Article Text

Download PDFPDF

Original research
Racial and economic disparities in the access to treatment of unruptured intracranial aneurysms are persistent problems
  1. Lorenzo Rinaldo1,
  2. Alejandro A Rabinstein2,
  3. Harry J Cloft2,
  4. John M Knudsen3,
  5. Giuseppe Lanzino1,3,
  6. Leonardo Rangel Castilla1,3,
  7. Waleed Brinjikji1,3
  1. 1 Department of Neurosurgery, Mayo Clinic, Rochester, Minnesota, USA
  2. 2 Department of Neurology, Mayo Clinic, Rochester, Minnesota, USA
  3. 3 Department of Radiology, Mayo Clinic, Rochester, Minnesota, USA
  1. Correspondence to Dr Waleed Brinjikji, Department of Neurosurgery, Mayo Clinic, Rochester, MN 55902, USA; Brinjikji.Waleed{at}mayo.edu

Abstract

Background and purpose Previous studies have documented disparate access to cerebrovascular neurosurgery for patients of different racial and socioeconomic backgrounds. We further investigated the effect of race and insurance status on access to treatment of unruptured intracranial aneurysms (UIAs) and compared it with data on patients with aneurysmal subarachnoid hemorrhage (aSAH).

Methods Through the use of a national database, admissions for clipping or coiling of an UIA and for aSAH were identified. Demographic characteristics of patients were characterized according to age, sex, race/ethnicity, and insurance status, and comparisons between patients admitted for treatment of an UIA versus aSAH were performed.

Results There were 10 545 admissions for clipping or coiling of an UIA and 33 166 admissions for aSAH between October 2014 and July 2018. White/non-Hispanic patients made up a greater proportion of patients presenting for treatment of an UIA than those presenting with aSAH (64.3% vs 48.2%; P<0.001), whereas black/Hispanic patients presented more frequently with aSAH than for treatment of an UIA (29.3% vs 26.1%; P=0.006). On multivariate linear regression analysis, the proportion of patients admitted for management of an UIA relative to those admitted for aSAH increased with the proportion of patients who were women (P<0.001) and decreased with the proportion of patients with a black/Hispanic background (P=0.010) and those insured with Medicaid or without insurance (P=0.003).

Conclusion For patients with UIAs, racial, ethnic, and socioeconomic backgrounds appear to continue to influence access to treatment.

  • aneurysm
  • subarachnoid
  • hemorrhage
View Full Text

Statistics from Altmetric.com

Introduction

Racial, ethnic, and socioeconomic disparities with regard to the accessibility of and outcomes after treatment of intracranial aneurysms is a documented phenomenon.1–5 Increasing awareness of this issue is critically important, not only to facilitate identification of shortcomings in healthcare available to minority and disadvantaged populations, but also to prompt correction of institutional biases, both at the individual and structural levels, that may influence quality of care.

While admission for aneurysmal subarachnoid hemorrhage (aSAH) is usually an emergency dictated by the severity and acuity of the disease, diagnosis and treatment of an unruptured intracranial aneurysm (UIA) is often purely elective. Thus identifying demographic, ethnic, and socioeconomic differences between patients receiving treatment of UIAs and those suffering from aSAH may disclose disparities in access to care across different groups.

In this study, we used the Vizient Clinical Data base/Resource Manager (CDB/RM) database to compare demographic characteristics of patients receiving treatment for UIAs and those treated for aSAH. We hypothesized that minorities and patients with poorer insurance (ie, uninsured/Medicaid) would have a higher representation in the aSAH group than in the UIA group, suggesting the presence of socioeconomic disparities in the treatment of UIAs.

Materials and methods

National database

The Vizient Clinical Data base/Resource Manager (CDB/RM) is a national database containing information on outcomes from inpatient admissions to over 400 healthcare institutions in the USA. Institutions voluntarily participate in the CDB/RM and send billing information for an unselected sample of inpatient admissions to the database on a continuous basis. After the delivered information is subjected to a quality control process, the CDB/RM quantifies various metrics associated with value and quality of healthcare delivery (eg, length of stay, mortality rate, etc), data which are then accessible to individuals at member institutions.

The CDB/RM can be queried to obtain information on admissions due to a specific diagnosis or to undergo a specific procedure using the International Classification of Disease, 9th and 10th editions' (ICD-9 and ICD-10) codes. The proportion of patients with a particular demographic background or comorbidity profile can also be determined. Results of a query to the CDB/RM include the number of admissions to each reporting institution matching the desired description over a specified time period along with mean values and/or total incidence of numerous outcomes, such as length of stay, complications, and mortality. The organization of query output by institution is advantageous as it allows a comparison of patients with different demographic backgrounds presenting to the same institution (see ’Statistical analysis' section below for details on methodology of comparative statistics). Institutional paradigms for the management of UIAs likely influence the decision to pursue UIA treatment versus observation. Moreover, practice patterns could also influence rates of aneurysm detection. These factors could bias an analysis on the effect of race on outcome of intracranial aneurysms if a particular demographic group is more likely than others to present to certain institutions. Comparing patients of different backgrounds treated at the same institution mitigates potential biases introduced by differences in institutional practice patterns, allowing us to better focus on the effect of patient demographics. This study did not require approval from our institutional review board given that the data were obtained exclusively from the database containing completely anonymized patient information.

Variables of interest

Patient admissions for clipping or coiling of an UIA occurring between October 2014 and July 2018 were identified using ICD-9 and ICD-10 codes. ICD-9 and ICD-10 codes for an UIA were 437.3 and I67.1, respectively. ICD-9 codes for clipping and coiling were 39.51 and 39.52, 39.72, 39.79, and ICD-10 codes were 03VG0CZ and 03LG3DZ, respectively. ICD-9 and ICD-10 codes for admission due to aSAH were 430 and variations of I60, respectively. Additional filters were applied to isolate admissions according to age, sex, race, ethnicity, and insurance status. Regarding age and sex, queries were performed to isolate admissions of female patients and those aged ≥65 years of age. Categories for race and ethnicity were white and non-Hispanic and black or Hispanic, while categories for insurance status included commercial or private insurance, Medicare, and Medicaid or uninsured patients. Finally, information on the prevalence of hypertension and cigarette smoking within populations of interest was collected given the association of these comorbidities with an increased risk of aneurysm rupture.6 7 ICD-9 and ICD-10 codes for hypertension and cigarette smoking were 401.0, 401.1, 401.9, and I10 and 305.1, 305.10, 305.11, 305.12, 305.13, and variations of F17, respectively. Hospital characteristics of interest were American Association of Medical Colleges teaching status and whether the hospital was designated as a critical access hospital. Comparisons were limited to data from institutions treating both patients with UIAs and aSAH.

Statistical analysis

The number of admissions for clipping or coiling of an UIA and aSAH were noted. Descriptive statistics for patient subgroups admitted for treatment of an UIA or aSAH were given as a frequency and percentage of total admissions (table 1). Comparative statistics were performed by comparing the proportion of total admissions at a given institution made up by different patient subgroups (tables 2 and 3). For example, in the comparison of white/non-Hispanic patients to black/Hispanic patients, the percentage of white/non-Hispanic patients who were women was compared with the percentage of Black/Hispanic patients who were women at each reporting institution. All such analyses were performed using the paired Student’s t test. Variables independently associated with admission for treatment of an UIA were identified using a multivariate linear regression model. The dependent variable in our model was the percentage of admissions for management of an UIA (with the sum of admissions for treatment of an UIA and aSAH serving as the denominator), while the independent variables were the percentage of individuals admitted to each institution meeting a specific criterion (eg, the percentage of patients who were black or Hispanic).

Variables that were significantly different between patients with an UIA and aSAH were included in the multivariate model, which was weighted according to the total number of admissions at each reporting institution. Results of the multivariate model were presented as a beta coefficient, describing the magnitude and directionality of change in the dependent variable for every 1 unit change in the independent variable, and t value, denoting the likelihood that the absolute value of the beta coefficient was >0 was statistically significant. A correlation coefficient (r2) was also provided for the overall model. The alpha level for statistical significance was 0.05. All analyses were performed using commercially available software (JMP 10.0.0, 2012 SAS Institute Inc, Cary, North Carolina, USA).

Table 1

Demographics of admissions for clipping or coiling of unruptured  intracranial  aneurysms and aneurysmal  subarachnoid   hemorrhage

Table 2

Comparison of white/non-Hispanic and black/Hispanic patients

Table 3

Comparison of patients treated with aneurysmal subarachnoid hemorrhage and treated unruptured intracranial aneurysms

Results

Demographics of patients admitted for clipping and coiling of an UIA and aSAH

There were 10 545 admissions for clipping or coiling of an UIA at 152 healthcare institutions over the time period of interest. Most reporting institutions were designated as American Association of Medical Colleges teaching hospitals (81.6%). A majority of patients were treated with coiling (52.3%). Most patients were women (73.1%), with a minority of patients were older than 65 years of age (29.5%). The most common type of insurance was commercial/private (42.6%). The percentage of total admissions made up by white/non-Hispanic and black/Hispanic patients were 65.6% and 22.5%, respectively.

At the same 152 institutions reporting outcomes for treatment of an UIA, there were 33 166 admissions for aSAH over the same time period. The percentage of total admissions made up by white/non-Hispanic and black/Hispanic patients in this population were 49.3% and 27.5%, respectively. Additional information on the demographics of admissions for treatment of an UIA and aSAH is shown in table 1.

Comparison of white/non-Hispanic and black/Hispanic patients

Characteristics of white/non-Hispanic and black/Hispanic patients were subsequently compared. Commercial/private insurance was less common among black/Hispanic patients (32.4% vs 42.9%; P<0.001), whereas Medicaid or uninsured status was more common among black/Hispanic patients (38.6% vs 16.5%; P<0.001). Both hypertension (58.6% vs 48.7%; P<0.001) and cigarette smoking (24.8% vs 21.2%; P=0.004) were more prevalent in black/Hispanic patients. Finally, admission for aSAH was more common among black/Hispanic patients than it was for white/non-Hispanic patients (85.5% vs 83.1%; P=0.008). Results of the comparisons between white/non-Hispanic and black/Hispanic patients are presented in table 2.

Comparison of patients admitted for treatment of UIA versus aSAH

We then compared the demographic makeup of patients admitted for treatment of an UIA versus those admitted for aSAH. The percentage of patients who were women was greater among patients admitted for treatment of an UIA relative to those admitted for aSAH (72.2% vs 60.8%; P<0.001). The percentage of patients who were ≥65 years of age was lower among patients admitted for an UIA (28.7% vs 32.0%; P=0.003), as was the percentage of patients with Medicaid or without insurance (21.8% vs 24.5%; P=0.006). Patients with a white/non-Hispanic background were less common among patients admitted for aSAH than those admitted for treatment of an UIA (48.2% vs 64.3%; P<0.001), whereas the opposite was true for black/Hispanic patients (29.3 vs 26.1%; P=0.006) (table 3).

Multivariate analyses

Variables independently associated with admission for treatment of an UIA were identified in a multivariate linear regression model. Female sex was positively correlated with admission for an UIA (P<0.001), while insurance with Medicaid or no insurance (P=0.003) and black/Hispanic race/ethnicity were negatively correlated with admission for an UIA. Results of both multivariate analyses are shown in table 4.

Table 4

Multivariate linear regression analysis identifying predictors of having a treated unruptured intracranial aneurysm*

Discussion

We investigated whether certain demographic characteristics were more prevalent in patients presenting for treatment of an UIA versus aSAH using a large national database. Black and/or Hispanic patients or those with limited or no insurance were more likely to present with aSAH as opposed to being admitted for treatment of an UIA. These findings suggest that minority and socioeconomically disadvantaged patients may have reduced access to the type of healthcare needed for UIA diagnosis and treatment.

The higher proportion of black/Hispanic patients and patients with Medicaid or no insurance among those presenting with aSAH relative to those presenting for treatment of an UIA may be explained by less frequent recognition of intracranial aneurysms before rupture in those subsets of patients or to a lower likelihood that these patients are offered elective treatment for UIAs. It would be of interest to investigate whether differences in the proportion of patients presenting with aSAH from a known UIA were present in different racial and ethnic groups. The preferential treatment of high risk aneurysms in certain demographic populations but not others would represent an indefensible disparity, the rectification of which would merit urgent prioritization. On the other hand, it is also likely that UIAs are simply discovered less often in minority populations. The most common indication for radiographic studies on which incidental aneurysms are detected is non-specific neurologic symptoms, such as headache.8 Patients with limited access to care, or with inadequate insurance, will likely undergo investigative radiographic evaluation less frequently, leading to lower aneurysm detection rates.

The disparities in treatment of UIAs observed in the current analysis of admissions taking place between 2014 and 2018 mirror those noted in earlier intervals (2000–20101 and 2001–20092). Unfortunately, this suggests that no progress has been made in reducing the gap in access to treatment of UIAs among minorities and those with limited or no insurance over the past decade. While it is incumbent on neurosurgical and neurointerventional providers to recognize and ameliorate healthcare disparities, these disparities likely arise at least in part at the primary care level prior to specialty referral. While the treatment of incidentally discovered aneurysms is controversial, metrics exist to help quantify the risk of rupture and determine appropriate candidates for treatment,7 and actuarial analyses suggest an overall benefit to the treatment of UIAs when considering both the risks of UIA rupture and treatment.9 This benefit should not be exclusive to traditionally advantaged populations.

In our analysis, patients insured with Medicaid and patients without insurance were grouped together, a combination that was deemed appropriate given the results of prior studies documenting the lower likelihood of UIA treatment in both of these groups.3 Nevertheless, contemporary studies comparing rate of aneurysm detection and treatment between patients insured with Medicaid and uninsured patients would be of interest, particularly given the recent expansion of Medicaid codified by the Affordable Care Act. While the effect of insurance with Medicaid over uninsured status on the access to intracranial imaging has not been directly studied, there is ample evidence to suggest that Medicaid increases access to other important diagnostic studies, for example cancer screening with mammography or pap smears,10 11 suggesting a potential benefit associated with obtaining Medicaid for patients with undetected UIAs. Relevant to our study, however, the benefit of Medicaid expansion may not be distributed evenly across all racial and ethnic groups,12 indicating additional barriers to care beyond insurance status.

Limitations

The limitations of our study are typical of those employing the use of national databases, including procedural and diagnosis coding errors.13 In addition, similarly to other databases, for example the Nationwide Inpatient Sample, the CDB/RM does not collect information on all admissions to a particular institution over a given time period, but rather an unselected sample. It is thus possible that selection bias affected our results. Our hope is that the large sample size of this study will mitigate these issues. Another limitation is the fact that little is known regarding differences in the natural history of UIAs in US minorities as well as the association between demographics and aneurysm natural history. It is possible that the reason why black and Hispanic race was independently associated with treatment of aSAH was due to differences in the natural history of unruptured aneurysms (ie, higher propensity to rupture).

Conclusion

Our study suggests the persistence of disparate access to UIA treatment for patients of different racial and ethnic backgrounds. Additional work is needed to understand the underlying cause of these disparities and how best to correct them.

References

View Abstract

Footnotes

  • Contributors LR: study conception and design, data collection, statistical analysis, drafting the manuscript, critical revision of the manuscript, and approved the final manuscript. AAR, HJC, JMK, GL, LRC, and WB: study conception and design, critical revision of the manuscript, and approved the final manuscript.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial, or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Patient consent for publication Not required.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.