Developing a Machine Learning Risk-adjustment Method for Hospitalizations and Emergency Department Visits of Nursing Home Residents With Dementia.

BACKGROUND: Long-stay nursing home (NH) residents with Alzheimer disease and related dementias (ADRD) are at high risk of hospital transfers. Machine learning might improve risk-adjustment methods for NHs. OBJECTIVES: The objective of this study was to develop and compare NH risk-adjusted rates of hospitalizations and emergency department (ED) visits among long-stay residents with ADRD using Extreme Gradient Boosting (XGBoost) and logistic regression. RESEARCH DESIGN: Secondary analysis of national Medicare claims and NH assessment data in 2012 Q3. Data were equally split into the training and test sets. Both XGBoost and logistic regression predicted any hospitalization and ED visit using 58 predictors. NH-level risk-adjusted rates from XGBoost and logistic regression were constructed and compared. Multivariate regressions examined NH and market factors associated with rates of hospitalization and ED visits. SUBJECTS: Long-stay Medicare residents with ADRD (N=413,557) from 14,057 NHs. RESULTS: A total of 8.1% and 8.9% residents experienced any hospitalization and ED visit in a quarter, respectively. XGBoost slightly outperformed logistic regression in area under the curve (0.88 vs. 0.86 for hospitalization; 0.85 vs. 0.83 for ED visit). NH-level risk-adjusted rates from XGBoost were slightly lower than logistic regression (hospitalization=8.3% and 8.4%; ED=8.9% and 9.0%, respectively), but were highly correlated. Facility and market factors associated with the XGBoost and logistic regression-adjusted hospitalization and ED rates were similar. NHs serving more residents with ADRD and having a higher registered nurse-to-total nursing staff ratio had lower rates. CONCLUSIONS: XGBoost and logistic regression provide comparable estimates of risk-adjusted hospitalization and ED rates.

Social determinants of health Z-codes and postoperative outcomes after colorectal surgery: A national population-based study.

BACKGROUND: We sought to evaluate the impact of social determinants of health (SDOH) on postoperative outcomes following colorectal surgery. METHODS: This retrospective cohort study used PearlDiver-Mariner, an all-payer insurance claims database. Patients who underwent colectomy or proctectomy between 2010 and 2020 were included. SDOH were identified using International Classification of Diseases diagnosis codes. Outcomes were compared using multivariable regression models. RESULTS: The 30-day postoperative complication rate among 333,387 patients (mean age, 59 years; 58% female) was 27%. Approximately 5% of patients reported at least one SDOH at baseline. SDOH were not associated with length of stay but were associated with higher odds of 30-day postoperative complications (OR:1.16, 95% CI:1.12-1.20), including urinary tract infection (OR:1.27, 95% CI:1.20-1.35) anastomotic leak (OR:1.22, 95% CI:1.16-1.28), pneumonia (OR:1.19, 95% CI:1.11-1.27), deep vein thrombosis (OR:1.13, 95% CI:1.02-1.23), sepsis (OR:1.12, 95% CI:1.07-1.18), disruption of wound (OR:1.12, 95% CI:1.03-1.21), and acute kidney injury (OR:1.04, 95% CI:0.99-1.10). CONCLUSIONS: SDOH Z-codes were associated with worse postoperative complications following colorectal surgery and may help target high-risk patients.

Disparities in patient and system factors explain racial/ethnic disparities in delayed time to treatment in muscle invasive bladder cancer.

PURPOSE: Treatment delays in muscle invasive bladder cancer (MIBC) have been shown to be associated with worse outcomes. While every attempt is made to provide adequate treatment expeditiously, Black and Hispanic patients often experience delays at a higher rate than their White counterparts. This study aims to quantify the mechanisms that contribute to this disparity in treatment delay. METHODS: Retrospective analysis of clinical T-stages II-IVa MIBC patients who underwent surgical resection from 2004 to 2017 in the National Cancer Database. A causal inference mediation analysis using the counterfactual framework was implemented to estimate the extent to which racial/ethnic disparities in patient and system factors explain the racial/ethnic disparities in time to treatment. Mediators included income, education, comorbidities, insurance, and hospital type. RESULTS: Among 22,864 patients who met inclusion criteria, 7%, 3%, 2% were of Black, Hispanic, and Other race/ethnicity, respectively. In multivariable models, compared to White patients, Black, and Hispanic patients were associated with 26% (odds ratio = 1.26, 95% confidence interval = 1.12-1.42) and 29% (odds ratio = 1.29, 95% confidence interval = 1.07-1.55) increased odds of having a treatment delay relative to White patients. Mediation analyses suggested that 49% and 26% the treatment delay among Black and Hispanic patients, respectively, could be removed if an intervention equalized the distribution of academic treatment, education, and insurance status to that of White patients. Treatment at an academic hospital and education were the mediators that explained the largest portion of the racial/ethnic disparity in treatment delay. CONCLUSION: Black and Hispanic MIBC patients experience treatment delays when compared to White patients. Intervening upon patient and system factors could reduce substantial treatment delays. Future research is needed to identify other causes of disparities in treatment delays and may help population health initiatives to address racial/ethnic disparities in clinical settings.

The monetary costs of pediatric upper urinary tract stone disease: Analysis in a contemporary United States cohort.

INTRODUCTION: The prevalence of upper urinary tract stone disease (USD) in the United States is rising among both adults and children. Studies on the contemporary economic burden of USD management in the pediatric population are lacking. OBJECTIVE: To comprehensively analyze the economic impact of USD in a contemporary United States pediatric cohort, and to evaluate drivers of cost. STUDY DESIGN: A retrospective cohort study of pediatric patients (aged 0-17), diagnosed with USD between 2011 and 2018 were identified from PearlDiver-Mariner, an all-payer claims database containing diagnostic, treatment and prescription data provided in all treatment settings. Relevant International Classification of Disease (ICD-9 and ICD-10) and Current Procedural Terminology (CPT) codes were used for identification, and only patients with claims recorded for at least one year before and after entry of a diagnosis code for USD were selected (N = 10,045). Patients were stratified into those undergoing operative vs. non-operative management and for each patient, total 1-year healthcare costs following USD diagnosis, including same day and non-same day encounters, were analyzed. Factors associated with increased spending, as well as economic trends were analyzed. RESULTS: Overall, 8498 (85%) patients were managed non-operatively, while 1547 (15%) underwent a total of 1880 procedural interventions. Total overall cost was $117.1 million, while median annual expenditure was $15.8 million. Proportion of spending for outpatient, inpatient and prescription services was 52%, 32% and 16%, respectively (Table). Outpatient management accounted for 67% of overall spending. The proportion of patients managed non-operatively increased significantly over time, in parallel with spending for non-operative care. Comorbidity burden, treatment year and geographic region were among predictors of costs. DISCUSSION: Our study is the first to report actual insurance reimbursements for pediatric USD management using actual reimbursement data, examined across all treatment settings. We found that majority of the costs were for outpatient services and for non-operative management, with a rising tendency toward non-operative management over time. Regional variation in expenditures was evident. Specific reasons underlying these observed patterns could not directly be discerned from our dataset, but merit further investigation. CONCLUSION: Non-operative and outpatient management for pediatric USD are increasingly common, resulting in parallel shifts in spending. Notably, 52% of overall spending was for outpatient care. These insights into the contemporary economic burden of pediatric USD could provide value in shaping future healthcare policy.

Contemporary Assessment of the Economic Burden of Upper Urinary Tract Stone Disease in the United States: Analysis of One-Year Health Care Costs, 2011-2018.

Background: The U.S. health care landscape has witnessed numerous changes since implementation of the Affordable Care Act coupled with rising prevalence of upper urinary tract stone disease (SD). Data on the economic burden of SD during this period are lacking, providing the objective of our study. Materials and Methods: Adults diagnosed as having SD from 2011 to 2018 were identified from PearlDiver Mariner, a national all-payer database reporting reimbursements and prescription costs for all health care encounters. Patients undergoing operative and nonoperative care were identified. Time trends in annual expenditures were evaluated. Multivariable analysis evaluated determinants of spending. Results: A total of $10 billion were spent on SD management between 2011 and 2018 (median overall annual expenditure = $1.4 billion) among 786,756 patients. Inpatient, prescription, and outpatient costs accounted for 34.7%, 20.7%, and 44.6% of expenditures, respectively. Seventy-eight percent of patients were managed nonoperatively (total cost = $6.9 billion). The average overall cost per encounter was $13,587 ($17,102 for surgical vs $11,174 for nonsurgical care). Expenditures on inpatient care decreased significantly over time, while expenditures on prescriptions and outpatient care increased significantly. On multivariable analysis, a higher Charlson Comorbidity Index (CCI) was associated with higher spending, while associations for age, insurance, and region varied by treatment modality. Conclusions: The economic burden of SD management is substantial, dominated by expenditure on nonoperative management and outpatient care. Expenditures for prescription and outpatient care are rising, with the only consistent predictor of higher spending being CCI. Spending variation according to demographic, clinical, and geographic factors was evident.

Variation in outcomes across surgeons meeting the Leapfrog volume standard for complex oncologic surgery.

BACKGROUND: A large body of evidence supports regionalization of complex oncologic surgery to high-volume surgeons at high-volume hospitals. However, whether there is heterogeneity of outcomes among high-volume surgeons at high-volume hospitals remains unknown. METHODS: Patients who underwent esophagectomy, lung resection, pancreatectomy, or proctectomy for primary cancer were identified within the Medicare 100% Standard Analytic File (2013-2017). Mixed-effects analyses assessed the association between Leapfrog annual volume standards for surgeons (esophagectomy ≥7, lung resection ≥15, pancreatectomy ≥10, proctectomy ≥6) and hospitals (esophagectomy ≥20, lung resection ≥40, pancreatectomy ≥20, proctectomy ≥16) relative to postoperative complications and 90-day mortality. Additional analyses using New York's all-payer Statewide Planning and Research Cooperative System (2004-2015) were performed. RESULTS: Among 112,154 Medicare beneficiaries, high-volume surgeons at high-volume hospitals were associated with lower adjusted odds of complications (esophagectomy: odds ratio [OR], 0.73 [95% CI, 0.61-0.86]; lung resection: OR, 0.88 [95% CI, 0.82-0.94]; pancreatectomy: OR, 0.73 [95% CI, 0.66-0.80]; proctectomy: OR, 0.92 [95% CI, 0.85-0.99]) and 90-day mortality (esophagectomy: OR, 0.60 [95% CI, 0.44-0.76]; lung resection: OR, 0.82 [95% CI, 0.73-0.93]; pancreatectomy: OR, 0.66 [95% CI, 0.56-0.76]; proctectomy: OR, 0.74 [95% CI, 0.65-0.85]). For the average patient at the average high-volume hospital, there was a 2-fold difference in the adjusted complication rate between the best-performing and worst-performing high-volume surgeon for all operations (esophagectomy, 28%-55%; lung resection, 7%-21%; pancreatectomy, 16%-35%; proctectomy, 16%-28%). Wide variation was also present in adjusted 90-day mortality for esophagectomy (3.5%-9.3%). Results from New York's all-payer database were similar. CONCLUSIONS: Even among high-volume surgeons meeting the Leapfrog volume standards, wide variation in postoperative outcomes exists. These findings suggest that volume alone should not be used as a quality indicator, and quality metrics should be continuously evaluated across all surgeons and hospital systems. LAY SUMMARY: Previous studies have demonstrated a surgical volume-outcome relationship for high-risk operations-that is high-volume surgeons and hospitals that perform a specific surgical procedure more frequently have better outcomes for that operation. Although most high-volume surgeons had better outcomes, this study demonstrated that some high-volume surgeons did not have better outcomes. Therefore, volume is an important factor but should not be the only factor considered when assessing the quality of a surgeon and a hospital for cancer surgery.

Variation in Adequate Lymph Node Yield for Gastric, Lung, and Bladder Cancer: Attributable to the Surgeon, Pathologist, or Hospital?

BACKGROUND: The Commission on Cancer recently released quality-of-care measures regarding adequate lymphadenectomy for colon, gastric, lung, and bladder cancer. There is currently little information regarding variation in adequate lymph node yield (ALNY) for gastric, lung, and bladder cancer. METHODS: The New York State Cancer Registry and Statewide Planning and Research Cooperative System were queried for stage I-III gastric, stage I-II lung, and stage II-III bladder cancer resections from 2004 to 2014. Hierarchical models assessed factors associated with ALNY (gastric ≥ 15; lung ≥ 10; bladder ≥ 2). Additionally, the proportions of variation attributable to surgeons, pathologists, and hospitals were estimated among Medicare patients. RESULTS: Among 3716 gastric, 18,328 lung, and 1512 bladder cancer resections, there were low rates of ALNY (gastric = 53%, lung = 36%, bladder = 67%). When comparing 2004-2006 and 2012-2014, there was significant improvement in ALNY for gastric cancer (39% vs. 68%), but more modest improvement for lung (33% vs. 38%) and bladder (65% vs. 71%) cancer. Large provider-level variation existed for each organ system. After controlling for patient-level factors/variation, the majority of variation was attributable to hospitals (gastric: surgeon = 4%, pathologist = 2.8%, hospital = 40%; lung: surgeon = 13.8%, pathologist = 1.5%, hospital = 18.3%) for gastric and lung cancer. For bladder cancer, most of the variation was attributable to pathologists (surgeon = 3.3%, pathologist = 10.5%, hospital = 6.2%). CONCLUSIONS: ALNY rates are low for gastric, lung, and bladder cancer, with only modest improvement over time for lung and bladder cancer. Given that the proportion of variation attributable to the surgeon, pathologist, and hospital is different for each organ system, future quality improvement initiatives should target the underlying causes, which vary by individual organ system.

Changes in Health Insurance Coverage Over Time by Immigration Status Among US Older Adults, 1992-2016.

Importance: Disparities in health insurance coverage by immigration status are well documented; however, there are few data comparing long-term changes in insurance coverage between immigrant and nonimmigrant adults as they age into older adulthood. Objective: To compare longitudinal changes in insurance coverage over 24 years of follow-up between recent immigrant, early immigrant, and nonimmigrant adults in the US. Design, Setting, and Participants: This population-based cohort study used data from the nationally representative Health and Retirement Study. Data were collected biennially from 1992 to 2016. The population included community-dwelling US adults born between 1931 and 1941 and aged 51 to 61 years at baseline. Statistical analysis was performed from February 3, 2017, to January 10, 2020. Exposures: Participants were categorized as nonimmigrants (born in the US), early immigrants (immigrated to the US before the age of 18 years), and recent immigrants (immigrated to the US from the age of 18 years onward). Main Outcomes and Measures: Self-reported data on public, employer, long-term care, and other private insurance were used to define any insurance coverage. Longitudinal changes in insurance coverage were examined over time by immigration status using generalized estimating equations accounting for inverse probability of attrition weights. The association between immigration status and continuous insurance coverage was also evaluated. Results: A total of 9691 participants were included (mean [SD] age, 56.0 [3.2] years; 5111 [52.6%] female). Nonimmigrants composed 90% (n = 8649) of the cohort; early immigrants, 2% (n = 201); and recent immigrants, 8% (n = 841). Insurance coverage increased from 68%, 83%, and 86% of recent immigrant, early immigrant, and nonimmigrant older adults, respectively, in 1992 to 97%, 100%, and 99% in 2016. After accounting for selective attrition, recent immigrants were 15% less likely than nonimmigrants to have any insurance at baseline (risk ratio, 0.85; 95% CI, 0.82-0.88), driven by lower rates of private insurance. However, disparities in insurance decreased incrementally over time and were eliminated, such that insurance coverage rates were similar between groups as participants attained Medicare age eligibility. Furthermore, recent immigrants were less likely than nonimmigrants to be continuously insured (risk ratio, 0.89; 95% CI, 0.85-0.94). Conclusions and Relevance: Among community-dwelling adults who were not age eligible for Medicare, recent immigrants had lower rates of health insurance, but this disparity was eliminated over the 24-year follow-up period because of uptake of public insurance among all participants. Future studies should evaluate policies and health care reforms aimed at reducing disparities among vulnerable populations such as recent immigrants who are not age eligible for Medicare.

Racial/Ethnic Disparities in Atrial Fibrillation Treatment and Outcomes among Dialysis Patients in the United States.

BACKGROUND: Because stroke prevention is a major goal in the management of ESKD hemodialysis patients with atrial fibrillation, investigating racial/ethnic disparities in stroke among such patients is important to those who could benefit from strategies to maximize preventive measures. METHODS: We used the United States Renal Data System to identify ESKD patients who initiated hemodialysis from 2006 to 2013 and then identified those with a subsequent atrial fibrillation diagnosis and Medicare Part A/B/D. Patients were followed for 1 year for all-cause stroke, mortality, prescription medications, and cardiovascular disease procedures. The survival mediational g-formula quantified the percentage of excess strokes attributable to lower use of atrial fibrillation treatments by race/ethnicity. RESULTS: The study included 56,587 ESKD hemodialysis patients with atrial fibrillation. Black, white, Hispanic, and Asian patients accounted for 19%, 69%, 8%, and 3% of the population, respectively. Compared with white patients, black, Hispanic, or Asian patients were more likely to experience stroke (13%, 15%, and 16%, respectively) but less likely to fill a warfarin prescription (10%, 17%, and 28%, respectively). Warfarin prescription was associated with decreased stroke rates. Analyses suggested that equalizing the warfarin distribution to that in the white population would prevent 7%, 10%, and 12% of excess strokes among black, Hispanic, and Asian patients, respectively. We found no racial/ethnic disparities in all-cause mortality or use of cardiovascular disease procedures. CONCLUSIONS: Racial/ethnic disparities in all-cause stroke among hemodialysis patients with atrial fibrillation are partially mediated by lower use of anticoagulants among black, Hispanic, and Asian patients. The reasons for these disparities are unknown, but strategies to maximize stroke prevention in minority hemodialysis populations should be further investigated.

Centralizing Rectal Cancer Surgery: What Is the Impact of Travel on Patients?

BACKGROUND: It is unclear what impact centralizing rectal cancer surgery may have on travel burden for patients. OBJECTIVE: This study aimed to determine the impact of centralizing rectal cancer surgery to high-volume centers on patient travel distance. DESIGN: This is a population-based study. SETTINGS: The New York State Cancer Registry and Statewide Planning and Research Cooperative System were queried for patients with rectal cancer undergoing proctectomy. PATIENTS: Patients with stage I to III rectal cancer who underwent surgical resection between 2004 and 2014 were included. MAIN OUTCOME MEASURES: The outcome of interest was travel distance calculated as the straight-line distance between the centroid of the patient residence zip code and the hospital zip code. Mean distance was compared by using the Student t test. RESULTS: A total of 5860 patients met inclusion criteria. The total number of hospitals performing proctectomies for rectal cancer decreased between 2004 and 2014. The average number of proctectomies performed at high-volume centers (20+ resections/year) increased from 16.6 to 24.4 during this time. The average number of miles traveled by patients was 12.1 miles in 2004, and this increased to 15.4 in 2014. If proctectomies were centralized to high-volume centers, there would be 11 facilities. The mean distance traveled would be 24.5 miles. LIMITATIONS: This study is subject to the limitations of an administrative data set. There are no patient preference or referral data. CONCLUSIONS: The number of hospitals performing rectal cancer resections in New York State is decreasing and volume by center is increasing. There was a statistically significant difference in the mean distance traveled by patients over time. If rectal cancer resections were centralized to high-volume centers, the mean travel distance would increase by 9.5 miles. There would be a 321% increase in the number of patients having to travel 50+ miles for surgery. Any plan for centralization in New York State will require careful planning to avoid placing undue travel burden on patients. See Video Abstract at http://links.lww.com/DCR/B138. CENTRALIZACIÓN DE LA CIRUGÍA DE CÁNCER RECTAL: ¿CUÁL ES EL IMPACTO DEL VIAJE PARA LOS PACIENTES?: No está claro qué impacto puede tener la centralización de la cirugía de cáncer rectal en la carga de viaje para los pacientes.Determinar el impacto de centralizar la cirugía de cáncer rectal en centros de alto volumen sobre la distancia de viaje del paciente.Este es un estudio basado en cohorte poblacional.El Registro de Cáncer del Estado de Nueva York y el Sistema Cooperativo de Planificación e Investigación Estatal fueron consultados para pacientes con cáncer rectal sometidos a proctectomía.Pacientes con cáncer rectal en estadio I-III que se sometieron a resección quirúrgica entre 2004-2014.El resultado de interés fue la distancia de viaje calculada como la distancia en línea recta entre el centroide de la residencia del paciente y el código postal del hospital. La distancia media se comparó mediante la prueba t de Student.Un total de 5,860 pacientes cumplieron los criterios de inclusión. El número total de hospitales que realizaron proctectomías para cáncer rectal disminuyó entre 2004-2014. El número promedio de proctectomías realizadas en centros de alto volumen (más de 20 resecciones/año) aumentó de 16.6 a 24.4 durante este tiempo. El número promedio de millas recorridas por los pacientes fue de 12.1 millas en 2004 y esto aumentó a 15.4 en 2014. Si las proctectomías se centralizaran en centros de alto volumen, habría 11 instalaciones. La distancia media recorrida sería de 24.5 millas.Limitaciones inherentes a un conjunto de datos administrativos. No existen datos sobre preferencia del paciente o sobre referencia de los mismos.El número de hospitales que realizan resecciones de cáncer rectal en Nueva York está disminuyendo y el volumen por centro está aumentando. Hubo una diferencia estadísticamente significativa en la distancia media recorrida por los pacientes a lo largo del tiempo. Si las resecciones por cáncer rectal se centralizaran en centros de gran volumen, la distancia media de viaje aumentaría 9.5 millas. Habría un aumento del 321% en el número de pacientes que tienen que viajar más de 50 millas para la cirugía. Cualquier plan de centralización en Nueva York requerirá una planificación cuidadosa para evitar imponer una carga de viaje excesiva a los pacientes. Consulte Video Resumen en http://links.lww.com/DCR/B138.

A Population-Based Study of 90-Day Hospital Cost and Utilization Associated With Robotic Surgery in Colon and Rectal Cancer.

BACKGROUND: The role of robotic surgery in colorectal cancer remains contentious with most data arising from small, single-institution studies. METHODS: Stage I-III colorectal cancer resections from 2008 to 2014 were identified in New York State. Propensity score-adjusted negative binomial models were used to compare cost and utilization between robotic, laparoscopic, and open resections. RESULTS: A total of 12,218 patients were identified. For colectomy, the robotic-to-open conversion rate was 3%, and the laparoscopic-to-open conversion rate was 13%. For rectal resection, the robotic-to-open conversion rate was 7% and the laparoscopic-to-open conversion rate was 32%. In intention-to-treat analysis, there was no significant difference in cost across the surgical approaches, both in overall and stratified analyses. Both laparoscopic and robotic approaches were associated with decreased 90-d hospital utilization compared with open surgery in intention-to-treat analyses. CONCLUSIONS: Robotic and laparoscopic colorectal cancer resections were not associated with a hospital cost benefit after 90 d compared with open but were associated with decreased hospital utilization. Conversion to open resection was common, and efforts should be made to prevent them. Future research should continue to measure how robotic and laparoscopic approaches can add value to the health care system.

Medicare Annual Wellness Visit association with healthcare quality and costs.

OBJECTIVES: Although use of the Medicare Annual Wellness Visit (AWV) is increasing nationally, it remains unclear whether it can help contain healthcare costs and improve quality. In the context of 2 primary care physician-led accountable care organizations (ACOs), we tested the hypothesis that AWVs can improve healthcare costs and clinical quality. STUDY DESIGN: A retrospective cohort study using propensity score matching and quasi-experimental difference-in-differences regression models comparing the differential changes in cost, emergency department (ED) visits, and hospitalizations for those who received an AWV versus those who did not from before until after the AWV. Logistic regressions were used for quality measures. METHODS: Between 2014 and 2016, we examined the association of an AWV with healthcare costs, ED visits, hospitalizations, and clinical quality measures. The sample included Medicare beneficiaries attributed to providers across 44 primary care clinics participating in 2 ACOs. RESULTS: Among 8917 Medicare beneficiaries, an AWV was associated with significantly reduced spending on hospital acute care and outpatient services. Patients who received an AWV in the index month experienced a 5.7% reduction in adjusted total healthcare costs over the ensuing 11 months, with the greatest effect seen for patients in the highest hierarchical condition category risk quartile. AWVs were not associated with ED visits or hospitalizations. Beneficiaries who had an AWV were also more likely to receive recommended preventive clinical services. CONCLUSIONS: In a setting that prioritizes care coordination and utilization management, AWVs have the potential to improve healthcare quality and reduce cost.

Variation in Hospital-Specific Rates of Suboptimal Lymphadenectomy and Survival in Colon Cancer: Evidence from the National Cancer Data Base.

BACKGROUND: Little is known about between-hospital differences in the rate of suboptimal lymphadenectomy. This study characterizes variation in hospital-specific rates of suboptimal lymphadenectomy and its effect on overall survival in a national hospital-based registry. METHODS: Stage I-III colon cancer patients were identified from the 2003-2012 National Cancer Data Base. Bayesian multilevel logistic regression models were used to assess the impact of patient- and hospital-level factors on hospital-specific rates of suboptimal lymphadenectomy (<12 lymph nodes), and multilevel Cox models were used to estimate the effect of suboptimal lymphadenectomy at the patient (yes vs. no) and hospital level (quartiles of hospital-specific rates) on overall survival. RESULTS: A total of 360,846 patients across 1345 hospitals in the US met the inclusion criteria, of which 25 % had a suboptimal lymphadenectomy. Wide variation was observed in hospital-specific rates of suboptimal lymphadenectomy (range 0-82 %, median 44 %). Older age, male sex, comorbidity score, no insurance, positive margins, lower tumor grade, lower T and N stage, and sigmoid and left colectomy were associated with higher odds of suboptimal lymphadenectomy. Patients treated at lower-volume and non-academic hospitals had higher odds of suboptimal lymphadenectomy. Patient- and hospital-level factors explained 5 % of the between-hospital variability in suboptimal lymphadenectomy, leaving 95 % unexplained. Higher suboptimal lymphadenectomy rates were associated with worse survival (quartile 4 vs. quartile 1: hazard ratio 1.19, 95 % confidence interval 1.16-1.22). CONCLUSION: Large differences in hospital-specific rates of suboptimal lymphadenectomy were observed, and this variation was associated with survival. Quality improvement initiatives targeting hospital-level adherence to the national standard may improve overall survival among resected colon cancer patients.

Patients With Adhesive Small Bowel Obstruction Should Be Primarily Managed by a Surgical Team.

OBJECTIVE: To evaluate the impact of a primary medical versus surgical service on healthcare utilization and outcomes for adhesive small bowel obstruction (SBO) admissions. SUMMARY BACKGROUND DATA: Adhesive-SBO typically requires hospital admission and is associated with high healthcare utilization and costs. Given that most patients are managed nonoperatively, many patients are admitted to medical hospitalists. However, comparisons of outcomes between primary medical and surgical services have been limited to small single-institution studies. METHODS: Unscheduled adhesive-SBO admissions in NY State from 2002 to 2013 were identified using the Statewide Planning and Research Cooperative System. Bivariate and mixed-effects regression analyses were performed assessing factors associated with healthcare utilization and outcomes for SBO admissions. RESULTS: Among 107,603 admissions for adhesive-SBO (78% nonoperative, 22% operative), 43% were primarily managed by a medical attending and 57% were managed by a surgical attending. After controlling for patient, physician, and hospital-level factors, management by a medical service was independently associated with longer length of stay [IRR = 1.39, 95% confidence interval (CI) = 1.24, 1.56], greater inpatient costs (IRR = 1.38, 95% = 1.21, 1.57), and a higher rate of 30-day readmission (OR = 1.32, 95% CI = 1.22, 1.42) following nonoperative management. Similarly, of those managed operatively, management by a medicine service was associated with a delay in time to surgical intervention (IRR = 1.84, 95% CI = 1.69, 2.01), extended length of stay (IRR=1.36, 95% CI = 1.25, 1.49), greater inpatient costs (IRR = 1.38, 95% CI = 1.11, 1.71), and higher rates of 30-day mortality (OR = 1.92, 95% CI = 1.50, 2.47) and 30-day readmission (OR = 1.13, 95% CI = 0.97, 1.32). CONCLUSIONS: This study suggests that management of patients presenting with adhesive-SBO by a primary medical team is associated with higher healthcare utilization and worse perioperative outcomes. Policies favoring primary management by a surgical service may improve outcomes and reduce costs for patients admitted with adhesive-SBO.