- Email: [email protected]
Does Insurance Effect the Outcome in Patients With Acute Coronary Syndrome?: An Insight from the Most Recent National Inpatient Sample
ARTICLE IN PRESS &
Does Insurance Effect the Outcome in Patients With Acute Coronary Syndrome?: An Insight from the Most Recent National Inpatient Sample Sandipan Chakraborty, MD, MBBS, Dhrubajyoti Bandyopadhyay, MBBS, MD, Birendra Amgai, MBBS, Jasdeep Singh Sidhu, MBBS, Rabin Paudel, MBBS, Soniya Koirala, MBBS, Adrija Hajra, MBBS, MD, Raktim K Ghosh, MBBS, MD, FACP, and Carl J Lavie, MD Abstract: Several studies have shown disparities in outcome in the patients with Acute coronary syndrome (ACS) based on several factors. Treatment might differ based on insurance type. Therefore, we retrospectively analyzed National Inpatient Sample (NIS 2016) data to identify the impact of different types of insurances on mortality outcome in patients admitted with ACS. ICD-CM-10 codes were used to identify hospital discharges with a principal diagnosis of ACS. Observations were stratified based on insurance (Medicare, Medicaid, Private, and No insurance). Primary and secondary outcomes were in-hospital mortality, length of stay and total cost. Any potential confounders were adjusted using multivariate logistic regression. STATA/IC 15.1 Stata Corp LLC was used for analysis. Conflict of interest statement: The undersigned author certifies that there is no conflict of interest. We also verifying that the article is original, is not under consideration by any other journal, and has not been previously published. All copyright ownership of the manuscript entitled (the title of the article) will be transferred to the publishers once the article is accepted for publication in the journal. All the authors have seen and approved the manuscript as well as the order of authors in the manuscript. The article represents original work, and we take full responsibility for the information provided. Sandipan Chakraborty and Dhrubajyoti Bandyopadhyay contributed equally. Curr Probl Cardiol 2019;00:1 10 0146-2806/$ see front matter https://doi.org/10.1016/j.cpcardiol.2019.02.003
Curr Probl Cardiol, 2019
1
ARTICLE IN PRESS A total of 8,01,195 hospitalizations with the primary diagnosis of ACS were identified, of which 59.2% had Medicare, 9.72% had Medicaid, 26.8% had Private insurance, and 4.3% had no insurance. Higher odds of mortality were seen in the patients with Medicare, Medicaid, and Noninsured group. Adjusted Odds ratio for mortality in Medicare was 1.01 (confidence interval [CI]: 0.94-1.1; P = 0.65), in Medicaid was 1.16 (CI: 1.03-1.30; P = 0.01) and in uninsured group was 1.46 (CI: 1.26-1.69; P 0.01). However, the patients with private insurance adjusted odds ratio for mortality were 0.77 (CI: 0.70-0.84; P 0.01) compared to the patients with other insurance groups. Above results show that the disparity exists in the outcome of patients admitted with ACS based on their insurance types, particularly for Medicaid patients. We need further studies to understand the root cause of this disparity. (Curr Probl Cardiol 2019;00:1 10.)
Introduction
A &
cute Coronary Syndrome (ACS) consists of ST-elevation myocardial infarction (STEMI), non-ST elevation Myocardial infarction (NSTEMI), and unstable angina (UA). The mainstay of treatment of STEMI involves early reperfusion strategies, either with percutaneous coronary intervention (PCI) or with fibrinolysis. On the other hand, risk stratification and aggressive risk factors modifications, low molecular weight heparin/ glycoprotein (GP) IIb/IIIa inhibitors and subsequent invasive strategies are the cornerstone of management for NSTEMI/UA. Treatment of patients with ACS as per American Heart Association/American College of Cardiology1-3 guidelines does not reach to all segments of the United States (US) population uniformly. Although Medicare and Medicaid cover significant US population with limited income, reimbursement differs with those with private insurance plans and patients with no insurance.4 Therefore, we aim to find out the effect of insurance on ACS outcome from the analysis of the large national database of US hospital discharges (2016).
Materials and Methods We used recently released National Inpatient Sample (NIS) database 2016 for our study. Population with age group 18 were included in the 2
Curr Probl Cardiol, 2019
ARTICLE IN PRESS study. Inpatient population with STEMI were identified using ICD10 codes (I21.01-I21.09, I21.1, I21.9, I21.21, I21.29, and I21.3). Inpatient population with NSTEMI was identified using ICD10 code I21.4 and UA inpatient population were identified using ICD 10 code I20.0. Principle diagnosis of ACS was established in the inpatient population having either STEMI or NSTEMI or UA. ACS population group based on individual insurance were separated. Patients carrying 2 different types of insurance were excluded from the study, and thus, patients with either Medicare, Medicaid, Private, and No insurance were created. The primary outcome of the study was inpatient mortality. The secondary outcome of the study was the length of stay (LOS), total charge, atrial fibrillation, cardiac arrest, respiratory failure, acute kidney injury, and cardiogenic shock. We compared baseline patient and hospital characteristics in patients with ACS. The univariate and multivariate regression model was used to analyze the primary and secondary outcomes—inpatient mortality adjusted for potential confounders that includes age, gender, hypertension, chronic kidney disease, dyslipidemia, smoking, metabolic syndrome, obesity, hypothyroidism, cerebrovascular accident, peripheral vascular disease, congestive heart failure, and hospital bed size, hospital location, teaching vs nonteaching hospital. After adjusting for confounders in univariate analysis, confounders with P < 0.2 were selected for multivariate analysis. A P value 0.05 was considered significant. Stata IC 15 was used to perform statistical analysis.
Results After excluding the missing values total of 801,195 patients were included in this study, of which 474,300 (59.2%) had Medicare plans. Medicaid covered 78,175 (9.75%) and 214,625 (26.8%) had private insurance, and the rest 34,095 (4.3%) had no insurance coverage. The overall study population was predominantly white (77.2%) and Male (62.3%), with the highest number of female patients represented under Medicare insurance plan (43.2%). The median age in all the study groups was as follows: in Medicare group 73.89 years [confidence interval (CI): 73.7874.00], Medicaid 54.7years(CI: 54.4-54.9), private 57.8 years (CI: 57.6258.02), noninsured 53.8 years(CI: 53.5-54.2). Majority of patients in each group (overall 30.84%) fell in the Low-income group ($ 1-38,999) except for patients having private insurance. Hospitals from the southern US had higher patient representation (41.15% of total), and most of the patients enrolled were from the large urban teaching Hospital setting. Medicaid group was noted to have the higher number of patients having Curr Probl Cardiol, 2019
3
ARTICLE IN PRESS cerebrovascular accident (4.42%), peripheral vascular disease (9.4%), congestive heart failure (22.1%), chronic kidney disease (30%), hypertension (53.2%), diabetes (43.6%), and hypothyroidism (15.74%). Also, the Medicaid group had more smoking burden (29.6%) as compared to the rest. However, the privately insured group had a higher proportion of patients with dyslipidemia (66.6%) metabolic syndrome (0.7%), and obesity (22.9%). All these differences were found to be statistically significant. (Table 1) When adjusted for different variables, the odds ratio (OR) for mortality in Medicare was 1.01 (CI: 0.94-1.1; P = 0.65) which was not statistically significant. The adjusted odds for mortality in Medicaid patients was 1.16 (CI: 1.03-1.30; P = 0.01) and in noninsured patients was 1.46 (CI: 1.261.69; P < 0.01). Patients in the privately insured group were found to have the lowest mortality odds (0.77, CI: 0.70-0.84; P < 0.01) following ACS.(Table 2) Patients in the Medicare group had a higher incidence of atrial fibrillation compared to other groups (OR: 1.22, CI: 1.17-1.26; P < 0.01). Medicaid patients were noted to have a higher rate of respiratory failure and acute renal failure. There was no statistically significant difference in the Incidence of cardiac arrest and shock among different groups. The privately insured patient group more likely to undergo PCI within 24 hours (OR: 1.3, CI: 1.25-1.33; P < 0.01) while patients are having Medicaid less likely got early PCI (OR: 0.8, CI: 0.75-0.82; P < 0.01). (Table 3) The median duration of LOS was highest in Medicare group (4.7 days, CI: 4.63-4.75; P < 0.01) closely followed by Medicare patient group (4.5 days, CI: 4.40-4.65; P = 0.02). Uninsured patients were more likely to be discharged sooner (LOS 3.81 days, CI: 3.67-3.95; P = 0.01), similar to the LOS for privately insured patients (3.82 days, CI: 3.75-3.89; P < 0.01). The median cost on hospitalization was significantly higher in Medicaid group $95349 (P < 0.01) followed by private insurance group ($93,359, P < 0.01). Medicare patients incurred $88,784(P < 0.01) as the median cost of hospitalization which was the lowest.
Discussion As per the Institute of Medicine, health care should be equitable besides being safe, effective, timely, and patient-centered. Appropriate and timely medical intervention is the cornerstone of improved outcome in patients with ACS and acute myocardial infarction (AMI) which is proved by many studies.5-7 However, in spite of the above evidence, multiple studies pointed out the disparities in the care of patients 4
Curr Probl Cardiol, 2019
Curr Probl Cardiol, 2019
TABLE 1. Baseline characteristics Medicare N = 474300 204,898(43.2)
P value Medicaid <0.01 <0.01
N = 78175 29,065(37.18)
P value Private 0.95 <0.01
N = 214625 58,915(27.45)
P value Noninsured <0.01 <0.01
370,381(78.1) 46,813(9.8) 32,584(6.8) 10,529(2.2) 23,72(0.50) 11,668(2.4) 73.89(73.78-74.00) <0.01
42,652(54.56) 14,978(19.16) 11,476(14.68) 3854(4.9) 711(0.90) 4511(5.77) 54.72(54.4-54.9) <0.01
163,287(76.08) 19,939(9.29) 16,698(7.78) 6482(3.02) 1009(0.47) 7211(3.36) 57.82(57.62-58.02) <0.01
<0.01
<0.01
<0.01
147,934(31.19) 13,090(27.6) 110,464(23.29) 84,995(17.92)
32,505(41.58) 21,084(26.97) 15,369(19.66) 9209(11.78) <0.01
87,935(18.54) 110,464(23.29) 193,277(40.75) 82,670(17.43)
<0.01 16,628(21.27) 17,120(21.9) 23,640(30.24) 207,887(26.59)
<0.01 75,177(15.85) 139,681(29.45)
52,690(24.55) 54,987(25.62) 56,661(26.4) 50,308(23.44)
<0.01 <0.01
21,262(62.36) 5524916.26) 4835(14.18) 829(2.43) 181(0.50) 1442(4.23) 53.84(53.51-54.17)
<0.01
<0.01
<0.01 2936(8.61) 5135(15.06) 23,151(67.9) 2874(8.43)
<0.01 32,859(15.31) 64,731(30.16)
<0.01
14,013(41.1) 9550(28.01) 6734(19.75) 3802(11.15)
39,491(18.4) 48,677(22.68) 89,649(41.77) 36,808(17.15) <0.01
10,538(13.48) 21,232(27.16)
P value
N = 34095 9424
0.61 5104(14.97) 9997(28.64)
5
(continued on next page)
ARTICLE IN PRESS
No of patients (N) Female (%) Race (%) White Black Hispanic Asian Native American Other Median age in y, (25th-75th percentile) Median annual income in patients zip code, US dollar (%) 1-38,999 39,000-47,999 48,000-62000 >63,000 Hospital characteristics Hospital region, no (%) Northeast Midwest South West Hospital bed size Small Medium
6
TABLE 1 (continued)
Large Urban location Teaching hospital CVA Peripheral vascular disease Congestive heart failure Chronic kidney disease Early PCI Hypertension Type 2 DM Dyslipidemia Smoking Metabolic syndrome Obesity Hypothyroidism
Medicare
P value Medicaid
259,442(54.7) 434,411(91.59) 301,323(63.53) 20,964(4.42) 44,584(9.4) 104,868(22.11) 142,290(30.0) 95,713 252,517(53.24) 207,032(43.65) 313,702(66.14) 140,393(29.6) 1707(0.36) 73,801(15.56) 74,655(15.74)
<0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01
46,405(59.36) 73,172(93.6) 54,003(69.08) 1790(2.29) 4628(5.92) 13,649(17.46) 12,258(15.68) 20404 46,842(59.92) 33,865(43.32) 47,140(60.3) 15,932(20.38) 414(0.53) 17,246(22.06) 5261(6.73)
P value Private <0.01 <0.01 <0.01 <0.01 0.1 <0.01 <0.01 <0.01 <0.01 <01 <0.01 0.3 <0.01 <0.01
117,035(54.53) 202,434(94.32) 144,722(67.43) 4829(2.25) 9615(4.48) 22,171(10.33) 23,137(10.78) 74,775 131,951(61.48) 74,453(34.69) 143,090(66.67) 51,553(24.02) 1524(0.7) 49,149(22.9) 17,685(8.24)
P value Noninsured <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01
19,226(56.39) 31,575(92.61) 21,909(64.26) 685(2.01) 1224(3.59) 4221(12.38) 3096(9.08) 10,300 21,101(61.89) 11,446(33.57) 18,803(55.15) 5970(17.51) 126(0.3) 6645(19.49) 1705(5)
P value 0.93 0.38 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 0.19 0.02 <0.01
CVA: cerebrovascular accident, DM: diabetes mellitus.
ARTICLE IN PRESS
Curr Probl Cardiol, 2019
ARTICLE IN PRESS TABLE 2. Primary outcomes of interest Mortality
Length of stay (in d)
Total Cost(in dollars)
Medicare Medicaid Private Noninsurance Medicare Medicaid Private Noninsurance Medicare Medicaid Private Noninsurance
OR: 1.01 (CI: 0.94-1.1) OR: 1.16 (CI: 1.03-1.30) OR: 0.77 (CI: 0.70-0.84) OR:1.46 (CI: 1.26-1.69) OR: 4.69 (CI: 4.63-4.75) OR: 4.53 (CI: 4.40-4.65) OR: 3.82 (CI: 3.75-3.89) OR: 3.81 (CI: 3.67-3.95) OR: 88,784 (CI: 86,381-91,185) OR: 95,349 (CI: 91,643-99,054) OR: 93,359 (CI: 90,699-96,018) OR: 89,296 (CI: 85,893-92,699)
P = 0.65 P = 0.01 P < 0.01 P < 0.01 P < 0.01 P = 0.018 P < 0.01 P < 0.01 P < 0.01 P < 0.01 P < 0.01 P = 0.36
TABLE 3. Secondary outcomes of interest Atrial fibrillation Medicare Medicaid Private Noninsurance
OR: 1.22 (CI: 1.17-1.26) OR: 0.90 (CI: 0.85-0.95) OR: 0.88 (CI: 0.85-0.91) OR: 0.76 (CI: 0.70-0.84)
P < 0.01 P < 0.01 P < 0.01 P < 0.01
Medicare Medicaid Private Noninsurance
OR: 0.97 (CI: 0.89-1.06) OR: 1.04 (CI: 0.93-1.17) OR: 0.98 (CI: 0.90-1.07) OR: 1.05 (CI: 0.90-1.22)
P = 0.54 P = 0.42 P = 0.76 P = 0.47
Medicare Medicaid Private Noninsurance
OR: 1.09 (CI: 1.0-1.18) OR: 1.24 (CI: 1.12-1.38) OR: 0.76 (CI: 0.70-0.83) OR: 1.14 (CI: 0.98-1.33)
P = 0.032 P < 0.01 P < 0.01 P = 0.07
Medicare Medicaid Private Noninsurance
OR: 0.98 (CI: 0.94-1.02) OR: 1.31 (CI: 1.24-1.39) OR: 0.83 (CI: 0.80-0.87) OR: 1.20 (CI: 1.11-1.31)
P = 0.52 <0.01 P < 0.01 P < 0.01
Medicare Medicaid Private Noninsurance
OR: 0.95 (CI: 0.89-1.01) OR: 1.01 (CI: 0.93-1.11) OR: 0.99 (CI: 0.94-1.05) OR: 1.15 (CI: 1.02-1.29)
P = 0.13 P = 0.67 P = 0.91 P = 0.38
Medicare Medicaid Private Noninsurance
OR: 0.87 ( CI: 0.84-0.89) OR: 0.79 (CI: 0.75-0.82) OR: 1.29 (CI: 1.25-1.33) OR: 0.90 (CI: 0.84-0.96)
P < 0.01 P < 0.01 P < 0.01 P = 0.002
Cardiac arrest
Respiratory failure
Acute Renal Failure
Shock
Early PCI (<24 h)
Curr Probl Cardiol, 2019
7
ARTICLE IN PRESS with ACS.8-11 Canto et al. demonstrated that Medicare patients are less likely to receive reperfusion therapy or invasive procedure in AMI.8 Our study also supported the fact that even in recent years, the odds of receiving an early PCI for AMI (< 24 hours) is lowest in Medicaid patients. Philbin et al. also confirmed the above findings.12 The drug prescription behavior is also influenced by insurance status, emphasizing the fact that patients with the better insurance coverage receive the newest and more expensive medications.13,14 In-hospital mortality also increased in patients with Medicaid payer status as described by Calvin et al.15 which corroborates with our findings. Our study demonstrated increased odds of dying in patients with no insurance and least odds for mortality in private payer; however, mortality is comparable in Medicare and Medicaid patients. Indeed, Medicaid patients are among the sicker population and having multiple comorbidities, thus emphasizing worse outcomes compared to private and Medicare patients. Even after adjustment of age, sex, hospital status, hospital location, and other potential confounders, Medicaid and noninsured patients are less likely to receive evidence-based care. Our study also demonstrated the lower LOS for private and noninsure patients, which is statistically significant. The total cost of care is even higher in private and Medicaid category. These findings could emphasize the importance of more resource utilization in private patients leading to increased cost and decreased the LOS compared to an increased LOS in other categories except noninsure population owing to possible delayed delivery of evidence-based care. The secondary outcomes of interest are lowest in the private insurance group, whereas the incidence of acute respiratory failure and acute renal failure is significantly highest in the Medicaid group. There could be multiple reasons for these discrepancies. First, Medicaid patients have numerous comorbidities making them a considerably sicker population; second, Medicaid payer status is associated with poor socioeconomic condition, making them less willing to seek evidence-based care when offered; and last, Medicaid patients receive care in hospitals less likely to deliver quality-based care. In 1 study, Medicaid patients are being managed less by the cardiologist who is more likely to provide evidence-based care.15 In our study, even after adjustment for potential confounders, there is a gap in care for Medicaid and noninsured patients. Our findings highlighted the importance of restructuring the Medicaid services and refinancing to ensure better patient care.16 Although ‘pay-for-performance’ initiatives evaluated efforts to monitor and improve quality of care, Medicaid service provided by individual states makes quality assurance programs difficult to operate.17-19 This is more contingent in the current situation where federal funds are being restricted to make health care more affordable, thus making access to evidence-based care difficult for many patients. 8
Curr Probl Cardiol, 2019
ARTICLE IN PRESS Limitations As our analysis is based on the large administrative NIS database and being a retrospective analysis, it does have some inherent limitations. It only represents 20% of all inpatient admission in the US. NIS records admissions, not the individual patient, so there is a chance to record rehospitalization as an independent observation, and also a difference in coding and coding errors are possible. Patient preferences towards selecting an invasive procedure cannot be determined. There is also a difference in severity of the coexisting medical condition. Last, the coverage for medications and procedures differ among Medicare and Medicaid patients, which cannot be taken into analysis.
Conclusions In the modern era also, a disparity exists in delivering evidence-based care based on insurance coverages. Medicaid patients experienced differences of care accounting for the higher risk of death and secondary complications. Patients with private insurance witnessed favorable outcome in comparison to other groups. As it has repeatedly been shown that the use of evidence-based care in ACS results in better outcomes, further steps need to be taken to identify the root cause of this disparity and to find effective strategies to bridge the gap in health care, yet making it more affordable and accessible for more patients.
REFERENCES 1. Braunwald E, Antman EM, Beasley JW, et al. ACC/AHA guidelines for the management of patients with unstable angina and non-ST-segment elevation myocardial infarction: executive summary and recommendations: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on the Management of Patients With Unstable Angina). Circulation 2000;102(10):1193–209. 2. Braunwald E, Antman EM, Beasley JW, et al. ACC/AHA 2002 guideline update for the management of patients with unstable angina and non-ST-segment elevation myocardial infarction summary article: a report of the American College of Cardiology/American Heart Association task force on practice guidelines (Committee on the Management of Patients With Unstable Angina). J Am Coll Cardiol 2002;40(7):1366–74. 3. Cannon CP, Weintraub WS, Demopoulos LA, et al. Comparison of early invasive and conservative strategies in patients with unstable coronary syndromes treated with the glycoprotein IIb/IIIa inhibitor tirofiban. N Engl J Med 2001;344(25):1879–87. 4. Kreindel S, Rosetti R, Goldberg R, et al. Health insurance coverage and outcome following acute myocardial infarction. A community-wide perspective. Arch Intern Med 1997;157(7):758–62. Curr Probl Cardiol, 2019
9
ARTICLE IN PRESS 5. Rosamond WD, Chambless LE, Heiss G, et al. Twenty-two-year trends in incidence of myocardial infarction, coronary heart disease mortality, and case fatality in 4 US communities, 1987 2008. Circulation 2012;125(15):1848–57. 6. Kaul P, Armstrong PW, Chang WC, et al. Long-term mortality of patients with acute myocardial infarction in the United States and Canada: comparison of patients enrolled in Global Utilization of Streptokinase and t-PA for Occluded Coronary Arteries (GUSTO)-I. Circulation 2004;110(13):1754–60. 7. Kleiman NS, Lincoff AM, Flaker GC, et al. Early percutaneous coronary intervention, platelet inhibition with eptifibatide, and clinical outcomes in patients with acute coronary syndromes. PURSUIT Investigators. Circulation 2000;101(7):751–7. 8. Canto JG, Rogers WJ, French WJ, Gore JM, Chandra NC, Barron HV. Payer status and the utilization of hospital resources in acute myocardial infarction: a report from the National Registry of Myocardial Infarction 2. Arch Intern Med 2000;160(6):817–23. 9. Carlisle DM, Leake BD, Shapiro MF. Racial and ethnic disparities in the use of cardiovascular procedures: associations with type of health insurance. Am J Public Health 1997;87(2):263–7. 10. Seddon ME, Ayanian JZ, Landrum MB, et al. Quality of ambulatory care after myocardial infarction among Medicare patients by type of insurance and region. Am J Med 2001;111(1):24–32. 11. Leape LL, Hilborne LH, Bell R, Kamberg C, Brook RH. Underuse of cardiac procedures: do women, ethnic minorities, and the uninsured fail to receive needed revascularization? Ann Intern Med 1999;130(3):183–92. 12. Philbin EF, McCullough PA, DiSalvo TG, Dec GW, Jenkins PL, Weaver WD. Underuse of invasive procedures among Medicaid patients with acute myocardial infarction. Am J Public Health 2001;91(7):1082–8. 13. Mott DA, Kreling DH. The association of insurance type with costs of dispensed drugs. Inquiry 1998;35(1):23–35. 14. Mamdani MM, Tu K, Austin PC, Alter DA. Influence of socioeconomic status on drug selection for the elderly in Canada. Ann Pharmacother 2002;36(5):804–8. 15. Calvin JE, Roe MT, Chen AY, et al. Insurance coverage and care of patients with non-ST-segment elevation acute coronary syndromes. Ann Intern Med 2006;145 (10):739–48. 16. Nelson AR. Unequal treatment: report of the Institute of Medicine on racial and ethnic disparities in healthcare. Ann Thorac Surg 2003;76(4):S1377–81. 17. Burwen DR, Galusha DH, Lewis JM, et al. National and state trends in quality of care for acute myocardial infarction between 1994-1995 and 1998-1999: the Medicare health care quality improvement program. Arch Intern Med 2003;163(12):1430–9. 18. The Cooperative Cardiovascular Project Best Practices Working Group. Improving care for acute myocardial infarction: experience from the Cooperative Cardiovascular Project. Jt Comm J Qual Improv 1998;24:480–90. 19. Jencks SF, Cuerdon T, Burwen DR, et al. Quality of medical care delivered to Medicare beneficiaries: a profile at state and national levels. JAMA 2000;284(13):1670–6.
10
Curr Probl Cardiol, 2019
Does Insurance Effect the Outcome in Patients With Acute Coronary Syndrome?: An Insight from the Most Recent National Inpatient Sample Sandipan Chakraborty, MD, MBBS, Dhrubajyoti Bandyopadhyay, MBBS, MD, Birendra Amgai, MBBS, Jasdeep Singh Sidhu, MBBS, Rabin Paudel, MBBS, Soniya Koirala, MBBS, Adrija Hajra, MBBS, MD, Raktim K Ghosh, MBBS, MD, FACP, and Carl J Lavie, MD Abstract: Several studies have shown disparities in outcome in the patients with Acute coronary syndrome (ACS) based on several factors. Treatment might differ based on insurance type. Therefore, we retrospectively analyzed National Inpatient Sample (NIS 2016) data to identify the impact of different types of insurances on mortality outcome in patients admitted with ACS. ICD-CM-10 codes were used to identify hospital discharges with a principal diagnosis of ACS. Observations were stratified based on insurance (Medicare, Medicaid, Private, and No insurance). Primary and secondary outcomes were in-hospital mortality, length of stay and total cost. Any potential confounders were adjusted using multivariate logistic regression. STATA/IC 15.1 Stata Corp LLC was used for analysis. Conflict of interest statement: The undersigned author certifies that there is no conflict of interest. We also verifying that the article is original, is not under consideration by any other journal, and has not been previously published. All copyright ownership of the manuscript entitled (the title of the article) will be transferred to the publishers once the article is accepted for publication in the journal. All the authors have seen and approved the manuscript as well as the order of authors in the manuscript. The article represents original work, and we take full responsibility for the information provided. Sandipan Chakraborty and Dhrubajyoti Bandyopadhyay contributed equally. Curr Probl Cardiol 2019;00:1 10 0146-2806/$ see front matter https://doi.org/10.1016/j.cpcardiol.2019.02.003
Curr Probl Cardiol, 2019
1
ARTICLE IN PRESS A total of 8,01,195 hospitalizations with the primary diagnosis of ACS were identified, of which 59.2% had Medicare, 9.72% had Medicaid, 26.8% had Private insurance, and 4.3% had no insurance. Higher odds of mortality were seen in the patients with Medicare, Medicaid, and Noninsured group. Adjusted Odds ratio for mortality in Medicare was 1.01 (confidence interval [CI]: 0.94-1.1; P = 0.65), in Medicaid was 1.16 (CI: 1.03-1.30; P = 0.01) and in uninsured group was 1.46 (CI: 1.26-1.69; P 0.01). However, the patients with private insurance adjusted odds ratio for mortality were 0.77 (CI: 0.70-0.84; P 0.01) compared to the patients with other insurance groups. Above results show that the disparity exists in the outcome of patients admitted with ACS based on their insurance types, particularly for Medicaid patients. We need further studies to understand the root cause of this disparity. (Curr Probl Cardiol 2019;00:1 10.)
Introduction
A &
cute Coronary Syndrome (ACS) consists of ST-elevation myocardial infarction (STEMI), non-ST elevation Myocardial infarction (NSTEMI), and unstable angina (UA). The mainstay of treatment of STEMI involves early reperfusion strategies, either with percutaneous coronary intervention (PCI) or with fibrinolysis. On the other hand, risk stratification and aggressive risk factors modifications, low molecular weight heparin/ glycoprotein (GP) IIb/IIIa inhibitors and subsequent invasive strategies are the cornerstone of management for NSTEMI/UA. Treatment of patients with ACS as per American Heart Association/American College of Cardiology1-3 guidelines does not reach to all segments of the United States (US) population uniformly. Although Medicare and Medicaid cover significant US population with limited income, reimbursement differs with those with private insurance plans and patients with no insurance.4 Therefore, we aim to find out the effect of insurance on ACS outcome from the analysis of the large national database of US hospital discharges (2016).
Materials and Methods We used recently released National Inpatient Sample (NIS) database 2016 for our study. Population with age group 18 were included in the 2
Curr Probl Cardiol, 2019
ARTICLE IN PRESS study. Inpatient population with STEMI were identified using ICD10 codes (I21.01-I21.09, I21.1, I21.9, I21.21, I21.29, and I21.3). Inpatient population with NSTEMI was identified using ICD10 code I21.4 and UA inpatient population were identified using ICD 10 code I20.0. Principle diagnosis of ACS was established in the inpatient population having either STEMI or NSTEMI or UA. ACS population group based on individual insurance were separated. Patients carrying 2 different types of insurance were excluded from the study, and thus, patients with either Medicare, Medicaid, Private, and No insurance were created. The primary outcome of the study was inpatient mortality. The secondary outcome of the study was the length of stay (LOS), total charge, atrial fibrillation, cardiac arrest, respiratory failure, acute kidney injury, and cardiogenic shock. We compared baseline patient and hospital characteristics in patients with ACS. The univariate and multivariate regression model was used to analyze the primary and secondary outcomes—inpatient mortality adjusted for potential confounders that includes age, gender, hypertension, chronic kidney disease, dyslipidemia, smoking, metabolic syndrome, obesity, hypothyroidism, cerebrovascular accident, peripheral vascular disease, congestive heart failure, and hospital bed size, hospital location, teaching vs nonteaching hospital. After adjusting for confounders in univariate analysis, confounders with P < 0.2 were selected for multivariate analysis. A P value 0.05 was considered significant. Stata IC 15 was used to perform statistical analysis.
Results After excluding the missing values total of 801,195 patients were included in this study, of which 474,300 (59.2%) had Medicare plans. Medicaid covered 78,175 (9.75%) and 214,625 (26.8%) had private insurance, and the rest 34,095 (4.3%) had no insurance coverage. The overall study population was predominantly white (77.2%) and Male (62.3%), with the highest number of female patients represented under Medicare insurance plan (43.2%). The median age in all the study groups was as follows: in Medicare group 73.89 years [confidence interval (CI): 73.7874.00], Medicaid 54.7years(CI: 54.4-54.9), private 57.8 years (CI: 57.6258.02), noninsured 53.8 years(CI: 53.5-54.2). Majority of patients in each group (overall 30.84%) fell in the Low-income group ($ 1-38,999) except for patients having private insurance. Hospitals from the southern US had higher patient representation (41.15% of total), and most of the patients enrolled were from the large urban teaching Hospital setting. Medicaid group was noted to have the higher number of patients having Curr Probl Cardiol, 2019
3
ARTICLE IN PRESS cerebrovascular accident (4.42%), peripheral vascular disease (9.4%), congestive heart failure (22.1%), chronic kidney disease (30%), hypertension (53.2%), diabetes (43.6%), and hypothyroidism (15.74%). Also, the Medicaid group had more smoking burden (29.6%) as compared to the rest. However, the privately insured group had a higher proportion of patients with dyslipidemia (66.6%) metabolic syndrome (0.7%), and obesity (22.9%). All these differences were found to be statistically significant. (Table 1) When adjusted for different variables, the odds ratio (OR) for mortality in Medicare was 1.01 (CI: 0.94-1.1; P = 0.65) which was not statistically significant. The adjusted odds for mortality in Medicaid patients was 1.16 (CI: 1.03-1.30; P = 0.01) and in noninsured patients was 1.46 (CI: 1.261.69; P < 0.01). Patients in the privately insured group were found to have the lowest mortality odds (0.77, CI: 0.70-0.84; P < 0.01) following ACS.(Table 2) Patients in the Medicare group had a higher incidence of atrial fibrillation compared to other groups (OR: 1.22, CI: 1.17-1.26; P < 0.01). Medicaid patients were noted to have a higher rate of respiratory failure and acute renal failure. There was no statistically significant difference in the Incidence of cardiac arrest and shock among different groups. The privately insured patient group more likely to undergo PCI within 24 hours (OR: 1.3, CI: 1.25-1.33; P < 0.01) while patients are having Medicaid less likely got early PCI (OR: 0.8, CI: 0.75-0.82; P < 0.01). (Table 3) The median duration of LOS was highest in Medicare group (4.7 days, CI: 4.63-4.75; P < 0.01) closely followed by Medicare patient group (4.5 days, CI: 4.40-4.65; P = 0.02). Uninsured patients were more likely to be discharged sooner (LOS 3.81 days, CI: 3.67-3.95; P = 0.01), similar to the LOS for privately insured patients (3.82 days, CI: 3.75-3.89; P < 0.01). The median cost on hospitalization was significantly higher in Medicaid group $95349 (P < 0.01) followed by private insurance group ($93,359, P < 0.01). Medicare patients incurred $88,784(P < 0.01) as the median cost of hospitalization which was the lowest.
Discussion As per the Institute of Medicine, health care should be equitable besides being safe, effective, timely, and patient-centered. Appropriate and timely medical intervention is the cornerstone of improved outcome in patients with ACS and acute myocardial infarction (AMI) which is proved by many studies.5-7 However, in spite of the above evidence, multiple studies pointed out the disparities in the care of patients 4
Curr Probl Cardiol, 2019
Curr Probl Cardiol, 2019
TABLE 1. Baseline characteristics Medicare N = 474300 204,898(43.2)
P value Medicaid <0.01 <0.01
N = 78175 29,065(37.18)
P value Private 0.95 <0.01
N = 214625 58,915(27.45)
P value Noninsured <0.01 <0.01
370,381(78.1) 46,813(9.8) 32,584(6.8) 10,529(2.2) 23,72(0.50) 11,668(2.4) 73.89(73.78-74.00) <0.01
42,652(54.56) 14,978(19.16) 11,476(14.68) 3854(4.9) 711(0.90) 4511(5.77) 54.72(54.4-54.9) <0.01
163,287(76.08) 19,939(9.29) 16,698(7.78) 6482(3.02) 1009(0.47) 7211(3.36) 57.82(57.62-58.02) <0.01
<0.01
<0.01
<0.01
147,934(31.19) 13,090(27.6) 110,464(23.29) 84,995(17.92)
32,505(41.58) 21,084(26.97) 15,369(19.66) 9209(11.78) <0.01
87,935(18.54) 110,464(23.29) 193,277(40.75) 82,670(17.43)
<0.01 16,628(21.27) 17,120(21.9) 23,640(30.24) 207,887(26.59)
<0.01 75,177(15.85) 139,681(29.45)
52,690(24.55) 54,987(25.62) 56,661(26.4) 50,308(23.44)
<0.01 <0.01
21,262(62.36) 5524916.26) 4835(14.18) 829(2.43) 181(0.50) 1442(4.23) 53.84(53.51-54.17)
<0.01
<0.01
<0.01 2936(8.61) 5135(15.06) 23,151(67.9) 2874(8.43)
<0.01 32,859(15.31) 64,731(30.16)
<0.01
14,013(41.1) 9550(28.01) 6734(19.75) 3802(11.15)
39,491(18.4) 48,677(22.68) 89,649(41.77) 36,808(17.15) <0.01
10,538(13.48) 21,232(27.16)
P value
N = 34095 9424
0.61 5104(14.97) 9997(28.64)
5
(continued on next page)
ARTICLE IN PRESS
No of patients (N) Female (%) Race (%) White Black Hispanic Asian Native American Other Median age in y, (25th-75th percentile) Median annual income in patients zip code, US dollar (%) 1-38,999 39,000-47,999 48,000-62000 >63,000 Hospital characteristics Hospital region, no (%) Northeast Midwest South West Hospital bed size Small Medium
6
TABLE 1 (continued)
Large Urban location Teaching hospital CVA Peripheral vascular disease Congestive heart failure Chronic kidney disease Early PCI Hypertension Type 2 DM Dyslipidemia Smoking Metabolic syndrome Obesity Hypothyroidism
Medicare
P value Medicaid
259,442(54.7) 434,411(91.59) 301,323(63.53) 20,964(4.42) 44,584(9.4) 104,868(22.11) 142,290(30.0) 95,713 252,517(53.24) 207,032(43.65) 313,702(66.14) 140,393(29.6) 1707(0.36) 73,801(15.56) 74,655(15.74)
<0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01
46,405(59.36) 73,172(93.6) 54,003(69.08) 1790(2.29) 4628(5.92) 13,649(17.46) 12,258(15.68) 20404 46,842(59.92) 33,865(43.32) 47,140(60.3) 15,932(20.38) 414(0.53) 17,246(22.06) 5261(6.73)
P value Private <0.01 <0.01 <0.01 <0.01 0.1 <0.01 <0.01 <0.01 <0.01 <01 <0.01 0.3 <0.01 <0.01
117,035(54.53) 202,434(94.32) 144,722(67.43) 4829(2.25) 9615(4.48) 22,171(10.33) 23,137(10.78) 74,775 131,951(61.48) 74,453(34.69) 143,090(66.67) 51,553(24.02) 1524(0.7) 49,149(22.9) 17,685(8.24)
P value Noninsured <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01
19,226(56.39) 31,575(92.61) 21,909(64.26) 685(2.01) 1224(3.59) 4221(12.38) 3096(9.08) 10,300 21,101(61.89) 11,446(33.57) 18,803(55.15) 5970(17.51) 126(0.3) 6645(19.49) 1705(5)
P value 0.93 0.38 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01 0.19 0.02 <0.01
CVA: cerebrovascular accident, DM: diabetes mellitus.
ARTICLE IN PRESS
Curr Probl Cardiol, 2019
ARTICLE IN PRESS TABLE 2. Primary outcomes of interest Mortality
Length of stay (in d)
Total Cost(in dollars)
Medicare Medicaid Private Noninsurance Medicare Medicaid Private Noninsurance Medicare Medicaid Private Noninsurance
OR: 1.01 (CI: 0.94-1.1) OR: 1.16 (CI: 1.03-1.30) OR: 0.77 (CI: 0.70-0.84) OR:1.46 (CI: 1.26-1.69) OR: 4.69 (CI: 4.63-4.75) OR: 4.53 (CI: 4.40-4.65) OR: 3.82 (CI: 3.75-3.89) OR: 3.81 (CI: 3.67-3.95) OR: 88,784 (CI: 86,381-91,185) OR: 95,349 (CI: 91,643-99,054) OR: 93,359 (CI: 90,699-96,018) OR: 89,296 (CI: 85,893-92,699)
P = 0.65 P = 0.01 P < 0.01 P < 0.01 P < 0.01 P = 0.018 P < 0.01 P < 0.01 P < 0.01 P < 0.01 P < 0.01 P = 0.36
TABLE 3. Secondary outcomes of interest Atrial fibrillation Medicare Medicaid Private Noninsurance
OR: 1.22 (CI: 1.17-1.26) OR: 0.90 (CI: 0.85-0.95) OR: 0.88 (CI: 0.85-0.91) OR: 0.76 (CI: 0.70-0.84)
P < 0.01 P < 0.01 P < 0.01 P < 0.01
Medicare Medicaid Private Noninsurance
OR: 0.97 (CI: 0.89-1.06) OR: 1.04 (CI: 0.93-1.17) OR: 0.98 (CI: 0.90-1.07) OR: 1.05 (CI: 0.90-1.22)
P = 0.54 P = 0.42 P = 0.76 P = 0.47
Medicare Medicaid Private Noninsurance
OR: 1.09 (CI: 1.0-1.18) OR: 1.24 (CI: 1.12-1.38) OR: 0.76 (CI: 0.70-0.83) OR: 1.14 (CI: 0.98-1.33)
P = 0.032 P < 0.01 P < 0.01 P = 0.07
Medicare Medicaid Private Noninsurance
OR: 0.98 (CI: 0.94-1.02) OR: 1.31 (CI: 1.24-1.39) OR: 0.83 (CI: 0.80-0.87) OR: 1.20 (CI: 1.11-1.31)
P = 0.52 <0.01 P < 0.01 P < 0.01
Medicare Medicaid Private Noninsurance
OR: 0.95 (CI: 0.89-1.01) OR: 1.01 (CI: 0.93-1.11) OR: 0.99 (CI: 0.94-1.05) OR: 1.15 (CI: 1.02-1.29)
P = 0.13 P = 0.67 P = 0.91 P = 0.38
Medicare Medicaid Private Noninsurance
OR: 0.87 ( CI: 0.84-0.89) OR: 0.79 (CI: 0.75-0.82) OR: 1.29 (CI: 1.25-1.33) OR: 0.90 (CI: 0.84-0.96)
P < 0.01 P < 0.01 P < 0.01 P = 0.002
Cardiac arrest
Respiratory failure
Acute Renal Failure
Shock
Early PCI (<24 h)
Curr Probl Cardiol, 2019
7
ARTICLE IN PRESS with ACS.8-11 Canto et al. demonstrated that Medicare patients are less likely to receive reperfusion therapy or invasive procedure in AMI.8 Our study also supported the fact that even in recent years, the odds of receiving an early PCI for AMI (< 24 hours) is lowest in Medicaid patients. Philbin et al. also confirmed the above findings.12 The drug prescription behavior is also influenced by insurance status, emphasizing the fact that patients with the better insurance coverage receive the newest and more expensive medications.13,14 In-hospital mortality also increased in patients with Medicaid payer status as described by Calvin et al.15 which corroborates with our findings. Our study demonstrated increased odds of dying in patients with no insurance and least odds for mortality in private payer; however, mortality is comparable in Medicare and Medicaid patients. Indeed, Medicaid patients are among the sicker population and having multiple comorbidities, thus emphasizing worse outcomes compared to private and Medicare patients. Even after adjustment of age, sex, hospital status, hospital location, and other potential confounders, Medicaid and noninsured patients are less likely to receive evidence-based care. Our study also demonstrated the lower LOS for private and noninsure patients, which is statistically significant. The total cost of care is even higher in private and Medicaid category. These findings could emphasize the importance of more resource utilization in private patients leading to increased cost and decreased the LOS compared to an increased LOS in other categories except noninsure population owing to possible delayed delivery of evidence-based care. The secondary outcomes of interest are lowest in the private insurance group, whereas the incidence of acute respiratory failure and acute renal failure is significantly highest in the Medicaid group. There could be multiple reasons for these discrepancies. First, Medicaid patients have numerous comorbidities making them a considerably sicker population; second, Medicaid payer status is associated with poor socioeconomic condition, making them less willing to seek evidence-based care when offered; and last, Medicaid patients receive care in hospitals less likely to deliver quality-based care. In 1 study, Medicaid patients are being managed less by the cardiologist who is more likely to provide evidence-based care.15 In our study, even after adjustment for potential confounders, there is a gap in care for Medicaid and noninsured patients. Our findings highlighted the importance of restructuring the Medicaid services and refinancing to ensure better patient care.16 Although ‘pay-for-performance’ initiatives evaluated efforts to monitor and improve quality of care, Medicaid service provided by individual states makes quality assurance programs difficult to operate.17-19 This is more contingent in the current situation where federal funds are being restricted to make health care more affordable, thus making access to evidence-based care difficult for many patients. 8
Curr Probl Cardiol, 2019
ARTICLE IN PRESS Limitations As our analysis is based on the large administrative NIS database and being a retrospective analysis, it does have some inherent limitations. It only represents 20% of all inpatient admission in the US. NIS records admissions, not the individual patient, so there is a chance to record rehospitalization as an independent observation, and also a difference in coding and coding errors are possible. Patient preferences towards selecting an invasive procedure cannot be determined. There is also a difference in severity of the coexisting medical condition. Last, the coverage for medications and procedures differ among Medicare and Medicaid patients, which cannot be taken into analysis.
Conclusions In the modern era also, a disparity exists in delivering evidence-based care based on insurance coverages. Medicaid patients experienced differences of care accounting for the higher risk of death and secondary complications. Patients with private insurance witnessed favorable outcome in comparison to other groups. As it has repeatedly been shown that the use of evidence-based care in ACS results in better outcomes, further steps need to be taken to identify the root cause of this disparity and to find effective strategies to bridge the gap in health care, yet making it more affordable and accessible for more patients.
REFERENCES 1. Braunwald E, Antman EM, Beasley JW, et al. ACC/AHA guidelines for the management of patients with unstable angina and non-ST-segment elevation myocardial infarction: executive summary and recommendations: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on the Management of Patients With Unstable Angina). Circulation 2000;102(10):1193–209. 2. Braunwald E, Antman EM, Beasley JW, et al. ACC/AHA 2002 guideline update for the management of patients with unstable angina and non-ST-segment elevation myocardial infarction summary article: a report of the American College of Cardiology/American Heart Association task force on practice guidelines (Committee on the Management of Patients With Unstable Angina). J Am Coll Cardiol 2002;40(7):1366–74. 3. Cannon CP, Weintraub WS, Demopoulos LA, et al. Comparison of early invasive and conservative strategies in patients with unstable coronary syndromes treated with the glycoprotein IIb/IIIa inhibitor tirofiban. N Engl J Med 2001;344(25):1879–87. 4. Kreindel S, Rosetti R, Goldberg R, et al. Health insurance coverage and outcome following acute myocardial infarction. A community-wide perspective. Arch Intern Med 1997;157(7):758–62. Curr Probl Cardiol, 2019
9
ARTICLE IN PRESS 5. Rosamond WD, Chambless LE, Heiss G, et al. Twenty-two-year trends in incidence of myocardial infarction, coronary heart disease mortality, and case fatality in 4 US communities, 1987 2008. Circulation 2012;125(15):1848–57. 6. Kaul P, Armstrong PW, Chang WC, et al. Long-term mortality of patients with acute myocardial infarction in the United States and Canada: comparison of patients enrolled in Global Utilization of Streptokinase and t-PA for Occluded Coronary Arteries (GUSTO)-I. Circulation 2004;110(13):1754–60. 7. Kleiman NS, Lincoff AM, Flaker GC, et al. Early percutaneous coronary intervention, platelet inhibition with eptifibatide, and clinical outcomes in patients with acute coronary syndromes. PURSUIT Investigators. Circulation 2000;101(7):751–7. 8. Canto JG, Rogers WJ, French WJ, Gore JM, Chandra NC, Barron HV. Payer status and the utilization of hospital resources in acute myocardial infarction: a report from the National Registry of Myocardial Infarction 2. Arch Intern Med 2000;160(6):817–23. 9. Carlisle DM, Leake BD, Shapiro MF. Racial and ethnic disparities in the use of cardiovascular procedures: associations with type of health insurance. Am J Public Health 1997;87(2):263–7. 10. Seddon ME, Ayanian JZ, Landrum MB, et al. Quality of ambulatory care after myocardial infarction among Medicare patients by type of insurance and region. Am J Med 2001;111(1):24–32. 11. Leape LL, Hilborne LH, Bell R, Kamberg C, Brook RH. Underuse of cardiac procedures: do women, ethnic minorities, and the uninsured fail to receive needed revascularization? Ann Intern Med 1999;130(3):183–92. 12. Philbin EF, McCullough PA, DiSalvo TG, Dec GW, Jenkins PL, Weaver WD. Underuse of invasive procedures among Medicaid patients with acute myocardial infarction. Am J Public Health 2001;91(7):1082–8. 13. Mott DA, Kreling DH. The association of insurance type with costs of dispensed drugs. Inquiry 1998;35(1):23–35. 14. Mamdani MM, Tu K, Austin PC, Alter DA. Influence of socioeconomic status on drug selection for the elderly in Canada. Ann Pharmacother 2002;36(5):804–8. 15. Calvin JE, Roe MT, Chen AY, et al. Insurance coverage and care of patients with non-ST-segment elevation acute coronary syndromes. Ann Intern Med 2006;145 (10):739–48. 16. Nelson AR. Unequal treatment: report of the Institute of Medicine on racial and ethnic disparities in healthcare. Ann Thorac Surg 2003;76(4):S1377–81. 17. Burwen DR, Galusha DH, Lewis JM, et al. National and state trends in quality of care for acute myocardial infarction between 1994-1995 and 1998-1999: the Medicare health care quality improvement program. Arch Intern Med 2003;163(12):1430–9. 18. The Cooperative Cardiovascular Project Best Practices Working Group. Improving care for acute myocardial infarction: experience from the Cooperative Cardiovascular Project. Jt Comm J Qual Improv 1998;24:480–90. 19. Jencks SF, Cuerdon T, Burwen DR, et al. Quality of medical care delivered to Medicare beneficiaries: a profile at state and national levels. JAMA 2000;284(13):1670–6.
10
Curr Probl Cardiol, 2019