Relative to nonminority patients, racial and ethnic minorities experience higher mortality from coronary heart disease.1,2 The underpinnings of this may partially relate to disparities in care delivery and cardiac testing. Patients from racial and ethnic minority groups presenting to the Emergency Department (ED) with chest pain experience a 32% lower odds of receiving stress testing compared with nonminorities.3 Accordingly, interventions to reduce disparities in cardiovascular care were identified by the Institute of Medicine as a high priority for comparative effectiveness research.4
We have recently tested a diagnostic pathway in 3 randomized clinical trials to assess the comparative effectiveness of observation unit strategies relative to usual care.5–7 We have named this protocol-driven chest pain care pathway ACES (Accelerated Chest pain Evaluation with Stress imaging); this protocol uses serial troponins with an expectation for stress testing to identify patients with intermediate to high risk findings for discharge or invasive management. The ACES care pathway expectation for stress testing involves less clinician judgment about whether stress testing should be performed. We hypothesized this care pathway would reduce racial disparities in objective cardiac testing. Accordingly, we sought to determine whether an ACES approach, relative to an existing customary care approach, would reduce racial disparities in objective cardiac testing among patients with acute chest pain.
MATERIALS AND METHODS
From 2008 through 2011, 3 randomized trials were performed involving patients with chest pain recruited from the ED of (name withheld for peer review; Registered with Clinicaltrials.gov as NCT00678639, NCT00869245, and NCT01035047). The medical center is a large tertiary care facility serving an urban, suburban, and rural population with an annual ED volume of 90,000–100,000 visits during the study period. Two of these trials randomized participants to ACES versus inpatient care and the third compared 2 ACES protocols that incorporated different imaging strategies. Inpatient care was provided by the Cardiology, Family Medicine, Internal Medicine, or Hospitalist services. ACES care was provided in the observation unit staffed by Emergency Physicians, Physician Assistants, and Nurse Practitioners. This analysis combined the data from these 3 trials into 1 dataset with participants categorized as receiving ACES or usual care.
Subject Selection Criteria
To be enrolled in the randomized trials, participants had symptoms prompting an evaluation for possible acute coronary syndrome and no contraindications for cardiac magnetic resonance testing. Exclusion criteria included new ST-segment elevation or depression, hypotension (systolic <90 mm Hg), and an initial troponin value above the institutional normal value. Low-risk patients were excluded, thus inclusion required a care provider assessment of intermediate to high risk chest pain using the American College of Cardiology/American Heart Association risk stratification framework, and/or a Thrombolysis in Myocardial Infarction (TIMI) risk score of ≥1.5–8 Race and ethnicity were self-reported to the study coordinator at the time of enrollment or obtained from the electronic medical record. Participants reporting race other than AA or White were excluded from this analysis (n = 5).
Outcome Measures and Covariates
The primary outcome for this analysis included the rate of objective cardiac testing during the index visit. Objective cardiac testing was defined as stress (exercise or dobutamine) echocardiography, cardiac nuclear stress imaging, coronary computed tomographic angiography, invasive catheter coronary angiography, stress cardiac magnetic resonance imaging, or stress treadmill testing.
For outcomes after the index visit, participants were followed over the course of a year from randomization using both record review and telephone contact. Secondary outcome measures included major adverse cardiac events at 90 days and cardiac-related hospital readmissions measured at 1 year. Major adverse cardiac events were the composite of death (all cause), myocardial infarction, and coronary revascularization. Myocardial infarction was defined and adjudicated by 2 investigators in accordance with the universal definition at the time the trials enrolled,9 requiring a gradual rise and fall of troponin, with appropriate clinical symptoms. In the case of any disagreement, the first 2 adjudicators discussed the discordant case with an attempt to reach agreement. If disagreement persisted, a third adjudicator was utilized. For hospital readmissions, cardiac relatedness was assessed using a definition previously described,5 requiring one of the following criteria: (1) cardiac procedures performed [cardiac imaging or stress testing (excluding resting echo), coronary revascularization, pacemaker/defibrillator placement], (2) primary reason for admission is concern for cute coronary syndrome (ACS), or (3) discharge diagnosis related to chest pain, MI, ACS, heart failure, or other cardiac disease.
Sample size was determined based on the available number of participants (n = 334). The 3 separate databases were combined into a single dataset. Our first objective was to examine whether testing disparities were observed between racial groups. χ2 and Fisher exact tests were used to assess differences between AA and White participants in demographics, test utilization, and outcomes, separately by care pathway. Next, separate logistic regression models for usual care and ACES groups were used to assess the association of race after adjusting for possible confounders. Covariates considered included age at enrollment, sex, insurance status, smoking status, hypertension, diabetes, hyperlipidemia, prior MI, prior coronary revascularization, and TIMI score. A backwards stepping algorithm was used to remove nonsignificant covariates from the model; race was forced to remain in all models.
From 2008 to 2011, 334 participants were enrolled in the 3 clinical trials, and 5 were excluded from this analysis based on reported race other than AA or White. The median age of the participants was 55 years (Q1 = 47 years, Q3 = 63 years; (Table 1), with 61/329 (19%) of participants being at least 65 years old. AA participants represented 111/329 (34%) of the total study population, 80/220 (36%) of the ACES group and 31/109 (28%) of the usual care group. Insurance was private (nongovernmental) in 42/111 (38%) of AA and in 110/218 (50%) of White participants. AA patients had a greater prevalence of hypertension (P < 0.001), diabetes (P = 0.020), and prior heart failure (P = 0.025), and they tended to have higher TIMI scores10 (P = 0.003).
Objective testing during the index visit is summarized in Table 2. In the usual care group, objective testing by at least one modality was performed in 22/31 (71%) AA patients and 69/78 (88%) White patients (P = 0.027). There was less use of cardiac catheterization among AA versus White patients (3% vs. 24%; P = 0.012). Nineteen patients had cardiac catheterization without noninvasive testing at any time during the index visit (18 White patients and 1 AA patient). There was little difference in the use of stress cardiac magnetic resonance, stress echo, or coronary computed tomographic angiography between AA and White patients [13% vs. 10% (P = 0.738), 55% vs. 60% (P = 0.604), and 3% vs. 1% (P = 0.490), respectively]. Unadjusted, the odds of having an objective test done were 3.14 times higher for Whites compared with AAs [95% confidence interval (CI): 1.11–8.88). After adjustment for age, sex, insurance status, smoking status, hypertension, diabetes, hyperlipidemia, prior MI, prior coronary revascularization, and TIMI score, the odds of having an objective test done were 3.03 times higher for Whites compared with AAs (95% CI: 0.92–10.0). However, none of these covariates were significantly associated with objective testing, and all fell out of the model using a backward stepping algorithm. Unadjusted, the odds of having a cardiac catheterization done were 9.66 times higher for Whites compared with AAs (95% CI: 1.23–75.6). After adjustment for age, sex, insurance status, smoking status, hypertension, diabetes, hyperlipidemia, prior MI, prior coronary revascularization, and TIMI score, the odds of having a cardiac catheterization done were 12.5 times higher for Whites compared with AAs (95% CI: 1.36–116). Only hypertension and prior coronary revascularization were significantly associated with objective testing, and the odds of having a cardiac catheterization done were 11.8 times higher for Whites compared with AA after adjusting for those 2 covariates (95% CI: 1.43–97.6).
In the ACES group, 78/80 (98%) AA participants and 138/140 (99%) White participants had objective testing (P = 0.623; odds ratio = 1.77; 95% CI: 0.24–12.8). No adjusted models were run due to the small number of patients without testing. Invasive angiography was performed in 4/80 (5%) AA ACES and 13/140 (9%) White ACES participants (P = 0.303). Unadjusted, the odds of having a cardiac catheterization done were 1.95 times higher for Whites compared with AA (95% CI: 0.61–6.18). After adjustment for age, sex, insurance status, smoking status, hypertension, diabetes, hyperlipidemia, prior MI, prior coronary revascularization, and TIMI score, the odds of having a cardiac catheterization done were 1.93 times higher for Whites compared with AAs (95% CI: 0.54–6.88). Only sex was significantly associated with objective testing, and the odds of having a cardiac catheterization done were 1.88 times higher for Whites compared with AAs after adjusting for sex (95% CI: 0.59–6.04).
No participants died during the 90-day clinical outcome assessment period (Table 3). Myocardial infarction occurred in 1/31 (3%) AA versus 8/78 (10%) White usual care participants (P = 0.441), and in 2/80 (3%) AA ACES and 4/140 (3%) White ACES participants (P = 1.0). In the White usual care group, there were 3 periprocedural myocardial infarction events (determined by adjudicators) and one other event that occurred after invasive angiography in a patient who went on to bypass surgery.
Revascularization across both groups was more common in Whites [1/111 (1%) vs. 18/218 (8%); P = 0.009). In the usual care group, revascularization was performed in 1/31 (3%) AA and 12/78 (15%) White participants (P = 0.104); no AA ACES (0/80) and 6/140 (4%) White ACES participants received revascularization (P = 0.089).
At 1 year, cardiac hospital readmissions occurred in 58/329 (18%) of the entire study group. By group and race, cardiac related hospital readmissions occurred in 12/31 (39%) AA usual care versus 19/78 (24%) White usual care participants (P = 0.134) and in 13/80 (16%) AA ACES and 14/140 (10%) White ACES participants (P = 0.169). Cardiac related hospital readmissions were more frequent among patients with ACS at 90 days (11/25, 44%) compared with those without ACS at 90 days (47/302, 16%), P = 0.001.
The results of our analysis suggest a protocol-driven care pathway with an expectation for stress imaging can eliminate the racial disparity in index visit cardiac testing observed in customary care practices. In our control group, White participants more frequently received cardiac testing, and more commonly underwent invasive angiography versus AA participants. These variations in care are consistent with the findings of others3,11 and were not observed in the ACES group. Further, White participants more commonly received revascularization procedures, also consistent with findings of others.12 Paradoxically, we found those procedures to be associated with an increase in postprocedural troponin elevations (type IV and V myocardial infarction). What is important here is that the ACES improved safety by reducing the number of catheterizations overall and thereby reducing the number of myocardial infarctions from a procedure that did not contribute to a reduction in deaths. One can work out the numbers, but it is possible that the lesser exposure to further testing in the AA group was actually a benefit—reduction is exposure to demonstrable (real) hazard.
In our prior work, we demonstrated that the protocol-driven ACES pathway reduced index visit cost, cost at 1 year, and invasive procedures when considering all races together.5,7,13 These earlier findings, combined with our new analysis, illustrate that eliminating the racial disparity in cardiovascular testing does not necessarily mean accumulating incremental healthcare expenditures. Rather, the ACES approach appears to achieve high rates of index visit noninvasive cardiac testing, which in turn reprioritizes the selection of patients for invasive angiography. The absence of ACS events between index visit discharge and 90 days suggest this reallocation of testing resources did not harm patient outcomes.
Recent investigations have demonstrated that many low-risk patients do not require stress imaging and can be effectively triaged using clinical decision aids and serum troponin measurements.14,15 However, no such decision aids exist to identify intermediate to high risk patients unlikely to benefit from objective cardiac testing, and guidelines recommend testing in all of these patients. Therefore, the ACES care pathway involves the ordering of stress imaging. Mechanistically, this effectively eliminates 2 clinical decisions: (1) whether testing should be obtained and (2) if testing should be obtained, should it be invasive or noninvasive testing? These decisions remain unresolved in usual hospital care until the receiving team conducts the initial evaluation. Our analysis suggests these decision nodes are in the causal pathway for index visit testing disparities.
Our findings suggesting an ACES pathway as a solution to index visit racial disparities in cardiovascular testing are tempered with limitations. Our analysis was conducted in a single institution and involved the collation of data from prior randomized clinical trials. The consistency of our findings with those of others suggest our baseline disparities observed in the usual care group are representative, and our intervention was randomized without respect to race, which should reduce the effect of these limitations. Further, our racial groups were limited to AA and White participants, and we cannot extrapolate to other racial or ethnic groups. Finally, we observed White participants to experience serum troponin elevations more commonly, particularly after revascularization, and these events were adjudicated as myocardial infarctions by a consensus of investigators. The impact of these events on clinical outcomes is not certain.
In conclusion, this analysis demonstrates the protocol-driven chest pain care pathway ACES decreases overall number of cardiac catheterizations and eliminates the racial disparity observed in the acquisition of index-visit cardiovascular testing observed among AA and White patients that present to emergency departments with chest pain at intermediate risk for ACS. Further research should be designed to prospectively test the effect of ACES on racial disparities.
2. Joynt KE, Orav EJ, Jha AKThirty-day readmission rates for Medicare beneficiaries by race and site of care. JAMA. 2011;305:675–681.
3. Napoli AM, Choo EK, Dai J, Desroches BRacial disparities in stress test utilization in an emergency department chest pain unit. Crit Pathw Cardiol. 2013;12:9–13.
4. Institute of Medicine (U.S.). Committee on Comparative Effectiveness Research Prioritization. Initial National Priorities for Comparative Effectiveness Research. 2009.Washington, DC: National Academies Press.
5. Miller CD, Hwang W, Case D, et alStress CMR imaging observation unit in the emergency department reduces 1-year medical care costs in patients with acute chest pain: a randomized study for comparison with inpatient care. JACC Cardiovasc Imaging. 2011;4:862–870.
6. Miller CD, Hoekstra JW, Lefebvre C, et alProvider-directed imaging stress testing
reduces health care expenditures in lower-risk chest pain patients presenting to the emergency department. Circ Cardiovasc Imaging. 2012;5:111–118.
7. Miller CD, Hwang W, Hoekstra JW, et alStress cardiac magnetic resonance imaging with observation unit care reduces cost for patients with emergent chest pain: a randomized trial. Ann Emerg Med. 2010;56:209–219.e2.
8. Anderson JL, Adams CD, Antman EM, et alAmerican College of Cardiology; American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 2002 Guidelines for the Management of Patients With Unstable Angina/Non ST-Elevation Myocardial Infarction
); American College of Emergency Physicians; Society for Cardiovascular Angiography and Interventions; Society of Thoracic Surgeons; American Association of Cardiovascular and Pulmonary Rehabilitation; Society for Academic Emergency Medicine. ACC/AHA 2007 guidelines for the management of patients with unstable angina/non ST-elevation myocardial infarction
: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 2002 Guidelines for the Management of Patients With Unstable Angina/Non ST-Elevation Myocardial Infarction
): developed in collaboration with the American College of Emergency Physicians, the Society for Cardiovascular Angiography and Interventions, and the Society of Thoracic Surgeons: endorsed by the American Association of Cardiovascular and Pulmonary Rehabilitation and the Society for Academic Emergency Medicine. Circulation. 2007;116:e148–e304.
9. Thygesen K, Alpert JS, White HD, et alJoint ESC/ACCF/AHA/WHF Task Force for the Redefinition of Myocardial Infarction
. Universal definition of myocardial infarction
. Circulation. 2007;116:2634–2653.
10. Antman EM, Cohen M, Bernink PJ, et alThe TIMI risk score for unstable angina/non-ST elevation MI: a method for prognostication and therapeutic decision making. JAMA. 2000;284:835–842.
11. Venkat A, Hoekstra J, Lindsell C, et alThe impact of race on the acute management of chest pain. Acad Emerg Med. 2003;10:1199–1208.
12. Freund KM, Jacobs AK, Pechacek JA, White HF, Ash ASDisparities by race, ethnicity, and sex in treating acute coronary syndromes. J Womens Health (Larchmt). 2012;21:126–132.
13. Miller CD, Case LD, Little WC, et alStress CMR reduces revascularization, hospital readmission, and recurrent cardiac testing in intermediate-risk patients with acute chest pain. JACC Cardiovasc Imaging. 2013;6:785–794.
14. Mahler SA, Riley RF, Hiestand BC, et alThe HEART Pathway randomized trial: identifying emergency department patients with acute chest pain for early discharge. Circ Cardiovasc Qual Outcomes. 2015;8:195–203.
15. Than M, Aldous S, Lord SJ, et alA 2-hour diagnostic protocol for possible cardiac chest pain in the emergency department: a randomized clinical trial. JAMA Intern Med. 2014;174:51–58.