Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Jul 1.
Published in final edited form as: J Am Coll Radiol. 2022 Apr 25;19(7):854–865. doi: 10.1016/j.jacr.2022.03.008

Updated trends, disparities, and clinical impact of neuroimaging utilization in ischemic stroke in the Medicare population: 2012-2019

Jason J Wang 1,2, Casey E Pelzl 3, Artem Boltyenkov 1,4, Jeffrey M Katz 2,5, Jennifer Hemingway 3, Eric W Christensen 3, Elizabeth Rula 3, Pina C Sanelli 1,2
PMCID: PMC9308737  NIHMSID: NIHMS1796889  PMID: 35483436

Abstract

Objective:

The purpose of this study was to update trends, investigate sociodemographic disparities, and evaluate the impact on mortality of stroke neuroimaging across the United States from 2012-2019.

Methods:

Retrospective cohort study using CMS Medicare 5% Research Identifiable Files, representing consecutive ischemic stroke emergency department/hospitalized patients aged ≥65 years. A total of 85,547 stroke episodes with demographic and clinical information were analyzed using Cochran-Mantel-Haenszel tests and logistic regression. Outcome measures were neuroimaging (CTA, CTP, MRI, and MRA) utilization, acute treatment (endovascular thrombectomy [EVT] and intravenous thrombolysis [IVT]), and mortality as in-hospital, 30-day, and 1-year post-discharge.

Results:

Significantly increasing utilization trends for CTA (250%), CTP (428%), and MRI (18%), and a decreasing trend for MRA (−33%) were observed from 2012-2019 (p<0.0001). Controlling for covariates in the logistic regression models, CTA and CTP were significantly associated with higher EVT and IVT utilization. While CTA, MRI, and MRA were associated with lower mortality, CTP was associated with higher mortality post-discharge. Less neuroimaging was performed in rural patients; older patients (≥80 years) had lower utilization of CTA, MRI, and MRA, female patients had lower rates of CTA, and Black patients had lower utilization of CTA and CTP.

Conclusions:

CTA and CTP utilization increased in the Medicare ischemic stroke population from 2012 to 2019 and both were associated with greater EVT and IVT use. However, disparities exist in neuroimaging utilization across all demographic groups, and further understanding of the root causes of these disparities will be crucial to achieving equity in stroke care.

Keywords: neuroimaging, angiography, perfusion, ischemic stroke, trend analysis

Summary Sentence:

A significant upward trend in CTA, CTP, and MRI utilization, and a simultaneous downtrend in MRA use was observed for ischemic stroke cases while disparities existed in neuroimaging utilization across all demographic groups.

Introduction

In the past decade, the United States population experienced demographic changes impacting stroke care, such as growth in the elderly population and continued increase in the prevalence of obesity and diabetes mellitus, which are stroke risk factors. At the same time, advances in stroke care made life-saving medical and endovascular treatments available to more patients than in the previous decade. Every year, approximately 795,000 patients suffer a stroke in the United States, with ischemic stroke accounting for 87% of these.1 Stroke risk increases with age, doubling every ten years after 55.1 Approximately three-quarters of all strokes occur in persons aged ≥65 years, which corresponds to the Medicare population.2

Intravenous thrombolysis (IVT) with recombinant tissue-type plasminogen activator and endovascular thrombectomy (EVT) are the only early treatments for ischemic stroke endorsed by the American Heart Association (AHA).3 Several studies have shown favorable clinical outcomes with IVT4-6 and EVT,7,8 using the results of advanced imaging. Computed tomography (CT) and magnetic resonance imaging (MRI), including angiography and perfusion studies, provide valuable information to select patients beyond the previously limited therapeutic time-windows for ischemic stroke treatment. In the United States, overall use of IVT and EVT in stroke cases increased from 7.0% in 2008 to 19.1% in 2018.9 Concomitantly, the utilization of CT and MRI increased markedly in ischemic stroke patients.10, 11 From 2006 to 2010, CT angiography (CTA) increased by 142% and CT perfusion (CTP) grew by 6,429%.10 Additionally, MRI use increased by 235% from 1999 to 2008.11 Published trends in nationwide neuroimaging and treatment utilization, particularly relating to other components of stroke care, do not exist beyond 2012.10,12 Furthermore, the literature does not address socioeconomic disparities in stroke imaging.10,12

Understanding recent trends and socioeconomic disparities in neuroimaging utilization among ischemic stroke patients, and how imaging relates to clinical treatment decisions and outcomes, is critically important for clinicians and healthcare policymakers as we strive to ensure optimal and equitable access to high quality stroke care across an increasingly diverse and aging population. We hypothesized that advanced neuroimaging (CTA, CTP, MRI, MR angiography [MRA]) utilization in ischemic stroke patients has continued to increase in the United States from 2012 to 2019. Furthermore, we hypothesized that (1) these trends were associated with increased acute treatment utilization and improved clinical outcomes, and (2) sociodemographic disparities associated with ischemic stroke imaging utilization exist. We focused our analyses on the Medicare population given the much higher incidence of stroke compared to commercially insured patients.13

Materials and Methods:

Study population and data collection

We performed a retrospective study to assess trends in the utilization of advanced neuroimaging in ischemic stroke care using the Medicare 5% Research Identifiable Files (RIF, 2012-2019). The RIF data contain individual level fee-for-service administrative claims data for a 5% nationally representative sample of Medicare beneficiaries across all places of service (i.e., emergency departments, inpatient and outpatient facilities, skilled nursing and hospice facilities, and home health agencies). Unique patient and facility identifiers were used to link claims of individual patients to produce episodes of care comprising all services associated with the stroke episode.

Using International Classification of Disease - Clinical Modification, 9th revision, (ICD-9CM) discharge diagnosis codes (International Classification of Disease 9/10: 433.x1, 434.x1, 436, I63.xx14) we identified patients aged ≥65 years with ischemic stroke episodes presenting in an emergency department or acute care inpatient setting. Patients with intracranial hemorrhage and transient ischemic attack (TIA) were excluded. Sociodemographic and clinical characteristics were categorized as: (a) sociodemographic: age (65-79 and ≥80 years), biological sex (female, male), race (Asian, Black, white, Hispanic, other, North American Native, unknown), median household income (MHI [<$60,000/year, ≥$60,000/year, unknown]) based on patient zip codes reported by the United States Census Bureau,15 supplemental insurance (yes, no), census region (Midwest, Northeast, South, West, other),16 and geography (urban [all metropolitan and micropolitan areas], rural [all areas not defined as metropolitan or micropolitan]);16 (b) clinical: Administrative Data Stroke Scale (ADSS [0, 1-2, >2]) was used as a surrogate for stroke severity,17 Charlson Comorbidity Index (CCI [0-1, ≥2]),18 arrival by transfer from another facility (transfer, non-transfer); (c) resource utilization: advanced neuroimaging (CTA, CTP, MRI, MRA), and acute treatment (IVT only, EVT with or without IVT, none), and (d) mortality: in-hospital, 30-day post-discharge, 1-year post-discharge. Race and ethnicity are combined in a single category in Medicare claims data and the individual groups are mutually exclusive. The data on race/ethnicity is self-reported by the patient and/or caregivers.

Statistical analyses

Patient sociodemographic and clinical characteristics, neuroimaging, treatment, and mortality were analyzed as categorical variables, and proportions were calculated at the episode level. Advanced neuroimaging utilization trends were assessed by year and stratified by patient sociodemographic characteristics. Cochran-Armitage and Cochran-Mantel-Haenszel tests were used to assess neuroimaging utilization trends over time.

Multivariable logistic regression analyses were performed to assess the associations and potential disparities between neuroimaging types. Each neuroimaging type served as individual dependent variables with patient characteristics and discharge years (2012-2019) as the independent variables. To assess associations between neuroimaging types and acute treatment (IVT only, EVT with/without IVT) utilization, chi-square tests were performed. Furthermore, we performed additional multivariable logistic regression analyses using treatment type as the dependent variable and patient characteristics, discharge years, and neuroimaging types as the independent variables. The associations between mortality and patient sociodemographic characteristics, neuroimaging and treatment types, and discharge years were analyzed in separate regression models. Odds ratios (OR), 99% confidence intervals (CI), and p-values were estimated. Given the large sample size, statistical significance was assigned at an alpha level of 0.01. All analyses were generated using SAS software version 9.4 (©2016, SAS Institute; Cary, NC).

Results:

During the 2012-2019 study period, there were 85,547 stroke episodes in patients aged 65 years and older. Sociodemographic and clinical characteristics of patients with stroke episodes during this period are shown in Table 1. CTA was performed in 29.7%, CTP in 4.2%, MRI in 68.6%, and MRA in 26.8% of stroke episodes. From 2012-2019, we observed significant increasing trends in utilization of CTA (15.7% to 54.9%, P<0.0001), CTP (2.2% to 11.4%, P<0.0001), and MRI (62.1% to 73.7%, P<0.0001), and a decreasing trend in utilization of MRA (29.7% to 19.8%, P<0.0001) (Table 2, Figure 1).

Table 1.

Sociodemographic and clinical characteristics, imaging and treatment utilization, and mortality for ischemic stroke in Medicare sample from 2012-2019

Characteristics Categories Total
N %
Total patient visits in Medicare 5% RIF 21,014,654 --
Total ischemic stroke episodes in Medicare cohort 85,547 0.41%
Age (years) 65-79 40,869 47.8%
≥80 44,678 52.2%
Sex Female 49,891 58.3%
Male 35,656 41.7%
Race White 70,821 82.8%
Black 9,762 11.4%
Other 1,142 1.3%
Asian 1,424 1.7%
Hispanic 1,393 1.6%
North American Native 408 0.5%
Unknown 597 0.7%
Census Region Midwest 20,549 24.0%
Northeast 15,999 18.7%
South 35,009 40.9%
West 13,632 15.9%
Othera 358 0.4%
Geography Urban 65,539 76.6%
Rural 20,008 13.8%
Median Household Income (2019 dollars) <$60,000 44,596 52.1%
≥$60,000 36,977 43.2%
Unknown 3,974 4.7%
Supplemental Insuranceb Yes 1,846 2.2%
No 83,701 97.8%
Transfer Status Arrived via transfer 2,223 2.6%
ADSS Score 0 48,050 56.2%
1-2 24,772 29.0%
>2 12,725 14.9%
Charlson Comorbidity Index 0-1 38,119 44.6%
≥2 47,428 55.4%
Neuroimaging CTA 25,436 29.7%
CTP 3,611 4.2%
MRI 58,667 68.6%
MRA 22,884 26.8%
Treatment EVT (+/− IVT) 2,112 2.5%
IVT Only 5,997 7.0%
No Treatment 77,438 90.5%
Mortality (In-hospital) Alive 81,506 95.3%
Died 4,041 4.7%
Mortality (30-day) Alive 72,527 84.8%
Died 13,020 15.2%
Mortality (1-year) Alive 59,857 70.0%
Died 25,690 30.0%
Open in a new tab
a

Census region: Other category encompasses Guam, Puerto Rico, and other U.S. outlying islands.

b

Supplemental insurance includes Medicare Part C or Private HMO.

RIF - Research Identifiable Files; ADSS - Administrative Data Stroke Scale; CTA - CT angiography; CTP - CT perfusion; MRI - magnetic resonance imaging; MRA - magnetic resonance angiography; EVT - endovascular therapy; IVT - intravenous thrombolysis.

Table 2.

Neuroimaging utilization trends for ischemic stroke in Medicare sample from 2012-2019

Categories Total
N		 Total
%		 Year p-value
2012 2013 2014 2015 2016 2017 2018 2019
Total Ischemic Stroke Episodes 85,547 100.0% 11347 11259 11180 11296 10847 10683 10015 8920
CTA 25,436 29.7% 15.7% 17.7% 19.9% 24.8% 30.5% 36.9% 44.8% 54.9% <0.0001*
CTP 3,611 4.2% 2.2% 2.3% 2.3% 2.7% 3.1% 4.6% 7.1% 11.4% <0.0001*
MRI 58,667 68.6% 62.1% 65.7% 66.7% 68.0% 70.4% 71.5% 72.3% 73.7% <0.0001*
MRA 22,884 26.8% 29.7% 30.4% 29.5% 27.7% 27.2% 25.3% 22.5% 19.8% <0.0001*
Open in a new tab
*

Denotes statistical significance at p<0.01. Cochran-Armitage and Cochran-Mantel-Haenszel tests were performed.

CTA - computed tomography angiography; CTP - CT perfusion; MRI - magnetic resonance imaging; MRA - magnetic resonance angiography

Figure 1. Trends in the neuroimaging utilization, treatment methodology, and mortality for ischemic stroke in Medicare sample from 2012-2019.

Figure 1.

Open in a new tab

CTA - computed tomography angiography; CTP - CT perfusion; MRI - magnetic resonance imaging; MRA - magnetic resonance angiography

Our analyses of neuroimaging trends stratified by patient sociodemographic characteristics (Table S1-S4) demonstrated significant increasing trends in the utilization of CTA, CTP, and MRI, with decreasing trends in MRA across all sociodemographic and clinical categories (P<0.001). Qualitatively, the gaps between age and sex categories narrowed within all neuroimaging types from 2012-2019 (Figure 2). Nevertheless, other disparities persisted through the study period. Black patients tended to have lower utilization of CTA and CTP than white patients, while also having higher MRI and MRA use. Although urban patients had higher rates of neuroimaging utilization compared to rural patients, the rapidly increasing rate of CTA and CTP in urban patients widened the gap between them and rural patients even further by 2019. Patients with moderate-to-severe stroke severity (i.e., ADSS 1-2 and ADSS >2) had increased rates of CTA and CTP utilization and decreased rates of MRI and MRA utilization compared to those with mild stroke severity (ADSS 0). Patients from lower MHI areas tended to have higher CTP use and lower MRA use compared to the higher MHI areas.

Figure 2. Trends in the neuroimaging utilization rates for CTA, CTP, MRI, and MRA stratified by patient characteristic for ischemic stroke in Medicare sample from 2012-2019.

Figure 2.

Open in a new tab

CTA - computed tomography angiography; CTP - CT perfusion; MRI - magnetic resonance imaging; MRA - MR angiography; MHI - median household income

After adjusting for the demographic and clinical characteristics in the logistic regression model, the yearly increasing trends in CTA and MRI utilization remained significant (Table 3). Despite the overall increasing trend for CTP and declining trend for MRA, the yearly increase in CTP utilization was only significant from 2017-2019 and the yearly decrease in MRA was only significant from 2015-2019, after controlling for patient characteristics. Older patients were less likely to have CTA (OR [99% CI]: 0.694[0.665-0.724]), MRI (0.679 [0.652-0.707]), and MRA (0.743[0.713-0.775]) than patients aged 65-79. Female patients were less likely to have CTA (0.851[0.816-0.888]) than male patients. Black patients were less likely to have CTA (0.820[0.7363-0.914]) and CTP (0.754 [0.592-0.961]) than patients of other races. Patients with MHI <$60,000 were more likely to have CTA (1.076 [1.027-1.127]) and CTP (1.452 [1.312-1.576]), but less likely to have MRI (0.953[0.911-0.996]) and MRA (0.823[0.786-0.861]) than patients with MHI ≥$60,000.

Table 3.

Patient characteristics associated with imaging utilization for ischemic stroke in Medicare sample from 2012-2019 using logistic regression

Imaging CTA CTP MRI MRA
99% CI 99% CI 99% CI 99% CI
Parameter OR Lower
Limit		 Upper
Limit		 p-value OR Lower
Limit		 Upper
Limit		 p-value OR Lower
Limit		 Upper
Limit		 p-value OR Lower
Limit		 Upper
Limit		 p-value
2012 Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref
2013 1.143 1.041 1.255 0.0002* 1.066 0.844 1.345 0.4824 1.168 1.086 1.256 <0.0001* 1.024 0.950 1.105 0.4140
2014 1.335 1.218 1.463 <0.0001* 1.041 0.823 1.316 0.6631 1.252 1.163 1.347 <0.0001* 0.992 0.919 1.070 0.7798
2015 1.649 1.509 1.802 <0.0001* 1.055 0.841 1.323 0.5447 1.301 1.208 1.407 <0.0001* 0.896 0.830 0.967 0.0002*
2016 2.105 1.929 2.298 <0.0001* 1.159 0.928 1.448 0.0871 1.425 1.321 1.537 <0.0001* 0.853 0.789 0.922 <0.0001*
2017 2.810 2.578 3.063 <0.0001* 1.774 1.442 2.182 <0.0001* 1.516 1.404 1.636 <0.0001* 0.773 0.714 0.836 <0.0001*
2018 3.927 3.600 4.279 <0.0001* 2.737 2.247 3.334 <0.0001* 1.604 1.483 1.735 <0.0001* 0.666 0.613 0.724 <0.0001*
2019 5.871 5.375 6.413 <0.0001* 4.539 3.752 5.491 <0.0001* 1.696 1.563 1.840 <0.0001* 0.564 0.516 0.616 <0.0001*
Age ≥80 0.694 0.665 0.724 <0.0001* 0.949 0.865 1.042 0.1514 1.679 0.652 0.707 <0.0001* 0.743 0.713 0.775 <0.0001*
Female sex 0.851 0.816 0.888 <0.0001* 0.946 0.863 1.038 0.1220 1.098 1.054 1.143 <0.0001* 0.989 0.948 1.031 0.4752
White race 1.210 1.107 1.323 <0.0001* 1.143 0.943 1.385 0.0738 0.979 0.899 1.067 0.5311 0.826 0.758 0.900 <0.0001*
Black race 0.820 0.736 0.914 <0.0001* 0.754 0.592 0.961 0.0027* 0.978 0.883 1.084 0.5769 0.939 0.848 1.040 0.1111
All other races Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref
Region: South Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref
Region: Midwest 1.039 0.985 1.097 0.0660 1.046 0.929 1.178 0.3313 0.864 0.822 0.909 <0.0001* 0.801 0.759 0.845 <0.0001*
Region: Northeast 1.057 0.989 1.130 <0.0315 1.457 1.272 1.670 <0.0001* 0.849 0.797 0.904 <0.0001* 1.144 1.074 1.219 <0.0001*
Region: West 1.321 1.244 1.404 <0.0001* 1.226 1.076 1.398 <0.0001* 0.869 0.819 0.922 <0.0001* 0.781 0.734 0.831 <0.0001*
Region: Other 0.120 0.062 0.232 <0.0001* 0.256 0.041 1.622 0.0574 0.154 0.112 0.210 <0.0001* 0.245 0.129 0.360 <0.0001*
Urban location 1.756 1.648 1.870 <0.0001* 3.334 2.804 3.964 <0.0001* 1.349 1.278 1.424 <0.0001* 1.367 1.287 1.451 <0.0001*
MHI ≥$60,000 Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref
MHI <$60,000 1.076 1.027 1.127 <0.0001* 1.452 1.312 1.607 <0.0001* 0.953 0.911 0.996 0.0050* 0.823 0.786 0.861 <0.0001*
MHI unlisted 1.683 1.549 1.828 <0.0001* 1.877 1.576 2.235 <0.0001* 1.214 1.119 1.317 <0.0001* 1.049 0.966 1.139 0.1319
Supp insurance 0.279 0.230 0.340 <0.0001* 0.278 0.159 0.487 <0.0001* 0.163 0.142 0.188 <0.0001* 0.429 0.358 0.513 <0.0001*
Transfer 0.755 0.656 0.870 <0.0001* 0.754 0.531 1.074 0.0386 0.557 0.496 0.625 <0.0001* 0.785 0.683 0.901 <0.0001*
ADSS 0 Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref
ADSS 1-2 1.553 1.482 1.627 <0.0001* 2.394 2.145 2.671 <0.0001* 0.831 0.795 0.869 <0.0001* 0.876 0.837 0.918 <0.0001*
ADSS >2 2.270 2.144 2.403 <0.0001* 4.198 3.747 4.704 <0.0001* 0.582 0.551 0.615 <0.0001* 0.626 0.588 0.668 <0.0001*
CCI 0-1 Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref Ref
CCI ≥2 0.894 0.857 0.932 <0.0001* 0.987 0.900 1.082 0.7120 0.836 0.804 0.870 <0.0001* 0.906 0.869 0.943 <0.0001*
Open in a new tab
*

Denotes statistical significance at p<0.01. Logistic regression analyses were performed.

CTA - computed tomography angiography; CTP - CT perfusion; MRI - magnetic resonance imaging; MRA - MR angiography; CI - confidence interval; OR - odds ratio; ADSS - Administrative Data Stroke Scale; CCI - Charlson Comorbidity Index; MHI - median household income.

We found significant neuroimaging utilization disparities based on census regions and geography (Table 3). The Northeastern and Western regions had significantly higher CTA and CTP utilization relative to the Southern region, which represents the Stroke Belt of the United States. In contrast, the Southern region had higher utilization of MRI and MRA than all other regions, except for the Northeast, which was associated with the highest MRA use. Furthermore, rural patients had uniformly lower neuroimaging utilization compared to urban patients.

Neuroimaging types were significantly associated with acute stroke treatment. CTA and CTP were significantly associated with increased IVT (CTA: OR=3.41 [99% CI, 3.19-3.65]; CTP: 3.56[3.19-3.98]) and EVT (CTA: 10.81[9.35-12.49]; CTP: 10.07[8.81-11.52]) use, while MRI and MRA were significantly associated with decreased IVT (MRI: 0.78[0.73-0.84]; MRA: 0.66[0.61-0.72]) and EVT (MRI: 0.69[0.61-0.77]; MRA: 0.29[0.24-0.35]) use (Figure 3). Furthermore, logistic regression analysis demonstrated that CTA and CTP were significantly associated with acute treatment using IVT (CTA: OR=2.831 [99% CI, 2.621-3.057] and CTP: 1.545 [1.372-1.740]) and EVT (CTA: 5.460 [4.636-6.431] and CTP: 2.567 [2.209-2.984]), after controlling for patient sociodemographic characteristics and discharge years (Table S5). MRI was associated with lower EVT (OR=0.655 [99% CI, 0.572-0.749]) and IVT (0.719 [0.665-0.779]) utilization, while MRA was associated with lower EVT utilization (0.743 [0.602-0.918]).

Figure 3. Associations between neuroimaging and treatment utilization for ischemic stroke in Medicare sample from 2012-2019.

Figure 3.

Open in a new tab

IVT - intravenous thrombolysis; EVT - endovascular therapy; CTA - computed tomography angiography; CTP - CT perfusion; MRI - magnetic resonance imaging; MRA - magnetic resonance angiography

Further logistic regression analyses showed that patients who underwent CTA were 26% less likely to die within 30 days post-discharge (OR=0.739 [99% CI, 0.690-0.792]) and 31% less likely to die within one-year post-discharge (0.0.692 [CI, 0.654-0.733]) (Table S6) than patients who did not have CTA. Similarly, patients who had MRA were 29% less likely to die within 30 days post-discharge (0.714 [0.659-0.733]) and 30% less likely to die within one-year post-discharge (0.696 [0.656-0.738]) than patients who did not have MRA. Patients who received MRI were 66% less likely to die in-hospital (0.358 [0.323-0.397]), 56% less likely to die within 30 days post-discharge (0.438 [0.412-0.466]), and 42% less likely to die within one-year post-discharge (0.575 [0.546-0.605]) than patients who had no MRI. However, CTP was associated with increased likelihood of mortality within 30 days (1.182 [1.041-1.341]).

Discussion:

Using Medicare claims data from 2012 to 2019, we identified a marked increase in advanced neuroimaging utilization of CTA (250%) and CTP (428%) in ischemic stroke patients similar to, and a continuation of, the upward trends reported from 2006 to 2010.10 It is likely that these utilization trends are a consequence of several landmark clinical trials, starting in 2015, which showed the significant benefit of EVT19-24 and led to the AHA3 and other society guideline changes.25,26 The broader acceptance and standardization of criteria for EVT, including recommendations for pre-treatment CTA in the 0-6-hour window and CTA and CTP in the extended 6-24-hour window,7 has likely contributed to these increasing neuroimaging trends. The growth in overall CT use may also be ascribed to the wide availability and ease-of-access to CT scanners in emergency departments. In addition, recent technological advances, including faster CT scanners and automated postprocessing software, have enabled broader imaging of a larger spectrum of suspected ischemic stroke patients within and beyond treatment time windows. Our results suggest that the previously reported observations of the rapid assimilation of CTA and CTP into clinical practice for acute stroke patients10, 27 has continued over the past decade. It is worth noting that the growth of CT imaging utilization for Medicare patients is not limited to the ischemic stroke population, as others have reported similar trends for various types of patients.28

Compared to CT-based imaging techniques, MR imaging lacks the associated risks of radiation exposure and contrast-associated nephrotoxicity.29 Despite this advantage, we observed only a modest increasing trend in MRI (18%) utilization and a downtrend in MRA (−33%) use over the study period. Our results are consistent with the previously reported growth in MRI utilization for stroke patients from a decade ago11. However, given the recent advances in IVT use in wake-up stroke patients based on MRI-selection criteria,6,30 the utilization of MRI over the next few years is uncertain and should be monitored. Nonetheless, a previously observed >20-fold increase in MRA utilization in a wider group of emergency department patients from 2001 to 201531 may be reversing, at least for Medicare recipients hospitalized with ischemic stroke from 2015-2019 in our study. Downward trends in MRA use in stroke patients with Medicare may represent appropriate avoidance of duplicative studies that provide similar information (i.e., CTA and MRA), as well as the higher sensitivity and specificity of CTA for detection of large artery atherosclerotic and non-atherosclerotic stenosis and occlusion in the cerebrovascular circulation compared to MRA.32,33

Our results suggest that increasing utilization trends for CTA and CTP, which are more widely available in emergency departments, may have a clinical impact. From 2012 to 2019, CTA increased 250% and CTP increased 428%, while EVT increased 688%, indicating that increased EVT use over the study period outpaced the increasing trends in CTA and CTP. Our study design was unable to assess for the appropriate utilization of imaging because the level of detail necessary to perform such an analysis is not available in Medicare claims data. Our study revealed a significant association between CTA and CTP utilization with both IVT and EVT treatment. This finding supports previous observations that CTA and CTP are increasingly performed in ischemic stroke patients who are eligible for treatment9,19,34,35,36 and suggests that CTA and CTP imaging information is utilized for acute stroke treatment decision-making. Prior evidence suggests that utilization of CT imaging leads to faster door-to-needle time,34 with early simultaneous utilization of CTA and CTP being more cost-effective,37 efficient, and associated with better early outcomes38 than sequential imaging. Furthermore, our study found that the odds of 30-day and one-year mortality were lower in those who underwent CTA, MRI, and MRA, but 30-day mortality was higher in those who had CTP, even after adjusting for patient characteristics such as age, stroke severity (ADSS), medical comorbidities (CCI), and acute treatment with IVT and/or EVT. However, other relevant clinical measures known to influence stroke outcomes, such as baseline functional status, stroke severity (National Institutes of Health Stroke Scale), presence of large vessel occlusion, and time from last known well to hospital arrival are not available in the Medicare data and therefore not included in our analyses. Further studies including these factors may refine our results.

Increased CT-based imaging utilization was not uniform across all sociodemographic groups; CTA and CTP utilization were lower in older (≥80 years), female, and Black patients. The most striking disparity we found was that rural patients had significantly lower neuroimaging utilization than their urban counterparts, with the gap increasing over time for CTA and CTP from 2012-2019. The causes of this disparity are uncertain, and while unlikely due to patient preferences or medical necessity, possible explanations include rural hospital closures,39 disparities in time-sensitive access to imaging across facilities, lack of resources and underfunding,40 longer travel time for stroke patients to rural hospitals,41 and other factors that lead patients in rural communities to present outside the recommended treatment time windows.42 The urban/rural disparity illustrated in our study has broader implications since previous studies have shown that low-income and marginalized populations are also more likely to seek care at low-volume hospitals.43,44 Many rural hospitals lack the angiographic equipment and personnel necessary to perform EVT procedures even if advanced CT imaging is performed. Our findings suggest that lower utilization of CT-based neuroimaging may translate to lower utilization of acute stroke treatment. Whether these disparities impact acute stroke treatment utilization and clinical outcomes for disadvantaged groups warrants further study.

We also found a significant disparity in neuroimaging utilization based on census regions with the highest CTA and CTP utilization in the Northeastern and Western regions. However, this finding is likely related to regional variability in practice patterns and resource utilization in stroke care. This is highlighted by the fact that even with high relative CTA imaging, the Northeastern region still had the highest MRA utilization, which may imply duplicative testing. Consequences of this finding in terms of acute treatment utilization and outcomes should be investigated further.

Strengths and limitations

Our study has several strengths and limitations. One strength is the large sample size that is nationally representative of the Medicare population, sufficient to observe macrotrends in neuroimaging utilization, which provided a bird’s-eye-view of a revolutionary era in ischemic stroke care over the past decade. Utilizing Medicare data also provided ample detail to enable a comprehensive analysis of imaging trends in terms of care disparities, impact on treatment utilization, and mortality. Using the Medicare claims data, however, has limitations, including potential misclassification of ischemic stroke patients when using ICD-9 and ICD-10 diagnostic codes and the inability to determine whether neuroimaging occurred before treatment if they were on the same calendar day.45 Furthermore, our study was limited to patients aged 65 years and older and was almost 83% white; therefore, our results may not be applicable to younger and more racially and sociodemographically diverse populations. In addition, the Medicare dataset has limited clinical measures relevant to acute stroke, such as baseline functional status, stroke severity (National Institutes of Health Stroke Scale), presence of large vessel occlusion, and time from last known well to hospital arrival. The exclusion of these factors is a limitation because these variables could influence physician selection of neuroimaging in the acute stroke setting. These factors may also be associated with post-stroke outcomes, so our ability to interpret our findings of association of CTP with increased mortality as well as CTA, MRI, and MRA with decreased mortality is limited. Therefore, further studies that include these pertinent variables will be useful in assessing the impact of increasing CT-based imaging on rates of acute treatment and outcomes.

One path for future research would be a study of the costs, consequences, and effectiveness of our observed trends in neuroimaging and acute treatment, given the projected rise in healthcare costs over the coming decades.46 It is unknown whether the observed higher rates of IVT and EVT associated with upward trends in advanced-CT imaging might yield higher acute care costs that may result in positive outcomes such as longer-term savings for post-stroke care, avoidance of lost income and decreased productivity, and mitigation of the need for long-term care. Previous research has shown growth in the cost of stroke care, with the cost of neuroimaging expanding at a faster rate than other components of stroke care.11 Further analyses quantifying the consequences of increased advanced-CT on downstream care decisions, employing methods such as time-driven activity-based costing47 at the single institutional level, may help explore the effect of our observed neuroimaging and treatment trends on the overall cost of stroke care.

Conclusion:

A significant upward trend in CTA, CTP, and MRI utilization, and a simultaneous downtrend in MRA use was observed for ischemic stroke cases in the Medicare population over the past decade. CTA and CTP imaging were associated with increased EVT and IVT use, with CTA, MRI and MRA being associated with lower post-discharge mortality after adjustment for measured potential confounders. However, disparities existed in neuroimaging utilization across all demographic groups. Given these associations, understanding the root causes of the observed disparities is paramount to achieving equity in stroke care across the population.

Supplementary Material

1

Take Home Points.

  1. From 2012 to 2019, we found significantly increased use of CTA, CTP, and MRI for ischemic stroke patients based on Medicare claims data. Over the same period, MRA use was on the decline.

  2. Overall, neuroimaging was less often performed in rural, older (≥80 years), Black, and female patients.

  3. CTA and CTP uptrends were significantly associated with increased IVT and EVT utilization, which had a positive impact on mortality.

Sources of support:

Funding support received from the Harvey L. Neiman Health Policy Institute through a research partnership.

Sources of Funding:

This work was supported by the National Institute of Neurological Disorders and Stroke (NINDS) of the National Institutes of Health, under award R56NS114275, and a research partnership between Siemens Healthineers and The Feinstein Institutes for Medical Research, Northwell Health.

Abbreviations keys

CTA

computed tomography angiography

CTP

computed tomography perfusion

EVT

endovascular therapy

IVT

intravenous thrombolysis

MRI

magnetic resonance imaging

MRA

magnetic resonance angiography

ADSS

Administrative Data Stroke Scale

CCI

Charlson Comorbidity Index

RIF

research identifiable files

NPI

national provider identifiers

CMS

Centers for Medicare & Medicaid Services

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.

Conflict of interest: The authors declare no conflict of interest.

Disclosures: Dr. Wang and Dr. Katz have received grants from Siemens Healthineers and NINDS during the conduct of the study. Dr Boltyenkov has received grants from Siemens Healthineers and NINDS during the conduct of the study and grants, personal fees, and nonfinancial support from Siemens Healthineers outside the submitted work. Dr Sanelli has received research grants from Siemens Healthineers and NINDS during the conduct of the study and research grants from the Harvey L. Neiman Health Policy Institute outside the submitted work.

Statement of Data Access and Integrity: The authors declare that they had full access to all of the data in this study and the authors take complete responsibility for the integrity of the data and the accuracy of the data analysis.

References

  • 1.Virani SS, Alonso A, Aparicio HJ, et al. Heart Disease and Stroke Statistics-2021 Update: A Report From the American Heart Association. Circulation. 2021;143(8):e254–e743. doi: 10.1161/CIR.0000000000000950 [DOI] [PubMed] [Google Scholar]
  • 2.Yousufuddin M, Young N. Aging and ischemic stroke. Aging (Albany NY). 2019;11(9):2542–2544. doi: 10.18632/aging.101931. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Powers WJ, Derdeyn CP, Biller J, et al. 2015 American Heart Association/American Stroke Association Focused Update of the 2013 Guidelines for the Early Management of Patients With Acute Ischemic Stroke Regarding Endovascular Treatment: A Guideline for Healthcare Professionals From the American Heart Association/American Stroke Association. Stroke. 2015;46(10):3020–3035. doi: 10.1161/STR.0000000000000074 [DOI] [PubMed] [Google Scholar]
  • 4.Psychogios K, Safouris A, Kargiotis O, et al. Advanced Neuroimaging Preceding Intravenous Thrombolysis in Acute Ischemic Stroke Patients Is Safe and Effective. J Clin Med. 2021;10(13):2819. Published 2021 Jun 26. doi: 10.3390/jcm10132819 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.McDonald JS, Fan J, Kallmes DF, Cloft HJ. Pretreatment advanced imaging in patients with stroke treated with IV thrombolysis: evaluation of a multihospital data base. AJNR Am J Neuroradiol. 2014;35(3):478–481. doi: 10.3174/ajnr.A3797 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Thomalla G, Simonsen CZ, Boutitie F, et al. MRI-Guided Thrombolysis for Stroke with Unknown Time of Onset. N Engl J Med. 2018;379(7):611–622. doi: 10.1056/NEJMoa1804355 [DOI] [PubMed] [Google Scholar]
  • 7.Nogueira RG, Jadhav AP, Haussen DC, et al. Thrombectomy 6 to 24 hours after stroke with a mismatch between deficit and infarct. N Engl J Med. 2018;378:11–21. [DOI] [PubMed] [Google Scholar]
  • 8.Albers GW, Marks MP, Kemp S, et al. Thrombectomy for stroke at 6 to 16 hours with selection by perfusion imaging. N Engl J Med. 2018. Feb 22;378:708–718. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Asaithambi G, Tong X, Lakshminarayan K, Coleman King SM, George MG. Current trends in the acute treatment of ischemic stroke: analysis from the Paul Coverdell National Acute Stroke Program. J Neurointerv Surg. 2020;12(6):574–578. doi: 10.1136/neurintsurg-2019-015133 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Vagal A, Meganathan K, Kleindorfer DO, Adeoye O, Hornung R, Khatri P. Increasing use of computed tomographic perfusion and computed tomographic angiograms in acute ischemic stroke from 2006 to 2010. Stroke. 2014;45:1029–1034. [DOI] [PubMed] [Google Scholar]
  • 11.Burke JF, Kerber KA, Iwashyna TJ, Morgenstern LB. Wide variation and rising utilization of stroke magnetic resonance imaging: data from 11 states. Ann Neurol. 2012;71:179–185. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Ng VT, Bayoumi AM, Fang J, et al. Temporal trends in the use of investigations after stroke or transient ischemic attack. Med Care. 2016;54:430–434. [DOI] [PubMed] [Google Scholar]
  • 13.Madsen TE, Khoury JC, Leppert M, et al. Temporal Trends in Stroke Incidence Over Time by Sex and Age in the GCNKSS [published correction appears in Stroke. 2020 Jul;51(7):e141]. Stroke. 2020;51(4):1070–1076. doi: 10.1161/STROKEAHA.120.028910 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Andrade SE, Harrold LR, Tjia J, et al. A systematic review of validated methods for identifying cerebrovascular accident or transient ischemic attack using administrative data. Pharmacoepidemiol Drug Saf. 2012;21 Suppl 1(Suppl 1):100–128. doi: 10.1002/pds.2312 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.United States Census Bureau. B19013 median household income in the past 12 months (in 2018 inflation-adjusted dollars). 2018 American Community Survey 5-year estimates. Available at: https://data.census.gov/cedsci/all?q=B19013%20median%20household%20income%20in%20the%20past%2012%20months%20%28in%202018%20inflation-adjusted%20dollars%29%205%20year%20estimates. Accessed April 22, 2021 [Google Scholar]
  • 16.Adeoye O, Albright KC, Carr BG, et al. Geographic access to acute stroke care in the United States. Stroke. 2014;45(10):3019–3024. doi: 10.1161/STROKEAHA.114.006293 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Patel PD, Salwi S, Liles C, et al. Creation and Validation of a Stroke Scale to Increase Utility of National Inpatient Sample Administrative Data for Clinical Stroke Research. J Stroke Cerebrovasc Dis. 2021. Apr;30(4):105658. doi: 10.1016/j.jstrokecerebrovasdis.2021.105658. Epub 2021 Feb 12. PMID: 33588186. [DOI] [PubMed] [Google Scholar]
  • 18.Austin SR, Wong YN, Uzzo RG, Beck JR, Egleston BL. Why Summary Comorbidity Measures Such As the Charlson Comorbidity Index and Elixhauser Score Work. Med Care. 2015;53(9):e65–e72. doi: 10.1097/MLR.0b013e318297429c [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Berkhemer OA, Fransen PS, Beumer D, et al. A randomized trial of intraarterial treatment for acute ischemic stroke [published correction appears in N Engl J Med. 2015 Jan 22;372(4):394]. N Engl J Med. 2015;372(1):11–20. doi: 10.1056/NEJMoa1411587 [DOI] [PubMed] [Google Scholar]
  • 20.Campbell BC, Mitchell PJ, Kleinig TJ, et al. Endovascular therapy for ischemic stroke with perfusion-imaging selection. N Engl J Med. 2015;372(11):1009–1018. doi: 10.1056/NEJMoa1414792 [DOI] [PubMed] [Google Scholar]
  • 21.Goyal M, Demchuk AM, Menon BK, et al. Randomized assessment of rapid endovascular treatment of ischemic stroke. N Engl J Med. 2015;372(11):1019–1030. doi: 10.1056/NEJMoa1414905 [DOI] [PubMed] [Google Scholar]
  • 22.Saver JL, Goyal M, Bonafe A, et al. Stent-retriever thrombectomy after intravenous t-PA vs. t-PA alone in stroke. N Engl J Med. 2015;372(24):2285–2295. doi: 10.1056/NEJMoa1415061 [DOI] [PubMed] [Google Scholar]
  • 23.Jovin TG, Chamorro A, Cobo E, et al. Thrombectomy within 8 hours after symptom onset in ischemic stroke. N Engl J Med. 2015;372(24):2296–2306. doi: 10.1056/NEJMoa1503780 [DOI] [PubMed] [Google Scholar]
  • 24.Bracard S, Ducrocq X, Mas JL, et al. Mechanical thrombectomy after intravenous alteplase versus alteplase alone after stroke (THRACE): a randomised controlled trial [published correction appears in Lancet Neurol. 2016 Nov;15(12):1203]. Lancet Neurol. 2016;15(11):1138–1147. doi: 10.1016/S1474-4422(16)30177-6 [DOI] [PubMed] [Google Scholar]
  • 25.Wahlgren N, Moreira T, Michel P, et al. Mechanical thrombectomy in acute ischemic stroke: Consensus statement by ESO-Karolinska Stroke Update 2014/2015, supported by ESO, ESMINT, ESNR and EAN. Int J Stroke. 2016;11(1):134–147. doi: 10.1177/1747493015609778 [DOI] [PubMed] [Google Scholar]
  • 26.Pierot L, Jayaraman MV, Szikora I, et al. Standards of practice in acute ischemic stroke intervention: international recommendations. J Neurointerv Surg. 2018;10(11):1121–1126. doi: 10.1136/neurintsurg-2018-014287 [DOI] [PubMed] [Google Scholar]
  • 27.Rao VM, Levin DC, Parker L, Frangos AJ, Sunshine JH. Trends in utilization rates of the various imaging modalities in emergency departments: nationwide Medicare data from 2000 to 2008. J Am Coll Radiol. 2011;8(10):706–709. doi: 10.1016/j.jacr.2011.04.004 [DOI] [PubMed] [Google Scholar]
  • 28.Hong AS, Levin D, Parker L, Rao VM, Ross-Degnan D, Wharam JF. Trends in Diagnostic Imaging Utilization among Medicare and Commercially Insured Adults from 2003 through 2016. Radiology. 2020;294(2):342–350. doi: 10.1148/radiol.2019191116 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Wintermark M, Meuli R, Browaeys P, et al. Comparison of CT perfusion and angiography and MRI in selecting stroke patients for acute treatment. Neurology. 2007;68(9):694–697. doi: 10.1212/01.wnl.0000255959.30107.08 [DOI] [PubMed] [Google Scholar]
  • 30.Powers WJ, Rabinstein AA, Ackerson T, et al. Guidelines for the Early Management of Patients With Acute Ischemic Stroke: 2019 Update to the 2018 Guidelines for the Early Management of Acute Ischemic Stroke: A Guideline for Healthcare Professionals From the American Heart Association/American Stroke Association [published correction appears in Stroke. 2019 Dec;50(12):e440-e441]. Stroke. 2019;50(12):e344–e418. doi: 10.1161/STR.0000000000000211 [DOI] [PubMed] [Google Scholar]
  • 31.Prabhakar AM, Gottumukkala RV, Hemingway J, Hughes DR, Patel SS, Duszak R Jr. Increasing utilization of emergency department neuroimaging in Medicare beneficiaries from 1994 to 2015. Am J Emerg Med. 2018;36(4):680–683. doi: 10.1016/j.ajem.2017.12.057 [DOI] [PubMed] [Google Scholar]
  • 32.Nguyen-Huynh MN, Wintermark M, English J, et al. How accurate is CT angiography in evaluating intracranial atherosclerotic disease?. Stroke. 2008;39(4):1184–1188. doi: 10.1161/STROKEAHA.107.502906 [DOI] [PubMed] [Google Scholar]
  • 33.Hirai T, Korogi Y, Ono K, et al. Prospective evaluation of suspected stenoocclusive disease of the intracranial artery: combined MR angiography and CT angiography compared with digital subtraction angiography. AJNR Am J Neuroradiol. 2002;23(1):93–101. [PMC free article] [PubMed] [Google Scholar]
  • 34.Macha K, Hoelter P, Siedler G, et al. Multimodal CT or MRI for IV thrombolysis in ischemic stroke with unknown time of onset. Neurology. 2020;95(22):e2954–e2964. doi: 10.1212/WNL.0000000000011059 [DOI] [PubMed] [Google Scholar]
  • 35.Hedlund F, Leighs A, Barber PA, Lundström E, Wu TY, Ranta A. Trends in stroke reperfusion treatment and outcomes in New Zealand. Intern Med J. 2020;50:1367–1372. [DOI] [PubMed] [Google Scholar]
  • 36.Khatri P, Kleindorfer DO, Devlin T, et al. Effect of Alteplase vs Aspirin on Functional Outcome for Patients With Acute Ischemic Stroke and Minor Nondisabling Neurologic Deficits: The PRISMS Randomized Clinical Trial. JAMA. 2018;320(2):156–166. doi: 10.1001/jama.2018.8496. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Martinez G, Katz JM, Pandya A, et al. Cost-Effectiveness Study of Initial Imaging Selection in Acute Ischemic Stroke Care. J Am Coll Radiol. 2021;18(6):820–833. doi: 10.1016/j.jacr.2020.12.013 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Katz JM, Wang JJ, Boltyenkov AT, et al. Rescan Time Delays in Ischemic Stroke Imaging: A Retrospective Observation and Analysis of Causes and Clinical Impact [published online ahead of print, 2021 Aug 12]. AJNR Am J Neuroradiol. 2021; 10.3174/ajnr.A7227. doi: 10.3174/ajnr.A7227 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.McCarthy S, Moore D, Smedley WA, et al. Impact of Rural Hospital Closures on Health-Care Access. J Surg Res. 2021;258:170–178. doi: 10.1016/j.jss.2020.08.055 [DOI] [PubMed] [Google Scholar]
  • 40.Slonim AD, See H, Slonim S. Challenges confronting rural hospitals accentuated during COVID-19. J Biomed Res. 2020;34(6):397–409. doi: 10.7555/JBR.34.20200112 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Yu CY, Blaine T, Panagos PD, Kansagra AP. Demographic Disparities in Proximity to Certified Stroke Care in the United States. Stroke. 2021;52(8):2571–2579. doi: 10.1161/STROKEAHA.121.034493 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Jauch EC, Schwamm LH, Panagos PD, et al. Recommendations for Regional Stroke Destination Plans in Rural, Suburban, and Urban Communities From the Prehospital Stroke System of Care Consensus Conference: A Consensus Statement From the American Academy of Neurology, American Heart Association/American Stroke Association, American Society of Neuroradiology, National Association of EMS Physicians, National Association of State EMS Officials, Society of NeuroInterventional Surgery, and Society of Vascular and Interventional Neurology: Endorsed by the Neurocritical Care Society [published correction appears in Stroke. 2021 Jul;52(7):e481]. Stroke. 2021;52(5):e133–e152. doi: 10.1161/STROKEAHA.120.033228 [DOI] [PubMed] [Google Scholar]
  • 43.Konety SH, Vaughan Sarrazin MS, Rosenthal GE. Patient and hospital differences underlying racial variation in outcomes after coronary artery bypass graft surgery. Circulation 2005; 111:1210–1216 [DOI] [PubMed] [Google Scholar]
  • 44.Lee SL, Yaghoubian A, Stark R, Sydorak RM, Kaji A. Are there differences in access to care, treatment, and outcomes for children with appendicitis treated at county versus private hospitals? Perm J. 2012;16(1):4–6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Patel PD, Salwi S, Liles C, et al. Creation and Validation of a Stroke Scale to Increase Utility of National Inpatient Sample Administrative Data for Clinical Stroke Research. J Stroke Cerebrovasc Dis. 2021;30(4):105658. doi: 10.1016/j.jstrokecerebrovasdis.2021.105658 [DOI] [PubMed] [Google Scholar]
  • 46.Ovbiagele B, Goldstein LB, Higashida RT, et al. Forecasting the future of stroke in the United States: a policy statement from the American Heart Association and American Stroke Association [published correction appears in Stroke. 2015 Jul;46(7):e179]. Stroke. 2013;44(8):2361–2375. doi: 10.1161/STR.0b013e31829734f2 [DOI] [PubMed] [Google Scholar]
  • 47.Yun BJ, Prabhakar AM, Warsh J, et al. Time-Driven Activity-Based Costing in Emergency Medicine. Ann Emerg Med. 2016;67(6):765–772. doi: 10.1016/j.annemergmed.2015.08.004. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1796889-supplement-1.docx (119.6KB, docx)

RESOURCES