Treatment at a high-volume academic research program mitigates racial disparities in pancreatic adenocarcinoma

Background In the present study, we aimed to describe, at the population level, patterns of adjuvant treatment use after curative-intent resection for pancreatic adenocarcinoma (pcc) and to identify independent predictors of adjuvant treatment use.Methods In this observational cohort study, patients undergoing pcc resection in the province of Ontario (population 13 million) during 2005–2010 were identified using the provincial cancer registry and were linked to administrative databases that include all treatments received and outcomes experienced in the province. Patients were defined as having received chemotherapy (ctx), chemoradiation (crt), or observation (obs). Clinicopathologic factors associated with the use of ctx, crt, or obs were identified by chi-square test. Logistic regression analyses were used to identify independent predictors of adjuvant treatment versus obs, and ctx versus crt.Results Of the 397 patients included, 75.3% received adjuvant treatment (27.2% crt, 48.1% ctx) and 24.7% received obs. Within a single-payer health care system with universal coverage of costs for ctx and crt, substantial variation by geographic region was observed. Although the likelihood of receiving adjuvant treatment increased from 2005 to 2010 (p = 0.002), multivariate analysis revealed widespread variation between the treating hospitals (p = 0.001), and even between high-volume hepatopancreatobiliary hospitals (p = 0.0006). Younger age, positive lymph nodes, and positive surgical resection margins predicted an increased likelihood of receiving adjuvant treatment. Among patients receiving adjuvant treatment, positive margins and a low comorbidity burden were associated with crt compared with ctx.Conclusions Interinstitutional medical practice variation contributes significantly to differential patterns in the rate of adjuvant treatment for pcc. Whether such variation is warranted or unwarranted requires further investigation.

Download Full-text

Impact of Treatment Facility Chracteristics on Patient (Pt) Outcomes in Acute Myeloid Leukemia (AML) Using Data from the National Cancer Database (NCDB)

Blood ◽

10.1182/blood-2020-139034 ◽

2020 ◽

Vol 136 (Supplement 1) ◽

pp. 22-22

Author(s):

Allison Taylor ◽

Kimberley Doucette ◽

Bryan Chan ◽

Xiaoyang Ma ◽

Jaeil Ahn ◽

...

Keyword(s):

Overall Survival ◽

Hazard Analysis ◽

Treatment Options ◽

Academic Research ◽

High Volume ◽

Rank Test ◽

Rural Residence ◽

Integrated Network ◽

Kaplan Meier ◽

Significant Difference

Introduction The literature suggests a widespread reduction in the availability and accessibility of newer treatment options among marginalized groups in AML. Studies from large national databases point to lower socio-economic status, Hispanic and African American race, Medicare or no insurance, being unmarried, treatment at non-academic centers, and rural residence as negatively impacting overall survival (OS) and rates of chemotherapy utilization in AML patients (Patel et al. 2015, Jaco et al. 2017, Bhatt et al. 2018, Master et al. 2016). We hypothesized that facility affiliation and pt volume would also have important effects on time to treatment (TTT) and OS in AML, even when these socioeconomic disparities were accounted for. Methods For this retrospective analysis, we used NCDB data that included 124,988 pts over the age of 18 with AML between the years 2004-2016. Variables analyzed included facility types described as community cancer programs (CP), comprehensive community cancer programs (CCP), academic/research center cancer programs (AC) and integrated network cancer programs (IN), and volume of facilities defined as high volume (HV) and low volume (LV). HV facilities had case volumes of ≥ 99th percentile and all other facilities were classified as LV. Multivariate analyses (MVA) included demographic and socioeconomic covariables. We used Cox proportional hazard analysis for both TTT and OS MVA. The Kaplan-Meier method was used to estimate median TTT and OS, and the log rank test used to compare TTT and OS across predictor variables. Results The median age of AML patients was 63 yrs (range 18-90) with 54% males, and 86% Caucasian. Five percent of patients were treated at CP, 30% at CCP, 44% at AC, and 10% at IN. 21% at HV facilities and 79% at LV facilities. Median TTT in days at CP facilities was 7, compared to 5 days in CCP and AC facilities versus 4 days at IN (p<0.0001). TTT was 5 days at HV facilities versus 4 days at LV facilities (p<0.0001). Kaplan-Meier curves showed that TTT was similar between HV and LV facilities(figure 1). The median OS was 3.25 months in CP compared to 4.34 months at CCP, 5.06 months at IN and 9.53 months at AC (p<0.0001). For facility volume, the median OS was 13.11 months in HV facilities compared to 6.93 months in LV facilities (p<0.0001). When sex, race, age, Hispanic Origin, education, urban/rural residence, Charlson-Deyo Comorbidity score and Great Circle Distance were adjusted for in MVA (table 1), the OS was higher in AC versus CP facilities (hazard ratio [HR] of 0.90 (0.87-0.93, p<0.0001), and there was no statistically significant difference with comparison of other facility types to CP. Similarly, there was a lower OS at LV versus HV facilities with a HR of 1.14 (1.12-1.16, p<0.0001). CCP facilities had a shorter TTT compared to CP with a HR of 1.21 (1.17-1.26, p<0.0001). AC had a shorter TTT than CP with a HR of 1.17 (1.13-1.22, p<0.0001), and IN had a shorter TTT compared to CP with a HR of 1.29 (1.24-1.34, p<0.0001). Additionally, TTT in the MVA for facility volume was shorter in LV facilities compared to HV facilities with HR of 1.05 (1.04-1.07, p<0.0001) [table 1]. Conclusion When adjusting for various socioeconomic factors, we found that TTT was longest in CP compared to CCP, AC, and IN. Treatment at a LV facility resulted in a decreased overall survival. LV facilities may be less familiar with treatment regimens for AML, less likely to use novel treatment options, and be less familiar with the disease. We showed that treatment at an AC compared to CP, CCP and IN facilities improved survival. Given poor outcomes for AML, these results show the importance of going to AC and HV facilities with more experience in treating AML for improved outcomes. Disclosures Lai: Astellas: Speakers Bureau; Jazz: Speakers Bureau; Abbvie: Consultancy; Agios: Consultancy; Macrogenics: Consultancy.

Download Full-text

A Novel Convolutional Neural Network for Classifying Indian Coins by Denomination

10.31224/osf.io/znxrg ◽

2021 ◽

Author(s):

Yash Chauhan ◽

Prateek Singh

Keyword(s):

Neural Network ◽

Convolutional Neural Network ◽

Academic Research ◽

High Volume ◽

Classification Problem ◽

General Purpose ◽

Training Data ◽

Slot Machines ◽

Validation Data ◽

Starting Point

Coins recognition systems have humungous applications from vending and slot machines to banking and management firms which directly translate to a high volume of research regarding the development of methods for such classification. In recent years, academic research has shifted towards a computer vision approach for sorting coins due to the advancement in the field of deep learning. However, most of the documented work utilizes what is known as ‘Transfer Learning’ in which we reuse a pre-trained model of a fixed architecture as a starting point for our training. While such an approach saves us a lot of time and effort, the generic nature of the pre-trained model can often become a bottleneck for performance on a specialized problem such as coin classification. This study develops a convolutional neural network (CNN) model from scratch and tests it against a widely-used general-purpose architecture known as Googlenet. We have shown in this study by comparing the performance of our model with that of Googlenet (documented in various previous studies) that a more straightforward and specialized architecture is more optimal than a more complex general architecture for the coin classification problem. The model developed in this study is trained and tested on 720 and 180 images of Indian coins of different denominations, respectively. The final accuracy gained by the model is 91.62% on the training data, while the accuracy is 90.55% on the validation data.

Download Full-text

Abstract 17027: Racial Disparities in Transcatheter Aortic Valve Replacement Access: Geographic Analysis Suggesting Disparate Access to Care

Circulation ◽

10.1161/circ.138.suppl_1.17027 ◽

2018 ◽

Vol 138 (Suppl_1) ◽

Author(s):

Nidal Ganim ◽

Dominique J Monlezun ◽

Enrique D Garcia-Sayan ◽

Prakash Balan

Keyword(s):

Aortic Valve ◽

Aortic Valve Replacement ◽

Racial Disparities ◽

Valve Replacement ◽

Transcatheter Aortic Valve Replacement ◽

Private Insurance ◽

High Volume ◽

Racial Minorities ◽

Transcatheter Aortic Valve ◽

Excess Costs

Background: Transcatheter aortic valve replacement (TAVR) has ample randomized trial evidence that it can reduce mortality and cost for patients with aortic stenosis. Yet racial disparities in procedure access are poorly understood. Methods: This case-control prospectively enrolled TAVR subjects at a single high-volume quaternary academic medical center in Houston, Texas, USA, from 11/8/11-3/28/18. Neural network machine learning-supported binomial probability testing was conducted comparing the Houston population versus the center’s TAVR rates by race, with mortality and cost extrapolations. The IOM definition of health inequities was applied using the rank and replace method for counterfactual comparison (matching subjects by insurance and Society of Thoracic Surgery [STS] risk score for TAVR eligibility). Results: Compared to the Houston population, TAVR subjects (N=1641) were significantly more likely to be Caucasians (51.93% vs 77.26%), and less likely to be African Americans (14.80% vs 6.02%), Hispanics (23.63% vs 15.02%), or other races (9.50% vs 1.70%), all p<0.001. Among TAVR subjects with private insurance, the large majority were Caucasian (832, 85.60%), with the minority being African American (34, 3.50%), Hispanic (96, 9.88%), and other (10, 37.04%) (private insurance by Caucasian versus non-Caucasian, p<0.001). Based on TAVR mortality and cost savings in the PARNTER trial, access disparities for racial minorities over 5 years may result in 858 excess deaths, $130,000 per patient excess costs, and $111.5 million excess costs per the overall sample of eligible presenting Houston subjects. The predicted versus actual racial distribution of TAVR for each minority group matched to Caucasians by insurance and STS score was significantly greater than the actual (each group comparison to Caucasians, p<0.001). Conclusion: Multi-year data from our high-volume center suggest Houston racial minorities are less likely to undergo TAVR, potentially translating into a growing number of preventable excess early deaths and costs as disease incidence increases. Additional studies are underway to determine and reduce the degree of preventable race-related disparities independent of known access predictors.

Download Full-text