The Role of Self-Monitoring in the Maintenance of Weight Loss Success


Material Information

The Role of Self-Monitoring in the Maintenance of Weight Loss Success
Physical Description:
1 online resource (51 p.)
Laitner, Melissa H
University of Florida
Place of Publication:
Gainesville, Fla.
Publication Date:

Thesis/Dissertation Information

Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Psychology, Clinical and Health Psychology
Committee Chair:
Committee Co-Chair:
Committee Members:


Subjects / Keywords:
obesity -- self-monitoring
Clinical and Health Psychology -- Dissertations, Academic -- UF
Psychology thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation


Self-monitoring has been shown to be a crucial part of initial weight loss success in behavioral interventions. However, less is known about the role of self-monitoring in the long run. Research suggests participants will gain back approximately one third or more of weight lost within the year following treatment. The current study examined the role of self-monitoring during an initial 6-month intervention phase (PH1) and a 12-month extended care phase (PH2) in a group of 167 obese older women (M: BMI = 37.0 kg/m2, age = 59.9 years). Cluster analysis identified three groups of participants with low, moderate, and high success rates for weight loss during PH1 and PH2. A one-way ANOVA revealed no significant differences in self-monitoring frequency between groups during PH1 (p = .645), but significant differences between all three groups during PH2 (p = .001). High success participants completed the most self-monitoring records, followed by the moderate group. The low success group completed the least number of records. Furthermore, self-monitoring during PH2 significantly mediated the relationship between follow-up session attendance and percent weight loss (95% CI -.004, -.001, p < .001), and this mediational relationship remained significant when controlling for age, education, race, and ethnicity (95% CI -.003, -.001). While previous research has shown the benefits of self-monitoring during behavioral interventions, these results highlight the importance of continuing self-monitoring after the initial phase of treatment.
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility:
by Melissa H Laitner.
Thesis (M.S.)--University of Florida, 2014.
Co-adviser: WIENS,BRENDA A.

Record Information

Source Institution:
Rights Management:
Applicable rights reserved.
lcc - LD1780 2014
System ID:

This item is only available as the following downloads:

Full Text




2 2014 Melissa Hope Laitner


3 To my mother, for inspiring me to pursue a graduate degree in a field I love, and to my father, for reminding me that some of the most important education occurs outside the classroom


4 ACKNOWLEDGMENTS I would like to thank my mentor, Dr. Michael Perri, for his dedication to his students and his encouragement throughout the past two years. I would also like to thank the members of my supervisory committee, Dr. Dei dre Pereira, Dr. Catherine Price, and Dr. Brenda Wiens, for their time and eff ort. I am grateful to the members of the University of Florida Weight Management Lab, for their endless advice and ability to make even data entry enjoyable. And finally, I would like to thank Frank Avery, for his support day in and day out.


5 TABLE OF CON TENTS page ACKNOWLEDGMENTS .................................................................................................. 4 LIST OF TABLES ............................................................................................................ 7 LIST OF FIGURES .......................................................................................................... 8 ABSTRACT ..................................................................................................................... 9 CHAPTER 1 INTRODUCTION .................................................................................................... 11 Overvi ew ................................................................................................................. 11 Prevalence and Health Risks ........................................................................... 11 Lifestyle Interventions for Obesity .................................................................... 12 Typical Response to Treatment ........................................................................ 14 Self Monitoring and the Model of Self Regulation ............................................ 16 The Utilization of Self Monitoring in Behavior al Weight Loss Treatment .......... 18 Specific Aims and Measures ................................................................................... 20 Specific Aim 1 ................................................................................................... 20 Specific Aim 2 ................................................................................................... 21 Specific Aim 3 ................................................................................................... 21 Exploratory Aim ................................................................................................ 22 2 MATERIALS AND METHODS ................................................................................ 23 Treatment of Obesity in Underserved Rural Settings (TOURS) Study .................... 23 Participants ............................................................................................................. 24 Procedures ............................................................................................................. 26 Measures ................................................................................................................ 26 Statistical Analysis .................................................................................................. 28 Specific Aim 1 ................................................................................................... 28 Specific Aim 2 ................................................................................................... 29 Specific Aim 3 ................................................................................................... 29 Exploratory Aim ................................................................................................ 30 3 RESULTS ............................................................................................................... 31 Baseline Character istics ......................................................................................... 31 Specific Aim 1 ......................................................................................................... 31 Specific Aim 2 ......................................................................................................... 32 Specific Aim 3 ......................................................................................................... 32 Exploratory Aim ...................................................................................................... 33


6 4 DISCUSSION ......................................................................................................... 38 APPENDIX SELF MONITORING RECORD ............................................................. 45 LIST OF REFERENCES ............................................................................................... 46 BIOGRAPHIC AL SKETCH ............................................................................................ 51


7 LIST OF TABLES Table page 3 1 Participant baseline characteristics .................................................................... 35 3 2 Cluster data ........................................................................................................ 36 3 3 Number of completed food records submitted by clusters .................................. 36


8 LIST OF FIGURES Figure page 3 1 Results of cluster analysis by participant weight change .................................... 37


9 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science THE ROLE OF SELF MONITORING IN THE MAINTENANCE OF WEIGHT LOSS SUCCESS By Melissa Hope Laitner May 2014 Chair: Michael G. Perri Major: Psychology Self monitoring has been shown to be a crucial part of initial weight loss success in behavioral interventions. However, less is known about the role of self monitoring in the long run. Research suggests participants will gain back approximately one third or more of weight lost within the year following treatment. The current study examined the role of self monitoring during an initial 6month intervention phase (PH1) and a 12month extended care phase (PH2) in a group of 167 obese older women (MSD: BMI = 37.05.1 kg/m2, age = 59.96.2 years). Cluster analysis identified three groups of participants with low, moderate, and high success rates for weight loss during PH1 and PH2. A one way ANOVA revealed no significant differences in self mo nitoring frequency between groups during PH1 ( p = .645), but significant differences between all three groups during PH2 ( p = .001). High success participants completed the most self monitoring records, followed by the moderate group. The low success group completed the least number of records. Furthermore, self monitoring during PH2 significantly mediated the relationship between follow up session attendance and percent weight loss (95% CI [ .004, .001], p < .001), and this mediational relationship remained significant when controlling for age, education, race, and ethnicity (95% CI [ .003, -


10 .001]). While previous research has shown the benefits of self monitoring during behavioral interventions, these results highlight the importance of continuing self mo nitoring after the initial phase of treatment.


11 CHAPTER 1 INTRODUCTION Overview Behavioral scientists continue to search for ef fective treatments for obesity, but it is clear that current treatments lack the results necessary in order to help the majority of individuals both lose clinically significant amounts of weight and maintain this weight loss long term. Furthermore, research suggests that participants respond differently to even the most effective types of treat ment. For the majority of overweight and obese individuals enrolled in behavioral treatment programs, however, we know that consistent self monitoring of caloric intake and weight change has been shown to be a vital method in achieving initial weight loss The effectiveness of self monitoring has yet to be fully assessed as a long term approach to weight management following a behavioral intervention. Therefore, the current study first sought to examine how participants could be categorized based on weight loss success during a behavioral intervention and a longterm extended care period, and then to identify the role of self monitoring during both phases. Prevalence and Health Risks The most recent data from the National Health and Nutrition Examination S urvey suggests that over a third of individuals aged 20 and above in the United States (34.9%) met criteria for obesity during the time period from 20112012, indicating that they have a Body Mass Index (BMI) greater than or equal to 30 kg/m2 (Ogden, Carroll, Kit, & Flegal, 2013). Other recent estimates suggest that an additional third of the population is defined as overweight, with a BMI greater than or equal to 25 kg/m2. These numbers indicate that a majority of the United Sates adult population fal l s ou tside of the normal


12 weight category (BMI = 1825 kg/m2; Flegal, Carroll, Kit, & Ogden, 2012). Adults who meet criteria for overweight and obesity are also at risk for various other chronic health conditions, including diabetes mellitus, cardiovascular disease and hypertension, gallbladder disease, arthritis, and some forms of cancer (Bray 2003). Generally, prevalence rates for obesity related diseases increase as body weight increases, and this is especially true for both diabetes mellitus and hypertension (Must, Spadano, Coakley, Field, Colditz, & Dietz, 1999). Moreover, as compared to individuals in the normal weight category, those who meet criteria for overweight and obesity are at greater risk for premature death (Flegal, Garubard, Williamson, & Gail, 2005). This may be in part due to the association between obesity and high risk diseases such as cardiovascular and kidney diseases, diabetes mellitus, and obesity related cancers (Mokdad, Marks, Stroup, & Gerberding, 2004). Lifestyle Interventions for Obes ity Research shows that overweight and obesity are risk factors for both morbidity and mortality, but it also demonstrates that treatment of obesity can significantly reduce risk factors associated with obesity related diseases. The National Heart, Lung, and Blood Institute suggests that a 10% reduction in body weight can help individuals improve their health by reducing cholesterol levels, triglycerides, and blood glucose levels (1998). However, even smaller reductions in weight can have a beneficial impact on health. Although magnitude of weight loss is positively correlated with health improvements, modest losses of 510% have been shown to significantly improve risk factors for cardiovascular disease (Wang et al., 2011; Douketis & Sharma 2005). Even los ses as small as 3 5% have been shown to decrease levels of low density lipoprotein


13 cholesterol, total cholesterol, triglycerides blood glucose, and insulin (Dattilo & Kris Etherton, 1992; Flechtner Mors, Ditschuneit, Johnson, Suchard, & Adler, 2000). Due to the significant benefits of weight loss, public and behavioral health research has focused on developing effective lifestyle and behavioral treatments for obesity. Based on the potential health risks and cost impact of pharmacotherapy and weight loss su rgery, this type of treatment may be the most effective way of achieving weight loss for the majority of obese individuals (McTigue et al., 2003). In terms of overall structure, a characteristic lifestyle intervention will involve weekly meetings with a gr oup of 1015 individuals, as group sessions have been shown to result in greater reductions in weight for all participants when compared to individual meetings with a trained interventionist (Renjilian, Perri, Nezu, McKelvey, Shermer, & Anton, 2001). Groups are led by professionals with training in nutrition, exercise, or psychology related fields. Sessions take place for 6090 minutes and involve a mix of health education, skills training, and the use of cognitivebehavioral strategies (Butryn, Webb, & Wadden, 2012). Initial treatment lasts four to six months, and more often, programs are offering maintenance sessions after the intervention that involve further group meetings on a biweekly or monthly basis. Behavioral or lifestyle intervention programs hav e several distinctive features: a process oriented approach designed to help individuals identify and adopt healthy behaviors; a focus on small, incremental steps toward weight loss; the use of goal setting to help participants change eating and exercise behaviors; and training in problem solving skills in order to help participants learn to overcome weight loss obstacles they may face outside of the group process (Foster, Makris, & Bailer, 2005).


14 In order to help participants make changes related to food i ntake and physical activity, treatment methods are based on the premise of shaping behaviors by setting small goals and reinforcing actions that move participants closer to the target behavior. The concept of shaping as applied to behavioral treatment was first propos ed by students of B.F. Skinner and is based on the principles of operant conditioning, in which actions are either reinforced or punished to bring about eventual behavior change (Ferster, Nurenberger, & Levitt, 1962). Other cognitive behavioral components incorporated in treatment include stimulus control, teaching individuals to control cues related to unhealthy eating behaviors; cognitive restructuring, helping participants to identify and reframe irrational or negative thoughts that may be un dermining weight loss efforts; and self monitoring, the process of maintaining daily records of food, physical activity, and weight change throughout the course of treatment (Wadden & Butryn, 2003). The use of these cognitive and behavioral strategies has been shown to be most effective when paired with development of a low calorie eating plan (e.g., a diet of 1,200 to 1,800 kcal per day) and the addition of regular physical activity (e.g., moderate intensity physical activity f or 150 minutes or more per week; Van Dorsten & Lindley, 2011). Typical Response to Treatment When analyzing the results of lifestyle interventions, the majority of studies present the mean response to treatment over time. The most typical response among participants is a loss of appr oximately 810% of weight over the course of a four to six month treatment period. This is similar to the amount that has been deemed clinically significant in helping participants attain the previously mentioned health benefits (Sarwer, von Sydow Green, V etter, & Wadden, 2013). Several recent studies indicate that severely obese (BMI 40 kg/m2) participants may demonstrate more noticeable


15 decreases in weight than overweight or obese individuals (Unick et al., 2011; Goodpaster et al., 2010), but the majority of studies continue to cite results that range from 5 10% weight los t (Yanovski & Yanovski, 2002). Despite these stated initial losses, long term weight reductions are often more difficult to maintain. Many studies confirm that par ticipants gain back approximately onethird of lost weight within one year following intervention, with slightly slower rates of regain ensuing over the next three to five years (Anderson, Konz, Frederich, & Wood, 2001). Over half of participants will have returned to their starting weight by the fiveyear mark (Wadden, Butryn, & Byrne 2004). Therefore, an increased number of studies are focusing on ways to improve long term weight maintenance. Current research highlights approaches such as increasing the i ntensity or extending the length of the preliminary intervention, improving long term motivation for weight control, and using both the primary intervention and a long term extended care period to teach maintenancespecific skills, above and beyond what is taught in a usual lifestyle program (Jeffrey et al., 2000). Such research represents an improvement upon earlier versions of treatment programs, as participants who are enrolled in extended care periods following treatment generally show improved weight m aintenance (Perri et al., 2008; Svetkey et al., 2008). However, even these additions to behavioral treatment have yet to provide the definitive formula for maintained weight loss over a long period of time. Mean weight loss and regain statistics demonstra te the pattern of weight change for a typical participant both during and after intervention. S tandard outcome studies for behavioral weight loss trials report these statistics and their associated level s of significance but showing only these responses t o treatment can mask the results of


16 individuals who may be more or less successful than the mean. One of the few research studies that does present categorical responses to treatment suggests that of participants enrolled in intense lifestyle modification interventions, who attended an average of 37 hours of treatment sessions during a yearlong period, 28% lost 10% or more weight, 26% lost 59.9% of their initial weight, and 38% lost 0.1 4.9% of initial weight (Christian, Tsai, & Bessesen, 2010). Recent r esearch has also seen a small trend toward examining the behaviors of those who are very successful at weight loss in order to discover patterns that may ultimately be most predictive of success. One study looked at participants enrolled in the National We ight Control Registry, who are required to have lost greater than or equal to 13.6 kg (30 lbs) and maintain ed this loss for over a year. Within the registry, r esearchers found a sample of 629 women and 155 men in the United States who had lost an average o f 30 kg and maintained at least a 13.6 kg weight loss for five years (Klem, Wing, McGuire, Seagle, Hill, 1997). Some, but not all individuals in the sample had participated in a structured behavioral treatment program. O ther studies suggest that of those who do take part in lifestyle interventions, about 20% are able to maintain 5% or greater weight reductions after five years (Wing & Hill, 2001). Examining only the mean responses to behavioral interventions therefore may make it difficult to judge components of treatment that are associated with the greatest rates of success. Future research that focuses on categorical responses to intervention will help to elucidate these key factors and help researchers understand differing responses to treatment. Self Monitoring and the Model of Self Regulation Self monitoring, or the recording of ones activities or behaviors, is rooted in the theory of self regulation. The process of self monitoring was originally proposed as a


17 tool for change by behaviorists such as Kanfer and Karoly, who stated that individuals could use their ability for self reflection to evaluate problematic healthrelated behaviors. They then are able to exercise control over their well being by monitoring (and altering) these behaviors ( Kanfer & Karoly, 1971). Scientists have linked this to the concept of weight reduction, as the process of losing weight is fundamentally an issue of consuming fewer calories t han one is able to expend. Therefore, changing behaviors related to food consumpt ion will help to alter this energy balance. Self monitoring serves as a trigger for executing new eating or exercise behaviors or for preventing the continuation of related unhealthy behaviors (Kanfer 1970). The act of monitoring is viewed as the first st ep in behavior change, by helping the individual to become aware of current actions and observing challenging patterns. The process allows individuals the opportunity to both establish goals for behavior change and track progress in achieving these goals ( Febbraro & Clum, 1998). Self monitoring is seen as the initial step in the three part feedback loop proposed by Kanfer. It requires both the careful observation and recording of the target behaviors (e.g., eating). This process naturally leads the indivi dual to the second step of self evaluation, during which they compare the results of their monitoring to what behaviors or actions are necessary in order to reach the desired outcome. This allows a person to assess whether what they are currently doing is adequate for achievement of their goals. Finally, during the self reinforcement stage, the individual is able to strengthen their current behavior, or alter behavior to more appropriately align with their goals based on the results of their evaluation (Mac e & Kratochwill 1985).


18 This model of self regulation is seen as a central part of the goal setting process used so frequently in behavioral weight loss theory. For participants, seeing what they have accomplished across a period of time is essential in providing them with the opportunity for reinforcement. Ultimately it is the measurement of the behavior via self monitoring that initializes this feedback loop, making the process of self monitoring fundamental for change. In the context of behavioral lifes tyle interventions for obesity, self monitoring can be done through the tracking of food and drink intake, physical activity, or weight change. It is typically seen as one of the foundations of behavioral treatment (Wadden, Crerand, & Brock, 2005). Further more, it has been shown to be correlated with long term success in weight loss (Head & Brookhart, 1997; Wing, Jeffery, Hellerstedt, & Burton, 1996). Individuals who are more consistent in this behavior have been shown to have higher rates of weight loss, a nd even individuals who struggle to self monitor consistently show greater rates of weight loss during periods of frequent monitoring when compared to periods of sporadic monitoring (Baker & Kirschenbaum, 1993). Boutelle and Kirschenbaum have suggested that self monitoring for at least 75% of days may be the optimal amount for enhanced weight loss success (1998). The Utilization of Self Monitoring in Behavioral Weight Loss Treatment The definitive goal of behavioral weight loss treatment is to reduce body w eight by a clinically significant amount, in order to improve health and reduce the risk of chronic disease. In order to do this, most programs aim for a weight loss of 0.4 kg (0.51 lb s) per week for overweight or obese individuals, and slightly higher lo sses of 0.8 kg (1 2 lbs) per week for individuals with a BMI equal to or greater than 35 kg/m2, in accordance with recommendations from the National Heart, Lung, and Blood Institute


19 (1998). To attain these rates of weight loss participants must aim to bur n more calories th an they expend on a daily basis. This energy imbalance is achieved by reducing caloric intake by 300500 kcal per day for overweight or obese individuals or by 5001,000 kcal per day for those who are more severely obese. Generally, the r eduction in caloric intake is done primarily through setting a calorie goal in combination with the additional of moderate physical activity. In behavioral treatment programs, participants first complete a week of tracking food intake without altering thei r diet in any way, in order to obtain a general estimate of baseline caloric intake. Group leaders then use these baseline records to help participants set goals for reducing the number of calories consumed per day. Participants are asked not only to track the type and amount of food eaten, but the associated caloric content, as well as information regarding time and location of eating, and emotions accompanying the eating (Wadden & Foster, 2000). Many programs will also request that participants track exer cise in order to make progress in increasing physical activity levels and calories burned. Finally, self monitoring may also take the form of periodic self weighing to help participants track weight over time. All three approaches have been shown to be ass ociated with weight loss, although self monitoring of caloric intake remains the most studied and perhaps most effective way of helping individuals to lose weight (Burke, Wang, & Sevick, 2011). When individuals self monitor on their own, without the help of a trained professional, the literature suggests that they tend to underestimate their caloric intake (Black & Cole, 2001). Most participants will have difficulties estimating both the amount of food eaten as well as caloric value of each food (Lansky & Brownell 1982). This


20 tends to be especially true for those who are more obese (Lichtman et al., 1992), such that under reporting of caloric intake is positively correlated with the BMI of the individual who is recording (Braam, Ock, BuenodeMesquita, & Seidell, 1998; Johansson, Wikman, Ahrn, Hallmans, & Johansson, 2001). Suggested explanations for underreporting include social desirability (Hebert, Clemow, Pbert, Ockene, & Ockene, 1995; Scagliusi 2003), a reluctance to report foods seen as unhealthy (Cook, Pryer, & Shetty 2000), dissatisfaction with body image (Novotny et al., 2003), and the inconvenience of detailed tracking (Macdiarmid & Blundell, 1998). However, despite the trend of underreporting, food records still serve as a method of helping part icipants monitor eating over the course of a day They also provide group leaders with an estimate of participant progress and remain a crucial measurement of whether a participant is adhering to their recommended calorie goal. Therefore, even when partici pants do underreport food records serve as a basis for the feedback loop that helps them to evaluate and adjust behavi or throughout the intervention. Despite the limitations of self monitoring participants who record what they eat are more likely to see clinically significant reductions in weight and thus, it remains a foundational element of lifestyle intervention programs. Specific Aims and Measures Specific Aim 1 Characterize participants in a behavioral treatment program by their weight loss both during the intervention and the extendedcare period. We hypothesized that participants would fall into clear groups based on their patterns of percent weight change over time, with some participants demonstrating results similar to the mean response (i.e., clinically significant weight loss followed by a regain during extended


21 care of onethird to one half the amount initially lost), and other participants showing results that are noticeably different from what is typically reported. S pecific Aim 2 Determine if the aforementioned groups differ in self monitoring frequency during both the intervention and follow up periods. Based on the previously mentioned research that suggests self monitoring is a key component of weight loss, we expected that participants in groups that demonstrated more significant success at weight loss would also have completed more records of food and drink intake during both phases of the study. For the purposes of this analysis, frequency of self monitoring was defined as the number of food records returned by participants that were either partially or fully completed. Specific Aim 3 Determine if self monitoring frequency mediated the relationship between group attendance and weight change during both phases of the study. This was examined as a follow up to the first two aims If, during either phase, the groups of participants were found to differ in self monitoring frequency, we believed that self monitoring would partially explain the relationship between trea tment attendance and weight change from 06 months (following the intervention) and 618 months (following the extended care phase). Treatment attendance during intervention was defined as participants who had arrived in person for the majority of the grou p session. For participants who were enrolled in the phone based extended care condition attendance was defined as completing a session check in on the phone with their group leader.


22 Exploratory Aim Determine if any potential mediational relationships rem ained significant with the inclusion of covariates. Previous research has found mixed results describing the effect of age, educational level, and race/ethnicity of lifestyle intervention participants. Some research suggests that older adults are more succ essful than younger adults and that patterns of weight loss vary among individuals of different races or ethnicities (Kumanyika, Obarzanek, Stevens, Hebert, & Whelton, 1991; Wadden et al., 2012). Other studies have shown that minorities initially have less success during lifestyle interventions, but that these differences disappear in the long term (Stevens et al., 2001). Finally, some studies have shown that educational attainment is related major weight gain or loss later in life (Kahn & Williamson, 1990) and that higher levels of education are associated with greater reductions in BMI (Halkjr, Holst, & Srensen, 2012). Therefore, we investigated the effect of controlling for these variables in any mediational relationships found to be significant in the previous aim. We hypothesized that the mediation would remain significant when age, education, and race/ethnicity variables were held constant.


23 CHAPTER 2 MATERIALS AND METHODS Treatment of Obesity in Underserved Rural Settings (TOURS) Study The TOURS study was a randomized controlled trial funded by the National Institutes of Health and conducted in six rural counties in northern Florida from 2003 through 2007 (Perri et al., 2008). The study investigated differing impacts of two types of extended care treatments that were designed to sustain weight loss following a behavioral intervention. The primary outcome was change in body weight during the extended care period, while secondary outcomes included changes in systolic and diastolic blood pressure, lipid profile, glycemic control, and physical fitness during the same time period. The study was conducted in two phases. During the initial intervention phase, or Phase 1, participants attended 24 weekly meetings in groups of 1014 women at local County Ext ension Offices. Group leaders were Family and Consumer Science (FCS) Agents, or trained leaders with bachelors or masters degrees in relevant fields The Phase 1 intervention program was based on the Diabetes Prevention Program (The Diabetes Prevention P rogram, 1999) but was modified slightly to address the special issues of women living in rural communities. Changes included education on ways to adapt Southern cooking recipes in order to reduce calories and grams of fat, as well as ways to exercise in rural areas where sidewalks or walking paths may not be available. Overall, the program focused on helping participants follow low calorie eating plans of 1 200 kcal/day and increase physical activity at home to 30 minutes per day of moderate intensity walki ng in order to create a caloric deficit that would result in approximately .04 kg per week of weight loss. Participants also received training in


24 cognitive behavioral strategies such as self monitoring and goal setting to help support diet and physical act ivity goals. Participants were asked to self monitor by recording food and drink intake on a daily basis using standardized forms designed to track eating (See Appendix). Participants also were provided with cooking demonstrations to increase knowledge of low fat, low calorie food preparation; stress management strategies; and techniques to help participants maintain a healthy diet when eating outside the home. During the extended care phase, Phase 2, participants were randomly assigned to one of three conditions: an officebased maintenance program, a telephonebased maintenance program, or an educational control group. Prior to the start of this follow up phase, every participant received written handouts on problem solving strategies. They were also provided with a 12month supply of eating and exercise logs in order to continue monitoring eating and activity levels. P articipants were encouraged to continue to self monitor three days per week, including at least one weekend day. After the start of the fo llow up sessions, participants in every group received contact two times per month. The officebased condition continued to meet in their initial weight loss groups, whereas participants in the telephonebased condition received individual telephone sessio ns and participants in the educational control group received newsletters in the mail. In total, the extended care phase lasted for 12 months. Participants The current study included 167 women, aged 5075, with a BMI 30 kg/m2, yet weighing less than 159 kg (350 lbs) in order to allow weight measurement on a standard balance beam scale. All women were required to live in one of six rural counties in northern Florida. Exclusion criteria included underlying diseases likely to limit lifespan,


25 such as cancer requiring treatment within the last five years, serious infectious disease, or cardiovascular events occurring within the last six months; abnormal lab values, including a fas ting blood glucose > 125 mg/dl and fasting serum triglycerides > 400 mg/dl or a r esting blood pressure > 140/90 mmHG despite appropriate drug treatment; and the use of certain medications, including antipsychotic agents, monoamine oxidase inhibitors, systemic corticosteroids, antibiotics for HIV or TB, chemotherapeutic drugs, or prescr iption weight loss drugs. In addition, individuals were excluded if they were unable or unwilling to provide informed consent, were unable to read English at a 5th grade level, were unwilling to accept random assignment, were unable to travel to an extensi on office for group meetings, or were currently participating in another randomized research project. Finally, individuals who reported a weight loss of greater than 10 pounds in the previous six months, excessive alcohol intake, or a diagnosis of a major psychiatric disorder were also excluded. A total of 559 women responded to study recruitment announcements, and 261 of these women were excluded prior to the start of the study. Of the 261 women who were found ineligible for participation, 82 were excluded for elevated blood pressure, 76 had abnormal lab values, 26 had contraindications based on medical history, 27 had other abnormal screening visit results, 17 had a BMI out of the required range, and 30 declined to participate. A total of 234 women compl eted Phase 1 of the study; however, participants randomized to the educational control group displayed significantly different weight change patterns during the extended care phase. These participants were therefore removed from the current study and a tot al of 167 participants were included in the final analysis.


26 Procedures Potential participants were contacted via methods of direct solicitation, including direct mailings and presentations at community events. Those who expressed interest in the program w ere encouraged to complete a brief telephone screening, and women who meet initial eligibility requirements were scheduled for an inperson screening appointment. At this appointment, the study was described in detail, informed consent was obtained, and de mographic information, a comprehensive medical history, and a current medication inventory were collected. Participants who continued to meet study requirements were scheduled for a second appointment, which included: questionnaires assessing dietary intak e, physical activity levels, health related quality of life, depressive symptoms and problem solving abilities ; height, weight, and waist measurements; resting heart rate and blood pressure; a fasting blood draw; a six minute walk test; and an electrocardiogram. Participants who were judged as meeting eligibility criteria after the second screening visits were randomized to a study group and extended care condition. Measures Body weight Body weight was measured at baseline, 6 months (following the initial intervention), and 18 months (following the extended care period). Participants were measured in light indoor c lothing without shoes and with pockets emptied by a study nurse who was masked to the treatment condition of the participants. Measures were taken using a calibrated and certified balance beam scale and were measured to the nearest tenth of a kilogram. Per cent change in weight over time was then calculated for all participants based on initial weight, 6 month weight, and 18 month weight.


27 Self monitoring Participants were given standardized food intake logs on which they were able to record all food and dri nk consumption on a daily basis Records were used to measure total caloric intake, to help participants set calorie goals and to evaluate their progress in meeting these goals. Each food record provided spaces for participants to log types of food eaten and the associated amount of food and calories consumed, as well as the time of day the eating occurred, activities completed while eating (e.g. driving, watching television), feelings while eating (e.g. tired, bored), and location where the eating occurre d (e.g. kitchen, car ). Participants also had an optional space to record grams of fat in each food consumed if they desired. For caloric and fat intake, participants had space to calculate subtotals for each meal (breakfast, lunch, dinner, planned snacks, and unplanned eating), as well as a grand total for the day. At the end of each week, participants were provided space to calculate an average caloric intake across the sevenday period. During the intervention, participants were asked to self monitor on a daily basis. The majority of participants were asked to set caloric goals of 1, 200 kcal per day. Records were returned to group leaders and reviewed at weekly group meetings. At the beginning of the extended care phase, all participants were provided wi th a years supply of food records. Participants were asked to monitor a minimum of two weekdays and one weekend day every week, although some chose to monitor more often. Caloric goals were adjusted for the extended care phase based on intervention weight change and long term weight maintenance plans. F ood records were collected in person by group leaders for those attending officebased follow up meetings. For those receiving


28 telephonebased contact, stamped and addressed envelopes were provided so food r ecords could be returned to the group leaders as necessary. Statistical Analysis The statistical software package SPSS 21.0 for Windows (SPSS Inc., IL) was used to conduct the statistical analysis for this research study. Specific Aim 1 A two step cluste r analysis was conducted in order to classify participants into separate groups based on weight change patterns across the course of the study. Cluster analysis is a statistical procedure which categorizes participants into groups while ensuring that withi n cluster differences are minimized and betweencluster differences are maximized. It is a technique that has been advocated as particularly useful to health psychologists in order to identify groups of individuals who may be most in need of intervention, resources, or specific services (Clatworty, Buick, Hankins, Weinman, & Horne, 2005). A twostep cluster analysis was chosen as most relevant for this data, as it is an exploratory technique that does not require the number of clusters to be predetermined; however, for the purposes of this study, the number of clusters was limited to three in order to ensure that clusters were of a meaningful size for further analyses. Clusters were analyzed based on individual participant weight change from baseline to 6 months and from 6 months to 18 months. The clustering solution was determined using log likelihood estimation and the squared Euclidean distance method was used to assess similarity between clusters in order to take into account elevation of weight change scores (Blashfield, 1980). Stability of clusters was assessed by random division of the sample into two halves, followed by repeated cluster analysis and visual inspection of the data (Clatworty, Buick, Hankins, Weinman, & Horne, 2005).


29 Specific Aim 2 We examined the hypothesis that clusters would differ based on self monitoring frequency across the intervention and extended care periods. Therefore, following the cluster analysis, the resulting groups were compared based on self monitoring frequency (i.e., how many food records were completed) during Phase 1 and Phase 2 using a oneway independent Analysis of Variance (ANOVA). Normality was assessed using tests of skewness kurtosis and visual inspection of Q Q plots. Posthoc tests were used to determine, w hen applicable, where specific differences between clusters could be found. Specific Aim 3 As a preliminary analysis for the third aim, correlational tests were used in order to test for significant relationships between group attendance, self monitoring frequency, and percent weight change. These correlational measures were completed separately for the variables from Phase 1 and Phase 2 of the study. Where significant correlations were found, the Preacher and Hayes model was used to determine if self mon itoring frequency mediated the relationship between treatment attendance and percent weight change. The number of treatment sessions attended was entered as the independent variable and percent weight change for the designated phase of the study was entered as the dependent variable. Number of self monitoring records was entered as the potential mediating variable. The Preacher and Hayes model for mediation was chosen in order to reduce the possibility of Type I error and increase statistical power. Results were bootstrapped to 5,000 samples, as bootstrapping is an effective technique for smaller sample sizes and avoids making parametric assumptions about the distribution


30 indirect effect to maximum possible indirect effect. Exploratory Aim Based on the results of the previous aim covariates were added to the Preacher and Hayes bootstrapped model in order to determine if any mediational relationships still held. The covar iates included in the analysis included: age at baseline, highest level of education obtained, and race/ethnicity variables.


31 CHAPTER 3 RESULTS Baseline Characteristics The study sample used for all analyses consisted of 167 obese older women from rural counties in Northern Florida, with a mean age of 59.9 years and a mean BMI of 37.0 kg/m2. Overall, the sample was predominantly White, nonHispanic (77.2%), married (74.3%), with a trade, vocational, or associates degree (44.9%). Participants were primari ly employed, either full time or part time (47.3%). Further baseline characteristics of the sample are summarized in Table 31. Specific Aim 1 The cluster analysis of participants percent weight change during intervention and follow up revealed three dist inct groups of participants, based on their patterns of weight loss and potential regain over time. The largest cluster consisted of 82 participants, or 49.10% of the sample. These participants demonstrated a mean weight change of 11.09% during Phase 1, but regained +4.78% mean weight during Phase 2. This cluster demonstrated results most typical of behavioral weight loss treatment, and was labelled the Moderate Success group. The next largest cluster evidenced low rates of achievement, and was labelled the Low Success group. This cluster contained 50 participants, or 29.94% of the sample. These individuals showed a mean weight change of 2.58% during Phase 1, and a regain of +1.50% mean weight during Phase 2. Finally, the remaining women fell within a thir d cluster of 35 participants, or 20.96% of the sample, and were labelled the High Success group based a pattern of continued weight loss over time. These participants initially showed a mean weight change of 13.99% during Phase 1, and an additional 7.44% mean weight change during Phase 2. Results


32 are presented visually in Figure 31 and descriptive statistics for the three clusters are shown in table 32. Visual inspection of the scatter plots for repeated analyses using split half data demonstrated stabi lity of clusters. Specific Aim 2 During Phase 1, the initial intervention, the three clusters of Low, Moderate, and High Success participants did not differ significantly in terms of number of food records completed, F (2, 164) = 1.93, p = .149. However, an ANOVA revealed significant differences between clusters for number of records completed during Phase 2, the extended care period, F (2, 164) = 22.91, p < .001. As Levenes test was significant for the Phase 2 data, Games Howell post hoc testing was used to correct for the potential differences of variance. The post hoc analyses revealed significant differences between the Low and Moderate Success groups ( p = .005), the Low and High Success groups ( p < .001), and the Moderate and High Success groups ( p < 001). The differences were such that, during Phase 2, the Moderate Success group completed a mean 52.43 additional records when compared to the Low Success group, and in turn, the High Success group completed a mean 91.59 more records when compared to the Moderate Success group during the year long extended care phase. The High Success group also completed a mean of 143.93 additional records when compared to the Low Success group. Mean number of food records submitted by group and associated significant dif ferences are presented in Table 33. Specific Aim 3 The final aim of this study was to examine th e relationship between Phase 2 treatment attendance, percent weight change, and number of self monitoring records completed. Phase 2 self monitoring frequency was seen as a potential mediator as it


33 correlated significantly with both Phase 2 attendance ( p < .001) and Phase 2 percent weight change ( p = .03). To estimate any direct and indirect effects of self monitoring, a Preacher and Hayes simple mediational analysis was conducted and 5,000 bootstrapped estimations were used in order to create a confidence interval for the indirect effect (Preacher & Hayes 2008). Results of the mediational analysis showed that the overall effect of treatment attendance on percent weight change (total effect = .0002, p < .001) became nonsignificant when self monitoring frequency was added to the model as a mediator (direct effect of self monitoring = .0005, p = .403). The analysis also indicated that the indirect effect of self monitoring during Phase 2 was significant, as the 95% confidence interval did not include zero (95% CI: .0026, .0008). Thus, the number of food records completed during Phase 2 fully mediated the relationship between session attendance and percent weight change. Participants who completed more records also showed greater weight loss than participants who completed fewer possible indirect effect) was .20, which suggests a moderate effect size. Exploratory Aim As an additional exploratory analysis, the mediational macro was run again with the inclusion of age, education, and race/ethnicity as covariates. When these three variables were held constant, results remained similar. The total effect of group attendance on percent weight change (total effect = .001, p = .040) again became nonsignificant with the inclusion of self monitoring as a mediator (direct effect of self monitoring = .0005, p = .440). The 95% confidence interval for the indirect effect also indicated mediation as the range still did not include zero (95% CI: .0026, .006). Therefore, results of the analysis suggest that self monitoring during the follow up


34 period of behavioral intervent ions plays an important role in weight maintenance or further weight loss and completely mediates the relationship between session attendance and weight change.


35 Table 3 1. Participant baseline characteristics Characteristics Age in years, M ( SD ) 59.9 (6.2) BMI in kg/m 2 M ( SD ) 37.0 (5.1) Ethnicity, n (%) Hispanic/Latino 2 (1.2) Other 165 (98.8) Race, n (%) White 131 (78.4) Black 34 (20.4) American Indian or Pacific Islander 2 (1.2) Highest Level of Education, n (%) Less than 12 years 62 (37.1) Trade, vocational, or associates degree 75 (44.9) Bachelors degree 18 (10.8) Advanced degree 29 (16.0) Marital Status, n (%) Never Married 1 (0.6) Divorced or separated 23 (13.8) Widowed 17 (10.2) Presently Married 124 (74.3) Living in marriage like relationship 2 (1.2) Employment Category, n (%) Not working 13 (7.8) Retired 42 (25.1) Homemaker 8 (4.8) Employed, full time or part time 79 (47.3) Disabled 2 (1.2) Other 3 (1.8) More than one category 20 (12.0) Total Annual Family Income, n (%) Less than $10,000 9 (5.4) $10,000 19,000 32 (19.2) $20,000 34,999 41 (24.6) $35,000 49,999 34 (20.4) $50,000 74,000 32 (19.2) $75,000 99,999 4 (2.4) $100,000 149,999 2 (1.2) Dont know 2 (1.2)


36 Table 3 2. Cluster d ata Participant Clusters Low Success Participants (n=50) MSD Age in years 59.45 6.54 Percent weight change, Phase 1 2.58 2.39 Percent weight change, Phase 2 1.50 3.42 Percent weight change, Total 1.07 4.57 Moderate Success Participants (n = 82) MSD Age in years 59.73 6.12 Percent weight change, Phase 1 11.09 3.92 Percent weight change, Phase 2 4.78 3.53 Percent weight change, Total 6.31 4.86 High Success Participants (n=35) MSD Age in years 60.87 6.03 Percent weight change, Phase 1 13.99 4.31 Percent weight change, Phase 2 7.44 3.90 Percent weight change, Total 21.43 5.19 Table 3 3. Number of completed food records submitted by clusters Study Phase Phase 1 MSD Low Success Participants 117.4 53.5 Moderate Success Participants 125.5 46.7 High Success Participants 123.2 45.6 Phase 2 MSD Low Success Participants 59.4 84.8 Moderate Success Participants 111.8 83.2 a High Success Participants 203.3 115.2 b a p = .005 versus the low group, b p < .001 versus the low and moderate groups


37 Figure 31. Results of cluster analysis by participant weight change


38 CHAPTER 4 DISCUSSION The purpose of the current study was both to identify different groups of participants based on their patterns of weight loss and to assess for differences between groups in terms of frequency of self monitoring. In addition, the mediational effect of self monitoring on the relationship between behavioral weight loss program attendance and weight change was examined The overall role of self monitoring was explored both during an initial lifestyle intervention and a long term extendedcare phase that focused on weight maintenance strategies. Results of statistical analyses dem onstrated clear patterns of weight loss that differed among intervention participants, such that individuals fell into three separate groups based on rates of success throughout both phases of the study. The largest group, classified as the Moderate Succes s group and accounting for about half of all participants, demonstrated results most typical of the mean an initial clinically significant loss of weight during the intervention, followed by a regain of almost half that weight during the year after inter vention. Two smaller groups, however, demonstrated very different patterns. A Low Success group lost less than three percent weight initially, and regained almost all of weight lost during the extended care phase. The smallest group, labeled the High Success cluster, demonstrated resul ts that may be closest to ideal for behavioral programs. These participants initially lost more than onefifth of their body weight. They also continued to lose clinically significant amounts during the extendedcare period, w hich is a substantial difference from results typically seen in lifestyle intervention programs.


39 When comparing these three groups based on self monitoring frequency during intervention and follow up, none of the groups demonstrated significant differences regarding the number of food records completed during the initial intervention period. However, each group differed significantly from the other groups during the extended care phase. Results showed that the most successful participants completed the highest number of self monitoring records, followed by the moderately successful participants (those demonstrating results closest to the mean). The group that showed the lowest rates of success also completed the smallest number of self monitoring records. In further inspecting the role self monitoring played during the extended care phase, the indirect effect of self monitoring frequency on the relationship between group attendance and weight loss was examined. A significant direct effect between attendance and weight change was found during follow up. After running a mediational analysis, results showed that self monitoring fully mediated the relationship between attendance and weight change. These results held constant with the inclusion of age, education, and race/ethnicity as covariates. Generally, these results suggest that individuals can be charact erized by patterns of weight loss and that approximately half of all participants in behavioral interventions do not follow the pattern of weight loss typically described in the literature. This study highlights an approach that researchers have not yet taken in observing response to treatment by categorizing weight loss participants using an empirically driven method, as opposed to only the use of descriptive statistics. The unique responses demonstrated by the three separate clusters indicates that there is room for future studies that more fully investigate the behaviors of those in the most and least


40 successful groups, in order to determine what components of treatment are most helpful in achieving weight loss. Perhaps by comparing behavioral patterns and program adherence between these groups, researchers will be able to find how best to encourage the low achievers. Furthermore, by establishing a recognized categorical response to behavioral treatments stepped care interventions can be designed to target those who may be at risk for low rates of weight loss or eventual higher rates of weight regain. These types of treatment methods are designed so that minimally intrusive interventions are the first to be implemented. Participants who are unable to meet weight loss goals are then provided with more intensive interventions. T his practice helps study staff preserve resources for those who may be in the most need while avoiding excessive treatment for those participants who may already be successful (Davis, 2000). Recent investigations on a stepped care approach suggest mixed results when compared to standard behavioral interventions, with some studies demonstrating equal amounts of weight loss and others demonstrating lesser rates of weight loss in the stepped care interventions When stepped care methods are effective, they also have been shown to significantly reduce the costs associated with implementing lifestyle interventions ( Carels et al., 2012; Jakicic et al., 2012). Results indicate that during the intervention phase, self monitoring may not be the only component of treatment that predicts success. During the initial program, none of the three groups differed significantly in self monitoring frequency. Rather, all groups showed fairly high levels of self monitoring during treatment despite their differing rates of weight loss, and only began to demonstrate significantly different rates of record keeping during extended care. These findings suggest that additional factors may be


41 contributing to the differing rates of success, such as comprehension of treatment materials, adherence to other aspects of treatment, or amount of support received from friends and family. Participants may also have varied physical responses to caloric deprivation due to underlying physiological mechanisms. Future research should focus on identifying limiting factors to weight loss early on during interventions in order address these issues in lower success participants. Depending on findings, it may be important to focus on increasing education or skills training for lower success participants, or screening participants for physiological traits that may thwart weight loss so that treatment can be altered appropriately. Alternatively, it will be i mportant to identify characteristics of the most successful participants that can be emphasized as important to treatment to help improve the results of both the moderate and the low success participants. Finally, results of this study suggest that self m onitoring remains important during extended care, and in fact fully explains the relationship between group attendance and weight loss. Thus, it seems crucial that participants be encouraged to continued tracking food and drink intake. Even minor amounts o f weight loss have been shown to be beneficial to individuals, and therefore it is important to both prevent weight regain and retain any amount of weight lost, no matter how small. By identifying individual obstacles to weight loss early on, we can perhaps prevent participants from becoming discouraged and stopping the effective behavioral change methods they have learned during treatment in order to maintain their weight or even continue to lose small amounts over time.


42 It is important to address some of the limitations of the current study. Primarily, the current analyses are correlational in nature. Therefore, the differences in self monitoring may be based on other aspects of treatment or individual differences that have yet to be explored. Additionall y, it is essential to point out that, due to the nature of penand paper record keeping, there is no way to determine intervals between eating and self monitoring. Research indicates that shorter intervals between eating and recording are more predictive of ultimate weight loss (Burke, Sereika, Music, Warziski, Styn, & Stone, 2008). This falls in line with the concept of the feedback loop mentioned earlier. Participants who hastily fill out a weeks worth of records before their weekly meeting, or even reco rd a full day of eating before bed, are denied the opportunity for immediate self evaluation that leads to the reinforcement or altering of observed behaviors. The effectiveness of self monitoring lies in the opportunity it provides individuals to consiste ntly remain an active participant in the feedback loop. Because paper records were utilized for the purposes of this study, it is impossible to determine if successful individuals were truly recording food and drink intake at the time of consumption. Final ly, the participants in this study were all older women living in rural southern communities, and therefore, results cannot be generalized to a broader population without further research. Given these limitations, it will be interesting to follow up on the results of the current study with the advent of new forms of self monitoring that take advantage of easy to use electronic devices. With the popularity of computer and cell phone applications that allow the efficient tracking of food intake, self monitori ng may become more appealing to a larger range of people. Recent studies have focused on


43 computerized software and online applications to help participants self monitor and have shown encouraging results (Yon, Johnson, Harvey Berino, Cold, & Howard, 2007; Glanz, Murphy, Moylan, Evensen, & Curb, 2006; Krukowski, Harvey Berino, Bursac, Ashikaga, & West, 2013). New forms of technol ogy are also being used to monitor other self regulated behavior, including physical activity and step goals (Kirwan, Duncan, Vand elanotte, & Kerry, 2012) and the management of blood glucose in diabetics (Anoop, Hou, Golnik, Flaherty, & Vu, 2010). These types of software and mobile applications may make it easier for participants to self monitor and for researchers to understand when and how frequently participants are tracking their eating. Furthermore, different forms of self monitoring are being explored that may address both the problems of efficiency and the interval between the behavior and the recording. Daily self weighing may be an effective way to help individuals track the results of eating and exercise behaviors on a regular basis in order to understand what behaviors lead to weight loss and weight gain, and thus reinforce or alter these behaviors in line with the theory of self regulation. Studies on self weighing have shown that more frequent weighing is associated with a lower BMI (Butryn, Phelan, Hill, & Wing, 2012), as well as increased weight loss or decreased weight regain (Linde, Jeffery, French, Pronk, & Boyle, 2005; Van Wormer et al., 2009). Focusing future research efforts on forms of self monitoring that are more efficient or more accessible to participants may motivate participants to self monitor more frequently, and can help to specifically address the problem s of individuals who are discouraged from continued self monitoring after the end of an intervention.


44 The results of the current study helped to determine differences between weight loss program participants in terms of patterns of weight loss and regain. The resulting clusters highlight the differences between the typical response to treatment shown by approximately half of participants, and patterns of little weight loss or exceptional weight loss found among smaller but still significant portions of the sample. Additional results helped to further define the role self monitoring plays following the conclusion of initial behavioral treatment, and the importance of continuing to track behaviors to maintain or even lose additional weight. During an extended care phase, the amount of self monitoring records completed helped to explain differences between participants evidencing low, moderate, and high rates of success. Moreover, it fully mediated the relationship between extended care treatment attendance and weight change after the end of the treatment program, even with the inclusion of age, education, and race/ethnicity as covariates. Overall, results both emphasize the need for expanding on the methods used to conceptualize patterns of weight loss among par ticipants in lifestyle interventions, and also suggest that ongoing vigilance via self monitoring is likely to help enhance long term weight management.




46 LIST OF REFERENCES Allison, D. B., & Weber, M. T. (2003). Treatme nt and prevention of obesity: What works, what doesnt work, and what might work. Lipids 38(2), 147 155. doi:10.1007/s11745003 10455 Anderson, J. W., Konz, E. C., Fredrich, R. C., & Wood, C. L. (2001). Long term weight loss maintenance: a metaanalysis of US studies. American Journal of Clinical Nutrition 74 579 584. Black, A.E., Cole, T.J. (2001). Biased over or under reporting is characteristic of individuals whether over time or by different assessment methods. Journal of the American Dietetic As sociation 101(1) 70 80. Bray, G. A. (2003). Risks of obesity. Primary Care: Clinics in Office Practice 30(2), 281 299. doi:10.1016/S00954543(03)00008 3 Braam, L.A., Ock, M.C., Bueno de Mesquita, H.B., Seidell, J.C. (1998). Determinants of obesity related underreporting of energy intake. American Journal of Epidemiology, 147(11), 10811086. Burke, L. E., Sereika, S. M., Music, E., Warziski, M., Styn, M. A., & Stone, A. (2008). Using instrumented paper diaries to document self monitoring pat terns in weight loss. Contemporary Clinical Trials 29(2), 182 193. doi:10.1016/j.cct.2007.07.004 Butryn, M. L., Phelan, S., Hill, J. O., & Wing, R. R. (2008). Consistent self monitoring of weight: A key component of successful weight loss m aintenance. Ob esity 15(12), 3091 3096. Butryn, M. L., Webb, V., & Wadden, T. A. (2011). Behavioral Treatment of Obesity. Psychiatric Clinics of North America 34(4), 841 859. doi:10.1016/j.psc.2011.08.006 Carels, R. A., Young, K. M., Hinman, N., Gumble, A., Koball, A ., Oehlhof, M. W., & Darby, L. (2012). Steppedcare in obesity treatment: Matching treatment intensity to participant performance. Eating Behaviors, 13(2), 112 118. doi: 10.1016/j.eatbeh.2012.01.002 Christian, J. G., Tsai, A. G., & Bessesen, D. H. (2009). Interpreting weight losses from lifestyle modification trials: using categorical data. International Journal of Obesity 34(1), 207 209. doi:10.1038/ijo.2009.213 Clatworthy, J., Buick, D., Hankins, M., Weinman, J., & Horne, R. (2005). The use and reporti ng of cluster analysis in health psychology: A review. British Journal of Health Psychology, 10, 329358.


47 Cook, A., Pryer, J., & Shetty, P. (2000). The problem of accuracy in dietary surveys: Analysis of the over 65 UK National Diet and Nutrition Survey. Journal of Epidemiology and Community Health, 54(8), 611 616. doi:10.1136/jech.54.8.611 Dattilo, A. M., & Kris Etherton, P. M. (1992). Effects of weight reduction on blood lipids and lipoproteins: a metaanalysis. American Journal of Clinical Nutrition, 5 6 320 328. Davison, G. C. (2000). Stepped care: Doing more with less? Journal of Counseling and Clinical Psychology, 68 (4), 580 585. Douketis, J. D., & Sharma, A. M. (2005). Obesity and cardiovascular disease: Pathogenic mechanisms and potential benefit s of weight reduction. Seminars in Vascular Medicine, 5 (01), 25 33. doi:10.1055/s 2005871739 Ferster, C.B., Nurenberger, J.I, & Levitt, E.B. (1962). The control of eating. Journal of Mathetics 1 87 109. Flegal, K.M., Graubard, B.I., Williamson, D.F., & Gail, M.H. (2005). Excess deaths associated with underweight, overweight, and obesity. Journal of the American Medical Association, 293(15), 18611867. Flegal, K. M., Carroll, M. D., Kit, B. K., & Ogden, C. L. (2012). Prevalence of obesity and trends i n the distribution of Body Mass Index among US adults, 19992010. Journal of the American Medical Association, 307 (5), 491. doi:10.1001/jama.2012.39 Fletchner Mors, M., Dischuneit, H.H., Johnson, T.D., Suchard, M.A., & Adler, G. (2000). Metabolic and weig ht loss effects of long term dietary intervention in obese patients: Four year results. Obesity Research, 8 (5), 399 402. Foster, G. D., Makris, A. P., & Bailer, B. A. (2005). Behavioral treatment of obesity. American Journal of Clinical Nutrition, 82(supp l), 230S 235S. Glanz, K., Murphy, S., Moylan, J., Evensen, D., & Curb, J. D. (2006). Improving dietary self monitoring and adherence with handheld computers: A pilot study. American Journal of Health Promotion, 20(3), 165 170. doi:10.4278/08901171 20.3. 165 Goodpaster, B. H., DeLany, J. P., Otto, A. D., Kuller, L., Vockley, J., South Paul, J. E., et al.,(2010). Effects of diet and physical activity interventions on weight loss and cardiometabolic risk factors in severely obese adults. Journal of the American Medical Association, 304 (16), 1795. doi:10.1001/jama.2010.1505 Halkjr, J., Holst, C., & Srensen, T. I. A. (2003). Intelligence test score and educational level in relation to BMI changes and obesity. Obesity 11(10), 1238 1245. doi:10.1038/oby. 2003.170


48 Hebert, J.R., Clemow, L., Pbert, L., Ockene, I.S., & Ockene, J.K. (1995). Social desirability bias in dietary self report may compromise the validity of dietary intake measures. International Journal of Epidemiology, 24(2), 389 398. Jakicic, J. M., Tate, D. F., Lang, W., Davis, K. K., Polzien, K., Rickman, A. D., ...Finkelstein, E. A. (2012). Effect of a stepped care intervention approach on weight loss in adults: A randomized clinical trial. Journal of the American Medical Association, 307(24), 26172626. Johansson, G., Wikman, A., Ahrn A.M., Hallmans, G., & Johansson, I. (2001). Underreporting of energy intake in repeated 24hour recalls related to gender, age, weight status, day of interview, educational level, reported food intake, smoking h abits, and area of living. Public Health Nutrition, 4 (4), 919 927. Kanfer, F.H, & Karoly, P. (1972). Self control: A behavioristic excursion into the lions den. Behavior Therapy, 3(3), 398 416. Klem, M. L., Wing, R. R., McGuire, M. T., Seagle, H. M., & Hill, J. O. (1997). A descriptive study of individuals successful at long term maintenance of substantial weight loss. American Journal of Clinical Nutrition, 66 239 246. Krukowski, R. A., Harvey Berino, J., Bursac, Z., Ashikaga, T., & West, D. S. (2013). Patterns of success: Online self monitoring in a webbased behavioral weight control program. Health Psychology 32(2), 164 170. doi:10.1037/a0028135 Kumanyika, S. K., Obarzanek, E., Stevens, V. J., Hebert, P. R., & Whelton, P. K. (2005). Weight l oss experience of black and white participants in NHLBI sponsored clinical trials. American Journal of Clinical Nutrition, 53(suppl), 161S 8S. Lichtman, S.W., Pisarska, K., Berman, E.R., Pestone, M., Dowling, H., Offenbacher, E., Heymsfield, S.B. (1992). Discrepancy between self reported and actual caloric intake and exercise in obese subjects. New England Journal of Medicine, 327(27), 18931898. Linde, J. A., Jeffery, R. W., French, S. A., Pronk, N. P., & Boyle, R. G. (2005). Self weighing in weight gai n prevention and weight loss trials. Annals of Behavioral Medicine, 30(3), 210 216. doi:10.1207/s15324796abm3003_5 Macdiarmid, J., & Blundell, J. (1998). Assessing dietary intake: Who, what and why of under reporting. Nutrition Research Reviews, 11, 231 2 53. McTigue, K. M., Harris, R., Hemphill, B., Lux, L., Sutton, S., Lohr, K. N., & Bunton, A. J. (2003). Screening and interventions for obesity in adults: Summary of the evidence for the U.S. Preventive Services Task Force. Annals of Internal Medicine, 13 9 (11), 933 34. doi:10.7326/00034819139 11 20031202000013


49 Mokdad, A. H., Marks, J. S., Stroup, D. F., & Gerberding, J. L. (2004). Actual causes of death in the United States, 2000. Journal of the American Medical Association 291(10), 1238 1245. Nation al Heart, Lung, and Blood Institute. (1998). Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults. NIH Publication 1 262. Novotny, J.A., Rumpler, W.V., Riddick, H., Hebert, J.R., Rhodes, D., Judd, J.T., Briefel, R. (2003). Personality characteristics as predictors of underreporting of energy intake on 24hour dietary recall interviews. Journal of the American Dietetic Association, 103(9), 11461151. Ogden, C. L., Carroll, M. D., Kit, B. K., & Flegal, K M. (2013). Prevalence of obesity among adults: United States, 2011 2012. NCHS Data Brief No. 131 1 8. Perri, M.G., Limacher, M.C., Durning, P.E., Janicke, D.M., Lutes, L.D., Bobroff, L.B., ...Martin D.M. (2008). Extendedcare programs for weight management in rural communities. Archives of Internal Medicine, 168(21), 23472354. Rao, A., Hou, P., Golnik, T., Flaherty, J., & Vu, S. (2010). Evolution of data management tools for managing self monitori ng of blood glucose results: A survey of iPhone applications. Journal of Diabetes Science and Technology 4 (4), 949 957. doi:10.1177/193229681000400426 Renjilian, D. A., Perri, M. G., Nezu, A. M., McKelvey, W. F., Shermer, R. L., & Anton, S. D. (2001). In dividual versus group therapy for obesity: Effects of matching participants to their treatment preferences. Journal of Consulting and Clinical Psychology, 69 (4), 717 721. doi:10.1037//0022006X.69.4.717 Scagliusi, F.B., Polacow, V.O., Artioli, G.G., Benatti, F.B., & Lancha, A.H. (2003). Selective underreporting of energy intake in women: M agnitude, determinants, and effect of training. Journal of the American Dietetic Association, 103(10), 13061313. Sarwer, D. B., von Sydow Green, A., Vetter, M. L., & Wa dden, T. A. (2009). Behavior therapy for obesity: W here are we now? Current Opinion in Endocrinology, Diabetes and Obesity 16 (5), 347 352. doi:10.1097/MED.0b013e32832f5a79 Stevens, V. J., Obar zanek, E., Cook, N. R., Lee, I., Appel L. J., West, D. S., ...Cohen, J. (2001). Long term weight loss and changes in blood pressure: Results of the trials of hypertension prevention, phase II. Annals of Internal Medicine, 134 (1), 1 12. Svetkey, L.P, Stevens, V.J., Brantley, P.J., Appel, L.J, Hollis, J.F., Loria, C.M., ...Aicher, K. (2008). Comparison of strategies for sustaining weight loss: The weight loss maintenance randomized control trial. Journal of the American Medical Association, 299(10), 11391148. doi:10.10001/jama.299.10.1139.


50 T h e Diabetes Prevention Program (1999). Design and methods for a clinical trial in the prevention of type 2 diabetes. Diabetes Care, 22(4), 623634. Unick, J. L., Beavers, D., Jakicic, J. M., Kitabchi, A. E., Knowle r, W. C., Wadden, T. A., ... Look Ahead Research Group (2011). Effectiveness of lifestyle interventions for individuals with severe obesity and type 2 diabetes: Results from the Look AHEAD trial. Diabetes Care, 34 (10), 2152 2157. doi:10.2337/dc110874 Van Dorsten, B., &, Lindley, E.M (2011). Cog nitive and behavioral approaches in the treatment of obesity. Medical Clinics of North America, 95(5), 971 988. doi:10.1016/j.mcna.2011.06.008 Van Wormer, J. J., Martinez, A. M., Martinson, B. C., Crain, A. L., Benson, G. A., Cosentino, D. L., & Pronk N. P. (2009). Self weighing promotes weight loss for obese adults. American Journal of Preventative Medicine, 36 (1), 70 73. doi:10.1016/j.amepre.2008.09.022 Wadden, T. A., & Butryn, M. L. (2003). Behavioral treatment of obesity. Endocrinology and Metabolism Clinics of North America 32 (4), 981 1003. doi:10.1016/S08898529(03)000720 Wadden, T. A., Butryn, M. L., & Byrne, K. J. (2004). Efficacy of lifestyle modification for long term weight control. Obesity Research 12(suppl), 151S 162S. Wadden, T. A., West, D. S., Neiberg, R. H., Wing, R. R., Ryan D. H., Johnson, K. C., ...Look Ahead Research Group (2012). Oneyear weight losses in the Look AHEAD study: Factors associated with success. Obesity 17(4), 713 722. doi:10.1038/oby.2008.637 Wing, R. R., La ng, W., Wadden, T. A., Safford, M., Knowler W. C., Bertoni, A. G., ...Look Ahead Research Group (2011). Benefits of modest weight loss in improving cardiovascular risk factors in overweight and obese individuals with type 2 diabetes. Diabetes Care, 34(7), 1481 1486. doi:10.2337/dc10 2415 Yanovski, S. Z., & Yanovski, J. A. (2002). Obesity. New England Journal of Medicine 346(8), 591 602. doi:10.1056/NEJMra012586 Yon, B. A., Johnson, R. K., Harvey Berino, J., Gold, B. C., & Howard, A. B. (2007). Personal digital assistants are comparable to traditional diaries for dietary self monitoring during a weight loss program. Journal of Behavioral Medicine, 30 (2), 165 175. doi:10.1007/s10865006 90921


51 BIOGRAPHICAL SKETCH Melissa Laitner was born in Lincoln, Nebr aska, but spent the majority of her childhood in Virginia. She graduated from West Potomac High School in 2007. Melissa went on to graduate from the University of South Carolina summa cum laude in 2011 with a B.A. in Experimental Psychology and a minor in Spanish. Following her g raduation, Melissa came to the University of Florida to continue her education in the fields of clinical and health psychology and public health. Melissa receiv ed her M.S. from the Universi ty of Florida in 2014 and is currently pursuing her research interests in the areas of obesity and health disparities.