Retrospective Analysis of Retroperitoneal-Abdominal-Pelvic Ganglioneuromas: An International Study by the Transatlantic Australasian Retroperitoneal Sarcoma Working Group (TARPSWG).

OBJECTIVE: The Transatlantic Australasian Retroperitoneal Sarcoma Working Group conducted a retrospective study on the disease course and clinical management of ganglioneuromas. BACKGROUND: Ganglioneuromas are rare tumors derived from neural crest cells. Data on these tumors remain limited to case reports and single-institution case series. METHODS: Patients of all ages with pathologically confirmed primary retroperitoneal, intra-abdominal, and pelvic ganglioneuromas between January 1, 2000, and January 1, 2020, were included. We examined demographic, clinicopathologic, and radiologic characteristics, as well as clinical management. RESULTS: Overall, 328 patients from 29 institutions were included. The median age at diagnosis was 37 years with 59.1% of patients being female. Symptomatic presentation comprised 40.9% of cases, and tumors were often located in the extra-adrenal retroperitoneum (67.1%). At baseline, the median maximum tumor diameter was 7.2 cm. One hundred sixteen (35.4%) patients underwent active surveillance, whereas 212 (64.6%) patients underwent resection with 74.5% of operative cases achieving an R0/R1 resection. Serial tumor evaluations showed that malignant transformation to neuroblastoma was rare (0.9%, N=3). Tumors undergoing surveillance had a median follow-up of 1.9 years, with 92.2% of ganglioneuromas stable in size. With a median follow-up of 3.0 years for resected tumors, 84.4% of patients were disease free after resections, whereas recurrences were observed in 4 (1.9%) patients. CONCLUSIONS: Most ganglioneuromas have indolent disease courses and rarely transform to neuroblastoma. Thus, active surveillance may be appropriate for benign and asymptomatic tumors particularly when the risks of surgery outweigh the benefits. For symptomatic or growing tumors, resection may be curative.

Co-Localization of Gastrointestinal Stromal Tumors (GIST) and Peritoneal Mesothelioma: A Case Series.

PURPOSE: Gastrointestinal stromal tumor (GIST) is associated with increased risk of additional cancers. In this study, synchronous GIST, and peritoneal mesothelioma (PM) were characterized to evaluate the relationship between these two cancers. METHODS: A retrospective chart review was conducted for patients diagnosed with both GIST and PM between July 2010 and June 2021. Patient demographics, past tumor history, intraoperative reports, cross-sectional imaging, peritoneal cancer index (PCI) scoring, somatic next-generation sequencing (NGS) analysis, and histology were reviewed. RESULTS: Of 137 patients who underwent primary GIST resection from July 2010 to June 2021, 8 (5.8%) were found to have synchronous PM, and 4 patients (50%) had additional cancers and/or benign tumors. Five (62.5%) were male, and the median age at GIST diagnosis was 57 years (range: 45-76). Seventy-five percent of GISTs originated from the stomach. Of the eight patients, one patient had synchronous malignant mesothelioma (MM), and the remaining had well-differentiated papillary mesothelioma (WDPM), which were primarily located in the region of the primary GIST (89%). The median PCI score was 2 in the WDPM patients. NGS of GIST revealed oncogenic KIT exon 11 (62.5%), PDGFRA D842V (25%), or SDH (12.5%) mutations, while NGS of the MM revealed BAP1 and PBRM1 alterations. CONCLUSIONS: One in 17 GIST patients undergoing resection in this series have PM, which is significantly higher than expected if these two diseases were considered as independent events. Our results indicate that synchronous co-occurrence of GIST and PM is an underrecognized finding, suggesting a possible relationship that deserves further investigation.

National Trends in Readmission Following Inpatient Surgery in the Hospital Readmissions Reduction Program Era.

OBJECTIVE: The aim of this study was to investigate whether the Hospital Readmissions Reduction Program, a national program that introduced financial penalties for high readmission rates for certain medical conditions, had a "spillover" effect on surgical conditions. SUMMARY BACKGROUND DATA: During the past decade, there have been multiple national efforts to improve surgical care. Readmission rates are a key metric for assessing surgical quality. Whether surgical readmission rates have declined, and whether the Hospital Readmissions Reduction Program has had an influence is unclear. METHODS: Using national Medicare data, we identified patients undergoing a range of procedures during the past decade. We examined whether certain procedures that would be targeted by the HRRP had a differential change in readmissions compared to other procedures. We used an interrupted time-series model to examine readmission trends in three time periods: pre-ACA, HRRP implementation, and HRRP penalty. RESULTS: Between 2005 and 2014, 17,423,106 patients underwent the procedures of interest; risk-adjusted rates of readmission across the 8 procedures declined from 12.2% to 8.6%. Pre-ACA rates of readmission were decreasing [-0.060% per quarter (-0.072%, -0.048%), P < 0.001]. During the HRRP implementation period, the rate of decline of readmissions increased [-0.129% (-0.142%, -0.116%), P < 0.001] and continued declining at a similar rate during the penalty period [-0.118% (-0.131%, -0.105%), P < 0.001]. Largest declines in surgical readmissions were seen among the nontargeted procedures. The hospitals with the greatest reductions in medical readmissions also had the greatest drop in surgical readmissions. CONCLUSIONS: Surgical readmission rates have fallen during the past decade and rates of decline have increased during the HRRP period.

Are Staging Computed Tomography (CT) Scans of the Chest Necessary in Pancreatic Adenocarcinoma?

BACKGROUND: There is no consensus on the use of chest imaging in pancreatic ductal adenocarcinoma (PDAC) patients. Among PDAC patients, we examined the use of chest computed tomography (CT) over time and determined whether the use of chest CT led to a survival difference or change in management via identification of indeterminate lung nodules (ILNs). METHODS: Retrospective clinical data was collected for patients diagnosed with PDAC from 1998 to 2014. We examined the proportion of patients undergoing staging chest CT scan and those who had ILN, defined as ≥ 1 well-defined, noncalcified lung nodule(s) ≤ 1 cm in diameter. We determined time to overall survival (OS) using multivariate Cox regression. We also assessed changes in management of PDAC patients who later developed lung metastasis only. RESULTS: Of the 2710 patients diagnosed with PDAC, 632 (23%) had greater than one chest CT. Of those patients, 451 (71%) patients had ILNs, whereas 181 (29%) had no ILNs. There was no difference in median overall survival in patients without ILNs (16.4 [13.6, 19.0] months) versus those with ILN (14.8 [13.6, 15.8] months, P = 0.18). Examining patients who developed isolated lung metastases (3.3%), we found that staging chest CTs did not lead to changes in management of the primary abdominal tumor. CONCLUSIONS: Survival did not differ for PDAC patients with ILNs identified on staging chest CTs compared with those without ILNs. Furthermore, ILN identification did not lead to changes in management of the primary abdominal tumor, questioning the utility of staging chest CTs for PDAC patients.

Unintended Consequences of the 30-Day Mortality Metric: Fact or Fiction.

OBJECTIVE: To assess if an incongruous increase in mortality occurs after postoperative day 30. BACKGROUND: In the current climate of public reporting and pay-for-performance, 30-day mortality after inpatient surgery has become a key metric to assess performance. Whereas the intent is to improve quality, there has been increasing concern that reporting 30-day mortality may influence providers' timing of treatment withdrawal. METHODS: We used national Medicare data to identify beneficiaries who underwent 1 of 19 major surgical procedures. We performed a survival analysis and calculated an adjusted daily hazard rate using all-cause mortality, accounting for patient comorbidities and case-mix. We ran linear regression models to examine discontinuity points around the 30-day mark, and conducted subgroup analyses for hospitals participating in the National Surgical Quality Improvement Program, which focuses on 30-day mortality reporting. RESULTS: We identified 872,968 patients who underwent 1 of 19 surgical procedures of interest; 71,583 of these patients (8.2%) died within 60 days of their index operation. We did not observe any statistically significant increases in mortality in the immediate period after day 30 compared with the immediate period before day 30. In fact, in each model, mortality rates tended to fall in the days after day 30, consistent with a general decreasing risk of death over time. These findings were similar among National Surgical Quality Improvement Program hospitals. CONCLUSIONS: We found no evidence of an increase in postoperative mortality after day 30. As payers move towards incorporating 30-day surgical mortality into pay-for-performance programs, these findings serve as a benchmark for measuring potential future unintended consequences of the metric.

Is Annual Volume Enough? The Role of Experience and Specialization on Inpatient Mortality After Hepatectomy.

OBJECTIVE: To investigate the effect of subspecialty practice and experience on the relationship between annual volume and inpatient mortality after hepatic resection. BACKGROUND: The impact of annual surgical volume on postoperative outcomes has been extensively examined. However, the impact of cumulative surgeon experience and specialty training on this relationship warrants investigation. METHODS: The New York Statewide Planning and Research Cooperative System inpatient database was queried for patients' ≥18 years who underwent wedge hepatectomy or lobectomy from 2000 to 2014. Primary exposures included annual surgeon volume, surgeon experience (early vs late career), and surgical specialization-categorized as general surgery (GS), surgical oncology (SO), and transplant (TS). Primary endpoint was inpatient mortality. Hierarchical logistic regression was performed accounting for correlation at the level of the surgeon and the hospital, and adjusting for patient demographics, comorbidities, presence of cirrhosis, and annual surgical hospital volume. RESULTS: A total of 13,467 cases were analyzed. Overall inpatient mortality was 2.35%. On unadjusted analysis, late career surgeons had a mortality rate of 2.62% versus 1.97% for early career surgeons. GS had a mortality rate of 2.98% compared with 1.68% for SO and 2.67% for TS. Once risk-adjusted, annual volume was associated with reduced mortality only among early-career surgeons (odds ratio 0.82, P = 0.001) and general surgeons (odds ratio 0.65, P = 0.002). No volume effect was seen among late-career or specialty-trained surgeons. CONCLUSIONS: Annual volume alone likely contributes only a partial assessment of the volume-outcome relationship. In patients undergoing hepatic resection, increased annual volume did not confer a mortality benefit on subspecialty surgeons or late career surgeons.

Reappraisal of Staging Laparoscopy for Patients with Pancreatic Adenocarcinoma: A Contemporary Analysis of 1001 Patients.

BACKGROUND: Recent advances in imaging and the increasing use of neoadjuvant therapy puts the contemporary utility of staging laparoscopy for patients with pancreatic adenocarcinoma (PDAC) into question. This study aimed to develop a prognostic score to optimize prevention of an unnecessary laparotomy and minimize the rate for unnecessary laparoscopy. METHODS: Clinicopathologic data were evaluated for all patients undergoing surgical intervention for PDAC between 2001 and 2015, who were stratified into group 1 (2001-2008) and group 2 (2009-2014). RESULTS: The study identified 1001 patients eligible for analysis, 331 (33%) of whom underwent a staging laparoscopy before exploration. An unnecessary laparotomy was prevented for 44.4% of the patients in period 1 and for 24% of the patients in period 2 (p < 0.001). Male gender [odds ratio (OR), 1.8; p < 0.05], preoperative resectability (borderline resectable OR 2.1; p < 0.019; locally advanced OR 7.6; p < 0.001), CA 19-9 levels higher than 394 U/L (OR 3.1; p < 0.001), no neoadjuvant chemotherapy (OR 2.7; p = 0.012), and pancreatic body or tail lesions (OR 1.8; p = 0.063) were predictive of occult metastatic disease. The developed scoring index demonstrated a c-statistic of 0.729. The observed-to-expected ratio for the index at every score level validated the index's model. A score cutoff at 4 was able to detect 76.1% of radiographically occult metastatic disease. CONCLUSION: The rate for unnecessary laparotomy among patients with PDAC has decreased in contemporary times, but unnecessary laparotomy still occurs for 1 in 4 patients. Using our scoring system, a cutoff of 4 allows 76% of radiographically occult metastases to be predicted, thereby selecting high-risk patients for laparoscopic biopsy and potentially avoiding a non-therapeutic laparotomy.

Racial disparities in colorectal cancer outcomes and access to care: a multi-cohort analysis.

Introduction: Non-Hispanic Black (NHB) Americans have a higher incidence of colorectal cancer (CRC) and worse survival than non-Hispanic white (NHW) Americans, but the relative contributions of biological versus access to care remain poorly characterized. This study used two nationwide cohorts in different healthcare contexts to study health system effects on this disparity. Methods: We used data from the Surveillance, Epidemiology, and End Results (SEER) registry as well as the United States Veterans Health Administration (VA) to identify adults diagnosed with colorectal cancer between 2010 and 2020 who identified as non-Hispanic Black (NHB) or non-Hispanic white (NHW). Stratified survival analyses were performed using a primary endpoint of overall survival, and sensitivity analyses were performed using cancer-specific survival. Results: We identified 263,893 CRC patients in the SEER registry (36,662 (14%) NHB; 226,271 (86%) NHW) and 24,375 VA patients (4,860 (20%) NHB; 19,515 (80%) NHW). In the SEER registry, NHB patients had worse OS than NHW patients: median OS of 57 months (95% confidence interval (CI) 55-58) versus 72 months (95% CI 71-73) (hazard ratio (HR) 1.14, 95% CI 1.12-1.15, p = 0.001). In contrast, VA NHB median OS was 65 months (95% CI 62-69) versus NHW 69 months (95% CI 97-71) (HR 1.02, 95% CI 0.98-1.07, p = 0.375). There was significant interaction in the SEER registry between race and Medicare age eligibility (p < 0.001); NHB race had more effect in patients <65 years old (HR 1.44, 95% CI 1.39-1.49, p < 0.001) than in those ≥65 (HR 1.13, 95% CI 1.11-1.15, p < 0.001). In the VA, age stratification was not significant (p = 0.21). Discussion: Racial disparities in CRC survival in the general US population are significantly attenuated in Medicare-aged patients. This pattern is not present in the VA, suggesting that access to care may be an important component of racial disparities in this disease.

Ratio of cesarean sections to total procedures as a marker of district hospital trauma capacity.

BACKGROUND: There are few established metrics to define surgical capacity in resource-limited settings. Previous work hypothesizes that the relative frequency of cesarean sections (CS) at a hospital, expressed as a proportion of total operative procedures (%CS), may serve as a proxy measure of surgical capacity. We attempted to evaluate this hypothesis as it specifically relates to hospital capacity for emergency interventions for injury. METHODS: We conducted a WHO survey of emergency surgical capacity at 40 Rwandan district hospitals in November 2010 and extracted annual operative volume for 2010 from the Ministry of Health centralized statistical system. We dichotomized the 40 hospitals into low and high %CS groups below and above the median proportion of CS performed. We compared low and high %CS groups across self-reported capabilities related to facility characteristics, trauma supplies, procedural capacity, and surgical training using bivariate χ(2) statistics with significance indicated at p ≤ 0.05. We evaluated herniorrhaphy proportion of total procedures (%Hernia) as a representative general surgery procedure in the same manner. RESULTS: High %CS hospitals were less likely to report capability related to blood banking (p = 0.05), amputation (p = 0.04), closed fracture repair (p = 0.04), inhalational anesthesia (p = 0.05), and chest tube insertion (p = 0.05). Availability of reliable electricity was the only measure that showed statistical significance with the %Hernia measure (p = 0.02). CONCLUSIONS: Cesarean section proportion shows some utility as a marker for district hospital injury-care capacity in resource-limited settings.

Patterns of Adjuvant Chemotherapy Use and Association With Survival in Adults 80 Years and Older With Pancreatic Adenocarcinoma.

IMPORTANCE: Patients 80 years and older with pancreatic ductal adenocarcinoma (PDAC) have not consistently received treatments that have established benefits in younger older adults (aged 60-79 years), yet patients 80 years and older are increasingly being offered surgery. Whether adjuvant chemotherapy (AC) provides additional benefit among patients 80 years and older with PDAC following surgery is not well understood. OBJECTIVE: To describe patterns of AC use in patients 80 years and older following surgical resection of PDAC and to compare overall survival between patients who received AC and those who did not. DESIGN, SETTING, AND PARTICIPANTS: Retrospective cohort study among patients 80 years or older diagnosed with PDAC (stage I-III) between 2004 to 2016 who underwent a pancreaticoduodenectomy at hospitals across the US reporting to the National Cancer Database. EXPOSURES: AC vs no AC 90 days following diagnosis of PDAC. MAIN OUTCOMES AND MEASURES: The proportion of patients who received AC was assessed over the study period. Overall survival was compared between patients who received AC and those who did not using Kaplan-Meier estimates and multivariable Cox proportional hazards regression. A landmark analysis was performed to address immortal time bias. A propensity score analysis was performed to address indication bias. Subgroup analyses were conducted in node-negative, margin-negative, clinically complex, node-positive, and margin-positive cohorts. RESULTS: Between 2004 and 2016, 2569 patients 80 years and older (median [IQR] age, 82 [81-84] years; 1427 were women [55.5%]) underwent surgery for PDAC. Of these patients, 1217 (47.4%) received AC. Findings showed an 18.6% (95% CI, 8.0%-29.0%; P = .001) absolute increase in the use of AC among older adults who underwent a pancreaticoduodenectomy comparing rates in 2004 vs 2016. Receipt of AC was associated with a longer median survival (17.2 months; 95% CI, 16.1-19.0) compared with those who did not receive AC (12.7 months; 95% CI, 11.8-13.6). This association was consistent in propensity and subgroup analyses. In multivariable analysis, receipt of AC (hazard ratio [HR], 0.72; 95% CI, 0.65-0.79; P < .001), female sex (HR, 0.88; 95% CI, 0.80-0.96; P < .001), and surgery in the more recent time period (≥2011) (HR, 0.90; 95% CI, 0.82-0.99; P = .02) were associated with a decreased hazard of death. An increased hazard of death was associated with higher pathologic stage (stage II: HR, 1.68; 95% CI, 1.43-1.97; P < .001; stage III: HR, 2.39; 95% CI, 1.88-3.04; P < .001), positive surgical margins (HR, 1.49; 95% CI, 1.34-1.65; P < .001), length of stay greater than median (10 days) (HR, 1.17; 95% CI, 1.07-1.28; P < .001), and receipt of oncologic care at a nonacademic facilities (Community Cancer Program: HR, 1.20; 95% CI, 1.07-1.35; P < .001; Integrated Network Cancer Program: HR, 1.25; 95% CI, 1.07-1.46; P < .001). CONCLUSIONS AND RELEVANCE: In this cohort study, the use of AC among patients who underwent resection for PDAC increased over the study period, yet it still was administered to fewer than 50% of patients. Receipt of AC was associated with a longer median survival.

Changes in Cancer Mortality by Race and Ethnicity Following the Implementation of the Affordable Care Act in California.

Although Affordable Care Act (ACA) implementation has improved cancer outcomes, less is known about how much the improvement applies to different racial and ethnic populations. We examined changes in health insurance coverage and cancer-specific mortality rates by race/ethnicity pre- and post-ACA. We identified newly diagnosed breast (n = 117,738), colorectal (n = 38,334), and cervical cancer (n = 11,109) patients < 65 years in California 2007-2017. Hazard rate ratios (HRR) and 95% confidence intervals (CI) were calculated using multivariable Cox regression to estimate risk of cancer-specific death pre- (2007-2010) and post-ACA (2014-2017) and by race/ethnicity [American Indian/Alaska Natives (AIAN); Asian American; Hispanic; Native Hawaiian or Pacific Islander (NHPI); non-Hispanic Black (NHB); non-Hispanic white (NHW)]. Cancer-specific mortality from colorectal cancer was lower post-ACA among Hispanic (HRR = 0.82, 95% CI = 0.74 to 0.92), NHB (HRR = 0.69, 95% CI = 0.58 to 0.82), and NHW (HRR = 0.90; 95% CI = 0.84 to 0.97) but not Asian American (HRR = 0.95, 95% CI = 0.82 to 1.10) patients. We observed a lower risk of death from cervical cancer post-ACA among NHB women (HRR = 0.68, 95% CI = 0.47 to 0.99). No statistically significant differences in breast cancer-specific mortality were observed for any racial or ethnic group. Cancer-specific mortality decreased following ACA implementation for colorectal and cervical cancers for some racial and ethnic groups in California, suggesting Medicaid expansion is associated with reductions in health inequity.

Patterns of adoption of robotic radical prostatectomy in the United States and England.

OBJECTIVES: To compare patterns of technological adoption of minimally invasive surgery for radical prostatectomy across the United States and England. DATA SOURCES: We examine radical prostatectomy in the United States and England between 2005 and 2017, using de-identified administrative claims data from the OptumLabs Data Warehouse in the United States and the Hospital Episodes Statistics in England. STUDY DESIGN: We conducted a longitudinal analysis of robotic, laparoscopic, and open surgery for radical prostatectomy. We compared the trends of adoption over time within and across countries. Next, we explored whether differential adoption patterns in the two health systems are associated with differences in volumes and patient characteristics. Finally, we explored the relationship between these adoption patterns and length of stay, 30-day readmission, and urology follow-up visits. DATA COLLECTION: Open, laparoscopic, and robotic radical prostatectomies are identified using Office of Population Censuses and Surveys Classification of Interventions and Procedures (OPCS) codes in England and International Classification of Diseases ninth revision (ICD9), ICD10, and Current Procedural Terminology (CPT) codes in the United States. PRINCIPAL FINDINGS: We identified 66,879 radical prostatectomies in England and 79,358 in the United States during 2005-2017. In both countries, open surgery dominates until 2009, where it is overtaken by minimally invasive surgery. The adoption of robotic surgery is faster in the United States. The adoption rates and, as a result, the observed centralization of volume, have been different across countries. In both countries, patients undergoing radical prostatectomies are older and have more comorbidities. Minimally invasive techniques show decreased length of stay and 30-day readmissions compared to open surgery. In the United States, robotic approaches were associated with lower length of stay and readmissions when compared to laparoscopic. CONCLUSIONS: Robotic surgery has become the standard approach for radical proctectomy in the United States and England, showing decreased length of stay and in 30-day readmissions compared to open surgery. Adoption rates and specialization differ across countries, likely a product of differences in cost-containment efforts.

Association of Teaching Status and Mortality After Cancer Surgery.

OBJECTIVE: To examine patient outcomes for nine cancer-specific procedures performed in teaching versus non-teaching hospitals. SUMMARY BACKGROUND DATA: Few contemporary studies have evaluated patient outcomes in teaching versus non-teaching hospitals across a comprehensive set of cancer-specific procedures. METHODS: Use of national Medicare data to compare 30-, 60-, and 90-day mortality rates in teaching and non-teaching hospitals for cancer-specific procedures. Risk-adjusted 30-day, all-cause, postoperative mortality overall and for each specific surgery, as well as overall 60- and 90-day mortality rates, were assessed. RESULTS: The sample consisted of 159,421 total cancer surgeries at 3,151 hospitals. Overall thirty-day mortality rates, adjusted for procedure type, state, and invasiveness of procedure were 1.3% lower at major teaching hospitals (95%CI=-1.6% to -1.1%; p<0.001) relative to non-teaching hospitals. After accounting for patient characteristics, major teaching hospitals continued to demonstrate lower mortality rates compared with non-teaching hospitals (-1.0% difference [95%CI -1.2% to -0.7%]; p<0.001). Further adjustment for surgical volume as a mediator reduced the difference to -0.7% (95%CI -0.9% to -0.4%, p<0.001). Cancer surgeries for four of the nine disease sites (bladder, lung, colorectal and ovarian) followed this overall trend. Sixty- and ninety-day overall mortality rates, adjusted for procedure type, state, and invasiveness of procedure showed that major teaching hospitals had a 1.7% (95%CI -2.1% to -1.4%; p<0.001) and 2.0% (95%CI -2.4 to -1.6%, p<0.001) lower mortality relative to non-teaching hospitals. These trends persisted after adjusting for patient characteristics. CONCLUSIONS: Among cancer-specific procedures for Medicare beneficiaries, major teaching hospital status was associated with lower 30-, 60-, and 90-day mortality rates overall and across four of the nine cancer types.

Association of race, ethnicity, and community-level factors with COVID-19 cases and deaths across U.S. counties.

The United States currently has one of the highest numbers of cumulative COVID-19 cases globally, and Latino and Black communities have been disproportionately affected. Understanding the community-level factors that contribute to disparities in COVID-19 case and death rates is critical to developing public health and policy strategies. We performed a cross-sectional analysis of U.S. counties and found that a 10% point increase in the Black population was associated with 324.7 additional COVID-19 cases per 100,000 population and 14.5 additional COVID-19 deaths per 100,000. In addition, we found that a 10% point increase in the Latino population was associated with 293.5 additional COVID-19 cases per 100,000 and 7.6 additional COVID-19 deaths per 100,000. Independent predictors of higher COVID-19 case rates included average household size, the share of individuals with less than a high school diploma, and the percentage of foreign-born non-citizens. In addition, average household size, the share of individuals with less than a high school diploma, and the proportion of workers that commute using public transportation independently predicted higher COVID-19 death rates within a community. After adjustment for these variables, the association between the Latino population and COVID-19 cases and deaths was attenuated while the association between the Black population and COVID-19 cases and deaths largely persisted. Policy efforts must seek to address the drivers identified in this study in order to mitigate disparities in COVID-19 cases and deaths across minority communities.