Training working memory to improve eating behavior after bariatric surgery

Obesity is a growing and worldwide health problem with a range of medical
comorbidities such as diabetes type 2, obstructive sleep apnea syndrome,
hypertension and hyperlipidemia, causing even more disability and medical costs
(Bray, 2004; Yu, Shen, Sun, Zhang, Wong, 2013). Besides physical comorbidities,
obesity is often associated with psychological distress, shame, depression and
social avoidance behavior (Lamers, et al., 2011). These physical and
psychological problems pose a significant burden on the health care system
(Bray, 2004) and make healthcare for these patients expensive (Neovius, et al.,
2012; Neff, Olbers & le Roux, 2013; Sampalis, Liberman, Auger & Christou,
2004).
Bariatric surgery is an effective treatment method for patients with morbid
obesity, resulting in clinically significant weight reduction, reduction of
comorbidity and improvement of psychosocial functioning and quality of life
(Cunneen, 2008; Herpertz, et al., 2004; Mathus-Vliegen & De Wit, 2007). Even
though bariatric surgery is the most effective treatment for morbid obesity,
insufficient results are found (< 50 % Excess Weight Loss (%EWL)) in 20-35% of
this group after bariatric surgery (Al-Bader, et al., 2015; Cooper, Simmons,
Webb, Burns & Kuscher, 2015; Magro, et al., 2008). Binge eating, grazing,
uncontrolled eating and postoperative loss of control have shown to predict
poorer weight loss and greater weight regain following bariatric surgery
(Ashton, Heinberg, Windover & Merrell, 2011; Canetti, Berry & Elizur, 2009).
Different studies also proved the importance of compliance to dietary rules in
terms of success after bariatric surgery (Van Hout & Leibbrandt, 2003; Elkins,
Whitfield, Mark, Symmonds, Rodiguez & Cook, 2005). In addition to the previous
described factors, obese patients often have poor regulation of eating habits,
less self-control and self-regulation than patients with normal weight (Jansen,
et al., 2009). In self-regulation mechanisms, executive functioning is
involved. With regard to obesity, problems in executive functioning are found
to play a role in experiencing difficulties maintaining weight loss and
adjusting and adhering to necessary lifestyle changes after bariatric surgery
(Boeka & Lokken, 2008). Even though an association is found between BMI and
executive functioning, it remains unclear whether higher BMI has a negative
effect on executive functioning or poor executive functioning leads to higher
BMI (Boeka & Lokken, 2008).
Recent research showed that working memory, which is an executive function,
contributes to the successful self-regulation of behavior, including eating
behavior (Hofmann, Schmeichel & Baddeley, 2012). Executive functions, for a
greater part, depend on WM capacity. It is therefore, that self-regulation
could profit from interventions that help to increase WM capacity. Training WM
has been found to be successful to improve other executive functions as well
(Klingberg, 2010).
Training WM could possibly be an effective strategy to improve eating behavior
(i.e., less emotional or external eating behavior, an increase of restrained
eating behavior) and improving self-regulation capacity (i.e. following diet
rules).

Bariatric surgery does not directly target the underlying psychological factors
that potentially contribute to the development and maintenance of obesity
(Cassin et al., 2013). Despite some psychosocial factors predicting
postoperative weight loss and weight regain, psychosocial interventions are not
routinely offered in bariatric surgery programs (Cassin et al., 2013). Even
though bariatric surgery is the most effective treatment (in terms of weight
reduction, reduction of comorbidity and improvement of psychosocial functioning
and quality of life for patients with morbid obesity, insufficient results are
found regarding behavior change and weight loss in the long term (Al-Bader, et
al., 2015; Cooper, et al., 2015; Magro, et al., 2008; Cassin, et al., 2013;
Cunneen, 2008; Herpertz, et al., 2004; Mathus-Vliegen & De Wit, 2007). An
important turning point is found around 18 to 24 months after surgery when a
part of the patients experience weight regain and relapse of obesity-related
comorbidities (Cassin, et al., 2013; Herpertz, et al., 2004). In the current
study, we aim to investigate whether training WM (by means of computerized
WM-tasks), as an important factor in executive functioning, will improve eating
behavior (that is less emotional, external or restrained eating behavior) and
self-regulation capacity (i.e., following diet-rules) and if improved WM
capacity will lead to the maintenance of BMI in the longer term. More specific,
the effect of training WM on eating behavior, BMI and self-regulation among
patients 14-18 months after bariatric surgery (before the important turning
point) will be studied.

A randomized controlled trial will be carried out to test the hypothesis that
WM-training will successfully improve eating behavior, self-regulation capacity
and maintenance of BMI. A comparison between 2 groups will be made: obese
participants complete either WM-training or control tasks for 25 consecutive
sessions, 14-18 months after bariatric surgery.

WM- training and control tasks: All participants (training and control
conditions) will complete 3 WM tasks: A visuospatial WM task, a backward digit
span task and a letter span task (adapted from Klingberg, Forssberg &
Westerberg, 2002). Each of the 3 tasks consists of 30 trials. In the training
condition, the difficulty level will automatically be adjusted on a
trial-by-trial basis. In the control condition, the difficulty level of the
WM-tasks will remain at the initial level throughout each task. The WM capacity
will be measured using the same tasks that were used during WM training. The
assessment version ends when participants are not able to reproduce a sequence
on two consecutive trials. The outcome measure for each task is the length of
the longest sequences that participants correctly reproduce on two consecutive
trials.

There are no risks and detrimental consequences associated with participating
in the study. Participants can stop their participation in the study at any
time. The burden of completing the questionnaires and to be weighted is low.
The burden of the training is higher, but it is expected that the training will
have beneficial consequences. Participants will be offered this intervention
free of charge.