Dichotomous Thinking Toward Food As A Mediator Between .

1y ago
933.16 KB
78 Pages
Last View : Today
Last Download : 1m ago
Upload by : Karl Gosselin

Dichotomous Thinking Toward Food as a Mediator BetweenEating Behavior and BMIbySohee LeeA Thesis Presented in Partial Fulfillmentof the Requirements for the DegreeMaster of ScienceApproved April 2018 by theGraduate Supervisory Committee:Elias Robles-Sotelo, ChairPerla VargasDeborah HallARIZONA STATE UNIVERSITYMay 2018

ABSTRACTLong-term results of dietary weight loss interventions are not promising, withrates of weight loss maintenance at a mere 20%. Psychological factors related toweight maintenance include setting unrealistic weight goals, poor problem-solvingskills, low self-efficacy, dichotomous thinking, and external locus of control. Theability to maintain a stable bodyweight over time has been associated with optimalhealth outcomes, lower stress levels, and higher general well-being. Dichotomousthinking has been associated with overeating and increased bodyweight. Cognitiverestraint, disinhibition, and hunger are three dimensions of human eating behaviorthat appear to be important to understanding weight loss maintenance. Individualswho attempt to maintain their bodyweight via dietary restraint mechanisms are moresusceptible to excessive eating episodes. Disinhibition has been found to be thestrongest predictor of weight gain, while the research on the association betweenhunger and bodyweight is mixed. This study sought to evaluate the relationshipbetween dichotomous thinking toward food and various eating behaviors (bingeeating, cognitive restraint, disinhibition, and hunger). A multiple regression analysisrevealed that binge eating, cognitive restraint, disinhibition, and hunger were eachsignificant unique predictors of higher body mass index (BMI). Higher levels ofhunger predicted lower BMI, controlling for cognitive restraint, disinhibition, andbinge eating. Mediation analyses revealed that dichotomous thinking mediated therelationships between binge eating and BMI, cognitive restraint and BMI, anddisinhibition and BMI. Further analysis revealed that binge eating mediated therelationship between dichotomous thinking and BMI, indicating that thinking of foodi

in black-and-white could lead to higher rates of binge eating, and the excess calorieconsumption could lead to increased BMI. The study findings suggest that a strongfocus should be made to promote a more flexible attitude toward food in an effort toimprove weight loss maintenance in the population.ii

To my parents: Thank you for fostering and always encouraging my love of learning.iii

ACKNOWLEDGMENTSA heartfelt thank you goes to my advisor, Dr. Elias Robles-Sotelo, who met withme on multiple occasions and helped guide the formation and creation of this thesis. I amalso grateful to Dr. Perla Vargas and Dr. Deborah Hall for all the time and energy theyspent combing through and providing feedback on my work.iv

TABLE OF CONTENTSPageLIST OF TABLES . viiLIST OF FIGURES . viiiCHAPTER1INTRODUCTION . 1Weight Loss Maintenance . 2Yo-Yo Dieting . 4Dichotomous Thinking Toward Food . 6Binge Eating . 9Cognitive Restraint, Disinhibition, and Hunger . 12Study Aims . 162METHODS . 16Participants . 16Measures . 17Procedure . 183RESULTS . 184DISCUSSION . 225LIMITATIONS . 276CONCLUSION . 29REFERENCES . 31v


LIST OF TABLESTablePage1. Participant Demographics and Descriptive Statistics . 512. Distribution of BMI Categories (N 3,893) . 533. Correlations for Variables in Hypotheses 1 and 2 . 554. Multiple Regression for Binge Eating, Cognitive Restraint, Disinhibition, andHunger Predicting BMI. 57vii

LIST OF FIGURESFigurePage1. Dichotomous Thinking Toward Food as a Mediator of the Relationship BetweenBinge Eating and BMI Controlling for Cognitive Restraint, Disinhibition, Hunger,and Yo-Yo Dieting. 592. Dichotomous Thinking Toward Food as a Mediator of the Relationship BetweenCognitive Restraint and BMI Controlling for Binge Eating, Disinhibition, Hunger,and Yo-Yo Dieting. 613. Dichotomous Thinking Toward Food as a Mediator of the Relationship BetweenDisinhibition and BMI Controlling for Binge eating, Cognitive Restraint, Hunger,and Yo-Yo Dieting. 634. Dichotomous Thinking Toward Food as a Mediator of Relationship BetweenHunger and BMI Controlling for Binge Eating, Cognitive Restraint, Disinhibition,and Yo-Yo Dieting. 655. Binge Eating as a Mediator of Relationship Between Dichotomous ThinkingToward Food and BMI Controlling for Cognitive Restraint, Disinhibition,Hunger, and Yo-Yo Dieting . 676. Dichotomous Thinking Toward Food as a Mediator of the Relationship BetweenYo-Yo Dieting and BMI Controlling for Binge Eating, Cognitive Restraint,Disinhibition, and Hunger. 69viii

INTRODUCTIONLong-term results of dietary weight loss interventions are not promising. Mostparticipants experience success in the short-term, yet the majority are not able to maintainweight loss in the long-term (Anderson, Konz, Frederich, & Wood, 2001). Rates ofweight loss maintenance, defined as losing at least 10% of initial bodyweight andmaintaining the loss for at least one year, are at a mere 20% (Befort et al., 2008; Stevens,Truesdale, McClain, & Cai, 2006; Wing & Hill, 2001; Wing & Phelan, 2005). Althoughlong-term weight loss maintenance appears to be a function of the extent to which healthbehavior is maintained, more research is needed to elucidate how, specifically,environmental and psychological factors contribute to weight maintenance.In order to understand how long-term weight loss maintenance is established, it isimportant to identify the specific health behaviors involved, which include continuedconsumption of a low-energy diet, high levels of physical activity, and greater frequencyof self-monitoring behaviors such as regular weighing (Butryn, Phelan, Hill, & Wing,2007; Mai et al., 2018; McGuire, Wing, Klem, Lang, & Hill,1999; Shick et al., 1998).The longer individuals are actively participating in a weight loss program, the higher theiradherence to weight loss behaviors, but as soon as treatment ends, they begin to regainthe weight (Perri, Nezu, Patti, & McCann, 1989; McGuire, Wing, Klem, Lang, & Hill,1999; Franz et al., 2007). To counter this, it has been suggested by Wing, Tate, Gorin,Raynor, and Fava (2006) that individuals who lose weight continue to participate inweight maintenance programs to help improve maintenance of weight loss.There is no agreed-upon definition of weight loss maintenance in the scientificliterature as the issue is complex and multi-faceted. Of course, the term suggests that1

some amount of weight has been lost, and subsequently the weight loss has beensustained over a period of time. One definition is, ‘losing at least 5% of baseline bodyweight between baseline and follow-up, and maintaining that weight or less for a furthertwo years,’ (Crawford, Jeffery, & French, 2000) and another is, ‘achieving an intentionalweight loss of at least 10% of initial body weight and maintaining this body weight for atleast one year’ (Wing & Hill, 2000). Still other researchers classify successful “losers” asthose who regain fewer than two body mass index points following weight loss (Cuntz,Leibbrand, Ehrig, Shaw, & Fichter, 2001). In recent years, MacLean and colleagues(2015) have identified successful weight loss to mean participants achieving a 5-10%weight loss at the end of the intervention. To further complicate the problem, differentstudies utilize different criteria to measure weight change following a period of weightloss: mean weight regain, percent regain, or percent still maintaining a given amount(Perri et al., 2008; Svetkey et al., 2008; Wing, Tate, Gorin, Raynor, & Fava, 2006).Understanding what factors contribute to making weight loss maintenance sodifficult is a key component to comprehending how to not only lose the weight but keepit off over the long-term. This knowledge can influence the advice given and strategiesemployed by practitioners during the weight loss and weight maintenance phases.Weight Loss MaintenanceNumerous studies have examined the behavioral variables associated with weightloss maintenance (Perri et al., 1984; Sciamanna et al., 2011; Wadden et al., 2011).Although weight regain appears to be in part a result of discontinued adherence to weightcontrol behaviors (Foster & Wadden, 1994; Jeffrey et al., 2000), there are other factors totake into consideration, including psychological factors such as setting unrealistic weight2

goals, poor problem-solving skills, low self-efficacy (Byrne, 2002), dichotomousthinking (Byrne, Cooper, & Fairburn, 2004), as well as an external locus of control(Williams, Grow, Freedman, Ryan, & Deci, 1996) and lack of use of positivereinforcement as a motivator (Williams, Earle-Richardson, Greeth, Scribani, & Monie,2016). Indeed, psychological factors associated with regaining weight after dietinginclude feelings of failure for not achieving the expected weight loss, dissatisfaction withthe weight achieved, a tendency to evaluate self-worth in terms of weight and shape, lackof vigilance with regard to weight control, a dichotomous black-and-white thinking style,and the tendency to use eating to regulate mood (Byrne, Cooper, & Fairburn, 2004). Incontrast, successful weight maintenance is related to realistic goal setting, consistentroutine and self-monitoring, minimizing deprivation, and effective coping skills (McKee,Ntoumanis, & Smith, 2013). Further, results from Brantley and colleagues (2014)suggested that perceptions of social support, functional and perceived physical health,functional and perceived mental health, and stress modestly predicted subjects’ weightloss maintenance.People’s decisions regarding behavior change partly depend on favorableexpectations of the future outcomes – specifically, expecting significant weight loss –while the maintenance of health behaviors depends strongly on the achieved results(Rothman, 2000). In both cases, however, the new behavior usually demands additionalcognitive effort and limits to otherwise preferred foods and activities. Importantly,feelings of deprivation adversely affect long-term weight loss maintenance by decreasingadherence to the dietary protocol (Urban, White, Anderson, Curry, & Kristal, 1992).3

Therefore, reducing dietary deprivation within any weight loss intervention may be aneffective strategy to promote lasting results.Carels and colleagues (2014) compared a weight loss program emphasizingreducing unhealthy relationships with food, body image dissatisfaction, and internalizedweight bias called New Perspectives, to a weight loss program emphasizingenvironmental modification and habit formation and disruption called Transforming YourLife. At the end of the 12-week dietary intervention, weight loss outcomes were similar;however, Transforming Your Life patients were significantly more effective atmaintaining weight loss through the 6-month post-intervention follow-up. This studysuggests that long-term weight loss success might depend not on finding the perfect dietprogram so much as on understanding the nuances of psychological factors and healthbehaviors involved, and then applying those lessons so that individuals can modify theirown lifestyle for the better.Yo-Yo DietingWeight cycling, or yo-yo dieting, is defined as intentionally losing weight throughenergy restrictions and then gaining the weight back over time (Field et al., 1999).Reports of weight cycling prevalence in the United States vary from anywhere between20-55% in women and 20-35% in men (Foreyt et al., 1995; Lahti-Koski, Männistö,Pietinen, & Vartiainen, 2005), largely because there is no operational definition of weightcycling (Atkinson et al., 1994).The research on the relationship between yo-yo dieting and adverse healthoutcomes is mixed. Wadden et al. (1992) found zero association between total number ofdieting attempts or total lifetime weight loss and self-reported depression, dietary4

restraint, disinhibition, hunger, or binge eating. Other studies support the finding that ahistory of weight cycling has no impact on subsequent weight loss (Wadden et al., 1992;van Dale & Saris, 1989; Beeson et al., 1989; Palm, Schram, Swarts, van Schothorst, &Keijer, 2017). In contrast, Miller and Parsonage (1975) suggested that a long history ofdieting made weight loss less likely in women, and other studies have similarly suggestedthat participation in weight loss programs was a negative predictor of subsequent weightloss (Jeffrey et al., 1984; Jeffrey et al., 1985). Smith and Wing (1991) found that weightloss decreased dramatically from the first diet cycle to the second in a year-longbehavioral weight reduction program. Other research has shown relationships betweenweight cycling and cardiovascular morbidity and mortality (Brownell et al., 1994; Lissneret al., 1990; Blair et al., 1993; Hamm et al., 1989; Kannel et al., 1991) and increased riskof hypertension (Guagnano et al, 2000).Wallner et al. (2004) showed that weight cycling induces a redistribution of bodyfat in women to a more android fat pattern, in which body fat is distributed around thetrunk and upper body, compared to a gynoid fat pattern, in which body fat is distributedaround the lower body, in normal weight, non-weight cycling women. These findings areconsistent with previous studies that found increased waist-to-hip ratio, which is the mostimportant and consistent factor associated with risk of hyperextension, and increasedupper body fat distribution due to weight cycling (Guagnano, Ballone, & Pace-Palitti,2000; Rodin, Radke-Sharpe, Rebuffe-Scrive, & Greanwood,1990). Of greater concern,Montani, Schutz, and Dulloo (2015) have identified numerous adverse healthconsequences associated with weight cycling, including but not limited to: increasedanxiety and depression as well as increased risk of morbidities such as type 2 diabetes,5

hypertension, and cancer. Most importantly, these health effects were seen in not justobese individuals but also those who were engaged in repeated dieting and weightcycling, including athletes and other individuals with normal body weight (Montani,Schutz, & Dulloo, 2015). More recently, in a 12-year cohort study, Madigan, Pavey,Daley, Jolly, & Brown (2018) found that weight cycling was associated with greaterlong-term weight gain and poorer mental health outcomes, and data collected byPacanowski and colleagues (2018) suggest that poorer psychological function precedesweight instability.Despite the conflicting research, the ability to maintain a stable bodyweight overtime has been associated with optimal health outcomes (van der Kooy, Leenen, Seidell,Deurenberg, & Hautvast, 1993) in addition to lower stress levels and higher general wellbeing (Foreyt et al., 1995). Thus, further increasing our understanding of how and whyweight cycling occurs is important.Dichotomous Thinking Toward FoodDichotomous thinking is characterized by the propensity to view situations inpolarized either-or categories. The ability to think in this manner can be helpful forexpedient, straightforward decision making in everyday life (Oshio, 2009). Indeed,dichotomous outcomes are common in society – vote yes or no, defendants are guilty ornot guilty, and so on. However, while this all-or-nothing approach can be helpful inreaching conclusions quickly, it can also have negative psychological outcomes and lessthan desirable impacts on eating behavior.This polarized thinking has also been found to be associated with depression(Teasdale et al., 2001), anxiety (Clark, 1986), and traits associated with eating disorders6

and perfectionistic tendencies (Shafran, Cooper, & Fairburn, 2002). Those who setunrealistically high expectations for themselves when it comes to their diet andbodyweight typically evaluate themselves based on their adherence to said standards(Shafran, Cooper, & Fairburn, 2002). Though research at this time is limited,dichotomous thinking has been implicated as a psychopathological process in eatingdisorders (Fairburn, Cooper, & Shafran, 2003; Garner & Bemis, 1982). On the mo

Yo-Yo Dieting Weight cycling, or yo-yo dieting, is defined as intentionally losing weight through energy restrictions and then gaining the weight back over time (Field et al., 1999). Reports of weight cycling prevalence in the United States vary from anywhere between 20-55% in women and 2