Diet and PA are the most important behaviors for energy balance, wherein they maintain and promote weight loss effects during the post–bariatric surgery period. This study evaluated the relationships and mediating effects between cognitive factors and diet or PA during the post–bariatric surgery period, explored the effects of diet and PA on subjective health outcomes, and attempted to identify the pathways hypothesized using CCAM through an empirical study. To the best of our knowledge, this is the first study to completely validate the hypotheses using CCAM. This model was originally proposed for the behavioral management of patients who have obesity or diabetes, with an intent to understand the relationship between diet and PA to help prevent or manage obesity and diabetes in a more effective manner[16]. All patients who have undergone bariatric surgery fall into the severely obese population, and a large proportion of them have diabetes[4]. Therefore, decoding the behaviors of such patients during the post–bariatric surgery period based on CCAM is of great clinical significance for postoperative lifestyle management. According to the results, the SEM showed an excellent fitting, and most of the path relationships between cognitive factors and behaviors in CCAM were confirmed to be effective. At the same time, the measurement of core cognitive factors using CCAM, namely, CB, TC, intention, and self-efficacy, was related to individual cognitions and attitudes toward diet and PA. This result suggested that it was feasible to explain the potential interaction between diet and PA through cognitive factors.
Our study also showed that the intention and self-efficacy of diet and PA can affect the corresponding behaviors, which was consistent with the conclusion of the studies based on the theory of behavioral change in the field of bariatric surgery. As for diet, Zhu et al.[33] explored the factors influencing dietary compliance in patients during the post-bariatric surgery period based on the Attitude-Social influence-Efficacy Model and reported that dietary intention, attitude, and self-efficacy were the main predictors of dietary compliance. Ren et al.[34] conducted a randomized controlled trial of exercise intervention in patients during the post-bariatric surgery period based on the Cross-theoretical Model. During the intervention, the patients’ exercise self-efficacy and intention were evaluated, and the authors accurately identified the stage of behavioral changes, with an intent to implement targeted intervention measures and consequently effectively improve patients’ exercise compliance.
As for the antecedent cognitions, both CB and TC mainly exerted their impact on intention and self-efficacy, but most of their influencing paths had no significant direct relationship with behaviors, and only CB had a negative effect on diet. Previous studies have reported the presence of behavioral compensation in diet[35] and several other behaviors (such as alcohol consumption, sedentary behavior, and smoking[21–22]). CB is a cognitive coping strategy developed by individuals during lack of self-management, wherein individuals use this belief to alleviate the guilt caused by practicing unhealthy behaviors through the execution of one compensatory healthy behavior[15]. Most patients undergoing bariatric surgery had unhealthy dietary behaviors such as overeating and grazing. The effects of weight loss caused by the surgery are usually extremely significant and remain only for a temporary period after the operation, and many patients are required to perform certain amounts of PA (such as walking or swimming) to increase energy expenditure, which may make them think that “the weight will keep dropping due to other factors, with occasional non-control of the diet”[36]; at this point, the CB level increases, leading to the deterioration of the dietary behavior. For this reason, the negative effect of CB on behaviors of patients during the post–bariatric surgery period was mainly reflected in the dietary pattern, and poor dietary habits may weaken the direct effect of CB on PA, which may be the reason why CB had only a negative effect on the intention of PA. TC originated from pedagogy research and has attracted increasing attention in the field of health behavior management in recent years[37]. A systematic review demonstrated that, when individuals are engaged in exercise at the best level, they would be more motivated to improve their eating behavior[38]. Based on this finding, Fleig explored the influence of TC on factors involved in the HAPA theory and found that TC would affect self-efficacy, and patients with a high level of TC reported frequent use of self-regulation strategies[39]. All patients in the aforementioned studies had one good behavior to activate the carry-over effect of TC, which improved the overall behavior mode. However, patients in the postoperative period usually had diet and PA behaviors below the average level, and behavior management requirements put forward by dietitians during follow-up will consume patients’ willpower, which may exert the positive effect of TC on behaviors only at self-efficacy, not directly on behaviors.
We also analyzed the mediating effects of antecedent cognitions on behaviors. Except for the relationship between CB and diet (H4a), intention and self-efficacy were confirmed to be complete mediators in the relationship between antecedent cognitions and behaviors in the other three hypotheses (H4b, H4c, and H4d). The complete mediating effects suggested that intention and self-efficacy were the key mediators that aided CB or TC to exert influence on behaviors, and interventions on individual CB or TC may affect behaviors only in the pre-contemplation or contemplation stage. Particularly, intention was a partial mediator of the relationship between CB and diet, which may be related to the poor dietary status of patients during the postoperative period, and CB was often activated when patients refused to correct their unhealthy dietary habits[35]. This result suggests that the dietary behavior during the post–bariatric surgery period can be improved by implementing intervention measures to inhibit CB. Additionally, because CB and TC can be indirectly associated with multiple behaviors simultaneously, interventions on these two components would benefit people who require multiple health behavioral changes, such as patients undergoing bariatric surgery. In the early stage after surgery, surgeons and nurses can educate patients about the universality and negative effects of “compensatory eating” by enumerating negative examples (e.g., patients often ate high-sugar snacks after exercise as compensation; the extra calorie intake may exceed the energy expenditure by PA, resulting in poor weight loss) to inhibit their CB[15]. TC can be achieved by adjusting the order of behavioral management. Performing exercise can promote individual initiative and confidence in dietary management, and increasing the intake of fruits and vegetables or reducing the intake of sweets can promote patients’ enthusiasm to exercise[40]. In the actual intervention, the carry-over effects of TC can be exerted as much as possible according to the preference of patients. For example, patients who tend to take a walk as exercise can be encouraged to consume vegetables and fruits in a timely manner after the activities, which may consolidate their enthusiasm for exercise.
As for the influence of behaviors on subjective health outcomes, diet exerted significant effects on perceived stress, well-being, and QOL, whereas PA exerted significant effects only on perceived stress. This may have been due to the immense changes in the gastrointestinal tract structure after bariatric surgery[41]. Patients who adhere to the norms of dietary guidelines not only achieve optimal weight loss within a short time but also demonstrate fewer gastrointestinal symptoms such as vomiting and acid reflux and gain a better weight loss experience[33]. The main purpose of PA after bariatric surgery was to increase energy expenditure and reduce muscle loss, which can improve patients’ body composition, but generally, it takes a long time for PA-induced weight loss to be effective, and most of the patients do not regularly engage in PA. The improvement in patients’ subjective feeling brought by dissipation of energy due to PA is limited[42]. This study focused on outcomes of psychological variables, and dietary behavior was also an important and main factor in obesity studies with physiological status as main outcomes. In the study conducted by Gils[43], for a short term after the surgery, regardless of whether the patients engaged in PA, those who adhered to the Mediterranean diet achieved more weight loss than those who adhered to PA alone. The factorial control trial of Brown[44] in obese patients with breast cancer demonstrated that, compared with patients with uncontrolled behaviors, those who adhered to a controlled diet alone or controlled diet and exercise simultaneously could achieve weight loss of a significant amount of fat, while exercise alone had limited effects on body weight.
To summarize the valid paths in CCAM, the relationship between cognitive factors and behaviors showed that conducting interventions with cognitive factors is easier and more effective through dietary modifications, and the promoting effects on the overall behavioral pattern activated by inhibiting CB possibly contributed more to PA when the status of dietary behavior had been improved; thus, the patients would have a better weight loss experience. The association between behaviors and subjective health outcomes also showed that diet exerted a more critical effect on the overall health than PA during the postoperative period. Overall, based on the analysis of CCAM and from the perspective of joint promotion of behaviors after bariatric surgery, dietary intervention had a higher priority than PA intervention. During intervention, in addition to the traditional facilitated on intention and self-efficacy, the management of antecedent cognitions including CB and TC may also play a positive role in the cooperative promotion of diet and PA, among which inhibiting the CB level may be more important.
As for behavioral interaction, previous studies have explored the relationship between diet and PA at the physiological level. Hazell et al.[45] reported that the release of cytokines, change in insulin concentration, and production of lactic acid due to exercise-induced muscle metabolism may mediate the changes in anorexia signal peptide tyrosine-tyrosine and glucagon-like peptide-1, while the change in appetite hormone will further inhibit ghrelin levels, resulting in the prolongation of eating latency and decrease in energy intake. A meta-analysis showed that gastric emptying was accelerated during low-intensity exercise, resulting in increased subsequent energy intake; when the exercise intensity exceeded 70% VO2 max, the rate of gastric emptying was delayed, which may be related to loss of appetite after exercise[46]. As for the effect of diet on PA, Castro et al’s[47] randomized controlled trial showed that obese patients who received a very low-calorie ketogenic diet had an increased amount of PA during the intervention. Diet composition also had an impact on subsequent PA time. Bray et al’s[48] study showed that when the extra energy intake was in the form of fats, the individual 24-hour energy expenditure (24EE) did not increase, whereas excessive dietary protein intake strongly stimulated 24EE and increased energy expenditure during sleep. Recently many studies have explained the mechanism of the effects of and mediating factors for PA on diet from the perspective of physiology, but the mechanism of the effects of diet on PA remains unclear. Our study explored the behavioral interaction at the psychological level, suggesting that CB mediated the effect of diet on PA, which may help explain the balance between energy intake and subsequent energy expenditure. Moreover, patients who underwent bariatric surgery had decreased gastric compliance, sensitive changes in intragastric pressure, and accelerated gastric emptying during the postoperative period[41]; thus, the interaction between diet and PA discussed in the aforementioned study may have been affected by surgery. Thus, explanation from the psychological perspective in this study will also enable the understanding.
This study has several limitations. Primarily, this was a cross-sectional study, and establishment of the sequence of cognitive factors, behaviors, and outcomes was mainly derived using CCAM. Additionally, the causal hypotheses in intermediary analysis were also based on inference; future longitudinal research should demonstrate specific causal relationships in a detailed manner. Second, for the behavioral factors in patients undergoing bariatric surgery, we measured only overall dietary compliance and different PA durations after the surgery, but we did not further distinguish dietary composition and types of PA. Third, CB and TC were the core variables of CCAM, and this study and previous research mainly focused on their influence on behaviors. However, the measurement of CB and TC was based on the patient’s personal attitude toward behaviors. Hence, it is necessary to further explore the influence of behaviors on CB and TC in future studies, to gain a better understanding of the behavioral pattern during the postoperative period, thereby implementing behavioral interventions in a better manner.