ANALYSIS OF A COGN'ITIVELY ORIENTED SELF- MANAGEMENT WEIGHT CONTROL PROGRAM USING AN ‘ INTENSIVE CASE STUDY WITHDRAWAL‘ DESIGN Dissertation for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY RANDALL DEAN GOLD 1976 N >.. m-:é:~‘:":<‘€~ :‘ warm-qnm-umruuq‘avg” .‘g um i. “@4'9W537311292fifl34232 fi;f;1;-r.~mz -'?".-‘ K97 « 19:9: «ML: fi;¢3!~5!5§§!fl3§m new mum “genus: 54,1134HWIR‘-f>~r~:,g..u~t:¢~:~uhg¢a_ I a“. k‘ , L1,. u. u’ 4‘ w :. 1.- ~ . y » _ w . . .' b ,t :3 (V -, * ",ge “iv: My, "av ., . an, :7 V ", i ii . ’4 2~. n l»: U‘ ‘» This is to certify that the thesis entitled ANALYSIS OF A COGNITIVELY ORIENTED SELF—MANAGEMENT WEIGHT CONTROL PROGRAM USING AN INTENSIVE CASE STUDY WITHDRAWAL DESIGN presented by Randall Dean Gold has been accepted towards fulfillment of the requirements for Ph . D. degree in Education Major professor Date July 22, 1976 _______________._—’—— 0-7 639 llll l l l llll Ill l M iii It'll l l 3 1293 ABSTRACT ANALYSIS OF A COGNITIVELY ORIENTED SELF-MANAGEMENT WEIGHT CONTROL PROGRAM.USING AN INTENSIVE CASE STUDY WITHDRAWAL DESIGN BY Randall Dean Gold weight control research strongly suggests that among the various procedures used, self-management appears to be the most promising. Self-management procedures, however, are not as effective as is desired since weight loss relapses or stabilizes in the absence of treatment. until recently, self-management approaches to weight con- trol were based totally on an operant conditioning model. The self- management treatment packages might be more powerful if they combined the relevant aspects of the operant model with a cognitive model. The cognitive model emphasizes the role of covert speech (self-statements) in self-control. Behavior change is facilitated by modifying what clients say to themselves. The purpose of this study was two-fold: (a) to evaluate the short-term effectiveness of a combined model of self-management for weight control which stressed cognitive concepts, and (b) to identify potentially relevant client variables that might be related to the success or failure of the clients to implement the treatment package. Twelve clients who fulfilled the selection criteria were included in the study. An intensive case study withdrawal design was used. The Randall Dean Gold study was divided into five phases which include baseline, two treatment phases and two withdrawal-of—treatment periods. The two treatment phases and two withdrawal phases were five weeks each. During each treatment phase the clients met individually with their counselor once per week for one hour. A total of 10 treatment sessions was held. During each withdrawal phase no contact between clients and their counselors occurred. If the self-management procedures were effective, clients should be able to lose a meaningful amount of weight during two withdrawal phases, as well as during two treatment phases. The withdrawal phases are the most critical periods since clients are learning to control their own behavior in the absence of a counselor. A self-management treatment package was developed that stressed the cognitive model. The treatment package emphasized the following concepts: (3) self-management; (b) model of eating behavior; (c) model of self-statements; (d) realistic weight loss and activity goals; (e) motivation to change; (f) well-balanced, nutritionally sound diet; (g) situational strategies; (h) ArB—C model of behavior; (1) eating as a choice; and (j) self-instruction. Results of this study did not support the efficacy of the self- management treatment package to produce meaningful weight loss in the absence of a counselor. Meaningful weight loss was defined as a minimum of five-pound or 2 1/2% weight reduction during each five-week phase. Only two clients lost five or more pounds (or 2 1/2% body weight) during the first withdrawal phase, and none of them lost five or more pounds (or 2 1/2% body weight) during the second withdrawal phase. Randall Dean Gold However, some encouraging results were noted. Eight clients lost weight over the total program, and five clients lost 10 or more pounds. Six clients lost three or more pounds during at least one of the with- drawal phases. The clients varied considerably in their responsiveness to the self-management weight control program, which suggests that client variables were important. It was the opinion of the researcher from observing these clients over a 20-week period that several variables were important in their success or failure. Client variables can be classified as follows: (a) motivation, (b) emotional stability, (c) reasons for wanting to lose weight, (d) cognitive set to fail, (e) understanding and implementing the treatment concepts, and (f) personal organization and record-keeping. These client variables might be help- ful in planning future research efforts. Implications for the treatment of obesity and for self-management weight control programs were discussed, as well as suggestions for future research efforts. ANALYSIS OF A COGNITIVELY ORIENTED SELF-MANAGEMENT WEIGHT CONTROL PROGRAM USING AN INTENSIVE CASE STUDY WITHDRAWAL DESIGN By Randall Dean Gold A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY . Department of Counseling, Personnel Services, and Educational Psychology 1976 © Copyright by RANDALL DEAN GOLD 1976 ii DEDICATION This dissertation is dedicated to my mother and brother who are always available for support and my father who died before its completion iii ACKNOWLEDGEMENTS In the preparation of this dissertation, several people have pro- vided assistance and support. I would like to acknowledge and thank i the following individuals for their contributions. I want to thank Dr. Norman R. Stewart, my major professor, for his guidance and help during the preparation of this disseration and during my two years at Michigan State University. I would also like to thank the other members of my committee as follows: to Dr. Herbert M. Burks for his editorial expertise and his friendship; and to Dr. Verda M. Scheifley and Dr. Mark Rilling for their willingness to help and advise me on an individual basis. I would like to thank the following individuals for their devoted efforts as counselors in this study: Margaret Beahan, Laura Caffrey, Mary Edens, Ray Husband, Ginger Lange, and Judith Taylor. I would also like to thank Nancy Martin and Bruce Walker for their time and work as raters. I am especially grateful to Judith Taylor for her help with the experimental design and the statistical analysis, as well as her help with a pilot study. Without the help of these individuals, I would not have been able to conduct this study. I would like to express my appreciation to Dr. Mary H. Ryan who acted as the consulting physician, and to Ms. Donna Riggs for her excellent typing and her knowledge of the tasks necessary to prepare this final copy. I would also like to thank Linda Cooper for proof iv reading the original copy. TABLE OF CONTENTS page LIST OF TABLES .. ..... ...... .......... . ......... .......... ..... ... x. LIST OF FIGURES .. ..... ........................................... x1 CHAPTER I ........................................................ 1 INTRODUCTION AND REVIEW OF THE LITERATURE ................... 1 Purpose and Rationale I I O O O O O O O O O O O O O O O O 0 O O O O O O O 0 O O O O O O O 1 Definition of Self-Management ..................... ..... 3 MOdels o f SE]. f-Mana gement O O O O O I O O O O O O O O O O I O O O O O O O 0 O O O O O 4 operant mdel O O O O O O O O O O O O O O O O O O O O .......... O O O O O O O 6 COgI‘i-tive mOdel . C C O O C C C O O O O O O O O ..... O O O O O O O O C O O O O O 8 Combination of the two models ......... ..... ....... 14 Review of Related Weight Control Literature ............ 15 Factors That Influence Eating Behavior ................. 24 Client Variables Related to Weight Loss ................ 26 Summary ................................................ 28 CHAPTER 2 O O O O O O O O I O O O O O O O O O O O O O O O O O O O O I O O O O O O O O O ..... O O O O O O O O O O O O 32 DESIGN AND PROCEDURES 0 O O 0 O O O O O O O O O O O O O O O O O O O O O O O O O I O O O O O O I O O 32 Clients 0 O O C O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O I O O O O O O 32 selection ......OOOOOOOOOOOOO......OOOOOOOOOOOOOOOO 32 Description 0 ...... 00.0.00.........OOOOOOOOCOOOOOOO 32 Counselors ............................................. 33 Description ....................................... 33 Counselor A ....................................... 34 Counselor B ....................................... 34 Counselor C ....................................... 34 Counselor D . 34 Counselor E ....................................... 35 Counselor F ....................................... 35 Training procedures ............................... 36 vi TABLE OF CONTENTS (cont'd.) P38? Assignment of clients to counselors ..... ..... ..... 36 Treatment Package ..... .............. . ..... ..... ...... .. 37 Self-management ..................... ....... ....... 38 Conceptual model of eating behavior ............... 39 Conceptual model of self-statements ............... 41 Setting realistic goals ........................... 41 Motivation to change .............................. 42 Diet plan ......................................... 43 Situational strategies ............................ 44 AeB-C model of behavior ........................... 45 Eating as an act of choice ........................ 46 Self-instruction strategy ......................... 47 Implementation of the treatment ................... 48 DeSign ......OOOOOOOOOOI. ....... ......OOOOOOOOOOOOOOO... 49 Intensive case study ... ...... . ...... .............. 49 Summary of the design ............................. 50 Dependent variable ................................ 51 Definition of meaningfulness ...................... 52 Hypotheses and Data Analysis ...........v............... 53 sumry ......OOOOOOIOOI......OOOOOOOOOOOO0.00.00.00.00. 54 mAPTERg......OOOOOOOOOOOOO...0...... ..... ......O... ....... .0... 56 RESULTS ......COOOOOOIOCO................OOOOOOOIOOOOOOO0.... 56 Counselor Training and Treatment Implementation ........ 56 IndiVidllal Clients ......OOOOOCOOO......OOOCOOOOO0...... 58 Ann (Counselor E) ................................. 59 Beth (Counselor C) ................................ 59 Ken (Counselor B) ................................. 63 Fran (Counselor D) ................................ 66 Gail (Counselor E) ................................ 66 Jan (Counselor F) ................................. 71 Kris (Counselor C) ................................ 76 Lisa (Counselor A) ................................ 76 Mary (Counselor A) ................................ 79 Pat (Counselor D) ................................. 79 Tina (Counselor B) ................................ 84 Ron (Counselor B) ................................. 87 Individual Client Hypotheses ........................... 93 Group Data and Hypotheses .............................. 95 Posttreatment Questionnaire ............................ 97 vii TABLE OF CONTENTS (cont'd.) page Summary ........................................... 99 CHAPTER 4 ........................................................ 101 DISCUSSION AND IMPLICATIONS ................................. 101 Summary ................................................ 101 Discussion ............................................. 102 Limitations ............................................ 112 Proposed Client Variables .............................. 113 Motivation ........................................ 114 Emotional stability ............................... 116 Reasons for wanting to lose weight ................ 117 Cognitive set to fail ............................. 118 Understanding and implementing the treatment concepts ........................................ 118 Personal organization and record-keeping .......... 119 Implications .0.....................OOOOOOOOOOOOOOOOOOO. 120 Treatment of obesity .............................. 120 Treatment paCkage 0.00.00.00.00.......OOOOOOOOOOOOO 123 Suggestions for Future Research ........................ 127 In Retrospect ..........OOOOOOOCOOOOOOO............0.... 130 \, APPENDIX A - Counselor Training Rating Forms Counselor Self-Rating (lst Treatment Phase) ................. 132 Counselor Self-Rating (2nd Treatment Phase) ................. 133 APPENDIX B - Baseline and First Treatment Phase Forms Information Sheet ........................................... 134 Research Consent Form and Audiotape Release Form ............ 135 Medical Clearance ........................................... 137 Questionnaire ............................................... 138 Baseline Weight Record ...................................... 140 Daily Eating Form ........................................... 141 Daily Eating Record Summary Form ............................ 142 Hunger Rating ............................................... 143 Checklist ................................................... 144 APPENDIX C - Handouts for Clients during Second Treatment Phase Eating isanActOfCh01ce O.........OOOOOOOOO...OOOOOOOOOOOO 145 Self-Defeating Behavior Worksheet ........................... 149 Ways to Defeat 3 Weight Reduction Program ................... 150 viii TABLE OF CONTENTS (cont‘d.) page Self—Instruction Strategy .................................... 151 APPENDIX D - Detailed Outline of Each Counseling Session Activities at Each Counseling Session ................... ..... 154 APPENDIX E - Treatment Implementation Rating Forms Counselor Implementation Self-Rating (lst Treatment Phase) ... 163 Counselor Implementation Self-Rating (2nd Treatment Phase) ... 165 Counselor Implementation Rating Form ......................... 167 APPENDIX F - Posttreatment Questionnaire Posttreatment Questionnaire .................................. 169 APPENDIX G - Supplemental Description of Each Client Supplemental Description of Each Client ...................... 172 REFERENCES .......OOOOOOOOOOOOOOOOOOOO.......00.0.0000...00.0.00... 180 ix LIST OF TABLES Table page 1 Intensive Case Study Withdrawal Design ..................... 51 2 Mean Rating of Counselor Understanding of the Treatment Concepts Following Training ................................ 56 3 Mean Self-Rating of How Well Treatment Concepts and Activities Were Implemented by the Counselors for Each Client during Each Treatment Phase ......................... 57 4 Judges' Mean Ratings of Each Counselor's Performance in Implementing Selected Components of the Treatment Package .. 58 5 Background Information Concerning Each Client's Age, Sex, Marital Status, and Weight Factors ......................... 6O 6 Pound and Percentage Weight Change during Each Phase for EaCh Client ......OOOOOOOOOOO......OOOOOOOOOOOOOOOOOO0...... 92 7 Total Pound and Percentage Weight Change for Each Client ... 93 8 Analysis of Hypotheses 1 through 5 for Each Client ......... 95 9 Pound and Percentage Weight Change for All Clients during EaCh Phase 00............OOOOOOOOOOOOOO......OOOOIOOOOOOOOOO 96 10 Number of Responses for Each Phase in Which Clients Thought They Would Succeed and Fail; Phase in Which Client's Highest and Lowest Motivation Occurred; and Mean Motivation Rating for Each Phase ...................................... 98 11 Number of Times the Clients' Motivation Was Rated Lower in a Phase of the Study Than It Had Been Rated in the Prev1ous Phase O..0.........OOOOOOOOOOOOO0.00.00.00.00..0... 99 12 Clients' Mean Self-Rating of How Well Five Treatment Concepts Were Implemented .................................. 100 LIST OF FIGURES Figure 1 WEIGHT CHANGE DURING THE FIVE PHASES FOR ANN (69% Over— weight). Each data point equals Ann's weight at that time during each phase. The change in pounds and per- centage body weight is presented at the bottom of each phase (e.g., a loss of 5 pounds and a reduction of 2.5% body weight during the first treatment). .................. WEIGHT CHANGE DURING THE FIVE PHASES FOR BETH (64% Over- weight). Each data point equals Beth's weight at that time during each phase. The change in pounds and per— centage body weight is presented at the bottom of each phase (e.g., a loss of 4 pounds and a reduction of 1.7% body weight during the first treatment). .................. WEIGHT CHANGE DURING THE FIVE PHASES FOR KEN (58% Over- weight). Each data point equals Ken's weight at that time during each phase. The change in pounds and per— centage body weight is presented at the bottom of each phase (e.g., a loss of 14 pounds and a reduction of 5.1% body weight during the first treatment). ................... WEIGHT CHANGE DURING THE FIVE PHASES FOR FRAN (56% Over- weight). Each data point equals Fran's weight at that time during each phase. The change in pounds and per- centage body weight is presented at the bottom of each phase (e.g., a loss of 12 1/2 pounds and a reduction of 6.6% body weight during the first treatment). .............. WEIGHT CHANGE DURING THE FIVE PHASES FOR GAIL (44% Over- weight). Each data point equals Gail's weight at that time during each phase. The change in pounds and per- centage body weight is presented at the bottom of each phase (e.g., a loss of 7 pounds and a reduction of 3.9% body weight during the first treatment). ................... WEIGHT CHANGE DURING THE FIVE PHASES FOR JAN (44% Over- weight). Each data point equals Jan's weight at that time during each phase. The change in pounds and per- centage body weight is presented at the bottom of each phase (e.g., a loss of 4 pounds and a reduction of 2.1% body weight during the first treatment). ................... xi page 62 68 7O 73 75 LIST OF FIGURES (cont'd.) Figure page 7 10 11 12 WEIGHT CHANGE DURING THE FIVE PHASES FOR KRIS (41% Over- weight). Each data point equals Kris's weight at that time during each phase. The change in pounds and per- centage body weight is presented at the bottom of each phase (e.g., a gain of 1 pound and an increase of 0.6% body weight during the first treatment). ................... 78 WEIGHT CHANGE DURING THE FIVE PHASES FOR LISA (37% Over- weight). Each data point equals Lisa's weight at that time during each phase. The change in pounds and per- centage body weight is presented at the bottom of each phase (e.g., a loss of 1 pound and a reduction of 0.6% body weight during the first treatment). ................... 81 WEIGHT CHANGE DURING THE FIVE PHASES FOR MARY (34% Over- weight). Each data point equals Mary's weight at that time during each phase. The change in pounds and per- centage body weight is presented at the bottom of each phase (e.g., a loss of 5 pounds and a reduction of 3.0% body weight during the first treatment). ................... 83 WEIGHT CHANGE DURING THE FIVE PHASES FOR PAT (33% Over- weight). Each data point equals Pat's weight at that time during each phase. The change in pounds and per- centage body weight is presented at the bottom of each phase (e.g., a loss of 6 pounds and a reduction of 3.4% body weight during the first treatment). ................... 86 WEIGHT CHANGE DURING THE FIVE PHASES FOR TINA (36% Over- weight). Each data point equals Tina's weight at that time during each phase. The change in pounds and per- centage body weight is presented at the bottom of each phase (e.g., a loss of 7 1/2 pounds and a reduction of 4.5% body weight during the first treatment). .............. 89 WEIGHT CHANGE DURING THE FIVE PHASES FOR RON (27%I0ver- weight). Each data point equals Ron's weight at that time during each phase. The change in pounds and per- centage body weight is presented at the bottom of each phase (e.g., a loss of 11 pounds and a reduction of 5,1% body weight during the first treatment). ................... 91 xii CHAPTER 1 INTRODUCTION AND REVIEW OF THE LITERATURE Purpose and Rationale Self-management procedures, to date, have shown considerable po- tential for the treatment of obesity, but they fall far short of being an ideal treatment package. Typically, self-management pro- cedures produce weight loss during the treatment phase, but during followbup periods the weight loss seems either not to continue or may even relapse. A potent self-management package that will be useful to the person over an extended period of time, after the contact with the counselor has ended, needs to be deve10ped. Most self-management procedures for weight control are based on an operant conditioning model which emphasizes stimulus control, self- monitoring, self-reward, and self-punishment. Recently, self-manage- ment treatment packages for weight control have been developed which stress a combination of an Operant model and a cognitive model. The cognitive model focuses on what people say to themselves and the role of these self-statements in eating behavior and weight control. The Operant model and cognitive model each have unique contributions for self-management which may complement each other. Consequently, a combined self-management treatment package may be more powerful in allowing people to lose weight and to keep the weight off once formal l treatment is over. To date, little research has been reported con- cerning the effectiveness of an operant and cognitive model of self- management for weight control. This study will involve an investigation of the short-term effectiveness of a combined self-management treatment package which stresses the cognitive approaches but also includes concepts from the operant model. The study is designed to determine how effectively clients can lose weight in the absence of a counselor, which is the ultimate goal of self-management. Treatment will take placein two five-week phases with a five-week withdrawal of treatment following each phase. During each withdrawal phase, the clients will have no contact with the counselors. If the self-management procedures are effective, then the clients should be able to continue losing weight on their own. Also, this study will attempt to identify potentially relevant, client variables that will influence the effectiveness of self-manage- ment procedures. Very little research has been done which attempts to isolate client variables that are related to success or failure in self-management weight control programs. Clients will vary consider- ably in their responsiveness to self-management weight control pro- cedures, which suggests that client variables are very important. In order to understand how client variables influence weight loss varia- bility, a factor must first be identified and then actively manipulated in a controlled setting. Most research in this area has not identified the important factors in the client variables. In this study, through an intensive case design, an attempt will be made to identify some client variables that may contribute to the success or failure of a self-management program. Definition of Self-Management All reinforcement paradigms are comprised of two discrete elements or organisms (Premack & Anglin, 1973). One person is the reinforcer or punisher, whose action is contingent on a second person performing some behavior. Premack and Anglin hypothesize that a process called internalization or socialization consists of the supervisoral organism being put into the performing organism. Through this process, one organism can now check, restrain, or reward itself in a way that dup- licates the powers that would ordinarily come from the second organism. In general, every operation that goes on at the level of two people could also be arranged with one person (self-reward, self-punishment, self-extinction, etc.). If it is true that all classic two-organism processes can be duplicated in one organism, this would have powerful, far—reaching implications. Researchers have used numerous definitions of self-management which contain very similar ideas (Cautela, 1969; Goldfried & Merbaum, 1973; Kanfer, 1975; Thoresen & Mahoney, 1974). Most definitions of self-management include the following ideas: (a) the client is the agent of his/her own behavior change, and this change is based on a conscious decision; (b) self-control involves the change in the proba- bility of behavior when there are conflicting alternatives available; (c) the immediate consequence of the self-controlled behavior is less pleasant than the alternatives; (d) there is an absence of immediate external controls operating on the person; (e) self-control is not a global personality trait, but a specific response; and (f) self-control can be facilitated through the learning process. The labeling of a piece of behavior as "self-control" has the same problems as labeling behavior as abnormal, mentally ill, or healthy (Thoresen & Mahoney, 1974). The difference between "self-con- trol" and "non—self-control" is not qualitative. Rather, it is em- bedded in the social context in which the behavior is displayed and in how obvious the external influences are. Also, behavior that is socially desirable is more likely to be labeled as self-control than an undesirable behavior. Thoresen and Mahoney (1974) reject the idea of a dichotomy of self-control versus external control of behavior. They would rather think of a continuum classification and speak of various degrees of self-control. Any behavior can be classified in relative terms of internal and external control. For purposes of simplification the term "self-management" as used in this paper, will denote any response made by individuals to change or maintain their own behavior (Jeffrey, in press). Models of Self-Management Behavioral models of self-management have considerable appeal since they complement the humanistic concerns of freedom to change as one desires and of mastering one's environment. Individuals are seen as active agents who influence their environment and take full respon- sibility for their own actions and treatment. It is maintained that self-management techniques allow individuals to maintain their be- havioral changes in the absence of the counselor better than more traditional approaches. Also, it is maintained that self—management procedures are more efficient than other techniques, since clients conduct much of their treatment on their own in the natural setting. This avoids some of the generalization and maintenance problems of counselor-centered techniques (Jeffrey, in press). The behavioral models of self-management can be separated into two basic categories. In the first category, Operant conditioning is utilized as a primary underlying assumption. The operant conditioning models are the most numerous (Kanfer, 1975; Premack & Anglin, 1973; Thoresen & Mahoney, 1974; Watson & Tharp, 1972). In the second category, the underlying assumption is the congitive approach as con- ceptualized by Ellis (1962, 1973) in his development of rational- emotive therapy. Mahoney (1974), Meichenbaum (1974, 1975; Meichenbaum & Cameron, 1974), and Goldfried and Goldfried (1975; Goldfried, Decenteceo, & Weinberg, 1974) have been the leaders in developing a cognitive self-management model. Both models of self-management make a distinction between the "controlled response" (CR) and the "self-controlling response" (SCR). The CR is the target behavior that the client wishes to change, whereas the SCR is the act that is necessary to bring about the change. Both the CR and SCR are learned behaviors. Self-management always involves the increasing or decreasing of one or more behaviors. In accelerative self-control, the CR has im- mediate negative or aversive elements, but its long-term consequences are positive. For example, physical exercise may be unpleasant as one is going through it, but its delayed consequences--such as better health, more energy, and the loss of weight-~may be very positive. In decelerative self-control, the CR usually has positive immediate effects but aversive long-term effects. For example, overeating and alcoholism are immediately satisfying, but their delayed consequences are debilitating. The greater the difference between the immediate effects of the CR and the long-term effects, the more difficult it will be to produce the SCRs. Operant model. Thoresen and Mahoney's (1974; Mahoney & Thoresen, 1974) model of self-management will be summarized below since it is the most complete one based on Operant conditioning principles. This model emphasizes the interdependence between one's environment and his/her behavior. Behavior is not only a function of one's environ- ment which can be manipulated by others, but also, the environment is a function of one's behavior. Individuals can alter their environment in ways which will change their behavior and, consequently, become active agents in their own behavior change. The act (SCR) of manipula- ting the environment in order to facilitate behavior change is a learned response which must be rewarded, like any other behavior, in order to be maintained. The operant model of self-management maintains that some antece- dent or initiating stimulus (cue) precede the controlled response (undesired behavior). For example, the sight or smell of cookies may be a cue for betweenemeal snacking, or the feeling of anxiety may be a cue to open the refrigerator and eat. In long-standing habits which have seemingly become automatic, the antecedent cue may either be no longer present or unspecified. For example, habitual smokers may un- consciously light-up a cigarette without any awareness of an antecedent cue. When the relationship between the antecedent cue and the controlled response (CR) is explicit, the individual can make a conscious decision which may modify the CR by various self-controlling responses (SCR). This model provides for two basic self-management strategies. The first involves environmental planning (stimulus control); the person alters the antecedent stimuli that seem to elicit the target behaviors (CR and SCR). The second strategy is behavioral programming, which involves self-administered reward or punishment contingent upon some designated behavior. Most applications of self-management involve a complex combination of the two strategies. Environmental planning strategies have been used successfully with many self-management problems such as obesity (Stuart & Davis, 1972), developing appropriate study skills (Beneke & Harris, 1972), and smoking (Roberts, 1969). They can involve changing the cues or elimi- nating and avoiding them. Environmental planning strategies can also include the prearrangement of behavioral consequences. Such arrange- ments may involve contingency contracting with others for rewards or punishments for some specificied behaviors. Behavioral programming involves self-administered consequences that follow either the CR or the SCR. This procedure may include self- observation, positive and negative self-reward, and positive and nega- tive self—punishment. Behavioral programming can involve verbal sym- bolic self-praise or self-criticism, as well as presenting him/herself with tangible consequences after a specified behavior. Symbolic mediating factors should not be overlooked, since the payoffs for many self-control efforts are greatly delayed (Thoresen & Mahoney, 1974). Thought processes are an important factor in mediating between the stimulus and the response (Bandura, 1971). Cognitive model. The cognitive model of self-management has its roots in the rationale and procedures of "Rational-Emotive Psycho- therapy" (RET), which was developed by Ellis (1962, 1973). RET assumes that when a person experiences self-defeating behavior and in- appropriate emotions such as guilt, depression, frustration, or anxiety, it is because of maladaptive, irrational thoughts. Psychological problems arise from misperceptions and mistaken cognitions about events. A person's emotional reactions and behavior are caused by con- scious and unconscious evaluations, interpretations and philosophies. Ellis symbolizes his approach by an A-B-C-D-E sequence. "A" refers to some environmental event; "B" represents the person's belief system about "A," which is symbolized by self-statements and implicit sets of premises; and "C" refers to the behavioral consequence of "A" and "B." If "B" represents a rational, realistic set of beliefs about "A," then the person will behave in a rational and reasonable manner. But if "B" represents an irrational, unrealistic set of beliefs about "A," then the person will react with maladaptive emo- tions and behavior. "D" represents the therapeutic intervention of actively disputing the irrational beliefs of the client. This is the point where the counselor has the greatest impact. Ellis (1973, p. 60) states that: ...the therapist actively deunnstrates to the client how, every time he experiences a dysfunctional emotion or behavior or CONSEQUENCE, at point C, it only in- directly stems from some ACTIVITY or AGENT that may be occurring (or about to occur) in his life, at point A, and it much more directly results from his interpreta- tions, philosophies, attitudes, or BELIEFS, at point B. The emphasis in RET is on the person learning how to dispute ("D") his/her own irrational beliefs. "E" represents the beneficial effects of disputing the irrational beliefs. Operating from the basic assumption of RET that emotional arousal and maladaptive behaviors are caused by one's self-statements about the situation, Goldfried and Goldfried (1975) have attempted to syste- matize Ellis' therapeutic approach within a general behavioral frame- work. Their approach provides clear, delineated steps that the thera— pist might take in training the client to modify the irrational belief system. Their approach is called systematic rational restructuring. Rational restructuring involves four steps: (a) presentation of the rationale, (b) an overview of irrational assumptions, (c) an analysis of the client's problem in rational terms, and (d) teaching the client to modify the statements he/she makes to him/herself (Goldfried, Decenteceo, & Weinberg, 1974). The individual's covert speech plays a major self-regulatory role, and people can deve10p self-control by learning to change what they say to themselves (Meichenbaum, 1974, 1975). It is not the environmen- tal consequences of one's behavior that are the most important, but what a person says to him/herself about those consequences (Meichenbaum & Cameron, 1974). Meichenbaum (1974, 1975) has developed a self—instructional train- ing method as a means of modifying clients' self-statements, which, in turn, mediate behavior change. The primary emphasis of self-instruc- tion training is that the counselor helps the client explore and become aware of negative self—statements which lead to self-defeating behavior, and to modify these self-statements in order to produce new, more 10 adaptive behavior. Mahoney (1974) refers to this process as cognitive ecology--a cleaning up what you say to yourself. Meichenbaum's (1975) self-instructional methods involve three phases. Phase 1 is called the educational phase or conceptualization of the problem. This phase is designed to provide the client with an explanatory scheme for understanding his/her behavior. It is important that the conceptual framework have face validity or an air of plausi- bility for the client, and acceptance of it should naturally lead to the practice of specific, cognitive and overt, coping behaviors. Phase 2 is called the rehearsal or "trying on" phase. The clients listen to their self-statements and identify the self-defeating ones. Then, the client's maladaptive behaviors, thoughts, and feelings become a cue to employ the coping technique that is learned and practiced in therapy. Meichenbaum has shown that a person's self-statements can be modified in the following ways: (a) by modeling appropriate self- statements and behavior, (b) by having the person cope with the problem by means of self-instruction during desensitization, (c) by having the person cognitively rehearse self-instruction, and (d) by pairing the expression of coping self-instruction with shock offset. Phase 3 is the application phase. Once the client has become proficient in employing the new cognitive coping skills in the therapy session, the client is instructed to try out and practice these skills in the real situation. Considerable empirical evidence supports the value of this thera- peutic approach (Meichenbaum, 1974, 1975). Self-instruction has been used successfully with impulsive children, with text-anxious college students, in creativity training, with snake-phobic clients, and in 11 reducing smoking behavior. These preliminary research findings suggest that significant behavioral changes can be produced by the cognitive approach with greater generalization and persistence of treatment effects than other methods. Cognitive self-control training could in- fluence the cognitive styles and strategies that people use in various problem situations. Mischel and Ebbesen (1970) and Mischel, Ebbesen, and Zeiss (1972) have demonstrated how cognitive and attentional processes can deter— mine voluntary delay of gratification in preschool children. An experimental situation was arranged where preschool children could obtain a less preferred reward immediately or continue waiting indef- initely for a more preferred, but delayed reward. Several experiments were performed to determine which factors influence delay of gratifi- cation. It was found that children would wait much longer for a pre- ferred reward when they were distracted from the rewards than when they attendedlto them directly. But it was also found that only cer- tain cognitive events served as effective ideational distractors. When the children were instructed to think about the rewards them- selves or to think "sad thoughts," they could only delay gratification for a short period of time. When the children were instructed to think "fun things," they could delay gratification for long periods of time.> Thinking about the rewards resulted in an average delay of less than one minute, whereas thinking "fun things" produced an average delay of almost 15 minutes. When the children could see the reward objects, they could not delay gratification for very long. It is maintained that seeing or thinking about the rewards increases one's frustrations, which makes it difficult to continue the delay. The 12 implications of these studies suggest that effective delay of gratifi- cation depends on cognitive and overt self-distractions in order to reduce the frustration. Any condition which will shift a person's attention from the reward object seems to facilitate voluntary delay, which could have important implications for self-management. The use of self-instruction in the resistance to temptation was further studied by Patterson and Mischel (1975, 1976). They explored the issue of the ability to maintain attention to a central task and to resist distraction. This ability could be crucial in many kinds of self-control situations. In their study, preschool children were offered attractive incentives for completing a long, repetitive task, and they were also warned that a "Clown Box" might tempt them to stop working. The experimenter suggested to some of the children that they employ one or more cognitive or self-instructional plans. The depen- dent variable assessed each child's success in resisting the "Clown Box" temptation in order to continue working. The results indicated that the children who had received the self- instructional plans were more successful in resisting temptation in order to continue working than the children who did not receive the self-instructional plans. Also, it was found that a temptation-inhib- iting self-instructional plan (e.g., "I'm not going to look at Mr. Clown Box") was more effective in facilitating the children's self- control than a task-facilitating self-instructional plan (e.g., "I'm going to look at my work"). These findings suggest that a self-in- structional plan is a useful cognitive process that leads to successful resistance to temptation but that its efficacy may depend on its specific content. 13 In another study, the effects of self-instructional training in 5 to 6 year-old children in relation to fear of the dark were examined (Kanfer, Karoly, & Newman, 1975). Children were trained with one of three types of self-instruction: (a) the competence group, which in- cluded sentences emphasizing the child's active control or competence (e.g., "I am a brave boy/girl. I can take care of myself in the dark"); (b) the stimulus group, which included sentences intended to reduce the aversive qualities of the stimulus situation (e.g., "The dark is a fun place to be. There are many good things in the dark"); and (c) the neutral group, which included sentences that were not relevant to the situation (e.g., "Mary had a little lamb. Its fleece was white as snow"). The children were trained in a well-lit room. Then each child was placed in total darkness and remained there until he/she decided to increase the illumination. The duration of tolerance of darkness was measured. The results indicated that the competence self-instructional group was superior to the stimulus group and the neutral group. These findings are consistent with those of Patterson and Mischel (1976) and suggest that the effectiveness of self-instruc- tional training is related to the specific content of the learned sen— tences. These recent studies strongly suggest that what people say to themselves makes a difference in their behavior where delay of gratifi- cation and resistance to temptation are concerned. These two areas are very important in the development of self-control. People are con- stantly thinking, and cognitions exert a powerful influence on one's behavior and emotions. 14 Combination of the two models. The operant and cognitive models of self-management are not mutually exclusive. Recently, a greater emphasis has been placed on combining the two models. Williams and Long (1975) include a combination of both in their model for self- control. They state that not only does one change the factors in the environment that are controlling behavior but also, through self-ver- balization, an individual internalizes the self-management behaviors. A cognitive-behavioral self-control procedure for cigarette smokers was developed which stressed changing one's thoughts about smoking (Conway, 1975). This procedure involved developing self- statements concerning the ultimate aversive consequences of continued smoking and the person's motivation for wanting to quit smoking. The urge to smoke was used as a cue to recite these self-statements. Stimulus control strategies were also included for limiting the cone ditions under which cigarettes were smoked, as well as procedures for goal setting, self-contracting, and self-reward. The concept of "cognitive ecology" was combined with operant techniques as part of a treatment package with overweight counseling trainees (Horan, Robb, & Hudson, 1975). Self-defeating thoughts and excuses were directly challenged, and thoughts dealing with the desirable aspects of losing weight were encouraged. K. Mahoney and M. J. Mahoney (1976) have made the greatest effort to combine the two models of self-management in the area of weight control. Their weight control program includes important fac- tors from the operant model such as self-monitoring, stimulus control, and self-reward, as well as emphasizing the major role selfestatements play in weight loss failures and successes. It seems that 15 self-management treatment packages may be more powerful if they combine the relevant aspects of the operant and cognitive models. Both models have independent contributions for self-management which can be inter- dependent and complementary Review of Related Weight Control Literature An ideal program for weight control would accomplish at least three goals: (a) it would attract and retain the participants, (b) it would assist the participants in losing a desired amount of weight, and (c) it would allow the participants to continue losing weight, or allow them to maintain their ideal weight after the formal program is over (Harris & Bruner, 1971). Stunkard (1972), in his review of several weight control studies, reports that behavior modification approaches are more effective than previous methods of treatment. Penick, Filion, Fox, and Stunkard (1971) treated 32 patients who were at least 20% overweight by using behavior modification approaches and a traditional group therapy method. Even though the individuals' responsiveness to the behavioral treatment varied widely, it was concluded that the self-control training was superior to the group therapy and that behavior modification represents a significant advance in the treatment of obesity. M. J. Mahoney and K. Mahoney (1976a) also state that behavioral approaches to weight control seem to be better than any other method presently used. Stuart's (1967) study is important because it demonstrated that stimulus control procedures and weight and food monitoring can be effectively used in producing weight loss. The results showed impres- sive weight losses in 8 of 10 clients over a combined treatment and 16 followbup period of one year. The weight loss ranged from 26 pounds to 47 pounds, with the number of therapeutic sessions ranging from 16 to 41. A subsequent study by Stuart (1971), which involved a more refined version of his treatment package, also produced impressive results. Very few studies, since these two by Stuart, have been as successful. Even though the behavioral approaches have shown great promise in the treatment of obesity, they are far from ideal. Hall and Hall's (1974) and Abramson's (1973) recent surveys of the behavioral approaches to weight control point out many problems in this area. Much of the research methodology that is used by the experimenters is inadequate. It is quite common to find high attrition rates for the subjects in the treatment groups, and when significant differences are found be- tween treatment and control groups, the actual weight loss can be quite small. For example, Manno and Marston (1972) found a significant dif- ference between covert sensitization treatments and no treatment, with an actual average weight loss of slightly more than four pounds. Most research studies, to date, have not moved beyond the demonstration stage. There is a need to identify relevant client and treatment var- iables that contribute to a meaningful weight loss. Hall and Hall (1974) devide behavior modification techniques for weight control into two classes: Experimenter-Managed (EM) and Self- Managed (SM). These two categories are not totally dichotomous, but they emphasize relative differences in counselor behavior (director vs. teacher), client behavior (passive vs. active), and location (therapy hour vs. natural environment). In Hall and Hall's (1974) review of numerous studies in both the EM and SM category, it is reported that, 17 even with the methodological problems of the studies, the self-manage- ment techniques appear to be the most promising. The effects of external control and self-control on the modifica- tion and maintenance of weight have been directly compared (Jeffrey, 1974). The external control group was shown previous research that indicated that weight loss is promoted if the counselor dispenses financial incentives, which are previously deposited by the clients, for successful attainment of weight-control goals. The implication was that the responsibility for weight loss was the therapist's control of the rewards. At the conclusion of each weekly meeting, the therapist paid the clients $1.75 if they had met their weight loss goal and $2.50 if they had met their eating-habit improvement goal. If either or both goals were not met, the money was deposited in a locked cash box and not refunded. The self-control group was told that each person was responsible for his/her own weight management and that weight loss is promoted if they learn to appropriately reward them- selves for success. The clients would reward themselves, by money previously deposited, if they had achieved their goals. The self- reward was at the same rate as in the external control group. The results indicated that the self-control and external control treatments were equally effective in producing weight loss during treatment. However, the self-control group was more effective than the external control group in maintaining the weight loss at a six-week followbup. Several other experimental studies (Bellack, 1976; Harris, 1969; Mahoney, Moura, & Wade, 1973; Wollersheim, 1970) have reported suc- cess using self-management procedures with the average weight loss 18 ranging from 7.5 pounds to 15 pounds. Follow-up periods ranged from 0 to 16 weeks. Even though self-management procedures seem to be more effective than experimenter-managed techniques, they are far from ideal. Jeffrey (in press) points out that no self-management study has demon- strated sustained weight loss, following the treatment phase, and continuing until the individual reaches his/her desired weight goal. Most studies with relatively short followbup periods indicate that while weight gain did not necessarily occur, neither was there con- tinued weight loss. If the purpose of self-management techniques is to teach clients how to control their own behavior, one would expect continued weight loss after the termination of formal treatment, rather than stabilization. On this point, self-management techniques have not been successful. The long-term effectiveness of self-management procedures, in terms of weight change after treatment, has been directly examined (Hall, Hall, Hanson, & Borden, 1974). Clients consisted of a univer- sity sample and a community sample. Four treatment groups were utilized. The combined self—management group included 10 weekly meetings of 75 minutes each, and the clients were taught weight and food monitoring, stimulus control over eating, application of depriva- tion and satiation, self-reinforcement, self-punishment, techniques to break the chain leading to eating, and the development of a prepo- tent repertory. The simple self-management group included 10 weekly meetings of 10-15 minutes each, and the clients were taught how to monitor the number of bites of ingested food per day and to record these data. The bites per day were systematically decreased until 19 the clients lost 1-2 pounds per week. The nonspecific group met for 75 minutes each session for 10 weeks and discussed how tension aggra- vates and instigates overeating. They also practiced relaxation and developed hierarchies of stressful situations. The fourth group was a no-treatment control. The clients' weights were recorded at pre- and posttreatment and at three- and six-month follow-up periods. The results showed no differences in the mean percentage body weight loss between the two self-management groups at posttreatment and followbup, but the self-management groups at posttreatment and followbup did significantly better than the nonspecific and no-treat- ment control groups. The most important finding was the failure of the self-management treatment to produce continued weight loss or even to produce maintenance of weight loss. At the six-month followbup, the self-management groups had gained back more than half of their lost weight. Almost all of the clients in the combined self-manage ment group reported, at the three-month follow-up, that they no longer used any of the techniques taught them. Almost three-fourths of the clients in the simple self-management condition reported abandoning the method. One conclusion that could be assumed is that self-manage- ment methods are effective if clients use them. The question becomes how to teach self-management so that the clients can continue to use the methods without the therapist. Hall et a1. (1974) raise the issue that weight loss during treat- ment may be related to the demand characteristics of the experimental situation. Since both self-management treatments produced similar results, the demands of pariticpation may compel the clients to carry out the techniques during treatment, but with the termination of 20 treatment, the demand to perform the techniques is removed. Another study found no reliable evidence that group self—control training will result in permanent weight loss (Murray, Davidoff, & Harrington, 1975). Two groups of nine overweight women received 12 weekly sessions of self-control training in the presence of fattening foods. One of the two groups received alternative response training along with self-control training. These two groups were compared with a motivated control group of people who had volunteered for the study and an unmotivated control group randomly chosen from TOPS members who did not volunteer for the experiment. All the subjects in the study were members of the national TOPS (Take Off Pounds Sensibly) organiza- tion. Weights were recorded at 12-week intervals: 12 weeks prior to the start of treatment, pre- and posttreatment, and 12- and 24-week follow~up periods. The results showed that, at posttreatment, the group with self- control and alternative response training lost significantly more weight (-8.89 pounds) than the other three groups. There was no dif- ference among the other three groups. But at the 24-week follow-up, almost all of the average weight loss of the self-control and alterna- tive response training group had disappeared (+7.73 pounds). This finding raises serious questions concerning the long-term usefulness of this self-control treatment package. 4 A similar study by Jeffrey, Christensen, and Katz (1975) showed that two out of four subjects were unable to maintain their weight loss at a six-month followbup period. In this study, four clients used a standard self-management procedure and had a total of 26 treatment sessions over a 24-week period. Sessions were scheduled so 21 that a high density of patient-therapist contact occurred early in treatment, with a gradual reduction in contact as treatment progressed. The four clients had a mean weight loss of 27 pounds during treatment, but at the six—month followbup, two of the clients regained 14 and 17 pounds. The other two clients were able to maintain their goal weight. Jeffrey et a1. (1975) suggest that long-term followbup data are necessary in order to guard against overly optimistic hopes as to the effective- ness of behavior therapy for obesity. Several conclusions could be drawn from the above studies con- cerning the ineffectiveness of self-management procedures to produce weight loss in the absence of the counselor, following formal treat- ment. One could conclude that treatment was not long enough for the clients to fully incorporate the self-management techniques into their own behavior. The self-controlling behavior (new eating habits) is a learned behavior and must replace some very strong, well-practiced pat— terns (old eating habits). This process of learning may take place slowly over an extended period of time. Also, the client's motiva- tion to change may decrease. The clients' understanding and acceptance of the self-management procedure is not assessed in the above studies. It could be assumed that clients must thoroughly understand the purpose, rationale, and pro- cedure of each technique and accept it as a useful tool for weight loss if they are going to continue using it in the absence of the counselor. A fourth conclusion is that the self-management procedures used in the above studies are not potent enough to produce long-term change in clients. All of the self-management procedures used in these studies were based on the Operant model. It could be that the 22 operant-based self-management procedures are not powerful enough for a complex problem like weight control. Little research has been done concerning the use of a cognitive model of self-management or a com- bination of the operant and cognitive model for weight control. Thorn and Boudewyns (1976) conducted a study which had only two treatment sessions and a one-month follow-up. Four groups were included: behavior therapy (operant model), rational therapy (cogni- tive model), discussion group, and no-treatment control. The rational therapy group emphasized that when trying to lose weight, people make irrational statements to themselves which will undermine their effort. The results indicated that, at followbup, the behavior therapy group had lost significantly more weight (-3.4 pounds) than the other three groups. One can question the meaningfulness of the data because of the limited treatment time, small weight loss, and short followbup period. But the study does indicate a move toward including a cogni- tive model of self—management with weight control problems. One might assume that a combination of operant and cognitive self- management procedures would be a more potent treatment package for weight control than either model by itself. Both models have indepen- dent contributions for the self-management of weight control problems. Used together, the two models could complement each other. Horan et al. (1975) used a combination of operant and cognitive self-management procedures in their treatment package. Their results showed a signifi- cant average weight 1088 (-6.54 pounds) for the seven weeks of treat- ment, and, at a two-month follow-up period, the weight loss was maintained. Again, the meaningfulness of the data can be questioned since the clients did not continue to lose weight during the follow-up 23 period, and the average weight loss is relatively small. There is also no indication of how well the clients understood and utilized the various components of the treatment package. But, the study is en- couraging, since it demonstrates that a comprehensive treatment package can be developed and implemented for weight control using procedures from both the operant and cognitive models. The most comprehensive treatment package for weight control which combines the operant and cognitive models of self-management was developed by M. J. Mahoney and K. Mahoney (1976a). Their weight con- trol program consists of eight components: (a) self-monitoring, (b) sound nutrition, (c) moderate exercise, (d) stimulus control, (e) relaxation training, (f) social support, (g) self-reward, and (h) cog- nitive ecology. They stress that successful self-control is an exercise in personal science which includes a seven-step process. The word SCIENCE is used as a mnemonic device, with each letter represen- ting one of the steps: S - Specify the general problem area; C - Collect data; I - Identify regularities and possible problem sources; E - Examine the various Options and possible solutions; N - Narrow the Options and experiment; C — Compare your data with your previous data; and E - Extend, revise, or replace your solution. They also strongly emphasize cognitive ecology--cleaning up what you say to yourself. Inappropriate self-talk contributes to inappro- priate behavior, and cognitions play a critical role in one's eating patterns. Eating is often a response to thoughts about situations and feelings. Cognitive ecology includes setting reasonable goals and replacing negative monologues with more appropriate ones. Little research has been reported on the short-term and long-term 24 effectiveness of a self-management treatment package which combines the operant and cognitive models. Factors That Influence Eating Behavior There is no human function so vital as eating, except for breathe ing, and there is none that is influenced by so many psychological factors, except for sex (Creedman, 1974). 'It appears that people, unless they are starving, eat what they like and not necessarily what their body needs. People "learn to like" certain flavors and foods, and these learned habits are strongly influenced by experiences early in life. What people eat and how much they eat is a result of a complex system of learned attitudes, ideas, and assumptions. These learned attitudes, ideas, and assumptions are developed because food is used for so many different purposes in our society. Food is used to promote friendship, attain status, achieve security, relieve ten- sion, and influence the behavior of others (e.g., children). Schachter (1971a, 1971b) presents substantial evidence that ene vironmental cues such as sight, smell, color, flavor, psychological setting, and the passage of time play a key role in conditioning the feeling of hunger in obese people. Obese people are relatively insen- sitive to internal cues but are highly sensitive to environmental, fooddrelated cues. Consequently, the eating behavior of obese people is more under situational control than that of normal-weight people. The eating behavior of obese and nonobese subjects in a natural- istic setting has been examined (Gaul, Craighead, & Mahoney, 1975). Differences were found in the eating styles of obese and non-obese individuals. Obese subjects took more bites, performed fewer chews 25 per bite, and spent less time chewing than did the normal-weight sub- jects. Also, the eating behavior of obese and nonobese people has been observed and recorded, via videotape, in a laboratory setting (Hill & McCutcheon, 1975). Obese individuals ate more highepreference and less lowhpreference food than nonobese individuals. Although the findings are not statistically significant, the obese people ate more food, ate faster, and took fewer bites, which is similar to the findings of Gaul et a1. (1975). Overeating can be a result of strong emotions (Berland, 1975). Many people eat in order to fulfill emotional needs such as attention, caring, and affection. Others may overeat in order to avoid personal involvement with other peeple, or obesity may be used as a form of hidden rebellion. In times of tension, eating may act as a tranquile izer, and in times of boredom, eating may be a stimulant. Individuals who regained weight following a weight loss program showed a strong association between eating and emotional arousal (Leon & Chamberlain, 1973a). Both positive and negative emotions were associated with eating. Also, these people showed a greater preference for high-calorie foods. It was concluded that the emotional states are discriminative stimuli for food intake. Research on personality traits of obese versus normal-weight people has been sparse and unproductive. Sikes and Singh (1974) studied 64 obese and 62 normal-weight students to determine whether obese and normal-weight people differ in motivational strength, self- esteem, and compliance. They found no significant differences. Gormanous and Lowe (1975) report no significant differences in Rotter's I-E (internal-external) scale between normal and obese female students. 26 It appears that obese people cannot be distinguished from normal-weight people on the basis of a single or small group of personality traits. The distinguishing factors may be behavioral measures. Client Variables Related to Weight Loss Clients will vary considerably in their responsiveness to self- management weight control procedures, and researchers should attempt to evaluate client variables in order to better understand weight loss variability (Jeffrey, in press). Surprisingly, little research has been done in order to determine what client variables, if any, contribute to success or failure in self-management programs. Where research has been conducted, conflicting results are often obtained. Motivation is a major consideration in any self-management pro- gram (Kanfer, 1975). Clients must be motivated to make a decision to change and to initiate and carry through on the program. A relapse may * be seen as the extinction of the commitment to change. Many factors influence people to make verbal statements that they want to change. People are more likely to say that they want to change under the fol- lowing conditions: (a) delay in the onset of the program; (b) a his- tory of positive reinforcement for promdse-making; (c) recent indul- gence to satiation; (d) where guilt, discomfort, and fear over one's behavior is high; (e) escape from social disapproval; (f) the pres- ence of others making promises; (g) where the behavior to be changed, is private and cannot be easily checked; and (h) where the promise is vaguely phrased. Other conditions make it more difficult to make a verbal commitment to change. These are: (a) the program begins immediately; (b) past failures to keep promises have been punished; 27 (c) problematic behavior is not perceived to be under the client's con- trol; (d) positive reinforcement for the problem behavior is high; (e) criteria for change are too high; (f) consequences of nonfulfill- ment are harsh; (g) behavior is publicly observable; and (h) support for change is not anticipated. One might assume that the commitment to change would involve a class of covert self-statements. People tell themselves certain statements about the aversiveness of the target behavior, the impor- tance of change, the anticipated effort required to change, the expec- tation of success, and the ultimate positive consequences. One factor that might influence whether people progress, maintain, or relapse in a self-management weight control program would be the quality of their self-statements. Certain self-statements may enhance one's ability to implement a self-management program. Other factors, such as program requirements and reinforcement for success, would also influence the execution of the program. Unfortunately, no systematic research has been reported that attempts to clarify the motivational variables which contribute to the success or failure of a client engaged in a self-management weight control program. Locus of control as measured by Rotter's I-E scale has yielded conflicting results in predicting success in weight reduction. Balch and Ross (1975) found significant correlations between internal Rotter I-E scores and both completion and success in a weight reduction pro- gram. Success was defined as those subjects showing a weight loss greater than the median. No follow-up data were provided. On the other hand, Leon and Chamberlain (1973b) found no significant 28 differences in the mean I-E score for subjects who had maintained their weight loss for one year and those who had regained the weight they had lost. Also, Murray et a1. (1975) found no significant cor- relations between weight loss and scores on Rotter's I-E scale. Data on relevant client variables that relate to the success or failure in self-management weight control programs are severely lacking. Identifying and systematically studying relevant client variables are an area that deserves attention. The success of self-management weight control programs may depend not only on the type of treatment package, but also on the kind of factors operating within the person. Summary The purpose of this study is two-fold: (a) to evaluate a combined model of self-management for weight control which includes operant and cognitive concepts, and (b) to identify potentially relevant client variables that are related to the success or failure of the client to implement the treatment package. Little research has been reported concerning either of these areas. The term "self-management," in this paper, denotes any response made by individuals in order to change or maintain their own behavior. The clients are seen as active agents who are responsible for their own behavior change in the absence of immediate external controls. The self-control behavior is viewed as a specific response which can be facilitated through the learning process. The behavioral models of self-management can be separated into an operant conditioning model and a cognitive, self-statement model. The operant model utilizes environmental planning (stimulus control) 29 strategies which involve altering the antecedent stimuli that elicit the target behaviors and prearrangement of behavioral consequences. Also, a behavioral programming strategy is used which involves self- administered rewards or punishments contingent upon some designated behavior. The cognitive model emphasizes the role of covert speech (self- statements) in self-control. Behavior change is facilitated by modi- fying what clients say to themselves (cognitive ecology). Systematic rational restructuring and self-instructional training are methods used to modify negative self-statements and monologues. The evidence presented indicates that self-instructional training can increase delay of gratification, resistance to temptation, and tolerance for darkness (fear object) in young children. Delay of gratification and resistance to temptation are very important areas in self-management approaches to weight control. Until recently, the self-management approaches to weight control were totally based on the operant model. But it seems that self-man- agement treatment packages may be more powerful if they combine the relevant aspects of the operant and cognitive models. Both models have independent contributions for self-management which would be com— plementary in a weight control program. Weight control research strongly suggests that self-management procedures are the most promising, but they are far from ideal. If the purpose of self-management procedures is to teach clients how to control their own behavior, one would expect that clients, following formal treatment, could continue to lose weight until their goal is reached and be able to maintain their new weight. Typically, clients 30 will lose weight during formal treatment using self-management pro- cedures, but they will fail to continue losing weight after treatment and, in many cases, will regain the lost weight. Possible reasons for the ineffectiveness of self-management pro- cedures at followbup are: (a) formal treatment was not long or com- plete enough; (b) clients fail to understand and accept the self-man- agement procedures, and, therefore, discontinue using what they had learned; (c) the client's motivation to change decreases; and (d) the present self-management procedures are not potent enough to produce long-term change in the client's behavior. A combined operant and cognitive self-management package for weight control may offer a more potent treatment. But little research has been reported on the short-term and long-term effectiveness of a combined self-management package. This study will investigate the short-term effectiveness of a self-management treatment package which stresses the cognitive approaches but which also includes concepts from the operant model. This study is designed to determine how effectively clients can lose weight in the absence of a counselor. Self-management procedures are taught in two five-week phases with a five-week withdrawal of treatment following each phase in which the clients have no contact with the counselors. If the self-management procedures are effective, the clients should be able to continue losing weight on their own. Research suggests that obese and normal-weight people cannot be distinguished on the basis of a single or small group of personality traits. The distinguishing factors may be related to learned atti- tudes, ideas, and assumptions about food; sensitivity to 31 environmental, food-related cues; and the interrelationship between eating and emotions. Research on relevant client variables that relate to the success or failure in self-management weight control programs is severely lacking. Identifying these variables is an area that deserves attene tion since clients vary considerably in their responsiveness to these programs. Research in this area must first identify potential client factors and then actively manipulate them in a controlled setting to determine how they influence weight loss variability. This study, through an intensive case design, will attempt to identify some client variables that may contribute to the success or failure of a self- management weight control program. CHAPTER 2 DESIGN AND PROCEDURES Clients Selection. Twenty-nine people either responded to an advertise- ment placed in the school newspaper or were referred by Dr. Mary H. Ryan, Michigan State University Health Center. To be eligible for the study, the clients had to fulfill the following criteria: (a) must be at least 20% overweight, as measured by the 1969 U.S. Department of Agri- culture desirable weights table (Stuart & Davis, 1972); (b) must state that they want to lose 20 or more pounds; (c) must be available for both the winter and spring quarter, 1976; (d) must not be presently engaged in any other weight control program; (e) must receive medical clearance from Dr. Mary H. Ryan; and (f) must have completed all assignment sheets during the baseline phase. At the end of the baseline phase, 14 clients met all of the above criteria. The 12 clients to be included in the study were chosen from these 14. The clients were ranked according to their percent over- weight, and the top 11 were automatically included. Of the remaining clients, the one whose schedule best matched the counselor's schedule was selected. Description. The 12 clients included 2 males and 10 females who ranged from 27% to 69% overweight, with a mean of 45%. The age 32 33 ranged from 18 to 24 years, with a mean of 21. All of the clients were students at Michigan State University. Three of the clients were married, and nine were single. Three of the clients lived in university dormitories, and nine lived in apartments or married housing. Eight of the 12 clients described themselves as being overweight since childhood, and the other four stated that they had been over- weight since either junior high school or high school. Ten of the 12 clients reported having tried three or more methods in order to lose weight. These methods primarily consisted of special diets such as Dr. Stillman's water diet, Dr. Atkins' diet, grapefruit and egg, pro- tein and water, Weight Watchers, and fasting. Diet pills had been used by two of the clients. Ten of the 12 clients stated that they had wanted to lose weight for five or more years. The stated weight loss goals ranged from.25 to 75 pounds, with a mean of 52. On a scale of 1 to 5, with 1 being slight and 5 being very strong, 11 of the 12 clients gave a self-rating of 4 or 5 to the questions, "How unpleasant is your weight problem to you?" and "How important is it that you lose weight?" Counselors Description. One male and five female counselors were trained and implemented the treatment package with the clients. Three of the coun- selors were doctoral students in counselor education, and the other three were master's degree-level counselors. The master's degree-level counselors had completed all the required counseling courses for the master's degree, which included two supervised practicums, but lacked one or two elective classes in order to receive their degree. Both the 34 doctoral program and the master's degree program, in which these counselors were enrolled, have a behavioral orientation. Counselor A. Counselor A was a female, first-year doctoral stue dent in counselor education. She had an M.A. degree in counseling and had worked eight years in student personnel work, including assign- ments in resident halls and financial aids counseling. She had taught psychology and was directly involved in personal counseling, both group and individual, for three years. At the time of the study, she was tutoring students who were not succeeding academically at Lansing Community College. Counselor A had also been involved in a doctoral- 1eve1 practicum. Counselor B. Counselor B was a male, second-year doctoral student in counselor education. He had an M.A. degree in counseling and was currently a head advisor of a university dormitory. He previously spent two years as director of enrollment and development at a small college and did ministry work with high school students involving both individual and group counseling. Also, he lost 20 pounds during the study. Counselor C. Counselor C was a female, second-year doctoral stu- dent in counselor education. She had an M.A. degree in counseling and had several years of teaching experience. She had been involved in a doctoral-level practicum and had previous experience in conducting weight control groups. Also, she lost 17 pounds during the study. Counselor D. Counselor D was a female and had completed all of the required counseling courses for the M.A. degree in school counseling, 35 including one practicum at a high school and one at a middle school. She had taught junior high school for one year and had served as a paraprofessional counselor for Parent Effectiveness Training (P.E.T.) groups. She had also conducted career orientation groups for adult women. At the time of this study, counselor D was employed as a staff advisor to volunteer programs at Michigan State University. She was rated in the top 10% of her class by her practicum supervisor. She lost seven pounds during the study. Counselor E. Counselor E was a female and had completed all of the required counseling courses for the M.A. degree in school counsel- ing, including two practicums at the community college level. Two elective courses remaining on her program were completed half-way through the study. This counselor had performed volunteer work at the Listening Ear for one year and had received extensive empathy training. Also, she had conducted career planning groups at Lansing Community College and tutored disabled adults through an adult basic education program. Counselor E was rated in the top 10% of her class by her practicum supervisor. She lost 10 pounds during the study. Counselor F. Counselor F was a female and had completed all of the required counseling courses for the M.A. degree in school counseling, including one practicum at the community college level and one at a middle school. She had conducted career planning groups and human potential seminars at Lansing Community College. Counselor F held a secondary teaching certificate and, at the time of the study, was employed as an assistant residence advisor in a university dormitory. Counselor F was rated in the top 10% of her class by her practicum 36 supervisor. Also, she lost six pounds during the study. Training procedures. The counselors read the following material prior to the training sessions: (a) "Foods and Flavors" (Creedman, 1974), (b) "Eat, Eat" (Schachter, 1971a), (c) "Emotions and Eating" (Berland, 1975, pp. 30-39), (d) Slim Chance in a Fat World (Stuart & Davis, 1972), (e) "Cognitive Change Methods" (Goldfried & Goldfried, 1975, pp. 89-100), and (f) "Self-Instructional Methods" (Meichenbaum, 1975, pp. 357-372). Four training sessions of approximately two hours each were held. The general concepts of the treatment package and the specific activi- ties during each counseling session were discussed. Following each training session, the counselors rated how well they understood the treatment concepts that were discussed during that session (see Appendix A for the rating forms). Also, the researcher rated each counselor. Both ratings had to indicate a very good understanding of the concepts and activities before implementation was allowed. The counselors were encouraged to contact the researcher if they had any problems, ques- tions, or misunderstandings concerning the implementation of the treat- ment package. All interviews between the clients and counselors were audiotaped and monitored by the researcher as a further check on proper implementation. Assignment of clients to counselors. The only variable utilized to assign clients to counselors was sex. Male clients were assigned to the male counselor, and the female clients were assigned to the female counselors. Chesler (1971) and Hill (1975) report that counselors and clients respond differently, depending upon the sex of each. It is ’37 suggested that counselors are able to identify more with people with similar experiences and that counselors have more difficulty being empathic with opposite-sex clients than with same-sex clients. Also, female clients reported more satisfaction.when paired with female therapists than with male therapists. In our society, it is quite common for females to talk freely with other females about their weight problems, but it is less frequent for a woman to discuss her weight problems with a man. A frequent motiva- tor for college-age people to lose weight is appearance and attraction to the opposite sex. It may be assumed that having the same-sex coun- selor and client, will make it easier for the client to discuss this issue. Also, male-female sexerole stereotypic behaviors are eliminated by having same-sex counselor and client. The clients were treated individually, with four of the counselors (A, B, C, and D) treating two clients each, counselor E treating three clients, and counselor F treating one client. Assignment of female counselors to female clients was based on compatibility schedules. The free time of the clients was matched to the time that the counselors had available. Treatment Package A self-management weight control package was developed which incor- porated concepts from the operant and cognitive models. The cognitive aspects were more heavily stressed. The treatment was divided into two five-week phases in which the clients met individually with their counselor for approximately one hour, once per week. Consequently, treatment time consisted of 10 one-hour sessions. At each session, the 38 client's weight was recorded to the nearest half-pound. During the baseline phase and prior to meeting with the coun- selors, the clients performed the following: (a) they signed a re- search consent and audiotape release form; (b) they obtained medical clearance from Dr. Mary H. Ryan, Michigan State University Health Center; (c) they completed a questionnaire; (d) they completed the baseline weight form for three weeks; (e) they completed a daily eating record for one week and a summary form of their eating habits; and (f) they read the condensed edition of Stuart and Davis' (1972) Slim Chance in a Fat World (see Appendix B for copies of the above forms). The following concepts were emphasized during the first five-week treatment phase. Self-management. It was stressed that the clients will learn techniques that make it easier to control their own behavior and that they are responsible for their behavior change. The role of the coun- selor is to teach the clients techniques that will allow them to reach their goal and exercise control over their eating behavior. The respon- sibility of the client is to implement the techniques and perform the necessary homework assignments. Weight control programs could be classified into two broad areas. The first category emphasizes weight loss by having one's eating behavior controlled by outside sources (other-managed). Extreme examples in this area include being hospitalized where food intake is highly controlled by the staff, or having one's jaws wired shut. Special or fad diets in which the total responsibility for weight loss is placed on the diet are included in this area. Also, if the person is 39 only losing weight for some reason outside of him/herself such as spouse, friend, or group, this is seen as other-managed. The second category emphasizes weight loss through changing one's eating habits, which characterizes the self-management approach. In order to lose weight and keep it off, a change in eating habits must occur. The clients were told to think of self-management techniques as leading to a permanent weight loss and that people who continually lose weight and regain it are producing more harm to their system than if they had stayed overweight. For self-management approaches to be effective, the primary motivation must come from within the individual. Outside factors can help a person lose weight, but in order to be suc- cessful over the long term, the person must lose weight for him/herself. For example, if people lose weight primarily because they are attending a weight control group and they do not want to be embarrassed by not being successful (other-managed), the group becomes the main motivator to lose weight. As soon as they stop attending the group and the moti- vation is removed, it is almost assured that they will regain the lost weight. Losing weight is hard work, requires considerable effort, and must be done primarily for oneself. It is a selfish activity. Conceptual model of eating behavior. Overeating has positive ime mediate effects but aversive long-term effects. In weight control, this delay in aversive effects makes it more difficult to reduce food intake. A conflict between the immediate positive effects and long- range aversive effects exists. Consequently, the more immediate the aversive effects can become, the easier it is to control eating behavior. 40 Eating habits are learned, and what, when, and how much people eat are determined by a complex systenlof attitudes, ideas, assumptions, and feelings. Situational and stimulus factors such as taste, flavor, texture, temperature, color, appearance, and setting are important psychological factors in controlling eating behavior. Research sug- gests that for obese people, the actual state of the body (stomach) has very little to do with the report of hunger and the eating of food. External cues such as taste, smell, sight, and time of day seem to have a major influence on eating behavior. Taste and flavor are important immediate reinforcers for eating. Also, food is used to promote friendship, relieve tension or boredom, influence the behavior of others, and achieve security. Overeating can be a result of strong feelings. Some people have learned to eat when they are in stressful situations. Eating becomes a tranquilizer for them. Eating can occur when people are bored, and it acts as a stimu- lant. The logical consequences of the above findings are: (a) to make the aversive consequences of overeating more immediate, which can be done through cognitive techniques; (b) to reduce the external control of a person's eating behavior by rearranging the environment; and (c) to develop alternative responses to problem situations and feelings. Understanding and acceptance of this model are important. Clients were asked to validate the model from their own experience and to analyze their own eating behavior in terms of this model. understanding and acceptance of this model should logically lead to implementing specific cognitive and behavioral strategies which will help change eating habits and weight loss. 41 Conceptual model of self-statements. Research suggests that what people say to themselves plays an important role in controlling their behavior. People talk to themselves in sentences and phrases, and these self-statements--in the form of expectations, misperceptions, and assumptions--have significant implications for overt behavior. For example, if people say to themselves that they are likely to fail on a particular diet, then they are more likely to give up early. If people think they can succeed, then they will persist longer. Irrational beliefs and self-defeating thoughts can keep people from reaching their goal. This model will be developed in greater detail in the second five-week treatment phase. The conceptual models should have face validity or an air of plausibility for the clients, and the clients' acceptance of them is very important. The acceptance of the above models should naturally lead to the practice of cognitive and behavioral coping behaviors. Setting realistic goals. It is very important in weight control that clients set realistic weekly, phase, and long—term goals. The goals not only involve weight loss but should also emphasize the activities that will lead to weight loss. The activity goals lead to a change in eating habits. Unrealistic weight loss goals and activity goals can lead to frustration and discouragement and often contribute to failure in a program. The following points were emphasized: (a) significant weight loss is a long—term goal, since it took a period of time to gain the weight and it will take a long period of time to lose the weight; (b) research shows that a one— to two-pound loss per week is a healthy rate, 42 and that too rapid loss of weight can be harmful to the body; (c) since weekly weight can fluctuate greatly because of body fluids, it is important that the person find goal satisfaction from performing the activities that will lead to long-range weight loss; and (d) a long- range activity goal is the changing of the person's eating habits so that a permanent weight loss can be obtained. Throughout the entire program, a checklist was employed in order to give the counselors and clients feedback on how well they were performing the activity goals (see Appendix B for the form). Motivation to change. The client's motivation or commitment to change is a key variable to the success of a weight control program. The clients were asked where losing weight fits into their life. What is more important in the client's life than losing weight, and what is less important? The clients were asked to discuss the following ques~ tions: (a) How aversive is being overweight for you? (b) How impor— tant is it that you lose weight? (c) How much effort do you feel that it will take to lose weight? (d) How much effort are you willing to expend? (e) What are the positive consequences of losing weight? (f) Where does losing weight fit into other values in your life? and (g) What is your expectation of success? From the material discussed, motivational self-statements were implemented. The clients developed self-statements which would help them at problem times to stay committed to the program and reach their goals. The motivational self-statements consisted of positive conse- quences of losing weight and/or negative consequences of not losing weight. The clients were instructed to use personal statements that 43 would have great significance for them. Also, these motivational self-statements were attached to recurring cues such as hunger, boredom, a particular food, or a problem situation. Diet plan. The diet plan found in Slim Chance in a Fat World (Stuart & Davis, 1972) was utilized. This plan involves an exchange list and, if used properly, guarantees a well-balanced, nutritionally sound diet. It was emphasized that a calorie deficit must be created in order to lose weight. But it was also stressed that extreme states of deprivation should be avoided because, if a person gets too hungry, it becomes easier to break the diet and give up the diet plan. In order to control the states of deprivation, three meals a day were recommended. Also, the more extreme the diet, the harder it is to stay on the diet for an extended period of time. Starvation diets should be avoided because a person will lose muscle mass, as well as fat, resulting in damage to vital organs. The protein of the body (muscle and organs) will break down with the fat to supply energy to the body. Most nutritionists recommend not going below 1200 calories a day and a balanced diet. If the diet is not balanced, specific deficiencies and hungers can occur. A hunger rating form (see Appendix B) was therefore used, for two purposes. First, the deprivation state that a diet is creating can be assessed. If the client is too hungry, then an increase in calories is needed. If the client experiences a low level of hunger, the diet may be reduced or a check can be made to determine if the client is implementing the diet properly. Also, the hunger rating form often indicates to clients that they have desires to eat when they are not actually very hungry. 44 Criteria for selecting a diet are as follows. A good diet is: (a) one that is nutritionally sound which provides all necessary nutri- ents and helps decrease body fat without damaging the body structure, (b) one that can become a basic pattern of eating for the rest of the person's life, (c) one that consists of a wide variety of foods that are appetizing and pleasant to eat, (d) one that helps train the appe- tite and develop a pattern of eating at regular intervals, (a) one that provides foods with staying power to prevent excessive hunger, (f) one that is built around a nucleus of familiar foods which can be adaptable to the person's living situation, and (g) one that can take into account individual differences and can be planned accordingly. A bad diet is: (a) one that has a limited choice of food, which leads to feelings of deprivation and monotony, and does not provide all the necessary nutrients; (b) one that requires special foods that make the diet difficult to follow in normal living situations; (c) one that leads to rapid weight loss which may be detrimental to health and damaging to the body structure; (d) one that does not establish a pattern to follow for the rest of the person's life; and (e) one that can be used for only a short time. Situational strategies. From the situational strategies which are discussed in Slim Chance in a Fat World (Stuart & Davis, 1972), two or three that were most relevant to the client were implemented. Each client, depending upon his/her situation, will find some of the situa- tional strategies more apprOpriate than others. The list of situational strategies includes: (a) arrange to eat in one room only, and only in one place in that room; (b) when you eat, 45 avoid other activities; (c) buy nonfattening foods when possible; and keep problem foods out of sight and out of reach; (d) if you eat problem foods, make sure that they need preparation; (e) do the gro- cery shopping after you have eaten, and always shop from a list; (f) train others to help you curb your eating; (g) measure all portions, make small portions of food appear large, and make second helpings hard to get; (h) take steps to avoid hunger, loneliness, depression, bore- dom, anger, and fatigue; (1) always keep on hand a variety of safe foods to use as snacks; (j) keep track of how much you have eaten and how much more you can eat within you diet at all times, everyday; (k) slow down the rate at which you eat; (1) keep a graph of how much you eat, how much you exercise, and how much weight you lose; (m) build in some payoff for following every step in the program; (n) do not serve high—calorie condiments at meals; (0) allow children and spouses to take their own sweets and desserts; and (p) clear plates directly into the garbage. The following concepts were emphasized during the second five-week treatment phase. AeB-C model of behavior. This model was used as a general intro- duction to the concept that self-statements and choices play a vital role in controlling people's eating behavior. "A" refers to some agent, activity, or action (e.g., problem food, situation, or feeling). "B" refers to the individual's belief system about "A" which is symbolized by self-statements and implicit sets of premises. "C" refers to the choice one makes as a consequence of "A" and "B." If "B" represents a rational, realistic set of beliefs about "A," then the person will make 46 a choice that leads to rational and reasonable behavior which allows goal attainment. But if "B" represents a misunderstanding, rationali— zation, or irrational set of beliefs about "A," then the person will make a choice that leads to self-defeating behavior (SDB). Therefore, SDB is caused by misperceptions and misevaluations of events which lead to inappropriate choices (Ellis, 1973). In order to get the client to accept this model, it was first pre- sented with non-diet and non-weight loss examples. After the client agreed that behavior is caused by what a person tells him/herself about the situation, the model was discussed in weight control terms. "A" is a problem food, situation, or feeling; "B" is self-statements about "A"; and "C" is the choice a person makes, such as maintaining the diet or breaking the diet. Eating as an act of choice. It was strongly emphasized that when, what, and how much people eat is clearly a choice on their part. Food is not flying off the plate and forcing itself into a person's mouth; rather, it is being placed there by his/her hands. People have a choice in their eating behavior (e.g., the eating habits of other cul- tures). Acceptance of this idea by the client is important, since it sets the stage for what is done during the second five weeks of treat- ment . The remainder of this concept was adapted from Cudney's (1975) work on self-defeating behavior. It was emphasized that overeating is a self-defeating behavior (SDB) which involves a number of choices by the person. These choices are mediated by self-statements. The factors which lead to SDB choices were discussed. These included fears, 47 techniques, disowning, and minimizing the negative consequences. The factors which lead to making goal—attainment choices were also dis- cussed. The clients were given a self-defeating behavior worksheet in order to become aware of how they implemented their self-defeating choices, as well as a handout on ways to defeat a weight reduction program. Self-instruction strategy. The purpose of self-instruction is to help the client, at choice points, make goals-attainment choices in- stead of self-defeating ones. The self-instruction package should con- tain statements from three basic areas: (a) confronting the problem (b) an elaboration of the negative effects of self-defeating behavior and/or the beneficial effects of goal-attainment choices, and (e) reinforcing self-statements. The elaboration of the negative effects of self—defeating behavior and/or the beneficial effects of goal-attain- ment choices is the most important area. In order for the unwanted effects or desired effects to have a greater impact on the client, it is necessary to make them as explicit, detailed, emotionally charged, and immediately meaningful as possible. The clients must make immedi- ate and repeated connections between their reasons for wanting to lose weight and their urge to engage in self-defeating behavior. Also, the need to practice "talking to yourself" in a positive way instead of a self-defeating way was stressed. The more the client practices the positive self-statements, the easier it will be to resist each desire for self-defeating behavior. The handouts given the clients concerning the two above concepts, the self-defeating behavior worksheet and ways to defeat a weight 48 reduction program, are included in Appendix C. A detailed outline of what occurred in each of the 10 counseling interviews is presented in Appendix D. Implementation of the treatment. In order to determine how well the treatment package was implemented by the counselors, two types of ratings were used. Using the counselor implementation self-rating form (see Appendix E), each counselor rated how well he/she implemented the various concepts and activities for each client, during each interview. A scale of l to 10 was used, with a rating of 1 indicating that the concept or activity was not implemented and a rating of 10 indicating that the concept or activity was implemented extremely well. A second rating was developed to determine how well the important components of the treatment package were implemented. Of the four coun- selors who had two or more clients for the entire study, one of their clients was randomly chosen for this rating. The two counselors who saw only one client for the entire study were rated on that client. Two independent raters were trained and rated the counselors using the counselor implementation rating form (see Appendix E). After train- ing, the interrater reliability was established by having each rater judge the same counselor implementing the concepts. A reliability coefficient was determined by a Pearson product-moment correlation After an interrater reliability of at least .85 was established, each rater judges three of the counselors on the following components of the treatment package: (a) model of eating behavior, (b) model of self-statements, (c) motivational self-statements, (d) attaching self- statements to recurring cues, (e) AeB-C model, (f) eating as an act of 49 choice, and (g) self-instruction strategy. Design Intensive case study. The intensive case study design appears to be most appropriate when one is trying to determine the variables that may cause a particular client's behavior and what treatment will work with that client (Miller & Warner, 1975). The kind of question the re- searcher seeks to answer should dictate the type of experimental design. With a between-groups design, the question is whether the average response across clients with one treatment is sufficiently greater than a corresponding group average on a second treatment (Chassan, 1967). But with a between-groups design, it is impossible to distinguish the particular client who improved because of the treatment, and the treat- ment effect cannot be specifically related to client characteristics and parameters. It is possible that particular treatments are only effective with some clients. The between-groups design adds little to the understanding of the causes of change in individual behavior. Thoresen and Anton (1974) state that "Exclusive concern with group means and variabilities impedes understanding of the treatment process and may lead to erroneous generalizations about treatment effects" (p. 553). Research in counseling should stress direct observation, careful description, and systematic planned intervention with individual clients (Thoresen, 1972). The intensive case study design offers several advantages for counseling research. These advantages include: (a) the specific behaviors of the individual client are the unit of analysis rather than group means; (b) the specific characteristics of the individual's behavior can be examined continuously during each phase 50 and between phases of the investigation; (c) individuals serve as their own control, and behavior change is compared to their baseline behavior; (d) experimental control of variables is greatly facilitated, which reduces the need for statistical control; (e) causal relation- ships can be established by replication of findings across clients; and (f) the intensive design offers a way of investigating the covert (internal) behavior of individuals. An intensive case study design was used in this study for the following reasons: (a) client responsiveness to self-management weight control programs will vary widely; (b) it is possible to identify which clients do well and which clients do poorly; (c) client variables that influence the success or failure of the treatment package for each client can be more readily identified; and (d) the effectiveness of the treatment is determined by plotting the client's progress on a repeated measures basis. Summary of the design. This study used an intensive case study withdrawal design (Leitenbeng, 1973). In the withdrawal design, the major concern is what happens to the target behavior (weight loss) when the self-management procedure is instated, withdrawn, then reinstated and withdrawn. With self-management procedures, the withdrawal phase is the most critical period since the clients are learning to control their own behavior in the absence of a counselor. The design for this study is presented in Table 1. The study was divided into five phases which include baseline, two treatment periods, and two withdrawal periods. During each treatment phase, the clients met once per week with their counselor. During each 51 withdrawal phase, the clients and counselors had no contact, but the clients reported, once per week, to be weighed on the researcher's scales. If self-management procedures are effective, the clients should be able to lose a meaningful amount of weight during the two withdrawal phases. Table 1 Intensive Case Study Withdrawal Design Phase I II III IV V Baseline Treatment 1 Withdrawal 1 Treatment 2 Withdrawal 2 6 weeks 5 weeks 5 weeks 5 weeks 5 weeks Dependent variable. Three criteria should be considered in the selection of a dependent variable for weight reduction studies (Bellack & Rozensky, 1975). These include: (a) compatibility with the data analysis procedures, (b) accurate representation of the relevant effects of treatment, and (c) comparability among studies. Each measure that is used in weight reduction studies has advantages and disadvantages. The most direct measure of change is pounds lost. The major dis- advantage of this measure is not taking initial weight into considera- tion. In order to deal with this problem, percent of body weight lost is used. But this measure does not separate body weight from amount of excess body fat. Both percentage change and raw change fail to take into account the weight-change goals of the client. A third measure that can be used is change in percentage over- weight, which incorporates both initial weight and a reasonable goal 52 in the calculation. Typically, ideal or desirable weight charts are used to determine percentage overweight and goal. However, the validity of ideal weight charts as a measure of obesity has been questioned. Also, this measure strongly favors lighter clients by making their losses appear disproportionately large. The dependent variable in this study was the change in pounds and the percentage of bocbrweight change during treatment and withdrawal phase for each client. These measures were determined by comparing the last weight of each phase with the last weight of the previous phase. Enough data are provided for the reader to determine change in percentage overweight. The clients were weighed at the beginning of the baseline phase and at the fifth and sixth week. During treatment and withdrawal phases the clients were weighed once per week. Their weight was recorded to the nearest half-pound on a Continental doctor's scale, Model 134. Definition of meaningfulness. Most research (M. J. Mahoney & K. Mahoney, 1976a; Stuart & Davis, 1972) strongly indicates that individuals should lose weight at a rate of one- to two-pounds per week and that people should not lose more than 1% of their body weight per week. Weight loss at a faster rate can be damaging to a person's health. In order to evaluate the treatment package, a minimum of one pound per week weight loss over each phase (five pounds or 2 1/2% body weight per phase) must be attained to be considered meaningful. There- fore, if the self-management procedures were effective, each client should be able to lose a minimum of five pounds (or 2 1/2% body weight) 53 during each withdrawal phase, as well as during each treatment phase. Hypotheses and Data Analysis The following hypotheses were tested in this study: H o 1 Each individual will show a 20-pound (or 10% weight reduction from the end of the baseline phase to the end of the second withdrawal phase. Each individual will show a five—pound (or 2 1/2%) weight reduction during the first treatment phase. Each individual will show a five-pound (or 2 l/2%) weight reduction during the first withdrawal phase. Each individual will show a_five-pound (or 2 1/2%) weight reduction during the second treatment phase. Each individual will show a five-pound (or 2 1/2%) weight reduction during the second withdrawal phase. The data for each client is plotted in graph form across the five phases. Each of the above hypotheses for each client was analyzed according to the data in each phase. H6 There will be no difference in the mean weight loss for all clients during the first treatment phase and the mean weight loss for all clients during the first withdrawal phase (T1 - W = 0). 1 There will be no difference in the mean weight loss for all clients during the second treatment phase and the mean weight loss for all clients during the second withdrawal phase (T2 - W2 - 0). There will be no difference in the mean weight loss for all 54 clients during the first treatment and the first withdrawal phases combined, as opposed to the second treatment and the second withdrawal phases combined [(T1 + W1) - (T2 + W2)] = 0. The data for all of the clients who completed the study was pooled, and the mean weight loss for the two treatment and the two withdrawal phases were analyzed by analysis of variance (ANOVA) with repeated measures. If the self-management procedures are effective, a meaningful weight loss should occur during each withdrawal phase (H3 and H5) and during each treatment phase (H2 and H4). Summary Twelve clients, two males and 10 females who fulfilled the selec- tion criteria, were included in the study. They were assigned, by sex, to six counselors who had been trained in the treatment methods. The counselors were either M.A. degree-level counselors or doctoral students in counselor education. A self-management treatment package was developed that stressed the cognitive model. The treatment package emphasized the following concepts: (a) self-management; (b) model of eating behavior; (c) model of self-statements; (d) realistic weight loss and activity goals; (e) motivation to change; (f) well-balanced, nutritionally sound diet; (g) situational strategies; (h) ArB-C model of behavior; (1) eating as a choice; and (j) self-instruction strategy. The treat- ment was divided into two five-week phases in which the clients met with their counselor once per week. An intensive case study withdrawal design was used. The study was 55 divided into five phases which include baseline, two treatment periods, and two withdrawal periods. If the self-management procedures were effective, the clients should lose a meaningful amount of weight during the two withdrawal phases, since they are learning to control their own behavior in the absence of a counselor. The purpose of the intensive case study design is to: (a) determine the effectiveness of the treatment, (b) identify which clients do well and which do not, and (c) identify possible client variables that influence success or failure. The dependent variable was change in pounds and the percentage of body weight change during treatment and withdrawal. A minimum weight loss of five pounds or 2 1/2% body weight during each treatment and withdrawal phase is considered meaningful. The hypotheses were analyzed according to the cirterion of meaningfulness. The data for all 12 clients were pooled and analyzed by analysis of variance with repeated measures . CHAPTER 3 RESULTS Counselor Training and Treatment Implementation \ In order to assess the counselors' understanding of the treatment package following training and prior to implementation, a self-rating and the researcher's rating of their understanding of the treatment concepts were utilized (see Appendix A for the rating form). A scale of l to 10 was used, with a rating of 1 indicating no understanding of the concept and a rating of 10 indicating that the concept was under- stood extremely well. Twenty-four specific treatment concepts were rated. The mean ratings for each counselor on the concepts used in each treatment phase are presented in Table 2. Table 2 Mean Rating of Counselor Understanding of the Treatment Concepts Following Training Self-Rating, Researcher's Ratigg Counselor T—l T-2 T-l T-2 A 9.28 8.77 8.55 8.62 B 7.64 7.93 8.37 8.16 C 9.00 9.00 9.19 8.93 D 8.19 8.23 8.28 8.15 E 9.28 10.00 8.91 9.16 F 9.64 9.62 8.91 9.39 Note. Maximum score . 10. T-l and T-2 represent first treatment phase and second treatment phase, respectively. 56 57 In order to assess how well the treatment package was implemented by the counselors, two types of ratings were used. The mean counselor implementation self-ratings for each client are presented in Table 3. The counselors rated how well various concepts and activities were implemented during each counseling session on a scale of l to 10 (see Appendix E for the rating forms). Table 3 Mean Self-Rating of How Well Treatment Concepts and Activities Were Implemented by the Counselors for Each Client during Each Treatment Phase Mean Self-Rating Treatment 1 Treatment 2 Total Counselor A Lisa 7.44 7.30 7.37 Mary 7.28 6.84 7.06 Counselor B Ken 7.39 7.13 7.26 Ron 7.56 6.87 7.22 Counselor C Beth 7.11 ---a 7.11 Kris 8.78 7.64 8.21 Counselor D Fran 8.11 7.50 7.81 Pat 7.33 6.56 6.95 Counselor E Ann 7.89 8.00b 7.95 Gail 9.56 8.87 9.22 Tina 9.78 9.48 9.63 Counselor F Jan 8.39 9.30 8.85 Note. Maximum score - 10. aClient terminated following the first treatment phase. bThis score is based on two counseling sessions. 58 The mean counselor ratings by independent judges on the implemen- tation of the important components of the treatment package (see Appendix E for the rating form) are presented in Table 4. A scale of l to 5 was used, with a rating of 1 indicating that the objective was not implemented and a rating of 5 indicating that the objective was implemented extremely well. Following the training of the judges, an interrater reliability for the components in the first treatment phase was .94 and the interrater reliability for the components in the second treatment phase was .87. The overall interrater reliability was .87. Table 4 Judges' Mean Ratings of Each Counselor's Performance in Implementing Selected Components of the Treatment Package Counselor Components A B C D E F First Treatment Phase 3.63 4.50 4.63 2.86 3.25 4.04 Second Treatment Phase 3.88 4.18 3.99 3.24 3.13 4.21 Grand Mean 3.76 4.34 4.31 3.05 3.19 4.13 Note. Maximum score = 5. The researcher was satisfied with the counselors' understanding of the treatment package and their implementation of the concepts and activities during the counseling sessions. Individual Clients Twelve clients participated in this study. Data from 11 clients were collected for the entire program. One client termdnated following the first treatment phase. Another client stapped seeing the counselor 59 after the second meeting in the second treatment phase, but agreed to provide her weights until the end of the program. A summary of the characteristics for the 12 clients is presented in Table 5. A supple- mental summary description of each client is presented in Appendix G. Ann (Counselor E). Ann, who was 69% overweight, had been some- what overweight all of her life and obese since her second year in college. Both parents, three of four grandparents, and all of her siblings were overweight. She had tried to lose weight several times using different methods, none of which had resulted in a permanent weight loss. Ann missed two counseling sessions during the first treatment phase which had to be rescheduled, and stopped attending altogether after the second meeting with the counselor in the second treatment phase. No reasons were given. She did, however, agree to weigh her- self at the end of the second treatment phase and at the end of the second withdrawal phase. Counselor E described their relationship as poor and stated that Ann would often act disinterested or even antago- nistic during the counseling sessions. After the counseling sessions, Ann would often want to discuss her boyfriend problems. Ann's weekly weight recording by the researcher throughout the study is presented in Figure 1. Changes and projections are noted. Beth (Counselor C). Beth, who was 64% overweight, stated that she had never been normal weight and had always been taller and heavier than the other children her age. Her father and brothers were also described as being overweight. She had tried numerous times to lose weight and reported that the most successful she used was Weight Watchers. 60 .Ammma .m«>mn w unmaumv manna munwfios oanmuwmoe ououaaowuw< mo uaoauumaon .m.= mead men he eoadaumuonm qu Nm Hm some NAN 05H omH co saw 2 x mm com “om NNH ems on sea 2 m Hm mafia Nmm «ma Hes mm osa m m ON one Nam mNH mNH cs won m a nu son: “an man man as ems m m NN mafia mas mNH and me can m m Hm mane Nee mma can on can m e me one use mNH ONH oo own m m ON dame “on man non mm can a m NN onus New msH HON mm ohm m 2 ON cox use wea was we mam m m am noon Noe mHH mNH an ass m m HN oo< unmeasuo>o moewsos pawns: Hooo mmoo sesame msumom xom owe uooaao unmoumm HmomH mouwmmn uanmB moumum mcHHmmmm Hmafim Houwumz muouomm uanmS mam .moumum Hmuwumz .xmm .mw< m.ucofiao zoom wawoumoooo cowumEHOMCH vcsouwxomm m mHaMH 61 Figure 1. WEIGHT CHANGE DURING THE FIVE PHASES FOR ANN (69% Overweight). Each data point equals Ann's weight at that time during each phase. The change in pounds and percentage body weight is presented at the bottom of each phase (e.g., a loss of 5 pounds and a reduction of 2.5% body weight during the first treatment). 62 H ouowwm some: 3 A38: 3 A33: 3 Games 8 Exam: 3 A£3Hm v3 rm Houm ... IIIIIO owemno uemHm3 ernuo< 1 SH SH mmH I: omH SCI NflOd 87 Figure 11. weight changes and projections are noted. Ron (Counselor B). Ron, who was 27% overweight, stated that the first time he felt overweight was when he became a lifeguard at age 17. He had tried to lose weight three times, with the last attempt being four years ago. During the first treatment phase, Ron was highly motivated and kept very careful records regarding his diet. He tended to see the diet as a challenge to him. He did well on the program until the fourth and fifth week of the first withdrawal phase, which coincided with final examinations and quarter break. At that time, Ron reported that he reverted back to his old eating habits. Ron also stated that the significant initial weight loss reduced his motivation for further weight loss. From that point, Ron slowly regained some of his lost weight. Ron's weekly weight recording throughout the study is presented in Figure 12. Changes and projections are noted. The pound and percentage weight changes for each client during the two treatment phases and the two withdrawal phases are presented in Table 6. The total pound and percentage weight changes for each client over the entire program are presented in Table 7. Figure 11. 88 WEIGHT CHANGE DURING THE FIVE PHASES FOR TINA (36% Over- weight). Each data point equals Tina's weight at that time during each phase. The change in pounds and percentage body weight is presented at the bottom of each phase (e.g., a loss of 7 1/2 pounds and a reduction of 4.5% body weight during the first treatment). 89 2 83$ A38: 3 32.83 3 38: e 332 3 A33: 8 H<3Hm vou onoum .7 IIIIIO owemnu uano3 43564 I II II a T. II 9: I. I... a: i 1 L o: SCINflOd Figure 12. 9O WEIGHT CHANGE DURING THE FIVE PHASES FOR RON (27% Over- weight). Each data point equals Ron's weight at that time during each phase. The change in pounds and percen- tage body weight is presented at the bottom of each phase (e.g., a loss of 11 pounds and a reduction of 5.1% body weight during the first treatment). 91 :33: 3 $3385: e283 A333 3 BZMZHHm voui owemso uanoz a so ...on ..- IIIIIO mmH com it as 1... SN Ir SN ONN SGNflOd 92 NoN o.H ~+ m.o H+ m.o HI H.m HHI GHN see one o.~ «I o.o o n.m m.mI m.s m.NI eon mafia m.mon N.N m.mI o.o o 0.0 HI e.m oI sen one «NH o.o HI a.~ m+ m.q N+ o.m mI men sum: ems o.o o e.o HI m.N N+ o.o HI man «man Nee m.o n+ m.o NI N.m ca+ 0.0 H+ can mane mNH e.o H+ N.N «I N.N mI H.N «I can emu eon N.H NI N.N NI N.H mI o.m NI omN Heme men o.q N+ q.a m.~+ m.~ «I e.o m.~HI can cane mom «.0 HI a.~ wI N.m mH+ N.m «HI GNN awe mmm II II II II II II N.H «I meN some Nan H.N «I H.o H+ H.~ «I m.N mI man ce< uumHoee No can N man N man N men N man newemz ucmnno -nm newnmz NI: NIB HIz NIH mannammm Hanan ommem ueoHHo eumm you ommem comm weHunv owemeo uanoB wwMueoonm vem venom c oHan 93 Table 7 Total Pound and Percentage Weight Change for Each Client Final Baseline Weight Change Percentage of Body Client Weight in Pounds Weight Change Ann 199 -12 -6.0 Beth 243 -4 -l.7 Ken 276 -8 -2.9 Fran 190 -7 -3.7 Gail 180 -14 -7.8 Jan 190 -12 -6.3 Kris 176 +11 +6.3 Lisa 158 0 0.0 Mary 168 +6 ' +3.6 Pat 176 -10.5 -6.0 Tina 166 -16 -9.6 Ron 216 -9 -4.2 Individual Client Hypotheses Hypotheses 1 through 5 predict a meaningful amount of weight loss for each client at different phases of the study and over the entire study. Each hypothesis was considered according to the number of clients who supported it. Hypothesis 1. Each individual will show a 20-pound or 10% weight reduction from the end of the baseline phase to the end of the second withdrawal phase. This hypothesis was accepted for 0 of 11 clients. Hypothesis 2. Each individual will show a five-pound or 2 1/2% 94 weight reduction during the first treatment phase. This hypothesis was accepted for 8 of 12 clients. Hypothesis 3. Each individual will show a five-pound or 2 1/2% weight reduction during the first withdrawal phase. This hypothesis was accepted for 2 of 11 clients. Hypothesis 4. Each individual will show a five-pound or 2 1/2% weight reduction during the second treatment phase. This hypothesis was accepted for l of 11 clients. Hypothesis 5. Each individual will show a five-pound or 2 1/2% weight reduction during the second withdrawal phase. This hypothesis was accepted for 0 of 11 clients. A summary of Hypothesis 1 through 5 for each client is presented in Table 8. 95 Table 8 Analysis of Hypotheses 1 through 5 for Each Client Client HI H2 H3 H: H: Ann no yes no no no Beth -- no -- -- -- Ken no yes no yes no Fran no yes no no no Gail no yes no no no Jan no no yes no no Kris no no no no no Lisa no no no no no Mary no yes no no no Pat no yes no no no Tina no yes yes no no Ron no yes no no no 3The client had to lose 20 pounds or 10% of body weight in order to be accepted. bThe client had to lose five pounds or 2 1/2% of body weight in order to be accepted. Group Data and Hypotheses The mean weight change, variance, and standard deviation during each phase for 10 clients are presented in Table 9. Beth, who termi- nated after the first treatment phase, and Kris, who had complicating health problems, are not included in the totals. 96 Table 9 Pound and Percent Weight Change for All Clients during Each Phase Phase T1 W1 T2 W2 Total Mean Pounds —7.30 +0.05 -0.55 -0.45 -8.25 . Percent -3.7 -0.2 -0.2 -0.3 -4.3 Variance Pounds 16.54 41.80 12.80 10.58 44.40 Percent 3.0 9.4 2.8 3.4 14.8 Standard Deviation Pounds 4.06 6.46 3.58 3.25 6.66 Percent 1.7 3.1 1.7 1.9 3.9 Note. Two clients are not included in these results. Beth ter— minated at the end of the first treatment phase, and Kris had complicating health problems. Analysis of variance (ANOVA) with repeated measures was performed on the mean pound weight change in order to test the following hy- potheses with an alpha level of .05. The multivariate test was sig- nificant, F (3,7) = 5.71,_p <.03. Hypothesis 6 . There will be no difference in the mean weight loss for all clients during the first treatment phase and the mean weight loss for all clients during the first withdrawal phase (T1 - W1 8 0). This hypothesis was rejected,_§ (l, 9) = 7.22, p <.02. The mean weight loss during the first treatment phase was significantly greater than during the first withdrawal phase. Hypothesis 7. There will be no difference in the mean weight loss for all clients during the second treatment phase and the mean weight loss for all clients during the second withdrawal phase 97 (T - W - 0). 2 This hypothesis was not rejected, F (l, 9) I .01, p_<.95. ggypothesis 8. There will be no difference in the mean weight loss for all clients during the first treatment and the first with- drawal phases combined, as opposed to the second treatment and the second withdrawal phases combined [(T1 + W1) - (T2 +‘W2)] I 0. This hypothesis was not rejected, F (1, 9) - 4.19, p <307. Posttreatment Questionnaire At the end of the second withdrawal phase, all of the clients except two (Ann and Beth) completed a questionnaire assessing their reactions to the program (see Appendix F for questionnaire). The pur- pose of the questionnaire was to assess changes in the clients' moti- vation and their understanding and reactions to various aspects of the program. The clients were also interviewed by the researcher in order to get their reaction to the treatment package and the procedures that were followed. They were asked the following questions: (a) What did you like and dislike about the program? (b) How did you respond to working alone as Opposed to working with a counselor? (c) What effects did the withdrawal phases have on you? (d) Were there any benefits or satisfactions from being on your own? (e) Has your approach to losing weight changed as a result of this program? If yes, how has it changed? (f) If you gained weight during the baseline phase, what was the reason? Did the anticipation of the program have any effect? and (3) Would you prefer working individually with a counselor or in a small group? These data were used to help under- stand the clients' behavior during this study and to evaluate the 98 treatment package. In order to get a better understanding of how the clients' motiva- tion specifically varied during the study, they were asked to rate their level of motivation during each phase. Data were collected concerning the phase of the study in which each client felt most likely to succeed and most likely to fail, as well as the phase in which the highest and lowest motivation was reported by the client. These data are presented in Table 10. The clients were also asked to explain what caused the change in their motivation during the different phases of the study. Table 10 Number of Responses for Each Phase in Which Clients Thought They Would Succeed and Fail; Phase in Which Clients' Highest and Lowest Motivation Occurred; and Mean Motivation Rating for Each Phase Expected Expected Highest a Lowest a Motzszzion Phase Success 'Failure Motivation Motivation Rating2_ Baseline 2 2 5 3 3.8 Treatment 1 3 0 3 0 4.2 Withdrawal l l 4 0 4 2.9 Treatment 2 2 2 l l 3.4 Withdrawal 2 1 l 0 1 3.2 Note. Kris is not included in the tally because of her complicating health problems. 8If two or more phases tied for highest or lowest ranking, the first phase in which the ranking occurred was tallied. bThe rating is based on a 5-point scale, with 1 indicating very low and 5 indicating very high. The number of times the clients rated their motivation in.a par- ticular phase as being lower than their motivation in the previous 99 phase is presented in Table 11. For example, two of nine clients rated their motivation as lower during the first treatment phase than during the baseline phase, and eight of nine clients rated their moti- vation as lower during the first withdrawal phase than during the first treatment phase. Table 11 Number of Times the Clients' Mbtivation Was Rated Lower in a Phase of the Study Than It Had Been Rated in the Previous Phase Treatment 1 Withdrawal 1 Treatment 2 Withdrawal 2 2 8 4 3 Note. Maximum score = 9. Kris is not included in the tally because of her complicating health problems. The clients' mean self-rating of how well they implemented five concepts of the treatment package is presented in Table 12. The five treatment concepts were: (a) the diet plan, (b) increased exercise, (c) self-statements, (d) exercising choice in their eating behavior, and (d) self-instruction. A rating scale of 1 to 5 was used, with 1 indicating the concept was not used and 5 indicating the concept was used extensively. The clients' rating is matched with their total weight change during the program. Summary Capsule summaries of each client were presented, along with graphs which indicated each client's weekly weight change throughout the study. The individual client hypotheses were considered according to the number of clients who lost a meaningful amount of weight, as 100 Table 12 Clients' Mean Self-Rating of How Well Five Treatment Concepts Were Implementeda Mean Total Weight Implementation Implementation Rating Client Change Self-Rating for Each Thirdb Tina -16 3.8 Gail -14 3.6 3.87 Jan -12 4.2 Pat -10.5 3.4 Ron -9 2.4 2.87 Ken -8 2.8 Fran -7 3.2 Lisa 0 2.4 2.80 Mary +6 2.8 Note. Maximum mean self-rating score = 5. Kris is not included in the scores because of her complicating health problems. 3These ratings are matched with the clients' total weight change. The nine clients were divided into thirds according to their ranked distribution of total weight change. as well naire. as the analysis of the group data and posttreatment question- The trends that appear in the data, relevant client variables, and the implications of these findings for the treatment of obesity and self-management procedures for weight control are discussed in Chapter 4. CHAPTER 4 DISCUSSION AND IMPLICATIONS Summary The purpose of this study was two-fold: (a) to evaluate the short-term effectiveness of a combined model of self—management for weight control which stressed cognitive concepts, and (b) to identify potentially relevant client variables that may be related to the success or failure of the clients to implement the treatment package. Little research has been reported on either the short- or the long-term effec- tiveness of a combined self-management package for weight control. It is assumed that self-management treatment packages may be more potent if they combine the relevant aspects of the operant and cognitive models. If the purpose of self-management procedures is to teach clients how to control their own behavior, one would expect that clients, following formal treatment, could continue to lose weight until their goal is reached and be able to maintain their new weight. Most studies indicate that clients will lose weight during formal treatment using self-management procedures, but they fail to continue losing weight after treatment and, in many cases, will regain the lost weight. Con- sequently, this study was designed to determine how effectively clients can lose weight on their own following contact with a counselor. The 101 102 self-management procedures were taught in two five-week treatment phases, with a five-week withdrawal of treatment following each phase in which the clients had no contact with the counselors. During the two five-week treatment phases the clients met individually with their coun- selor once per week. A total of 10 treatment sessions were held. It is assumed that, in order for clients to be successful in the long term, they first must be successful in the short term. A minimum weight loss of five pounds or 2 1/2% body weight during each treatment and withdrawal phase was considered meaningful. Research on relevant client variables that relate to the success or failure in self-management weight control programs is severely lack? ing. Clients vary considerably in their responsiveness to self-manage- ment weight control programs, which suggests that client variables are important and deserve attention. This study, through an intensive case design, was undertaken to identify some client variables that may con- tribute to the success or failure of this self-management weight control program. Discussion The results of this study did not support the efficacy of the self- management treatment package to produce meaningful weight loss in the absence of a counselor. Only two of 11 clients lost five or more pounds (or 2 1/2% body weight) during the first withdrawal phase, and none of them lost five or more pounds (or 2 1/2% body weight) during the second withdrawal phase. If the purpose of self-management procedures is to teach clients how to control their own behavior, weight loss during the withdrawal phases would be expected. These results are consistent with 103 Jeffrey's (in press) observations that no self-management study has demonstrated continued weight loss following the treatment phase. However, some encouraging results were noted. Eight of 10 clients lost weight over the total program, and five clients lost 10 or more pounds. During the first withdrawal phase, two clients (Jan and Tina) were successful in losing a meaningful amount of weight, and two clients (Ann and Fran) missed losing a meaningful amount by one pound. Five of 11 clients lost three or more pounds during the first withdrawal phase. Several clients did well at the beginning of the first withdrawal phase and then lost their momentum. Several reasons might account for this change in behavior. The five weeks of the first withdrawal phase coin- cided with the last three weeks of classes during the winter term, a week of final examinations, and a week of quarter break. Since students typically experience more pressure during the last few weeks of a term with papers, projects, and final examinations, and since quarter break is associated with vacation, the clients may have found it more diffi- cult under these conditions to maintain the weight control program on their own. This may be especially true after only five treatment ses- sions. Also, contact with a counselor during treatment may be a motiva- ting factor to the clients. As the withdrawal phase progressed, the contact with the counselor became more remote and the clients' motiva- tion or enthusiasm may have waned. In support of these comments, several clients indicated during a posttreatment interview that it was motivating for them to meet with a counselor and to be accountable to someone. During the second withdrawal phase, no clients lost a meaningful amount of weight, but three clients (Ann, Pat, and Tina) lost three or 104 more pounds on their own. Ken and Jan did well for the first week of the second withdrawal phase and then started to slowly gain weight until the last week in which they both showed a weight loss. Six of 11 clients lost three or more pounds during at least one of the with- drawal phases. When the first withdrawal phase is compared to the second withdrawal phase for the group, little difference exists in mean weight loss, but a larger difference was found in the variance. A greater variation appeared among the clients in the first withdrawal phase than in the second. This finding may indicate that, after more treatment sessions, weight change becomes more stable. Several possible reasons for the ineffectiveness of the self-man- agement procedures in previous studies were reported in Chapter 1. One important cause seemed to be the decrease in the client's motivation to change. Evidence of this factor may be found in observing the dif- ference in weight loss between the first treatment phase and the second treatment phase. The average difference in weight loss for the 10 sub- jects was 6.75 pounds. Meaningful weight loss occurred during the first treatment phase (H = —7.3 pounds), and very little weight loss occurred during the second treatment phase (K = -0.55 pounds). Most of the total weight loss (i = -8.25 pounds) occurred during the first treatment phase, with eight of 12 clients losing five or more pounds. One possible factor that may have contributed to this difference in weight loss between the first and second treatment phases is the decrease in the clients' motivation. The data collected from the post- treatment questionnaire suggest that motivation was highest during the first treatment phase (see Table 10), which may reflect the clients' enthusiasm for starting a new program and their hope that they would be 105 successful. Three of nine clients rated the first treatment phase as the time of highest motivation and the time they thought they would succeed. No clients rated this phase as the time of lowest motivation, and no clients thought that they would fail. During the second treat- ment phase, one client rated that phase as the time of highest motiva- tion and one client as the time of lowest motivation. Also, two clients (Pat and Tina) thought that they would succeed, and two clients (Lisa and Ron) thought that they would fail during the second treatment phase. Interestingly, neither Pat nor Tina lost any weight during the .second treatment phase, but they had experienced some success during the previous phases. Pat attributed her increase in motivation to the fact that she was changing other problems which clouded her chances for success. Tina stated that the increase in motivation was directly related to her acceptance of the responsibility for the weight loss. Both Pat and Tina lost weight during the next withdrawal phase. Both Lisa and Ron had experienced a weight gain during the latter part of the first withdrawal phase, which may have influenced their motivation during the second treatment phase. Even though this information was collected at the end of the study and may be unreliable, it suggests that motivation may fluctuate as a function of previous perceived suc- cesses or failures. But one can also conclude that the decrease in motivation that was reported by the clients may be a function of time. Five of nine clients rated the baseline phase as a time of highest motivation, and three clients rated the first treatment phase as the time of highest motivation. These two phases also have the highest mean motivation rating. As the clients' initial enthusuunndecreased, their motivation may also have decreased. 106 Another reason for the ineffectiveness of the self-management pro- cedures is that the formal treatment may be too short or may not be complete. In this study, the experimental procedures may have had an influence on the clients' behavior during the two treatment phases. The first treatment phase may not have been long enough for the clients to incorporate the new behaviors they were learning, and/or the first withdrawal phase may have been too long for the clients to maintain their new behavior. A lack of success during the first withdrawal phase may have reinforced the clients' cognitive set to fail and, con- sequently, reduced their motivation. Only one client (Jan) lost more weight during the first withdrawal phase than during the first treat- ment phase. All of the other clients either lost considerably less weight or gained weight during the first withdrawal phase. The clients may not have been ready to try on their own for such a long period of time. Six of the 10 clients stated that the first withdrawal phase was too long, and three stated that they would have liked the first treat- ment phase to be longer. The mean motivation rating (see Table 10) was the lowest during the first withdrawal phase, and four clients thought that they were most likely to fail during this phase. Four clients rated the first withdrawal phase as the time of lowest motivation. Also, eight of nine clients rated their motivation as lower during the first withdrawal phase than during the first treatment phase (see Table 11). The lack of success during the first withdrawal may have made it harder for the clients to get involved again in the second treatment phase. For those clients who did lose weight during the second treatment phase, weight loss did not occur until after the second treatment session. 107 If only the first treatment and first withdrawal phase are con- sidered, the group results are very consistent with most research in this area (Hall et al., 1974; Jeffrey, 1974; Murray et al., 1975). Clients will lose weight during treatment and will maintain or slowly regain it after treatment. An average weight loss of 7.3 pounds occurred during the first treatment phase, and an almost complete stabilization occurred during the first withdrawal phase. Therefore, the first withdrawal phase, because of its length and the clients' per? ceived lack of success, may have undermined the benefits established during the first treatment phase and had an adverse effect on the second treatment phase. Support for the thought that the first withdrawal phase may have had an adverse effect on the rest of the program is found in results from two weight reduction groups that were conducted by the researcher. Direct comparisons cannot be made between these groups and the 12 clients in the present study because samples were drawn from different popula- tions. Nonetheless, the results are worthy of note. The two groups consisted of four clients each, with an average percent overweight of 52.4% and an average age of 37 years. These clients received the same treatment package but had 14 treatment sessions spaced over a 20bweek period. The longest withdrawal period was two weeks. Six clients finished the study with an average weight loss of 17.5 pounds. One possible reason for their success was that the short withdrawal periods did not adversely affect their motivation. The ineffectiveness of self-management procedures may be a result of these techniques not being potent enough to produce long-term change in the clients. Likewise, the difference in the two treatment phases 108 may be attributed to the potency of the material in each phase. The material in the first treatment phase may be more effective than the material in the second treatment phase. But from the posttreatment interview, eight clients indicated a strong preference for the material presented during the second treatment phase and some stated that it should be introduced earlier in the program. This preference may be a result of the clients gaining more insight into their own behavior even though they were less successful in losing weight. Hall et al. (1974) raised the issue that the demand characteristics of the treatment phase may compel clients to implement the techniques during treatment, but with the termination of treatment, the demand to perform the techniques is removed. Therefore, clients lose weight during treatment but do not during withdrawal. The large difference in weight loss between the two treatment phases of this study-does not support the notion of demand characteristics being a primary factor in the clients' behavior. It is difficult to assume that the demands of participation in the program were greater during the first treatment phase than during the second treatment phase. Hall et al. (1974) also reported that a group using a simple self-management approach was as effective in losing weight during treatment as a group using more com- plex self-management techniques. Therefore, other factors must be con- sidered in order to account for the difference in the two treatment phases and the ineffectiveness of self-management procedures during withdrawal. The effectiveness of the self-management procedures may depend less on the specific treatment techniques and more on client variables. The wide variability of client response to self-management weight 109 control programs which is found in this study and in other studies sug— gests that client variables may be highly significant factors to con- sider. Eight of the 12 clients in this study described themselves as being overweight since childhood. Stunkard and Mahoney (1976) state that people who develOp obesity in childhood differ from people who become obese as adults. Juvenile-onset obesity tends to be more severe, more resistant to treatment, and more likely to be associated with emotional problems. A person who became overweight as a child and did not reduce his/her weight during adolescence stands little chance of becoming a normal-weight adult. Stundard and Mahoney (1976) esti- mate this would probably occur only once in 28 times. Therefore, the majority of clients in this study do not have a good prognosis for becoming normal weight. Self-management programs may not work for all clients. A major task would be to determine which clients are most likely to succeed in a self-management program and which are most likely to fail. One fac- tor to consider may be the age at which overweight becomes a problem. Adult-onset obesity may respond to one set of techniques, whereas juvenile—onset may respond to a second set of techniques. Also, emo- tional stability should be considered. People with emotional problems underlying and/or associated with their weight problems may have to be treated in a different manner than people who are relatively free of emotional problems. Motivation appears to be one of the major factors relating to suc- cess. As long as clients are motivated to change, they are successful. Ron and Fran are two excellent examples of how clients are successful when motivated, but as soon as the motivation decreases, their success 110 at weight loss disappears. The major question is what changes one's level of motivation and how to keep motivation high over a long period of time. It would seem that success at weight loss would strengthen the person's motivation, but apparently motivation to lose weight is closely interrelated.with many other factors that are operating in the person's life. For example, Ron and Fran were very successful during the first eight weeks of the program, losing 16 and 15 1/2 pounds, respectively. But at that point, their motivation changed and they began to gain weight. At present, the factors that influence motiva- tion are poorly understood. More effort should be focused on identi- fying the factors that make up motivation and the factors that change it. Several other interesting findings appear in the data. Even though the results in Table 12 are based upon self-report and were collected at the end of the study, they suggest that those clients who implement the treatment concepts are most likely to succeed, whereas those who do not are less likely to succeed. The three most successful clients (Tina, Gail, and Jan) had a mean implementation rating of 3.87, and the three least successful clients (Fran, Lisa, and Mary) had a mean implementation rating of 2.80. If these data are reliable, they suggest that the treatment techniques may be useful if they are imple- mented. This finding is consistent with the observations of Hall et al. (1974). The consistent, long-term implementation of treatment concepts and techniques is a major issue with self-management procedures. Also, within the concepts that were rated (question 3 on the posttreatment questionnaire, Appendix F), the three least successful clients gave a mean rating of 1 for the motivational self-statements, which meant 111 that they did not use them. The three most successful clients gave a mean rating of 4 for the motivational self-statements, which meant that they used this concept rather extensively (maximum score = 5). Since this one concept is the major contributor to the overall difference in implementation rating, motivational self-statements may be a major factor that contributed to their success. Increasing exercise is another area with which most clients had difficulty. Some theories of obesity suggest that physical inactivity is a major cause (Stunkard & Mahoney, 1976). Little research is reported in the weight control literature that considers why people have difficulty in increasing their exercise. One major factor may be embarrassment. Most overweight people have more difficulty with sports, and several clients stated they wished to lose weight in order to be better at sports. This is an area that deserves further attention in weight control programs. Seven of 12 clients gained five or more pounds during the six-week baseline phase. This phase coincided with the Christmas holidays and the first week of the winter term. During the posttreatment interview these clients were asked if the knowledge that they were going to start a weight control program contributed to their weight gain. Only one client (Tina) indicated that it did. All of the other clients stated that it was normal for them to gain weight over the holidays. Without a control group it is impossible to determine if this weight gain during the baseline phase had any effect on the subsequent treatment. Two clients terminated the program before the end of the second treatment phase. Beth terminated at the end of the first treatment phase, and Ann dropped out following the second session of the second 112 treatment phase. An attrition rate of 17% is normal for weight control programs. It is the impression of the researcher that the attrition rate might have been higher if the clients had met in groups. During the posttreatment interview, all but two clients strongly favored meeting individually with a counselor, as opposed to meeting in a small group of four or five. The other two saw advantages in both individual counseling sessions and group methods. The individual counseling sessions allowed clients to gain some benefit in related areas, as well as weight control. Often, other problem areas are associated directly or indirectly with the clients' weight problems. This was especially true of Lisa, Mary, and Pat. Limitations The nature of self-management studies places limitations on the conclusions that can be drawn from them. Since the treatment techniques are implemented by the clients in their own environment, the validity of the self-report data can be questioned. When the clients state that they maintained their diet throughout the week, the only confirmation is their weight loss, which is very indirect. Also, the validity of the self-report data in the posttreatment questionnaire could be ques- tioned. Was the level of motivation during the various phases of the study the same as the level reported at the end of the study? The population from which the clients were obtained is limiting. The clients in this study were all college-age individuals attending a large state university. The clients either responded to an advertise- ment or were referred by the consulting physician. Consequently, the clients in this study may differ in certain respects from peOple in the 113 general population who are overweight. The results and conclusions of this study can be generalized to the extent that these clients resemble other people who are overweight. The intensive case study design allows generalization of treatment effects through replication across subjects. The extent to which the treatment package was uniformly implemented across clients by the counselors can be questioned. Differences in clients and differences in the abilities of the counselors may create variation in the imple- mentation of the treatment package. This variation in implementation may limit the conclusions concerning the treatment package. The proposed client variables were derived from the observations of the researcher and represent his opinions concerning the factors influencing the clients' success or failure in this study. Since little research has been performed in this area, these client variables are intended to be viewed as observations that might be helpful in planning future research efforts. Proposed Client Variables From observing the 12 clients over a 20-week period in their attempt to implement the treatment package and lose weight, it is the Opinion of the researcher that several variables were important in the success or failure of these clients. The client variables can be classified according to the following categories: (a) motivation, (b) emotional stability, (c) reasons for wanting to lose weight, (d) cognitive set to fail, (e) understanding and implementing the treat- ment concepts, and (f) personal organization and record-keeping. These six categories are somewhat interwoven. 114 Motivation. It is of no surprise that the motivation of clients is a key variable in weight loss. Motivation to change is considered very important in any type of self—management program. The important aspect of motivation is the specific factors that contribute to it and maintain it over a long period of time. A major factor influencing motivation to change is the payoffs the clients receive for both being overweight and for losing weight. It could be assumed that people receive positive consequences for being overweight, as well as negative consequences. The positive consequences could be small and immediate, such as the taste of good food, or they could be large and important, such as serving major personality functions. Ken had a large part of his identity associated with being overweight, and Lisa used her weight problem to influence her father. Also, there are both perceived positive and negative consequences for losing weight. If the perceived positive effects of losing weight are not much greater than the positive consequences of being overweight, then peoples' motivation to change will be low. Likewise, if the nega- tive effects of being overweight are not much greater than the negative consequences of losing weight, motivation will be low. Therefore, it is necessary to find out what the real payoffs are for the client in order to assess the level of motivation. If peOple are getting a large payoff for being overweight, then this problem would have to be resolved before the person could be successful in losing weight. Keeping motivation to change high over a long period of time is another problem area. Decreases in motivation occurred in several clients. One factor that might contribute to maintaining motivation is the view that weight loss is a selfish activity. The primary reward 115 for weight loss should be intrinsic. If the primary reward for weight loss is extrinsic, such as pleasing one's spouse, getting more dates, buying new clothes, or being better at sports, the amount of motiva- tion to change will depend upon the power of the outside agents. For example, if clients want to lose weight for a specific occasion such as graduation or job interviews (extrinsic), the motivation to change will greatly decrease as soon as the special occasion passes. Also, if the special occasion is not immediate or the goal will not be reached, the motivation to change will decrease. With many extrinsic rewards, clients fear that if the weight is lost, the extrinsic reward will not be more available (e.g., more dates). If the primary reward for weight loss is intrinsic, such as better health, then there is a better chance of sustaining the motivation to change over a longer period of time and a greater chance to achieve permanent weight reduction. Environmental factors will have a smaller effect on one's motivation. Another factor related to keeping motivation to change high over a long period of time is that the client must find satisfaction from performing the activities and achieving the goals that lead to weight loss, as well as satisfaction from the weight loss itself. If people only gain satisfaction from the weight loss itself, then motivation will greatly decrease when a plateau is reached. The motivation of many clients is strongly influenced by what the scales read every week. The slow rate of weight loss and its delayed positive conse- quences can greatly affect motivation through discouragement if the primary reward is weight loss itself. Characteristics of individuals with high need-achievement motiva- tion are that they set challenging personal goals that are moderately 116 high and receive satisfaction when those goals are attained. It could be that the level of need-achievement motivation may be a factor in the success of self-management weight control programs. Those clients with high need-achievement motivation may be able to persist at the weight loss activities longer than a person with low need-achievement motivation. Emotional stability. If people are experiencing many other prob- lems at the time of a weight loss program, their attempts to lose weight will probably be disrupted by these problems. Two areas associated with emotional stability appear to be important to weight loss. First, clients must care about themselves. If clients do not care about themselves, as was the case with Mary, then they will do little to attain their goal of losing weight. Often people who do not care about themselves will express their dislike through overeating. This idea is closely associated with viewing weight loss as a selfish activity. If a person does not like him/herself, then he/she will do little to improve. Second, clients must be able to tolerate frustration in order to lose weight. Frustration can come from several sources. In order to lose weight, people must experience some hunger. If people view the state of hunger as very unpleasant and cannot tolerate this form of frustration, then their chances of permanent weight loss are probably low. Frustration can come from feelings of being deprived. People in the client's environment may eat a lot of food while he/she can only have a small amount, and this is a punishing experience. Social situa- tions can be a source of frustration. Friends may exert pressure to 117 drink or eat certain high-calorie foods. Also, frustration can arise from attempting to learn new ways to handle problem feelings like . depression and boredom. Some peeple have learned to handle these feelings by eating, and the process of learning new ways can be dif- ficult. Reasons for wanting to lose weight. Often, the stated reasons for wanting to lose weight are very general and vague. These reasons may include better health, better appearance, and more self-confidence. Sometimes clients have accepted these reasons because they are very common and socially acceptable. When asked to elaborate upon the value they personally assign to better health, appearance, or self-confidence, they have difficulty. The reasons for wanting to lose weight need to be very concrete, immediate, and emotionally meaningful to the client. Without such reasons, it is difficult for the client to counteract the powerful, immediate reinforcing effect of food. Some people feel that they should lose weight, but it is something they really do not want to do. Because of this "should," people will feel guilty if they are not trying to lose weight. Actual weight loss is not important, but the act of trying to lose weight is important. Consequently, there are numerous people who are always on a diet but never lose any weight. Beth and Pat seem to have this "should" Operating in them. In order to be successful, clients must fully understand their reasons for wanting to lose weight and then decide if those reasons are strong enough for them to perform the necessary activities that will lead to weight loss. With vague, unclear reasons, it is 118 difficult to change a strong behavior like eating habits. Cognitive set to fail. Most people who have tried many times to lose weight and who have not been successful will enter a weight con- trol program with a cognitive set toward failure. Most people are looking for a guaranteed, easy method to lose weight and will try any method that comes along. Many people have the attitude that they will try a program or diet and hope that it works, even though they have their doubts. Permanent weight control requires much effort and hard work over an extended period of time. When the weight control program becomes hard work, the attitude of failure is an excuse for not trying and giving up. It becomes a self-fulfilling prophecy. This attitude was displayed by several clients in this program. It must be confronted if the clients are going to be successful on their own over an extended period of time. Understanding and implementing the treatment concepts. If healthy people implement the concepts of the treatment package, they will lose weight. The problem becomes one of implementation over an extended period of time. Understanding and acceptance of two concepts seem crucial for the long-term success of the client. First, clients must accept the full responsibility for controlling their own eating behavior. If clients place the control for their eating behavior in some agent outside of themselves (e.g., spouse or counselor), their success will probably be short-term. This was clearly evident in the case of Fran. Second, clients must understand and accept that they have a choice of when, what, and how much they eat. If they do not accept this 119 choice, then they will always have an excuse for not performing the necessary activities. If clients do accept this choice, then they can allow many factors to enter their decision-making process and will be more likely to make choices that will lead to weight loss. Also, by accepting this choice over their eating behavior, guilt feelings for breaking the diet can be reduced. Many people have guilt feelings when they break their diet, and these guilt feelings often lead to greater frustration and low self-concept, which may make it even harder to con- tinue a weight control program. But if pe0p1e consciously choose to go off the diet and are willing to accept the consequences, then they tend not feel guilty about their behavior. People can intellectually accept the idea of choice in their eating behavior very quickly, but it takes more time for them to learn how to exercise that choice on a behavioral level. Jan and Tina were starting to exercise that choice in their behavior. Personal organization and record-keeping. In order to implement a weight control program, pe0p1e must be able to plan ahead and organize their time. In order to keep deprivation states manageable, people should eat three meals a day which include the proper food. The person who cannot plan ahead, schedule meals, and have the proper food avail- able will have more difficulty in handling these deprivation states. If people do not organize their time, they will have difficulty main- taining a regular exercise program. If clients' lives are very dis- organized, it may be helpful to spend time developing some organization. Accurate record—keeping is very important. Often, as soon as the record-keeping decreases, so does the weight loss. Without accurate 120 records, people may think they are maintaining the diet when, in reality, they have slowly increased their calorie intake. Implications Treatment of obesity. Weight problems have multiple origins, and within the individual they may be highly complex in their development and maintenance (M. J. Mahoney & K. Mahoney, 1976b; Stunkard & Mahoney, 1976). Unfortunately, all of the causes of obesity are poorly understood. Human weight problems range from those having simple, easy-to-manage causes to those with highly complex, difficult-to-manage causes. Most people's weight problems probably result from a combina- tion of problems that are unique to the individual. The causes of obesity might be placed on a continuum ranging from simple reasons, such as poor diet, to complex reasons, such as over- weight serving important personality functions for the person. Within the extremes of the continuum could be such factors as lack of exercise, lack of nutritional knowledge, poor eating habits, craving for sweets, boredom, nervous tension, and biological factors. Any individual could have a combination of reasons, and it could be assumed that the more factors involved and the more complex those factors, the more difficult it will be to treat the problem. Likewise, the methods of treatment should be dictated by the nature of the causes. Therefore, it seems very unlikely that one treatment package will be useful for everyone. This hypothesis may account for the wide variability in client response which is typically found in weight control programs. Treatment packages should be highly flexible so that, after determining the cause of the overweight problem, they can be individualized. Some people may need 121 extensive diet information; others may need a structured exercise pro- gram; still others may need help with problem feelings and alternative ways to reinforce themselves. Some pe0p1e may respond best in an indi- vidual setting, others in a group setting, and some in a combination of both. How does one determine the cause of people's weight problems in order to individualize the treatment program? If there is no biologi- cal or genetic component to an individual's weight problem, then it can be assumed that the person's overweight is an acquired condition and that the person is receiving some payoff (reward) for his/her behavior. The overweight person is consuming more calories than he/she is expend- ing. Therefore, since the individual's body does not need the extra food, the person must be receiving something positive for his/her eating behavior. The first step in a treatment program might be a thorough assessment of what the payoffs are for an individual. This may take considerable time and may be an ongoing consideration through- out a program, since people typically are not aware of their payoffs until they start to give them up. For example, one client mentioned that she did not realize how she had used food to cover up her feelings of rejection until she stapped her between-meal snacking. A lot of feelings surfaced which she did not realize were there. Obviously, the payoff was that food and eating helped cover up her negative feelings. A thorough assessment of peoples' reasons for wanting to lose weight is essential. Many people have very vague reasons that do not have much real meaning to them. The person's reasons for wanting to lose weight will have to be stronger than the payoffs received for being overweight in order for him/her to be successful. After a 122 complete analysis of a person's reasons for wanting to lose weight, he/ she may decide that it is not worth attempting to lose weight at this time. A thorough assessment should also be made of people's attitudes toward weight control in general, dieting, exercise, basic food groups, the feeling of hunger, their own self-control, past failures, expecta- tion of success, their self-concept, caring about themselves, self- improvement, and the role of food in their lives. Also, peoples' know- ledge of various areas such as nutrition, the role of exercise in weight control, factors that influence eating behavior, and the handling of problem feelings should be assessed. Any myths that people maintain about weight control and dieting should be located and discussed. In- formation in areas such as those listed above may make it easier to personalize treatment programs and may increase the likelihood of success. One reason for lack of success in permanent weight control may be that treatment programs ask people to give up their immediate payoffs for overeating in return for the promise of vague, future benefits. Instead, people need to handle the payoffs they are getting from over- eating in a different, less self-destructive way. If the payoff is the immediate pleasure of the taste of the food, then the person must be helped to find other immediate pleasures that can be a substitute for food. If the payoff for eating is the temporary removal of problem feelings, then the client must be helped to find alternative ways of handling those feelings. If the payoff is an avoidance of close inter- personal relationships or a fear of failure, the the client must be helped in those areas. Asking the client to give up the payoffs with 123 nothing in return except some vague, future benefit will probably result in efforts which last for only a short period of time. Permanent weight control programs should deal with the alteration of benefits so that the weight problem can be handled on a continuing basis, rather than imposing temporary demands. Unfortunately, most at present do not. In order to increase the likelihood of long-term success, it seems that certain attitude changes must also occur. The most important attitude change may be the confidence that one can control his/her eat— ing behavior and the acceptance of the idea of choice in his/her eating behavior. Also, peoples' attitude toward the role that food plays in their lives often needs to be changed. For example, the role that food and drink play in social situations often needs to be changed. In the area of attitude change, the cognitive model of self-management can exert a strong influence. A major part of any attitude is the person's belief system, which involves self-statements. More work should be done to determine the role of attitudes on peoples' eating behavior and on their ability to lose weight. Most people have an attitude that a special diet is the factor that causes them to lose weight. Instead, their attitude should be redirected toward the idea of permanently changing eating habits in order to effect a loss in weight. Treatment packggg. In the posttreatment questionnaire and inter- view, every client, even the unsuccessful ones, had very favorable comments about the treatment package. The most typical comments were that it was a very practical and realistic approach and it allowed them to gain a better understanding of their own behavior. Almost all of the clients stated that they liked the concepts in the second 124 treatment phase and found them most beneficial. The second treatment phase emphasized the cognitive aspects, and these may be the most beneficial over the long-term. The researcher thinks that the cognitive components of the treatment package have the potential of being the most helpful to clients. This position coincides with M. J. Mahoney and K. Mahoney's (1976b) observations from their recent studies. But it seems that much more time should be given to these concepts if they are going to have their maximum benefit. The clients who were not very successful did not blame the treatment methods, but rather accepted the responsibility themselves. How much of the favorable comments were a result of the clients' investing 20 weeks of their time in the program and the demand characteristics of the interview is impossible to assess. For most clients, 10 treatment sessions are not enough, eSpecially when major payoffs are being obtained for their overweight. Ken was a good example of a client needing many more treatment sessions. If he had had more sessions, he might have had a better chance of succeeding. It takes time for a client to fully understand his/her own self-defeat- ing behavior and how he/she implements it. For example, most clients immediately state that they have no fears associated with losing weight, but after a period of time and thought they realize that their fears have been suppressed. Both the concept of eating as a choice and the technique of self—instruction required more treatment time than was given in this study. It takes time for clients to learn how to exercise choice in their eating behavior and to talk positively to themselves. It is the opinion of the researcher that most clients would require a minimum of 20 treatment sessions, with some clients requiring more. In Stuart's (1967) original study, the number of sessions ranged from 16 125 to 41. Old eating habits have been present for years and learning new eating habits takes time and considerable effort. Also, the concepts in the second treatment phase should be introduced earlier so that they can be referred to throughout the treatment. More time should be spent on assessing the clients' motivation and reasons for wanting to lose weight. If these issues are clarified early in the treatment, possibly before the dieting begins, clients will be in a better position to assess what kind of commitment they are willing to make to the program. By comparing benefits of losing weight with the sacrifices made while engaging in weight loss activities, clients can make a more conscious commitment to what they want to do. Some means should be developed of assessing the clients' attitudes toward weight control, dieting, exercise, self-control, failure, self- reinforcement, basic food groups and frustration, as well as an assess— ing of the payoffs that clients receive for being overweight and for losing weight. Value clarification exercises might be useful for this purpose. A classification system for the cause of overweight should be developed. This classification system may help in personalizing the treatment methods in terms of techniques used and time required. ~ Because of the overwhelming preference by these clients for indi- vidual counseling as opposed to group counseling, an effort should be made to assess whether a particular client would do better in individual counseling or group counseling. Both individual and group sessions have advantages as well as disadvantages. Individual sessions offer a more flexible personalization of the treatment strategies and allow a more intensive effort with the client's problems. The group approach 126 may add incentives and group pressure and allow the clients to gain insights and ideas from other members of the group. A combination of individual and group methods is possible and perhaps desirable, especially since group methods are more economically feasible. Some parts of the treatment program may be more effective if introduced and learned on an individual basis, while other parts may be effectively acquired in group sessions. Most of the clients stated that they liked the idea of a with- drawal phase where they could attempt the weight loss strategy on their own and then return to see a counselor and discuss what happened. The withdrawal of treatment seems to be a good idea, but it might be more effective if the meetings with the counselor were faded out gradually. Possibly, failures could be caught early and averted through this process of gradual withdrawal. Also, before clients are sent on their own, a criterion of consistency should be attained. For example, clients could be required to maintain the diet for three consecutive weeks before being allowed to go a week on their own. In some cases it may be necessary to meet with a client more than once per week in order for the client to learn how to maintain a diet for a full week. This would have been appropriate for Ken. He maintained his diet on the days closest to meeting the counselor. On the days furthest from his meeting with the counselor, he would not maintain the diet. Fre- quent meetings early in the prognam,.with a gradual decrease in meetings, might be beneficial in such cases. More time should have been spent on developing alternative behaviors to eating in problem situations. For example, some people reward themselves primarily through eating and have learned very few 127 alternative ways of self-reward. One client stated that when she reached her goal in a previous program at Weight Watchers, she went home and ate everything in the house as a reward. In other situations when she did well, she would reward herself through eating, which is very self-defeating. Another client stated that when she felt depressed she would eat, and the food would make her feel better. Assertive training could be included in a treatment package. Many people have trouble saying "n0" when they are offered food in social situations. Clients should have the right to refuse someone's food without feeling guilty. Many people are "food pushers,’ and the over- weight person should learn how to deal with those situations. Suggestions for Future Research The results of research in the area of self-management weight con- trol programs suggest a great need for well-designed studies in this area. The programs themselves are far from being refined, let alone researched. Much of the preceding sections has indicated directions for future research. Several specific areas will be identified that seem to be particularly promising areas for future research. The greatest need is in the development of instruments that could measure client variables and assess the relevant attitudes. Also, instruments that will assess the payoffs clients are receiving from their overweight and the reasons for wanting to lose weight (motiva- tion) need to be developed. With reliable measuring instruments, a classification system could be develOped which might aid in the individualization of treatment methods. With reliable measuring instruments, it may be possible to isolate predictor variables which 128 would indicate which clients are more likely to profit from a self—man- agement weight control program and which clients are not. This might help in determining a "readiness" concept for pe0p1e who want to lose weight. Some people may be more ready to perform the necessary activities in order to lose weight than other people. A second research area involves the components of the treatment package and its implementation. Further research should be performed in order to isolate the most effective and the least effective compo- nents of the treatment package. Efficacy of the operant components and the cognitive components should be explored. Data in this study suggest that the cognitive components (motivational self-statements) may have been a crucial variable in the success of these clients. But this hypothesis should be empirically investigated. Also, some compo- nents of a treatment package may be more effective with some clients, and other components may be more effective with other clients. The cause of the overweight problem may relate to which components of a treatment package will be most effective for each client. Some clients may find some components of a treatment package easier to implement than other components. This possibility should be explored. Another area that needs further research is the role of the coun- selor. Hundreds of how-to-lose-weight books and articles appear every year in the mass media. The implication of these books and articles is that people can lose weight on their own and stay thin forever if they follow the few simple suggestions. Each article or book has its own newly discovered secret. However, results are generally disappointing. Another source of disappointing results is the many doctors who will recommend that their patients lose weight, give them a diet or some 129 form of medication, and then send them out on their own to accomplish this task. It is the researcher's opinion that, in most cases, clients cannot be successful unless they have a professional helper to assist them in learning habits and new ways to cope with problem areas. This is especially true for clients who have complex reasons for their overweight. Unless clients have an accurate understanding of the factors that influence their eating behavior and have mastered some techniques to change their eating habits, their chances of long-term success are probably minimal. However, this is a researchable ques- tion. Can clients who have access to the same information be as suc- cessful as those who have the information plus discussions with a counselor? Closely related to the role of a counselor in weight reduction efforts is the effectiveness of the treatment packages being implemented in a group situation or on an individual basis, or with a combination of both group and individual sessions. Many treatment packages are introduced in group settings, but the preference for individual counsel- ing from the clients in this study suggests that this is an area that deserves some attention. Groups are much more economical, but if individual treatment is more effective, the cost effectiveness of this treatment may be less expensive in the long term. A combination of group and individual sessions may be the most economical and effective. The idea of developing a criterion of consistency in performing weight loss activities is another area for research. Clients must meet that criterion before they are allowed to attempt weight loss on their own for a period of time. The number of counseling sessions would vary from client to client, and if clients could not meet the criterion 130 of consistency, then a reevaluation of their reasons for wanting to lose weight and the payoffs they were receiving from their overweight would be necessary. Clients who cannot maintain a diet and exercise program and maintain other activities that are related to weight loss may need other kinds of help before they enter a weight control program. The researchable question is whether a set number of treatment sessions before clients attempt to lose weight on their own is more effective than the clients' fulfilling a criterion of consistency, regardless of the number of treatment sessions. In Retrospect From experiences gained through conducting this study, the re- searcher has identified several changes that are recommended for subse- quent efforts. Each will be briefly described. 1. A more extensive pretreatment questionnaire could be used in order to collect more detailed background information concerning each client's weight problem. A questionnaire similar to the one used at Stanford University could be developed (Agras, Ferguson, Greaves, Qualls, Rand, Ruby, Stunkard, Taylor, Werne, & Wright, 1976). 2. The clients' reasons for wanting to lose weight could have been assessed more thoroughly. A simple profile form could be developed which would list the clients' reasons for wanting to lose weight, and each reason could be rated for its strength; This form could be matched with a similar form which listed the payoffs for being overweight and their strengths. 3. The number of treatment sessions could have been increased, with a decrease in the length of the first withdrawal phase. For 131 example, the first treatment phase could have been seven weeks and the first withdrawal phase three weeks. 4. The concept of eating as an act of choice, and the technique of self-instruction could be introduced earlier in the treatment pro- cess. These concepts seemed to be effective in allowing clients to gain some understanding concerning their eating behavior, but more time is needed for them to be implemented more completely. By expanding the first treatment phase, these concepts could easily be introduced then. 5. A method should be developed to make clients more accountable for reporting their activities. A checklist form was utilized, but several clients did not use it or were not willing to complete it every day. A few clients stated that the checklist form became a chore to fill in. 6. A simple instrument could be developed that would assess the clients' motivation at various times during the day. This may be a more reliable procedure than asking the clients at the end of the study to rate their motivation during the different phases. APPENDICES APPENDIX A Counselor Training Rating Forms APPENDIX A WEIGHT REDUCTION PROGRAM TRAINING SESSIONS (Ist Treatment Phase) Counselor Self-Rating Name: According to the scale below, rate how well you understand the following concepts. Do Not Extremely Understand Slight Moderate Very ‘Well 1 2 3 II S 6 7 8 9 10 1. The concept of self-management: 2. The conceptual model of eating behavior: 3. The purpose of the daily eating records: b. Using the diet plan: 5. Setting weekly and phase goals: 6. Situational strategies: 7. Purpose of the exercise program: 8. Developing motivational self-statements: 9. The conceptual model of self-verbalization: 10. Attaching self-statements to recurring cues: 11. Developing alternative responses to problem situations: 132 133 WEIGHT REDUCTION PROGRAM TRAINING SESSIONS (2nd Treatment Phase) Counselor Self-Rating Name: According to the scale below, rate how well you understand the following concepts. Do Not Extremely Understand Slight Moderate Very Well 1 2 3 II S 6 7 8 9 1O 1. The A-B-C model of behavior: 2. The role of self-statements in the A-B-C model: 3. Eating as an act of choice concept: h. The necessity of having the client accept the A-B-C model: 5. Factors that lead to SDB - Fear: Techniques: Disowning: Prices: 6. Factors that lead to GAS: 7. Purpose of the SDB worksheet: 8. The purpose of the self-instruction package: 9. The three components of the self-instruction package: 10. The purpose of rehearsal of the self-instruction package: APPENDIX B Baseline and First Treatment Phase Forms APPENDIX B WEIGHT REDUCTION PROGRAM Information Sheet Date: Name: Address: Phone Number: When is the best time to be reached? Age: Sex: Present weight: Height: Total pounds you want to lose? Which evenings are you free during the Winter Quarter? Do you have any present health problems? If yes, please explain: Are you presently involved in a weight control program? If yes, please explain: 134 135 Research Consent Form 1. I have freely consented to take part in a scientific study being conducted by Randy Gold under the supervision of Dr. Norman R. Stewart, Professor of Education. 2. The study has been explained to me and I understand the explana- tion that has been given and what my participation will involve. 3. I understand that I am free to discontinue my participation in the study at any time without penalty. 4. I understand that the results of the study will be treated in strict confidence and that I will remain anonymous. Within these restrictions, results of the study will be made available to me at my request. 5. I understand that my participation in the study does not guarantee any beneficial results to me. 6. I understand that, at my request, I can receive additional explana- tion of the study after my participation is completed. Signed: Date: Audiotape Release Form 1, , hereby agree to permit audiotape recordings of the counseling interviews in which I appear. I under- stand that the confidentiality of the material presented will be pro- tected. I likewise authorize the researcher to use such materials for instructional purposes with professional groups so long as they also agree to protecting the confidentiality of the material. The materials recorded will be stored and protected as confidential material by the researcher. The Specific method for maintaining confidentiality and storage is determined by the professional supervisor and the re- searcher. When the materials are no longer useful for research or 136 instructional purposes, or at my written request, they will be with- drawn from use, mechanically erased, or destroyed. Signed: Date: Witness: 137 ‘WEIGHT REDUCTION PROGRAM Medical Clearance has been examined and no gross endocrine problems or gross illnesses which may be complicated by dieting were found. The above named person may participate in the self-management weight reduction program conducted by Randy Gold. Er. Mary H:_Ryan Date: 138 WEIGHT REDUCTION PROGRAM Questionnaire Name: Age: Marital Status: 3. 7. 9. What is your present weight? Height? What is your goal weight? Total pounds you want to lose? List your reasons for wanting to lose weight. Be as specific as possible and include as many as possible. How long have you wanted to lose weight? Have you attempted to lose weight in the past? If yes, how many times? What have you tried (e.g., special diets, weight watchers, exercise, etc.)? Be specific. When was the last time you tried? Is there anyone in your social environment who would be willing to help you with your weight control program? If yes, who? How much money would you be willing to bet that you will achieve your goal? $10 $25 $50 $75 $100 How unpleasant is your weight problem to you? slight moderate very 1 2 3 II 5 How important is it that you lose weight? slight moderate very 1 2 3 LI 10. 11. 12. 13. 1h. 139 List as many positive consequences as possible that will result if you lose weight. List any negative consequences that will result if you lose weight. How much effort do you feel will be required in order to lose weight? slight moderate very 1 2 3 II What are your expectations that: a) you will be able to implement the weight reduction program? low medium high 1 2 3 II S b) you will be able to continue with the program after the treatment phase is over? low medium high 1 2 3 II 5 ‘Write a brief history of your weight problems. Please include such things as approximate length of time you have been overweight, the possible reasons for weight problems, and any other relevant information (use back of page or other paper). 140 WEIGHT REDUCTION PROGRAM Baseline Weight Record December 1bth to January 10th It is important to have an extended record of your weight in order to determine normal fluctuations over time. Please record your weight at approximately the same time on each of the following days: Name: Mon. Wed. Fri. Mon. Wed. Fri. Mon. Wed. Mon. Wed. Fri. Dec. Dec. Dec. Dec. Dec. Dec. Dec. Dec. Jan, Jan. Jan Jan. 15 17 fifiv 19 22 2h 26 29 31 \OQVIN 141 .Hmoe emomemH one new on voanvoA oEHv mo enmeoH one veooom emu-wee "—‘ -..‘5 .fl- meme