Fostering physical activity-related health competence after bariatric surgery with a multimodal exercise programme: A randomised controlled trial

Bariatric surgery is an effective obesity treatment achieving substantial body weight reduction and improvement of obesity-associated comorbidities (Mingrone et al., 2015; Peterli et al., 2018; Schauer et al., 2017). However, there is increasing evidence that over the longer term, many patients experience weight regain (Sjöström, 2013). Due to the well-established positive effects of physical activity (PA) on weight management, cardiovascular risk, bone health, and wellbeing, PA support should be an integral part of follow-up care packages after bariatric surgery (Josbeno et al., 2010). However, implementation of PA support in clinical practice is still limited (James et al., 2022). Many patients do not increase their PA level after surgery and remain overly sedentary (Barbosa et al., 2019; Herring et al., 2016). Thus, there is a need for theory-based interventions, that clinicians can use to help patients to change their PA behaviour.

The physical activity-related health competence (PAHCO) model offers a theoretical base for developing PA interventions (Sudeck & Pfeifer, 2016). It describes a specific set of competences required to initiate and maintain a healthy, active lifestyle. The PAHCO model can be conceptually located at the interface between health literacy (Sørensen et al., 2012) and physical literacy (Edwards et al., 2017; Giblin et al., 2014). The PAHCO model seems particularly appropriate for the current study for at least two reasons: Firstly, because of its PA-specifity. It includes PA-typical constructs, which are often not represented adequately in more general models of health behaviour (e.g., the Health Action Process Approach; Schwarzer, 2012). Secondly, because of its interdisciplinarity. Cognitive, motivational-emotional, and physical aspects are considered to explain not only PA, but also health and wellbeing.

Two of the PAHCO competences, control competence and self-regulation competence, are of particular importance in this study (see Fig. 1). Control competence describes an individual’s ability to gear their own activity to achieve positive effects in health and wellbeing. Two facets are differentiated here: (a) control competence for physical training allows people to model physical load in a health enhancing way, both by applying fitness knowledge (e.g., to choose longer but less intense bouts of exercise for continuous training) and by perceiving body signals (e.g., to use heart rate to control physical load in endurance training; Sudeck & Pfeifer, 2016); (b) PA-specific affect regulation helps people to plan and carry out activities in order to optimise benefits in wellbeing (e.g., to choose jogging in the forest to relax after a stressful day at work; Sudeck & Pfeifer, 2016). Self-regulation competence refers to the motivation and volition that enables individuals to be regularly active. Here too, two facets are distinguished: (a) Motivational competence that allows people to choose a specific activity type that matches their preferences (e.g., to join a community-based hiking group because social contact is important for you; Schorno et al., 2021); (b) PA-specific self-control, which helps people to translate intentions into action by using behavioral change techniques, such as action planning (e.g., to plan when, where, how, and with whom you will be physically active in the future) and coping planning (e.g., to identify potential barriers in daily life and strategies to overcome them; see also Dishman & Ickes, 1981; Gollwitzer & Oettingen, 2016).

Various studies showed that the facets of PAHCO are not only related to the actual and habitual amount of PA, but also linked with health and wellbeing. This empirical evidence was found both in healthy adults (Carl, Grüne, et al., 2020; Schorno et al., 2021; Sudeck et al., 2018; Sudeck & Pfeifer, 2016) and in individuals with musculoskeletal, metabolic, cardiologic, oncologic, and chronic obstructive pulmonary diseases (Carl et al., 2021; Schmid et al., 2020; Sudeck & Pfeifer, 2016). However, the model has not yet been addressed in bariatric surgery patients even though it seems reasonable that its basic assumptions are also valid for this population.

To promote PAHCO, interventions should include not only a practical component (e.g., participating in PA) but also a theoretical one (e.g., teaching of knowledge about fitness). Furthermore, it is required to transfer these two components not in isolation but to combine them in a programme (Carl et al., 2020b; Haible et al., 2019). The instructional principle “reflection in or on action” is an effective method for combining practical and theoretical components (Baartman & Bruijn, 2011; Schön, 1983) but so far has received relatively little attention in the field of health promotion. One exception is the study by Volk et al. (2021), in which an intervention programme was designed and evaluated that particularly aimed to promote control competence for physical training in the context of physical education. Students performed specific activities (e.g., running with different intensities whilst monitoring their heart rate, creating a scatter plot diagram of heart rates during the activities), reflected, and discussed the experiences made together with other students and the teacher (e.g., group discussion about factors influencing heart rate; for more details see, Haible et al., 2019). Positive short-term effects of the described intervention on fitness knowledge and control competence for physical training were found (Volk et al., 2021). The instructional principle reflection on action was also applied in a study by Schorno et al. (2022), however they concentrated specificly on promoting the two facets of self-regulation competence in less active adults. Schorno et al. (2022) developed an exercise counselling approach that aims to help people find and regularly engage in a PA that suits them. Individuals gained diverse exercise experiences in trial sessions on-site (e.g., fitness activities vs. games vs. dance activities) and reflected on these experiences immediately after (e.g., discussion about which activity one liked and why). Afterwards, suitable exercise activities were deduced, and the counsellor assisted in developing a concrete action and coping plan. These two behaviour change techniques have already been found to be effective in adults with obesity (Göhner et al., 2012) and bariatric surgery patients (Bond et al., 2017). Positive short- and medium-term effects of the exercise counselling on motivational competence, PA-specific self-control, and self-reported exercise and sport volume were found (Schorno et al., 2022).

The aim of the present study was to examine the efficacy of a 3-month multimodal exercise programme focused on the promotion of PAHCO in a sample of bariatric surgery patients. The programme stands out by combining experiences with guided reflections (Schön, 1983; for more details see left box of Fig. 1), meaning that individuals do not only train, but they get more involved by discussing their exercise experiences. This study investigated (1) the impact of the programme on control competence for physical training, PA-specific affect regulation, motivational competence, and PA-specific self-control (primary outcomes; see Fig. 1). In accordance with previous findings (Göhner et al., 2012; Schorno et al., 2022; Volk et al., 2021), albeit partly in different populations, we expected positive treatment effects on all four facets of PAHCO mentioned above. Furthermore, this study examined (2) the impact on PA behaviour and subjective vitality as an indicator for wellbeing (secondary outcomes; see Fig. 1). Again, we expected positive treatment effects.

In this work, we designed and evaluated a novel 3-month multimodal exercise programme focussing on specific facets of PAHCO in patients who recently underwent bariatric surgery. We found that the multimodal exercise programme was successful in enhancing control competence for physical training and PA-specific self-control and positively influenced self-reported exercise and subjective vitality. Conversely, we found no statistically significant positive effects on PA-specific affect regulation, motivational competence, and device-based PA.

The observed positive impact on participants’ control competence for physical training can be considered as large and in line with a previous intervention study (Volk et al., 2021). Of note, the effects were not only short-term, but persisted until the follow-up assessment three months after completion of the programme. This means that the intervention empowered participants to apply training-specific knowledge and effectively select and structure activities for promoting physical health in daily life. The present research revealed larger treatment effects than in a recent study published by Volk et al. (2021). This could be due to differences in the study populations such as age (adolescents vs. adults) and/or biopsychosocial health status (healthy individuals vs. bariatric surgery individuals). It is possible that adults with complex biopsychosocial needs, such as the population under investigation, particularly benefit from such a programme as they appreciate its personal utility and relevance and therefore are more willing to improve their PAHCO than healthy adolescents in compulsory education. Overall, the observed success in our study population suggests that the PAHCO model with the intervention sessions developed by Haible et al. (2019) are well applicable to complex clinical settings and patients groups such as those after bariatric surgery.

A possible explanation for the lack of a positive treatment effect on the PA-specific affect regulation may be that it was not sufficiently addressed by the exercise programme (see Table 1 in the supplementary material). It is important to note that promoting affect regulation, or in other words, the competence to steer PA in order to optimise psychological wellbeing, remains an under-researched area. Future research should focus on which personal, environmental, and activity-related factors influence intra-individual differences in affective response to PA (e.g., valence, arousal; see Bourke et al., 2020) thereby setting the ground for more effective intervention.

Although the present study revealed medium effects of the exercise programme on motivational competence, they remained non-significant, probably influenced by our rather small sample size. However, compared to a previous study with less active, healthy adults (Schorno et al., 2022), this research revealed slightly smaller treatment effects. One can speculate that this is related to differences in characteristics in the study population or the delivery mode of the individual exercise counselling. In the study by Schorno et al. (2022) the trial exercise sessions and the reflection about the experiences made took place on the same day. In contrast, the trial exercise sessions and the reflection in the present study were spread over several days for logistic reasons. It is therefore possible, that lack of interweaving between practical experience and reflection in terms of time and space diminished the impact on individual learning (Schön, 1983).

With regard to PA-specific self-control, this study showed a positive, medium treatment effect at the end of the exercise programme. This means that participants felt more competent in implementing regular PA in daily life. This result is in line with prior studies with less active but healthy adults (Schorno et al., 2022), individuals with obesity (Göhner et al., 2012), and bariatric surgery patients (Bond et al., 2016), which also revealed medium to high treatment effects on related volitional outcomes. However, the present findings at the follow-up assessment three months after completion of the programme looked somewhat different. Although the participants of the exercise programme maintained their increased level of PA-specific self-control on average at t₃, the participants of the CG were also able to improve their competence between t₂ and t₃, resulting in a not significant treatment effect at follow-up. Consequently, a question warranting further research is, whether the exercise programme needs to be prolonged or supplemented with a booster session after its completion to maximize its benefits in the long run.

A secondary outcome of this trial was self-reported exercise behaviour. The exercise programme resulted in increased self-reported exercise for members of the IC compared to those of the CG. In absolute terms, members of the IG were able to raise their exercise level by 28 min/week. This is an increase that has shown to be associated with positive health effects (e.g., a reduction mortality; see Wen et al., 2011). However, this large effect on the self-reported exercise volume found in our study need to be interpreted with caution, as no treatment effect was found on device-based moderate to vigorous PA. Our results are consistent with the few existing RCT, which also found no effects of exercise programmes on accelerometer-based PA in bariatric surgery patients (Bellicha et al., 2018). The fact that self-reports differ from device-based assessed PA has been observed and discussed in other studies with adults who underwent bariatric surgery (Herring et al., 2017). Discrepancies may be explained by several reasons: Firstly, the two measures have different scopes. Whereas self-reported data reflect structured exercise activities performed during leisure time, accelerometer-based data reflect moderate to vigorous PA, including structured and unstructured PA, not only performed during leisure time, but also at work or for transportation (Burchartz et al., 2020; Fuchs et al., 2015; Strath et al., 2013). Secondly, a potential underestimation of acelerometer-based PA in the IG. Despite technical advances in recent years, accelerometers are not (yet) able to measure all physical activities. For instance, cycling, a session on a cross trainer, or strength training can not be adequately captured with a device (Burchartz et al., 2020). A closer look at the data shows that the above activities were frequently reported specifically by patients of the IG at t₂ and t₃. Thirdly, an overestimation of self-reported exercise in the IG. Due to social desirability, patients of the IG might have reported more or longer exercise activities after they participated in the programme (Nigg et al., 2020).

A large effect was found for the multimodal exercise programme on subjective vitality at follow-up, meaning that members of the IG felt more energised and vital compared to the CG. This result is in line with findings of previous observational studies (Carl, Grüne, et al., 2020; Schmid et al., 2020). The positive finding on subjective vitality supports the idea that exercise programme should also focus on qualitative aspects of PA and not simply quantitative aspects and teaching the formula “the more, the better” (Sudeck & Pfeifer, 2016).

The strengths of the present RCT include its rigorous design and the novel application of the PAHCO model in the bariatric surgery population. However, we acknowledge some limitations. Firstly, participants included mostly had a rather high PA level at baseline (see Table 4 in supplementary material), which may limit the generalisability of the present findings to all bariatric surgery patients. Secondly, we did not examine the effect of the exercise programme on movement competence, which is also integrated in the PAHCO model alongside control competence and self-regulation competence (Sudeck & Pfeifer, 2016). Movement competence allows people to participate in diverse leisure activities and to master motor challenges of daily life (e.g., motor abilities and motor skills for riding a bike or gymnastics) and is therefore an important competence worth to consider in further research. Thirdly, our study design does not allow a concrete statement about which programme element has which effect on PAHCO, PA behavior, and subjective vitality, as we evaluated the exercise programme as a whole. Finally, the sample size of the study can be considered to be limited, resulting in low power to detect small effects.

Findings from this RCT overall revealed that a multimodal exercise programme was effective in improving single facets of PAHCO, self-reported exercise, and wellbeing among bariatric surgery patients. This study therefore provides a foundation for future research for the use of the PAHCO model (Sudeck & Pfeifer, 2016) to optimise long-term post bariatric surgery outcomes. In particular, larger scale RCTs and longer follow-up periods to determine maintenance of outcomes from such a novel programme are needed. Additionally, optimal timing, frequency, and feasibility of implementation in routine care need to be defined.