Application of an Integrated Model to Predict Behavior in the Bariatric Surgery Population

Introduction

Obesity is considered a complex chronic disease with multiple causative factors, including social, economic, genetic, behavioral, physiologic, psychological, and metabolic components. ¹ Obesity and overweight conditions have reached epidemic levels in America with over 60% of adults reaching overweight or obese standards.^2,3 The American Obesity Association and the Centers for Disease Control and Prevention define overweight as having a Body Mass Index (BMI) ≥25 kg/m², obese as a BMI ≥30 kg/m², and extremely obese as a BMI ≥40 kg/m².^1,4 The number of U.S. adults who are classified as being obese or extremely obese has steadily increased since 1976; the percentage of obese adults rose from 15% in 1976 to 34.3% in 2008, while the percentage of extremely obese adults rose from 1.4% in 1976 to 6% in 2008. ³ This increase is regardless of gender, education level, or region of the country. ¹ The implications of the rise in obesity levels are staggering.

Obesity has been linked to increases in the incidence of 30 serious medical diseases, increases in mortality from all diseases, and impaired mobility, and is now considered by some authorities as the number one preventable cause of morbidity and mortality.^1,4,5 Nearly 80% of obese persons have diabetes, coronary artery disease, hypercholesterolemia, hypertension, gallbladder disease, or osteoarthritis, and 40% of these obese adults have two or more of these conditions. ⁶ In the year 2000, the total direct and indirect costs contributed to obesity were $117 billion. ⁷ The yearly number of deaths directly related to obesity and its comorbid conditions were almost equal to those connected to cigarette smoking in 2000; the latest statistics may show that poor diet and physical inactivity have overtaken tobacco as the leading cause of mortality in the United States.^8,9 Obese persons also suffer emotional consequences, such as stigmatization and discrimination, on a daily basis in their employment and academic settings.^5,10,11

Weight loss surgery (WLS) provides an option for long-term weight control to those morbidly obese patients for whom other diet and exercise programs have failed. ¹² WLS has been performed since 1954, although it has become more common in the past decade. ⁷ WLS is indicated for those individuals whose BMI is ≥40 or a BMI ≥35 in the presence of at least one comorbid condition. ¹² In 2008, approximately 220,000 persons underwent WLS, more than a tenfold increase in the number of surgeries performed a decade earlier, ¹³ yet a mere 1% of the 15 million clinically eligible morbidly obese persons in the United States. ¹⁴

Studies have shown that persons who undergo WLS can expect to lose 47–80% of excess body weight.^12,15,16 WLS surgery has been shown to decrease, or even eliminate, many of the comorbid conditions associated with obesity, such as diabetes, sleep apnea, hypercholesterolemia, and hypertension.^14,15 The procedure can decrease the risk of developing coronary artery disease by up to one half ¹⁴ and reduce the prevalence of cancer by up to 80% within 5 years post-operatively. ¹⁵ However, up to 20% of patients cannot maintain their dramatic weight loss after 2 years post surgery. ¹⁷ The reasons for this weight regain are numerous and individualized, ranging from the physiologic (stretching of the gastric pouch, resuming one's pre-surgical diet, and grazing behaviors) to the behavioral and sociocultural (not exercising, not following prescribed food restrictions, skipping post-surgical appointments, and expecting the surgery to do all the work).^7,18–20 Therefore, any framework that can help explain behaviors and predict adherence to the post-operative regimen is crucial to incorporate into programs for WLS clients. Current studies using theories or models have shown only little to moderate success in predicting post-surgical adherence; an integrated model of behavior prediction that seeks to expose all of the many factors influencing behavior may have more success.

Previous Use of Models in WLS Clients

The transtheoretical model of change (TM) has been used in provider practice, such as by family practice physicians and nurse practitioners. The TM can be used to determine if the client is in a stage of change, and the provider can incorporate interventions based on this stage of change aimed at weight management, including referrals for WLS; ²¹ however, this model has not been used to predict adherence post-operatively in the WLS population.

Protection motivation theory (PMT) is a widely researched model that can explain how cognitive processes impact behavior. ¹³ PMT was used by Boeka, Prentice-Dunn, and Lokken ¹³ to develop a pre-operative intervention focusing on the importance of adhering to the post-operative eating guidelines and how adherence could best be accomplished. The control group received the usual education from the bariatric surgeon. While the entire PMT intervention did not prove to be statistically significant, secondary analyses concluded that two components of the PMT intervention, perceived self-efficacy and perceived threat of not following the guidelines, did predict the patients' intentions of complying with the regimen. ¹³

The Health Belief Model (HBM) was developed to help predict adherence to preventative health recommendations. ²² The HBM reasons that when a person's perceived value of an outcome is added to the expectation that a behavior will lead to that outcome, then the behavior will be performed and change occurs. ²² The authors used HBM in a small (n = 40) study of women and concluded that costs of the surgery were well worth the perceived benefits; however, these results may not be generalizable due to the small sample size and homogeneity of the sample.

Social cognitive theory, and its construct of self-efficacy, has been used in the bariatric surgery population. McAllen, in a doctoral dissertation, found that self-efficacy was the only variable that was found to be associated with positive maintenance of weight loss. ²³ Another study compared the eating self-efficacy of WLS patients to obese nonsurgical patients and concluded that the weight loss associated with WLS was associated with higher levels of eating self-efficacy than it was for those patients who had not had surgery. ²⁴ No other variable besides eating habits related to WLS has been studied using social cognitive theory.

The theory of reasoned action (TRA) and the theory of planned behavior (TPB) were compared in a study of WLS patients to determine which framework best predicted exercise adherence after surgery. ¹⁶ These theories are the two most commonly tested theories used in predicting exercise behaviors. ¹⁶ The TPB was determined to be better at predicting exercise in the WLS sample, and perceived behavioral control was the best single predictor of both exercise intent and behavior, while subjective norm and attitudes toward exercise were more predictive of behavior than hypothesized. ¹⁶ This study states the usefulness of the TPB in this population but only looks at exercise adherence, a single factor in the maintenance of weight loss post-operatively.

Conceptual Framework

An integrated behavioral model, derived from the TRA, the TPB, and other theories can be used to determine salient behaviors for this population. ²⁵ According to this integrated model, the most important factor that determines behavior is the intention to perform the behavior. Motivation is key; individuals are unlikely to perform a behavior if there is no motivation to do so. ²⁵ Other factors that play a role in the performance of a behavior are the needed knowledge and skills to perform the behavior, the “salience”^25,26 of the behavior, environmental constraints that may affect the behavior, and habit. ²⁵

It stands to reason that if a person must learn skills or devise a way around environmental constraints, then the intent to perform the behavior must be strong. There are three factors that affect the intention to perform the behavior: attitude (consisting of experiential attitude and instrumental attitude), perceived norm (consisting of injunctive norm and descriptive norm), and personal agency (consisting of perceived control and self-efficacy). ²⁵ Attitude toward performing a behavior is simply whether the person feels performing the behavior is good or bad, and is for or against performance of the behavior.^25,27,28 Experiential attitude is affective; it is the individual's emotional response to the idea of performing the behavior, while instrumental attitude is cognitive in nature, based on beliefs regarding the outcome of the behavior. ²⁵

Attitude, both experiential and instrumental, can be viewed in the light of emotional eating, body image disturbance, and depression. If a person engages in emotional eating, such as binge eating, stress eating, or using food for comfort or mood control, the risk of post-operative grazing and snacking is higher, based on the attitude that the surgery is preventing future weight gain. ²⁹ Body image disturbance can occur when massive amounts of weight are lost quickly and hanging skin is left, when one's self-perception still feels obese despite weight loss, and if significant others reject the “new” person after the surgery; some patients react negatively and attempt to gain weight in order to fill out the hanging skin or appease the significant other who is not happy with the results. ²⁹ This negative attitude can lead to significant post-operative weight gain. While self-image usually improves after surgery, some patients experience increasing levels of depression post-operatively when the surgery doesn't fix all of life's problems. ²⁹ Depression can lead to a negative attitude toward following the post-surgical regimen and lead to increased weight gain.

In two qualitative phenomenologic studies by LePage ³⁰ and by Sutton, Raines, and Murphy, ³¹ themes of finding balance and paradox, or change, were identified, which could be considered evidence of experiential and instrumental attitude. The physical changes occurred quite rapidly, leaving the emotional side of the person feeling as if he or she was still a fat person.^30,31 While health improved, along with socialization and mobility and energy levels, some sacrifices had to be made also; one participant lost his family, his sobriety, and his job because he never dealt with the reasons he overate and became obese. ³⁰ The void still needed to be filled, and food was no longer available to fill the emptiness within. Some participants turned to shopping, bulimia, and drug addiction to ease their pain and give them a sense of comfort. ³⁰ These studies provide concrete examples of how important attitude can be post-operatively. Post-operative psychological support, in the form of support groups or individual sessions with the bariatric psychologist, is also important in figuring out what caused patients to overeat, why they overeat, and how they can substitute healthy, adaptive behaviors for eating whenever they have those eating triggers. Some bariatric practices do provide this intense post-operative support; hopefully, in the future, this practice will become the norm.

Second, perceived norm is the pressure an individual feels socially to perform or not perform the behavior.^25,32 The injunctive norm can be described as those beliefs about what others think an individual should do and the motivation to comply with the others' wishes. ²⁵ Also, the descriptive norm can be a strong influence on intention; descriptive norms can be thought of as the individual's perceptions about what others are doing. ²⁵ Obese persons have and still experience a great deal of discrimination and stigma. It is important when designing interventions intended to help post-surgical WLS patients adhere to dietary and exercise regimens that this societal pressure is taken into account. Some WLS patients lie about their procedures and travel out of state to have the procedure performed to avoid further stigma and embarrassment; this indicates that social support could be lacking for many pre-operative patients. ³³ Assessing how important the perceived norm is for these persons could have a large impact on the design of the intervention aimed at increasing performance of adhering to the post-operative regimen. Some patients experience conflict within their spousal relationships as the patient starts receiving attention from others and the physical body changes. ³¹ Some patients attempt to regain the weight, based on their injunctive norm beliefs that their significant others prefer them heavy. ²⁹

Last, personal agency is a construct adopted from Bandura ³⁴ that has been proposed to be a significant factor in influencing intention. ²⁵ Personal agency in this model is composed of perceived control and self-efficacy. Perceived control is the amount of control an individual perceives himself to have over performing the behavior, based on how easy or hard the individual thinks the behavior is, while self-efficacy is one's confidence in the ability to perform the behavior regardless of barriers or difficulties. ²⁵ Many obese individuals have tried to lose weight before and failed; therefore, it is reasonable to assume that their feelings of self-efficacy to stick to the plan post-operatively may be low. In order to fully capture how strong a factor personal agency is in adhering to the regimen, it must be assessed. Research has shown that post-operative patients have increased self-efficacy regarding eating habits when a large amount of weight is lost ²⁴ and that levels of self-efficacy are associated with maintenance of weight loss in this population. ²³ While these constructs have been shown as important in adhering to the surgical recommendations, they alone do not fully predict adherence.

Discussion

The three determinants of behavioral intention described in this article may have differing values for different behaviors and for different populations; ²⁵ one behavior may be predicted best by attitude, while another behavior may be poorly predicted by attitude and better predicted by personal agency. One population may consider the perceived norm to have the greatest influence on intention to perform the behavior, while another population deems attitude more important. Therefore, in constructing interventions to influence intentions to perform behaviors post-operatively, first, each intention must be assessed for the degree of influence by attitude, perceived norm, and personal agency. ²⁵

Due to this difference in strengths of the constructs based on the behavior or population, the investigator has to go out into the population being studied to identify the correct, or salient, beliefs based on that behavior relevant to that population.^25,27,28,35 Interviews with the population must be conducted to elicit information about the attitudes, beliefs, control, and efficacy feelings regarding the behaviors; ²⁵ this is called an elicitation study. Once the salient beliefs have been established for that population, then an instrument can be constructed using these measures, and analyses of the data can determine which beliefs can best predict the performance of the behavior. ²⁵ While WLS patients have been interviewed for research studies in the past, elicitation studies have the explicit purpose of determining what is important to them post-operatively, specifically related to attitude, perceived norm, and personal agency.

This framework can be used to examine the perceived normative beliefs, the injunctive norm and the descriptive norm, or others' behaviors and expectations, and how they influence the intention to adhere to the post-WLS prescribed dietary and exercise regimen. Using this framework can shed some light on what behaviors are important after WLS. Just as the reasons behind weight gain are as individual as the persons who experience the weight gain, so are the attitudes, beliefs, and feelings about adhering to the dietary and exercise regimen after WLS. Using this integrated behavioral model gives meaning to all the various factors that influence behavior, which can lead to individualized interventions aimed at maintaining maximum weight loss after bariatric surgery.