Cost of Care for New-Onset Acute Coronary Syndrome Patients Who Undergo Coronary Revascularization

Several U.S.-based studies have estimated the overall cost of illness for acute coronary syndrome (ACS).^1–5 However, these studies either do not address the costs of care for the growing group of patients who undergo coronary revascularization, or examine the typically younger, commercially-insured population.^1–4 The therapeutic approach for the ACS patient is the subject of numerous clinical studies and guidelines.^6–8 Consensus guidelines now recommend early invasive therapy for the high-risk ACS patients.⁸ The frequency of an early invasive strategy in ACS is reflected in data from the American College of Cardiology National Cardiovascular Data Registry indicating that over 60% of percutaneous coronary interventions (PCI) are performed in patients with ACS.⁹ Guidelines for drug use in patients undergoing revascularization, with or without ACS, are available.^8,10 The 2004 update of coronary artery bypass graft (CABG) surgery guidelines broadly recommends long-term aspirin and statin therapy for appropriate patients in the post-operative phase.¹⁰ Similarly, the 2002 ACS therapy guideline update recommends antiplatelet agents, aggressive lipid and blood pressure management, beta-blockers, and angiotensin-converting enzyme inhibitors in appropriate patients.⁸ If patients go on to PCI and receive a stent, there is general agreement that clopidogrel should be initiated and continued for up to 12 months.¹¹ However, data from the clinical practice setting shows gaps between treatment guidelines and actual practice.^12–14 The purpose of this study was to examine healthcare utilization in managed care patients with new-onset ACS who also underwent coronary revascularization within the first year following the index event. The study objectives were to: 1) learn details of the revascularization, such as type, rate, and when it was performed; 2) estimate total health services utilization and cost; and 3) understand patterns of drug therapy use. Methods Study design and data source. This was an observational, retrospective study that used administrative claims data from a large U.S. managed care organization (MCO). Patients were selected from a research database that maintains enrollment, provider, facility, and pharmacy records for a managed care population. The MCO is a large, national organization that consists of regional health plans throughout the United States, with a discounted fee-for-service model. The data source for this study was de-identified and linked enrollment, provider, facility, and pharmacy records for approximately 5 million individuals. Patient selection. We utilized the definition of ACS outlined by Grech and Ramsdale, which encompasses unstable angina (UA), non-ST segment elevation myocardial infarction (NSTEMI), and ST-segment elevation myocardial infarction (STEMI).15 Patients included in the study experienced their index ACS event during the time period from 7/1/1999 to 6/30/2001, were > 18 years old, and had a subsequent revascularization procedure during the first year of follow-up. Patients were included if they had a medical claim for ICD-9 code 410.xx (acute myocardial infarction) or 411.1x (intermediate coronary syndrome), and an in-patient facility or emergency room (ER) admission as the site of service for the index claim. The definition of new-onset ACS was the absence of ACS claims diagnoses during the previous 6 months, and the index claim could not be for 410.x2 (which indicates a subsequent episode of care). If patients did not have 6 months of continuous enrollment in the health plan prior to their index claim, they were excluded from the study. Patients were followed until they disenrolled from the health plan, or for up to 12 months following study entry, or death was recorded in the medical claims (whichever occurred first). The final inclusion criteria were a claim for a CABG, percutaneous transluminal coronary angioplasty (PTCA), or stent procedure during the study period. Patients were identified by ICD-9 procedure/CPT codes: PTCA = 36.01, 36.02, 36.05, 92982; stent = 36.06, 36.07, 92980; CABG = 36.1x, 33510–33514, 33516, 33533–33536. The date the procedure occurred was also recorded in order to categorize it as associated with the index event or post-index event. Study measurements. Patient characteristics and outcome measures of interest in this study were: demographics, comorbidities, medical and pharmacy utilization data, and costs. Demographic information such as age and gender were recorded and comorbidities were identified from the medical claims using ICD-9 diagnosis codes. The University of Washington Diagnostic Clusters were used to group diagnosis codes for identification of comorbidities.¹⁶ This method categorizes comorbidities into 120 clusters for analytic purposes. The total number and cost of health services utilized during the study period (including the index event) were collected for each patient. Medical utilization was categorized as: hospitalizations, office/out-patient visits, ER visits, lab services, nursing home visits, and surgery center visits based on the site of service, type of provider, or procedure code on the claim. The total cost of all pharmacy claims during the follow-up period was also collected. For each type of utilization, cost was calculated as that incurred by the health plan and the patient via the amount paid by the health plan and any copays or deductibles incurred by the patient. The use of selected cardiac medications was also examined. The receipt of at least 1 prescription in each of the following classes during the baseline and follow-up periods was recorded: non-statin lipid lowering therapy, HMG CoA-reductase inhibitor (statin), beta-blocker, angiotensin-converting enzyme inhibitor (ACE-I), angiotensin II receptor inhibitor (AT-II blocker), calcium channel blocker, and clopidogrel. Aspirin utilization was not examined since it is non-prescription and therefore not typically recorded in pharmacy claims data. The length of clopidogrel therapy was defined as the number of days between the first prescription in the follow-up period through the run-out date of the last prescription. Patients were designated as newly initiating therapy if they did not have a prescription claim in the same class during the 6-month baseline period, otherwise, they were designated as continuing therapy. The total number of patients receiving therapy during the study is the number of “new” plus “continuing” users. Analyses. All analyses presented are descriptive in nature. Results are expressed on a per patient-month of follow-up basis. Due to the high proportion of patients with the full 12 months of follow-up, overall percents and means are also presented where appropriate. Results Study population. Of 13,731 patients who met the inclusion criteria for ACS, 6,929 patients had a revascularization procedure during the first year after the index event and are reported here (Table 1). Seventy-three percent were male and 77.0% were between the ages of 45 and 64. Nearly 38% of patients had an index diagnosis of UA, 48.9% had a diagnosis of MI, and 13.5% had both UA and MI diagnoses. Nearly all patients were admitted to the hospital during their index event (96.2%). A large portion of patients had multiple claims diagnoses for cardiac-related comorbidities. The average number of comorbidities as categorized by the Washington Clusters Method was 13.3. Overall, 87% had a diagnosis of lipid disorder and 77% had hypertension. In addition, 31% had diabetes and 38% had cardiac arrhythmias. Importantly, these rates reflect medical claims at any time during the baseline and follow-up period, and the high rates of diagnoses may reflect the increased contact with the healthcare system due to the ACS event. The majority of patients had a stent as their revascularization procedure (67.5%), while 4.7% had PTCA alone. The remaining 27.8% of revascularizations were CABG. Sixty-nine percent of all procedures occurred during the initial ACS hospitalization, typically on the first day. Health services utilization. Total healthcare costs (health plan plus patient) for all 6,929 patients during the study period were $211 million, or $30,402 per patient ($3,010 per patient-month) (Table 2, Figure 1). Of these, $127 million were incurred by the 5,002 PCI patients ($25,411 per patient), and $84 million are attributable to the 1,927 CABG patients ($43,355 per patient). The largest component of cost was hospitalizations, which accounted for 77% ($161.7 million) of total expenditures. The mean number of in-patient days per patient during the follow-up was 7.4 (0.73 per patient per month). A total of 2,314 patients had more than 1 hospitalization during the follow-up period, and the average number of in-patient stays per patient was 1.51. The second largest component of cost was office/out-patient visits, accounting for $21.2 million. On average, there were 1.86 office/out-patient visits per patient per month of follow-up. The most frequent type of provider encounter was for cardiologist visits, which accounted for 28% of all office/out-patient visits. Pharmacy expenditures were the third largest component of cost at $12.8 million. While ER visits accounted for only 3% of the total cost, 2,294 patients had more than 1 ER visit during the follow-up period. Over 59% of the total cost was incurred during the index event ($125 million). Patients with an index hospitalization had a mean (SD) index cost of $18,802 (19,561). Patients with an index event of an ER visit had a mean (SD) index cost of $698 (1,542). Overall, patients with an index diagnosis of UA appeared to have a lower mean index cost than those with an index diagnosis of MI, or both UA and MI ($15,221, $19,373, and $21,641, respectively). Medication use in the follow-up period. A majority of patients received a cholesterol-lowering medication during follow-up (75.5% received a statin and 13.6% received a non-statin cholesterol medication). Overall, 75.7% received a prescription for a beta-blocker, 46.4% received an ACE-I, 7.0% received an AT-2 blocker, and 22.7% received a calcium channel blocker (Figure 2). A total of 63.5% of all revascularization patients received at least 1 prescription for clopidogrel during the follow-up period. Among PCI patients, the value was 84.8%. The most common pattern of use was “newly initiating therapy,” at 97% (Figure 2). The mean length of clopidogrel therapy was 83.5 days (median 30 days), while the average length of follow-up for clopidogrel users was 310 days. Discussion This analysis provides new insight into ACS patients who undergo coronary revascularization in the MCO setting. We describe the therapeutic strategies, in-patient and out-patient resource utilization, and patterns of drug use for the first year of follow-up. Sixty-nine percent of patients underwent revascularization during the initial ACS presentation, and PCI accounted for the majority of the revascularization procedures. Myocardial infarction was the most frequent index diagnosis. One of the clearest differences in our study population compared to other large datasets is patient age. Only 11.3% of our sample were > 65 years old, but the National Hospital Discharge Survey data indicate that over half of all coronary revascularizations were performed in patients > 65 years old.¹⁷ While the inclusion criteria for our study and that from Bhatt et al. differ somewhat (we included all types of MI, whereas they included only NSTEMI), the average patient in our study was 8 years younger than those in the Bhatt et al. study (55 versus 63 years of age).¹⁴ Slightly more patients underwent CABG in our study (27.8 versus 21%), and our patients also had higher rates of diagnosed hypertension and hyperlipidemia. The rate of diagnosed diabetes was similar (30.5% versus 27.7%).¹⁴ Total healthcare costs in our study were $211 million. This was distributed as 6% pharmacy and 94% medical. Medical is further distributed as: 77% in-patient, 10% out-patient, and 4% other medical. The majority of medical cost (59%) was also attributable to the index hospitalization. Contrasting these 6,929 patients who had a revascularization procedure to the initial group of 13,731 patients with ACS, the revascularization group accounts for roughly 50% of patients (6,929/13,731) and 68% of dollars ($211 million/$309 million).⁵ Although the timeframe of the data are more remote, one study that may be useful as a cost benchmark is the BARI (Bypass Angioplasty Revascularization Investigation) trial.¹⁸ While our study differs from BARI (not all of our patients underwent revascularization during an index event), BARI estimated mean first year costs in the range of $34,000–40,000 (in 1995 dollars) for PTCA or CABG, respectively. In our study, mean first year costs per patient were $24,411 and 43,455 (in 2001 dollars) for PTCA or CABG, respectively. These differences can be explained by multiple factors, such as the dates the studies were performed, protocol-driven care, payer type, etc. However, a consistent finding among cost studies is the general link of disease severity, complications and comorbidities to cost.¹⁹ An example in our study was patients with UA had somewhat lower index episode costs versus patients with MI, or both UA and MI. These revascularization patients were also frequent users of outpatient provider services, with a mean of 1.86 office/out-patient visits per patient per month. Cardiologist visits were the most frequent single provider-type encounter. It is also important to note the relatively high proportion of patients with more than 1 in-patient stay (33%) or an ER visit (33%) during the study period. Further research on these issues is warranted to better understand the reasons behind these re-admissions, costs of recurrent events, and possible methods to decrease these episodes of care. Total mean pharmacy costs were $1,841 ($182 per patient per month) and the majority of patients received a statin, beta-blocker, and clopidogrel in their cardiovascular drug regimen (Figure 2). There were several interesting differences in baseline drug therapy in our study compared to earlier trials in PCI such as the Clopidogrel for the Reduction of Events During Observation (CREDO) trial.²⁰ CREDO was a PCI study that examined the effects of clopidogrel on atherothrombotic events. In the CREDO baseline period, approximately 55% of patients were on statin therapy, and approximately 33% were on an ACE-I. In our study, the number of patients was much lower, only 27% and 18%, respectively, which suggests differences in the type of patient encountered in randomized clinical trials compared to the usual practice setting. While retrospective studies such as ours are unable to examine all possible contraindications to drug therapy, it appears that statin, beta-blocker, and clopidogrel therapy achieved reasonable utilization levels in these revascularization patients. Figure 2 indicates a large number of patients began therapy after the index event, which suggests that cardiology specialty care may have influenced the better adherence to treatment guidelines. However, as seen in Table 1 and Figure 2, there were a high proportion of patients who may have been unrecognized prior to the index event since they were not on medication for known risk factors. For the antiplatelet clopidogrel, the relatively short duration of treatment is consistent with use in the stent patient. However, recent treatment guidelines for PCI patients suggest a longer duration of clopidogrel therapy.¹¹ Our data collection period precede these guidelines and should be re-examined at later dates. Limitations. As with any claims database study, the limitation of miscoding and missing data is possible. In-patient pharmacy utilization is not recorded in these data. It is possible that patients received medications of interest prior to discharge that was not taken into account. While a large number of patients across the United States are represented, the results are obtained from a fee-for-service model MCO and may not be generalizable to a capitated health plan or Preferred Provider Organization (PPO) model. Conclusion Early revascularization appears to be a common therapeutic strategy in these ACS patients. Total healthcare costs (health plan plus patient) were dominated by medical costs, most of which occurred during the index hospitalization. Three medications (statin, beta-blocker, and clopidogrel) were prescribed to a reasonable number of patients in accordance with practice guidelines. This could be a reflection of more intensive cardiologist interactions since all patients underwent revascularization.

1. Kong DF, Blazing MA, O’Connor CM, et al. The health care burden of unstable angina. Cardiology Clinics 1999;17:247–261. 2. Eisenstein EL, Shaw LK, Anstrom KJ, et al. Assessing the clinical and economic burden of coronary artery disease: 1986–1998. Medical Care 2001;39:824–835. 3. Russell MW, Huse DM, Drowns S, et al. Direct medical costs of coronary artery disease in the United States. Am J Cardiol 1998;81:1110–1115. 4. Bloom BS, Tibi-Lévy Y, Harari A, Fendrick AM. Direct medical care costs of unstable angina pectoris in a defined population. J Man Care Pharm 1999;5:39–44. 5. Etemad LR, McCollam P. Total cost of acute coronary syndromes patients in a managed care population during the one-year following initial presentation. J Man Care Pharm 2004;10:202. 6. Wallentin L, Lagerqvist B, Husted S, et al. Outcome at 1 year after an invasive compared with a non-invasive strategy in unstable coronary artery disease: The FRISC II invasive randomized trial. Lancet 2000;356:9–16. 7. 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;344:1879–1887. 8. 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 Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on the Management of Patients With Unstable Angina). J Am Coll Cardiol 2002;40:1366–1374. 9. Anderson HV, Shaw RE, Brindis RG, et al. A contemporary overview of percutaneous coronary interventions The American College of Cardiology - National Cardiovascular Data Registry (ACC–NCDR). J Am Coll Cardiol 2002;39:1096–1103. 10. Eagle KA, Guyton RA, Davidoff R, et al. ACC/AHA 2004 guideline update for coronary artery bypass graft surgery: A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Update the 1999 Guidelines for Coronary Artery Bypass Graft Surgery). American College of Cardiology Web Site. http://www.acc.org/clinical/guidelines/cabg/cabg.pdf accessed 12/1/2004. 11. Levine GN, Kern MJ, Berger PB, et al. for the American Heart Association Diagnostic and Interventional Catheterization Committee and Council on Clinical Cardiology. Management of patients undergoing percutaneous coronary revascularization. Ann Intern Med 2003;139:123–136. 12. Stafford RS, Radley DC. The underutilization of cardiac medications of proven benefit, 1990 to 2002. J Am Coll Cardiol 2003;41:56–61. 13. Miller RR, Li Y-F, Sun H, et al. Underuse of cardioprotective medications in patients prior to acute myocardial infarction. Am J Cardiol 2003;92:209–211. 14. Bhatt DL, Roe MT, Peterson ED, et al. Utilization of early invasive management strategies for high-risk patients with non–ST-segment elevation acute coronary syndromes: results from the CRUSADE quality improvement initiative. JAMA 2004;292:2096–2104. 15. Grech ED, Ramsdale DR. ABC of interventional cardiology. Acute coronary syndrome: unstable angina and non-ST segment elevation myocardial infarction. Br Med J 2003;326:1259–1261. 16. Schneeweiss R, Cherkin DC, Hart LG, et al. Diagnosis clusters adapted for ICD-9-CM and CDHPPC-2. J Fam Pract 1986;22:69–72. 17. Kozak LJ, Hall MJ, Owings MF. National Hospital Discharge Survey: 2000 annual summary with detailed diagnosis and procedure data. National Center for Health Statistics. Vital Health Stat 2002;153:1–194. 18. Hlatky MA, Rogers WJ, Johnstone I, et al. Medical care costs and quality of life after randomization to coronary angioplasty or coronary bypass surgery. N Engl J Med 1997;336:92–99. 19. Hlatky MA, Boothroyd D, Brooks MM, et al. Clinical correlates of the initial and long-term cost of coronary bypass surgery and coronary angioplasty. Am Heart J 1999;138:376–383. 20. Steinhubl SR, Berger PB, Mann JT, et al. Early and sustained dual oral antiplatelet therapy following percutaneous coronary intervention. JAMA 2002;288:2411–2420.