Background: Quality of metabolic bariatric surgery (MBS) care is often monitored by national registries using quality indicators (QIs), but data collection takes up considerable time and costs. QIs are mostly introduced merely based on expert opinion. Therefore, the study’s aim was to systematically evaluate whether all QIs from the Dutch Audit for Treatment of Obesity (DATO) are still relevant and useful to initiate quality improvement initiatives. Methods: Twenty-four QIs were evaluated using hospital data from 2022 to 2023. To test whether QIs measured the same quality of care aspect (parsimony and relevance), correlations of QI pairs were examined using Pearson correlation coefficients. Usefulness to identify improvement opportunities was considered limited when variance is ≤ 0.001 without any outliers identified, indicating that the QI could be retired. Actionability was assessed through line graphs of hospital performance over the years. Results: Eleven QIs were highly correlated to other QIs, particularly the follow-up and weight loss indicators at 2 and 4 years, and therefore lacked added value. The weight loss QIs showed minimal variance and were adjusted by increasing the threshold to achieving ≥ 25% total weight loss. Multiple QIs showed improving trends and thereby their actionability, most pronounced for postoperative complications. The final QI set measured three constructs with good validity: Cronbach’s alpha values 0.53 (safety), 0.70 (treatment effectiveness), and 0.43 (follow-up process). Conclusion: Through a systematic evaluation of the DATO QI set, a smaller set of 13 QIs was shown to capture the same relevant information to improve MBS care.

Background

Metabolic bariatric surgery (MBS) quality registries monitor various outcomes, enabling the assessment of hospital performance in comparison with national benchmarks. However, if there is considerable between-surgeon outcome variation, surgeon-level feedback may be better suited. The aim of this study was to assess the extent to which patient-, surgeon-, and hospital-level factors contribute to the variation in outcomes after MBS. 

Methods

All primary procedures registered in the Dutch MBS quality registry between 1 January 2020 and 31 December 2023 were included. Outcomes included severe postoperative complications, reoperation, prolonged length of stay (LOS), readmission, textbook outcome, and achieving ≥25% total weight loss within 1 year. Multilevel logistic regression models were built for each outcome, including all available patient characteristics, operating surgeon, and hospital, to determine the variance explained by patient-, surgeon-, and hospital-level factors. 

Results

In total, 30 610 patients were included, operated on by 144 surgeons in 19 hospitals. Hospital-level factors contributed most to the explained variance for all outcomes, ranging from 59.6% for reoperation to 90.3% for prolonged LOS. Surgeon-level factors explained less variance, ranging from 3.2% for prolonged LOS to 28.2% for reoperation. Patient characteristics explained the least, ranging from 4.4% for textbook outcome to 13.1% for severe postoperative complications. 

Conclusion

Variation in outcomes is mostly explained by hospital factors, rather than surgeon factors, supporting hospital-based performance feedback. The results suggest that the pre- and postoperative trajectory and perioperative care may affect MBS outcomes more than patient characteristics or surgical team performance.

Introduction: Some patients undergoing metabolic bariatric surgery (MBS) may transfer to another hospital for subsequent procedures. Due to legal constraints imposed by privacy regulations on inter-hospital data sharing, limited research has examined the characteristics and outcomes of these patients. This study aimed to identify the frequency and factors associated with hospital transfer using a novel privacy-enhancing approach based on secure multi-party computation (MPC). Methods: All primary and secondary MBS procedures registered in the Dutch Audit for Treatment of Obesity between January 1, 2014, and December 31, 2022, were considered. MPC enabled privacy-preserving linkage of surgeries across different hospitals. Patients undergoing secondary surgery in the same or a different hospital were compared on patient and treatment characteristics and outcomes of primary MBS to investigate associations with hospital transfer. Results: Two thousand three hundred eighty-two patients with data on both primary and secondary MBS were identified. A minority (n = 275; 11.5%) underwent their second procedure elsewhere. At baseline, these patients on average were younger (37.9 vs. 42.5, p < 0.001), less often had hypertension or GERD, and had similar BMI (43.9 vs 43.9, p = 0.89) compared with those who stayed. At secondary surgery, the BMI of patients transferring hospitals on average was lower (39.0 vs. 43.0, p < 0.001), and the indication was more often recurrent weight gain (49.0% vs. 23.0%, p < 0.001). Conclusion: A minority of patients (1 in 9) transferred to a different hospital for secondary MBS. These patients were generally younger and had fewer obesity-related diseases. Although they presented with a lower BMI at the time of secondary surgery, they more frequently sought surgery for recurrent weight gain.

The name of author Perla J. Marang-van de Mheen was tagged incorrectly. “Perla J.” should be tagged as the given name and “Marang-van de Mheen” should be tagged as the family name.

Introduction: Bariatric surgery aims for optimal patient outcomes, often evaluated through the percentage total weight loss (%TWL). Quality registries employ funnel plots for outcome comparisons between hospitals. However, funnel plots are traditionally used for dichotomous outcomes, requiring %TWL to be dichotomized, potentially limiting feedback quality. This study evaluates whether a funnel plot around the median %TWL has better discriminatory performance than binary funnel plots for achieving at least 20% and 25% TWL. Methods: All hospitals performing bariatric surgery were included from the Dutch Audit for Treatment of Obesity. A funnel plot around the median was constructed using 5-year %TWL data. Hospitals positioned above the 95% control limit were colored green and those below red. The same hospitals were plotted in the binary funnel plots for 20% and 25% TWL and colored according to their performance in the funnel plot around the median. We explored the hospital’s procedural mix in relation to %TWL performance as possible explanatory factors. Results: The median-based funnel plot identified four underperforming and four outperforming hospitals, while only one underperforming and no outperforming hospitals were found with the binary funnel plot for 20% TWL. The 25% TWL binary funnel plot identified two underperforming and three outperforming hospitals. The proportion of sleeve gastrectomies performed per hospital may explain part of these results as it was negatively associated with median %TWL (β = − 0.09, 95% confidence interval [− 0.13 to − 0.04]). Conclusion: The funnel plot around the median discriminated better between hospitals with significantly worse and better performance than funnel plots for dichotomized %TWL outcomes. Graphical Abstract: (Figure presented.)

Background: Risk-prediction tools can support doctor–patient (shared) decision making in clinical practice by providing information on complication risks for different types of bariatric surgery. However, external validation is imperative to ensure the generalizability of predictions in a new patient population. Objective: To perform an external validation of the risk-prediction model for serious complications from the Michigan Bariatric Surgery Collaborative (MBSC) for Dutch bariatric patients using the nationwide Dutch Audit for Treatment of Obesity (DATO). Setting: Population-based study, including all 18 hospitals performing bariatric surgery in the Netherlands. Methods: All patients registered in the DATO undergoing bariatric surgery between 2015 and 2020 were included as the validation cohort. Serious complications included, among others, abdominal abscess, bowel obstruction, leak, and bleeding. Three risk-prediction models were validated: (1) the original MBSC model from 2011, (2) the original MBSC model including the same variables but updated to more recent patients (2015–2020), and (3) the current MBSC model. The following predictors from the MBSC model were available in the DATO: age, sex, procedure type, cardiovascular disease, and pulmonary disease. Model performance was determined using the area under the curve (AUC) to assess discrimination (i.e., the ability to distinguish patients with events from those without events) and a graphical plot to assess calibration (i.e., whether the predicted absolute risk for patients was similar to the observed prevalence of the outcome). Results: The DATO validation cohort included 51,291 patients. Overall, 986 patients (1.92%) experienced serious complications. The original MBSC model, which was extended with the predictors “GERD (yes/no),” “OSAS (yes/no),” “hypertension (yes/no),” and “renal disease (yes/no),” showed the best validation results. This model had a good calibration and an AUC of .602 compared with an AUC of .65 and moderate to good calibration in the Michigan model. Conclusion: The DATO prediction model has good calibration but moderate discrimination. To be used in clinical practice, good calibration is essential to accurately predict individual risks in a real-world setting. Therefore, this model could provide valuable information for bariatric surgeons as part of shared decision making in daily practice.

Background: Literature remains scarce on patients experiencing weight recurrence after initial adequate weight loss following primary bariatric surgery. Therefore, this study compared the extent of weight recurrence between patients who received a Sleeve Gastrectomy (SG) versus Roux-en-Y gastric bypass (RYGB) after adequate weight loss at 1-year follow-up. Methods: All patients undergoing primary RYGB or SG between 2015 and 2018 were selected from the Dutch Audit for Treatment of Obesity. Inclusion criteria were achieving ≥ 20% total weight loss (TWL) at 1-year and having at least one subsequent follow-up visit. The primary outcome was ≥ 10% weight recurrence (WR) at the last recorded follow-up between 2 and 5 years, after ≥ 20% TWL at 1-year follow-up. Secondary outcomes included remission of comorbidities at last recorded follow-up. A propensity score matched logistic regression analysis was used to estimate the difference between RYGB and SG. Results: A total of 19.762 patients were included, 14.982 RYGB and 4.780 SG patients. After matching 4.693 patients from each group, patients undergoing SG had a higher likelihood on WR up to 5-year follow-up compared with RYGB [OR 2.07, 95% CI (1.89–2.27), p < 0.01] and less often remission of type 2 diabetes [OR 0.69, 95% CI (0.56–0.86), p < 0.01], hypertension (HTN) [OR 0.75, 95% CI (0.65–0.87), p < 0.01], dyslipidemia [OR 0.44, 95% CI (0.36–0.54), p < 0.01], gastroesophageal reflux [OR 0.25 95% CI (0.18–0.34), p < 0.01], and obstructive sleep apnea syndrome (OSAS) [OR 0.66, 95% CI (0.54–0.8), p < 0.01]. In subgroup analyses, patients who experienced WR after SG but maintained ≥ 20%TWL from starting weight, more often achieved HTN (44.7% vs 29.4%), dyslipidemia (38.3% vs 19.3%), and OSAS (54% vs 20.3%) remission compared with patients not maintaining ≥ 20%TWL. No such differences in comorbidity remission were found within RYGB patients. Conclusion: Patients undergoing SG are more likely to experience weight recurrence, and less likely to achieve comorbidity remission than patients undergoing RYGB.