To the Editor:

The authors of this study [1] are accurate in noting that there is significant variability in preoperative psychological evaluation practices, as well as limited success to date in identifying preoperative psychosocial variables that predict weight loss outcomes after weight loss surgery (WLS). In this study, the authors used a “traffic light” system for classifying patients after psychological assessment for bariatric surgery, with “green light” (GL) indicating no apparent psychosocial contraindications, “yellow light” (YL) for individuals about whom there were concerns and thus psychological intervention was required prior to surgery, and “red light” (RL) for patients who were deemed unsuitable for WLS for psychosocial reasons. This classification system is appealing in its simplicity and straightforwardness. However, the study does raise some issues related to both its methodology and the potential conclusions that might be drawn from the findings. There are two very different potential interpretations of the study’s findings, and while there is room for debate regarding which of these is most accurate or appropriate, it is worth noting that both potential interpretations highlight important issues related to the complex interplay of clinical practice and research within the field of bariatric psychology. The two interpretations are discussed below.

Interpretation 1: Given that the YL and GL groups were not significantly different in their postoperative early weight loss or adherence to postoperative clinic follow-up appointments, it may be that presurgical psychosocial assessment is not useful in predicting variations in outcomes following WLS.

Discussion: Despite numerous studies examining the relationship between preoperative mood and anxiety symptoms/diagnoses and WLS outcomes, findings have been strikingly mixed, yielding limited evidence that these indices are consistent predictors of either medical or psychosocial WLS outcomes. Indeed, in the current study, the rationale for the authors’ emphasizing mood issues above other factors in determining patients’ suitability for surgery, along with requiring that these were addressed prior to surgery (thereby delaying access to a medically necessary procedure), is not clear. One of the key aims of psychosocial evaluation for WLS is to determine not just the presence of symptoms, but, more usefully, whether they may impact upon the individual’s ability to implement and sustain long-term behavior changes necessary for optimal long-term WLS outcome. It is important to shift our focus away from simply documenting the presence of psychiatric disorders in the psychosocial evaluation for WLS. This narrow focus promotes discrimination against individuals purely on the basis of their mental health status, which may or may not relate to WLS outcomes. Obviously, if mood issues are impeding the individual’s ability to engage in appropriate self-care, this is a legitimate basis for concern, but this does not seem to have been examined in the classification system described by the authors of this study. Given that even within a group of individuals with mood and anxiety symptoms, there is wide heterogeneity in the extent to which these affect the patient’s functioning and self-care, it is not surprising that no differences in the outcomes of YL and GL patients were found, given that the classification was made on the basis of the presence, rather than the impact, of mood symptoms.

Another potential explanation for the equivalent outcomes observed between GL and YL patients in this study is that WLS may, in and of itself, have a powerful effect on mood and anxiety symptoms. Preoperative anxiety and depression are often a reflection of the impact that the individual’s weight has had on self-image, health-related quality of life, self-efficacy, practical and social functioning, and life opportunities. Research clearly shows that as individuals lose weight, there are associated positive changes in mood, and it has been found that these improvements correspond proportionally to the degree of weight loss achieved [2].

Not only do we question the emphasis placed on mood and anxiety symptoms to determine group classification, but we also have concerns about the fact that this study, like many empirical papers to date, focused almost exclusively on weight loss as the primary outcome (with some attention also paid to postoperative visit adherence). We propose that it may be time to broaden the definition of surgical “success.” Notably, the psychosocial evaluation may be useful in highlighting other important domains of successful outcome (e.g., lessened functional disability, improved psychosocial functioning, etc.) that were not examined in this study.

Interpretation 2: Despite significant mood and/or anxiety symptoms at baseline, the YL group was able to achieve equivalent outcomes to the GL group because potential issues were identified and remediated as a result of the psychosocial evaluation process.

Discussion: It seems quite plausible that the reason that the YL patients, who were more psychosocially compromised at the time of presurgical evaluation, were able to achieve outcomes commensurate to those in the GL group, was that potential risk factors were identified in the psychosocial evaluation, and the YL patients received appropriate treatment to minimize these risk factors. In essence, as a result of the psychosocial evaluation process, the YL patients were effectively “turned into” GL patients. As the authors do not provide data regarding whether there were changes in the YL patients’ anxiety and depression symptoms following the required psychological intervention, we cannot definitively conclude that YL and GL groups were equivalent in terms of psychosocial status by the time of surgery. However, this was ostensibly the intended aim of the required recommended preoperative psychological treatment and a plausible explanation for the similarity in postsurgical outcomes. Notably, if the authors’ intention in this study was to determine whether preoperative mood problems were associated with poorer weight loss outcomes, a control group consisting of YL individuals who did not receive psychological intervention but still proceeded to surgery would have been needed. A sizeable proportion of the total study population (23 %) was advised to undergo psychotherapy prior to surgery, but most did not complete this requirement and thus did not have surgery. If these patients had been allowed to proceed to surgery, comparisons of their outcomes to the outcomes of the GL patients would have been informative. This would also have yielded interesting data regarding whether mood improves as a consequence of WLS, rather than specific psychological intervention delivered before surgery.

To conclude, we would suggest that it is time to shift our focus away from crystal-ball-gazing predictions about who will, and will not, have a successful weight loss outcome following surgery, and instead, focus our efforts and skills on providing interventions throughout the surgical process that can enhance all domains of WLS outcome for each individual patient.