Abstract
Objective:
To evaluate if receipt of complex cancer surgery at high quality hospitals is associated with a reduction in disparities between individuals living in the most and least deprived neighborhoods.
Background:
The association between social risk factors and worse surgical outcomes for patients undergoing high-risk cancer operations is well documented. To what extent neighborhood socioeconomic deprivation as an isolated social risk factor known to be associated with worse outcomes can be mitigated by hospital quality is less known.
Methods:
Using 100% Medicare fee-for-service claims, we analyzed data on 212,962 Medicare beneficiaries >age 65 undergoing liver resection, rectal resection, lung resection, esophagectomy and pancreaticoduodenectomy for cancer between 2014 and 2018. Clinical risk-adjusted 30-day post-operative mortality rates were used to stratify hospitals into quintiles of quality. Beneficiaries were stratified into quintiles based on census tract Area Deprivation Index. The association of hospital quality and neighborhood deprivation with 30-day mortality was assessed using logistic regression.
Results:
There were 212,962 patients in the cohort including 109,419(51.4%) men with mean (SD) age of 73.8(5.9) years old. At low-quality hospitals, patients living in the most deprived areas had significantly higher risk-adjusted mortality than those from the least deprived areas for all procedures; esophagectomy: 22.3% vs. 20.7%; P<0.003, liver resection 19.3% vs. 16.4%; P<0.001, pancreatic resection 15.9% vs. 12.9%; P<0.001, lung resection 8.3% vs. 7.8%; P<0.001, rectal resection 8.8% vs. 8.1%; P<0.001. Surgery at a high-quality hospitals was associated with no significant differences in mortality between individuals living in the most compared to least deprived neighborhoods for esophagectomy, rectal resection, liver resection and pancreatectomy. For example, the adjusted odds of mortality between individuals living in the most deprived compared to least deprived neighborhoods following esophagectomy at low quality hospitals (OR 1.22; 95% CI 1.14–1.31; P<0.001) was higher than at high quality hospitals (OR 0.98, 95%CI 0.94–1.02; P=0.03).
Conclusion and Relevance:
Receipt of complex cancer surgery at a high-quality hospital was associated with no significant differences in mortality between individuals living in the most deprived neighborhoods compared to least deprived. Initiatives to increase access referrals to high quality hospitals for patients from high deprivation levels may improve outcomes and contribute to mitigating disparities.
Mini Abstract:
We used Medicare claims 2014–2018 to assess if improved hospital quality was associated with significant change in the differences in mortality between individuals living in the most and least deprived neighborhoods. We found that surgery at a high-quality hospital was associated with no significant differences in mortality between individuals living in the most compared to least deprived neighborhoods for esophagectomy, rectal resection, liver resection and pancreatectomy.
Introduction:
The socioeconomic status of a patient’s neighborhood has a significant influence on the receipt of high-quality cancer care, including surgery [1,2]. An individual’s neighborhood context is a key determinant of health and reflects social capital, economic positioning, community relationships and built environments that contribute to health and healthcare delivery [3–5]. Patients who require surgical resection for cancer and live in socioeconomically disadvantaged or deprived neighborhoods are less likely to receive surgical intervention, undergo surgery at a high-volume hospital, and achieve optimal postoperative outcomes [6–10]. Mitigation of these disparities and achieving equity in complex surgical oncologic care has become a priority for the American College of Surgeons, National Institutes of Health, and the American Society of Clinical Oncology [11,12]. Furthermore, to date there has been significant policy and health-system debates about the improving access of medically complex patients requiring cancer surgery to high quality hospitals, yet it remains unclear if socially-at-risk patients may also benefit from these efforts [13–15].
Despite studies demonstrating that higher neighborhood deprivation is associated with worse surgical outcomes, it is unclear if receiving care at hospitals with higher quality may be associated with changes in this disparity. On the one hand, patients living in areas with more deprivation have more comorbidities, complex social needs, and potential barriers to care which may contribute to higher post-operative mortality, complications, and readmissions irrespective of care delivered within a given hospital [16–18]. However, it could be that receiving surgical care at high quality hospitals by individuals from deprived neighborhoods increases their access to with more resources leading to lower failure to rescue rates and higher specialty care leading to improved outcomes [19]. Despite studies demonstrating that hospital level factors, such as operative volume, medical specialist services, and advanced technology, are associated with higher quality hospitals for high-risk cancer surgery, it remains unclear how much a hospital’s quality can mitigate the known disparities based on an individual’s neighborhood [20,21].
The purpose of this analysis was to assess the combined effect of neighborhood socioeconomic deprivation and hospital quality on mortality following high risk cancer operations in a nationally representative cohort. In this study, we used 100% Medicare claims to evaluate if receipt of rectal resection, liver resection, lung resection, pancreaticoduodenectomy, and esophagectomy at high quality centers mitigated the disparity in surgical outcomes between individuals living in the most and least deprived neighborhoods.
Methods:
Study Design and Participants
We used data from the Medicare Provider Analysis and Review (MED-PAR) file, which included 100% claims for beneficiaries undergoing surgery from 2014–2018. We included Medicare fee for service beneficiaries between the ages of 65 and 99 years old with complete payment data. All beneficiaries included were enrolled in Medicare for at least 3 months before and 12 months after surgery. All patient data including age, gender, race and ethnicity and geographic residence including census tract information, co-morbidities and geographic information were extracted from the MedPAR file. The American Hospital Association Annual Survey, which includes hospital characteristics including size, surgical volume, teaching status, ownership and nurse to patient ratio, was linked to the MedPAR files using unique hospital identifiers.
Diagnosis and Procedure Identification
Patients undergoing high-risk cancer operations including rectal resection, liver resection, lung resection, pancreaticoduodenectomy, and esophagectomy for cancer were included. These represent operations that have been included in policies centered on volume standards and assessments on centralizing cancer care to improve surgical quality[13–15]. Each procedure was identified using the International Classification of Disease Ninth or Tenth Revision, Clinical Modification (ICD-9 or ICD-10) codes. Each individual cancer diagnosis was collected from ICD-9 or ICD-10 diagnosis codes. This study was deemed exempt by the University of Michigan Institutional Review Board.
Neighborhood Deprivation:
Area Deprivation Index (ADI) is a composition measure, which includes data in the domains of education, employment, housing quality and poverty, that determines the degree of socioeconomic disadvantage in an individual’s neighborhood[22,23]. The index is a relative measure with scores ranging from 1 to 100. An ADI score of 1 represents the lowest level of disadvantage in the nation and a score of 100 represents the highest level of disadvantage. For each beneficiary, we extracted their 9-digit ZIP code and crossed walked to the census tract level and calculated their ADI. We then created neighborhood quintiles based on overall ADI scores with the least deprived neighborhoods (scores 1–20) and most deprived (scores 81–100).
Hospital Quality:
We used rates of mortality within 30 days of the index operation as the primary outcome to identify high- and low-quality hospitals. Hospitals were identified in rank order by rates of 30-day mortality rates and divided into quartiles. Hospitals in the quintile with highest mortality rates were designated as “Low Quality”. “High Quality” hospitals included those within the quintile with the lowest mortality rates.
Outcomes:
The primary outcome of interest was risk-adjusted, 30-day mortality defined as death occurring within 30 days of the index procedure. We defined disparities as significant differences in risk-adjusted 30-day mortality between beneficiaries living in the most deprived and least deprived neighborhood within each quintile of hospital quality.
Statistical Analysis:
The goal of our analysis was to understand the relationship between neighborhood deprivation and hospital quality on post-operative mortality. First, we compared the baseline characteristics of beneficiaries in each quintile of neighborhood deprivation using t-test and Chi-squared test as appropriate. Of note, race and ethnicity data is captured through Social Security data with categories designated by the Office of Management and Budget. In this analysis, we used American Indian/Native American, Asian and Pacific Islander, Hispanic, non-Hispanic Black, non-Hispanic White racial and ethnic categories. The patient characteristics included age, gender, comorbidities, operation performed, admission type (elective versus unplanned) and discharge location. We used a total of 27 Elixhauser comorbidities as done in prior analyses of surgical cohorts using Medicare claims data [24] Hospital characteristics compared included hospital size, ownership, hospital region, median surgical volume per hospital, teaching hospital status and patient to nurse ratio.
Next, 30 day mortality rates were risk-adjusted using a multivariable regression model that included patient age, gender, Elixhauser comorbidities, admission type (elective vs. unplanned), procedure and all hospital characteristics. Additionally, year was included as a categorical variable to account for secular trends. We applied robust standard errors to account for clustering within hospitals.
All reported P values were 2 sided and a value of less than .05 was used as threshold for significance. All statistical analyses were completed with Stata, version 14 (Stata Corp).
Results:
Patient and Hospital Characteristics
This study included 212,962 patients with mean (SD) age 73.8(5.9). There were significant differences between patients depending on the level of their neighborhood deprivation (Table 1). Patients living in the most deprived neighborhoods compared to the least were more likely to be non-Hispanic Black (17.1% vs. 4.4%, P<0.001), Hispanic/Latino (1.5% vs. 1.3%, P<0.001), or Native American (0.7% vs. 0.1%, P<0.001). They also were more likely to have more comorbidities including hypertension (73.3% vs. 65.9%, P<0.001), chronic pulmonary disorder (42.5% vs. 29.1%) and diabetes (30.6% vs. 22.6%, p<0.001). Patients living in the most deprived neighborhoods had higher rates of unplanned admissions (11.5% vs. 7.5%, P<0.001) and lower rates of being discharged home (78.4% vs. 84.4%, P<0.001).
Table 1.
Patient characteristics by level of Neighborhood Deprivation.
| Total | Least Deprivation | Below Average Deprivation | Average Deprivation | Above Average Deprivation | Most Deprivation | P-Value | |
|---|---|---|---|---|---|---|---|
|
| |||||||
| No of. Patients | 212,962 | 45,575 | 48,886 | 47,125 | 40,276 | 31,100 | |
|
| |||||||
| Age, mean (SD) | 73.8(5.9) | 74.4(6.0) | 73.9(5.9) | 73.7(5.8) | 73.5(5.8) | 73.3(5.8) | <0.001 |
|
| |||||||
| Men, n (%) | 109,419(51.4%) | 22,756(49.9%) | 25,084(51.3%) | 24,549(52.1%) | 20,981(52.1%) | 16,049(51.6%) | <0.001 |
|
| |||||||
| Race | |||||||
| White | 185,202(87.0%) | 38,207(83.8%) | 43,741(89.5%) | 42,854(90.9%) | 35,779(88.8%) | 24,621(79.2%) | <0.001 |
| Black | 15,701(7.45%) | 2,006(4.4%) | 2,533(5.2%) | 2,638(5.6%) | 3,200(7.9%) | 5,3242(17.1%) | <0.001 |
| Asian | 4,377(2.1%) | 2,612(5.7%) | 948(1.9%) | 421(0.9%) | 236(0.6%) | 160(0.5%) | <0.001 |
| Hispanic | 2,758(1.3%) | 631(1.4%) | 681(1.4%) | 512(1.1%) | 458(1.1%) | 476(1.5%) | <0.001 |
| Native American | 735(0.3%) | 24(0.1%) | 112(0.2%) | 160(0.3%) | 212(0.5%) | 227(0.7%) | <0.001 |
| Other | 4,189(2.0%) | 2,095(4.6%) | 871(1.8%) | 540(1.1%) | 391(1.0%) | 292(0.9%) | <0.001 |
|
| |||||||
| Elixhauser Comorbidities | |||||||
| 0 | 11,998(5.6%) | 3,182(7.0%) | 2,910(6.0%) | 2,571(5.5%) | 1,967(4.9%) | 1,368(4.4%) | <0.001 |
| 1 | 32,846 (15.4%) | 8,156(17.9%) | 7,866(16.1%) | 7,092(15.0%) | 5,722(14.2%) | 4,010(12.9%) | <0.001 |
| >2 | 168,118(78.9%) | 34,237(75.1%) | 38,110(78.0%) | 37,462(79.5%) | 32,587(80.9%) | 25,722(82.7%) | <0.001 |
|
| |||||||
| Discharge Location | |||||||
| Home | 173,860(81.6%) | 38,452(84.4%) | 40,192(82.2%) | 38,274(81.2%) | 32,550(80.8%) | 24,392(78.4%) | <0.001 |
| Skilled Nursing Facility | 31,718(14.9%) | 5,935(13.0%) | 7,288(14.9%) | 7,170(15.2%) | 6,139(15.2%) | 5,186(16.7%) | <0.001 |
| Transferred | 878(0.4%) | 183(0.4%) | 168(0.3%) | 197 (0.4%) | 170(0.4%) | 160(0.5%) | 0.008 |
| Other | 5,479 (2.6%) | 849(1.9%) | 1,022(2.1%) | 1,235(2.6%) | 1,212(3.0%) | 1,161(3.7%) | <0.001 |
| Hospice | 1,027 (0.5%) | 156(0.3%) | 216(0.4%) | 249(0.5%) | 205(0.5%) | 201(0.6%) | <0.001” |
|
| |||||||
| Admission Type | |||||||
| Elective | 193,889(91.0%) | 42,155(92.5%) | 44,816(91.7%) | 43,009(91.3%) | 36,389(90.3%) | 27,520(88.5%) | <0.001 |
| Unplanned | 19,073(9.0%) | 3,420(7.5%) | 4,070(8.3%) | 4,116(8.7%) | 3,887(9.7%) | 3,580(11.5%) | <0.001 |
|
| |||||||
| Type of Operation | |||||||
| Esophagectomy | 9,981(4.7%) | 2,022(4.4%) | 2,482(5.1%) | 2,329(4.9%) | 1,824(4.5%) | 1,324(4.3%) | <0.001 |
| Liver Resection | 14,011(6.6%) | 3,585(7.9%) | 3,329(6.8%) | 2,767(5.9%) | 2,425(6.0%) | 1,905(6.1%) | <0.001 |
| Lung Resection | 122,034(57.3%) | 26,040(57.1%) | 28,102(57.5%) | 27,309(58.0%) | 23,081(57.3%) | 17,502(56.3%) | <0.001 |
| Pancreatic Resection | 24,455(11.5%) | 6,002(13.2%) | 5,831(11.9%) | 5,327(11.3%) | 4,171(10.4%) | 3,124(10.0%) | <0.001 |
| Rectal Resection | 42,481(19.9%) | 7,926(17.4%) | 9,142(18.7%) | 9,393(19.9%) | 8,775(21.8%) | 7,245(23.3%) | <0.001 |
There were significant differences in hospital characteristics related to hospital quality for each of the four cancer operations evaluated (Table 2). For each of the operations, low and below average quality hospitals were more likely to be for-profit and located in the South. High and above average quality hospitals had higher surgical volumes and nurse ratios.
Table 2.
Risk Adjusted 30-Day Mortality among the Most and Least Deprived Medicare Beneficiaries at all Hospital Quality Levels.
| Risk-Adjusted 30-Day Mortality Rates % (95%CI) | |||||
|---|---|---|---|---|---|
| Least Deprived | Most Deprived | Absolute Difference | Odds Ratio (95%CI) | P-value | |
| Esophagectomy | |||||
| High Quality Hospitals | 3.84(3.33–4.35) | 3.50(3.05–3.96) | −0.34(−1.30–0.63) | 0.98(0.94–1.02) | 0.3 |
| Above Average Quality Hospitals | 6.21(5.68–6.74) | 6.21(5.71–6.72) | 0.00(−1.03–1.04) | 0.97(0.92–1.02) | 0.22 |
| Average Quality Hospitals | 8.54(7.97–9.12) | 9.15(8.51–9.79) | 0.61(−0.60–1.83) | 1.07(1.03–1.11) | <0.001 |
| Below Average Quality Hospitals | 11.91(11.03–12.79) | 11.22(10.38–12.07) | −0.69(−2.42–1.04) | 0.98(0.93–1.04) | 0.52 |
| Low Quality Hospitals | 20.67(18.95–22.39) | 22.31(20.30–24.33) | 1.64(−2.09–5.38) | 1.22(1.14–1.31) | <0.001 |
| Liver Resection | |||||
| High Quality Hospitals | 1.79(1.49–2.09) | 2.22(1.91–2.52) | 0.43(−0.18–1.103) | 1.03(0.96–1.10) | 0.39 |
| Above Average Quality Hospitals | 3.46(3.17–3.75) | 3.50(3.19–3.81) | 0.04(−0.56–0.64) | 1.07(1.02–1.13) | 0.003 |
| Average Quality Hospitals | 5.14(4.78–5.50) | 5.21(4.81–5.60) | 0.07(−0.68–0.83) | 1.09(1.03–1.15) | 0.003 |
| Below Average Quality Hospitals | 8.05(7.54–8.56) | 8.54(7.90–9.19) | 0.49(−0.66–1.65) | 1.14(1.09–1.18) | <0.001 |
| Low Quality Hospitals | 16.39(15.04–17.74) | 19.25(17.62–20.88) | 2.86(−0.13–5.84) | 1.42(1.30–1.54) | <0.001 |
| Lung Resection | |||||
| High Quality Hospitals | 1.72(1.63–1.82) | 1.94(1.85–2.03) | 0.22(0.03–0.40) | 1.08(1.06–1.10) | <0.001 |
| Above Average Quality Hospitals | 2.36(2.28–2.44) | 2.69(2.60–2.78) | 0.33(0.16–0.50) | 1.16(1.13–1.18) | <0.001 |
| Average Quality Hospitals | 3.23(3.13–3.33) | 3.53(3.41–3.64) | 0.30(0.09–0.51) | 1.12(1.10–1.14) | <0.001 |
| Below Average Quality Hospitals | 4.56(4.41–4.70) | 4.98(4.82–5.13) | 0.42(0.12–0.72) | 1.12(1.10–1.14) | <0.001 |
| Low Quality Hospitals | 7.79(7.47–8.10) | 8.31(8.01–8.62) | 0.52(−0.1–1.14) | 1.07(1.04–1.09) | <0.001 |
| Pancreatic Resection | |||||
| High Quality Hospitals | 2.88(2.60–3.17) | 3.08(2.80–3.35) | 0.19(−0.37–0.75) | 1.10(1.05–1.14) | <0.001 |
| Above Average Quality Hospitals | 4.86(4.58–5.14) | 4.77(4.51–5.03) | −0.09(−0.63–0.45) | 0.98(0.96–1.01) | 0.26 |
| Average Quality Hospitals | 6.14(5.82–6.46) | 6.27(5.92–6.63) | 0.13(−0.54–0.81) | 1.04(1.01–1.08) | 0.01 |
| Below Average Quality Hospitals | 8.65(8.15–9.15) | 9.37(8.82–9.92) | 0.72(−0.33–1.77) | 1.09(1.05–1.14) | <0.001 |
| Low Quality Hospitals | 12.93(12.02–13.85) | 15.87(14.76–16.97) | 2.93(0.91–4.95) | 1.26(1.21–1.32) | <0.001 |
| Rectal Resection | |||||
| High Quality Hospitals | 1.51(1.33–1.70) | 1.41(1.30–1.52) | −0.10(−0.39–0.19) | 0.92(0.85–0.98) | 0.02 |
| Above Average Quality Hospitals | 2.38(2.23–2.53) | 2.35(2.21–2.49) | −0.02(−0.32–0.27) | 1.04(1.01–1.06) | 0.001 |
| Average Quality Hospitals | 3.30(3.13–3.47) | 3.58(3.40–3.76) | 0.28(−0.07–0.63) | 1.08(1.06–1.11) | <0.001 |
| Below Average Quality Hospitals | 4.68(4.46–4.90) | 4.93(4.69–5.16) | 0.25(−0.21–0.71) | 1.07(1.04–1.09) | <0.001 |
| Low Quality Hospitals | 8.16(7.69–8.62) | 8.83(8.33–9.33) | 0.67(−0.29–1.64) | 1.06(1.03–1.09) | <0.001 |
Post-Operative Mortality
Overall, the adjusted 30-day mortality for beneficiaries living in the least deprived and most deprived neighborhoods significantly lower within higher quality hospitals for each operation. However, hospital quality and its association with 30-day mortality between individuals living in the most and least deprived neighborhoods varied by procedure. For beneficiaries undergoing resection for esophageal cancer at low quality hospitals, the difference in mortality between those living in the least and most deprived neighborhoods was 22% vs. 21% (Abs difference 1.64, OR 1.22, 95%CI 1.14–1.31; P<0.001) (Table 3 and Figure 1). However, there was no statistically significant difference in mortality at high quality hospitals between beneficiaries from the most and least deprived neighborhoods (3.5% vs. 3.8%, Abs difference −0.34, OR 0.98 95%CI 0.94–1.02; p=0.30). A similar reduction in the absolute difference in mortality between those living in the most and least deprived neighborhoods with improving hospital quality was demonstrated for liver and pancreatic resection for cancer.
Table 3.
Risk Adjusted 30-Day Mortality among the Most and Least Deprived Medicare Beneficiaries at all Hospital Quality Levels.
| Risk-Adjusted 30-Day Mortality Rates % (95%CI) | |||||
|---|---|---|---|---|---|
| Least Deprived | Most Deprived | Absolute Difference | Odds Ratio (95%CI) | P-value | |
| Esophagectomy | |||||
| High Quality Hospitals | 3.84(3.33–4.35) | 3.50(3.05–3.96) | −0.34(−1.30–0.63) | 0.98(0.94–1.02) | 0.3 |
| Above Average Quality Hospitals | 6.21(5.68–6.74) | 6.21(5.71–6.72) | 0.00(−1.03–1.04) | 0.97(0.92–1.02) | 0.22 |
| Average Quality Hospitals | 8.54(7.97–9.12) | 9.15(8.51–9.79) | 0.61(−0.60–1.83) | 1.07(1.03–1.11) | <0.001 |
| Below Average Quality Hospitals | 11.91(11.03–12.79) | 11.22(10.38–12.07) | −0.69(−2.42–1.04) | 0.98(0.93–1.04) | 0.52 |
| Low Quality Hospitals | 20.67(18.95–22.39) | 22.31(20.30–24.33) | 1.64(−2.09–5.38) | 1.22(1.14–1.31) | <0.001 |
| Liver Resection | |||||
| High Quality Hospitals | 1.79(1.49–2.09) | 2.22(1.91–2.52) | 0.43(−0.18–1.103) | 1.03(0.96–1.10) | 0.39 |
| Above Average Quality Hospitals | 3.46(3.17–3.75) | 3.50(3.19–3.81) | 0.04(−0.56–0.64) | 1.07(1.02–1.13) | 0.003 |
| Average Quality Hospitals | 5.14(4.78–5.50) | 5.21(4.81–5.60) | 0.07(−0.68–0.83) | 1.09(1.03–1.15) | 0.003 |
| Below Average Quality Hospitals | 8.05(7.54–8.56) | 8.54(7.90–9.19) | 0.49(−0.66–1.65) | 1.14(1.09–1.18) | <0.001 |
| Low Quality Hospitals | 16.39(15.04–17.74) | 19.25(17.62–20.88) | 2.86(−0.13–5.84) | 1.42(1.30–1.54) | <0.001 |
| Lung Resection | |||||
| High Quality Hospitals | 1.72(1.63–1.82) | 1.94(1.85–2.03) | 0.22(0.03–0.40) | 1.08(1.06–1.10) | <0.001 |
| Above Average Quality Hospitals | 2.36(2.28–2.44) | 2.69(2.60–2.78) | 0.33(0.16–0.50) | 1.16(1.13–1.18) | <0.001 |
| Average Quality Hospitals | 3.23(3.13–3.33) | 3.53(3.41–3.64) | 0.30(0.09–0.51) | 1.12(1.10–1.14) | <0.001 |
| Below Average Quality Hospitals | 4.56(4.41–4.70) | 4.98(4.82–5.13) | 0.42(0.12–0.72) | 1.12(1.10–1.14) | <0.001 |
| Low Quality Hospitals | 7.79(7.47–8.10) | 8.31(8.01–8.62) | 0.52(−0.1–1.14) | 1.07(1.04–1.09) | <0.001 |
| Pancreatic Resection | |||||
| High Quality Hospitals | 2.88(2.60–3.17) | 3.08(2.80–3.35) | 0.19(−0.37–0.75) | 1.10(1.05–1.14) | <0.001 |
| Above Average Quality Hospitals | 4.86(4.58–5.14) | 4.77(4.51–5.03) | −0.09(−0.63–0.45) | 0.98(0.96–1.01) | 0.26 |
| Average Quality Hospitals | 6.14(5.82–6.46) | 6.27(5.92–6.63) | 0.13(−0.54–0.81) | 1.04(1.01–1.08) | 0.01 |
| Below Average Quality Hospitals | 8.65(8.15–9.15) | 9.37(8.82–9.92) | 0.72(−0.33–1.77) | 1.09(1.05–1.14) | <0.001 |
| Low Quality Hospitals | 12.93(12.02–13.85) | 15.87(14.76–16.97) | 2.93(0.91–4.95) | 1.26(1.21–1.32) | <0.001 |
| Rectal Resection | |||||
| High Quality Hospitals | 1.51(1.33–1.70) | 1.41(1.30–1.52) | −0.10(−0.39–0.19) | 0.92(0.85–0.98) | 0.02 |
| Above Average Quality Hospitals | 2.38(2.23–2.53) | 2.35(2.21–2.49) | −0.02(−0.32–0.27) | 1.04(1.01–1.06) | 0.001 |
| Average Quality Hospitals | 3.30(3.13–3.47) | 3.58(3.40–3.76) | 0.28(−0.07–0.63) | 1.08(1.06–1.11) | <0.001 |
| Below Average Quality Hospitals | 4.68(4.46–4.90) | 4.93(4.69–5.16) | 0.25(−0.21–0.71) | 1.07(1.04–1.09) | <0.001 |
| Low Quality Hospitals | 8.16(7.69–8.62) | 8.83(8.33–9.33) | 0.67(−0.29–1.64) | 1.06(1.03–1.09) | <0.001 |
Figure 1.
Adjusted Odds of 30-Day Mortality for Beneficiaries Living in the Most compared to Least Deprived Neighborhoods, 2014–2018.
For instance, at low quality hospitals there was a significant difference in mortality between patients living in the least and most deprived neighborhoods (19.25% vs. 16.39%, Abs difference:1.65, OR 1.42; P<0.001). However, at high quality hospitals there were no significant differences (2.22% vs. 1.79%, Abs Diff 0.43, OR 1.03, 95%CI 0.96–1.10, P=0.39). For resection for pancreatic cancer, there was a reduction in the difference in mortality between individuals living in the most and least deprived neighborhoods but there remained a disparity at the highest quality hospitals (3.08% vs. 2.88%, Abs Diff 0.19, OR 1.10, 95%CI 1.05–1.14, P<0.001).
For resection for rectal cancer, patients from the most deprived neighborhoods had higher mortality rates at low quality hospitals (8.83% vs. 8.16%, Abs difference: 0.67, OR 1.06, 95%CI 1.03–1.09, P<0.001). However, at high quality, beneficiaries from the most deprived neighborhoods had significantly lower mortality rates compared to the least deprived (1.41% to 1.51%, Abs difference −0.10, OR 0.92, 95%CI 0.85–0.98, P=0.02). Last for resection for lung cancer, we found that differences in 30-day mortality between those living in the most compared to least deprived neighborhoods remained statistically significant at high quality hospitals.
Discussion:
Although disparities in mortality following complex cancer surgery are well known, the potential role of hospital quality in reducing these disparities has been understudied. In this study, we found that improving hospital quality was associated with no significant difference in mortality between beneficiaries from the least and most deprived neighborhoods. As healthcare systems and policymakers attempt to identify targeted interventions to reduce disparities in care, a focus on improving access to high quality hospitals for socially-at-risk patients, who live in socioeconomically deprived areas, may be warranted.
Prior work evaluating the association of improved hospital quality and better surgical outcomes for high-risk Medicare beneficiaries has primarily focused solely on medical risk factors. For example, investigation of high-risk surgical candidacy, defined by age, procedure, and co-morbidities, and surgical outcomes found that when high risk surgical candidates received care at high quality hospitals for common elective procedures their overall complications were significantly reduced [25]. In a similar retrospective study by Smith et al, medically high risk patients undergoing resection for pancreatic surgery had reduced mortality and serious complications when they received their operation at a high quality hospital[26]. The present study extends both of these works by evaluating how beneficiary neighborhood deprivation, determined by Area Deprivation Index and a measure of social risk used by the Institute of Medicine, may also be used to identify patients that may benefit from targeted referral and access to high quality hospitals [27].
Efforts to examine neighborhood deprivation and surgical outcomes have raised concern that individuals living in socioeconomically deprived areas have worse access to high quality hospitals and outcomes but have not identified how hospital quality mediates disparities in surgical care in national cohorts. For instance, Diaz et al demonstrated in a retrospective study in California that patients residing in the highest socially vulnerable counties were less likely to receive complex oncologic surgery at a high-volume hospital. Additional studies have demonstrated that neighborhood social vulnerability is associated with higher mortality, complications and readmissions following complex cancer operations [28,29]. Our present findings extend these studies by demonstrating in a nationally representative cohort of patients undergoing resection for cancer, that receipt of surgery in high quality hospitals is associated with reduction in disparities in mortality between individuals living in the most and least deprived neighborhoods.
Our findings of no significant differences in mortality at high quality hospitals between levels of neighborhood deprivation have important policy implications. Referring patients requiring complex cancer surgery to high quality centers to reduce post-operative outcomes is central to ongoing centralization policy debates[30,31]. One of the often-cited limitations of plans for centralization efforts for high-risk cancer surgeries has been the potential to increase barriers to care particularly for socially-at-risk populations due to increased travel times, reduced access to social support and lack of financial resources [32]. Yet our study demonstrates that socially-at-risk patients, particularly those living in the most deprived areas, may uniquely benefit from selective referral to high quality hospitals for complex cancer surgery. This finding supports recent calls for “Ethical centralization” for high-risk surgery, which prioritizes efforts to ensure equitable access to high quality hospitals particularly for racial and ethnic minorities as well as individuals with socioeconomic barriers, including the neighborhoods in which they live [33]. However, this should be considered in conjunction with performance improvement at hospitals that may disproportionately care for socioeconomically deprived patient populations. The focus on neighborhood deprivation is particularly important given the historic nature federal and state policies that have led to profound levels of racial residential segregation in the United States. For instance, in our cohort racial and ethnic minorities were more likely to live in the most deprived neighborhoods reflecting the longstanding effects historic policies allowing for red-lining, housing discrimination, disinvestment in communities, tribal lands and Indian reservations. Therefore, policy efforts to target individual’s living in the most deprived neighborhoods may also aid in the goal of improving cancer care for racial and ethnic minorities. Prior assessments have demonstrated that individuals with higher socioeconomic disadvantage would travel to a more specialized center for cancer care if financial barriers for travel were removed [32]. Taken together, initiatives to increase receipt of care at high quality hospitals for individuals living in neighborhoods with high socioeconomic disadvantage should be prioritized to improve care and mitigate disparities but will also require healthcare systems and payers to address the social and financial needs of patients to relocate location of care.
Our study has several limitations. First, the MEDPAR file contains only administrative claims data for Medicare beneficiaries which may not be applicable to other patient populations. However, given that Medicare represents a geographically diverse patient cohort and is a large payer currently implementing strategies to reduce disparities, our analysis will be directly informative for future policy. Given that it is a claims database relying on accurate and complete coding, it is possible that specific patient characteristics may be incompletely captured. However, given that our outcome of interest in this analysis was 30-day mortality alone, it is less likely to be impacted by coding error compared to other outcomes such as complications. Additionally, the Medicare claims data in this analysis did not include oncologic stage, neo-adjuvant therapy history or whether the operation was for an incident or recurrent cancer diagnosis. However, the phenomenon of increasing hospital quality being associated with a reduction in mortality and disparities persisted across different cancer operations suggesting that hospital quality has an important association with disparities. Finally, Area Deprivation Index is a measure of community level socioeconomic disadvantage that may not fully capture an individual’s social risk or poverty level, unlike other proxy measures such as dual enrollment in Medicaid and Medicare. However, in this analysis we used 9-digit ZIP code correlating to a beneficiary’s census tract rather than ZIP code or county geographic levels, which provides a more detailed representation of a beneficiaries-built environment.
Conclusions:
In summary, our results demonstrate that receipt of surgery at high quality hospitals was associated with no significant difference in mortality between in individuals living in the most and least deprived neighborhoods. Our results suggest that healthcare systems and policy makers should prioritize efforts to improve access to high quality hospitals not only for both medically and socially high-risk surgical candidates. Furthermore, routine assessment of social risk factors in addition to medical risk should be incorporated into research and policy centered on improvement of surgical care delivery.
Supplementary Material
Funding Sources:
Sidra Bonner receives funding from the NIH T32 Multidisciplinary Program in Lung Disease at the University of Michigan.
Justin B. Dimick received grant funding from the National Institutes of Health (R01AG039434).
References:
- 1.Gomez SL, Shariff-Marco S, DeRouen M, et al. The impact of neighborhood social and built environment factors across the cancer continuum: Current research, methodological considerations, and future directions. Cancer. 2015;121(14):2314–2330. doi: 10.1002/cncr.29345 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Beyer KM, Malecki KM, Hoormann KA, Szabo A, Nattinger AB. Perceived Neighborhood Quality and Cancer Screening Behavior: Evidence from the Survey of the Health of Wisconsin. J Community Health. 2016;41(1):134–137. doi: 10.1007/s10900-015-0078-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Adie Y, Kats DJ, Tlimat A, et al. Neighborhood Disadvantage and Lung Cancer Incidence in Ever-Smokers at a Safety Net Health-Care System: A Retrospective Study. Chest. 2020;157(4):1021–1029. doi: 10.1016/j.chest.2019.11.033 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Zhang D, Matthews CE, Powell-Wiley TM, Xiao Q. Ten-year change in neighborhood socioeconomic status and colorectal cancer. Cancer. 2019;125(4):610–617. doi: 10.1002/cncr.31832 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.White K, Haas JS, Williams DR. Elucidating the role of place in health care disparities: the example of racial/ethnic residential segregation. Health Serv Res. 2012;47(3 Pt 2):1278–1299. doi: 10.1111/j.1475-6773.2012.01410. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Azap RA, Hyer JM, Diaz A, Paredes AZ, Pawlik TM. Association of County-Level Vulnerability, Patient-Level Race/Ethnicity, and Receipt of Surgery for Early-Stage Hepatocellular Carcinoma. JAMA Surg. 2021;156(2):197–199. doi: 10.1001/jamasurg.2020.5554 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Oliphant R, Nicholson GA, Horgan PG, et al. Deprivation and colorectal cancer surgery: longer-term survival inequalities are due to differential postoperative mortality between socioeconomic groups. Ann Surg Oncol. 2013;20(7):2132–2139. doi: 10.1245/s10434-013-2959-9 [DOI] [PubMed] [Google Scholar]
- 8.Diaz A, Hyer JM, Barmash E, Azap R, Paredes AZ, Pawlik TM. County-level Social Vulnerability is Associated With Worse Surgical Outcomes Especially Among Minority Patients. Ann Surg. 2021;274(6):881–891. doi: 10.1097/SLA.0000000000004691 [DOI] [PubMed] [Google Scholar]
- 9.Bhattacharyya O, Li Y, Fisher JL, et al. Low neighborhood socioeconomic status is associated with higher mortality and increased surgery utilization among metastatic breast cancer patients. Breast. 2021;59:314–320. doi: 10.1016/j.breast.2021.08.003Bhattacharyya [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Li Y O, Fisher JL, et al. Low neighborhood socioeconomic status is associated with higher mortality and increased surgery utilization among metastatic breast cancer patients. Breast. 2021;59:314–320. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Patel MI, Lopez AM, Blackstock W, Reeder-Hayes K, Moushey EA, Phillips J, Tap W. Cancer Disparities and Health Euqity: A Policy Statement from the American Society of Clinical Oncology. J Clin Oncol. 2020; 38(29):3439–3448. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Haider AH, Danwa-Mullan I, Maragh-Bass AC et al. Setting a National Agenda for Surgical Disparties Research: Recommendations from the National Institutes of Health and American College of Surgeons Summit, JAMA Surg. 2016;151(6):554–63. [DOI] [PubMed] [Google Scholar]
- 13.Sheetz KH, Dimick JB, Nathan H. Centralization of High-Risk Cancer Surgery Within Existing Hospital Systems. J Clin Oncol. 2019;37(34):3234–3242. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Urbach DR. Pledging to Eliminate Low-Volume Surgery. N Engl J Med. 2015;373(15):1388–1390. doi: 10.1056/NEJMp1508472 [DOI] [PubMed] [Google Scholar]
- 15.Vonlanthen R, Lodge P, Barkun JS, et al. Toward a Consensus on Centralization in Surgery. Ann Surg. 2018;268(5):712–724. [DOI] [PubMed] [Google Scholar]
- 16.Hu J, Kind AJH, Nerenz D. Area Deprivation Index Predicts Readmission Risk at an Urban Teaching Hospital. Am J Med Qual. 2018;33(5):493–501. doi: 10.1177/1062860617753063 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Diez Roux AV, Merkin SS, Arnett D, et al. Neighborhood of residence and incidence of coronary heart disease. N Engl J Med. 2001;345(2):99–106. doi: 10.1056/NEJM200107123450205 [DOI] [PubMed] [Google Scholar]
- 18.Yen IH, Kaplan GA. Neighborhood social environment and risk of death: multilevel evidence from the Alameda County Study. Am J Epidemiol. 1999;149(10):898–907. doi: 10.1093/oxfordjournals.aje.a009733 [DOI] [PubMed] [Google Scholar]
- 19.Reames BN, Birkmeyer NJ, Dimick JB, Ghaferi AA. Socioeconomic disparities in mortality after cancer surgery: failure to rescue. JAMA Surg. 2014;149(5):475–481. doi: 10.1001/jamasurg.2013.5076 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Wakeam E, Hevelone ND, Maine R, et al. Failure to rescue in safety-net hospitals: availability of hospital resources and differences in performance. JAMA Surg. 2014;149(3):229–235. doi: 10.1001/jamasurg.2013.3566 [DOI] [PubMed] [Google Scholar]
- 21.Schrag D, Panageas KS, Riedel E, et al. Hospital and surgeon procedure volume as predictors of outcome following rectal cancer resection. Ann Surg. 2002;236(5):583–592. doi: 10.1097/00000658-200211000-00008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Kind AJH, Buckingham W. Making Neighborhood Disadvantage Metrics Accessible: The Neighborhood Atlas. New England Journal of Medicine, 2018. 378: 2456–2458. DOI: 10.1056/NEJMp1802313. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Kind AJ, Jencks S, Brock J, et al. Neighborhood socioeconomic disadvantage and 30-day rehospitalization: a retrospective cohort study. Ann Intern Med 2014; 161: 765–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Iezzoni LI, Daley J, Heeren T, et al. Identifying complications of care using administrative data. Med Care. 1994;32(7):700–715. [DOI] [PubMed] [Google Scholar]
- 25.Smith ME, Shubeck SP, Nuliyalu U, Dimick JB, Nathan H. Local Referral of High-risk Patients to High-quality Hospitals: Surgical Outcomes, Cost Savings, and Travel Burdens. Ann Surg. 2020;271(6):1065–1071. doi: 10.1097/SLA.0000000000003208 [DOI] [PubMed] [Google Scholar]
- 26.Smith ME, Nuliyalu U, Dimick JB, Nathan H. Local Referral of High-risk pancreatectomy patients to improve surgical outcomes and minimize travel burden. J Gasrointest Surg. 2020;24(4):882–889. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.National academies of sciences, engineering, and medicine. 2016. Accounting for social risk factors in Medicare payment: Identifying social risk factors. Washington, DC: The National Academies Press. [PubMed] [Google Scholar]
- 28.Hyer JM, Tsilimigras DI, Diaz A, et al. High Social Vulnerability and “Textbook Outcomes” after Cancer Operation. J Am Coll Surg. 2021;232(4):351–359. doi: 10.1016/j.jamcollsurg.2020.11.024 [DOI] [PubMed] [Google Scholar]
- 29.Azap RA, Paredes AZ, Diaz A, Hyer JM, Pawlik TM. The association of neighborhood social vulnerability with surgical textbook outcomes among patients undergoing hepatopancreatic surgery. Surgery. 2020;168(5):868–875. doi: 10.1016/j.surg.2020.06.032 [DOI] [PubMed] [Google Scholar]
- 30.Stitzenberg KB, Sigurdson ER, Egleston BL, Starkey RB, Meropol NJ. Centralization of cancer surgery: implications for patient access to optimal care. J Clin Oncol. 2009;27(28):4671–4678. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Sheetz KH, Dimick JB, Nathan H. Centralization of High-Risk Cancer Surgery Within Existing Hospital Systems. J Clin Oncol. 2019;37(34):3234–3242. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Resio BJ, Chiu AS, Hoag JR, et al. Motivators, Barriers, and Facilitators to Traveling to the Safest Hospitals in the United States for Complex Cancer Surgery. JAMA Netw Open. 2018;1(7):e184595. Published 2018 Nov 2. doi: 10.1001/jamanetworkopen.2018.4595 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Binkley CE, Kemp DS. Ethical Centralization of High-risk Surgery Requires Racial and Economic Justice. Ann Surg. 2020;272(6):917–918. doi: 10.1097/SLA.0000000000004460 [DOI] [PubMed] [Google Scholar]
- 34.Williams DR, Collins C. Racial residential segregation: a fundamental cause of racial disparities in health. Public Health Rep. 2001;116(5):404–416. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Landrine H, Corral I, Lee JGL, Efird JT, Hall MB, Bess JJ. Residential Segregation and Racial Cancer Disparities: A Systematic Review. J Racial Ethn Health Disparities. 2017;4(6):1195–1205. [DOI] [PubMed] [Google Scholar]
- 36.Kruse G, Lopez-Carmen VA, Jensen A, Hardie L, Sequist TD. The Indian Health Service and American Indian/Alaska Native Health Outcomes. Annu Rev Public Health. 2022;43:559–576. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.