Predicting Screening Factors for an Obesity Team’s Decision: A large Retrospective Cohort study

Background The magnitude of the effect of bariatric surgery depends on the willingness and the ability of patients to adapt a healthier lifestyle. Therefore, it is worthwhile to know how to select patients for surgery and/or preoperative interventions. Aim To investigate the predictors for a disapproval at the time, direct approval or a prehabilitation program for bariatric surgery candidates after multidisciplinary evaluation. Of the 2,063 included patients, 627 (30.4%) were approved for surgery, 1,275 (61.8%) received a preoperative trajectory advise and 161 (7.8%) were denied. Eleven variables appeared to be significant predictors for the result of the multidisciplinary consultation; gender, age, BMI, RAND-36 total score, SQ-48 total score, binging score, craving score, consciously eating, psychological help score, excessive drug use in the past and the impulsiveness score. Although the predicting model might not be applicable to other centers, it could attribute to enhance patient selection and thereby improve healthcare logistics.


Introduction
Bariatric surgery is the only treatment with proven long-term results for patients with severe obesity [1][2][3].
However, the extent of surgery's success depends on the willingness and ability to adapt to a healthier lifestyle. According to the interdisciplinary European guidelines on metabolic and bariatric surgery, patients eligible for bariatric surgery should follow an interdisciplinary evaluation [1]. This evaluation should be performed by an obesity team consisting of a surgeon, dietitian, psychologist, and a nurse. The team decides if a bariatric surgery candidate is psychologically and physically ready to make the required lifestyle changes for sustainable weight loss.
The challenge for the multidisciplinary obesity team is to select patients with morbid obesity who are most likely to adhere to the necessary adjustments. Although this evaluation is widely accepted and practiced, there is no standardized assessment procedure described in international guidelines [1][2][3]. In addition to preoperative evaluation, some bariatric centers provide a preoperative prehabilitation program on exercise, diet and/or cognitive behavioral therapy. The positive effects of exercise and diet programs have been described in the enhanced recovery after surgery guidelines, such as reduced postoperative complications and reduced liver volume. However, the evidence is insufficient to standardize prehabilitation before bariatric surgery in clinical practice guidelines [3]. Additionally, a recent 12-year follow-up study showed that patients who were denied for surgery never sought weight loss surgery again although they eventually would qualify based on National Institutes of Health consensus criteria [4]. On the other hand, the extensive preoperative nutritional and psychological evaluation does not prevent the risk of weight regain [5,6]. To enhance the screening process and thus patient selection for an additional program, the retrospective cohort study aimed to investigate the predictors for disapproval at the time, approval. or a prehabilitation program for bariatric surgery candidates after multidisciplinary evaluation.

Data Analysis
Depending on normality, differences for continuous variables between multidisciplinary outcome groups were measured using the one-way analysis of variance

Results
In total 2,304 patients were extracted from the hospital database, of which 241 were excluded due to missing data on more than two questionnaires (n = 188) or applied for revisional surgery (n = 53   (Figure 1).

Preoperative trajectory versus direct approval for surgery
The adjusted odds ratio of the model with direct

Equation 1
The variables gender, consciously eating and excessive drug use are dichotomous data and must be filled in as 0 if female, eating consciously and used excessive drugs in the past and as 1 if male, not eating consciously and did not use excessive drugs in the past.

Denied for surgery versus direct approval for surgery
The adjusted odds ratio of the model with direct approval for surgery as reference category shows that the odds of being denied for surgery is 2.2 times higher for male individuals and 3.3 times higher for patients who excessively used drugs in the past ( The variables gender, consciously eating and excessive drug use are dichotomous data and must be filled in as 0 if female, eating consciously and used excessive drugs in the past and as 1 if male, not eating consciously and did not use excessive drugs in the past.

Discussion
A multinomial logistic regression model was built using three screening questionnaires and basic characteristics from 2,063 patients to investigate the predicting factors for the obesity team's decision: approval for surgery, denial at the time, or a preoperative prehabilitation program. Eleven variables appeared to be significant predictors, i.e., gender, age, BMI, RAND-36 total score, SQ-48 total score, binging score, craving score, consciously eating, psychological help score, excessive drug use in the past and the impulsiveness score. All variables were positively correlated with a higher odds for receiving prehabilitation advice or being denied for surgery compared to the direct approval group, except for age.
Men had a significantly higher chance for either a preoperative program advice or refusal for surgery.
This finding is consistent with that of Wee et al. who examined patients' consideration for surgery and showed that weight loss was more desired by women than men and therefore women might be more motivated [11]. Furthermore, compared to women, men were less likely to have accurate weight perception, weight dissatisfaction, and attempted weight loss [12].
In our study, denied patients were significantly heavier and younger (  [13]. Other studies also showed that older age was associated with less successful weight loss one year after surgery [14,15].
On the other hand, Altieri et al. reported that patients between 30 and 50 years of age and with higher preoperative weights may be more successful in losing weight in a 3 to 6-months during preoperative weight loss program including behavioral modifications and nutrition counseling [16]. Also, higher preoperative weights appeared to be positively correlated to postoperative weight loss up to 60 months [17]. On the contrary, other authors found that greater preoperative BMI was negatively associated with postoperative weight loss up to 135 months [6,14].
Current drug abuse is a contraindication for bariatric surgery, prior abuse is less obvious [1]. Tedesco et al. showed that patients with substance abuse achieve equivalent weight loss at 6 and 12 months postoperative compared to patients without alcohol or drug abuse [18]. A review of 2019 by Kanji et al. supported this belief, but it may contribute to increased substance abuse after bariatric surgery [19]. New-onset abuse is also evident after surgery, the follow-up in our center favorably endures for 5 years [20,21]. In our model, the odds of being denied for surgery was 3.3 times higher for patients who abused drugs in the past compared to patients who did not.

Conclusion
A multinomial logistic regression was built using three screening questionnaires and basic characteristics from 2,063 patients to investigate the predicting factors for the obesity team's decision. Although the exact model might not apply to other bariatric centers, the elements found are likely to be universal. This predictive model could attribute to enhance patient selection and thereby improve healthcare logistics.

Ethical approval
No formal ethical approval was necessary in line with legislation on non-WMO research in the Netherlands, and as the accessed data was fully anonymized. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Conflict of interests
The authors declare that they have no conflict of interest.