A multifaceted intervention to narrow the evidence-based gap in the treatment of acute coronary syndromes: Rationale and design of the Brazilian Intervention to Increase Evidence Usage in Acute Coronary Syndromes (BRIDGE-ACS) cluster-randomized trial

(1)

A multifaceted intervention to narrow the

evidence-based gap in the treatment of acute coronary

syndromes: Rationale and design of the Brazilian

Intervention to Increase Evidence Usage in Acute

Coronary Syndromes (BRIDGE-ACS)

cluster-randomized trial

Otávio Berwanger, MD, PhD,a Hélio P. Guimarães, MD, PhD,aLigia N. Laranjeira, MS,a

Alexandre B. Cavalcanti, MD,aAlessandra Kodama, MD,aAna Denise Zazula, MD,aEliana Santucci, MS,a Elivane Victor, MS,aUri A. Flato, MD,aMarcos Tenuta, MD,aVitor Carvalho, PhD,aVera Lucia Mira, MS,a Karen S. Pieper, MS,bLuiz Henrique Mota, MD,aEric D. Peterson, MD, MPH,band Renato D. Lopes, MD, PhDb,c São Paulo, Brazil; and Durham, NC

Translating evidence into clinical practice in the management of acute coronary syndromes (ACS) is challenging. Few ACS quality improvement interventions have been rigorously evaluated to determine their impact on patient care and clinical outcomes. We designed a pragmatic, 2-arm, cluster-randomized trial involving 34 clusters (Brazilian public hospitals). Clusters were randomized to receive a multifaceted quality improvement intervention (experimental group) or routine practice (control group). The 6-month educational intervention included reminders, care algorithms, a case manager, and distribution of educational materials to health care providers. The primary end point was a composite of evidence-based post-ACS therapies within 24 hours of admission, with the secondary measure of major cardiovascular clinical events (death, nonfatal myocardial infarction, nonfatal cardiac arrest, and nonfatal stroke). Prescription of evidence-based therapies at hospital discharge were also evaluated as part of the secondary outcomes. All analyses were performed by the intention-to-treat principle and took the cluster design into account using individual-level regression modeling (generalized estimating equations). If proven effective, this multifaceted intervention would have wide use as a means of promoting optimal use of evidence-based interventions for the management of ACS. (Am Heart J 2012;163:323-329.e1.)

Cardiovascular diseases, especially acute coronary

syndromes (ACS), represent the leading cause of

morbidity and mortality globally.

1,2

Large-scale

random-ized trials and systematic reviews have established the

ef

ﬁ

cacy and safety of several interventions for the

management of patients with ACS, including antiplatelet

therapy, anticoagulation, reperfusion strategies for

patients with ST-segment elevation myocardial infarction

(MI), and secondary prevention with

β

-blockers, statins,

and angiotensin-converting enzyme inhibitors.

3-7

Never-theless, registries have consistently demonstrated that

the translation of such research

ﬁ

ndings into clinical

practice is suboptimal.

8,9

_{The reasons for this gap}

between evidence and practice are complex and include

barriers related to knowledge (lack of awareness and

familiarity), behavior (external barriers), and attitudes

(lack of agreement with current evidence, outcome

expectancy, and the inertia of previous practices in

changing behavior).

10

Prior studies have demonstrated that certain quality

improvement strategies are associated with better quality

of care. These include reminder systems, academic

detailing (educational outreach visits), audit and

feed-back, case management, and distribution of educational

materials to health care providers.

11

Combined strategies

targeting different barriers are more likely to be effective

From thea_{Research Institute Hcor}

—Hospital do Coração, São Paulo, Brazil,bDuke Clinical

Research Institute, Duke University Medical Center, Durham, NC, andc_{Brazilian Clinical}

Research Institute, Federal University of São Paulo, Paulista School of Medicine, São Paulo, Brazil.

Clinicaltrials.gov:NCT00958958.

Neal S. Kleiman, MD served as guest editor for this article. Submitted January 3, 2012; accepted February 7, 2012.

Reprint requests: Otavio Berwanger, MD, PhD, Research Institute HCor–Hospital do

Coração, Abilio Soares Street, 250, 12th Floor, São Paulo-SP, Brazil. E-mail:[email protected]

(2)

than single interventions.

12

_{However, quality}

improve-ment interventions have rarely been rigorously evaluated.

We designed a cluster-randomized trial, Brazilian

Intervention to Increase Evidence Usage in Acute

Coronary Syndromes (BRIDGE-ACS), to assess the

effec-tiveness of a quality improvement initiative in patients

with ACS. The study has important features such as

concealed randomization (including reminders and

dis-tribution of educational materials and case management

at the professional practice or health care organization

level, which represents an ideal design for evaluating

dissemination and implementation strategies

13

_{) and}

evaluation of the use of evidence-based ACS treatment

and the impact on patient outcomes.

Methods

Study design and objectives

BRIDGE-ACS was a pragmatic 2-arm, cluster-randomized (concealed), controlled trial (Figure 1) with blinded adjudica-tion of outcomes and intenadjudica-tion-to-treat analysis. The main objective was to evaluate whether a multifaceted quality improvement intervention can improve the prescription of proven efﬁcacious therapies for patients with ACS within the

ﬁrst 24 hours and at hospital discharge and reduce the incidence of major cardiovascular events.

Participants

Cluster eligibility criteria and recruitment.

Clus-ters were eligible for the BRIDGE-ACS trial if they were general teaching or nonteaching tertiary public hospitals from major urban areas with an emergency department (ED) that receives patients with ACS (Table I). We excluded private hospitals, public cardiology institutes, and hospitals located in rural areas. A list of potential eligible clusters was provided by the Brazilian Ministry of Health. Letters were sent to eligible hospitals inviting them to participate; hospitals not responding to 2 letters were contacted by telephone.

Patient eligibility criteria and recruitment.

At

each participating cluster, all consecutive eligible patients with ACS were enrolled according to the following standardized deﬁnitions (Table I)15:

•ST-segment elevation myocardial infarction (STEMI): New or presumed new ST-segment elevation seen in any location or new left bundle-branch block on the index or qualifying electrocardiogram (ECG) with at least 1 positive cardiac biochemical marker of necrosis (including troponin).

• Non–ST-segment elevation myocardial infarction

(NSTEMI): Presence of at least 1 positive cardiac biochemical marker of necrosis without new ST-segment elevation seen on the index or qualifying ECG.

•Unstable angina (UA): Absence of ST-segment elevation on the ECG and serum biochemical markers indicative of myocardial necrosis within each hospital laboratory's

Figure 1

(3)

normal range but with a discharge diagnosis of ACS. Patients originally admitted for unstable angina but in whom MI occurred during the hospital stay were classiﬁed as having an MI.

Centers were instructed to enroll patients as soon as they presented in the ED. We excluded patients transferred from other hospitals withinN12 hours, patients with non–type I MIs, and patients for whom the presumptive admission diagnosis was ACS but were subsequently shown to have some other cardiac or noncardiac cause for their presentation.

Randomization and allocation concealment

Eligible clusters were randomly allocated (1:1) to a multifac-eted quality improvement intervention (experimental group) or to routine practice (control group). Randomization was stratiﬁed by the following characteristics: teaching versus nonteaching hospitals, presence or absence of a 24-hour cardiac catheterization laboratory, and capability for cardiac surgery. To guarantee concealment of allocation, all clusters were random-ized at once by a blinded statistician using a central Web-based randomization system developed by the Research Institute HCor (São Paulo, Brazil).

Blinding

Because of the nature of the intervention in the BRIDGE-ACS trial, only the independent outcome assessors were blinded to the intervention.

Quality improvement intervention

The multifaceted quality improvement intervention included reminders, a checklist, case management, and educational materials implemented during a 6-month period. Clusters randomized to receive the quality improvement intervention received on-site training visits complemented by Web-based and telephone training. In addition, 2 health care providers from all clusters (a physician who acted as the local leader and a nurse who acted as case manager) attended a 2-day workshop on how to implement the BRIDGE-ACS quality improvement interven-tion. These training sessions used simulation-based learning techniques. These 2 professionals trained the health care staff involved with the care of ACS patients at their site, provided formal documentation of training sessions to the coordinating center, and guaranteed adequate implementation of the quality improvement tools. At least 80% of the research medical staff from each site were trained for this study.

Reminders and checklist.

The reminders and the

checklist were designed to be implemented in sequence during the management of ACS patients. First, a printed reminder (“Chest Pain”label) was attached to the clinical evaluation form. This reminder serves as a rapid triage tool upon patient arrival in the ED (Figure 2). Second, once a patient has been given a potential diagnosis of ACS, the nurse gave the attending ED physician the clinical evaluation form with the “Chest Pain”

label and an attached checklist (Figure 3). The checklist contains an algorithm for risk stratiﬁcation (based on clinical presenta-tion, ECG analysis, and cardiac enzymes) and recommended evidence-based therapies for each risk category. The algorithm divides patients into 3 color categories: red for patients with ST-segment elevation ACS, yellow for patients with non–ST-segment

Table I. Eligibility criteria

Participants

Cluster eligibility criteria and recruitment Inclusion criteria

General teaching or nonteaching tertiary public hospitals from capitals or major urban areas with an ED that typically assists patients with ACS Exclusion criteria

Private for-profit or not-for-profit hospitals, public cardiology institutes, and hospitals located in rural areas

Patients eligibility criteria and recruitment Inclusion criteria

STEMI: new or presumed new ST-segment elevation seen in any location or new LBBB on the index or qualifying ECG with at least 1 positive cardiac biochemical marker of necrosis (including troponin) NSTEMI: presence of at least 1 positive cardiac biochemical marker of

necrosis without new ST-segment elevation seen on the index or qualifying ECG

Unstable angina: absence of ST-segment elevation on the ECG and serum biochemical markers indicative of myocardial necrosis within each hospital laboratory's normal range, but with a discharge diagnosis of ACS; patients originally admitted for unstable angina but in whom MI occurred during the hospital stay were classified as having an MI Exclusion criteria

Patients transferred from others hospitals within more than 12 hours and for whom the presumptive admission diagnosis was ACS but were subsequently shown to have some other cardiac or noncardiac cause for their presentation

LBBB, Left bundle-branch block.

Figure 2

Reminder of“chest pain”patient enrolled in study.

Berwanger et al325

(4)

elevation ACS, and green for patients with a normal ECG and cardiac enzymes. The checklist requires that the attending physician check and confirm the use (or no use in case of contraindications) of all suggested evidence-based interventions. Finally, once patients have been classified into 1 of the 3 catego-ries, they received a colored bracelet (red, yellow, or green) according to the risk stratification category (Figure 4). This bracelet helps to promptly identify ACS patients in the ED to avoid delays in initiating recommended evidence-based therapies.

Case manager.

A trained nurse who works at the hospital (in the ED or coronary care unit) acted as a case manager and followed up all patients during their hospital stay. The role of the case manager includes interacting with physicians once gaps in the incorporation of evidence-based interventions are identiﬁed, ensuring that all components of the quality improvement intervention are being used for every ACS patient, identifying barriers for the implementation of the quality improvement tools and consequently of evidence-based therapies, and continuous training of health care staff involved with the care of ACS patients.

Educational materials

We provided educational materials for all clusters randomized to the experimental group, including pocket guidelines, an interactive Web site containing presentations about ACS, instructional videos on how to implement the quality improve-ment intervention, and posters containing evidence-based recommendations for the management of ACS to be displayed in the ED, coronary care unit, and clinical wards.

Outcomes

An independent, blinded clinical events committee adjudicat-ed all suspectadjudicat-ed primary and secondary outcomes basadjudicat-ed on standardized and predeﬁned deﬁnitions.

The primary end point was 100% adherence to evidence-based therapies (aspirin; clopidogrel; anticoagulation with enoxaparin, unfractionated heparin, or fondaparinux; and

statins), in patients without contraindications, during thefirst 24 hours. Secondary outcomes included individual components of the primary combined end point (prescription rate of each of the evidence-based medications during the first 24 hours); prescription rates of aspirin,β-blockers, statins, and angiotensin-converting enzyme inhibitors (in combination and individually) at discharge; combined end point of 100% adherence to evidence-based therapies (acute and discharge); reperfusion rates for patients with STEMI and 100% adherence to evidence-based therapies for patients with STEMI (aspirin, reperfusion, clopidogrel, anticoagulation, and statins) during the first 24 hours; major cardiovascular events, including a combined end point of cardiovascular mortality, nonfatal MI, nonfatal cardiac arrest, and nonfatal stroke at discharge and at 30 days; all-cause mortality during the initial hospitalization and at 30 days; and major bleeding (in-hospital). The following definitions were used for clinical outcomes:

•Mortality: Classified as being cardiovascular or nonvascular because of other specified causes (or of unknown etiology). Cardiovascular deathis defined as any death with a vascular cause and includes death after an MI, cardiac arrest, stroke, cardiac revascularization procedure (ie, percutaneous coro-nary intervention or corocoro-nary artery bypass graft surgery), pulmonary embolus, hemorrhage, or death caused by an unknown cause.Nonvascular deathis defined as any death caused by a clearly documented nonvascular cause (eg, trauma, infection, malignancy).

•Myocardial infarction: Patients had to fulﬁll at least 2 of the following: (1) typical prolonged severe chest pain or related symptoms or signs (eg, ST-segment changes of T-wave inversion on ECG) suggestive of MI; (2) elevation of troponin or creatine kinase–MB to more than the upper limit of normal, or if creatine kinase–MB was elevated at baseline, reelevation of at least 50% above the previous level; (3) development of signiﬁcant Q waves in at least 2 adjacent ECG leads.

•Stroke: Deﬁned as an acute onset of a focal neurologic deﬁcit of presumed vascular origin lasting for≥_{24 hours or resulting}

Figure 3

(5)

in death. Stroke is categorized as ischemic, hemorrhagic, or unknown cause (based on computed tomographic or magnetic resonance scanning or autopsy). Fatal stroke is deﬁned as death from any cause within 30 days of stroke.

•Nonfatal cardiac arrest: Defined as successful resuscitation from either documented or presumed ventricularfibrillation, sustained ventricular tachycardia, asystole, or pulseless electrical activity requiring cardiopulmonary resuscitation, pharmacologic therapy, or cardiac defibrillation.

•Major bleeding: Deﬁned as a decrease in hemoglobin levels of≥_{20 g/L (2 g/dL) or more, a blood transfusion requiring}≥₂ U of whole blood or red cells, intracerebral bleeding, retroperitoneal bleeding, pericardial bleeding, bleeding into a major joint, bleeding into the eye, and fatal bleeding.

Sample size

We performed a prerandomization survey at participating sites and found that the primary end point rates were in the range of 40%. To detect a 20% improvement in the prescription of 100% evidence-based therapies within thefirst 24 hours (primary end point), considering 80% power, 5% significance level, and an intracluster correlation coefficient of 0.21 (value based on a prerandomization pilot phase), we needed to randomize at least 34 clusters and approximately 1,020 patients (considering a median of 30 ACS patients per cluster). The final population enrolled in the BRIDGE-ACS was 1,150 patients from 34 clusters.

Statistical analysis plan

All analyses will follow the intention-to-treat principle. Baseline characteristics will be analyzed to evaluate cluster imbalances between treatment and control groups and to

identify possible control variables to be included in theﬁnal models.16,17Continuous variables with normal distribution will be summarized using mean and SD, and those with skewed distribution will be described using median and 25th and 75th percentiles. Dichotomous variables will be described by pro-portions. All estimates at the cluster or patient level will be weighted by the number of clusters and the number of patients in each cluster.

Comparisons between groups at the cluster level will be conducted using the Student t test or Mann-Whitney U test (when appropriate) for continuous variables and the χ2 or Fisher exact test for comparisons of proportions. For variables with skewed distribution, a logarithmic transformation will be used, and all tests will be weighted by the number of patients at each cluster. Comparisons between the intervention and control groups at the patient level will be done using a generalized estimation equation for modeling proportions and quantitative variables, taking into account the dependence between observations within a cluster. The intervention effects will be expressed as a population average odds ratio in the case of qualitative variables or the mean difference in the case of quantitative variables, both with 95% CIs. The effect of intervention on cardiovascular events at discharge and mortality at 30 days will be illustrated by Kaplan-Meier curves and compared using proportional hazards Cox models accounting for dependence between patients from the same cluster (shared

γ frailty models). A sensitivity analysis will be performed including patients with contraindications for evidence-based therapies in the denominator of the overall population. Subgroup analysis of the primary end point of the trial among patients with ST-segment elevation and non–ST-segment eleva-tion will also be performed. All analyses will be conducted using STATA SE 11 for Windows (STATA Corp LP, College Station, TX)18 or R 2.13 software.19 No interim analyses will be performed. The main statistical analyses will be performed by the Research Institute HCor and validated by the Duke Clinical Research Institute (Durham, NC).

Ethical considerations

This study was conducted in accordance with the Declaration of Helsinki and good clinical practice guidelines. All participat-ing clusters submitted the study protocol for approval by their research ethics board; written consent was obtained at the cluster level. This is a common and well-accepted approach; the objective of such an approach is to avoid selection bias that may arise from different consent refusals rates between clusters.20

Organizational structure

The BRIDGE-ACS trial was led by an academic steering committee composed of 2 co-chairs and 1 principal investigator (see online Appendix). This committee provided scientiﬁc direction and input, addressed policy issues regarding the protocol, and met periodically to assess the trial progress. The executive committee is composed of a subset of senior leaders from the steering committee. This committee was responsible for evaluating the progress and safety of the trial and made any decisions regarding early termination or continuation of the trial. The trial was conducted by 2 coordinating centers: the Research Institute HCor and the Brazilian Clinical Research Institute, both in Sao Paulo, Brazil. The Duke Clinical Research

Figure 4

Patient bracelet.

Berwanger et al327

(6)

Institute will assist with the statistical analysis validation of the main results. Some of the coordinating centers' activities include selecting and training trial centers; assisting trial centers with regulatory submissions; distributing and supplying study sites with the tools and forms, if appropriate; monitoring recruitment and follow-up at trial centers; data management, testing, and maintenance of the electronic data capture system, and data quality control; and data analysis.

Data collection, quality control, and clinical data

management system.

In all participating clusters, data will be collected prospectively by a trained independent health professional not involved in the care of ACS patients. Data will be entered using an electronic Web-based data capture system developed on a Microsoft SQL (Microsoft Corp, Redmond, WA) platform by a team of programmers at the Research Institute HCor. Data quality control will be guaranteed by automated data entry checks, weekly contact with investigators, on-site moni-toring, and central statistical monitoring.21_{General feedback will}

be provided at investigator meetings and in periodic newsletters.

This trial is registered at clinicaltrials.gov (NCT00958958).14 Funding for this study was provided by the Brazilian Ministry of Health (Projeto Hospitais de Excelencia a Serviço do SUS–

PROADI). The authors are solely responsible for the design and conduct of this study, all study analyses, the drafting and editing of the manuscript, and itsﬁnal contents.

Discussion

Previous studies using a before-after design

9,22,23

have

shown absolute improvements in the uptake of

evidence-based therapies of 5% to 15%. Before-after studies are

superior to observational studies; however, they may be

prone to limitations because secular trends or sudden

changes make it dif

ﬁ

cult to attribute observed changes to

the intervention.

13

Furthermore, in such studies, the

intervention is confounded by the Hawthorne effect,

which could lead to an overestimate of the effectiveness

of an intervention. To prevent such types of bias,

cluster-randomized trials are recommended as the optimal design

for evaluating quality improvement interventions. In

addition, cluster-randomized trials allow adequate control

of contamination, which would be challenging in a trial

with randomization at the patient level. Previous

cluster-randomized studies

24–26

in the setting of ACS using

different quality improvement tools have had mixed

results, and none evaluated clinical end points or were

conducted in emerging economies and developing

countries, where approximately 80% of the health care

burden is attributable to cardiovascular diseases. In

addition, educational programs such as the one tested

in the BRIDGE-ACS trial can be expensive. Thus, studying

these interventions to prove that they can work is

necessary to sustain them. Finally, if the quality

improve-ment interventions tested in our study are not effective,

then more intensive efforts, with more associated costs,

would be necessary to improve patient care.

In summary, validated quality improvement programs

represent an important way to bridge the gap between

research evidence and clinical practice. If proven

effective, the multifaceted intervention proposed in the

BRIDGE-ACS trial can become a routine part of clinical

practice to maximize the use of evidence-based

in-terventions for the management of ACS.

Disclosures

Role of the funding source: This study is funded by the

Brazilian Ministry of Health in Partnership with Hospital

do Coracao (HCor)

—

Programa Hospitais de Excelencia a

Servico do Sus (PROADI-SUS). The funding source has no

role in the design, execution, analysis, and decision to

publish the results.

Competing interests: The author(s) declare that they

have no individually competing interests.

References

1. Yusuf S, Reddy S, Ounpuu S, et al. Global burden of cardiovascular diseases: part I: general considerations, the epidemiologic transition, risk factors, and impact of urbanization. Circulation 2001;104: 2746-53.

2. Yusuf S, Reddy S, Ounpuu S, et al. Global burden of cardiovascular diseases: part II: variations in cardiovascular disease by specific ethnic groups and geographic regions and prevention strategies. Circulation 2001;104:2855-64.

3. Fibrinolytic Therapy Trialists' (FTT) Collaborative Group. Indications for fibrinolytic therapy in suspected acute myocardial infarction: collaborative overview of early mortality and major morbidity results from all randomized trials of more than 1000 patients. Lancet 1994; 343:311-22.

4. Keeley EC, Boura JA, Grines CL. Primary angioplasty versus intravenous thrombolytic therapy for acute myocardial infarction: a quantitative review of 23 randomised trials. Lancet 2003;361: 13-20.

5. Freemantle N, Cleland J, Young P, et al. Beta blockade after myocardial infarction: systematic review and meta regression analysis. BMJ 1999;318:1730-7.

6. Sposito AC, Caramelli B, Fonseca FA, et al. IV Brazilian guideline for dyslipidemia and atherosclerosis prevention: Department of Athero-sclerosis of Brazilian Society of Cardiology. Arq Bras Cardiol 2007; 88(Suppl1):2-19.

7. Chen ZM, Jiang LX, Chen YP, et al. Addition of clopidogrel to aspirin in 45,852 patients with acute myocardial infarction: randomised placebo-controlled trial. Lancet 2005;336:1607-21.

8. Fox KA, Goodman SG, Klein W, et al. Management of acute coronary syndromes. Variations in practice and outcome: findings from the Global Registry of Acute Coronary Events (GRACE). Eur Heart J 2002;23:1177-89.

9. Hoekstra JW, Pollack Jr CV, Roe MT, et al. Improving the care of patients with non–ST-elevation acute coronary syndromes in the emergency department: the CRUSADE initiative. Acad Emerg Med 2002;9:1146-55.

10. Cabana MD, Rand CS, Powe NR, et al. Why don't physicians follow clinical practice guidelines? A framework for improvement. JAMA 1999;282:1458-65.

(7)

12. Grimshaw JM, Shirran L, Thomas R, et al. Changing provider behavior: an overview of systematic reviews of interventions. Med Care 2001;39:II2-45.

13. Grimshaw J, Campbell M, Eccles M, et al. Experimental and quasi-experimental designs for evaluating guideline implementation strategies. Fam Pract 2000;17(Suppl1):S11-6.

14. BRIDGE-ACS (Brazilian Intervention to Increase Evidence Usage in Acute Coronary Syndromes) trial: NCT00958958. ClinicalTrials.gov Web sitehttp://clinicaltrials.gov/ct2/results?term=NCT00958958

Accessed December 29, 2011.

15. Anderson JL, Adams CD, Antman EM, et al. 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 Management of Patients With Unstable Angina/Non ST-Elevation Myocardial Infarction): developed in collaboration with the American College of Emergency Physicians, the Society for Cardiovascular Angiography and Interventions, and the Society of Thoracic Surgeons: endorsed by the American Association of Cardiovascular and Pulmonary Rehabilitation and the Society for Academic Emergency Medicine. Circulation 2007;116:e148-304.

16. Campbell MJ, Donner A, Klar N. Developments in cluster randomized trials and statistics in medicine. Stat Med 2007;26:2-19.

17. Hayes RL, Moulton LH. Cluster randomised trials. Boca Raton, FL: Chapman & Hall/CRC; 2009.

18. Kreuter F, Valliant R. A survey on survey statistics: what's done and can be done in Stata. Stata J 2011;7(Suppl1):1-21.

19. Development Core Team R. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2008.

20. Edwards SJ, Braunholtz DA, Lilford RJ, et al. Ethical issues in the design and conduct of cluster randomised controlled trials. BMJ 1999;318:1407-9.

21. Buyse M, George SL, Evans S, et al. The role of biostatistics in the prevention, detection and treatment of fraud in clinical trials. Stat Med 1999;18:3435-51.

22. Mehta RH, Montoye CK, Gallogly M, et al. Improving quality of care for acute myocardial infarction: the Guidelines Applied in Practice (GAP) Initiative. JAMA 2002;287:1269-76.

23. Fonarow GC, Gawlinski A, Moughrabi S, et al. Improved treatment of coronary heart disease by implementation of a Cardiac Hospitali-zation Atherosclerosis Management Program (CHAMP). Am J Cardiol 2001;87:819-22.

24. Flather MD, Babalis D, Booth J, et al. Cluster-randomized trial to evaluate the effects of a quality improvement program on manage-ment of non–ST-elevation acute coronary syndromes: the European Quality Improvement Programme for Acute Coronary Syndromes (EQUIP-ACS). Am Heart J 2011;62:700-7.

25. Carlhed R, Bojesting M, Wallentin L, et al. Improved adherence to Swedish national guidelines for acute myocardial infarction: the Quality Improvement in Coronary Care (QUICC) Study. Am Heart J 2006;152:1175-81.

26. Flather M, Booth J. Prospective registry of outcomes and management in ischaemic syndromes. Presented at World Congress of Cardiology; September 2006; Barcelona, Spain; 2006.

Berwanger et al329

(8)