Faculty Bibliography

IMPORTANCE:Sleeve gastrectomy is the most widely used bariatric operation; however, its long-term safety is largely unknown. OBJECTIVE:To compare the risk of mortality, complications, reintervention, and health care use 5 years after sleeve gastrectomy and gastric bypass. DESIGN, SETTING, AND PARTICIPANTS:This retrospective cohort study included adult patients in a national Medicare claims database who underwent sleeve gastrectomy or gastric bypass from January 1, 2012, to December 31, 2018. Instrumental variables survival analysis was used to estimate the cumulative incidence of outcomes up to 5 years after surgery. EXPOSURES:Laparoscopic sleeve gastrectomy and laparoscopic Roux-en-Y gastric bypass. MAIN OUTCOMES AND MEASURES:The main outcome was risk of mortality, complications, and reinterventions up to 5 years after surgery. Secondary outcomes were health care use after surgery, including hospitalization, emergency department (ED) use, and total spending. RESULTS:Of 95 405 patients undergoing bariatric surgery, 57 003 (60%) underwent sleeve gastrectomy (mean [SD] age, 57.1 [11.8] years), of whom 42 299 (74.2%) were women; 124 (0.2%) were Asian; 10 101 (17.7%), Black; 1951 (3.4%), Hispanic; 314 (0.6%), North American Native; 43 194 (75.8%), White; 534 (0.9%), of other race or ethnicity; and 785 (1.4%), of unknown race or ethnicity. A total of 38 402 patients (40%) underwent gastric bypass (mean [SD] age, 55.9 [11.7] years), of whom 29 050 (75.7%) were women; 109 (0.3%), Asian; 6038 (15.7%), Black; 1215 (3.2%), Hispanic; 278 (0.7%), North American Native; 29 986 (78.1%), White; 373 (1.0%), of other race or ethnicity; and 404 (1.1%), of unknown race or ethnicity. Compared with patients undergoing gastric bypass, at 5 years after surgery, patients undergoing sleeve gastrectomy had a lower cumulative incidence of mortality (4.27%; 95% CI, 4.25%-4.30% vs 5.67%; 95% CI, 5.63%-5.69%), complications (22.10%; 95% CI, 22.06%-22.13% vs 29.03%; 95% CI, 28.99%-29.08%), and reintervention (25.23%; 95% CI, 25.19%-25.27% vs 33.57%; 95% CI, 33.52%-33.63%). Conversely, patients undergoing sleeve gastrectomy had a higher cumulative incidence of surgical revision at 5 years (2.91%; 95% CI, 2.90%-2.93% vs 1.46%; 95% CI, 1.45%-1.47%). The adjusted hazard ratio (aHR) of all-cause hospitalization and ED use was lower for patients undergoing sleeve gastrectomy at 1 year (hospitalization, aHR, 0.83; 95% CI, 0.80-0.86; ED use, aHR, 0.87; 95% CI, 0.84-0.90) and 3 years (hospitalization, aHR, 0.94; 95% CI, 0.90-0.98; ED use, aHR, 0.93; 95% CI, 0.90-0.97) after surgery but similar between groups at 5 years (hospitalization, aHR, 0.99; 95% CI, 0.94-1.04; ED use, aHR, 0.97; 95% CI, 0.92-1.01). Total health care spending among patients undergoing sleeve gastrectomy was lower at 1 year after surgery ($28 706; 95% CI, $27 866-$29 545 vs $30 663; 95% CI, $29 739-$31 587), but similar between groups at 3 ($57 411; 95% CI, $55 239-$59 584 vs $58 581; 95% CI, $56 551-$60 611) and 5 years ($86 584; 95% CI, $80 183-$92 984 vs $85 762; 95% CI, $82 600-$88 924). CONCLUSIONS AND RELEVANCE:In a large cohort of patients undergoing bariatric surgery, sleeve gastrectomy was associated with a lower long-term risk of mortality, complications, and reinterventions but a higher long-term risk of surgical revision. Understanding the comparative safety of these operations may better inform patients and surgeons in their decision-making.

OBJECTIVE:We sought to measure the extent of variation in episode spending around total hip replacement within and across hospital systems. SUMMARY OF BACKGROUND DATA:Bundled payment programs are pressuring hospitals to reduce spending on surgery. Meanwhile, many hospitals are joining larger health systems with the stated goal of improved care at lower cost. METHODS:Cross-sectional study of fee-for-service Medicare patients undergoing total hip replacement in 2016 at hospital systems identified in the American Hospital Association Annual Survey. We calculated risk- and reliability-adjusted average 30-day episode payments at the hospital and system level. RESULTS:Average episode payments varied nearly as much within hospital systems ($2515 between the lowest- and highest-cost hospitals, 95% confidence interval $2272-$2,758) as they did between the lowest- and highest-cost quintiles of systems ($2712, 95% confidence interval $2545-$2879). Variation was driven by post-acute care utilization. Many systems have concentrated hip replacement volume at relatively high-cost hospitals. CONCLUSIONS:Given the wide variation in surgical spending within health systems, we propose tailored strategies for systems to maximize savings in bundled payment programs.

OBJECTIVE:The aim of this study was to understand relationships among insurance plan type, out-of-pocket cost sharing, and the utilization of bariatric surgery among commercially insured patients. BACKGROUND:Only 1% of eligible persons undergo bariatric operations, and this underutilization is often attributed to lack of insurance coverage. But even among the insured, underinsurance is now recognized as a major barrier to accessing medical care. The relationships among commercial insurance design, out-of-pocket cost sharing, and elective surgery utilization, particularly in bariatrics, are not well understood. METHODS:Retrospective review of 73,002 commercially insured members of the IBM MarketScan commercial claims database who underwent bariatric surgery from 2014 to 2017. The exposure variables were insurance plan type and out-of-pocket cost sharing. The outcome was utilization of bariatric surgery. We also examined seasonal trends in bariatric surgery utilization stratified by average levels of cost sharing. RESULTS:Utilization of bariatric surgery was higher in plans with lower cost sharing, such as PPOs (20 operations/100,000 enrollees) than in HDHPs (high-deductible health plans, 12.1 operations/100,000 enrollees). Overall, every $1000 increase in cost sharing was associated with 5 fewer bariatric operations per 100,000 insured lives; this association was strongest in plans with high cost sharing (high-deductible and consumer-directed health plans). Members of all plan types had higher surgical utilization in quarter 4 relative to quarter 1 of each year; these seasonal variations were also most pronounced in plans with high cost sharing. CONCLUSIONS:Insurance plan types with higher cost sharing have lower utilization of bariatric surgery. Underinsurance may represent a newly identified barrier to surgical care that should be addressed by advocates and policymakers.

BACKGROUND:Hiatal hernia repair performed at the time of laparoscopic sleeve gastrectomy (LSG) may reduce post-operative reflux symptoms. It is unclear whether intra-operative diagnosis of hiatal hernia varies among surgeons or if it affects outcomes. STUDY DESIGN:Surgeons (n = 38) participating in a statewide bariatric surgery quality improvement collaborative reviewed 33 videos of LSG in which no hiatal hernia repair was performed. Reviewers were blinded to patient information and were asked whether they perceived a hiatal hernia. Surgeon characteristics and surgeon-specific patient outcomes for LSG were compared between surgeons who identified at least one hiatal hernia during video review and those who did not. RESULTS:Ten surgeons (26%) identified at least one hiatal hernia after reviewing the videos. There were no significant differences in operative experience or practice type between surgeons who did and did not identify hiatal hernias. Surgeons who identified a hiatal hernia more often performed concurrent hiatal hernia repair in their practice when compared to those who did not (43.0% versus 36.5%, p < 0.001). Although complication rates were similar between surgeon groups, there were higher rates of de novo reflux symptoms (13.6% versus 11.1%, p = 0.032) and lower rates of antacid discontinuation at one-year (71.0% versus 77.2%, p = 0.043) among surgeons who identified hiatal hernias. CONCLUSION:Surgeons who identified hiatal hernias during video review had a higher rate of concurrent hiatal hernia repairs in their practice. This was not associated with improved patient-reported reflux symptoms after LSG. Standardizing identification and management of hiatal hernias during bariatric surgery may help improve reflux outcomes post-operatively.

BACKGROUND:Approximately 70% of hospitals today are part of larger health systems. Proponents of hospital consolidation tout its potential to reduce health spending and improve outcomes, but to the authors' knowledge the available evidence has suggested that this promise is unrealized. Variations in costs and outcomes within systems may highlight opportunities for collaborative quality improvement and practice standardization. To assess this potential, the authors sought to measure variations in episode spending within and across hospital systems among Medicare beneficiaries undergoing complex cancer surgery. METHODS:Using 100% Medicare claims data, the authors identified fee-for-service Medicare patients who were undergoing elective pancreatectomy, lung resection, or colectomy for cancer from 2014 through 2016. Risk-adjusted, price-standardized payments for the surgical episode from admission through 30 days after discharge were calculated. The authors then assessed the reliability-adjusted variations at the hospital and system levels. RESULTS:Average episode payments varied nearly as much within hospital systems for pancreatectomy ($1946 between the lowest and highest spending systems; 95% CI, $1910-$1972), lung resection ($625 between the lowest and highest spending systems; 95% CI, $621-$630), and colectomy ($813 between the lowest and highest spending systems; 95% CI, $809-$817) as they did between the lowest and highest spending hospitals (pancreatectomy: $2034; lung resection: $1789; and colectomy: $770). For pancreatectomy, this variation was driven by index hospitalization spending whereas both index hospitalization and postacute care use drove variations for lung resection and colectomy. CONCLUSIONS:In this analysis of Medicare patients undergoing complex cancer surgery, wide variations in surgical episode spending were noted both within and across hospital systems. System leaders may seek to better understand variations in practices among their hospitals to standardize care and reduce variations in outcomes, use, and costs.

IMPORTANCE:In any surgical procedure, various aspects of technique may affect patient outcomes. As new procedures enter practice, it is difficult to evaluate the association of each aspect of technique with patient outcomes. OBJECTIVE:To examine the associations between technique and outcomes in laparoscopic sleeve gastrectomy. DESIGN, SETTING, AND PARTICIPANTS:In this cohort study of bariatric surgery programs participating in a statewide surgical quality improvement collaborative, 30 surgeons submitted intraoperative videos from representative sleeve gastrectomies performed on 6915 patients with morbid obesity. These videos were reviewed by blinded peer surgeons on key technical elements, and 605 reviews were linked to sleeve gastrectomy outcomes of all of the surgeons' patients from January 1, 2015, to December 31, 2016. EXPOSURES:Surgeons' technical approaches to 5 controversial aspects of laparoscopic sleeve gastrectomy: dissection of the proximal stomach, sleeve caliber, sleeve anatomy, staple line reinforcement, and leak testing. MAIN OUTCOMES AND MEASURES:The 30-day outcomes were rate of postoperative hemorrhage and staple line leak. The 1-year outcomes were percentage of total weight lost and reflux severity (Gastroesophageal Reflux Disease Health-Related Quality of Life instrument). RESULTS:A total of 30 surgeons submitted 46 videos of operations performed on 6915 patients (mean [SD] age, 45.4 [11.7] years; 5494 [79.5%] female; 4706 [68.1%] White). Complete dissection of the proximal stomach was associated with reduced hemorrhage rates (higher ratings for complete mobilization of fundus were associated with a decrease in hemorrhage rate from 2.1% [25th percentile] to 1.0% [75th percentile], P = .01; higher ratings for visualization of the left crus were associated with a decrease in hemorrhage rate from 1.5% to 0.94%, P = .006; and higher ratings for complete division of the short gastrics were associated with a decrease in hemorrhage rate from 2.8% to 1.2%, P = .03). The reduction in hemorrhage rates came at the expense of higher leak rates (higher ratings for complete mobilization of fundus were associated with an increase in leak rate from 0.05% [25th percentile] to 0.16% [75th percentile], P < .001; higher ratings for visualization of the left crus were associated with an increase in leak rate from 0.1% to 0.2%, P = .003; and higher ratings for complete division of the short gastrics were associated with an increase in leak rate from 0.02% to 0.1%, P = .01). Surgeons who stapled more tightly to the bougie had smaller decreases in reflux than those who stapled less tightly (-2.0 to -1.3 on a 50-point scale, P = .002). Staple line reinforcement (buttressing and oversewing) was associated with a small (2 of 1000 cases) decrease in hemorrhage rates. Staple line buttressing was also associated with a similarly small increase in leak rates (1 of 1000 cases). Leak testing was associated with a statistically insignificant change in the staple line leak rate (0.16%-0.22%, P = .47). CONCLUSIONS AND RELEVANCE:Variations in surgical technique can be measured by video review and are associated with differences in patient outcomes.

This cohort study analyzes changes in financial risk protection associated with implementation of the Patient Protection and Affordable Care Act (ACA) across income strata and insurance types.