but have been among the fastest *Correspondence: RD Mattes, Purdue University, Department of Foods and Nutrition, Stone Hall, Room 212, West Lafayette, IN 479076-1264, USA. E-mail: mattesr@cfs.purdue.edu Received 19 April 1999; revised 27 October 1999; accepted 6 January 2000 International Journal of Obesity (2000) 24, 794±800 ß 2000 Macmillan Publishers Ltd All rights reserved 0307±0565/00 $15.00 www.nature.com/ijo