- Email: [email protected]
Both a higher number of sips and a longer oral transit time reduce ad libitum intake
Food Quality and Preference 32 (2014) 234–240
Contents lists available at ScienceDirect
Food Quality and Preference journal homepage: www.elsevier.com/locate/foodqual
Both a higher number of sips and a longer oral transit time reduce ad libitum intake Dieuwerke P. Bolhuis a,⇑, Catriona M.M. Lakemond a, Rene A. de Wijk b, Pieternel A. Luning a, Cees de Graaf c a b c
Food Quality and Design, Wageningen University, P.O. Box 17, 6700 AA Wageningen, The Netherlands Food and Biobased Research, Consumer Science & Intelligent Systems, Wageningen University and Research, P.O. Box 17, 6700 AA Wageningen, The Netherlands Division of Human Nutrition, Wageningen University, P.O. Box 8129, 6700 EV Wageningen, The Netherlands
a r t i c l e
i n f o
Article history: Received 25 January 2013 Received in revised form 2 October 2013 Accepted 2 October 2013 Available online 9 October 2013 Keywords: Orosensory exposure Oral transit time Satiation Sip size Food intake
a b s t r a c t Background: A higher eating rate leads to a higher food intake, possibly through shorter orosensory exposure to food. The transit time in the oral cavity and the number of bites or sips per gram (inversely related to bite or sip size) are main contributors that affect eating rate. The separate role of these two aspects on satiation and on orosensory exposure needs further clarification. Objective: The objective of the first study was to investigate contributions of the number of sips per gram (sips/g) and oral transit time per gram (s/g) on ad libitum intake. The objective of the second study was to investigate both aspects on the total magnitude of orosensory exposure per gram food. Methods: In study 1, 56 healthy male subjects consumed soup where the number of sips and oral transit time differed by a factor three respectively: 6.7 vs. 20 sips/100 g, and 20 vs. 60 s/100 g (2 2 cross-over design). Eating rate of 60 g/min was kept constant. In study 2, the effects of number of sips and oral transit time (equal as in study 1) on the total magnitude of orosensory exposure per gram soup were measured by time intensity functions by 22 different healthy subjects. Results: Higher number of sips and longer oral transit time reduced ad libitum intake by respectively 22% (F(1, 157) = 55.9, P < 0.001) and 8% (F(1, 157) = 7.4, P = 0.007). Higher number of sips led to faster increase in fullness per gram food (F(1, 157) = 24.1, P < 0.001) (study 1). Higher number of sips and longer oral transit time both increased the orosensory exposure per gram food (F(1, 63) = 23.8, P < 0.001) and (F(1, 63) = 19.0, P < 0.001), respectively (study 2). Conclusion: Higher number of sips and longer oral transit time reduced food intake, possibly through the increased the orosensory exposure per gram food. Designing foods that will be consumed with small sips or bites and long oral transit time may be effective in reducing energy intake. Crown Copyright Ó 2013 Published by Elsevier Ltd. All rights reserved.
1. Introduction The current food supply consists of a majority of highly processed foods that support fast intake of energy and minimal oral processing, like energy-yielding beverages and foods low in fiber content (Cordain et al., 2005; Malik, Schulze, & Hu, 2006; Slimani et al., 2009). Foods that can be consumed quickly (i.e., fast eating rate, grams or energy per time unit) may facilitate over-consumption. A number of studies have shown that higher eating rate leads to higher energy intake (Coelho et al., 2006; De Wijk, Zijlstra, Mars, De Graaf, & Prinz, 2008; Haber, Heaton, Murphy, & Burroughs, 1977; Melanson, Greene, & Petty, 2011; Ponte, Melanson, & Greene, 2011; Viskaal-van Dongen, Kok, & de Graaf, 2011; Zijlstra, Mars, De ⇑ Corresponding author. Address: Bomenweg 2, 6703 HD Wageningen, The Netherlands. Tel.: +31 317482063. E-mail address: [email protected] (D.P. Bolhuis).
Wijk, Westerterp-Plantenga, & De Graaf, 2008). Moreover, several studies suggest a positive relationship between eating rate and body weight status (Hill & McCutcheon, 1984; Llewellyn, van Jaarsveld, Boniface, Carnell, & Wardle, 2008; Maruyama et al., 2008; Otsuka et al., 2006; Sasaki, Katagiri, Tsuji, Shimoda, & Amano, 2003). Hogenkamp, Mars, Stafleu, and de Graaf (2010) compared ad libitum intake of yoghurts low and high in energy density with different eating rates. Food intake was primarily influenced by eating rate, and not by energy density. We propose that shorter sensory exposure of food in the oral cavity (i.e., orosensory exposure) explains the increased food intake at a fast eating rate. A number of studies show the importance of orosensory exposure to food in establishing feedback signals of satiation (Davis & Smith, 2009; Lavin, French, Ruxton, & Read, 2002b; Weijzen, Smeets, & de Graaf, 2009; Wijlens et al., 2012; Zijlstra, De Wijk, Mars, Stafleu, & De Graaf, 2009). Direct infusions of food into the stomach or duodenum, thus bypassing the orosensory exposure, give much
0950-3293/$ - see front matter Crown Copyright Ó 2013 Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.foodqual.2013.10.001
D.P. Bolhuis et al. / Food Quality and Preference 32 (2014) 234–240
weaker responses of satiation compared to oral intake (Cecil, Francis, & Read, 1998; Cecil, Francis, & Read, 1999; French & Cecil, 2001; Lavin, French, & Read, 2002a). There are two ways to slow down the eating rate (g/s), in order to increase the orosensory exposure per gram food and thereby decrease the amount of food intake, which are prolonging the transit time of the food in the oral cavity (i.e., oral transit time) or decreasing the bite/sip sizes. Previous studies have shown that bite size positively affected food intake and oral transit time negatively affected food intake (Bolhuis, Lakemond, de Wijk, Luning, & de Graaf, 2011; Weijzen et al., 2009; Zijlstra et al., 2009). However, in these studies the effects of bite size and oral transit time were confounded which did not allow the verification of individual contributions. In a natural way of eating, bite or sip size and oral transit time influence each other. Forde, van Kuijk, Thaler, de Graaf, and Martin (2012) showed that foods that are naturally consumed with smaller bite sizes are associated with longer oral transit times than foods that are consumed with larger bite sizes. Also within the same food, smaller bite/sip sizes lead to relatively longer oral transit time (Tracy et al., 1989). The separate contributions of oral transit time and bite/sip size to satiation are not known. Longer oral transit time is expected to increase the orosensory exposure to food and decrease food intake. However, smaller bites/sips may also increase the total magnitude of orosensory exposure. For example, three sips of 5 g instead of one sip of 15 g, means a more pulsating exposure to food, which may increase orosensory exposure per gram food, and thereby influencing satiation. We use the term number of sips per gram (inversely related to sip size: g/sip) in this paper, because sip size influences the oral transit time in a natural way of eating. Understanding the individual contributions of number of sips and oral transit time per gram food on intake will lead to more insight in the process of satiation. The aim of the first study is to investigate the separate effects of the number of sips and oral transit time on ad libitum intake at a constant eating rate. Possibly, a greater orosensory exposure to food, as a result of higher number of sips and longer oral transit time, may explain the effects on satiation. Therefore, a second study was executed to investigate whether and to what extent the number of sips and oral transit time affect the total magnitude of orosensory exposure to food.
2. Subjects and methods 2.1. Study 1 2.1.1. Subjects Fifty-nine male subjects were recruited for participation. Fiftysix subjects completed the study, two subjects dropped out before the start of the study and one subject missed three ad libitum intake sessions. Subjects were healthy, had a normal weight (BMI 18.5–25 kg/m2, mean ± SD: 22 ± 2 kg/m2), were aged between 18 and 35 years (mean ± SD: 22 ± 3 years) and liked creamy tomato soup (pleasantness score >5 on a 9-point hedonic scale). Exclusion criteria were restrained eating behavior (Dutch eating behavior questionnaire (DEBQ) score >2.89), following an energy-restricted diet during the last two months, gained or lost >5 kg weight during the last year, having a lack of appetite, smoking, suffering from gastrointestinal illness, diabetes, thyroid disease or any other endocrine disorder, hypertension and kidney diseases. Subjects were informed that the aim of the research was to investigate the effect of sip size on flavor perception of soup. This study was conducted according to the guidelines laid down in the Declaration of Helsinki and all procedures involving human subjects were approved by the Medical Ethical Committee of Wageningen University. All subjects signed an informed consent form before participation.
235
2.1.2. Experimental design The study consisted of a 2 2 crossover design. Subjects came five times to the lab, including a ‘‘practice session’’ (first session), to consume soup in each of the four conditions (Fig. 1). The eating rate was set in all four conditions at 60 g/min. The oral transit time was three times longer in the ‘‘long’’ conditions (Long-LS and LongHS) compared to the ‘‘short’’ conditions (Short-LS and Short-HS), 60 s/100 g vs. 20 s/100 g, respectively. The number of sips per gram was three times higher in the ‘‘high number of sips’’ conditions (HS) compared to the ‘‘low number of sips’’ conditions (LS), 20 vs. 6.7 sips/100 g, respectively. 2.1.3. Test foods Tomato soup was used as test product in this study. One kilogram of soup was made from 333 g sieved tomatoes (Heinz, Elst, The Netherlands), 662.7 g water, and 4.7 g salt (NaCl). The mixture was heated until 60 °C. The calculated nutrient composition from the used ingredients was: 0.57 g protein, 1.6 g carbohydrates, 0.03 g fat, 253 mg sodium and 38 kJ (9.1 kcal) energy per 100 g soup. Raisin buns (local bakery) were used as preload. The nutrient composition was: 8 g protein, 52 g carbohydrates, 3 g fat, 300 mg sodium and 1120 kJ (268 kcal) energy per 100 g, according to the Dutch Food Composition Database (NEVO, version 2009/1.0). Each raisin bun weighed 22 g (246 kJ). The number of raisin buns as preload was calculated for each subject at half of the energy provided by an average lunch in the Netherlands (Hulshof et al., 2003), that is equal to 11% energy of the daily energy need. The daily energy need for each subject was estimated by the Schofield I equation (WHO, 1985), taking into account: gender, age, weight and a physical activity level of 1.6. Thirty subjects received 5 buns and 27 subjects received 6 buns. Subjects were instructed to eat all the raisin buns they were served. 2.1.4. Control of sip sizes, intervals and swallowing To control the sips and intervals, subjects consumed the soup through a food-grade silicon tube that was connected to a peristaltic pump (Watson-Marlow, type 323Du, Watson-Marlow Bredel, Wilmington, MA, USA). The tube ended in a pan of soup that was placed on a balance (Kern, type 440-49A, KERN & Sohn GmbH, Balingen, Germany) to record the amount consumed. The pump, the pan and the balance were all located at the experimenters’ side of the sensory booths, thus subjects did not see the experimental setup. The moment the pump started driving, subjects heard an auditory signal to prepare them that they would receive soup in their mouths (Fig. 1). They heard a double auditory signal at the moment they had to swallow. Before the start of each session, subjects were instructed that it was very important to swallow at the double auditory signal. 2.1.5. Ad libitum intake sessions Subjects came four times during lunch for the ad libitum intake of soup, with one week in between sessions. The four conditions were presented at random. Subjects started with consumption of the preload that consisted of raisin buns. A preload was used so that subjects would be in a less hungry state prior to soup consumption. A hungry state may overrule sensory factors to terminate consumption (Bolhuis, Lakemond, de Wijk, Luning, & de Graaf, 2012). They were instructed to consume all served raisin buns and they were allowed to drink water. After that, subjects paused for 30 min. Subjects were allowed to study or read, but were not allowed to leave the sensory room. After the pause, subjects received a tube from which they had to consume soup. Subjects received instructions and questions via a computer screen. After answering several appetite
236
D.P. Bolhuis et al. / Food Quality and Preference 32 (2014) 234–240
Time (s) Short-LS: Long-LS: Short-HS: Long-HS:
0
5
10
15
15g
15g
15g
15g
5g
5g
5g
5g
5g
5g
5g
5g
Fig. 1. Sips and intervals in the four experimental conditions. In the ‘‘Short-LS’’ (short transit time, low number of sips) condition, sips of 15 g were exposed in 3 s (from the start of the administration of soup until swallowing) in pulses of 15 s. In the ‘‘Long-LS’’ (long transit time, low number of sips) condition, sips of 15 g were exposed in 9 s in pulses of 15 s. In the ‘‘Short-HS’’ (short transit time, high number of sips) condition, sips of 5 g were exposed in 1 s in pulses of 5 s. In the ‘‘Long-HS’’ (long transit time, high number of sips) condition, sips of 5 g were exposed in 3 s in pulses of 5 s. Subjects heard an auditory signal when they received the soup and a double auditory signal when they had to swallow.
and hedonic questions, as described below, subjects pushed a button on the screen to start soup consumption. They were instructed to terminate consumption when they had enough. The mean (±SD) initial temperature of the soup was 56 ± 5 °C and the mean end temperature was 51 ± 3 °C. Subjects were instructed to stay in the sensory booths for at least ten minutes after they started consuming the soup. After ten minutes, a visual warning signal popped up on the laptop screen to inform subjects the ten minutes had passed. This was done to prevent subjects from leaving the research area for other reasons than being satiated with the soup. 2.1.6. Appetite, hedonic ratings and questionnaires Just before soup consumption, subjects rated their feelings of hunger, fullness, prospective consumption and thirst. After that, subjects were served a small sample of 10 g soup and rated pleasantness and desire-to-eat the soup. The same questions were answered again at the end of the ad libitum intake. In addition, the same questions, except for thirst, were rated at random after consumption of every 75 g soup. All questions were answered by using a 100 mm VAS, on a scale from ‘‘not at all’’ at the left end to ‘‘very much’’ at the right end. At the end of the session, subjects indicated the reasons for terminating soup consumption. The subjects were asked to what extent they agreed with the propositions: ‘‘I terminated consumption because I was full’’; ‘‘I terminated consumption because the flavor of the soup was not pleasant anymore’’; and ‘‘I terminated consumption because I did not like the manner of consumption’’. The propositions were answered on a 5-point scale from ‘‘totally disagree’’ (1) to ‘‘completely agree’’ (5). 2.1.7. Standardization of the satiety state To standardize the satiety state, subjects always started the lunch session at the same time. They were instructed to consume the same breakfast and not to eat and only drink water before the lunch started. Moreover, they were asked to refrain from drinking one hour before the test started. After each test lunch, subjects answered questions about what they ate for breakfast and whether they ate or drank between breakfast and test lunch. To make sure subjects would consume the soup until they felt satiated; they were instructed not to eat until one hour after the test. 2.2. Study 2 2.2.1. Subjects Twenty-two different subjects (12 male, BMI 18.5–25 kg/m2, mean ± SD: 22 ± 2 kg/m2), aged between 18 and 35 years
(mean ± SD: 22 ± 2 years) participated in the second part of the study. The same exclusion and inclusion criteria as in the first study were used to recruit subjects, except for gender. These subjects were used to consume via a tube and peristaltic pump with controlled sip sizes and intervals, because they also participated in another study that used a similar experimental set up. 2.2.2. Time intensity measurements Time intensity (TI) measurements allow to measure the perceived taste intensity over a certain period of time (Busch, Tournier, Knoop, Kooyman, & Smit, 2009). TI measurements were made in each of the four different conditions used in study 1. The area under the curve (AUC) of the time intensity measurements was used as an indication about the total perceived taste, which is considered to represent the total magnitude of orosensory exposure to the taste. The conditions were offered in random order. Subjects had one practice session to train the procedure of the TI measurements. Subjects were instructed to rate their perceived taste intensity, inclusive aftertaste intensity, continuously for one minute on a VAS from 0 to 100 mm. Subjects were instructed not to rate the first sip to have a short time to adapt to the procedure, therefore, the AUC was calculated between t = 30 s and t = 60 s, in which 30 g of soup (53 ± 2 °C) was consumed. The area under the curve (AUC) represents the total magnitude of orosensory exposure. 2.2.3. Statistical analyses Statistical analyses were performed using SAS version 9.1.4 (SAS Institute Inc., Cary, NC, USA). Data are presented as means ± SDs, P-values <0.05 were considered significant. Effects of number of sips, oral transit time, and their interaction, on most outcome parameters were assessed in a mixed model ANOVA that included order and had subject as random factor. To assess effects on ad libitum intake and on ratings after and during intake, ratings of initial hunger and ‘‘I terminated consumption because I did not like the manner of consumption’’ were added to the mixed model ANOVA. Changes in appetite and hedonic ratings between before (initial) vs. after (post) ad libitum intake were assessed by two-tailed paired t-tests. The changes in appetite and hedonic ratings during ad libitum intake were fitted per subject in a linear model: y = a + b ⁄ intake. A linear model was chosen because this was the best fit in most individual curves. The curves shown in the results section were calculated from the mean intercepts and mean slopes of the individual plots. Effects of number of sips, oral transit time, and their interaction on the AUC of the taste intensity were assessed in two-way ANOVA
237
D.P. Bolhuis et al. / Food Quality and Preference 32 (2014) 234–240
that included subject. Bonferroni adjustments were used for all post hoc comparisons in the present study.
Table 1 Appetite and hedonic ratings from before and after ad libitum intake of soup in the four conditions1,2,3. Short-LS
Long-LS
Short-HS
Long-HS
Hunger Initial Post4
60 ± 20 18 ± 18
61 ± 21 19 ± 19
62 ± 19 21 ± 19
62 ± 19 21 ± 19
Fullness Initial Post4
33 ± 21 75 ± 20
32 ± 19 73 ± 18
30 ± 18 73 ± 20
31 ± 19 76 ± 17
Prospective consumption Initial 62 ± 24 24 ± 21 Post4
64 ± 20 25 ± 23
65 ± 17 29 ± 22
64 ± 19 27 ± 18
Thirst Initial Post5
59 ± 21 54 ± 27
54 ± 21 54 ± 25
55 ± 21 54 ± 23
57 ± 19 57 ± 24
Pleasantness Initial Post4
64 ± 20 51 ± 23
65 ± 18 51 ± 18
64 ± 18 50 ± 22
66 ± 18 52 ± 22
Desire-to-eat Initial Post4
65 ± 19 29 ± 21
64 ± 18 28 ± 21
65 ± 19 31 ± 20
67 ± 19 32 ± 22
3. Results 3.1. Study 1 3.1.1. Ad libitum intake of soup The number of sips (F(1, 157) = 55.9, P < 0.001) affected ad libitum intake; it was 22% lower in the ‘‘high number of sips’’ conditions compared to the ‘‘low number of sips’’ conditions (Fig. 2). Oral transit time also affected ad libitum intake (F(1, 157) = 7.4, P = 0.007); it was 8% lower in the ‘‘Long’’ conditions compared to the ‘‘Short’’ conditions. There was no interaction effect between number of sips and oral transit time on ad libitum intake (F(1, 157) = 0, P = 1.0). 3.1.2. Appetite and hedonic ratings before and after ad libitum intake Initial appetite ratings (i.e., hunger, fullness and prospective consumption) and hedonic ratings (i.e., pleasantness and desireto-eat the soup) did not differ between conditions (all P-values >0.59) (Table 1). Initial thirst did also not differ between conditions (P = 0.17). In the preload phase, subjects drank: 146 ± 38 g and this were not different between conditions (P = 0.20). After ad libitum intake, appetite and hedonic ratings changed in each condition compared to the initial ratings (all P-values <0.001), except for thirst (all P-values >0.10) (Table 1). Appetite, thirst and hedonic ratings after intake were not affected by oral transit time (all P-values >0.44), or by number of sips (all P-values >0.21), and there were no interaction effects (all P-values >0.26). 3.1.3. Appetite and hedonic ratings during ad libitum intake Fig. 3 shows the linear curves of the changes in rated hunger (A) and fullness (B) as a function of intake. Higher number of sips led to faster decrease in hunger (F(1, 157) = 9.1, P = 0.003) (Fig. 3A); faster increase in fullness (F(1, 157) = 24.1, P < 0.001) (Fig. 3B); faster decrease in ratings of prospective consumption (F(1, 157) = 6.6, P = 0.011); and faster decrease in desire-to-eat (F(1, 157) = 5.7, P = 0.018). Oral transit time did not significantly affect the decrease in hunger (F(1, 157) = 1.1, P = 0.31) (Fig. 3A); the increase in fullness (F(1, 157) = 2.2, P = 0.14) (Fig. 3B); and the decrease in desire-to-eat (F(1, 157) = 1.0, P = 0.17). However, longer oral transit time resulted in faster decrease in prospective consumption (F(1, 157) = 4.5, P = 0.035). Pleasantness was not affected by number of sips or oral transit time (both P-values >0.33). None of the changes in appetite and hedonic ratings showed an interaction effect between number of sips and oral transit time (all P-values >0.30).
1 Values are means ± SDs. Values in a row with different superscript letters are significantly different (P < 0.05). 2 Rated on a 100 mm VAS. 3 Short-LS = ‘‘Short transit time, low number of sips’’ condition. Long-LS = ‘‘Long transit time, low number of sips’’ condition. Short-HS = ‘‘Short transit time, high number of sips’’ condition. Long-HS = ‘‘Long transit time, high number of sips’’ condition. 4 Significant differences between initial and post appetite and hedonic ratings in all conditions: all P-values <0.001. 5 No significant difference between initial and post thirst ratings.
3.1.4. Reasons to terminate consumption ‘‘I terminated consumption because I was full’’ was the most important reason in all conditions for terminating consumption (Table 2). The reasons for terminating consumption: ‘‘I was full’’ or ‘‘flavor not pleasant anymore’’ were not affected by number of sips or oral transit time (all P-values >0.38). Ratings of ‘‘I terminated consumption because I did not like the manner of consumption’’ were higher in high number of sips conditions compared to low number of sips conditions (F(1, 159) = 14.8, P < 0.001), but were not affected by oral transit time (F(1, 159) = 1.6, P = 0.21). 3.2. Study 2 3.2.1. Area under the curve of perceived taste intensity of 30 g soup Fig. 4 shows the mean perceived taste during 30 s, which is equal to consumption of 30 g in each condition. The mean AUC of each condition is 953 ± 528 for Short-LS; 1199 ± 544 for Long-LS; 1230 ± 561 for Short-HS; and 1494 ± 658 for Long-HS. Both higher number of sips (F(1, 63) = 23.8, P < 0.001) and longer oral transit time (F(1, 63) = 19.0, P < 0.001) led to a greater AUC, suggesting more orosensory exposure per consumed gram food. The taste of the soup is highest in intensity directly after swallowing, shown by the top of the peaks (Fig. 4). The mean heights of the peak were 71 ± 19 mm in the ‘‘Short-LS’’ condition; 71 ± 17 mm in the ‘‘Long-LS’’ condition; 63 ± 22 mm in the ‘‘Short-HS’’ condition; and 65 ± 22 mm in the ‘‘Long-HS’’ condition (calculated from the individual curves). Higher number of sips resulted in lower heights of the peaks (F(1, 63) = 17.8, P < 0.001). Longer oral transit time did not affect the mean height of the peaks (F(1, 63) = 0.5, P = 0.47). 4. Discussion
Fig. 2. Means (+SD) of ad libitum intakes of soup. A higher number of sips (P < 0.001), and longer oral transit time (P = 0.007), reduced ad libitum intake. Mean values with different letters are significantly different between conditions.
Previous studies indicated that a higher eating rate leads to a higher food intake when eating in a natural way (Coelho et al.,
238
D.P. Bolhuis et al. / Food Quality and Preference 32 (2014) 234–240
Short-LS
Hunger (mm)
100
(a)
Long-LS Short-HS
75
Long-HS
50 25 0 0
100 200 300 400 500 600 700 800
Intake (g)
Fullness (mm)
100
(b)
75 50 25 0 0
100 200 300 400 500 600 700 800
Intake (g) Fig. 3. Linear functions of rated hunger (A) and fullness (B) against intake in the four conditions. Higher number of sips led to a faster decrease in hunger (P = 0.003), and a faster increase in fullness (P < 0.001) per consumed gram food. Longer oral transit time did not significantly affect the appetite ratings. The dots visualize the mean ad libitum intake in each condition. Short-LS = ‘‘Short transit time, low number of sips’’ condition. Long-LS = ‘‘Long transit time, low number of sips’’ condition. Short-HS = ‘‘Short transit time, high number of sips’’ condition. LongHS = ‘‘Long transit time, high number of sips’’ condition.
Table 2 The reasons to terminate soup consumption in the four conditions1,2,3.
‘‘I was full’’ ‘‘Flavor not pleasant anymore’’ ‘‘Manner of consumption’’
Short-LS3
Long-LS
Short-HS
Long-HS
4.0 ± 1.1 2.6 ± 1.2
4.0 ± 1.1 2.6 ± 1.1
4.0 ± 1.1 2.6 ± 1.1
3.9 ± 1.0 2.5 ± 1.3
2.4 ± 1.1a
2.6 ± 1.1ab
2.9 ± 1.2b
3.0 ± 1.4b
1 Values are means ± SDs. 1Values in a row with different superscript letters are significantly different (P < 0.05). 2 The propositions were answered on a 5-point scale from ‘‘totally disagree’’ (1) to ‘‘completely agree’’ (5). 3 Short-LS = ‘‘Short transit time, low number of sips’’ condition. Long-LS = ‘‘Long transit time, low number of sips’’ condition. Short-HS = ‘‘Short transit time, high number of sips’’ condition. Long-HS = ‘‘Long transit time, high number of sips’’ condition.
2006; Haber et al., 1977; Melanson et al., 2011; Ponte et al., 2011; Viskaal-van Dongen et al., 2011). We have investigated the underlying mechanisms on how the factors that influence eating rate (e.g., number of sips, oral transit time) affect satiation. The results show that both higher number of sips per gram food and longer oral transit time per gram food independently reduced food intake by 22% and 8%, respectively (Fig. 2). In accordance, consumption with higher number of sips per gram food was associated with a faster increase in subjective fullness (Fig. 3). Higher number of sips and longer oral transit time both contribute independently to greater total magnitude of sensory exposure per gram food (Fig. 4). Longer transit time of food in the oral cavity
is associated with sensing the taste over a longer period of time, which explains the increased orosensory exposure (greater AUC, Fig. 4). In this study, oral transit time was varied within the same food. Within our food range, oral transit time from different taste qualities and textures may differently affect oral sensory exposure, e.g., bitter compounds often lead to longer after taste. In theory, equal oral transit times of different taste qualities may differently affect orosensory exposure and possibly satiation. Interestingly, higher number of sips per gram (thus smaller sips), also result in greater orosensory exposure per gram food. One sip of 5 g almost reached the same taste intensity as one sip of 15 g (top of the peaks, Fig. 4). This means that three sips of 5 g result in a greater orosensory exposure than one sip of 15 g (greater AUC, Fig. 4). In addition, the orosensory exposure may not only be affected by the taste but also by aroma release. Three sips of 5 g may lead to more sensing of the aroma in the retronasal area compared to one sip of 15 g. After each swallow, aroma compounds release into the retronasal area (Buettner, Beer, Hannig, Settles, & Schieberle, 2002). More sips and more swallowing result in more pulses of retronasal aroma release. The increased orosensory exposure may explain the reduction in food intake, as orosensory exposure to food has been found to be important in feedback signaling of satiation (Davis & Smith, 2009; Lavin et al., 2002b; Zijlstra et al., 2009). As far as we know, the effect of number of sips per gram on food intake was not studied before. The impact of a threefold increase in number of sips on food intake was greater than that of a threefold increase in the oral transit time. The results of this study suggest that reducing the sip size is probably more effective in reducing food intake than only prolonging the oral transit time. Consumption with relatively higher number of sips led to higher ratings for ‘‘I terminated consumption because I did not like the manner of consumption’’ (Table 2). This may be due to the increased effort that is associated with a higher number of smaller sips compared to a lower number of larger sips. The ‘‘manner of consumption’’ may also have influenced the size of the effect of number of sips on intake; subjects may have terminated consumption earlier in the ‘‘high number of sips’’ conditions because they did not like the manner of consumption. Nevertheless, the results of the present study suggest that a higher number of sips per gram are more satiating, because it led to a faster increase in fullness per consumed gram food (Fig. 3). It seems that subjects in the present study consumed until a certain state of satiety, as the mean ‘‘endpoint’’ for hunger was 20 mm and for fullness 75 mm on a 100 mm VAS (Fig. 3, Table 1). This certain state of satiety was faster reached when subjects consumed with a higher number of sips per gram, which may consequently have led to a lower food intake. In theory, eating rate is also affected by sip/bite frequency; the latter may also influence satiation. To keep the eating rate constant in this study, the sip frequency was three times higher for 5 g sips (HS conditions) than for 15 g sips (LS conditions). Results of a previous study (Bolhuis et al., 2012) clearly showed a positive contribution of sip size on intake, while there was hardly any contribution of sip frequency on intake when using Bayesian Modelling (unpublished results). This study (Bolhuis et al., 2012) used the same soup as test food, but subjects consumed the soup in a natural manner, with a spoon. Moreover, Zijlstra et al. (2009) investigated effects of bite size when both the bites of 5 and 15 g were administered at the same bite frequency, effectively increasing the eating rate by a factor three. Zijlstra et al. (2009) showed 30% lower intake when the food was consumed with smaller bites compared to larger bites, this indicates a clear effect of bite size when bite frequency was held constant. In accordance, studies that lowered the eating rate by lowering bite frequency failed to find a reduction in food intake (Lemmens, Martens, Born, Martens, & Westerterp-Plantenga, 2011; Yeomans, Gray, Mitchell, & True,
D.P. Bolhuis et al. / Food Quality and Preference 32 (2014) 234–240
239
Fig. 4. Mean rated taste intensity during 30 s, thus 30 g, of soup in each of the four conditions. The grey areas (AUC) represent the total perceived taste and thereby the total magnitude of orosensory exposure. In the ‘‘low number of sips’’ conditions, two sips of 15 g were administered in 30 s, at t = 30 and t = 45, each sip was swallowed after 3 s in the ‘‘Short-LS’’ condition and after 9 s in the ‘‘Long-LS’’ condition. In the ‘‘high number of sips’’ conditions, 6 sips were administered in 30 s (in pulses of 5 s), each sip was swallowed after 1 s in the ‘‘Short-HS’’ condition and after 3 s in the ‘‘Long-HS’’ condition. The maxima of the peaks were reached directly after swallowing. Short-LS = ‘‘Short transit time, low number of sips’’ condition. Long-LS = ‘‘Long transit time, low number of sips’’ condition. Short-HS = ‘‘Short transit time, high number of sips’’ condition. LongHS = ‘‘Long transit time, high number of sips’’ condition.
1997). Overall, decreasing the bite frequency in order to decrease the eating rate might be less effective in reducing the energy intake. Strengths of this study are the within subjects design and the tightly controlled experimental design, which allowed us to investigate underlying mechanisms of eating rate on food intake. However, this tightly controlled design is also a limitation of the study. In a natural way of eating, visual cues play an important role in satiation (Wansink, 2004), whereas in the present study, subjects were not able to see how much they consumed. Moreover, controlling the sip size and frequency, and the unnatural way of soup consumption, may have influenced the results to some extent. In a natural way of eating, eating rate is not linear but decelerate over the meal (Kissileff, Klingsberg, & Vanitallie, 1980; Zandian, Ioakimidis, Bergh, Brodin, & Sodersten, 2009). Nevertheless, ‘‘fullness’’ was the most important reason to terminate consumption in all conditions, which indicates that subjects consumed until they were satiated (Table 2). Lower food intake resulting from smaller bites/sips or longer oral transit time is probably not compensated during the rest of the day or over more days. It seems unlikely that people accurately compensate their energy intake as a result of changes in energy intake from a meal or a day as demonstrated by other studies (Caputo & Mattes, 1992; Levitsky & DeRosimo, 2010; Levitsky, Obarzanek, Mrdjenovic, & Strupp, 2005). Two studies (McGee et al., 2012; Walden, Martin, Ortego, Ryan, & Williamson, 2004), that used an oral device to decrease the bite sizes, showed that the device led to a reduction in meal size without changes on subjective ratings of satiety between meals compared to normal intake. More research, however, is needed to investigate effects of bite size and oral transit time on longer term energy intake. Textural food properties as viscosity and hardness influence the oral transit time and bite/sip size. A higher viscosity and harder
foods decrease the eating rate (De Wijk et al., 2008; Forde et al., 2012; Viskaal-van Dongen et al., 2011), and these properties are negatively correlated with sip/bite size and oral transit time (De Wijk et al., 2008; Forde et al., 2012). Forde et al. (2012), found a strong correlation (r = 0.98) between oral transit time and number of chews. Thus, harder foods that need more chews elicit a longer oral transit time, and have a higher satiating capacity, as also showed by Forde et al. (2012). In general, beverages are consumed with larger sips and shorter oral transit times than solid foods (De Wijk et al., 2008). The differences in oral processing characteristics may explain that beverages have weaker satiety responses than solid foods (Cassady, Considine, & Mattes, 2012; Houchins et al., 2012; Wolf, Bray, & Popkin, 2008; Zijlstra et al., 2008). Further research is needed to investigate how textural food properties influence oral processing and thereby satiation. In conclusion, higher number of sips and longer oral transit time per gram food resulted in reduced food intake, where the number of sips showed the greatest effect. In addition, a higher number of sips per gram food, thus smaller sips, led to faster increase in fullness per consumed gram food. This means that reducing the sip size is probably more effective in reducing food intake than only prolonging the oral transit time. Advices to consume with smaller sips or bites and to prolong the oral transit time may be helpful in body weight management. Moreover, designing foods that will be consumed with small sips or bites and long oral transit times may also be an effective tool to reduce food intake. More insight is needed in associations between food properties and oral processes in order to influence the satiating capacity of foods. Acknowledgements This study was supported by the Science and Technology Foundation of the Netherlands Organization for Scientific Research
240
D.P. Bolhuis et al. / Food Quality and Preference 32 (2014) 234–240
(NWO-STW), with co-financers: Unilever, Danone Nederland, Royal FrieslandCampina, and Top Institute Food and Nutrition (TIFN). We thank all who assisted in conducting the study; especially Inge van der Wurff, Janet Ottens and Els Siebelink; and the technical department of Wageningen University for the software used in this study, especially Hans Meijer. References Bolhuis, D. P., Lakemond, C. M. M., de Wijk, R. A., Luning, P. A., & de Graaf, C. (2011). Both longer oral sensory exposure to and higher intensity of saltiness decrease ad libitum food intake in healthy normal-weight men. Journal of Nutrition, 141(12), 2242–2248. Bolhuis, D. P., Lakemond, C. M. M., de Wijk, R. A., Luning, P. A., & de Graaf, C. (2012). Effect of salt intensity in soup on ad libitum intake and on subsequent food choice. Appetite, 58(1), 48–55. Buettner, A., Beer, A., Hannig, C., Settles, M., & Schieberle, P. (2002). Physiological and analytical studies on flavor perception dynamics as induced by the eating and swallowing process. Food Quality and Preference, 13(7–8), 497–504. Busch, J., Tournier, C., Knoop, J. E., Kooyman, G., & Smit, G. (2009). Temporal contrast of salt delivery in mouth increases salt perception. Chemical Senses, 34(4), 341–348. Caputo, F. A., & Mattes, R. D. (1992). Human dietary responses to covert manipulations of energy, fat, and carbohydrate in a midday meal. American Journal of Clinical Nutrition, 56(1), 36–43. Cassady, B. A., Considine, R. V., & Mattes, R. D. (2012). Beverage consumption, appetite, and energy intake: What did you expect? American Journal of Clinical Nutrition, 95(3), 587–593. Cecil, J. E., Francis, J., & Read, N. W. (1998). Relative contributions of intestinal, gastric, oro-sensory influences and information to changes in appetite induced by the same liquid meal. Appetite, 31(3), 377–390. Cecil, J. E., Francis, J., & Read, N. W. (1999). Comparison of the effects of a high-fat and high-carbohydrate soup delivered orally and intragastrically on gastric emptying, appetite, and eating behavior. Physiology & Behavior, 67(2), 299–306. Coelho, M. S., Passadore, M. D., Gasparetti, A. L., Bibancos, T., Prada, P. O., Furukawa, L. L., et al. (2006). High- or low-salt diet from weaning to adulthood: Effect on body weight, food intake and energy balance in rats. Nutrition Metabolism and Cardiovascular Diseases, 16(2), 148–155. Cordain, L., Eaton, S. B., Sebastian, A., Mann, N., Lindeberg, S., Watkins, B. A., et al. (2005). Origins and evolution of the western diet: Health implications for the 21st century. American Journal of Clinical Nutrition, 81(2), 341–354. Davis, J. D., & Smith, G. P. (2009). The conditioned satiating effect of orosensory stimuli. Physiology & Behavior, 97(3–4), 293–303. De Wijk, R. A., Zijlstra, N., Mars, M., De Graaf, C., & Prinz, J. F. (2008). The effects of food viscosity on bite size, bite effort and food intake. Physiology & Behavior, 95(3), 527–532. Forde, C. G., van Kuijk, N., Thaler, T., de Graaf, C., & Martin, N. (2012). Oral processing characteristics of solid savoury meal components, and relationship with food composition, sensory attributes and expected satiation. Appetite, 60, 208–219. French, S. J., & Cecil, J. E. (2001). Oral, gastric and intestinal influences on human feeding. Physiology & Behavior, 74(4–5), 729–734. Haber, G. B., Heaton, K. W., Murphy, D., & Burroughs, L. F. (1977). Depletion and disruption of dietary fibre: Effects on satiety, plasma-glucose, and seruminsulin. The Lancet, 310(8040), 679–682. Hill, S. W., & McCutcheon, N. B. (1984). Contributions of obesity, gender, hunger, food preference, and body size to bite size, bite speed, and rate of eating. Appetite, 5(2), 73–83. Hogenkamp, P. S., Mars, M., Stafleu, A., & de Graaf, C. (2010). Intake during repeated exposure to low- and high-energy-dense yogurts by different means of consumption. American Journal of Clinical Nutrition, 91(4), 841–847. Houchins, J. A., Burgess, J. R., Campbell, W. W., Daniel, J. R., Ferruzzi, M. G., McCabe, G. P., et al. (2012). Beverage vs. solid fruits and vegetables: Effects on energy intake and body weight. Obesity, 20(9), 1844–1850. Hulshof, K. F. A. M., Ocke, M. C., Van Rossum, C. T. M., Buurma-Rethans, E. J. M., Brants, H. A. M., Drijvers, J. M. M., et al. (2003). Results of the food consumption survey (No. RIVM report 350030002/2004). Bilthoven. Kissileff, H. R., Klingsberg, G., & Vanitallie, T. B. (1980). Universal eating monitor for continuous recording of solid or liquid consumption in man. American Journal of Physiology, 238(1), R14–R22. Lavin, J. H., French, S. J., & Read, N. W. (2002a). Comparison of oral and gastric administration of sucrose and maltose on gastric emptying rate and appetite. International Journal of Obesity, 26(1), 80–86.
Lavin, J. H., French, S. J., Ruxton, C. H. S., & Read, N. W. (2002b). An investigation of the role of oro-sensory stimulation in sugar satiety. International Journal of Obesity, 26(3), 384–388. Lemmens, S. G., Martens, E. A., Born, J. M., Martens, M. J., & Westerterp-Plantenga, M. S. (2011). Staggered meal consumption facilitates appetite control without affecting postprandial energy intake. Journal of Nutrition, 141(3), 482– 488. Levitsky, D. A., & DeRosimo, L. (2010). One day of food restriction does not result in an increase in subsequent daily food intake in humans. Physiology & Behavior, 99(4), 495–499. Levitsky, D. A., Obarzanek, E., Mrdjenovic, G., & Strupp, B. J. (2005). Imprecise control of energy intake: Absence of a reduction in food intake following overfeeding in young adults. Physiology & Behavior, 84(5), 669–675. Llewellyn, C. H., van Jaarsveld, C. H. M., Boniface, D., Carnell, S., & Wardle, J. (2008). Eating rate is a heritable phenotype related to weight in children. American Journal of Clinical Nutrition, 88(6), 1560–1566. Malik, V. S., Schulze, M. B., & Hu, F. B. (2006). Intake of sugar-sweetened beverages and weight gain: A systematic review. American Journal of Clinical Nutrition, 84(2), 274–288. Maruyama, K., Sato, S., Ohira, T., Maeda, K., Noda, H., Kubota, Y., et al. (2008). The joint impact on being overweight of self reported behaviors of eating quickly and eating until full: Cross sectional survey. British Medical Journal, 337. McGee, T. L., Grima, M. T., Hewson, I. D., Jones, K. M., Duke, E. B., & Dixon, J. B. (2012). First australian experiences with an oral volume restriction device to change eating behaviors and assist with weight loss. Obesity, 20(1), 126–133. Melanson, K. J., Greene, G. W., & Petty, A. J. (2011). Laboratory and free-living eating rates in young adults reporting to be slow, medium, or fast paced eaters. Obesity, 19, S101. Otsuka, R., Tamakoshi, K., Yatsuya, H., Murata, C., Sekiya, A., Wada, K., et al. (2006). Eating fast leads to obesity: Findings based on self-administered questionnaires among middle-aged japanese men and women. Journal of Epidemiology, 16(3), 117–124. Ponte, E. E., Melanson, K. J., & Greene, G. W. (2011). Characteristics associated with eating rates in young adults. Obesity, 19, S101. Sasaki, S., Katagiri, A., Tsuji, T., Shimoda, T., & Amano, K. (2003). Self-reported rate of eating correlates with body mass index in 18-y-old japanese women. International Journal of Obesity, 27(11), 1405–1410. Slimani, N., Deharveng, G., Southgate, D. A. T., Biessy, C., Chajes, V., van Bakel, M. M. E., et al. (2009). Contribution of highly industrially processed foods to the nutrient intakes and patterns of middle-aged populations in the european prospective investigation into cancer and nutrition study. European Journal of Clinical Nutrition, 63, S206–S225. Tracy, J. F., Logemann, J. A., Kahrilas, P. J., Jacob, P., Kobara, M., & Krugler, C. (1989). Preliminary observations on the effects of age on oropharyngeal deglutition. Dysphagia, 5, 90–94. Viskaal-van Dongen, M., Kok, F. J., & de Graaf, C. (2011). Eating rate of commonly consumed foods promotes food and energy intake. Appetite, 56(1), 25– 31. Walden, H. M., Martin, C. K., Ortego, L. E., Ryan, D. H., & Williamson, D. A. (2004). A new dental approach for reducing food intake. Obesity Research, 12(11), 1773–1780. Wansink, B. (2004). Environmental factors that increase the food intake and consumption volume of unknowing consumers. Annual Review of Nutrition, 24, 455–479. Weijzen, P. L. G., Smeets, P. A. M., & de Graaf, C. (2009). Sip size of orangeade: Effects on intake and sensory-specific satiation. British Journal of Nutrition, 102(7), 1091–1097. WHO (1985). Principles for the estimation of energy requirments. Energy and protein requirements. Report of a joint FAO/WHO/UNU expert consultation. Geneva: WHO, pp: 34–53. Wijlens, A. G. M., Erkner, A., Alexander, E., Mars, M., Smeets, P. A. M., & de Graaf, C. (2012). Effects of oral and gastric stimulation on appetite and energy intake. Obesity, 20(11), 2226–2232. Wolf, A., Bray, G. A., & Popkin, B. M. (2008). A short history of beverages and how our body treats them. Obesity Reviews, 9(2), 151–164. Yeomans, M. R., Gray, R. W., Mitchell, C. J., & True, S. (1997). Independent effects of palatability and within-meal pauses on intake and appetite ratings in human volunteers. Appetite, 29(1), 61–76. Zandian, M., Ioakimidis, L., Bergh, C., Brodin, U., & Sodersten, P. (2009). Decelerated and linear eaters: Effect of eating rate on food intake and satiety. Physiology & Behavior, 96(2), 270–275. Zijlstra, N., De Wijk, R. A., Mars, M., Stafleu, A., & De Graaf, C. (2009). Effect of bite size and oral processing time of a semisolid food on satiation. American Journal of Clinical Nutrition, 90(2), 269–275. Zijlstra, N., Mars, M., De Wijk, R. A., Westerterp-Plantenga, M. S., & De Graaf, C. (2008). The effect of viscosity on ad libitum food intake. International Journal of Obesity, 32(4), 676–683.
Contents lists available at ScienceDirect
Food Quality and Preference journal homepage: www.elsevier.com/locate/foodqual
Both a higher number of sips and a longer oral transit time reduce ad libitum intake Dieuwerke P. Bolhuis a,⇑, Catriona M.M. Lakemond a, Rene A. de Wijk b, Pieternel A. Luning a, Cees de Graaf c a b c
Food Quality and Design, Wageningen University, P.O. Box 17, 6700 AA Wageningen, The Netherlands Food and Biobased Research, Consumer Science & Intelligent Systems, Wageningen University and Research, P.O. Box 17, 6700 AA Wageningen, The Netherlands Division of Human Nutrition, Wageningen University, P.O. Box 8129, 6700 EV Wageningen, The Netherlands
a r t i c l e
i n f o
Article history: Received 25 January 2013 Received in revised form 2 October 2013 Accepted 2 October 2013 Available online 9 October 2013 Keywords: Orosensory exposure Oral transit time Satiation Sip size Food intake
a b s t r a c t Background: A higher eating rate leads to a higher food intake, possibly through shorter orosensory exposure to food. The transit time in the oral cavity and the number of bites or sips per gram (inversely related to bite or sip size) are main contributors that affect eating rate. The separate role of these two aspects on satiation and on orosensory exposure needs further clarification. Objective: The objective of the first study was to investigate contributions of the number of sips per gram (sips/g) and oral transit time per gram (s/g) on ad libitum intake. The objective of the second study was to investigate both aspects on the total magnitude of orosensory exposure per gram food. Methods: In study 1, 56 healthy male subjects consumed soup where the number of sips and oral transit time differed by a factor three respectively: 6.7 vs. 20 sips/100 g, and 20 vs. 60 s/100 g (2 2 cross-over design). Eating rate of 60 g/min was kept constant. In study 2, the effects of number of sips and oral transit time (equal as in study 1) on the total magnitude of orosensory exposure per gram soup were measured by time intensity functions by 22 different healthy subjects. Results: Higher number of sips and longer oral transit time reduced ad libitum intake by respectively 22% (F(1, 157) = 55.9, P < 0.001) and 8% (F(1, 157) = 7.4, P = 0.007). Higher number of sips led to faster increase in fullness per gram food (F(1, 157) = 24.1, P < 0.001) (study 1). Higher number of sips and longer oral transit time both increased the orosensory exposure per gram food (F(1, 63) = 23.8, P < 0.001) and (F(1, 63) = 19.0, P < 0.001), respectively (study 2). Conclusion: Higher number of sips and longer oral transit time reduced food intake, possibly through the increased the orosensory exposure per gram food. Designing foods that will be consumed with small sips or bites and long oral transit time may be effective in reducing energy intake. Crown Copyright Ó 2013 Published by Elsevier Ltd. All rights reserved.
1. Introduction The current food supply consists of a majority of highly processed foods that support fast intake of energy and minimal oral processing, like energy-yielding beverages and foods low in fiber content (Cordain et al., 2005; Malik, Schulze, & Hu, 2006; Slimani et al., 2009). Foods that can be consumed quickly (i.e., fast eating rate, grams or energy per time unit) may facilitate over-consumption. A number of studies have shown that higher eating rate leads to higher energy intake (Coelho et al., 2006; De Wijk, Zijlstra, Mars, De Graaf, & Prinz, 2008; Haber, Heaton, Murphy, & Burroughs, 1977; Melanson, Greene, & Petty, 2011; Ponte, Melanson, & Greene, 2011; Viskaal-van Dongen, Kok, & de Graaf, 2011; Zijlstra, Mars, De ⇑ Corresponding author. Address: Bomenweg 2, 6703 HD Wageningen, The Netherlands. Tel.: +31 317482063. E-mail address: [email protected] (D.P. Bolhuis).
Wijk, Westerterp-Plantenga, & De Graaf, 2008). Moreover, several studies suggest a positive relationship between eating rate and body weight status (Hill & McCutcheon, 1984; Llewellyn, van Jaarsveld, Boniface, Carnell, & Wardle, 2008; Maruyama et al., 2008; Otsuka et al., 2006; Sasaki, Katagiri, Tsuji, Shimoda, & Amano, 2003). Hogenkamp, Mars, Stafleu, and de Graaf (2010) compared ad libitum intake of yoghurts low and high in energy density with different eating rates. Food intake was primarily influenced by eating rate, and not by energy density. We propose that shorter sensory exposure of food in the oral cavity (i.e., orosensory exposure) explains the increased food intake at a fast eating rate. A number of studies show the importance of orosensory exposure to food in establishing feedback signals of satiation (Davis & Smith, 2009; Lavin, French, Ruxton, & Read, 2002b; Weijzen, Smeets, & de Graaf, 2009; Wijlens et al., 2012; Zijlstra, De Wijk, Mars, Stafleu, & De Graaf, 2009). Direct infusions of food into the stomach or duodenum, thus bypassing the orosensory exposure, give much
0950-3293/$ - see front matter Crown Copyright Ó 2013 Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.foodqual.2013.10.001
D.P. Bolhuis et al. / Food Quality and Preference 32 (2014) 234–240
weaker responses of satiation compared to oral intake (Cecil, Francis, & Read, 1998; Cecil, Francis, & Read, 1999; French & Cecil, 2001; Lavin, French, & Read, 2002a). There are two ways to slow down the eating rate (g/s), in order to increase the orosensory exposure per gram food and thereby decrease the amount of food intake, which are prolonging the transit time of the food in the oral cavity (i.e., oral transit time) or decreasing the bite/sip sizes. Previous studies have shown that bite size positively affected food intake and oral transit time negatively affected food intake (Bolhuis, Lakemond, de Wijk, Luning, & de Graaf, 2011; Weijzen et al., 2009; Zijlstra et al., 2009). However, in these studies the effects of bite size and oral transit time were confounded which did not allow the verification of individual contributions. In a natural way of eating, bite or sip size and oral transit time influence each other. Forde, van Kuijk, Thaler, de Graaf, and Martin (2012) showed that foods that are naturally consumed with smaller bite sizes are associated with longer oral transit times than foods that are consumed with larger bite sizes. Also within the same food, smaller bite/sip sizes lead to relatively longer oral transit time (Tracy et al., 1989). The separate contributions of oral transit time and bite/sip size to satiation are not known. Longer oral transit time is expected to increase the orosensory exposure to food and decrease food intake. However, smaller bites/sips may also increase the total magnitude of orosensory exposure. For example, three sips of 5 g instead of one sip of 15 g, means a more pulsating exposure to food, which may increase orosensory exposure per gram food, and thereby influencing satiation. We use the term number of sips per gram (inversely related to sip size: g/sip) in this paper, because sip size influences the oral transit time in a natural way of eating. Understanding the individual contributions of number of sips and oral transit time per gram food on intake will lead to more insight in the process of satiation. The aim of the first study is to investigate the separate effects of the number of sips and oral transit time on ad libitum intake at a constant eating rate. Possibly, a greater orosensory exposure to food, as a result of higher number of sips and longer oral transit time, may explain the effects on satiation. Therefore, a second study was executed to investigate whether and to what extent the number of sips and oral transit time affect the total magnitude of orosensory exposure to food.
2. Subjects and methods 2.1. Study 1 2.1.1. Subjects Fifty-nine male subjects were recruited for participation. Fiftysix subjects completed the study, two subjects dropped out before the start of the study and one subject missed three ad libitum intake sessions. Subjects were healthy, had a normal weight (BMI 18.5–25 kg/m2, mean ± SD: 22 ± 2 kg/m2), were aged between 18 and 35 years (mean ± SD: 22 ± 3 years) and liked creamy tomato soup (pleasantness score >5 on a 9-point hedonic scale). Exclusion criteria were restrained eating behavior (Dutch eating behavior questionnaire (DEBQ) score >2.89), following an energy-restricted diet during the last two months, gained or lost >5 kg weight during the last year, having a lack of appetite, smoking, suffering from gastrointestinal illness, diabetes, thyroid disease or any other endocrine disorder, hypertension and kidney diseases. Subjects were informed that the aim of the research was to investigate the effect of sip size on flavor perception of soup. This study was conducted according to the guidelines laid down in the Declaration of Helsinki and all procedures involving human subjects were approved by the Medical Ethical Committee of Wageningen University. All subjects signed an informed consent form before participation.
235
2.1.2. Experimental design The study consisted of a 2 2 crossover design. Subjects came five times to the lab, including a ‘‘practice session’’ (first session), to consume soup in each of the four conditions (Fig. 1). The eating rate was set in all four conditions at 60 g/min. The oral transit time was three times longer in the ‘‘long’’ conditions (Long-LS and LongHS) compared to the ‘‘short’’ conditions (Short-LS and Short-HS), 60 s/100 g vs. 20 s/100 g, respectively. The number of sips per gram was three times higher in the ‘‘high number of sips’’ conditions (HS) compared to the ‘‘low number of sips’’ conditions (LS), 20 vs. 6.7 sips/100 g, respectively. 2.1.3. Test foods Tomato soup was used as test product in this study. One kilogram of soup was made from 333 g sieved tomatoes (Heinz, Elst, The Netherlands), 662.7 g water, and 4.7 g salt (NaCl). The mixture was heated until 60 °C. The calculated nutrient composition from the used ingredients was: 0.57 g protein, 1.6 g carbohydrates, 0.03 g fat, 253 mg sodium and 38 kJ (9.1 kcal) energy per 100 g soup. Raisin buns (local bakery) were used as preload. The nutrient composition was: 8 g protein, 52 g carbohydrates, 3 g fat, 300 mg sodium and 1120 kJ (268 kcal) energy per 100 g, according to the Dutch Food Composition Database (NEVO, version 2009/1.0). Each raisin bun weighed 22 g (246 kJ). The number of raisin buns as preload was calculated for each subject at half of the energy provided by an average lunch in the Netherlands (Hulshof et al., 2003), that is equal to 11% energy of the daily energy need. The daily energy need for each subject was estimated by the Schofield I equation (WHO, 1985), taking into account: gender, age, weight and a physical activity level of 1.6. Thirty subjects received 5 buns and 27 subjects received 6 buns. Subjects were instructed to eat all the raisin buns they were served. 2.1.4. Control of sip sizes, intervals and swallowing To control the sips and intervals, subjects consumed the soup through a food-grade silicon tube that was connected to a peristaltic pump (Watson-Marlow, type 323Du, Watson-Marlow Bredel, Wilmington, MA, USA). The tube ended in a pan of soup that was placed on a balance (Kern, type 440-49A, KERN & Sohn GmbH, Balingen, Germany) to record the amount consumed. The pump, the pan and the balance were all located at the experimenters’ side of the sensory booths, thus subjects did not see the experimental setup. The moment the pump started driving, subjects heard an auditory signal to prepare them that they would receive soup in their mouths (Fig. 1). They heard a double auditory signal at the moment they had to swallow. Before the start of each session, subjects were instructed that it was very important to swallow at the double auditory signal. 2.1.5. Ad libitum intake sessions Subjects came four times during lunch for the ad libitum intake of soup, with one week in between sessions. The four conditions were presented at random. Subjects started with consumption of the preload that consisted of raisin buns. A preload was used so that subjects would be in a less hungry state prior to soup consumption. A hungry state may overrule sensory factors to terminate consumption (Bolhuis, Lakemond, de Wijk, Luning, & de Graaf, 2012). They were instructed to consume all served raisin buns and they were allowed to drink water. After that, subjects paused for 30 min. Subjects were allowed to study or read, but were not allowed to leave the sensory room. After the pause, subjects received a tube from which they had to consume soup. Subjects received instructions and questions via a computer screen. After answering several appetite
236
D.P. Bolhuis et al. / Food Quality and Preference 32 (2014) 234–240
Time (s) Short-LS: Long-LS: Short-HS: Long-HS:
0
5
10
15
15g
15g
15g
15g
5g
5g
5g
5g
5g
5g
5g
5g
Fig. 1. Sips and intervals in the four experimental conditions. In the ‘‘Short-LS’’ (short transit time, low number of sips) condition, sips of 15 g were exposed in 3 s (from the start of the administration of soup until swallowing) in pulses of 15 s. In the ‘‘Long-LS’’ (long transit time, low number of sips) condition, sips of 15 g were exposed in 9 s in pulses of 15 s. In the ‘‘Short-HS’’ (short transit time, high number of sips) condition, sips of 5 g were exposed in 1 s in pulses of 5 s. In the ‘‘Long-HS’’ (long transit time, high number of sips) condition, sips of 5 g were exposed in 3 s in pulses of 5 s. Subjects heard an auditory signal when they received the soup and a double auditory signal when they had to swallow.
and hedonic questions, as described below, subjects pushed a button on the screen to start soup consumption. They were instructed to terminate consumption when they had enough. The mean (±SD) initial temperature of the soup was 56 ± 5 °C and the mean end temperature was 51 ± 3 °C. Subjects were instructed to stay in the sensory booths for at least ten minutes after they started consuming the soup. After ten minutes, a visual warning signal popped up on the laptop screen to inform subjects the ten minutes had passed. This was done to prevent subjects from leaving the research area for other reasons than being satiated with the soup. 2.1.6. Appetite, hedonic ratings and questionnaires Just before soup consumption, subjects rated their feelings of hunger, fullness, prospective consumption and thirst. After that, subjects were served a small sample of 10 g soup and rated pleasantness and desire-to-eat the soup. The same questions were answered again at the end of the ad libitum intake. In addition, the same questions, except for thirst, were rated at random after consumption of every 75 g soup. All questions were answered by using a 100 mm VAS, on a scale from ‘‘not at all’’ at the left end to ‘‘very much’’ at the right end. At the end of the session, subjects indicated the reasons for terminating soup consumption. The subjects were asked to what extent they agreed with the propositions: ‘‘I terminated consumption because I was full’’; ‘‘I terminated consumption because the flavor of the soup was not pleasant anymore’’; and ‘‘I terminated consumption because I did not like the manner of consumption’’. The propositions were answered on a 5-point scale from ‘‘totally disagree’’ (1) to ‘‘completely agree’’ (5). 2.1.7. Standardization of the satiety state To standardize the satiety state, subjects always started the lunch session at the same time. They were instructed to consume the same breakfast and not to eat and only drink water before the lunch started. Moreover, they were asked to refrain from drinking one hour before the test started. After each test lunch, subjects answered questions about what they ate for breakfast and whether they ate or drank between breakfast and test lunch. To make sure subjects would consume the soup until they felt satiated; they were instructed not to eat until one hour after the test. 2.2. Study 2 2.2.1. Subjects Twenty-two different subjects (12 male, BMI 18.5–25 kg/m2, mean ± SD: 22 ± 2 kg/m2), aged between 18 and 35 years
(mean ± SD: 22 ± 2 years) participated in the second part of the study. The same exclusion and inclusion criteria as in the first study were used to recruit subjects, except for gender. These subjects were used to consume via a tube and peristaltic pump with controlled sip sizes and intervals, because they also participated in another study that used a similar experimental set up. 2.2.2. Time intensity measurements Time intensity (TI) measurements allow to measure the perceived taste intensity over a certain period of time (Busch, Tournier, Knoop, Kooyman, & Smit, 2009). TI measurements were made in each of the four different conditions used in study 1. The area under the curve (AUC) of the time intensity measurements was used as an indication about the total perceived taste, which is considered to represent the total magnitude of orosensory exposure to the taste. The conditions were offered in random order. Subjects had one practice session to train the procedure of the TI measurements. Subjects were instructed to rate their perceived taste intensity, inclusive aftertaste intensity, continuously for one minute on a VAS from 0 to 100 mm. Subjects were instructed not to rate the first sip to have a short time to adapt to the procedure, therefore, the AUC was calculated between t = 30 s and t = 60 s, in which 30 g of soup (53 ± 2 °C) was consumed. The area under the curve (AUC) represents the total magnitude of orosensory exposure. 2.2.3. Statistical analyses Statistical analyses were performed using SAS version 9.1.4 (SAS Institute Inc., Cary, NC, USA). Data are presented as means ± SDs, P-values <0.05 were considered significant. Effects of number of sips, oral transit time, and their interaction, on most outcome parameters were assessed in a mixed model ANOVA that included order and had subject as random factor. To assess effects on ad libitum intake and on ratings after and during intake, ratings of initial hunger and ‘‘I terminated consumption because I did not like the manner of consumption’’ were added to the mixed model ANOVA. Changes in appetite and hedonic ratings between before (initial) vs. after (post) ad libitum intake were assessed by two-tailed paired t-tests. The changes in appetite and hedonic ratings during ad libitum intake were fitted per subject in a linear model: y = a + b ⁄ intake. A linear model was chosen because this was the best fit in most individual curves. The curves shown in the results section were calculated from the mean intercepts and mean slopes of the individual plots. Effects of number of sips, oral transit time, and their interaction on the AUC of the taste intensity were assessed in two-way ANOVA
237
D.P. Bolhuis et al. / Food Quality and Preference 32 (2014) 234–240
that included subject. Bonferroni adjustments were used for all post hoc comparisons in the present study.
Table 1 Appetite and hedonic ratings from before and after ad libitum intake of soup in the four conditions1,2,3. Short-LS
Long-LS
Short-HS
Long-HS
Hunger Initial Post4
60 ± 20 18 ± 18
61 ± 21 19 ± 19
62 ± 19 21 ± 19
62 ± 19 21 ± 19
Fullness Initial Post4
33 ± 21 75 ± 20
32 ± 19 73 ± 18
30 ± 18 73 ± 20
31 ± 19 76 ± 17
Prospective consumption Initial 62 ± 24 24 ± 21 Post4
64 ± 20 25 ± 23
65 ± 17 29 ± 22
64 ± 19 27 ± 18
Thirst Initial Post5
59 ± 21 54 ± 27
54 ± 21 54 ± 25
55 ± 21 54 ± 23
57 ± 19 57 ± 24
Pleasantness Initial Post4
64 ± 20 51 ± 23
65 ± 18 51 ± 18
64 ± 18 50 ± 22
66 ± 18 52 ± 22
Desire-to-eat Initial Post4
65 ± 19 29 ± 21
64 ± 18 28 ± 21
65 ± 19 31 ± 20
67 ± 19 32 ± 22
3. Results 3.1. Study 1 3.1.1. Ad libitum intake of soup The number of sips (F(1, 157) = 55.9, P < 0.001) affected ad libitum intake; it was 22% lower in the ‘‘high number of sips’’ conditions compared to the ‘‘low number of sips’’ conditions (Fig. 2). Oral transit time also affected ad libitum intake (F(1, 157) = 7.4, P = 0.007); it was 8% lower in the ‘‘Long’’ conditions compared to the ‘‘Short’’ conditions. There was no interaction effect between number of sips and oral transit time on ad libitum intake (F(1, 157) = 0, P = 1.0). 3.1.2. Appetite and hedonic ratings before and after ad libitum intake Initial appetite ratings (i.e., hunger, fullness and prospective consumption) and hedonic ratings (i.e., pleasantness and desireto-eat the soup) did not differ between conditions (all P-values >0.59) (Table 1). Initial thirst did also not differ between conditions (P = 0.17). In the preload phase, subjects drank: 146 ± 38 g and this were not different between conditions (P = 0.20). After ad libitum intake, appetite and hedonic ratings changed in each condition compared to the initial ratings (all P-values <0.001), except for thirst (all P-values >0.10) (Table 1). Appetite, thirst and hedonic ratings after intake were not affected by oral transit time (all P-values >0.44), or by number of sips (all P-values >0.21), and there were no interaction effects (all P-values >0.26). 3.1.3. Appetite and hedonic ratings during ad libitum intake Fig. 3 shows the linear curves of the changes in rated hunger (A) and fullness (B) as a function of intake. Higher number of sips led to faster decrease in hunger (F(1, 157) = 9.1, P = 0.003) (Fig. 3A); faster increase in fullness (F(1, 157) = 24.1, P < 0.001) (Fig. 3B); faster decrease in ratings of prospective consumption (F(1, 157) = 6.6, P = 0.011); and faster decrease in desire-to-eat (F(1, 157) = 5.7, P = 0.018). Oral transit time did not significantly affect the decrease in hunger (F(1, 157) = 1.1, P = 0.31) (Fig. 3A); the increase in fullness (F(1, 157) = 2.2, P = 0.14) (Fig. 3B); and the decrease in desire-to-eat (F(1, 157) = 1.0, P = 0.17). However, longer oral transit time resulted in faster decrease in prospective consumption (F(1, 157) = 4.5, P = 0.035). Pleasantness was not affected by number of sips or oral transit time (both P-values >0.33). None of the changes in appetite and hedonic ratings showed an interaction effect between number of sips and oral transit time (all P-values >0.30).
1 Values are means ± SDs. Values in a row with different superscript letters are significantly different (P < 0.05). 2 Rated on a 100 mm VAS. 3 Short-LS = ‘‘Short transit time, low number of sips’’ condition. Long-LS = ‘‘Long transit time, low number of sips’’ condition. Short-HS = ‘‘Short transit time, high number of sips’’ condition. Long-HS = ‘‘Long transit time, high number of sips’’ condition. 4 Significant differences between initial and post appetite and hedonic ratings in all conditions: all P-values <0.001. 5 No significant difference between initial and post thirst ratings.
3.1.4. Reasons to terminate consumption ‘‘I terminated consumption because I was full’’ was the most important reason in all conditions for terminating consumption (Table 2). The reasons for terminating consumption: ‘‘I was full’’ or ‘‘flavor not pleasant anymore’’ were not affected by number of sips or oral transit time (all P-values >0.38). Ratings of ‘‘I terminated consumption because I did not like the manner of consumption’’ were higher in high number of sips conditions compared to low number of sips conditions (F(1, 159) = 14.8, P < 0.001), but were not affected by oral transit time (F(1, 159) = 1.6, P = 0.21). 3.2. Study 2 3.2.1. Area under the curve of perceived taste intensity of 30 g soup Fig. 4 shows the mean perceived taste during 30 s, which is equal to consumption of 30 g in each condition. The mean AUC of each condition is 953 ± 528 for Short-LS; 1199 ± 544 for Long-LS; 1230 ± 561 for Short-HS; and 1494 ± 658 for Long-HS. Both higher number of sips (F(1, 63) = 23.8, P < 0.001) and longer oral transit time (F(1, 63) = 19.0, P < 0.001) led to a greater AUC, suggesting more orosensory exposure per consumed gram food. The taste of the soup is highest in intensity directly after swallowing, shown by the top of the peaks (Fig. 4). The mean heights of the peak were 71 ± 19 mm in the ‘‘Short-LS’’ condition; 71 ± 17 mm in the ‘‘Long-LS’’ condition; 63 ± 22 mm in the ‘‘Short-HS’’ condition; and 65 ± 22 mm in the ‘‘Long-HS’’ condition (calculated from the individual curves). Higher number of sips resulted in lower heights of the peaks (F(1, 63) = 17.8, P < 0.001). Longer oral transit time did not affect the mean height of the peaks (F(1, 63) = 0.5, P = 0.47). 4. Discussion
Fig. 2. Means (+SD) of ad libitum intakes of soup. A higher number of sips (P < 0.001), and longer oral transit time (P = 0.007), reduced ad libitum intake. Mean values with different letters are significantly different between conditions.
Previous studies indicated that a higher eating rate leads to a higher food intake when eating in a natural way (Coelho et al.,
238
D.P. Bolhuis et al. / Food Quality and Preference 32 (2014) 234–240
Short-LS
Hunger (mm)
100
(a)
Long-LS Short-HS
75
Long-HS
50 25 0 0
100 200 300 400 500 600 700 800
Intake (g)
Fullness (mm)
100
(b)
75 50 25 0 0
100 200 300 400 500 600 700 800
Intake (g) Fig. 3. Linear functions of rated hunger (A) and fullness (B) against intake in the four conditions. Higher number of sips led to a faster decrease in hunger (P = 0.003), and a faster increase in fullness (P < 0.001) per consumed gram food. Longer oral transit time did not significantly affect the appetite ratings. The dots visualize the mean ad libitum intake in each condition. Short-LS = ‘‘Short transit time, low number of sips’’ condition. Long-LS = ‘‘Long transit time, low number of sips’’ condition. Short-HS = ‘‘Short transit time, high number of sips’’ condition. LongHS = ‘‘Long transit time, high number of sips’’ condition.
Table 2 The reasons to terminate soup consumption in the four conditions1,2,3.
‘‘I was full’’ ‘‘Flavor not pleasant anymore’’ ‘‘Manner of consumption’’
Short-LS3
Long-LS
Short-HS
Long-HS
4.0 ± 1.1 2.6 ± 1.2
4.0 ± 1.1 2.6 ± 1.1
4.0 ± 1.1 2.6 ± 1.1
3.9 ± 1.0 2.5 ± 1.3
2.4 ± 1.1a
2.6 ± 1.1ab
2.9 ± 1.2b
3.0 ± 1.4b
1 Values are means ± SDs. 1Values in a row with different superscript letters are significantly different (P < 0.05). 2 The propositions were answered on a 5-point scale from ‘‘totally disagree’’ (1) to ‘‘completely agree’’ (5). 3 Short-LS = ‘‘Short transit time, low number of sips’’ condition. Long-LS = ‘‘Long transit time, low number of sips’’ condition. Short-HS = ‘‘Short transit time, high number of sips’’ condition. Long-HS = ‘‘Long transit time, high number of sips’’ condition.
2006; Haber et al., 1977; Melanson et al., 2011; Ponte et al., 2011; Viskaal-van Dongen et al., 2011). We have investigated the underlying mechanisms on how the factors that influence eating rate (e.g., number of sips, oral transit time) affect satiation. The results show that both higher number of sips per gram food and longer oral transit time per gram food independently reduced food intake by 22% and 8%, respectively (Fig. 2). In accordance, consumption with higher number of sips per gram food was associated with a faster increase in subjective fullness (Fig. 3). Higher number of sips and longer oral transit time both contribute independently to greater total magnitude of sensory exposure per gram food (Fig. 4). Longer transit time of food in the oral cavity
is associated with sensing the taste over a longer period of time, which explains the increased orosensory exposure (greater AUC, Fig. 4). In this study, oral transit time was varied within the same food. Within our food range, oral transit time from different taste qualities and textures may differently affect oral sensory exposure, e.g., bitter compounds often lead to longer after taste. In theory, equal oral transit times of different taste qualities may differently affect orosensory exposure and possibly satiation. Interestingly, higher number of sips per gram (thus smaller sips), also result in greater orosensory exposure per gram food. One sip of 5 g almost reached the same taste intensity as one sip of 15 g (top of the peaks, Fig. 4). This means that three sips of 5 g result in a greater orosensory exposure than one sip of 15 g (greater AUC, Fig. 4). In addition, the orosensory exposure may not only be affected by the taste but also by aroma release. Three sips of 5 g may lead to more sensing of the aroma in the retronasal area compared to one sip of 15 g. After each swallow, aroma compounds release into the retronasal area (Buettner, Beer, Hannig, Settles, & Schieberle, 2002). More sips and more swallowing result in more pulses of retronasal aroma release. The increased orosensory exposure may explain the reduction in food intake, as orosensory exposure to food has been found to be important in feedback signaling of satiation (Davis & Smith, 2009; Lavin et al., 2002b; Zijlstra et al., 2009). As far as we know, the effect of number of sips per gram on food intake was not studied before. The impact of a threefold increase in number of sips on food intake was greater than that of a threefold increase in the oral transit time. The results of this study suggest that reducing the sip size is probably more effective in reducing food intake than only prolonging the oral transit time. Consumption with relatively higher number of sips led to higher ratings for ‘‘I terminated consumption because I did not like the manner of consumption’’ (Table 2). This may be due to the increased effort that is associated with a higher number of smaller sips compared to a lower number of larger sips. The ‘‘manner of consumption’’ may also have influenced the size of the effect of number of sips on intake; subjects may have terminated consumption earlier in the ‘‘high number of sips’’ conditions because they did not like the manner of consumption. Nevertheless, the results of the present study suggest that a higher number of sips per gram are more satiating, because it led to a faster increase in fullness per consumed gram food (Fig. 3). It seems that subjects in the present study consumed until a certain state of satiety, as the mean ‘‘endpoint’’ for hunger was 20 mm and for fullness 75 mm on a 100 mm VAS (Fig. 3, Table 1). This certain state of satiety was faster reached when subjects consumed with a higher number of sips per gram, which may consequently have led to a lower food intake. In theory, eating rate is also affected by sip/bite frequency; the latter may also influence satiation. To keep the eating rate constant in this study, the sip frequency was three times higher for 5 g sips (HS conditions) than for 15 g sips (LS conditions). Results of a previous study (Bolhuis et al., 2012) clearly showed a positive contribution of sip size on intake, while there was hardly any contribution of sip frequency on intake when using Bayesian Modelling (unpublished results). This study (Bolhuis et al., 2012) used the same soup as test food, but subjects consumed the soup in a natural manner, with a spoon. Moreover, Zijlstra et al. (2009) investigated effects of bite size when both the bites of 5 and 15 g were administered at the same bite frequency, effectively increasing the eating rate by a factor three. Zijlstra et al. (2009) showed 30% lower intake when the food was consumed with smaller bites compared to larger bites, this indicates a clear effect of bite size when bite frequency was held constant. In accordance, studies that lowered the eating rate by lowering bite frequency failed to find a reduction in food intake (Lemmens, Martens, Born, Martens, & Westerterp-Plantenga, 2011; Yeomans, Gray, Mitchell, & True,
D.P. Bolhuis et al. / Food Quality and Preference 32 (2014) 234–240
239
Fig. 4. Mean rated taste intensity during 30 s, thus 30 g, of soup in each of the four conditions. The grey areas (AUC) represent the total perceived taste and thereby the total magnitude of orosensory exposure. In the ‘‘low number of sips’’ conditions, two sips of 15 g were administered in 30 s, at t = 30 and t = 45, each sip was swallowed after 3 s in the ‘‘Short-LS’’ condition and after 9 s in the ‘‘Long-LS’’ condition. In the ‘‘high number of sips’’ conditions, 6 sips were administered in 30 s (in pulses of 5 s), each sip was swallowed after 1 s in the ‘‘Short-HS’’ condition and after 3 s in the ‘‘Long-HS’’ condition. The maxima of the peaks were reached directly after swallowing. Short-LS = ‘‘Short transit time, low number of sips’’ condition. Long-LS = ‘‘Long transit time, low number of sips’’ condition. Short-HS = ‘‘Short transit time, high number of sips’’ condition. LongHS = ‘‘Long transit time, high number of sips’’ condition.
1997). Overall, decreasing the bite frequency in order to decrease the eating rate might be less effective in reducing the energy intake. Strengths of this study are the within subjects design and the tightly controlled experimental design, which allowed us to investigate underlying mechanisms of eating rate on food intake. However, this tightly controlled design is also a limitation of the study. In a natural way of eating, visual cues play an important role in satiation (Wansink, 2004), whereas in the present study, subjects were not able to see how much they consumed. Moreover, controlling the sip size and frequency, and the unnatural way of soup consumption, may have influenced the results to some extent. In a natural way of eating, eating rate is not linear but decelerate over the meal (Kissileff, Klingsberg, & Vanitallie, 1980; Zandian, Ioakimidis, Bergh, Brodin, & Sodersten, 2009). Nevertheless, ‘‘fullness’’ was the most important reason to terminate consumption in all conditions, which indicates that subjects consumed until they were satiated (Table 2). Lower food intake resulting from smaller bites/sips or longer oral transit time is probably not compensated during the rest of the day or over more days. It seems unlikely that people accurately compensate their energy intake as a result of changes in energy intake from a meal or a day as demonstrated by other studies (Caputo & Mattes, 1992; Levitsky & DeRosimo, 2010; Levitsky, Obarzanek, Mrdjenovic, & Strupp, 2005). Two studies (McGee et al., 2012; Walden, Martin, Ortego, Ryan, & Williamson, 2004), that used an oral device to decrease the bite sizes, showed that the device led to a reduction in meal size without changes on subjective ratings of satiety between meals compared to normal intake. More research, however, is needed to investigate effects of bite size and oral transit time on longer term energy intake. Textural food properties as viscosity and hardness influence the oral transit time and bite/sip size. A higher viscosity and harder
foods decrease the eating rate (De Wijk et al., 2008; Forde et al., 2012; Viskaal-van Dongen et al., 2011), and these properties are negatively correlated with sip/bite size and oral transit time (De Wijk et al., 2008; Forde et al., 2012). Forde et al. (2012), found a strong correlation (r = 0.98) between oral transit time and number of chews. Thus, harder foods that need more chews elicit a longer oral transit time, and have a higher satiating capacity, as also showed by Forde et al. (2012). In general, beverages are consumed with larger sips and shorter oral transit times than solid foods (De Wijk et al., 2008). The differences in oral processing characteristics may explain that beverages have weaker satiety responses than solid foods (Cassady, Considine, & Mattes, 2012; Houchins et al., 2012; Wolf, Bray, & Popkin, 2008; Zijlstra et al., 2008). Further research is needed to investigate how textural food properties influence oral processing and thereby satiation. In conclusion, higher number of sips and longer oral transit time per gram food resulted in reduced food intake, where the number of sips showed the greatest effect. In addition, a higher number of sips per gram food, thus smaller sips, led to faster increase in fullness per consumed gram food. This means that reducing the sip size is probably more effective in reducing food intake than only prolonging the oral transit time. Advices to consume with smaller sips or bites and to prolong the oral transit time may be helpful in body weight management. Moreover, designing foods that will be consumed with small sips or bites and long oral transit times may also be an effective tool to reduce food intake. More insight is needed in associations between food properties and oral processes in order to influence the satiating capacity of foods. Acknowledgements This study was supported by the Science and Technology Foundation of the Netherlands Organization for Scientific Research
240
D.P. Bolhuis et al. / Food Quality and Preference 32 (2014) 234–240
(NWO-STW), with co-financers: Unilever, Danone Nederland, Royal FrieslandCampina, and Top Institute Food and Nutrition (TIFN). We thank all who assisted in conducting the study; especially Inge van der Wurff, Janet Ottens and Els Siebelink; and the technical department of Wageningen University for the software used in this study, especially Hans Meijer. References Bolhuis, D. P., Lakemond, C. M. M., de Wijk, R. A., Luning, P. A., & de Graaf, C. (2011). Both longer oral sensory exposure to and higher intensity of saltiness decrease ad libitum food intake in healthy normal-weight men. Journal of Nutrition, 141(12), 2242–2248. Bolhuis, D. P., Lakemond, C. M. M., de Wijk, R. A., Luning, P. A., & de Graaf, C. (2012). Effect of salt intensity in soup on ad libitum intake and on subsequent food choice. Appetite, 58(1), 48–55. Buettner, A., Beer, A., Hannig, C., Settles, M., & Schieberle, P. (2002). Physiological and analytical studies on flavor perception dynamics as induced by the eating and swallowing process. Food Quality and Preference, 13(7–8), 497–504. Busch, J., Tournier, C., Knoop, J. E., Kooyman, G., & Smit, G. (2009). Temporal contrast of salt delivery in mouth increases salt perception. Chemical Senses, 34(4), 341–348. Caputo, F. A., & Mattes, R. D. (1992). Human dietary responses to covert manipulations of energy, fat, and carbohydrate in a midday meal. American Journal of Clinical Nutrition, 56(1), 36–43. Cassady, B. A., Considine, R. V., & Mattes, R. D. (2012). Beverage consumption, appetite, and energy intake: What did you expect? American Journal of Clinical Nutrition, 95(3), 587–593. Cecil, J. E., Francis, J., & Read, N. W. (1998). Relative contributions of intestinal, gastric, oro-sensory influences and information to changes in appetite induced by the same liquid meal. Appetite, 31(3), 377–390. Cecil, J. E., Francis, J., & Read, N. W. (1999). Comparison of the effects of a high-fat and high-carbohydrate soup delivered orally and intragastrically on gastric emptying, appetite, and eating behavior. Physiology & Behavior, 67(2), 299–306. Coelho, M. S., Passadore, M. D., Gasparetti, A. L., Bibancos, T., Prada, P. O., Furukawa, L. L., et al. (2006). High- or low-salt diet from weaning to adulthood: Effect on body weight, food intake and energy balance in rats. Nutrition Metabolism and Cardiovascular Diseases, 16(2), 148–155. Cordain, L., Eaton, S. B., Sebastian, A., Mann, N., Lindeberg, S., Watkins, B. A., et al. (2005). Origins and evolution of the western diet: Health implications for the 21st century. American Journal of Clinical Nutrition, 81(2), 341–354. Davis, J. D., & Smith, G. P. (2009). The conditioned satiating effect of orosensory stimuli. Physiology & Behavior, 97(3–4), 293–303. De Wijk, R. A., Zijlstra, N., Mars, M., De Graaf, C., & Prinz, J. F. (2008). The effects of food viscosity on bite size, bite effort and food intake. Physiology & Behavior, 95(3), 527–532. Forde, C. G., van Kuijk, N., Thaler, T., de Graaf, C., & Martin, N. (2012). Oral processing characteristics of solid savoury meal components, and relationship with food composition, sensory attributes and expected satiation. Appetite, 60, 208–219. French, S. J., & Cecil, J. E. (2001). Oral, gastric and intestinal influences on human feeding. Physiology & Behavior, 74(4–5), 729–734. Haber, G. B., Heaton, K. W., Murphy, D., & Burroughs, L. F. (1977). Depletion and disruption of dietary fibre: Effects on satiety, plasma-glucose, and seruminsulin. The Lancet, 310(8040), 679–682. Hill, S. W., & McCutcheon, N. B. (1984). Contributions of obesity, gender, hunger, food preference, and body size to bite size, bite speed, and rate of eating. Appetite, 5(2), 73–83. Hogenkamp, P. S., Mars, M., Stafleu, A., & de Graaf, C. (2010). Intake during repeated exposure to low- and high-energy-dense yogurts by different means of consumption. American Journal of Clinical Nutrition, 91(4), 841–847. Houchins, J. A., Burgess, J. R., Campbell, W. W., Daniel, J. R., Ferruzzi, M. G., McCabe, G. P., et al. (2012). Beverage vs. solid fruits and vegetables: Effects on energy intake and body weight. Obesity, 20(9), 1844–1850. Hulshof, K. F. A. M., Ocke, M. C., Van Rossum, C. T. M., Buurma-Rethans, E. J. M., Brants, H. A. M., Drijvers, J. M. M., et al. (2003). Results of the food consumption survey (No. RIVM report 350030002/2004). Bilthoven. Kissileff, H. R., Klingsberg, G., & Vanitallie, T. B. (1980). Universal eating monitor for continuous recording of solid or liquid consumption in man. American Journal of Physiology, 238(1), R14–R22. Lavin, J. H., French, S. J., & Read, N. W. (2002a). Comparison of oral and gastric administration of sucrose and maltose on gastric emptying rate and appetite. International Journal of Obesity, 26(1), 80–86.
Lavin, J. H., French, S. J., Ruxton, C. H. S., & Read, N. W. (2002b). An investigation of the role of oro-sensory stimulation in sugar satiety. International Journal of Obesity, 26(3), 384–388. Lemmens, S. G., Martens, E. A., Born, J. M., Martens, M. J., & Westerterp-Plantenga, M. S. (2011). Staggered meal consumption facilitates appetite control without affecting postprandial energy intake. Journal of Nutrition, 141(3), 482– 488. Levitsky, D. A., & DeRosimo, L. (2010). One day of food restriction does not result in an increase in subsequent daily food intake in humans. Physiology & Behavior, 99(4), 495–499. Levitsky, D. A., Obarzanek, E., Mrdjenovic, G., & Strupp, B. J. (2005). Imprecise control of energy intake: Absence of a reduction in food intake following overfeeding in young adults. Physiology & Behavior, 84(5), 669–675. Llewellyn, C. H., van Jaarsveld, C. H. M., Boniface, D., Carnell, S., & Wardle, J. (2008). Eating rate is a heritable phenotype related to weight in children. American Journal of Clinical Nutrition, 88(6), 1560–1566. Malik, V. S., Schulze, M. B., & Hu, F. B. (2006). Intake of sugar-sweetened beverages and weight gain: A systematic review. American Journal of Clinical Nutrition, 84(2), 274–288. Maruyama, K., Sato, S., Ohira, T., Maeda, K., Noda, H., Kubota, Y., et al. (2008). The joint impact on being overweight of self reported behaviors of eating quickly and eating until full: Cross sectional survey. British Medical Journal, 337. McGee, T. L., Grima, M. T., Hewson, I. D., Jones, K. M., Duke, E. B., & Dixon, J. B. (2012). First australian experiences with an oral volume restriction device to change eating behaviors and assist with weight loss. Obesity, 20(1), 126–133. Melanson, K. J., Greene, G. W., & Petty, A. J. (2011). Laboratory and free-living eating rates in young adults reporting to be slow, medium, or fast paced eaters. Obesity, 19, S101. Otsuka, R., Tamakoshi, K., Yatsuya, H., Murata, C., Sekiya, A., Wada, K., et al. (2006). Eating fast leads to obesity: Findings based on self-administered questionnaires among middle-aged japanese men and women. Journal of Epidemiology, 16(3), 117–124. Ponte, E. E., Melanson, K. J., & Greene, G. W. (2011). Characteristics associated with eating rates in young adults. Obesity, 19, S101. Sasaki, S., Katagiri, A., Tsuji, T., Shimoda, T., & Amano, K. (2003). Self-reported rate of eating correlates with body mass index in 18-y-old japanese women. International Journal of Obesity, 27(11), 1405–1410. Slimani, N., Deharveng, G., Southgate, D. A. T., Biessy, C., Chajes, V., van Bakel, M. M. E., et al. (2009). Contribution of highly industrially processed foods to the nutrient intakes and patterns of middle-aged populations in the european prospective investigation into cancer and nutrition study. European Journal of Clinical Nutrition, 63, S206–S225. Tracy, J. F., Logemann, J. A., Kahrilas, P. J., Jacob, P., Kobara, M., & Krugler, C. (1989). Preliminary observations on the effects of age on oropharyngeal deglutition. Dysphagia, 5, 90–94. Viskaal-van Dongen, M., Kok, F. J., & de Graaf, C. (2011). Eating rate of commonly consumed foods promotes food and energy intake. Appetite, 56(1), 25– 31. Walden, H. M., Martin, C. K., Ortego, L. E., Ryan, D. H., & Williamson, D. A. (2004). A new dental approach for reducing food intake. Obesity Research, 12(11), 1773–1780. Wansink, B. (2004). Environmental factors that increase the food intake and consumption volume of unknowing consumers. Annual Review of Nutrition, 24, 455–479. Weijzen, P. L. G., Smeets, P. A. M., & de Graaf, C. (2009). Sip size of orangeade: Effects on intake and sensory-specific satiation. British Journal of Nutrition, 102(7), 1091–1097. WHO (1985). Principles for the estimation of energy requirments. Energy and protein requirements. Report of a joint FAO/WHO/UNU expert consultation. Geneva: WHO, pp: 34–53. Wijlens, A. G. M., Erkner, A., Alexander, E., Mars, M., Smeets, P. A. M., & de Graaf, C. (2012). Effects of oral and gastric stimulation on appetite and energy intake. Obesity, 20(11), 2226–2232. Wolf, A., Bray, G. A., & Popkin, B. M. (2008). A short history of beverages and how our body treats them. Obesity Reviews, 9(2), 151–164. Yeomans, M. R., Gray, R. W., Mitchell, C. J., & True, S. (1997). Independent effects of palatability and within-meal pauses on intake and appetite ratings in human volunteers. Appetite, 29(1), 61–76. Zandian, M., Ioakimidis, L., Bergh, C., Brodin, U., & Sodersten, P. (2009). Decelerated and linear eaters: Effect of eating rate on food intake and satiety. Physiology & Behavior, 96(2), 270–275. Zijlstra, N., De Wijk, R. A., Mars, M., Stafleu, A., & De Graaf, C. (2009). Effect of bite size and oral processing time of a semisolid food on satiation. American Journal of Clinical Nutrition, 90(2), 269–275. Zijlstra, N., Mars, M., De Wijk, R. A., Westerterp-Plantenga, M. S., & De Graaf, C. (2008). The effect of viscosity on ad libitum food intake. International Journal of Obesity, 32(4), 676–683.