No differences in post-surgical physical activity levels in RYGB vs control group

Swedish researchers have reported that there were no differences in physical activity levels in women two years after Roux-en-Y gastric bypass compared to a control group. The researchers believe this is the first dissonance-based intervention targeting women undergoing RYGB surgery. The findings were reported in the paper, ‘Physical activity in women attending a dissonance-based intervention after Roux-en-Y Gastric Bypass: A 2-year follow-up of a randomized controlled trial’, published in PLOSOne.

The paper’s authors explained that dissonance-based interventions are built on cognitive dissonance theory and aim to change a person’s behaviour by altering their cognitions. However, dissonance-based interventions have never previously been conducted in patients undergoing RYGB-surgery.

They established the Wellbeing after Gastric Bypass (WELL-GBP), a dissonance-based intervention for female RYGB patients conducted at five hospitals in Sweden. The aim was to examine if there were any differences in objectively-measured physical activity levels, between the two groups. The researchers hypothesised that such an intervention could also influence various wellbeing-related outcomes after RYGB surgery.

In total, 259 participants provided were randomised to the intervention group (60%, n=156) consisting of four group sessions three months post-surgery (during which a modified version of the Stice dissonance-based intervention model is used) or control group (40%, n=103) who received usual follow-up care.


The mean pre-surgery BMI was 40.8 (SD 4.5) and the mean age 44.7 years (SD 10.3). Pre-surgery, 54% met the recommended physical activity guidelines of ≥150 minutes of moderate-to-vigorous physical activity (MVPA) per week. At the two-year follow-up, 167 of the 259 participants had valid accelerometer measurements (intervention group n=99 and control group n=68) - a loss to follow-up rate of 35.5% (36.5% in intervention group and 34.0% in control group).

Of the participants in the intervention group, 61.6% (n=61) had attended the session where physical activity had been discussed. The only pre-RYGB characteristic that were significantly different between those included in the two-year follow-up (n=167) compared to those not included (n=92) were age at surgery (p=0.0097). There were no significant differences in pre-RYGB characteristics between intervention group and control group.

Seventy-three percent of included participants in the intervention group (n=99) had attended at least one group session (any of the four sessions), and 51% had received the intervention according to protocol (i.e. attended ≥3 group sessions). There were no statistical differences in pre-RYGB characteristics between those receiving the intervention compared to both control group and participants in the intervention group who did not receive the intervention.

At two years, there were no statistically significant differences in any of the physical activity levels between intervention and control group. At two-years, the intervention group spent 29.0 min/day (SE=1.8) in MVPA and the control group 27.1 min/day (SE=2.5). Participants were sedentary for 493.3 min/day (SE=12.0) and 458.8 min/day (SE=12.3) in the intervention group and control group, respectively. Percentage of women who met the recommended physical activity guidelines (non-bouts) were 66.7% in the intervention group and 54.4% in the control group. No difference between groups were statistically significant.

Those who received the intervention were more sedentary (522.5 min/day, SE = 17.2) than the control group (458.8 min/day, SE = 12.3) (p = 0.002, d = -0.58) at two-years post-RYGB. However, this difference was also significantly different at pre-RYGB (p = 0.039), and therefore this difference may not be relevant from a clinical perspective, the authors noted.

An analysis of the participants attending the session about physical activity (n=61) versus participants in the intervention group who did not attend that session (n=38), showed no significant differences of clinical importance in any of the physical activity outcomes that changed any of the conclusions. There were also no differences in any of the pre-RYGB characteristics between these two groups.

“Our study therefore adds further knowledge on physical activity patterns among women from before to long-term post-RYGB. Finally, the WELL-GBP trial can easily be implemented in healthcare settings, as various healthcare personnel, such as nurses, dietitians and physiotherapists would be able to deliver the intervention after appropriate training,” they concluded. “Dissonance-based interventions focusing exclusively on the topic of physical activity might be more suitable and should be considered for future dissonance-based interventions aiming to increase physical activity levels among bariatric patients.”

Further information

To access this paper, please click here