Research reportSweetened drink and snacking cues in adolescents. A study using ecological momentary assessment☆
Introduction
The proportion of adolescents in the US who are overweight or obese is a critical public health concern (Ogden, Carroll, Curtin, Lamb, & Flegal, 2010). Nationwide, the prevalence of being overweight and obese (BMI ⩾ 85 percentile) is 34.2% among all adolescents 12–19 years of age (Ogden et al., 2010). The prevalence of obesity is especially high among lower income families. In California, where this study was conducted, teens 12–17 years of age from lower income families (<300% Federal Poverty Level) are at elevated levels of risk of being overweight (20.7%) or obese (15.4%) compared to teens from higher income families (>300% FPL: 11.8% overweight and 7.5% obese) according to the California Health Interview Survey (2012). Adolescents with high BMI are at increased risk for chronic diseases including cardiovascular disease and type 2 diabetes mellitus among others (Freedman et al., 2007, Knight, 2011).
Obesity is, of course, closely related to diet (Baranowski et al., 2000, Goran, 2001, Hill et al., 2000, Mendlein et al., 2000), including snacking and sweetened beverage consumption. Evidence is accumulating that consumption of sugar-sweetened beverages is linked to increased body weight (Malik, Schulze, & Hu, 2006) and increased risk of medical problems including diabetes (Centers for Disease Control and Prevention, 2011, Malik et al., 2010, Vartanian et al., 2007). In addition, there is evidence that excess consumption of energy-dense snack foods is associated with an unhealthy weight gain (Piernas and Popkin, 2011, Swinburn et al., 2004). The current study uses real-world, real-time data collected via Ecological Momentary Assessment (EMA: (Shiffman, 2009, Stone and Shiffman, 1994)) techniques to identify environmental and intrapersonal cues associated with habitual consumption of high calorie snacks and sweetened beverages.
Over time, some dietary behaviors may evolve through learning into habits that are initiated by situational cues (stimulus-driven habits). Research in neuroscience (Knowlton et al., 1996, Yin and Knowlton, 2006a, Yin and Knowlton, 2006b), memory (Nelson & Goodmon, 2003), social psychology (Bargh and Williams, 2006, Dijksterhuis et al., 2005), and research on appetitive behavior (LaBar et al., 2001) have consistently shown the importance of cues in spontaneously triggering habits and related cognitions. A situation such as a location, social setting, or mood may become a cue for a behavior after repeated co-occurrence with that behavior, especially when the behavior has immediate rewards such as consumption of palatable foods. A friend, for example, may not be associated with having a snack initially, but after repeatedly meeting the friend after school to buy chips and a soda, the sight of the friend may become a cue for the behavior. The current study was especially interested in stimulus–response (S–R) habits formed by this type of instrumental learning that may be highly resistant to modification (Yin & Knowlton, 2006b). After a strong (S–R) habit is formed, the cue (stimulus) can initiate the behavior (response) regardless of anticipated outcomes (Wood and Neal, 2007, Yin and Knowlton, 2006a). The habit is likely to persist even after the outcome contingency has changed (i.e., negative consequences are encountered due to excessive weight gain) and despite learning new facts about obesity (e.g., through traditional education). Habit is supported by neural systems that reflect a set of processes classified as procedural memory, which is independent from declarative or explicit memory (e.g., memory for new facts through education), as documented in a series of studies on multiple brain systems (Knowlton et al., 1996, Ryan and Cohen, 2003). Frameworks incorporating these findings and non-declarative processes have been increasingly applied to a range of appetitive behaviors (for recent reviews, see (Stacy et al., 2010, Stacy and Wiers, 2010)). Poor dietary habits that are allowed to continue unchecked can lead to a lifetime struggle with obesity and related chronic diseases. It is vitally important therefore to identify cues that trigger maladaptive dietary habits to facilitate the design of interventions that will disrupt the cue–behavior link and encourage healthy diets.
Participants may not be aware of the cues that trigger their habits (Dijksterhuis et al., 2005, Knowlton et al., 1996, Wood and Neal, 2007). Self-reports on the causes of behavior may not fully reflect the cue-habit process (Bargh, 2005, Dijksterhuis et al., 2005), and this makes it difficult to assess the cue-habit link with traditional surveys where participants are asked to introspect about their behavior in the previous days or months. It is possible that assessing behavior in real time may better identify cue–behavior patterns that are not captured by conventional, retrospective questionnaires (Shiffman, 1993, Shiffman et al., 2008). Real time assessment may be particularly helpful in identifying these linkages if the procedures simply record cue and behavior co-occurrences, facilitating the study of their empirical linkages without requiring participant awareness of the link.
EMA (Shiffman et al., 2008) is an assessment technique with several key features: (a) participants respond to questions during their typical activities in the real-world environment, which permits researchers to generalize the findings to the real lives of the participants, (b) responses of the participants focus on their current situation, activities, and feelings, which can eliminate recall bias associated with assessments that ask for recall of events over longer periods of time, (c) questions are asked at strategically selected times to capture situations, activities, and feelings during target events such as eating and, for purposes of comparison, during random times when participants may not be doing the target activity (i.e., non-eating situations), and (d) data are typically collected multiple times in a day and over several days to capture how behavior changes across situations and to accumulate multiple instances of the events of interest. The design and technology may differ by study question and behavior, but all EMA studies collect data repeatedly from participants on their current state or situation in their natural environment. Researchers then may examine how situations, activities, and feeling states influence the behavior of interest.
EMA has been widely used over a period of more than 20 years to measure health behaviors and antecedents related to smoking (Shiffman, 2005), exercise (Dunton et al., 2007, Gorely et al., 2007, Hausenblas et al., 2008), and diet (Glanz and Murphy, 2007, Greeno et al., 2000, le Grange et al., 2001), with considerable evidence supporting its validity and utility among adults and youth. EMA procedures have shown less recall bias than retrospective questionnaires among adolescents and children as young as 7 years (van den Brink, Bandell-Hoekstra, & Abu-Saad, 2001), and previous EMA studies among children and adolescents include, for example, physical activity (Dunton et al., 2011, Dunton et al., 2007) smoking cessation (Gwaltney, Bartolomei, Colby, & Kahler, 2008), and mood (Weinstein and Mermelstein, 2007, Weinstein et al., 2007). Prior EMA studies have successfully examined dietary behavior but largely among participants recruited from obese populations and/or those with eating disorders (Engel et al., 2009, Greeno et al., 2000, Hilbert and Tuschen-Caffier, 2007, Smyth et al., 2009).
The objective of the current study was to empirically identify situations or cues associated with unhealthy snacking and sugar-sweetened beverage consumption among participants recruited from public high schools in lower SES neighborhoods. We anticipated that, although the development of habits is likely to be idiosyncratic, common life experiences across participants would result in some common cue–behavior associations that can be identified using EMA. We also anticipated that multiple situations may cue snacking behaviors and that some of those situations might be related but not co-occurring. We reasoned, for example, that development of a habitual response to one cue (e.g., feeling happy) would not necessarily exclude the development of the same habitual response to a related cue (e.g., feeling sad). The analyses contrasted situational factors associated with consumption of these unhealthy drinks and snacks with those associated with non-sweetened drinks, healthy snacks, meals, and non-eating or drinking occasions. The study focused on cues associated with the consumption of sugar-sweetened drinks, sweet snacks, and salty snacks, which are associated with weight gain and related medical problems (Carels et al., 2001, Centers for Disease Control and Prevention, 2011, Malik et al., 2010, Piernas and Popkin, 2011, Vartanian et al., 2007). In addition, consumption of these food items is more likely to be under the control of adolescents, compared to meals prepared by adults in the home, making sweetened drinks and energy dense snacks ideal targets for behavioral interventions among adolescents.
Section snippets
Participants
Participants were recruited from high schools that met the following criteria: (a) minimum of 25% of students in a free or reduced price meal program, (b) minimum of 25% Hispanic students, (c) maximum of 25% Asian students, (d) minimum enrollment of 100, (e) included students between 14 and 17 years of age, and (f) were within 30 miles of the assessment site in San Dimas, CA. The intention was to recruit a sample of students from lower income families at elevated risk of being overweight or obese
Assessments and compliance
Participants (n = 158) were monitored for an average of 6.70 (SD = 0.25) days; 98.73% were monitored for the full 7 days. A total of 3992 momentary assessments were recorded: 1868 random prompts (1.69 per participant day), 2124 eating event reports including having a drink only, eating a snack, or having a meal (1.92 per participant day), and 1043 evening reports were also recorded (0.94 per participant day). Participants completed 71% of the assessments solicited by random prompting, and 95% of
Discussion
This is the first study, of which we are aware, to use Ecological Momentary Assessment (EMA) to identify contexts and cues associated with the consumption of sweetened drinks and sweet and salty snacks in a non-clinical sample of adolescents. The objective was to find cue–behavior links that over time might have become stimulus–response (S–R) habits; the identification of these cue–behavior associations has practical implications for the development of dietary behavior interventions. Relatively
References (77)
- et al.
Do implementation intentions help to eat a healthy diet? A systematic review and meta-analysis of the empirical evidence
Appetite
(2011) - et al.
Physical activity and nutrition in children and youth. An overview of obesity prevention
Preventive Medicine
(2000) - et al.
On the road to obesity. Television viewing increases intake of high-density foods
Physiology & Behavior
(2006) - et al.
Ecological momentary assessment of temptation and lapse in dieting
Eating Behaviors
(2001) - et al.
Effects of food-cue exposure on dieting-related goals. A limitation to counteractive-control theory
Appetite
(2008) - et al.
The unconscious consumer. Effects of environment on consumer behavior
Journal of Consumer Psychology
(2005) - et al.
Eating behavior in obese bed, obese non-bed, and non-obese control participants. A naturalistic study
Behaviour Research and Therapy
(2009) - et al.
The specificity of restrained versus unrestrained eaters’ responses to food cues. General desire to eat, or craving for the cued food?
Appetite
(2003) - et al.
Cardiovascular risk factors and excess adiposity among overweight children and adolescents. The bogalusa heart study
The Journal of Pediatrics
(2007) Metabolic precursors and effects of obesity in children. A decade of progress, 1990–1999
American Journal of Clinical Nutrition
(2001)
Dietary fat intake and regulation of energy balance. Implications for obesity
Journal of Nutrition
Daily hassles and emotional eating in obese adolescents under restricted dietary conditions. The role of ruminative thinking
Appetite
Social norms and diet in adolescents
Appetite
Intake of sugar-sweetened beverages and weight gain. A systematic review
American Journal of Clinical Nutrition
Physical activity and nutrition in children and youth. Opportunities for performing assessments and conducting interventions
Preventive Medicine
How visibility and convenience influence candy consumption
Appetite
Factors associated with sugar-sweetened beverage intake among United States high school students
The Journal of Nutrition
Self-reported academic grades and other correlates of sugar-sweetened soda intake among US adolescents
Journal of the Academy of Nutrition and Dietetics
Increased portion sizes from energy-dense foods affect total energy intake at eating occasions in US children and adolescents. Patterns and trends by age group and sociodemographic characteristics, 1977–2006
The American Journal of Clinical Nutrition
Caloric restriction in the presence of attractive food cues. External cues, eating, and weight
Physiology and Behavior
Binge eating in adolescents. Its relation to behavioural problems and family-meal patterns
Eating Behaviors
Tween sex differences in snacking preferences during television viewing
Journal of the American Dietetic Association
Peer influence on snacking behavior in adolescence
Appetite
Breaking habits with implementation intentions. A test of underlying processes
Personality & Social Psychology Bulletin
When planning is not enough. Fighting unhealthy snacking habits by mental contrasting with implementation intentions (MCII)
European Journal of Social Psychology
Bypassing the will. Toward demystifying the nonconscious control of social behavior
The automaticity of social life
Current Directions in Psychological Science
Characteristics associated with older adolescents who have a television in their bedrooms
Pediatrics
Impact of television viewing patterns on fruit and vegetable consumption among adolescents
Pediatrics
An ecological momentary assessment of relapse crises in dieting
Journal of Consulting & Clinical Psychology
A global measure of perceived stress
Journal of Health and Social Behavior
Investigating children’s physical activity and sedentary behavior using ecological momentary assessment with mobile phones
Obesity (Silver Spring, Md.)
Mapping the social and physical contexts of physical activity across adolescence using ecological momentary assessment
Annals of Behavioral Medicine
The essential guide to effect sizes. Statistical power, meta-analysis, and the interpretation of research results
Dietary assessment and monitoring in real time
Cited by (0)
- ☆
Acknowledgments: Support for this research was provided by the National Institutes of Health (U01 HL097839-01). We especially wish to thank James Pike and Kim Massie for their excellent management of the project and all of the research assistants that help recruit participants and collect data.