Surgeon-level Variation Research Articles

Surgeon-Level Variation in Utilization of Local Staging and Neoadjuvant Therapy for Stage II-III Rectal Adenocarcinoma

IntroductionNeoadjuvant therapy (NT) is the standard of care for clinical stage II-III rectal adenocarcinoma, but utilization remains suboptimal. We aimed to determine the underlying reasons for omission of local staging and NT. MethodsWe conducted a retrospective study of patients with clinical stage II-III or undocumented clinical stage/pathologic stage II-III rectal adenocarcinoma who were treated in 2010–2016 in one of nine Intermountain Healthcare hospitals. The outcomes of omission of local staging and NT were examined with multivariable models. Risk- and reliability-adjusted rates of local staging and NT were calculated for surgeons who treated ≥ 3 patients. Pathologic and long-term outcomes were examined after excluding patients who were not resected or who underwent local excision (N = 11). ResultsLocal staging was omitted in 43/240 (17.9%) patients and NT was omitted in 41/240 (17.1%). The strongest risk factors for local staging and NT omission were upper rectal tumors and surgeons who treated ≤ 3 cases/year. Thirty-six of 41 (87.8%) cases of omitted NT had local staging omitted. Adjusted surgeon-specific local staging rates varied 1.6-fold (56.3–92.4%) and NT rates varied 2.8-fold (34.1–97.1%). Surgeon local staging and NT rates were strongly correlated (r = 0.92). NT was associated with lower rates of positive circumferential radial margins (7.9 vs. 20.0%; P = 0.02), node positivity (33.3 vs. 55.0%; P = 0.01), and local recurrences (7.6 vs. 14.9% at 5 years; P = 0.0176). ConclusionsNT omission should be understood as a consequence of surgeon failure to perform local staging in most cases. Quality improvement efforts should focus on improving utilization of local staging.

Surgeon-driven variability in emergency general surgery outcomes: Does it matter who is on call?

BackgroundHospital-level variation has been found to influence outcomes in emergency general surgery. However, whether the individual surgeon plays a role in this variation is unknown. MethodsWe performed an analysis of the Florida State Inpatient Database (2010–2014), which is linked to the American Hospital Association's Annual Survey Database, including patients who emergently underwent 1 or more of 7 procedures (laparotomy, adhesiolysis, small bowel resection, colectomy, repair of a perforated gastric ulcer, appendectomy, or cholecystectomy). We used multilevel random effects modeling to quantify the amount of variation in mortality, complications, and 30-day readmissions attributable to surgeons. Patient clinical and demographic factors, as well as hospital-level factors, were introduced into the model in a forward stepwise fashion, and the percent of the variation attributable to surgeons was derived. ResultsOur study included 2,149 surgeons across 224 hospitals, with a total of 569,767 emergency general surgery cases. The overall unadjusted mortality rate was 3.8%, and the complication and readmission rates were 12.7% and 27.7%, respectively. Surgeon-level variation had the greatest impact on mortality, explaining 32.77% of the overall variability in mortality risk compared with 0.08% and 2.28% for complications and readmissions, respectively. Peptic ulcer disease operations were most susceptible to surgeon-level variation in mortality and readmissions, whereas appendectomies and cholecystectomies were least susceptible to surgeon-level variation for all outcomes. ConclusionsSurgeon-level variation contributes to a significant portion of mortality in EGS. This variation is most pronounced in surgery for peptic ulcer disease, a high-risk, low-frequency surgical condition. Programs to reduce mortality in emergency general surgery should address reducing variability in practice with attention to high-risk, low-frequency procedures.

416 - Surgeon-Level Variation in Utilization of Neoadjuvant Therapy for Locally Advanced Rectal Adenocarcinoma

416 - Surgeon-Level Variation in Utilization of Neoadjuvant Therapy for Locally Advanced Rectal Adenocarcinoma

The impact of individual physicians on outcomes after trauma: is it the system or the surgeon?

BackgroundBenchmarking of mortality outcomes across the country has revealed major differences in survival based on the trauma center at which a patient receives care. The role of the individual surgeon in determining trauma outcomes is unknown. Most believe that differences in outcomes are primarily driven by system- and process-based variations. Our objective was to determine if variation in individual surgeon outcomes could help explain difference in survival after trauma. MethodsAnalysis of trauma patients in the Florida State Inpatient Database from 2010 to 2014. The presence of unique physician identifiers, in addition to hospital identifiers, rendered this data set ideal for performance of multilevel analysis. The amount of the variation attributable to surgeon-level variation was calculated using multilevel random-effects models controlling for patient clinical factors (such as injury severity and comorbidities/age) and hospital-level factors, such as case mix and bed size. ResultsThere were 31 hospitals, 175 surgeons, and 65,706 admissions. The overall mortality rate was 5.6%. The average mortality rate across surgeons ranged from 0% to 17.4% (mean 0.4%, standard deviation 1.85). At the individual surgeon level, when controlling for clinical and hospital-level factors, 9% of this variation was attributable solely to the surgeon. ConclusionsAt the state level, we found that differences in outcomes among trauma centers are impacted by individual surgeon-level variation. Implementation of protocolized, system-based trauma care is useful for improving the overall quality of care for injured patients but does not entirely negate surgeon-specific variations in management.

Trends in Surgeon-Level Utilization of Sacral Nerve Stimulator Implantation for Fecal Incontinence in New York State.

There is a paucity of real-world data regarding surgeon utilization of sacral nerve stimulation for fecal incontinence compared with anal sphincteroplasty. This study aims to examine trends in sacral nerve stimulation use compared with sphincteroplasty for fecal incontinence and surgeon-level variation in progression to implantation of the pulse generator. This is a population-based study. Patients with fecal incontinence between 2011 and 2014 in New York who underwent stage 1 of the sacral nerve stimulation procedure were selected. For the comparison with sphincteroplasty, patients with fecal incontinence who underwent anal sphincteroplasty between 2008 and 2014 were included. The main outcomes after sacral nerve stimulation generator placement were unplanned 30-day admission, emergency department visit within 30 days, revision or explant of leads or generator, and 30-day mortality. Six hundred twenty-one patients with fecal incontinence underwent a stage 1 procedure with 79.7% progressing to stage 2. There has been an increase in the number of sacral nerve stimulation cases per year as well as the number of surgeons performing the procedure. The rate of progression to stage 2 among patients treated by colorectal surgeons was 80.2% compared with 77.0% among those treated by noncolorectal surgeons. Among those who completed stage 2, there were 3 (0.5%) unplanned 30-day admissions, 24 (4.4%) emergency department visits within 30 days, and 0 mortalities within 30 days. Thirty-two (6.5%) patients had their leads or pulse generator revised or explanted. There was a significant decrease in annual sphincteroplasty cases and the number of providers performing the procedure starting in 2011. We lacked data regarding patient and physician decision making and the severity of disease. Sacral nerve stimulation for fecal incontinence is increasing in popularity with an increasing number of surgeons utilizing sacral nerve stimulation for fecal incontinence rather than sphincteroplasty. See Video Abstract at http://links.lww.com/DCR/A450.

Explaining variation in ventral and inguinal hernia repair outcomes: A population-based analysis

No study has evaluated the relative importance of patient, surgeon, and hospital-level factors on surgeon and hospital variation in hernia reoperation rates. This population-based retrospective cohort study evaluated factors associated with variation in reoperation rates for recurrence after initial ventral hernia repair and inguinal hernia repair. The Statewide Planning and Research Cooperative System identified initial ventral hernia repairs and inguinal hernia repairs in New York state from 2003-2009. Mixed-effects Cox proportional hazards analyses were performed assessing factors associated with surgeon/hospital variation in 5-year reoperation rates for hernia recurrence. Among 78,267 ventral hernia repairs and 124,416 inguinal hernia repairs, the proportion of total variation in reoperation rates attributable to individual surgeons compared with hospitals was 87% for ventral hernia repairs and 92% for inguinal hernia repairs. In explaining variation in ventral hernia repair reoperation between surgeons, 19% was attributable to patient-level factors, 4% attributable to mesh placement, and 10% attributable to surgeon volume and type of board certification. In explaining variation in inguinal hernia repair reoperation between surgeons, 1.1% was attributable to mesh placement and 10% was attributable to surgeon volume and years of experience. However, 67% of the variation between surgeons for ventral hernia repair and 89% of the variation between surgeons for inguinal hernia repair remained unexplained by factors in the models. The majority of variation in hernia reoperation rates is attributable to surgeon-level variation. This suggests that hernia recurrence may be an appropriate surgeon quality metric. While modifiable factors such as mesh placement and surgeon characteristics play roles in surgeon variation, future research should focus on identifying additional surgeon attributes responsible for this variation.

Abstract P34: Variation in Transfusion Practice and Associated Postoperative Complications in Abdominally-Based Autologous Breast Reconstruction Surgery

PURPOSE: Blood transfusion is the most frequently performed procedure in US hospitals. Despite its life-saving potential when clinically indicated, evidence shows that transfusion in some cases does not provide a significant clinical benefit to patients and thus does not justify its associated costs and risks. Increasing emphasis has been placed on reducing these potentially unnecessary transfusions, demonstrated by recently published guidelines from the American Association of Blood Banks (AABB) recommending restrictive hemoglobin thresholds for transfusion (≤7 g/dL). In plastic and reconstructive surgery, free flap procedures are associated with high utilization of blood transfusion, reported in the literature to be as high as 42%. The high volume of flap-based breast reconstructions makes this procedure an ideal target for quality improvement interventions related to blood transfusion practices. In this study, we assessed variation in transfusion practice and its associated clinical outcomes in a large cohort of breast reconstruction patients to identify potential targets for quality improvement. METHODS: After obtaining IRB approval, we extracted perioperative blood utilization data and hemoglobin transfusion triggers from two prospectively-collected anesthesia and blood management databases (Metavision and IMPACT Online) for all patients who underwent abdominally-based autologous breast reconstruction at the Johns Hopkins Hospital between 2009 and 2015. We defined hemoglobin transfusion triggers as the lowest measured hemoglobin level preceding a blood transfusion. We used ANOVA, Chi-squared, and linear regression to examine patient-level and surgeon-level variation in the use of overall blood transfusions, potentially unnecessary blood transfusions, and hemoglobin transfusion triggers. RESULTS: Of 653 patients, 65 (10%) received perioperative blood transfusions. Risk factors for increased blood utilization were higher ASA class (OR: 2.4; p=0.015) and younger age (OR: 1.2 for every 5-year decrease in age; p=0.008), with a trend for the presence of rheumatic comorbidities (OR: 3.7; p=0.098). Use of perioperative blood transfusions varied by surgeon (range: 5% to 24%; p=0.001), suggesting the presence of variation in transfusion practices. Mean hemoglobin trigger was 6.6 g/dL (±0.83 g/dL; range 3.4–8.2 g/dL). Hemoglobin triggers varied by surgeon (range: 6 g/dL to 7.5 g/dL; p<0.001) and patient age (p=0.031), with a trend for Charlson comorbidity index (p=0.093). Of the 65 patients that received blood transfusions, 16 patients (25%) had potentially unnecessary transfusions (hemoglobin triggers ≥7g/dL). Potentially unnecessary utilization of blood transfusion did not vary by surgeon (range: 1% to 4%; p=0.142), but was higher in TRAM flap reconstructions (OR: 9; p=0.001) and showed a trend for higher ASA class (OR: 3; p=0.076). Patients who received blood transfusions experienced worse clinical outcomes in terms of postoperative infections (p=0.006), Clostridium difficile infection (p=0.003), sepsis (p<0.001), and 30-day readmission (p<0.001). CONCLUSION: Use of perioperative blood transfusion varied among surgeons. However, there was no evidence for surgeon-level variation in potentially unnecessary perioperative transfusion. Transfusions were significantly associated with higher risk of perioperative infections and postoperative readmissions. These findings emphasize the importance of standardizing transfusion practices with the goal of minimizing unnecessary transfusions and their potential negative consequences.

Surgeon and Hospital Level Variation in the Costs of Robot-Assisted Radical Prostatectomy

We assessed surgeon and hospital level variation in robot-assisted radical prostatectomy costs and predictors of high and low cost surgery. The study population consisted of a weighted sample of 291,015 men who underwent robot-assisted radical prostatectomy for prostate cancer by 667 surgeons at 197 U.S. hospitals from 2003 to 2013. We evaluated 90-day direct hospital costs (2014 USD) in the Premier Hospital Database. High costs per robot-assisted radical prostatectomy were those above the 90th percentile and low costs were those below the 10th percentile. Mean hospital cost per robot-assisted radical prostatectomy was $11,878 (95% CI $11,804-$11,952). Mean cost was $2,837 (95% CI $2,805-$2,869) in the low cost group vs $25,906 (95% CI $24,702-$25,490) in the high cost group. Nearly a third of the variation in robot-assisted radical prostatectomy cost was attributable to hospital characteristics and more than a fifth was attributable to surgeon characteristics (R-squared 30.43% and 21.25%, respectively). High volume surgeons and hospitals (90th percentile or greater) had decreased odds of high cost surgery (surgeons: OR 0.24, 95% CI 0.11-0.54; hospitals: OR 0.105, 95% CI 0.02-0.46). The performance of robot-assisted radical prostatectomy at a high volume hospital was associated with increased odds of low cost robot-assisted radical prostatectomy (OR 839, 95% CI 122-greater than 999). This study provides insight into the role of surgeons and hospitals in robot-assisted radical prostatectomy costs. Given the substantial variability, identifying and remedying the root cause of outlier costs may yield substantial benefits.

Surgeon-Level Variation in Postoperative Complications

BackgroundVariation in surgical outcomes is often attributed to patient comorbidities and the severity of underlying disease, but little is known about the extent of variation in outcomes by surgeon and the surgeon factors that are associated with quality. MethodsUsing the Maryland Health Services Cost Review Commission database, we evaluated risk-adjusted postoperative events by surgeon. Operations studied were elective laparoscopic and open colectomy procedures for colon cancer performed over a 2-year period (July 2012–September 2014). Postoperative events were defined using the Agency for Healthcare Research and Quality Patient Safety Indicators. Surgeons performing fewer than ten procedures during the study period were excluded. Logistic regression and post-estimation were used to calculate an observed-to-expected (O/E) ratio of postoperative complications for each surgeon, adjusting for patient and surgeon characteristics. ResultsA total of 2525 patients underwent an elective colectomy during the study period by 276 surgeons at 44 hospitals. Postoperative complications varied more by surgeon (range 0 to 30.0 %) than by hospital (range 0 to 18.2 %). Surgeon-level use of laparoscopic surgery to perform colectomy ranged from 0 to 100 %. After risk adjustment with patient factors, surgeon experience, surgeon medical school, surgeon gender, and annual surgeon colectomy volume were not associated with postoperative complications. Surgeon use of laparoscopy was the strongest predictor of lower complications (vs fourth quartile of surgeons, first quartile OR = 0.47 (0.26–0.85); second quartile OR = 0.41 (0.22–0.73); and third quartile OR = 0.84 (0.52–1.36). ConclusionsQuality metrics in health care have been measured at the hospital level, but a greater quality improvement potential exists at the surgeon level. Awareness of this variation could better inform patients undergoing elective surgery and their referring physicians.

PI-04 SURGEON AND HOSPITAL VARIATION IN THE COSTS OF ROBOT-ASSISTED RADICAL PROSTATECTOMY IN THE UNITED STATES

You have accessJournal of UrologyBest Abstract1 Apr 2016PI-04 SURGEON AND HOSPITAL VARIATION IN THE COSTS OF ROBOT-ASSISTED RADICAL PROSTATECTOMY IN THE UNITED STATES Alexander Cole, Jeffrey Leow, Steven Chang, Christian Meyer, Nawar Hanna, Adam Kibel, Mani Menon, Jesse Sammon, Firas Abdollah, Benjamin Chung, Maxine Sun, and Quoc-Dien Trinh Alexander ColeAlexander Cole More articles by this author , Jeffrey LeowJeffrey Leow More articles by this author , Steven ChangSteven Chang More articles by this author , Christian MeyerChristian Meyer More articles by this author , Nawar HannaNawar Hanna More articles by this author , Adam KibelAdam Kibel More articles by this author , Mani MenonMani Menon More articles by this author , Jesse SammonJesse Sammon More articles by this author , Firas AbdollahFiras Abdollah More articles by this author , Benjamin ChungBenjamin Chung More articles by this author , Maxine SunMaxine Sun More articles by this author , and Quoc-Dien TrinhQuoc-Dien Trinh More articles by this author View All Author Informationhttps://doi.org/10.1016/j.juro.2016.02.1209AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookTwitterLinked InEmail INTRODUCTION AND OBJECTIVES In a cost-conscious healthcare environment, eliminating unwarranted costs and understanding cost variation are paramount. Prostate cancer is the most common malignancy among men and surgical treatment with radical prostatectomy (RP) is associated with substantial costs, particularly with the recent rapid adoption of robot-assisted RP (RARP). We sought to assess individual surgeon-level variation in costs of RARP and identify independent predictors of high and low-cost surgery. METHODS We identified 291,015 men in the Premier all-payer discharge database who underwent RARP for prostate cancer by 667 unique surgeons at 197 different hospitals across the United States from 2003 to 2013. We evaluated 90-day direct hospital costs (2014 US$). High costs were defined as costs superior to the 90th percentile of all patient costs (≥$18,564/RARP), while low costs were defined as the costs inferior to the 10th percentile of all patient costs (≤$5,585/RARP). Multivariable logistic regression models identified independent predictors of high and low costs, while multilevel hierarchal models determined variability of various characteristics on costs RESULTS Mean costs for each RARP patient was $11,878 (95% CI 11,804-11,952), Mean costs per RARP in the low-cost group was $2837 (95% CI: $2805-$2869) compared to $25,906 (95% CI: $24702-$25490) in the high-cost group. Nearly a third of the variation in RARP cost was attributable to hospital characteristics, and over a fifth was attributable to surgeon characteristics (R-squared for each 30.43% and 21.25%, respectively). Factors associated with high-cost surgeries included Black race (vs. Whites; OR: 1.42, 95% CI: 1.12 to 1.33), Charlson comorbidity≥2 (vs. 0; OR 1.88 95% CI 1.55-2.29) and teaching hospitals (vs. non-teaching; OR 2.27, 95% CI 1.28-4.04). High-volume providers (≥90th percentile) was associated with significantly lower odds of high-cost surgery (surgeons: OR 0.24, 95% CI 0.11-0.54; hospitals: OR 0.105 95% CI 0.02-0.46). As for predictors for low-cost surgeries, high-volume surgeons were found to be less likely to have low-cost RARPs (OR: 0.41, 95% CI: 0.20-0.84). Conversely, high-volume hospitals were highly associated with low-cost RARPs (OR: 839, p<0.0001). CONCLUSIONS This study provides new insight into the key role played by individual surgeons performing RARP across a large nationally representative sample. Given the substantial variability in RARP costs contributed by hospital and surgeon characteristics, future steps to identify and remedy the root cause of outliers costs, and to reduce unwarranted variability in costs may yield great benefits for patients, physicians and the health system as a whole. © 2016FiguresReferencesRelatedDetails Volume 195Issue 4SApril 2016Page: e410 Advertisement Copyright & Permissions© 2016MetricsAuthor Information Alexander Cole More articles by this author Jeffrey Leow More articles by this author Steven Chang More articles by this author Christian Meyer More articles by this author Nawar Hanna More articles by this author Adam Kibel More articles by this author Mani Menon More articles by this author Jesse Sammon More articles by this author Firas Abdollah More articles by this author Benjamin Chung More articles by this author Maxine Sun More articles by this author Quoc-Dien Trinh More articles by this author Expand All Advertisement Advertisement PDF downloadLoading ...

PD39-07 SURGEON VARIATION IN THE COSTS OF RADICAL CYSTECTOMY

You have accessJournal of UrologyBladder Cancer: Invasive VI1 Apr 2016PD39-07 SURGEON VARIATION IN THE COSTS OF RADICAL CYSTECTOMY Jeffrey Leow, Alexander Cole, Steven Chang, Christian Meyer, Adam Kibel, Mani Menon, Thomas Seisen, Jesse Sammon, Mark Preston, Benjamin Chung, Maxine Sun, and Quoc-Dien Trinh Jeffrey LeowJeffrey Leow More articles by this author , Alexander ColeAlexander Cole More articles by this author , Steven ChangSteven Chang More articles by this author , Christian MeyerChristian Meyer More articles by this author , Adam KibelAdam Kibel More articles by this author , Mani MenonMani Menon More articles by this author , Thomas SeisenThomas Seisen More articles by this author , Jesse SammonJesse Sammon More articles by this author , Mark PrestonMark Preston More articles by this author , Benjamin ChungBenjamin Chung More articles by this author , Maxine SunMaxine Sun More articles by this author , and Quoc-Dien TrinhQuoc-Dien Trinh More articles by this author View All Author Informationhttps://doi.org/10.1016/j.juro.2016.02.1513AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookTwitterLinked InEmail INTRODUCTION AND OBJECTIVES In a cost-conscious healthcare environment, identifying unwarranted costs and cost variability is paramount. Bladder cancer (BCa) incurs the highest lifetime treatment costs per patient of all cancers and radical cystectomy (RC), as a cornerstone of BCa treatment, is a major contributor. Under these premises, the aim of the current study was to assess individual surgeon-level variations and identify independent predictors of high- and low-cost RC. METHODS We queried the Premier all-payer discharge database to accrue a study cohort of a weighted sample of 11,225 individuals who underwent RC by 292 unique surgeons at 144 different hospitals between 2003 and 2013. Surgeon and hospital volume as well as 90-day direct hospital costs (2014 US$) were ranked and divided into high and low volume, and high and low costs. High volume and costs were defined as values above the 90th percentile while low volume and costs corresponded to costs and volume below the 10th percentile. Multivariable logistic regression models identified independent predictors of high and low costs, while multilevel hierarchal models determined variability of various characteristics on costs. RESULTS Mean 90-day direct hospital costs for each RC patient was $32,261 (95% CI $31220 - $33,302). There were 70 surgeons whose mean costs associated with each RC fell in the lowest decile of costs (<$16,278), with mean costs of $13,654 (95% CI: $13,191 to $14,116). There were 113 surgeons with mean costs for each RC in the top decile (>$51,285), with mean costs of $82,642 (95% CI: $76541 to $88744). Factors associated with high costs include worse comorbidity profile (vs. no comorbidity: OR 3.45, 95% CI 2.21- 5.38, p<0.0001) and laparoscopic approach (vs. open: 2.83, 95% CI: 1.52 to 5.27, p=0.0004). We found that patient, hospital and surgeon characteristics had a modest effect on the variability of costs (4.9%, 2.1%, 0.6% respectively). However, the presence of a 90-day major complication after RC and a prolonged postoperative length of stay contributed substantially to the overall cost variability (18.1% and 18.2% respectively). CONCLUSIONS There was a nearly 6-fold difference in costs between the most and least costly surgeons performing RC. There appears to be a key role played by individual surgeons across a large nationally representative sample. Given the modest variability in costs contributed by hospital and surgeon characteristics, the avoidance and prompt management of major complications may be key to reduce unwarranted variability in BCa/RC costs. © 2016FiguresReferencesRelatedDetails Volume 195Issue 4SApril 2016Page: e932 Advertisement Copyright & Permissions© 2016MetricsAuthor Information Jeffrey Leow More articles by this author Alexander Cole More articles by this author Steven Chang More articles by this author Christian Meyer More articles by this author Adam Kibel More articles by this author Mani Menon More articles by this author Thomas Seisen More articles by this author Jesse Sammon More articles by this author Mark Preston More articles by this author Benjamin Chung More articles by this author Maxine Sun More articles by this author Quoc-Dien Trinh More articles by this author Expand All Advertisement Advertisement PDF downloadLoading ...

Reliability of surgeon-specific reporting of complications after colectomy.

We sought to determine the reliability of surgeon-specific postoperative complication rates after colectomy. Conventional measures of surgeon-specific performance fail to acknowledge variation attributed to statistical noise, risking unreliable assessment of quality. We examined all patients who underwent segmental colectomy with anastomosis from 2008 through 2010 participating in the Michigan Surgical Quality Collaborative Colectomy Project. Surgeon-specific complication rates were risk-adjusted according to patient characteristics with multiple logistic regression. Hierarchical modeling techniques were used to determine the reliability of surgeon-specific risk-adjusted complication rates. We then adjusted these rates for reliability. To evaluate the extent to which surgeon-level variation was reduced, surgeons were placed into quartiles based on performance and complication rates were compared before and after reliability adjustment. A total of 5033 patients (n = 345 surgeons) undergoing partial colectomy reported a risk-adjusted complication rate of 24.5%. Approximately 86% of the variability of complication rates across surgeons was explained by measurement noise, whereas the remaining 14% represented true signal. Risk-adjusted complication rates varied from 0% to 55.1% across quartiles before adjusting for reliability. Reliability adjustment greatly diminished this variation, generating a 1.2-fold difference (21.4%-25.6%). A caseload of 168 colectomies across 3 years was required to achieve a reliability of more than 0.7, which is considered a proficient level. Only 1 surgeon surpassed this volume threshold. The vast majority of surgeons do not perform enough colectomies to generate a reliable surgeon-specific complication rate. Risk-adjusted complication rates should be viewed with caution when evaluating surgeons with low operative volume, as statistical noise is a large determinant in estimating their surgeon-specific complication rates.

Sources of variation in follow-up expenditure after radical cystectomy

BackgroundFollow-up care after radical cystectomy is poorly defined, with extensive variation in practice patterns. We sought to determine sources of these variations in care as well as examine the economic effect of standardization of care to guideline-recommended care. MethodsUsing linked Surveillance, Epidemiology, and End Results (SEER)-Medicare data from 1992 to 2007, we determined follow-up care expenditures (time and geography standardized) for 24 months after surgery. Accounted expenditures included office visits, imaging studies, urine tests, and blood work. A multilevel model was implemented to determine the effect of region, surgeon, and patient factors on care delivery. We then compared the actual expenditures on care in the Medicare system (interquartile range) with the expenditures if patients received care recommended by current clinical guidelines. ResultsExpenditures over 24 months of follow-up were calculated per month and per patient. The mean and median total expenditures per patient were $1108 and $805 respectively (minimum $0, maximum $9,805; 25th–75th percentile $344–$1503). Variations in expenditures were most explained at the patient level. After accounting for surgeon and patient levels, we found no regional-level variations in care. Adherence to guidelines would lead to an increase in expenditures by 0.80 to 10.6 times the expenditures exist in current practice. ConclusionAlthough some regional-level and surgeon-level variations in care were found, the most variation in expenditure on follow-up care was at the patient level, largely based on node positivity, chemotherapy status, and final cancer stage. Standardization of care to current established guidelines would create higher expenditures on follow-up care than current practice patterns.

494 Wide Surgeon-Level Variation in Complications After Colectomy

admissions were defined as admissions occurring on Saturday and Sunday. Results: We identified 1,564,277 patients with a primary diagnosis of PUB. Weekend admissions (WA) accounted for 24% and non-weekend (NWA) 76% of the discharges (p=<0.0001). The percentage of male and female patients for WA and NWA was approximately 53% and 46% (p=<0.0001). The mean age for WA and NWA was approximately 67 years (p=0.07) and the mean length of stay for WA was 5.68 and for NWA was 5.43 (p=<0.0001). The overall percentage mortality for PUB for NWA was 3% and for WA was 4% (p=<0.001). ICU admissions accounted for 12% of the WA and 11% of the NWA (p=<0.0001). Approximately 62% had EGD, 18% had small bowel enteroscopy (SBE) only and 5% had both EGD and SBE. Only 15% did not receive inpatient endoscopy at discharge. Among non ICU cases, WA had higher inpatient mortality when compared with NWA, odds ratio (OR=1.27); 95% confidence interval (CI) 1.08-1.49, p=0.0043). Patients admitted on the weekend were more likely to have delayed endoscopy when compared with NWA (OR= 2.22; 95%CI, 2.082.37, p=<0.0001). There was no difference in the rates of inpatient endoscopy between WA and NWA (OR= 0.90; 95% CI 0.791-1.03, p=0.14). There was no difference in mortality between ICU WA and NWA admissions (OR, 0.96; 95% CI, 0.86-1.13), p=0.835. There was no delay in endoscopy between ICU WA and NWA patients (OR, 1.12; 95% CI, 0.8651.46), p=0.38. Finally, there was no difference in rates of inpatient endoscopy for both ICU admissions on the weekend and NWA (OR=1.052; 95% CI 0.967-1.15, p=0.23). Conclusions: Non ICU, weekend admission with PUB had a significantly higher mortality and were two times more likely to have delayed endoscopy when compare with non ICU non-weekend admissions. The two cohorts had similar rates of inpatient endoscopy. There is no difference in rates of mortality, delayed endoscopy or inpatient endoscopy among ICU admissions.

Sources of variation in follow-up expenditure after radical cystectomy.

335 Background: Follow-up care after radical cystectomy is poorly defined with extensive variation in practice patterns. We sought to determine sources of these variations in care as well as examine the economic impact of standardization of care to guideline recommended care. Methods: Using linked SEER-Medicare data from 1992 to 2007, we determined follow-up care expenditures (time and geography standardized) for 24 months after surgery. Accounted costs included office visits, imaging studies, urine tests and blood work. A multilevel model was implemented to determine the impact of region, surgeon, and patient factors on care delivery. We then compared the actual expenditures on care in the Medicare system (interquartile range) to the expenditures if patients received care recommended by current clinical guidelines. Results: Expenditures over 24 months of follow-up were calculated per month and per patient. The mean and median monthly expenditures were $33 and $21 respectively (minimum $0, maximum $429, 25th to 75th percentile $9 to $43). The total variance of expenditure situated at the surgeon-level and SEER region-level was 9.9% and 4.0% respectively. After accounting for the region, the total variance of expenditure situated at the patient-level and surgeon-level was 14.95% and 7.81% respectively. The assessed cost of guideline follow-up recommendations varied from 0.78-9.05 times the calculated Medicare costs. The guideline recommended cost of follow-up was higher than actual Medicare expenditures in all but one category. Conclusions: While some regional and surgeon-level variations in care were found, most variation in expenditure on follow-up care was at the patient-level, largely based on comorbidity, node positivity, chemotherapy status, readmission rates, and final cancer stage. Standardization of care to current established guidelines would create larger expenditures for the Medicare system than current practice patterns.

Variation in Warfarin Thromboprophylaxis After Mitral Valve Repair: Does Equipoise Exist and Is a Randomized Trial Warranted?

There are limited data available to inform decision making regarding warfarin thromboprophylaxis early after mitral valve repair. We studied 13,082 patients from The Society of Thoracic Surgeons Adult Cardiac Surgery Database (STS ACSD) who underwent primary mitral valve repair between January 1, 2008, and June 30, 2010. Excluded were those having other major concomitant operations or with an indication/contraindication to warfarin. The predictors of warfarin administration at dismissal were evaluated. In this cohort (median age 58 years; 59% male), warfarin was prescribed at hospital dismissal for 46% (5,963) of patients. Median postoperative length of stay was 5 days overall (6 days warfarin versus 5 days, p < 0.0001). Substantial surgeon and center variation existed, and multivariable analysis identified that warfarin use was more common among patients with postoperative atrial fibrillation (odds ratio [OR] 4.04, 95% confidence interval [CI]: 3.57 to 4.58), postoperative neurologic events (stroke OR 1.72, 95% CI: 1.08 to 2.71; transient ischemic attack/reversible ischemic neurologic deficit OR 6.29, 95% CI: 2.67 to 14.84), and preoperative arrhythmia (OR 2.49, 95% CI: 1.84 to 3.38). Warfarin use was less common among patients having surgery in the more recent era (OR 0.92, 95% CI: 0.89 to 0.96, per half-year increase in date of surgery), those requiring intraoperative transfusion of red blood cells (OR 0.82, 95% CI: 0.71 to 0.96), and patients with advanced heart failure (New York Heart Association functional class IV OR 0.77, 95% CI: 0.59 to 1.00). At present, half of patients are prescribed warfarin after isolated mitral valve repair in North American cardiac surgical practice which may impact the length of hospital stay. Although patient-level predictors of warfarin prescription exist, center- and surgeon-level variations are prominent. There is a pressing need for a randomized trial both to guide therapy and to ascertain the potential for resource conservation.

Surgeon-Level Variation in Patients’ Appraisals of Their Breast Cancer Treatment Experiences

While variation in breast cancer quality indicators has been studied, to date there have been no studies examining the degree of surgeon-level variation in patient-reported outcomes. The purpose of this study is to examine surgeon-level variation in patient appraisals of their breast cancer care experiences. Survey responses and clinical data from breast cancer patients reported to Detroit and Los Angeles Surveillance, Epidemiology and End Results registries from 6/2005 to 2/2007 were merged with attending surgeon surveys (1,780 patients, 291 surgeons). Primary outcomes were patient reports of access to care, care coordination, and decision satisfaction. Random-effects models examined variation due to individual surgeons for these three outcomes. Mean values on each patient-reported outcome scale were high. The amount of variation attributable to individual surgeons in the unconditional models was low to modest: 5.4% for access to care, 3.3% for care coordination, and 7.5% for decision satisfaction. Few factors were independently associated with patient reports of better access to or coordination of care, but less-acculturated Latina patients had lower decision satisfaction. Patients reported generally positive experiences with their breast cancer treatment, though we found disparities in decision satisfaction. Individual surgeons did not substantively explain the variation in any of the patient-reported outcomes.

Factors Associated with the Adoption of Minimally Invasive Radical Prostatectomy in the United States

Minimally invasive radical prostatectomy has supplanted radical retropubic prostatectomy in popularity despite the absence of strong comparative effectiveness data demonstrating its superiority. We examined the influence of patient, surgeon and hospital characteristics on the use of minimally invasive radical prostatectomy vs radical retropubic prostatectomy. Using SEER (Surveillance, Epidemiology and End Results)-Medicare linked data we identified 11,732 men who underwent radical prostatectomy from 2003 to 2007. We assessed the contribution of patient, surgeon and hospital characteristics to the likelihood of undergoing minimally invasive radical prostatectomy vs radical retropubic prostatectomy using multilevel logistic regression mixed models. Patient factors (36.7%) contributed most to the use of minimally invasive radical prostatectomy vs radical retropubic prostatectomy, followed by surgeon (19.1%) and hospital (11.8%) factors. Among patient specific factors Asian race (OR 1.86, 95% CI 1.27-2.72, p = 0.001), clinically organ confined tumors (OR 2.71, 95% CI 1.60-4.57, p <0.001) and obtaining a second opinion from a urologist (OR 3.41, 95% CI 2.67-4.37, p <0.001) were associated with the highest use of minimally invasive radical prostatectomy while lower income was associated with decreased use of minimally invasive radical prostatectomy. Among surgeon and hospital specific factors, higher surgeon volume (OR 1.022, 95% CI 1.015-1.028, p <0.001), surgeon age younger than 50 years (OR 2.68, 95% CI 1.69-4.24, p <0.001) and greater hospital bed size (OR 1.001, 95% CI 1.001-1.002, p <0.001) were associated with increased use of minimally invasive radical prostatectomy, while solo or 2 urologist practices were associated with decreased use of minimally invasive radical prostatectomy (OR 0.48, 95% CI 0.27-0.86, p = 0.013). The adoption of minimally invasive radical prostatectomy vs radical retropubic prostatectomy is multifactorial, and associated with specific patient, surgeon and hospital related factors. Obtaining a second opinion from another urologist was the strongest factor associated with opting for minimally invasive radical prostatectomy.

Repeat surgery after lumbar decompression for herniated disc: the quality implications of hospital and surgeon variation

Repeat lumbar spine surgery is generally an undesirable outcome. Variation in repeat surgery rates may be because of patient characteristics, disease severity, or hospital- and surgeon-related factors. However, little is known about population-level variation in reoperation rates. To examine hospital- and surgeon-level variation in reoperation rates after lumbar herniated disc surgery and to relate these to published benchmarks. Retrospective analysis of a discharge registry including all nonfederal hospitals in Washington State. We identified adults who underwent an initial inpatient lumbar decompression for herniated disc from 1997 to 2007. We then performed generalized linear mixed-effect logistic regressions, controlling for patient characteristics and comorbidity, to examine the variation in reoperation rates within 90 days, 1 year, and 4 years. Our cohort included 29,529 patients with a mean age of 47.5 years, 61% privately insured, and 15% having any comorbidity. The age-, sex-, insurance-, and comorbidity-adjusted mean rate of reoperation among hospitals was 1.9% at 90 days (95% confidence interval [CI], 1.2-3.1), with a range from 1.1% to 3.4%; 6.4% at 1 year (95% CI, 3.9-10.6), with a range from 2.8% to 12.5%; and 13.8% at 4 years (95% CI, 8.8-19.8), with a range from 8.1% to 24.5%. The adjusted mean reoperation rates of surgeons were 1.9% at 90 days (95% CI, 1.4-2.4) with a range from 1.2% to 4.6%, 6.1% at 1 year (95% CI, 4.8-7.7) with a range from 4.3% to 10.5%, and 13.2% at 4 years (95% CI, 11.3-15.5) with a range from 10.0% to 19.3%. Multilevel random-effect models suggested that variation across surgeons was greater than that of hospitals and that this effect increased with long-term outcomes. Even after adjusting for patient demographics and comorbidity, we observed a large variation in reoperation rates across hospitals and surgeons after lumbar discectomy, a relatively simple spinal procedure. These findings suggest uncertainty about indications for repeat surgery, variations in perioperative care, or variations in quality of care.