The value of surgical experience: excess costs associated with the Roux-en-Y gastric bypass learning curve

Machine Learning Models for Predicting Postoperative Outcomes following Skull Base Meningioma Surgery

Objective  While predictive analytic techniques have been used to analyze meningioma postoperative outcomes, to our knowledge, there have been no studies that have investigated the utility of machine learning (ML) models in prognosticating outcomes among skull base meningioma patients. The present study aimed to develop models for predicting postoperative outcomes among skull base meningioma patients, specifically prolonged hospital length of stay (LOS), nonroutine discharge disposition, and high hospital charges. We also validated the predictive performance of our models on out-of-sample testing data. Methods  Patients who underwent skull base meningioma surgery between 2016 and 2019 at an academic institution were included in our study. Prolonged hospital LOS and high hospital charges were defined as >4 days and >$47,887, respectively. Elastic net logistic regression algorithms were trained to predict postoperative outcomes using 70% of available data, and their predictive performance was evaluated on the remaining 30%. Results  A total of 265 patients were included in our final analysis. Our cohort was majority female (77.7%) and Caucasian (63.4%). Elastic net logistic regression algorithms predicting prolonged LOS, nonroutine discharge, and high hospital charges achieved areas under the receiver operating characteristic curve of 0.798, 0.752, and 0.592, respectively. Further, all models were adequately calibrated as determined by the Spiegelhalter Z -test ( p >0.05). Conclusion  Our study developed models predicting prolonged hospital LOS, nonroutine discharge disposition, and high hospital charges among skull base meningioma patients. Our models highlight the utility of ML as a tool to aid skull base surgeons in providing high-value health care and optimizing clinical workflows.

Predicting High-Value Care Outcomes After Surgery for Skull Base Meningiomas

BACKGROUND

Although various predictors of adverse postoperative outcomes among patients with meningioma have been established, research has yet to develop a method for consolidating these findings to allow for predictions of adverse health care outcomes for patients diagnosed with skull base meningiomas. The objective of the present study was to develop 3 predictive algorithms that can be used to estimate an individual patient's probability of extended length of stay (LOS) in hospital, experiencing a nonroutine discharge disposition, or incurring high hospital charges after surgical resection of a skull base meningioma.

METHODS

The present study used data from patients who underwent surgical resection for skull base meningiomas at a single academic institution between 2017 and 2019. Multivariate logistic regression analysis was used to predict extended LOS, nonroutine discharge, and high hospital charges, and 2000 bootstrapped samples were used to calculate an optimism-corrected C-statistic. The Hosmer-Lemeshow test was used to assess model calibration, and P < 0.05 was considered statistically significant.

RESULTS

A total of 245 patients were included in our analysis. Our cohort was mostly female (77.6%) and white (62.4%). Our models predicting extended LOS, nonroutine discharge, and high hospital charges had optimism-corrected C-statistics of 0.768, 0.784, and 0.783, respectively. All models showed adequate calibration (P>0.05), and were deployed via an open-access, online calculator: https://neurooncsurgery3.shinyapps.io/high_value_skull_base_calc/.

CONCLUSIONS

After external validation, our predictive models have the potential to aid clinicians in providing patients with individualized risk estimation for health care outcomes after meningioma surgery.

Earlier Provision of Gastric Bypass Surgery in Canada Enhances Surgical Benefit and Leads to Cost and Comorbidity Reduction

Background: Effective provision of bariatric surgery for patients with obesity may be impeded by concerns of payers regarding costs or perceptions of patients who drop out of surgical programs after referral. Estimates of the cost and comorbidity impact of these inefficiencies in gastric bypass surgery in Canada are lacking but would aid in informing healthcare investment and resource allocation.

Objectives: To estimate total and relative public payer costs for surgery and comorbidities (diabetes, hypertension, and dyslipidemia) in a bariatric surgery population.

Methods: A decision analytic model for a 100-patient cohort in Canada (91% female, mean body mass index 49.2 kg/m², 50% diabetes, 66% hypertension, 59% dyslipidemia). Costs include surgery, surgical complications, and comorbidities over the 10-year post-referral period. Results are calculated as medians and 95% credibility intervals (CrIs) for a pathway with surgery at 1 year (“improved”) compared with surgery at 3.5 years (“standard”). Sensitivity analyses were performed to test independent contributions to results of shorter wait time, better post-surgical weight loss, and randomly sampled cohort demographics.

Results: Compared to standard care, the improved path was associated with reduction in patient-years of treatment for each of the three comorbidities, corresponding to a reduction of $1.1 (0.68–1.6) million, or 34% (26-41%) of total costs. Comorbidity treatment costs were 9.0- and 4.7-fold greater than surgical costs for the standard and improved pathways, respectively. Relative to non-surgical bariatric care, earlier surgery was associated with earlier return on surgical investment and 2-fold reduction in risk of prevalence of each comorbidity compared to delayed surgery.

Conclusions: Comorbidity costs represent a greater burden to payers than the costs of gastric bypass surgery. Investments may be worthwhile to reduce wait times and dropout rates and improve post-surgical weight loss outcomes to save overall costs and reduce patient comorbidity prevalence.