Clopidogrel Use and Hospital Quality in Medically Managed Patients With Non ST-Segment Elevation Myocardial Infarction

Transcription

1 Clopidogrel Use and Hospital Quality in Medically Managed Patients With Non ST-Segment Elevation Myocardial Infarction Thomas M. Maddox, MD, MSc; P. Michael Ho, MD, PhD; Thomas T. Tsai, MD, MSc; Tracy Y. Wang, MD, MHS, MSc; Shuang Li, MS; S. Andrew Peng, MS; Stephen D. Wiviott, MD; Fredrick A. Masoudi, MD, MSPH; John S. Rumsfeld, MD, PhD Background Clopidogrel prescription is a class I guideline recommendation for medically managed patients with non ST-segment elevation myocardial infarction (NSTEMI). However, clopidogrel has historically been underused in this population. We evaluated contemporary rates of its use and evaluated associated factors, with a particular focus on hospital quality of myocardial infarction (MI) care. Methods and Results We examined clopidogrel prescription rates among patients with NSTEMI discharged from 382 US hospitals between October 2009 and March Associations between clopidogrel prescription and various patient and hospital factors, including hospital quality of MI care, were determined with regression modeling. Of the sample, 54.9% of eligible patients with NSTEMI received clopidogrel prescription at hospital discharge. Variation in rate by hospital was large, ranging from 22% to 97%. A variety of patient and hospital factors were associated with clopidogrel prescription. Hospital quality of MI care demonstrated modest association with clopidogrel prescription (odds ratio, 0.68; 95% CI, ) between the lowest and highest hospital quality quartile) and accounted for 5.7% of the variation in prescription rates. Conclusions Clopidogrel prescription is significantly underused in the medically managed NSTEMI population and demonstrates wide variability by hospital. Although hospital quality of MI care is associated with its use, the findings suggest that it only has a modest effect. Therefore, efforts to improve clopidogrel use likely will require measures beyond improving the overall hospital quality of MI care. (Circ Cardiovasc Qual Outcomes. 2012;5:00-00.) Key Words: myocardial infarction prevention coronary artery disease In the Clopidogrel in Unstable Angina to Prevent Recurrent Events (CURE) trial, antiplatelet therapy with clopidogrel and aspirin was associated with an improved 1-year primary outcome of cardiovascular death, nonfatal myocardial infarction (MI), or stroke among patients with non ST-segment elevation myocardial infarction (NSTEMI) compared with aspirin alone. 1 This treatment benefit was similar among revascularized patients and those managed medically. Accordingly, clopidogrel prescription for all patients with NSTEMI at hospital discharge is a class I recommendation in the most recent American College of Cardiology/American Heart Association (ACC/AHA) guidelines. 2 In addition, clopidogrel prescription at discharge for patients with NSTEMI was added as a test measure in the recently updated ACC/AHA performance measures for acute MI (AMI) care. 3 Despite these measures, clopidogrel use among the medically managed NSTEMI population is suboptimal, with more than one half of these patients failing to receive its prescription at hospital discharge. 4 This gap in care represents a target for quality improvement. Understanding factors associated with this gap can inform efforts to improve current prescription practices of clopidogrel. In addition, these insights may inform optimal dissemination of the newer antiplatelet therapies, such as prasugrel and ticagrelor, among medically managed patients with acute coronary syndrome. Accordingly, we assessed clopidogrel prescription rates and their variation by hospital in a large national clinical registry of patients with NSTEMI. We then determined patient and hospital factors associated with clopidogrel pre- Received August 30, 2011; accepted April 16, From the VA Eastern Colorado Health Care System, Denver, CO (T.M.M., P.M.H., T.T.T., J.S.R.); University of Colorado Denver, Aurora, CO (T.M.M., P.M.H., T.T.T., F.A.M.); Duke Clinical Research Institute, Durham, NC (T.Y.W., S.L., S.A.P.); and TIMI Study Group, Cardiovascular Division, Brigham and Women s Hospital, Harvard Medicine School, Boston, MA (S.D.W.). Drs Masoudi and Rumsfeld, Associate Editors of Circulation: Quality and Outcomes, are coauthors of the article but had no role in the evaluation of the article or the decision about its acceptance. This article was handled independently by Guest Editor David J. Moliterno, MD. Correspondence to Thomas M. Maddox, MD, VA Eastern Colorado Health Care System; Cardiology (111B), 1055 Clermont St; Denver, CO thomas.maddox@va.gov 2012 American Heart Association, Inc. Circ Cardiovasc Qual Outcomes is available at DOI: /CIRCOUTCOMES Downloaded from 1 by guest on May 12, 2016

2 2 Circ Cardiovasc Qual Outcomes July 2012 scription, with a particular focus on overall hospital quality of AMI care. By providing a better understanding of these relationships, we can inform quality improvement initiatives to improve clopidogrel prescription rates. WHAT IS KNOWN Clopidogrel, in conjunction with aspirin, significantly reduces the rate of recurrent cardiac events and mortality among medically managed patients with myocardial infarction. However, clopidogrel has been consistently underused in this population. WHAT THE STUDY ADDS In a contemporary, national sample of cardiac patients, only 54.9% of eligible patients with non ST-segment elevation myocardial infarction were discharged from the hospital with a clopidogrel prescription. There was significant variation in clopidogrel discharge rates by hospital, ranging from 22% to 97%. Hospital performance on quality-of-care measures for patients with myocardial infarction was modestly associated with clopidogrel prescription rates and accounted for only 5.7% of the variation in these rates. Methods Study Population We evaluated patients in the National Cardiovascular Data Registry s (NCDR s) Acute Coronary Treatment and Intervention Outcomes Network Registry-Get With The Guidelines (ACTION Registry-GWTG). The registry records patient and treatment characteristics of individuals with MI hospitalized in a variety of practice settings throughout the United States. 5 Trained abstractors at each site collect patient, presentation, treatment, and hospital characteristics from the medical record in a standardized fashion. Once the registry receives the data, the data undergo extensive quality review with required completeness standards, logic checks, and onsite data accuracy audits. Only data meeting these data quality standards are included in the research registry. Analyses are conducted at the ACTION Registry-GWTG analytic center located at the Duke Clinical Research Institute (Durham, NC). Full details of the registry are located at Study Cohort In this study, we evaluated patients with NSTEMI who were medically managed without percutaneous or surgical revascularization. We focused on patients with NSTEMI both because of the recent addition of the clopidogrel test performance measure for medically treated patients with MI and because patients with ST-segment elevation MI have high rates of percutaneous coronary intervention treatment and clopidogrel use. Exclusion criteria included patients who had a documented contraindication to clopidogrel, died during hospitalization, were transferred to a different facility, were discharged on comfort measures or to hospice, left against medical advice, had missing discharge clopidogrel information, or were discharged on ticlopidine or prasugrel. Figure 1. Study cohort creation. NSTEMI indicates non ST-segment elevation myocardial infarction. Variables The primary dependent (outcome) variable of interest was clopidogrel prescription at hospital discharge as recorded from the ACTION Registry-GWTG case report form containing the abstracted patient chart list of discharge medications. Independent variables of interest were selected on the basis of prior studies and clinical judgment and included demographic, clinical history, presentation and treatment, and hospital factors. Demographic factors included age (per 10-year increase), sex, race (white versus other), and insurance type (health maintenance organization versus government versus self-pay or none). Clinical history factors included prior MI, prior percutaneous coronary intervention, prior coronary artery bypass graft, peripheral arterial disease (PAD), cerebrovascular disease, diabetes mellitus, hypertension, hyperlipidemia, smoking, body mass index per 5-kg/m 2 increase, renal failure requiring dialysis, and medications before admission (aspirin, -blockers, statins, angiotensin-converting enzyme inhibitors or angiotensin receptor blockers, clopidogrel, and warfarin). Presentation and treatment factors included admission heart rate (per 100-beats/min increase), admission blood pressure (per 10-mm Hg increase), admission renal function (categorized as creatinine clearance 60 ml/min, ml/min, and 30 ml/min), admission hematocrit level (categorized as 30% versus 30%), ejection fraction (EF) (categorized as 40% versus 40%), troponin level (categorized as 3 times the upper limit of normal of the assay measurement versus 3 times the upper limit of normal), receipt of diagnostic catheterization during hospitalization, and in-hospital major and minor bleeding (see later for definitions). Hospital factors included number of beds (per 100-bed increase); presence of diagnostic catheterization, percutaneous coronary intervention, and cardiothoracic surgical service facilities; teaching hospital (versus community hospital), and overall hospital quality of MI care (see later for definition). In-hospital major bleeding was defined as an absolute hematocrit drop in level of 12%, an intracranial hemorrhage, a retroperitoneal bleed, an admission hematocrit level of 28% plus a red blood cell transfusion, or an admission hematocrit level of 28% plus a red blood cell transfusion and a witnessed bleeding event. Minor bleeding was defined as a suspected bleeding event that did not meet a major bleeding criterion. Hospital quality of MI care was defined as the percentage of fulfilled quality opportunities for each hospital, as suggested by the Joint Commission. 6 Quality opportunities were based on the 2008 ACC/AHA AMI performance measures, which did not include clopidogrel prescription at discharge. 3 They included aspirin administration within 24 hours of arrival, aspirin prescription at discharge, -blocker prescription at discharge, statin prescription at discharge, evaluation of EF before discharge, angiotensin-converting enzyme

3 Maddox et al Clopidogrel and Hospital Quality 3 Table 1. Baseline Characteristics for Patients Discharged With or Without Clopidogrel Prescription Clopidogrel at Discharge (n ) No Clopidogrel at Discharge (n ) P Value Patient factors Age, y 71 (60 82) 72 (60 83) Male sex 6730 (53.9) 5332 (49.9) White race 9987 (80.0) 8318 (77.8) Insurance HMO 6780 (54.3) 5655 (52.9) Government 4929 (39.5) 4200 (39.3) Self, none, missing 781 (6.3) 841 (7.9) Clinical history factors Prior MI 5185 (41.5) 2845 (26.6) Prior PCI 4444 (35.6) 1658 (15.5) Prior CABG 3846 (30.8) 2044 (19.1) PAD 2309 (18.5) 1325 (12.4) CVD 1770 (14.2) 1341 (12.5) Diabetes 5406 (43.3) 3848 (36.0) HTN (85.3) 8617 (80.6) Dyslipidemia 8920 (71.4) 6306 (59.0) Current/recent smoker 2996 (24.0) 2394 (22.4) BMI, kg/m ( ) 27.4 ( ) Renal failure requiring dialysis 542 (4.3) 434 (4.1) 0.28 Home aspirin 7424 (59.4) 4931 (46.1) Home -blocker 7235 (57.9) 4976 (46.5) Home statin 7103 (56.9) 4596 (43.0) Home ACEI 4722 (37.8) 3522 (32.9) Home ARB 1724 (13.8) 1309 (12.2) Home clopidogrel 5031 (40.3) 325 (3.0) Home warfarin 613 (4.9) 1479 (13.8) Presentation and treatment factors Admission HR, beats/min 85 (72 100) 89 (75 107) Admission systolic BP, mm Hg 147 ( ) 143 ( ) Admission renal function, Cockcroft-Gault (51.0) 5074 (49.4) 30, (34.7) 3597 (35.1) (13.5) 1524 (14.9) Admission Hgb 9 g/dl 368 (3.0) 541 (5.1) Ejection fraction 40% 2722 (24.0) 2415 (25.0) 0.11 Troponin 3 ULN (85.6) 8757 (81.9) Diagnostic catheterization during hospitalization 7537 (60.3) 5966 (55.8) In-hospital major bleeding 910 (7.3) 1028 (9.6) Suspected bleeding event 272 (2.2) 343 (3.2) Hospital factors No. hospital beds 395 ( ) 384 ( ) Hospital facilities 0.03 No catheterization laboratory 182 (1.5) 197 (1.8) Diagnostic catheterization only 284 (2.3) 278 (2.6) PCI without cardiothoracic surgery 928 (7.4) 758 (7.1) PCI and cardiothoracic surgery (88.8) 9463 (88.5) Teaching hospital 3154 (25.3) 2528 (23.6) Delivery of quality measures Quartile 1 ( 96%) 3757 (30.1) 2671 (25.0) Quartile 2 (92% 96%) 3465 (27.7) 2871 (26.8) Quartile 3 (86% 91%) 3078 (24.6) 2972 (27.8) Quartile 4 (53% 85%) 2105 (16.9) 2117 (19.8) Data are presented as median (25th 75th percentile) or n (%). HMO indicates health maintenance organization; MI, myocardial infarction; PCI, percutaneous coronary intervention; CABG, coronary artery bypass graft; PAD, peripheral arterial disease; CVD, cerebrovascular disease; HTN, hypertension; BMI, body mass index; ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; HR, heart rate; BP, blood pressure; Hgb, hemoglobin; ULN, upper limit of normal.

4 4 Circ Cardiovasc Qual Outcomes July 2012 Figure 2. Rates of clopidogrel prescription by hospital quality quartile of quality opportunities fulfilled (quartile 1, 93%; quartile 2, 87% 93%; quartile 3, 80% 86%; quartile 4, 80%). inhibitor/angiotensin receptor blocker prescription for patients with an EF 40%, smoking cessation counseling provided for current smokers, and cardiac rehabilitation referral at discharge. Hospital quality was then calculated by dividing the number of these quality measures that occurred by the number of quality opportunities that the hospital had to provide quality MI care. After these calculations were performed, hospitals were categorized by quartile of quality performance (quartile 1, highest quality; quartile 4, lowest quality). Statistical Analysis Rates of clopidogrel prescription for the overall cohort and by hospital were calculated. The cohort was then stratified by clopidogrel prescription at hospital discharge, and we compared demographic, clinical history, presentation, treatment, and hospital factors. Continuous variables are presented as medians and compared using either t tests or Wilcoxon rank sum tests, depending on the normality of the data distribution. Categorical variables were presented as percentages and compared with 2 tests. For the hospital rates, only those hospitals treating 30 patients were included to ensure stable estimates. Observed rates of clopidogrel prescription by hospital were calculated, and exact 95% binomial CIs were generated. To determine associations of the various patient and hospital factors with the likelihood of clopidogrel prescription, we specified multivariable logistic regression models using generalized estimating equations to control for clustering of patients by hospital. To properly specify the models, patients who were treated at hospitals that contributed 30 patients to the ACTION Registry-GWTG or who were treated at hospitals with missing hospital data were excluded from analysis. A C statistic for the full model was calculated to determine its discriminatory ability. To understand the relationship between hospital quality of MI care and hospital rates of clopidogrel, we generated a scatterplot of these 2 measures and calculated a Pearson correlation coefficient to determine the correlation between the 2 variables. We also calculated the R 2 variable to determine the amount of clopidogrel prescription variation that could be explained by variation in hospital quality of MI care. Because clopidogrel use before MI admission was strongly associated with clopidogrel prescription at discharge, we conducted a secondary analysis removing those patients from the analysis to explore factors associated with clopidogrel prescription among those without prior use. Similarly, because warfarin use before admission was strongly associated with a lack of clopidogrel prescription at discharge (likely because of the uncertainty and risks surrounding triple therapy), we also excluded those patients from the secondary analysis. Thus, the cohort for the secondary analysis comprised patients without clopidogrel or warfarin use on admission. Finally, because providers may restrict their use of clopidogrel only to those patients who would have qualified for inclusion in the CURE trial, we conducted a secondary analysis excluding those patients from the present primary study cohort who would have been excluded from the CURE trial. 1 Our exclusion criteria were patient contraindication to antiplatelets or anticoagulants, patients with a suspected bleeding event or hemorrhagic cerebrovascular accident, patients with heart failure or shock on presentation or during hospitalization, and patients on home or discharge warfarin. Similar to our primary analysis, rates of clopidogrel prescription and predictors of its use in this cohort were calculated. For all analyses, P 0.05 was considered significant. All analyses were performed using SAS version 10 (SAS Institute) statistical software. Results Rates of Clopidogrel Prescription at Discharge Between October 2009 and March 2011, medically managed patients with NSTEMI at 387 hospitals were rec- Figure 3. Rates of clopidogrel prescription by hospital.

5 Maddox et al Clopidogrel and Hospital Quality 5 orded in the ACTION Registry-GWTG. After applying our exclusion criteria, at 382 hospitals remained in the study cohort (Figure 1). In the study cohort, (53.9%) patients received a clopidogrel prescription at hospital discharge. Table 1 lists the patient and hospital factors by receipt of clopidogrel prescription at discharge. Among demographic factors, younger age, male sex, white race, and the presence of health maintenance organization or government insurance were associated with higher rates of clopidogrel prescription. Among clinical history factors, higher rates of prior cardiovascular disease, cardiac risk factors, and cardiac medication use before hospital admission for MI were associated with higher rates of clopidogrel prescription. The sole exception was warfarin use before admission, where higher rates of its use were associated with lower rates of clopidogrel prescription. Among treatment factors, higher rates of diagnostic catheterization and lower rates of in-hospital bleeding were associated with higher rates of prescription. Among hospital factors, hospitals with catheterization laboratories or with higher quality of MI care were associated with higher rates of clopidogrel prescription. Figure 2 illustrates the differences in clopidogrel prescription rates as a function of hospital quality of MI care. To calculate hospital rates of clopidogrel prescription, we excluded 141 hospitals with 30 patients (total number of patients excluded, 2 101). Among this cohort of 241 hospitals, rates of clopidogrel prescription varied widely, ranging from 22% to 97% (Figure 3). Table 2. Factors Associated With Clopidogrel Prescription at Discharge 2 OR (95% CI) Demographic factors White race ( ) Clinical history factors Dyslipidemia ( ) Diabetes ( ) Current/recent smoker ( ) Prior MI ( ) Prior PCI ( ) Prior CABG ( ) PAD ( ) Home clopidogrel ( ) Home warfarin ( ) Presentation and treatment factors Admission systolic BP (per ( ) 10-mm Hg increase) Diagnostic catheterization ( ) Troponin 3 ULN ( ) Admission HR (per 10-beats/min ( ) increase) Admission Hgb 9 g/dl ( ) Ejection fraction 40% ( ) In-hospital major bleeding ( ) Hospital factors Delivery of quality measures (relative to quartile 1 96% ) Quartile 2 (92% 96%) 0.83 ( ) Quartile 3 (86% 91%) 0.7 ( ) Quartile 4 (53% 85%) 0.68 ( ) OR indicates odds ratio; MI, myocardial infarction; PCI, percutaneous coronary intervention; CABG, coronary artery bypass graft; PAD, peripheral arterial disease; BP, blood pressure; ULN, upper limit of normal; HR, heart rate; Hgb, hemoglobin. Associations Between Patient and Hospital Factors and Clopidogrel Prescription After exclusion of 150 patients with missing hospital data or treated at hospitals with 30 ACTION Registry-GWTG patients, patients treated at 350 hospitals were included in the primary multivariable model. A variety of factors were associated with clopidogrel prescription at discharge (Table 2). The only demographic factor associated with higher rates of clopidogrel prescription was white race. Among clinical history factors, dyslipidemia, diabetes, current or recent smoking, prior coronary artery disease (indicated by prior MI, percutaneous coronary intervention, or coronary artery bypass graft), PAD, and clopidogrel use before admission were associated with higher rates of clopidogrel prescription, whereas warfarin use before admission was associated with lower rates. Among presentation and treatment factors, higher admission blood pressure, receipt of diagnostic catheterization, and higher troponin values were associated with higher rates of clopidogrel prescription, whereas higher admission heart rate, admission hemoglobin level of 9 g/dl, EF of 40%, and in-hospital major bleeding were associated with lower rates of clopidogrel prescription. Among hospital factors, quality of MI care significantly affected clopidogrel prescription rates. Patients treated at a hospital in the lowest quartile of quality performance, relative to hospitals in the highest quartile, had 32% lower odds to receive clopidogrel prescription (odds ratio, 0.68; 95% CI, ) (Table 2). The full model demonstrated good discrimination between those with and those without clopidogrel prescription (C statistic, 0.78). There was a modest positive correlation between hospital quality of MI care and hospital-level clopidogrel prescription rates (Pearson correlation coefficient, 0.24; P ) (Figure 4). The R 2 value between the 2 factors was 0.057, meaning that 5.7% of the variation in the clopidogrel prescription rates was explained by variation in hospital quality of MI care. Results of our secondary analysis, which excluded patients taking clopidogrel or warfarin before hospital admission (n 5356 and n 1789, respectively) are shown in Table 3. In contrast to the primary analysis, significant associations with clopidogrel prescription and older age, male sex, and larger hospitals were seen, and associations with diabetes and PAD were no longer present. Results of the secondary analysis replicating the CURE trial are shown in Table 4. Among the subset of patients who closely resembled trial participants, rates of clopidogrel prescription at discharge increased to 58.8%. As with the other secondary analysis, significant

6 6 Circ Cardiovasc Qual Outcomes July 2012 Figure 4. Scatterplot of hospital quality of myocardial infarction care and clopidogrel prescription at discharge rates. associations with clopidogrel prescription and older age, male sex, and larger hospitals were seen. In addition, associations with diabetes, smoking, PAD, and an EF of 40% were no longer present. Discussion In this large national cohort of medically managed patients with NSTEMI, only 56% of eligible patients received clopidogrel prescription at discharge, with wide variability in rates of prescription by hospital. Patient-level factors such as male sex, white race, cardiac risk factors, receipt of diagnostic catheterization, treatment by a cardiologist, and higher troponin levels were associated with higher prescription rates. Other patient-level factors such as anemia, in-hospital bleeding events, and warfarin use before admission were associated with lower rates of clopidogrel prescription. Among hospital-level factors, higher quality of MI care was associated with higher rates of clopidogrel prescription, although this relationship was only modest. Furthermore, the variation in hospital quality of MI care only explained 12% of the variation in clopidogrel prescription rates. Among patients who resembled the CURE trial population, the rates of clopidogrel were higher (58.8%), but still suboptimal. These findings demonstrate a significant gap in clopidogrel prescription to medically managed patients with NSTEMI and a wide variability in prescription rates by hospital and suggest that other measures beyond focusing on overall hospital quality of MI care are needed to improve clopidogrel prescription rates. Based on the CURE trial, clopidogrel prescription at discharge is a class I guideline and is currently a test performance measure for AMI care. 1 3 However, the uptake of clopidogrel in routine clinical practice for patients with NSTEMI has been suboptimal. Tricoci and colleagues 7 examined patients with NSTEMI eligible for clopidogrel prescription from 461 hospitals between January 2002 and December 2003 in the CRUSADE (Can Rapid Risk Stratification of Unstable Angina Patients Suppress Adverse Outcomes with Early Implementation of the ACC/AHA Guidelines) Registry, which was a precursor to the NCDR ACTION Registry-GWTG. They demonstrated that only 42.8% of medically managed patients received clopidogrel prescription at discharge. Amsterdam and colleagues, 4 who also used CRUSADE data, showed that prescription rates among the medically managed NSTEMI population increased to 53.9% by The present study finds that there has been little change in these rates for medically managed patients with NSTEMI, despite the guideline and performance measure emphasis on this metric. Why has clopidogrel use remained practically unchanged for the medically managed NSTEMI population? The present study suggests that multiple factors are at play, including several that may be targets for quality improvement. Clopidogrel prescription, like any medication or therapy, requires evaluation of its benefits and risks to the individual patient. Accordingly, factors that confer higher cardiac risk, such as dyslipidemia, diabetes, prior revascularization, PAD, and higher troponin levels may influence providers to prescribe clopidogrel to minimize this risk. Similarly, factors that confer a higher risk of bleeding, such as hemodialysis, anemia, concurrent warfarin use, or major bleeding during hospitalization, may prohibitively increase the risk of clopidogrel therapy, thus influencing providers to avoid clopidogrel prescription. The association between receipt of diagnostic catheterization and higher rates of clopidogrel has been demonstrated previously 4 and may represent more cardiology involvement during hospitalization, with a subsequent focus on optimal cardiac therapies, and may reflect selection of patients believed to have a more optimal risk/treatment benefit for clopidogrel therapy. Similarly, the relationship between cardiologist care and higher rates of evidence-based cardiac therapies, including clopidogrel, is a well-studied association and likely demonstrates cardiologists greater familiarity and comfort with cardiac therapies as well as some selection bias in which patients receive cardiology care. 7,8 Finally, disparities of treatment by sex and race persist, a troubling trend that

7 Maddox et al Clopidogrel and Hospital Quality 7 Table 3. Factors Associated With Clopidogrel Prescription Among Patients Without Clopidogrel or Warfarin Use Before Admission 2 OR (95% CI) Demographic factors Age (per 10-y increase) ( ) Male sex ( ) White race ( ) Clinical history factors Dyslipidemia ( ) Current or recent smoker ( ) Prior MI ( ) Prior PCI ( ) Prior CABG ( ) Presentation and treatment factors Admission systolic BP (per ( ) 10-mm Hg increase) Diagnostic catheterization ( ) Troponin 3 ULN ( ) Admission HR (per 10-beats/min ( ) increase) Admission Hgb 9 g/dl ( ) Ejection fraction 40% ( ) In-hospital major bleeding ( ) Hospital factors No. of hospital beds (per 100) ( ) Delivery of quality measures (relative to quartile 1 96% ) Quartile 2 (92% 96%) 0.79 ( ) Quartile 3 (86% 91%) 0.72 ( ) Quartile 4 (53% 85%) 0.65 ( ) OR indicates odds ratio; MI, myocardial infarction; PCI, percutaneous coronary intervention; CABG, coronary artery bypass graft; BP, blood pressure; ULN, upper limit of normal; HR, heart rate; Hgb, hemoglobin. Table 4. Factors Associated With Clopidogrel Prescription Among Patients Without CURE Exclusion Criteria 2 OR (95% CI) Demographic factors Age (per 10-y increase) ( ) Male sex ( ) White race ( ) Clinical history factors Dyslipidemia ( ) Prior MI ( ) Prior PCI ( ) Prior CABG ( ) Home clopidogrel ( ) Presentation and treatment factors Admission systolic BP (per ( ) 10-mm Hg increase) Diagnostic catheterization ( ) Troponin 3 ULN ( ) Admission HR (per 10-beats/min ( ) increase) Admission Hgb 9 g/dl ( ) In-hospital major bleeding ( ) Hospital factors No. of hospital beds (per 100) ( ) Delivery of quality measures (relative to quartile 1 96% ) Quartile 2 (92% 96%) 0.8 ( ) Quartile 3 (86% 91%) 0.71 ( ) Quartile 4 (53% 85%) 0.68 ( ) CURE indicates Clopidogrel in Unstable Angina to Prevent Recurrent Events; OR, odds ratio; MI, myocardial infarction; PCI, percutaneous coronary intervention; CABG, coronary artery bypass graft; BP, blood pressure; ULN, upper limit of normal; HR, heart rate; Hgb, hemoglobin. has been demonstrated in prior studies 9 and warrants further investigation. How can rates of clopidogrel prescription be improved? The positive correlation between hospital quality of MI care and clopidogrel prescription rates is encouraging, but the low R 2 value between the 2 factors indicates that simply improving overall hospital quality of MI care would have little effect on improving clopidogrel prescription. Rather, a more successful approach may be the incorporation of clopidogrel prescription as a performance measure for hospitals. The 2008 MI performance measures, released after the dates of the present study, included clopidogrel prescription rates for this population as a test measure, indicating that the writing committee believed that this measure was appropriate for internal quality improvement measures but not for external reporting until further validation and testing could be performed. 3 Future studies should investigate the impact of this test measure designation on reducing hospital variation in clopidogrel prescription rates. If significant improvement has not occurred and further studies can demonstrate reduction in death and recurrent MI, specifically among contemporary, medically managed patients with NSTEMI treated with clopidogrel, then consideration should be given to making clopidogrel prescription rates a full performance measure to provide external pressure to improve rates. Insights into factors associated with clopidogrel prescription can also inform the emerging use of the newer antiplatelet therapies prasugrel and ticagrelor because both have demonstrated salutary effects on patients with acute coronary syndromes and are likely to become part of routine clinical practice. Prasugrel is currently being compared with clopidogrel for medically managed patients with acute coronary syndromes in the TRILOGY-ACS (A Comparison of Prasugrel and Clopidogrel in Acute Coronary Syndrome Subjects With Unstable Angina/Non-ST-Elevation Myocardial Infarction Who Are Medically Managed) trial. 10 Ticagrelor demonstrated greater efficacy than clopidogrel for medically managed patients with NSTEMI in the PLATO (Platelet Inhibition and Patient Outcomes) trial, and recently received Food and Drug Administration approval for treatment in this population. 11,12 As these agents make the transition from clinical trials to the real world, the factors that affect clopidogrel prescription are also likely to affect their use, suggesting that quality improvement initiatives are likely to

8 8 Circ Cardiovasc Qual Outcomes July 2012 benefit prescription rates of all agents. Furthermore, despite the emergence of these newer agents, clopidogrel would be expected to remain a mainstay of antiplatelet therapy for patients with MI given the long-term experience with its use and its lower cost. The present study has several limitations. First, the ACTION Registry-GWTG only records prescription rates and does not track actual fill and use rates of the medication. Furthermore, the registry does not track care after discharge. However, prior studies have demonstrated that gaps in both prescription and use of secondary prevention medications, such as clopidogrel, at hospital discharge, are often magnified rather than reduced in the outpatient setting. 13,14 Second, incomplete or inaccurate documentation of clopidogrel contraindication could cause our characterization of a gap to be overstated. However, data abstraction for the ACTION Registry-GWTG is subject to data quality methods, which minimizes this potential bias. Third, the ACTION Registry- GWTG does not contain granular information on hospital quality processes. Thus, the general associations illustrated in this study will need confirmation and exploration among individual hospitals and delivery systems to identify specific hospital and provider factors that contribute to this gap. Fourth, the ACTION Registry-GWTG is an observational cohort and, thus, subject to unmeasured variables that may confound the reported associations. However, the size of the data set combined with robust modeling techniques that include multiple covariates for clopidogrel use reduces this source of error. Nonetheless, residual confounding can never be fully eliminated, and this may affect the findings. Fifth, the CURE trial took place 10 years ago. More contemporary treatments (eg, early invasive catheterization and intervention) may lessen the impact of the benefit of clopidogrel among the current NSTEMI population. Accordingly, providers may be appropriately deciding that the benefit of clopidogrel does not justify its risks. The data did not capture the nuances of this clinical consideration. Finally, participation in the ACTION Registry-GWTG is voluntary and likely represents hospitals delivering higher-quality care. Thus, the findings likely underreport the true extent of clopidogrel underuse in US hospitals. In conclusion, we found that significant gaps in clopidogrel prescription among medically managed patients with NSTEMI persist in contemporary practice, despite clinical practice guideline recommendations and the inclusion of clopidogrel prescription as a test performance measure for AMI. Clopidogrel prescription rates vary widely by hospital and are associated with a variety of patient and hospital factors. Although hospital quality of MI care is associated with its use, the present findings suggest that it only has a modest effect. Thus, other measures beyond improving the overall hospital quality of MI care, such as directly assessing clopidogrel prescription as a quality measure, are necessary to improve clopidogrel prescription rates and reduce adverse outcomes among medically managed patients with NSTEMI. Acknowledgments Dr Maddox is supported by a Veterans Administration Health Services Research and Development Career Development Award. He had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. This research was supported by the ACC Foundation s NCDR. The views expressed in this article represent those of the authors and do not necessarily represent the official views of the NCDR or its associated professional societies identified at Sources of Funding ACTION Registry-GWTG is an initiative of the ACC Foundation and the AHA with partnering support from the Society of Chest Pain Centers, the American College of Emergency Physicians, and the Society of Hospital Medicine. The registry is funded in part by an independent grant from Merck and by Bristol-Myers Squibb (who manufacture clopidogrel). The sponsors had no role in the conception, design, analysis, or writing of this article. None. Disclosures References 1. Yusuf S, Zhao F, Mehta SR, Chrolavicius S, Tognoni G, Fox KK. Effects of clopidogrel in addition to aspirin in patients with acute coronary syndromes without ST-segment elevation. N Engl J Med. 2001;345: Anderson JL, Adams CD, Antman EM, Bridges CR, Califf RM, Casey DE, Chavey WE, Fesmire FM, Hochman JS, Levin TN, Lincoff AM, Peterson ED, Theroux P, Wenger NK, Wright RS, Smith SC, Jacobs AK, Adams CD, Anderson JL, Antman EM, Halperin JL, Hunt SA, Krumholz HM, Kushner FG, Lytle BW, Nishimura R, Ornato JP, Page RL, Riegel B. 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. J Am Coll Cardiol. 2007;50:e1 e Krumholz HM, Anderson JL, Bachelder BL, Fesmire FM, Fihn SD, Foody JM, Ho PM, Kosiborod MN, Masoudi FA, Nallamothu BK. ACC/AHA 2008 performance measures for adults with ST-elevation and non-st-elevation myocardial infarction: a report of the American College of Cardiology/American Heart Association Task Force on Performance Measures (Writing Committee to Develop Performance Measures for ST-Elevation and Non-ST-Elevation Myocardial Infarction) developed in collaboration with the American Academy of Family Physicians and American College of Emergency Physicians endorsed by the American Association of Cardiovascular and Pulmonary Rehabilitation, Society for Cardiovascular Angiography and Interventions, and Society of Hospital Medicine. J Am Coll Cardiol. 2008;52: Amsterdam EA, Peterson ED, Ou F-S, Newby LK, Pollack CV, Gibler WB, Ohman EM, Roe MT. Comparative trends in guidelines adherence among patients with non-st-segment elevation acute coronary syndromes treated with invasive versus conservative management strategies: results from the CRUSADE quality improvement initiative. Am Heart J. 2009;158: Chin CT, Chen AY, Wang TY, Alexander KP, Mathews R, Rumsfeld JS, Cannon CP, Fonarow GC, Peterson ED, Roe MT. Risk adjustment for in-hospital mortality of contemporary patients with acute myocardial infarction: the Acute Coronary Treatment and Intervention Outcomes Network (ACTION) Registry-Get With The Guidelines (GWTG) acute myocardial infarction mortality model and risk score. Am Heart J. 2011; 161: Kahn CN III, Ault T, Isenstein H, Potetz L, Van Gelder S. Snapshot of hospital quality reporting and pay-for-performance under Medicare. Health Aff (Millwood). 2006;25: Tricoci P, Roe MT, Mulgund J, Newby LK, Smith SC, Pollack CV, Fintel DJ, Cannon CP, Bhatt DL, Gibler WB, Ohman EM, Peterson ED, Harrington RA. Clopidogrel to treat patients with non-st-segment elevation

9 Maddox et al Clopidogrel and Hospital Quality 9 acute coronary syndromes after hospital discharge. Arch Intern Med. 2006;166: Ho PM, Luther SA, Masoudi FA, Gupta I, Lowy E, Maynard C, Sales AE, Peterson ED, Fihn SD, Rumsfeld JS. Inpatient and follow-up cardiology care and mortality for acute coronary syndrome patients in the Veterans Health Administration. Am Heart J. 2007;154: Chan MY, Becker RC, Harrington RA, Peterson ED, Armstrong PW, White H, Fox KAA, Ohman EM, Roe MT. Noninvasive, medical management for non-st-elevation acute coronary syndromes. Am Heart J. 2008;155: Chin CT, Roe MT, Fox KAA, Prabhakaran D, Marshall DA, Petitjean H, Lokhnygina Y, Brown E, Armstrong PW, White HD, Ohman EM. Study design and rationale of a comparison of prasugrel and clopidogrel in medically managed patients with unstable angina/non-st-segment elevation myocardial infarction: the TaRgeted platelet Inhibition to clarify the Optimal strategy to medically manage Acute Coronary Syndromes (TRILOGY ACS) trial. Am Heart J. 2010;160: Wiviott SD, Braunwald E, McCabe CH, Montalescot G, Ruzyllo W, Gottlieb S, Neumann F-J, Ardissino D, De Servi S, Murphy SA, Riesmeyer J, Weerakkody G, Gibson CM, Antman EM. Prasugrel versus clopidogrel in patients with acute coronary syndromes. N Engl J Med. 2007;357: Wallentin L, Becker RC, Budaj A, Cannon CP, Emanuelsson H, Held C, Horrow J, Husted S, James S, Katus H, Mahaffey KW, Scirica BM, Skene A, Steg PG, Storey RF, Harrington RA, Freij A, Thorsén M. Ticagrelor versus clopidogrel in patients with acute coronary syndromes. N Engl J Med. 2009;361: Ho PM, Tsai TT, Maddox TM, Powers JD, Carroll NM, Jackevicius C, Go AS, Margolis KL, DeFor TA, Rumsfeld JS, Magid DJ. Delays in filling clopidogrel prescription after hospital discharge and adverse outcomes after drug-eluting stent implantation: implications for transitions of care. Circ Cardiovasc Qual Outcomes. 2010;3: Ho PM, Tsai TT, Wang TY, Shetterly SM, Clarke CL, Go AS, Sedrakyan A, Rumsfeld JS, Peterson ED, Magid DJ. Adverse events after stopping clopidogrel in post-acute coronary syndrome patients: insights from a large integrated healthcare delivery system. Circ Cardiovasc Qual Outcomes. 2010;3:

10 Clopidogrel Use and Hospital Quality in Medically Managed Patients With Non ST-Segment Elevation Myocardial Infarction Thomas M. Maddox, P. Michael Ho, Thomas T. Tsai, Tracy Y. Wang, Shuang Li, S. Andrew Peng, Stephen D. Wiviott, Fredrick A. Masoudi and John S. Rumsfeld Circ Cardiovasc Qual Outcomes. published online May 8, 2012; Circulation: Cardiovascular Quality and Outcomes is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX Copyright 2012 American Heart Association, Inc. All rights reserved. Print ISSN: Online ISSN: The online version of this article, along with updated information and services, is located on the World Wide Web at: Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in Circulation: Cardiovascular Quality and Outcomes can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office. Once the online version of the published article for which permission is being requested is located, click Request Permissions in the middle column of the Web page under Services. Further information about this process is available in the Permissions and Rights Question and Answer document. Reprints: Information about reprints can be found online at: Subscriptions: Information about subscribing to Circulation: Cardiovascular Quality and Outcomes is online at: