Abstract
Objective:
Studies examining the relationship between socioeconomic disparities and peripheral artery disease (PAD) often focus on individual social health determinants and fail to account for the complex interplay between factors that ultimately impact disease severity and outcomes. Area Deprivation Index (ADI), a validated measure of neighborhood adversity, provides a more comprehensive assessment of social disadvantage. Therefore, we examined the impact of ADI on PAD severity and its management.
Methods:
We identified all patients who underwent infrainguinal revascularization (open or endovascular) or amputation for symptomatic PAD in the VQI registry between 2003-2020. An ADI score of 1-100 was assigned to each patient based on their residential zip code, with higher ADI scores corresponding with increasing adversity. Patients were categorized by ADI quintiles (Q1-Q5). Outcomes of interest included indication for procedure (claudication, rest pain, or tissue loss) and rates of revascularization (vs primary amputation). Multinomial logistic regression was used to evaluate for an independent association between ADI quintile and these outcomes.
Results:
Among the 79,973 patients identified, 9,604 (12%) were in the lowest ADI quintile (Q1), 14,961 (18.7%) in Q2, 19,800 (24.8%) in Q3, 21,735 (27.2%) in Q4 and 13,873 (17.4%) in Q5. There were significant trends towards lower rates of claudication (Q1 39% vs Q5 34%, p<0.001), and higher rates of rest pain (Q1 12.4% vs Q5 17.8%, p<0.001) as the indication for intervention, and lower rates of revascularization (Q1 80% vs Q5 69%, p<0.001) with increasing ADI quintiles. In adjusted analyses, there was a progressively higher likelihood of presenting with rest pain vs claudication, with patients in Q5 having the highest probability when compared to those in Q1 (RR 2.0 95%CI 1.8-2.2, p<0.001). Patients in Q5, when compared to those in Q1, also had a higher likelihood of presenting with tissue loss vs claudication, (RR 1.4 95%CI 1.3-1.6; p<0.001). Compared with patients in Q1, patients in Q2-5 had a lower likelihood of undergoing any revascularization procedure.
Conclusions:
Among patients who underwent infrainguinal revascularization or amputation in the VQI, those with higher neighborhood adversity had more advanced disease at presentation and lower rates of revascularization. Further work is needed to better understand neighborhood factors that are contributing to these disparities in order to identify community-level targets for improvement.
Keywords: PAD, Peripheral Artery Disease, PVD, Amputation, Revascularization, Lower Extremity, Disparity, VQI, Equity, Area Deprivation Index, Disadvantage, Neighborhood
Table of Contents Summary:
In this retrospective study of patients who underwent infrainguinal revascularization or amputation in the Vascular Quality Initiative, patients living in neighborhoods with high deprivation were more likely to undergo amputation rather than revascularization, and more likely to undergo revascularization for rest pain or tissue loss rather than claudication. Policies addressing barriers to care in these high deprivation neighborhoods may lead to improvement in the management of PAD.
Introduction
The relationship between social economic status (SES) and surgical outcomes has been widely explored in recent years. Low socioeconomic status, classically measured by lack of insurance and low income, has been associated with lower revascularization rates in peripheral artery disease (PAD) and higher likelihood of undergoing amputation.1,2 Studies examining the relationship of race and/or ethnicity (other commonly used SES proxies) and PAD have demonstrated similar disparities in outcomes of the management of PAD. Black and Hispanic patients have been shown to have higher rates of amputation compared to white patients, and experience higher rates of mobility loss and higher costs following lower extremity revascularization.3-6 These differences are often attributed to increase comorbidity burden or lack of access to high-performing centers, but the impact of community factors on disease severity and outcomes is unknown .7,8 Factors such as education, geography, housing status, food access, safety, and other community level factors may play an integral role in presentation and management of PAD, and therefore need to be adequately explored.
Area Deprivation Index (ADI), a measure of social determinants of health, was developed to better understand the environmental impact experienced at the community level.9 ADI was created by the Health Resources and Services Administration utilizing census data to assess 17 education, employment, housing quality, and poverty measures.10 It has since been validated and used to study the relationship between neighborhood disadvantage and multiple surgical outcomes.11,12,13 Using ADI to analyze the impact of communal social determinants of health has resulted in a better understanding of factors that contribute to the severity of disease and outcomes following intervention. We, therefore, aimed to examine the impact of neighborhood social disadvantage, utilizing area deprivation index, on the severity and management of PAD utilizing the Vascular Quality Initiative (VQI).
Methods
Data Source
We performed a retrospective cohort study using the VQI, a clinical registry established as a collaboration between regional quality groups to improve patient care. At the time of this study, the VQI included 18 regions and over 550 participating centers in three countries. However, for the purposes of this study only centers in the U.S were included. More information about the VQI can be found at www.vascularqualityinitiative.org. This study was designed in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.14 The Institutional Review Board at Beth Israel Deaconess Medical Center approved this study and informed consent was waived, given the de-identified nature of this study.
Patient Population
We identified all patients undergoing infrainguinal open or endovascular intervention or amputation for peripheral artery disease between 2003 and 2020 (n=270,829). An ADI score of 1-100 was assigned to each patient based on their residential zip code, with higher ADI scores corresponding with increasing deprivation, according to data available via the Neighborhood Atlas.15 Patients were then further divided into ADI-based quintiles. Patients without zip code data were excluded (n=28,723). Patients presenting with acute limb ischemia or asymptomatic disease (eg. aneurysm) were also excluded (n=11,064). Additionally, patients who had previously undergone a revascularization procedure, open or endovascular, or amputation were also excluded (n=151,069).
Definitions and variables
Preoperative creatinine value was used to estimate the glomerular filtration rate (eGFR) for each patient using the Chronic Kidney Disease Epidemiology Collaboration equation, which accounts for the patient's sex and age. Smoking history was defined as any previous or current cigarette, cigar, or pipe use. Coronary artery disease (CAD) was defined as prior history of stable or unstable angina, MI, or a prior coronary intervention. Tissue loss severity was defined as mild (shallow ulcer), moderate (full thickness ulcer with exposed bone, joint or tendon), or severe (extensive deep ulcer or gangrene with exposed bone, joint or tendon). Tissue loss severity was only available for patients undergoing endovascular peripheral vascular intervention (PVI). Revision of intervention was defined as return to the OR for reoperation/revision as defined in the VQI. Hospital volume was measured within each procedure type over the 365-day period preceding each operation. Volume was then divided into quintiles, and cut-offs for low (quintile 1), medium (quintiles 2-4), and high volume (quintile 5) were selected. Centers were determined to be high volume if they performed more than 110 PVI or 33 infrainguinal bypasses, low volume if less than 9 PVIs or bypasses, or medium volume if between these case numbers in the defined 365-day time period. Surgeons were high volume if they performed 29 PVI or 12 infrainguinal bypass, medium if performing 7-28 PVI or 3-11 bypasses, and low volume if less than 7 PVI or 3 bypasses annually. Revascularization rate was defined as the proportion of endovascular or open revascularization procedures performed of the total number of procedures for PAD (PVI, bypass, and amputation), and not necessarily directly correlated to center or physician overall volume. When calculating revascularization rate, only centers that contributed to all three databases were included in the analysis. High revascularization rate, utilizing similar quintiles cutoffs as volume, was assigned to centers and physicians with a revascularization rate >99%, medium to centers with a rate between 79-99% or physicians with 80-99%, and low to centers with less than 79% revascularization rate and physicians with less than 80%.
Outcomes
The primary outcomes of interest were severity of disease at presentation (claudication vs. rest pain vs. tissue loss) and rate of revascularization (vs amputation) stratified by ADI quintile. Secondary outcomes included analysis of distribution of patient presentation based on center and surgeon experience, severity of tissue loss, and perioperative outcomes.
Statistical Analysis
Categorical variables were presented as counts and percentages. Univariate differences between cohorts were assessed using χ2 and Fisher's exact tests. Cuzick’s test for trend was performed to analyze the association of a particular variable with increasing ADI quintiles. Baseline demographics, preoperative medical therapy, and comorbidities of the different ADI cohorts were described. Multinomial logistic regression with clustering at the center level was performed to determine the effect of ADI on the odds of undergoing intervention for rest pain or tissue loss compared to claudication and undergoing revascularization procedure rather than amputation. The following variables were included in these models: age, race/ethnicity, sex, insurance, hypertension, insulin-dependent diabetes, chronic obstructive pulmonary, chronic kidney disease, preoperative dialysis, coronary artery disease, congestive heart failure, smoking status, BMI, preoperative aspirin therapy, preoperative statin therapy, preoperative anticoagulant use, ambulatory status, overall center volume and surgeon volume. For the likelihood of revascularization model, surgeon and center revascularization rate was also included.
Statistical analysis was conducted using Stata version 16.1 (StataCorp LP, College Station, Texas). P ≤0.05 was considered statistically significant.
Results:
Demographics
Of the 79,973 patients identified, 12% were in the first ADI quintile (least deprivation), 19% in the second, 25% in the third, 27% in the fourth, and 17% in the fifth ADI quintile (most deprivation) (Table 1). Patients in the fifth ADI quintile were younger, with most patients being between ages 60-69 (33%), while those in the first ADI quintile were more likely to be between ages of 70-79 (31%). Compared to those in the first ADI quintile, patients in the fifth ADI quintile were more likely to be female (41% vs 28%, p<0.001), Black (42% vs 18%, p<0.001), and uninsured (2.7% vs 1.5%, p<0.001). Patients in the fifth ADI quintile were also more likely to have comorbidities such as insulin dependent diabetes (36% vs 28%, p<0.001), hypertension (89% vs 87%, p<0.001), COPD (25% vs 16%, p<0.001), and CAD (30% vs 27%, p<0.001). Patients in the fifth ADI quintile were also more likely to be on dialysis preoperatively (12% vs 10%, p<0.001), and be current smokers (45% vs 23%, p<0.001) compared to those in the first ADI quintile (Table 1). There was no difference in preoperatively statin use between patients in the fifth ADI quintile and the ones in the first ADI quintile (66% vs 68%, p=0.018), but patients in the fifth ADI quintile were more likely to be on aspirin preoperatively (66% vs 63%, p<0.001) preoperatively.
Table 1 – Patient Characteristics and Hospital Volume.
Characteristics of patients undergoing lower extremity revascularization (open and endovascular) or amputation in the VQI from 2003-2020. * p values referent to Cuzick’s trend test.
| ADI Quintile | ||||||
|---|---|---|---|---|---|---|
| 1 (Lowest Deprivation) |
2 | 3 | 4 | 5 (Highest Deprivation) |
p value* | |
| Total N (%) | 9,604 (12%) | 14,961 (19%) | 19,800 (25%) | 21,735 (27%) | 13,873 (17%) | |
| AGE (%) | <0.001 | |||||
| <60 | 16 | 20 | 21 | 25 | 27 | |
| 60-69 | 26 | 28 | 30 | 32 | 33 | |
| 70-79 | 31 | 29 | 28 | 26 | 24 | |
| 80-89 | 21 | 19 | 18 | 14 | 13 | |
| 90+ | 6.0 | 4.0 | 3.6 | 2.8 | 2.4 | |
| Female (%) | 38 | 36 | 38 | 38 | 41 | <0.001 |
| Race (%) | <0.001 | |||||
| White (Non-Hispanic) | 77 | 81 | 83 | 77 | 56 | |
| Black (Non-Hispanic) | 18 | 17 | 16 | 22 | 42 | |
| Asian | 4.7 | 1.5 | 0.5 | 0.3 | 0.3 | |
| Other | 1.3 | 0.9 | 0.9 | 0.8 | 1.5 | |
| Ethnicity (%) | 0.684 | |||||
| Hispanic | 7.8 | 5.8 | 4.8 | 5.3 | 7.6 | |
| Insurance (%) | <0.001 | |||||
| Medicare | 58 | 57 | 55 | 53 | 52 | |
| Medicaid | 7 | 6 | 6 | 8 | 11 | |
| Private Insurance | 33 | 34 | 36 | 34 | 33 | |
| Other Insurance | 0.6 | 1.1 | 1.3 | 1.4 | 1.0 | |
| No insurance | 1.5 | 2.3 | 2.4 | 2.9 | 2.7 | |
| Ambulatory (%) | 71 | 74 | 75 | 75 | 73 | <0.001 |
| In SNF prior to admission (%) | 5.3 | 5.3 | 5.0 | 4.6 | 4.4 | <0.001 |
| Insulin Dependent DM (%) | 28 | 31 | 32 | 35 | 36 | <0.001 |
| DM (%) | 53 | 55 | 55 | 57 | 60 | <0.001 |
| CHF (%) | 21 | 21 | 20 | 20 | 21 | 0.626 |
| HTN (%) | 87 | 87 | 87 | 88 | 89 | <0.001 |
| eGFR (%) | 0.135 | |||||
| <30 | 6.9 | 5.9 | 6.2 | 6.5 | 7.3 | |
| 30-60 | 34 | 34 | 33 | 33 | 33 | |
| >60 | 59 | 60 | 61 | 61 | 60 | |
| Preoperative Dialysis (%) | 9.8 | 9.2 | 8.6 | 9.1 | 12 | 0.001 |
| COPD (%) | 16 | 21 | 23 | 25 | 25 | <0.001 |
| CAD (%) | 27 | 27 | 29 | 30 | 30 | <0.001 |
| Smoking History (%) | 62 | 71 | 73 | 74 | 72 | <0.001 |
| Current Smoker (%) | 23 | 33 | 37 | 43 | 45 | <0.001 |
| Preoperative Medication (%) | ||||||
| Aspirin | 63 | 65 | 66 | 67 | 66 | <0.001 |
| Statin | 68 | 66 | 65 | 65 | 66 | 0.018 |
| P2Y12 | 26 | 27 | 26 | 28 | 28 | <0.001 |
| Anticoagulant | 7.0 | 6.6 | 6.2 | 5.7 | 5.1 | <0.001 |
| Preoperative ABI | ||||||
| ≥0.80 | 32 | 29 | 26 | 25 | 24 | <0.001 |
| 0.6-0.79 | 26 | 24 | 25 | 25 | 24 | 0.362 |
| 0.4-0.59 | 25 | 27 | 29 | 30 | 29 | <0.001 |
| ≤0.39 | 17 | 20 | 20 | 21 | 22 | <0.001 |
Perioperative Outcomes
There were no significant differences in perioperative mortality following intervention across ADI quintiles in patients undergoing lower extremity revascularization or amputation (Table 2). Patients in the fifth ADI quintile were more like to undergo amputation following initial intervention (17% vs 15%, P=0.007) or revision of their intervention (15% vs 13%, p=0.022). There were no other differences in perioperative outcomes.
Table 2. Perioperative Outcomes and Patient Distribution by Center and Surgeon Revascularization rate.
Perioperative outcomes of patients undergoing lower extremity revascularization (open and endovascular) or amputation, and patient distribution stratified by center and surgeon revascularization in the VQI from 2003-2020. † Only available for patients undergoing endovascular revascularization (PVI). * p values referent to Cuzick’s trend test.
| ADI Quintile | ||||||
|---|---|---|---|---|---|---|
| 1 (Lowest Deprivation) |
2 | 3 | 4 | 5 (Highest Deprivation) |
p value* | |
| 9,604 | 14,961 | 19,800 | 21,735 | 13,873 | ||
| Any Complication (%) | 27 | 28 | 27 | 27 | 26 | 0.101 |
| Postop Amputation (%) | 15 | 15 | 15 | 17 | 17 | 0.007 |
| SSI (%) | 0.7 | 0.7 | 0.7 | 0.7 | 0.8 | 0.826 |
| Postop CHF (%) | 1.2 | 1.4 | 1.2 | 1.2 | 1.6 | 0.435 |
| Postop MI (%) | 0.8 | 0.6 | 0.6 | 0.7 | 0.8 | 0.309 |
| Respiratory Complication (%) | 1.3 | 1.0 | 1.2 | 1.2 | 1.3 | 0.217 |
| Stroke (%) | 0.4 | 0.6 | 0.8 | 0.8 | 0.8 | 0.51 |
| Revision (%) | 13 | 12 | 12 | 12 | 15 | 0.022 |
| Renal Complication (%) | 1.4 | 1.6 | 1.6 | 1.7 | 1.8 | 0.081 |
| Mean Length Treated† cm (SD) | 13 (12) | 13 (13) | 14 (12) | 14 (13) | 15 (12) | <0.001 |
| Endovascular Revascularization | 82 | 78 | 78 | 77 | 78 | 0.689 |
| Center Revascularization Rate (%) | ||||||
| Low (<79%) | 14 | 23 | 17 | 20 | 21 | 0.010 |
| Medium (79-99%) | 21 | 20 | 20 | 14 | 13 | <0.001 |
| High (>99%) | 65 | 56 | 64 | 66 | 66 | 0.189 |
| Surgeon Revascularization Rate (%) | ||||||
| Low (<80%) | 15 | 19 | 15 | 18 | 19 | 0.013 |
| Medium (80-99%) | 13 | 13 | 10 | 8.4 | 8.5 | 0.001 |
| High (>99%) | 72 | 68 | 74 | 74 | 72 | 0.098 |
Revascularization Approach
Patients in the first ADI quintile had a higher rate of endovascular (vs open) revascularization compared to the remaining ADI quintiles, but there were no significant trends suggesting decreasing endovascular usage with increasing ADI quintile (Table 2). In adjusted analysis although patients in ADI quintiles 2-5 were less likely to undergo endovascular revascularization (vs open) compared to those in the first ADI quintile, there were no clear differences between these groups (Table 3).
Table 3. Adjusted Relative Risk of Presenting with Rest Pain or Tissue Loss Versus Claudication, and Odds Ratio of Undergoing Revascularization Versus Amputation.
Adjusted analysis of procedural indication (rest pain or tissue loss vs claudication) and type of intervention (revascularization vs amputation), type of revascularization (endovascular versus open), and revascularization approach (endovascular vs open) based on ADI quintiles. ADI – Area Deprivation index; RR–Risk Ratio (calculated using multinomial logistic regression), OR – Odds Ratio. Analysis adjusted for age, race, gender, insurance, smoking status, hypertension, COPD, diabetes, insulin dependent diabetes, CAD, CHF, dialysis dependence, BMI, center and surgeon volume, preoperative aspirin, statin, p2y12 and anticoagulation use. For odds of revascularization vs amputation, center yearly revascularization rate and surgeon yearly revascularization rate also included; CI – Confidence Interval; Ref – Reference
| Indication | ADI Quintile | RR | 95% CI | p value | |
|---|---|---|---|---|---|
| Rest Pain | 1 (Lowest Deprivation) | Ref | |||
| 2 | 1.4 | 1.2 | 1.6 | <0.001 | |
| 3 | 1.7 | 1.5 | 1.9 | <0.001 | |
| 4 | 1.9 | 1.7 | 2.1 | <0.001 | |
| 5 (Highest Deprivation) | 2 | 1.8 | 2.2 | <0.001 | |
| Tissue Loss | 1 (Lowest Deprivation) | Ref | |||
| 2 | 1.2 | 1.1 | 1.4 | <0.001 | |
| 3 | 1.3 | 1.2 | 1.5 | <0.001 | |
| 4 | 1.4 | 1.3 | 1.5 | <0.001 | |
| 5 (Highest Deprivation) | 1.4 | 1.3 | 1.6 | <0.001 | |
| Procedural Intervention | ADI Quintile | OR | 95% CI | p value | |
| Revascularization | 1 (Lowest Deprivation) | Ref | |||
| 2 | 0.75 | 0.65 | 0.86 | <0.001 | |
| 3 | 0.75 | 0.65 | 0.87 | <0.001 | |
| 4 | 0.60 | 0.52 | 0.69 | <0.001 | |
| 5 (Highest Deprivation) | 0.59 | 0.51 | 0.70 | <0.001 | |
| Endovascular vs. Open | 1 (Lowest Deprivation) | Ref | |||
| 2 | 0.76 | 0.70 | 0.82 | <0.001 | |
| 3 | 0.76 | 0.70 | 0.82 | <0.001 | |
| 4 | 0.77 | 0.71 | 0.83 | <0.001 | |
| 5 (Highest Deprivation) | 0.78 | 0.72 | 0.85 | <0.001 | |
Distribution of presentation based on center and physician revascularization rates
All patients regardless of ADI were more likely to present to center and physicians with high revascularization rate compared to those with low revascularization rates (Table 2). However, when compared to patients in the first ADI quintile, a higher proportion of patients in the fifth ADI quintile presented to centers (21% vs 14%, p=0.010) and surgeons (19% vs 15%, p=0.013) that were classified as having a low revascularization rate.
Severity of disease presentation across ADI quintiles
When comparing indication for surgical intervention, patients in the fifth ADI quintile were less likely to have claudication as their primary indication for intervention (34% vs 39%, p<0.001), and more likely to have rest pain as their indication for intervention (18% vs 12%, p<0.001) compared to those in the first ADI quintile (Figure 1). There was no difference in the rates of tissue loss as the indication for intervention among ADI quintiles in unadjusted analysis (48% vs 48%, p=0.587). Patients in the fifth ADI quintile were also more likely to present with an ABI ≤0.39 compared to patients in the first ADI quintile (22% vs 17%, p<0.001), and subsequently less likely to present with an ABI ≥0.80 (24% vs 32%, p<0.001). After adjustment with the first ADI quintile as the referent group, patients in all other quintiles had increasing odds of rest pain as their indication for intervention as opposed to claudication, with patients in the fourth and fifth ADI quintile having the highest odds (4th quintile: OR 1.9; 95%CI [1.7 – 2.1], p<0.001; 5th quintile: OR 2.0; 95%CI [1.8 – 2.2], p<0.001) (Table 3). In adjusted analysis increasing ADI quintile was also associated with increase odds of undergoing intervention for tissue loss rather than claudication. Patients in the fifth quintile were 40% more like to undergo intervention for tissue loss rather than claudication compared to their counterparts in the first quintile (OR 1.4; 95%CI [1.3 – 1.6], p<0.001) (Table 3). In patients undergoing endovascular interventions, patients in the fifth ADI quintile were more likely to have severe (6.9% vs 5.9%, p<0.001), and less likely to have mild (63% vs 67%, p<0.001) degrees of tissue loss (Figure 2). Patients in the fifth ADI quintile also had a greater mean treated arterial length compared to patients in the first ADI quintile (15cm vs 13cm, p<0.001) (Table 2).
Figure 1.
Proportion of patients undergoing revascularization (open or endovascular) or amputation stratified by indication among ADI quintiles in the VQI between 2003-2020. * p values referent to cuzick’s trend test.
Figure 2.
Tissue loss severity among ADI quintiles in patients undergoing endovascular revascularization (PVI). * p values referent to Cuzick’s trend test.
Revascularization rate among ADI quintiles
A total of 29,244 patients were identified who underwent intervention at centers who report to PVI, Bypass and Amputation databases. Among these patients, those in the fifth ADI quintile had a revascularization rate of 69%, while those in the first ADI quintile had a rate of 80% (p<0.001) (Figure 3). For patients undergoing intervention for chronic limb-threatening ischemia (CLTI) only, those in the fifth ADI quintile continued to have lower revascularization rates compared to those in the first ADI quintile (61% vs. 72%, p<0.001) (Figure 3). In adjusted analysis these differences persistent with increasing ADI quintile correlating to a decrease in odds of undergoing revascularization versus amputation, with patients in the 5th ADI quintile have the lowest odds, being 41% less likely to undergo revascularization compared to those in the first ADI quintile (OR 0.59; 95%CI [0.50 – 0.69], p<0.001) (Table 3).
Figure 3.
Proportion of patients undergoing revascularization (vs ampuation) among ADI quintiles in the VQI between 2003-2020. * p values referent to cuzick’s trend test; CLTI – Chronic Limb-Threatening Ischemia.
Discussion
In this national sampling of patients undergoing infrainguinal revascularization or amputation for PAD from 2003-2020 in the VQI, we found that neighborhood deprivation impacts the severity of PAD disease presentation and rate of revascularization versus primary amputation. Compared to patients in less deprived neighborhoods, patients living in more deprived neighborhoods are more likely to present with more advanced disease (rest pain or tissue loss rather than claudication) and more likely to undergo primary amputation. These findings suggest complex interplay between community factors that ultimately impact disease severity and outcomes.
The association between SES and perioperative outcomes following lower extremity revascularization procedures have been extensively explored in recent years, however the association between the severity of PAD and SES, particularly at the community level, has not been adequately explored. 3,16-18 Our study demonstrates an association between neighborhood deprivation and higher severity of disease, with patients in high deprivation areas being more likely to undergo initial intervention for rest pain rather than claudication, having more extensive vessel involvement, as indicated by longer vessel treatment length, as well having more advanced tissue loss and lower ABI at presentation compared to their counterparts in less deprived neighborhoods. These differences may arise due to the higher rate of comorbidities, smoking rates, and lack of optimal medical care in patients in neighborhoods of high deprivation. Our data demonstrate that patients in the highest ADI quintile are more likely to have comorbidities such as insulin dependent diabetes, coronary artery disease, and chronic kidney disease, as well as being more likely to be current smokers, all significant risk factors for PAD that have been shown to accelerate disease progression. 19,20 Furthermore, despite the higher comorbidities burden, according to our data patients in the highest ADI quintile were not more likely to be on a statin preoperatively, which is known to decrease cardiovascular events and PAD associated mortality.21,22 Although there was no significant difference in perioperative mortality found in this study, these findings still emphasize the need for the creation of public health strategies that target specific barriers surrounding preventive healthcare that are found in these neighborhoods. For instance, neighborhood stress level and disorder have been associated with failure to comply with to public health initiatives such as smoking cessation.23 Programs addressing neighborhood characteristics that contribute to these factors would be helpful in resolving the disparities in PAD described in our study.
Comorbidity burden and optimization of medical care are often cited as the explanation for disparities in the diagnosis and management of PAD, however, they are only partially responsible for the neighborhood-related disparities encountered in our study.3,4,24After adjusting for comorbidities, smoking status, and preoperative medical management, patients from more deprived neighborhoods continue to demonstrate a significant increase in the likelihood of undergoing initial intervention for rest pain or tissue loss rather than claudication. Additionally, individual measures of SES, such as race and insurance status that have been previously associated with severity of PAD, also do not fully explain these differences.25,26 This highlights the importance of utilizing ADI to explore and address community-level factors that are likely at play, which are often missed due to the preferential utilization of individual measures of social disadvantage in the current literature. Although further studies are needed to identify the specific barriers to the management of PAD in these high deprivation neighborhoods, recent studies have indicated food and housing insecurity as factors that are associated with an increased risk of PAD and related complications and are therefore potential targets for intervention. 27-29 Additionally, it is likely that the higher severity of disease present at the time of intervention is also associated with delayed diagnosis. Higher neighborhood deprivation has been associated with a lower likelihood of an established source of medical care, as well as low adherence to screening guidelines, independent of individual characteristics.30-33 Neighborhood conditions such as inconsistent public transport and lack of safety have been implicated in the medical underutilization in these neighborhoods, and are likely contributing to delayed presentation resulting in the higher severity of PAD demonstrated in our study in patients of high deprivation neighborhoods. Policies addressing these environmental factors are likely to improve a neighborhood’s deprivation level and provide improvement in the care of PAD in patients living in these areas.
Lower extremity revascularization was also impacted by neighborhood deprivation. Patients in the highest ADI quintile were less likely to undergo a revascularization procedure (vs amputation) at initial intervention compared to patients in the lowest ADI quintile. The impact of SES in revascularization rate has been widely reported. Black, uninsured, or patients of low-income quartiles have all been shown to have a lower likelihood of undergoing limb-saving revascularization.4,18,34 Our study adds to this growing literature surrounding SES and revascularization disparities, but unlike previous studies, also highlights neighborhood resources as an important factor to explore when attempting to address these disparities independent of race, gender, or insurance status. Even after adjusting for individual markers of social disadvantage, patients in the high deprivation neighborhoods were more than 40% less likely to undergo a revascularization procedure rather than amputation. Higher severity disease often found in these patients, is also often postulated to be the driving force behind this revascularization disparity.4 Our data certainly appear to corroborate this in part. As discussed, patients in high deprivation neighborhoods are more likely to present with higher severity of PAD at initial intervention; however, even after adjusting for disease severity in our study, patients in high deprivation neighborhoods continue to undergo revascularization at lower rates compared to patients in low deprivation neighborhoods. A potential explanation for this persistent finding may be disparities present in the referral to cardiovascular specialists. In our study, we found a significantly higher percentage of patients in the highest ADI quintile undergoing intervention at centers and physicians with low revascularization rates compared to patients in the lowest ADI quintile. Studies surrounding specialist access in arteriovenous fistula have demonstrated a geographic variation in vascular specialist, with more specialist physicians being present in areas with low unemployment rates and higher percentage of white population.35 A similar phenomenon may be at play here, with patients in high deprivation neighborhoods having less access to surgeons with higher revascularization rates, leading to the disparities demonstrated here. Interventions aimed to improve the geographic variation in the supply of qualified vascular surgeons may result in a significant improvement in the rates of revascularization seen in patients in high deprivation neighborhoods.
This study must be interpreted in the context of the design. One important limitation is that ADI was assigned based on 5-digit zip codes. Given that 9-digit zip codes are not available in the VQI, 5-digit codes were used instead, reducing the granularity of the data and therefore limiting the ability to account for neighborhood deprivation variability within a single zip code. The VQI database is a voluntary quality initiative that represents a self-selected subset of the total population of hospitals which may limit its generalizability. Furthermore, we are limited by the variables and their definitions available in the VQI, including information medical optimization of comorbidities and completion of medical management of PAD. VQI provides limited data on screening, preoperative surveillance, primary care physician access, or referral data, which constrains our ability to analyze the true impact of health care access and referral patterns in the diagnosis and management of PAD. There is also no ability to distinguish between rural and urban centers or teaching and nonteaching hospitals, which may be important when addressing outcomes associated with neighborhood deprivation given the potential differences in resources and surgical techniques available in these different settings. Furthermore, we cannot analyze the distance to the nearest health care center, which is undoubtedly an important factor leading to delayed medical care. Additionally, the omission of non-VQI centers or those which do not contribute to all three databases when analyzing revascularization rate likely skews the distribution of the type of centers where patients in the highest ADI quintiles are receiving their care. Additional studies utilizing datasets that include this information can help identify additional areas of intervention that will help decrease the disparity in severity and management of PAD highlighted in our study. Lastly, the VQI only captures data from patients who underwent an intervention. Therefore, any disparities due to neighborhood deprivation, which results in death or an inoperable pathology would be missed in this study.
Conclusion:
Among patients who underwent revascularization or amputation in the VQI, those with higher neighborhood deprivation had significantly higher rates of rest pain or tissue loss (vs claudication) as their indication for initial intervention, and lower rates of revascularization (vs amputation). Our study indicates the potential of utilizing ADI to identify high-risk patients and develop policies that aim to improve their outcomes. Further work is needed to better understand the neighborhood factors that contribute to these disparities in order to identify specific community-level targets for improvement.
Article Highlights:
Type of Research:
Retrospective review of prospectively collected Vascular Quality Initiative (VQI) data.
Key Findings:
Among patients who underwent infrainguinal revascularization or amputation for peripheral artery disease (PAD), those living in areas of higher neighborhood deprivation had more advanced disease at presentation and lower revascularization rates. These disparities persisted regardless of center or surgeon volume, and patient factors such as race, sex, insurance status, or comorbidities.
Take home Message:
Patients living in neighborhoods with high deprivation have more severe PAD and less access to standard-of-care for PAD.
Sources of Funding:
LM is supported by the Harvard-Longwood Research Training in Vascular Surgery NIH T32 Grant 5T32HL007734. CM is supported by grant number F32HS027285 from the Agency for Healthcare Research and Quality. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Agency for Healthcare Research and Quality.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Disclosures: None
This work has been presented at the 2022 Vascular Annual Meeting on June 17, 2022 in Boston, MA.
References
- 1.Henry AJ, Hevelone ND, Belkin M, Nguyen LL. Socioeconomic and Hospital-Related Predictors of Amputation for Critical Limb Ischemia. J Vasc Surg, 2011; 53:330–339. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Hughes K, Mota L, Nunez M, Sehgal N, Ortega G. The effect of income and insurance on the likelihood of major leg amputation. J Vasc Surg. 2019. Aug;70(2):580–587. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Hughes K, Olufajo OA, White K, Roby DH, Fryer CS, Wright JL, et al. The influence of socioeconomic status on outcomes of lower extremity arterial reconstruction. J Vasc Surg. 2022. Jan;75(1):168–176. [DOI] [PubMed] [Google Scholar]
- 4.Durazzo TS, Frencher S, Gusberg R. Influence of Race on the Management of Lower Extremity Ischemia: Revascularization vs Amputation. J Vasc Surg. 2013;58(6):1719. [DOI] [PubMed] [Google Scholar]
- 5.Traven SA, Synovec JD, Walton ZJ, Leddy LR, Suleiman LI, Gross CE. Notable Racial and Ethnic Disparities Persist in Lower Extremity Amputations for Critical Limb Ischemia and Infection. J Am Acad Orthop Surg. 2020;28(21):885–892. [DOI] [PubMed] [Google Scholar]
- 6.McDermott MM, Polonsky TS, Kibbe MR, Tian L, Zhao L, Pearce WH, et al. Racial differences in functional decline in peripheral artery disease and associations with socioeconomic status and education. J Vasc Surg. 2017;66(3):826–834. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Regenbogen SE, Gawande AA, Lipsitz SR, Greenberg CC, Jha AK. Do differences in hospital and surgeon quality explain racial disparities in lower-extremity vascular amputations? Ann Surg. 2009;250:424–31. [DOI] [PubMed] [Google Scholar]
- 8.Frieden TR. A framework for public health action: the health impact pyramid. Am J Public Health. 2010;100(4):590–595. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Kind AJH, Buckingham WR. Making neighborhood-disadvantage metrics accessible—The Neighborhood Atlas. N Engl J Med. 2018;378(26):2456–2458. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Knighton AJ, Savitz L, Belnap T, Stephenson B, VanDerslice J. Introduction of an Area Deprivation Index Measuring Patient Socioeconomic Status in an Integrated Health System: Implications for Population Health. EGEMS. 2016;4(3):1238. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Mora J, Krepline AN, Aldakkak M, Christians KK, George B, Hall WA, et al. Adjuvant therapy rates and overall survival in patients with localized pancreatic cancer from high Area Deprivation Index neighborhoods. Am J Surg. 2021;222(1):10–17. [DOI] [PubMed] [Google Scholar]
- 12.Boutrous ML, Tian Y, Brown D, Freeman CA, Smeds MR. Area Deprivation Index Score is Associated with Lower Rates of Long Term Follow-up after Upper Extremity Vascular Injuries. Ann Vase Surg. 2021;75:102–108. [DOI] [PubMed] [Google Scholar]
- 13.Phillips AR, Andraska EA, Reitz KM, Habib S, Martinez-Meehan D, Dai Y, et al. Association between neighborhood deprivation and presenting with a ruptured abdominal aortic aneurysm before screening age. J Vasc Surg. 2022; 18: S0741-5214(22)00449-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. BMJ. 2007;335(7624):806–808. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Health UoWSoMaP. Neighborhood Atlas. Univeristy of Wisconsin School of Medicine and Public Health; 2020. Accessed https://www.neighborhoodatlas.medicine.wisc.edu [Google Scholar]
- 16.Soden PA, Zettervall SL, Deery SE, Hughes K, Stoner MC, Goodney PP, et al. Black patients present with more severe vascular disease and a greater burden of risk factors than white patients at time of major vascular intervention. J Vase Surg. 2018. Feb;67(2):549–556.e3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Arya S, Binney Z, Khakharia A, Brewster LP, Goodney P, Patzer R, et al. Race and Socioeconomic Status Independently Affect Risk of Major Amputation in Peripheral Artery Disease. J Am Heart Assoc. 2018. Jan 12;7(2):e007425. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Demsas F, Joiner MM, Telma K, Flores AM, Teklu S, Ross EG. Disparities in peripheral artery disease care: A review and call for action. Semin Vasc Surg. 2022. Jun;35(2):141–154. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Criqui MH, Aboyans V. Epidemiology of peripheral artery disease. Circ Res. 2015. Apr 24;116(9):1509–26. [DOI] [PubMed] [Google Scholar]
- 20.Barnes JA, Eid MA, Creager MA, Goodney PP. Epidemiology and Risk of Amputation in Patients With Diabetes Mellitus and Peripheral Artery Disease. Arterioscler Thromb Vasc Biol. 2020. Aug;40(8):1808–1817. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Harris SK, Roos MG, Landry GJ. Statin use in patients with peripheral arterial disease. J Vasc Surg. 2016. Dec;64(6):1881–1888. [DOI] [PubMed] [Google Scholar]
- 22.Bonaca MP, Hess CN. ASCVD Risk and Statin Use in PAD: Implementing a New Approach to an Old Problem. J Am Coll Cardiol. 2020. Jul 21;76(3):265–267. [DOI] [PubMed] [Google Scholar]
- 23.Latkin CA, Kennedy RD, Davey-Rothwell MA, Tseng TY, Czaplicki L, Baddela A, et al. The Relationship Between Neighborhood Disorder and Barriers to Cessation in a Sample of Impoverished Inner-City Smokers in Baltimore, Maryland, United States. Nicotine Tob Res. 2018. Nov 15;20(12):1451–1456. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Ultee KH, Bastos Gonçalves F, Hoeks SE, Rouwet EV, Boersma E, Stolker RJ, et al. Low Socioeconomic Status is an Independent Risk Factor for Survival After Abdominal Aortic Aneurysm Repair and Open Surgery for Peripheral Artery Disease. Eur J Vasc Endovasc Surg. 2015. Nov;50(5):615–22. [DOI] [PubMed] [Google Scholar]
- 25.de Jager E, Gunnarsson R, Ho YH. Disparities in Advanced Peripheral Arterial Disease Presentation by Socioeconomic Status. World J Surg. 2022. Jun;46(6):1500–1507. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Jelani QU, Jhamnani S, Spatz ES, Spertus J, Smolderen KG, Wang J, et al. Financial barriers in accessing medical care for peripheral artery disease are associated with delay of presentation and adverse health status outcomes in the United States. Vasc Med. 2020. Feb;25(1):13–24. [DOI] [PubMed] [Google Scholar]
- 27.Redmond ML, Dong F, Goetz J, Jacobson LT, Collins TC. Food Insecurity and Peripheral Arterial Disease in Older Adult Populations. J Nutr Health Aging. 2016;20(10):989–995. [DOI] [PubMed] [Google Scholar]
- 28.Smith EJT, Ramirez JL, Wu B, Zarkowsky DS, Gasper WJ, Finlayson E, et al. Living in a Food Desert is Associated with 30-day Readmission after Revascularization for Chronic Limb-Threatening Ischemia. Ann Vasc Surg. 2021. Jan;70:36–42. [DOI] [PubMed] [Google Scholar]
- 29.To MJ, Brothers TD, Van Zoost C. Foot Conditions among Homeless Persons: A Systematic Review. PLoS One. 2016. Dec 9;11(12):e0167463. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Kirby JB, Kaneda T. Neighborhood socioeconomic disadvantage and access to health care. J Health Soc Behav. 2005;46(1):15–31. [DOI] [PubMed] [Google Scholar]
- 31.Prentice JC. Neighborhood effects on primary care access in Los Angeles. Soc Sci Med. 2006;62(5):1291–1303. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Zhang Y, Ancker JS, Hall J, Khullar D, Wu Y, Kaushal R. Association Between Residential Neighborhood Social Conditions and Health Care Utilization and Costs. Med Care. 2020;58(7):586–593. [DOI] [PubMed] [Google Scholar]
- 33.Vemulapalli S, Greiner MA, Jones WS, Patel MR, Hernandez AF, Curtis LH. Peripheral arterial testing before lower extremity amputation among Medicare beneficiaries, 2000 to 2010. Circ Cardiovasc Qual Outcomes. 2014. Jan;7(1):142–50. [DOI] [PubMed] [Google Scholar]
- 34.Hughes K, Sehgal N. Racial/ethnic Disparities in Lower Extremity Amputation Vs Revascularization: A Brief Review. J Natl Med Assoc. 2018;110(6):560–563. [DOI] [PubMed] [Google Scholar]
- 35.Lee SD, Xiang J, Kshirsagar AV, Steffick D, Saran R, Wang V. Supply and Distribution of Vascular Access Physicians in the United States: A Cross-Sectional Study. Kidney360. 2020. Aug;1(8):763–771. [DOI] [PMC free article] [PubMed] [Google Scholar]