Despair Over Disparities

By Walter Alexander

In recent years, there has been some despair over disparities in care. Early concern focused on uneven distribution of care based on gender or ethnicity. Other evidence emphasized inherent physiological differences or inadequate understanding of racial variations in known CV risk factors.

For example, although the prevalence of diabetes has increased globally, there are disproportionally higher rates of new-onset diabetes in racial and ethnic minority populations. In a 2013 article in JACC, Prakash Deedwania, MD, noted that people of South-Asian origin (both in their native countries as well as migrant populations) now have one of the highest rates of developing diabetes mellitus in the world.1 At the recent Cardiovascular Research Technologies (CRT) meeting, the Association of Black Cardiologists (ABC) had a featured session with presentations noting that African Americans have longer time to presentation after MI, higher rates of bleeding post-MI, higher risk-adjusted mortality after ACS, high rates of stent thrombosis, and more. From ACC.13, other papers continued to clarify the issue based on data from other ethnic populations.

One more major factor influencing care: insurance. Kapoor et al. analyzed data from almost 100,000 HF hospitalizations and found significant differences across four payer categories: private, Medicare, Medicaid, and no insurance.2 The Medicaid and no-insurance groups were less likely to receive a number of proven treatments, notably evidence-based beta-blockers for patients with reduced LVEF. Compared with private insurance patients, Medicaid patients had a higher adjusted in-hospital mortality rate, as did the no-insurance subgroup with reduced LVEFs.

Given all the contributing factors, what can we learn from recent studies? And, while disparities remain and knowledge gaps persist, can model remedial strategies (like one successfully deployed in North Carolina) point the way toward establishing quality care for all?

ABC Insights

In an analysis of STEMI patients treated with fibrinolysis presented at CRT 2013, Sara D. Collins, MD, from Washington Hospital Center, reported that moderate-to-severe bleeding after thrombolysis was more frequent in African Americans than in Caucasians (16.3% vs. 14.1%) and mortality was higher in those with bleeding (HR = 2.83; 95% CI 2.08–3.86).3 In a second study, this time using data from the National Registry of Myocardial Infarction (NRMI), bleeding was higher in African Americans following either thrombolytic therapy (OR = 1.21) or PCI (OR = 1.33).4 Baseline differences among the African-American population included younger age, higher preponderance of female gender, higher incidence of diabetes, chronic renal insufficiency, hypertension, HF, and more—all of which contributed to more compromised health at presentation compared to Caucasians.

Beyond comorbidities and demographics, African Americans present later after MI symptom onset than Caucasians and face significantly longer treatment delays once in-hospital, both in time-to-fibrinolysis administration and time-to-primary PCI.4 Medicare records from 1.2 million beneficiaries showed that—after adjustment for demographics, comorbidities, and illness severity—African Americans were transferred after MI less often to facilities offering coronary revascularization (25.2% vs. 31.0%), and they underwent fewer revascularizations (34.3% vs. 50.2%).5

Signs of Progress

A comparison of 10-year CV risk contrasting NHANES data from 1988–1994 against 2005–2008 showed mean MI risk dropping by 30.4% (5.4% to 4.1%),6 and the ARIC study (1987–2008) including about 31,000 acute MI patients revealed decreases in annual incident MI rates by 4.3% for Caucasian men, 3.8% for Caucasian women, and 3.4% for African-American men.7

"Unfortunately, the rate for African-American women," Dr. Collins said, "was way down at the bottom, falling only by 1.5% over 20 years. That's quite concerning." While disparities in outcomes after MI continue to exist along ethnic lines, overall incidence of MI and mortality after MI has decreased. Dr. Collins concluded, "Overall, we've made a lot of headway in regards to mortality and MI events—the concern is that the disparity still exists."

"Bad Actor"

Hypertension is well-established as an indicator of increased CV risk. According to Vasilios Papademetriou, MD, professor of medicine, Georgetown University, high BP is more prevalent in African Americans and confers higher stroke risk. Among 27,748 individuals in the REGARDS study, the black-to-white HR for stroke in patients with stage 1 hypertension was 2.38 in patients aged 45–64 years. Among normotensives it was 0.87.

Dr. Papademetriou's 10-year multicenter study of 582,881 veterans revealed improvement in BP control rates between 2000 and 2010, but the control rate among African Americans (25.1%) was half that of Caucasians (52.3%), with Hispanics faring even worse (21.1%).8

Dr. Papademetriou's list of additional "bad actor" traits associated with hypertension in African Americans is long and grim. In them, hypertension is:

  • more severe
  • more difficult to control
  • less responsive to medications
  • more likely to be resistant
  • accompanied by more complications and by activation of the sympathetic nervous system

While we lack data on renal denervation in African Americans, Dr. Papademetriou said, it would seem that the African-American population is "optimally suited for interventional techniques," such as renal denervation, to better control hypertension.

Asian Americans and Hypertension

When it comes to CHD, Asian Americans have higher rates and hypertension may be the culprit, per an analysis of Pan Asian Cohort Study (PACS) data. Powell Jose, MD, research physician at the Palo Alto Medical Foundation Research Institute, reported at ACC.13 that the Asian-American population is expected to swell to 34 million by 2050. While higher CHD rates have been identified in Asian Indians and Asian Filipinos, major CV risk factors, including hypertension, have not been studied adequately among Asian Americans.

Using electronic health records from a major San Francisco Bay Area outpatient HMO, Dr. Jose and colleagues compared hypertension rates among several Asian-American subgroups (Indian, Chinese, Filipino, Japanese, Korean, or Vietnamese), with non-Hispanic white rates. Overall, the adjusted hypertension rate for the aggregated Asian-American group was lower than that of non-Hispanic whites (34.9% vs. 38.9%). Rates varied, however, among individual subgroups: the hypertension rate was lower among Chinese (29.8%), Koreans (30.7%), Vietnamese (30.8%), and Asian Indians (36.9%); for those of Japanese descent, their rate was comparable to non-Hispanic whites (38.2%), while Filipinos had a markedly higher rate of hypertension (51.9%).

Next Stop: South Asian and African-Caribbean Populations

In a recent issue of JACC, Tillin et al. examined the influence of diabetes mellitus and other metabolic risk factors on CVD in the SABRE (Southall and Brent Revisited) study.9 About half of the patients were of European descent, with the rest split between origins in South Asia and the African-Caribbean region. Over 20 years, there was a nearly three-fold higher prevalence of diabetes in South Asians and African Caribbeans, along with more central obesity and atherogenic dyslipidemia in South Asians than the African-Caribbean population. Both had higher stroke rates than those of European descent.

In the accompanying editorial commentary, Dr. Deedwania noted that one of the study's most important findings was that both South Asians and African Caribbeans, despite lower body mass index, had more diabetes and hypertension than Europeans. More prevalent central obesity, as a "harbinger of insulin resistance," might account for the difference. However, the wide range of uncontrolled variables and long duration of the study precluded reliable conclusions.

Dr. Deedwania underscored the urgent need for further study of ethnicity-based differences in CV risk factors and control rates, to help prevent "the oncoming tsunami of chronic, debilitating, and deadly disorders like diabetes and cardiovascular disease around the world."

Disparities Cure: Evidence-Based Protocols?

The way forward may be fairly direct. According to J. Jeffrey Marshall, MD, cardiac cath lab director at Northeast Georgia Medical Center and president of SCAI, many of the race-based disparities in outcomes still relate to access. Dr. Marshall reviewed a study evaluating in-hospital cardiac arrest based on data from the AHA National Registry of CPR Investigators.10 The analysis covered three study outcomes: survival to hospital discharge, return of spontaneous circulation for 20 minutes (ROSC), and post-resuscitation survival among ROSC patients.

Compared to Caucasians, African Americans were younger, less likely to be male, much sicker (more sepsis, pneumonia, renal failure), and less likely to receive high levels of monitoring. They were more likely to be treated in large urban hospitals and less likely to survive until discharge (25.2% vs. 37.4%). ROSC was less likely for African Americans (55.8% vs. 67.4%), and post-resuscitation survival was lower (45.2% vs. 55.5%).

While delayed defibrillation was higher in African-American patients with in-hospital cardiac arrest (22% vs. 17.4%), aggressive CPR was performed similarly in both groups. Differences were largely explained by the hospital site at which patients received post-resuscitation care.

Dr. Marshall said, "Seventy percent of black patients with in-hospital cardiac arrest were concentrated in the two quintiles of hospitals that had the worst probability for survival to discharge." Among the key hospital factors: lower-quality ICUs; less use of hypothermia, catheterization, ICU, and telemetry monitoring; and lower RN-patient ratios with longer time to first defibrillation.

Another factor generally affecting care equity is the transfer process to PCI/CABG hospitals from non-PCI–capable hospitals. Urban, safety-net hospitals with financial constraints, he said, have poor-quality improvement processes, leading to slower transfer.

Now for the Good News...

Dr. Marshall reviewed the success of the Reperfusion of Acute Myocardial Infarction in North Carolina Emergency Departments (RACE) study, a program organized by local physicians to improve STEMI care across the state. At CRT, he compared pre-RACE data to 1-year RACE data, based on 2,063 patients treated at 65 hospitals. Within this group, 1,140 were treated at PCI centers and 923 at non-PCI centers. Door-to-balloon times were reduced equally among African-American (83 to 74 minutes) and Caucasian patients (84 to 73 minutes). Reperfusion rates, which exceeded 95% both pre- and post-RACE, were similar for both groups.

"RACE shows that the implementation of standardized treatment protocols for STEMI can eliminate the health care disparities found in NRMI," Dr. Marshall told CardioSource WorldNews. "The data show that certain socioeconomic groups tend to present at safety-net hospitals without a good nurse-patient ratio or at critical access rural hospitals—hospitals that are less capable of providing the best care. The fix will be in developing processes that accomplish what RACE did for STEMI."

STEMI protocols utilize processes of care that can be applied to every patient in an even, equitable fashion. Attention to process details like the door-in to door-out times for transfer patients and having resuscitation and cooling protocol teams ready for immediate action will minimize time to treatment and maximize treatment efficiencies.

He added that individual physicians are the drivers for change in CV care. Since processes in STEMI care fail patients at certain hospitals, he offered a new goal: "To identify and eliminate health care disparities one cath lab at a time."


1. Deedwania P. J Am Coll Cardiol. 2013;61:1787-9.
2. Kapoor JR, Kapoor R, Hellkamp AS, et al. J Am Coll Cardiol. 2011;58:1465-71.
3. Mehta RH, Stebbins A, Lopes RD, et al. Am J Med. 2011;124:48-57.
4. Mehta RH, Parsons L, Rao SV, et al. Circulation. 2012;125:1727-34.
5. Popescu I, Vaughan-Sarrazin MS, Rosenthal G. JAMA. 2007;297:2489-95.
6. Kramarow E, Lubitz J, Francis R Jr. Ann Epidem. 2013;23:31-4.
7. Rosamond WD, Chambless LE, Heiss G, et al. Circulation. 2012;125:1848-57.
8. Fletcher RD, Amdur RL, Kolodner R, et al. Circulation. 2012;125:2462-68.
9. Tillin T, Hughes AD, Mayet J, et al. J Am Coll Cardiol. 2013;61:1777-86.
10. Chan PS, Nichol G, Krumholz HM, et al. JAMA. 2009;302:1195-1201.

Keywords: Hospital Mortality, European Continental Ancestry Group, Medicaid, Comorbidity, Healthcare Disparities, Risk Factors, Asian Continental Ancestry Group, Electronic Health Records, Prevalence, Incidence, Time-to-Treatment, Quality Improvement, Registries, Clinical Protocols, Cardiovascular Diseases, Medicare, African Continental Ancestry Group, Nutrition Surveys, Health Maintenance Organizations, Diabetes Mellitus, Safety-net Providers

< Back to Listings