0
We're unable to sign you in at this time. Please try again in a few minutes.
Retry
We were able to sign you in, but your subscription(s) could not be found. Please try again in a few minutes.
Retry
There may be a problem with your account. Please contact the AMA Service Center to resolve this issue.
Contact the AMA Service Center:
Telephone: 1 (800) 262-2350 or 1 (312) 670-7827  *   Email: subscriptions@jamanetwork.com
Error Message ......
Original Investigation |

Better Outcomes for Patients Treated at Hospitals That Participate in Clinical Trials FREE

Sumit R. Majumdar, MD, MPH; Matthew T. Roe, MD, MHS; Eric D. Peterson, MD, MPH; Anita Y. Chen, MS; W. Brian Gibler, MD; Paul W. Armstrong, MD
[+] Author Affiliations

Author Affiliations: Department of Medicine, University of Alberta, Edmonton, Alberta, Canada (Drs Majumdar and Armstrong); Duke Clinical Research Institute, Duke University Medical Center, Durham, North Carolina (Drs Roe and Peterson and Ms Chen); and Department of Emergency Medicine, University of Cincinnati School of Medicine, Cincinnati, Ohio (Dr Gibler).


Arch Intern Med. 2008;168(6):657-662. doi:10.1001/archinternmed.2007.124.
Text Size: A A A
Published online

Background  Barriers to institutions participating in clinical trials include concerns about harms and costs. However, we hypothesized that patients treated at hospitals participating in trials would have better outcomes than patients treated at nonparticipating hospitals. We tested this hypothesis in 494 CRUSADE (Can Rapid Risk Stratification of Unstable Angina Patients Suppress Adverse Outcomes With Early Implementation of the American College of Cardiology/American Heart Association Guidelines) hospitals treating 174 062 patients with non–ST-segment elevation acute coronary syndrome.

Methods  Hospitals were classified into tertiles by percentage of patients concurrently enrolled in non–ST-segment elevation acute coronary syndrome trials. Outcomes were use of composite guideline-indicated care and in-hospital mortality. Multivariate regression was used to examine the association between hospital trial participation and outcomes.

Results  Overall, 4590 patients (2.6%) were enrolled in trials, ranging from 0% (145 hospitals) to low-enrollment tertile (1.0%; interquartile range [IQR], 0.5%-1.4%; n = 226) to high-enrollment tertile (4.9%; IQR, 3.5%-9.7%; n = 123). The composite guideline adherence score increased with increasing tertiles of trial participation: 76.9% (IQR, 71.8%-81.3%) vs 78.3% (IQR, 73.2%-82.4%) vs 81.1% (IQR, 76.2%-84.1%) (adjusted P = .008). Hospitals that participated in trials had higher adjusted guideline adherence than nonparticipating hospitals (low enrollment, 0.8% greater [95% confidence interval {CI}, −0.9% to 2.6%]; and high enrollment, 2.5% greater [95% CI, 0.5%-4.5%]). In-hospital mortality decreased with increasing trial participation: 5.9% vs 4.4% vs 3.5% (adjusted P = .003). Patients treated at hospitals that participated in trials had significantly lower mortality than patients treated at nonparticipating hospitals (low enrollment adjusted odds, 0.9 [95% CI, 0.8-1.0]; and high enrollment adjusted odds, 0.8 [95% CI, 0.7-0.9]).

Conclusions  The CRUSADE hospitals enrolled less than 3% of their patients with non–ST-segment elevation acute coronary syndrome into trials, and one-third never participated in trials. Compared with hospitals that do not participate in trials, those hospitals that do participate in trials seem to provide better care and to have lower mortality.

Figures in this Article

The quality of care for common conditions treated in hospitals, such as acute coronary syndromes, has improved over time in response to new evidence, clinical guidelines, performance measurement and reporting, and quality improvement efforts.13 Nevertheless, the rate of improvement has been slow and nonuniform, with marked variability between hospitals.13 Furthermore, the link between better processes of care (eg, provision of guideline-concordant care) and better outcomes (eg, mortality) has only recently been convincingly demonstrated.4,5 A better understanding of the structures and characteristics of hospitals that deliver high-quality care, beyond easy-to-measure attributes such as size, caseload, or teaching status, is, thus, desirable.16

Bradley et al1 examined 1 performance measure (use of β-blockers after myocardial infarction) and determined that 4 hospital characteristics were associated with better performance: physician leadership, presence of shared goals, administrative (infrastructure) support, and credible feedback. While it is difficult to characterize hospitals according to such attributes for multiple measures or conditions, we have previously observed that 3 of these characteristics (physician leadership, shared goals, and infrastructure support) are also essential to the successful conduct of hospital-based clinical trials.

Indeed, much of the clinical trial enterprise has been developed to ensure that trial subjects receive safe, high-quality, protocol-driven care from highly trained research personnel overseen by experienced and well-informed investigators.713 We hypothesized that these same elements required for hospitals to participate in trials could induce beneficial changes in the hospital environment, thereby leading to better processes and outcomes of care for patients treated outside the trial setting. To test this hypothesis, we studied 174 062 patients treated at 494 hospitals that were part of the CRUSADE (Can Rapid Risk Stratification of Unstable Angina Patients Suppress Adverse Outcomes With Early Implementation of the American College of Cardiology/American Heart Association Guidelines) Initiative.4,6,10 Specifically, we hypothesized that patients with non–ST-segment acute coronary syndromes treated at hospitals that participated in trials would receive better care (ie, greater adherence to American College of Cardiology/American Heart Association treatment guidelines) and have better outcomes (ie, lower mortality) than patients treated at hospitals that did not participate.

SETTING AND SUBJECTS

The CRUSADE Initiative has previously been described.4,6,10 In brief, CRUSADE is an ongoing, voluntary, observational data collection and quality improvement initiative that involves 557 hospitals in the United States. It was initiated in November 2001, with retrospective data collection starting in January 2001. The institutional review board of each hospital approved participation, and because unique identifiers were not collected, individual consent was waived.

CRUSADE included only patients with high-risk non–ST-segment elevation acute coronary syndrome with unstable angina and non–ST-segment elevation myocardial infarction. All CRUSADE patients presented with acute ischemic symptoms (lasting for at least 10 minutes at rest) within 24 hours of hospital arrival and had at least 1 of the following diagnostic features to distinguish those with an increased risk of adverse outcomes: ST-segment depression of 0.5 mm or greater, transient ST-segment elevation of 0.5 to 1.0 mm (lasting for <10 minutes), and/or a positive result for cardiac biomarkers.

We analyzed patients included in CRUSADE from January 1, 2001, through June 30, 2006. Of 180 842 patients with non–ST-segment elevation acute coronary syndromes treated at 557 hospitals, we excluded patients with missing data regarding trial enrollment (5510 [3.0%]) and low-volume centers (1270 patients cared for at 63 hospitals). This left a final study sample of 174 062 patients with non–ST-segment elevation acute coronary syndrome treated at 494 hospitals.

DATA COLLECTION AND VALIDATION

Participating hospitals collected detailed process-of-care and in-hospital outcome data through medical record reviews of consecutive eligible patients. Data collected include use of medications within 24 hours of presentation, use of invasive cardiac procedures, laboratory results, and discharge therapies. In 2002, CRUSADE assessed reliability of data abstraction with randomly selected sites and records.4 The overall accuracy of audited records was 95%, and the degree of missing data was 5% across all collected data elements.4

EXPLANATORY VARIABLE OF INTEREST

We abstracted the dichotomous variable “clinical trial enrollment during the index hospitalization” for every patient.10 Clinical trial referred to any trial for the management of acute coronary syndrome. Beyond this, we did not gather information on whether patients were enrolled in a drug or device trial and we did not collect any information with respect to trial participation outside the field of cardiology. We conceptualized “trial participation” as a hospital-level characteristic and defined it by the proportion of CRUSADE patients who were concomitantly enrolled in a clinical trial during hospitalization. For analysis, we initially aggregated hospitals into quartiles of trial participation; for ease of presentation and because the second and third quartiles were virtually indistinguishable, we decided to aggregate hospitals into tertiles of trial participation (no trial participation vs low participation vs high participation).

OUTCOMES

To determine whether hospitals that participated in trials delivered better quality care, we examined a previously validated “composite guideline adherence score.”4 This was derived by determining adherence to 9 individual class 1 guideline-recommended therapies among patients eligible to receive these therapies.4Eligible was defined, for denominator purposes, as being alive, not being transferred to another hospital, and having no contraindications to the treatment. Therapies included 4 short-term measures (aspirin, β-blockers, heparin, and glycoprotein IIb/IIIa inhibitors) and 5 discharge measures (aspirin, β-blockers, clopidogrel, angiotensin-converting enzyme inhibitors, and statins).4 Definitions, indications, and contraindications according to established guidelines have been previously published,4,6,10 and further details are available from the authors on request. For each patient, a composite adherence score was calculated as the sum of correct care according to each patient's total number of eligible opportunities for care. Thus, each patient could have received up to 9 therapies (4 within the first 24 hours and 5 by the time of discharge) and would have been eligible to receive anywhere from 1 to 9 therapies. The ratio of the number of treatments received to the number of treatments the patient was eligible for, multiplied by 100, represents that patient's composite adherence score. The individual results were then summated to derive a hospital-level composite guideline adherence score. In the original development and validation of these composite scores, hospital-level guideline adherence ranged from 35% to 90%.4

To determine whether hospitals that participated in trials had better outcomes, we examined all-cause (in-hospital) mortality. A priori, we chose all-cause mortality because it would not be susceptible to bias with respect to determining the cause of death, it provided an integrated outcome that captured all serious adverse events, and it provided the most events.

ANALYSIS

We stratified hospital and patient characteristics, patterns of care, and outcomes by tertiles of hospital-level trial participation. Continuous variables are described as medians with 25th and 75th percentiles (ie, the interquartile range [IQR]), while categorical variables are described as frequencies. We used Spearman rank correlation coefficients to describe the relationship between hospital-level trial participation (as a continuous measure) and the composite guideline adherence scores and percentage of mortality.

To examine the association between hospital-level trial participation and hospital-level composite guideline adherence scores, we performed hospital-level analyses using linear regression models that accounted for the number of patients per hospital. These multivariate linear regression models adjusted for all available hospital characteristics, including total number of hospital beds, geographic region (West, Northeast, Midwest, or South), revascularization capabilities (no services, diagnostic catheterization only, and percutaneous coronary intervention without on-site cardiac surgery vs percutaneous coronary intervention with on-site cardiac surgery), and teaching status (academic vs nonacademic).

To examine the association between hospital-level trial participation and in-hospital mortality, we conducted patient-level analyses. For these analyses, to reduce bias, as has been recommended by others,8 we excluded patients who were transferred to another hospital (20 835 patients [12.0%]). We accounted for within-hospital clustering, whereby patients at the same hospital were more likely to have similar responses to each other relative to patients treated at other hospitals (eg, within-center correlation), by using logistic generalized estimating equation models to estimate marginal effects of percentage of hospital-level trial participation.14,15 This method produces estimates similar to those from ordinary logistic regression, but the variances of the estimates are adjusted for within-hospital clustering of responses. Variables entered into our model included sociodemographic factors (age [continuous], sex, race [white vs nonwhite], and insurance status), family history of coronary disease, medical history (hypertension, diabetes mellitus, smoking, hyperlipidemia, myocardial infarction, percutaneous coronary intervention, coronary artery bypass grafting, heart failure, stroke, and renal insufficiency), features of the initial clinical presentation (ST-segment depression, transient ST-segment elevation, signs of heart failure, heart rate [continuous], systolic blood pressure [continuous], body mass index, and specialty of attending physician [cardiology vs noncardiology]), and all hospital-level characteristics. All analyses were performed using SAS statistical software, version 9.1.3 (SAS Institute Inc, Cary, North Carolina).

Our final study sample consisted of 494 US hospitals. Most hospitals were large (median number of beds, 265; IQR, 137-457); 79.3% had capacity for surgical revascularization, and 29.1% were academic teaching centers. Fully, 71.0% of hospitals were clinical trial sites (ie, they enrolled at least 1 patient in a clinical trial during the observation period). Despite this, only 4590 patients (2.6%) were ever concurrently enrolled in a clinical trial, with hospital-level trial participation ranging from 0% to 71.4% of all patients with non–ST-segment elevation acute coronary syndrome. Tertiles of hospital-level trial participation were as follows: no enrollment (0% enrolled in trials [145 sites]), low tertile (median, 1.0% of patients enrolled; IQR, 0.5%-1.4% [226 sites]), and high tertile (median, 4.9% of patients enrolled; IQR, 3.5%-9.7% [123 sites]). Hospitals that enrolled patients in trials were larger, tended to have surgical revascularization capacity, and were more likely to be academic centers than hospitals that did not participate in trials (Table 1).

Table Graphic Jump LocationTable 1. Characteristics of 494 CRUSADE Hospitals by Tertiles of Clinical Trial Participation

Characteristics of the 174 062 patients in this analysis included a median age of 68 years (IQR, 56-78 years); 40.2% were female, 82.2% were white, and all but 5.9% had some health insurance. Because of the large sample, there were many statistically significant differences across tertiles of trial participation, but few, if any, of these differences were clinically meaningful (Table 2).

Table Graphic Jump LocationTable 2. Characteristics of 174062 CRUSADE Patients by Tertiles of Clinical Trial Participation
ADHERENCE TO EVIDENCE-BASED GUIDELINES

In general, the hospitals that participated in clinical trials had better adherence to each of the 9 individual recommendations that constituted our guideline adherence score and other indexes of the quality of care that we measured (Table 3). The overall median composite guideline adherence score was 78.4% (IQR, 73.5%-82.6%). There was a modest direct correlation between the proportion of patients enrolled in a trial by a hospital and composite guideline adherence score (Spearman rank correlation coefficient, 0.2; P < .001). Median guideline adherence scores increased across tertiles of trial participation: 76.9% (IQR, 71.8%-81.3%) for no-enrollment hospitals vs 78.3% (IQR, 73.2%-82.4%) for low-tertile hospitals vs 81.1% (IQR, 76.2%-84.1%) for high-tertile hospitals (adjusted P value for trend = .008) (Figure 1). In weighted multivariate analyses adjusted for hospital-level characteristics, low-tertile trial participation hospitals achieved 0.8% higher (95% confidence interval [CI], −0.9% to 2.6%; P = .36) adherence scores and high-tertile hospitals achieved 2.5% higher (95% CI, 0.5%-4.5%; P = .01) adherence scores than did no-enrollment hospitals. The 1.7% adjusted difference in adherence scores between the high- and low-tertile hospitals was also significant (95% CI, 0.7%-3.3%; P = .03).

Place holder to copy figure label and caption
Figure 1.

Adherence to evidence-based guideline recommendations at 494 CRUSADE (Can Rapid Risk Stratification of Unstable Angina Patients Suppress Adverse Outcomes With Early Implementation of the American College of Cardiology/American Heart Association Guidelines) hospitals, according to tertiles of hospital-level clinical trial participation. Tertiles of hospital-level trial participation include no participation tertile (0% of patients enrolled in trials) vs low trial participation tertile (1.0% enrolled [interquartile range, 0.5%-1.4%]) vs high trial participation tertile (4.9% enrolled [interquartile range, 3.5%-9.7%]). P = .36 for the difference between the no- and low-enrollment groups, P = .01 for the difference between the no- and high-enrollment groups, and P = .03 for the difference between the low- and high-enrollment groups. P values are weighted by the number of patients at each site and adjusted for all hospital-level variables presented in Table 1 (see the “Analysis” subsection of the “Methods” section).

Graphic Jump Location
Table Graphic Jump LocationTable 3. Measures of Processes of Care According to Tertiles of Clinical Trial Participation
MORTALITY

Overall, 6665 of 174 062(3.8%) patients died during hospitalization. There was a modest inverse correlation between the proportion of patients enrolled in a trial by a hospital and mortality (Spearman rank correlation coefficient, −0.2; P < .001). Mortality decreased across tertiles of trial participation: 5.9% for the no-enrollment hospitals vs 4.4% for the low-tertile hospitals vs 3.5% for the high-tertile hospitals (adjusted P value for trend = .003) (Figure 2). This relationship persisted and was statistically significant in multivariate analyses that accounted for hospital and patient characteristics. Compared with the no-enrollment hospitals, the adjusted odds of mortality in low-tertile hospitals was 0.9 (95% CI, 0.8-1.0; P = .04) and it was 0.8 (95% CI, 0.7-0.9; P = .003) in the high-tertile hospitals. The adjusted odds of mortality was also lower at the high-tertile trial participation hospitals when compared with the low-tertile hospitals (0.9; 95% CI, 0.8-1.0), although this was of borderline statistical significance (P = .06).

Place holder to copy figure label and caption
Figure 2.

Short-term mortality in 174 062 patients with acute coronary syndrome treated at 494 CRUSADE (Can Rapid Risk Stratification of Unstable Angina Patients Suppress Adverse Outcomes With Early Implementation of the American College of Cardiology/American Heart Association Guidelines) hospitals, according to tertiles of hospital-level clinical trial participation. Tertiles of hospital-level trial participation include no participation tertile (0% of patients enrolled in trials) vs low trial participation (1.0% enrolled [interquartile range, 0.5%-1.4%]) vs high trial participation tertile (4.9% enrolled [interquartile range, 3.5%-9.7%]). P = .04 for the difference between the no- and low-enrollment groups, P = .003 for the difference between the no- and high-enrollment groups, and P = .06 for the difference between the low- and high-enrollment groups. P values are adjusted for all hospital- and patient-level variables presented in Table 1 and Table 2, respectively (see the “Analysis” subsection of the “Methods” section).

Graphic Jump Location

In a large cohort of US hospitals treating acute coronary syndromes, we found that only 2.6% of patients were ever enrolled in a clinical trial for that condition and that over more than 4 years of observation, almost one-third of hospitals never included a patient in a clinical trial. We also found that hospitals that participated in trials seemed to deliver better quality of care and have significantly lower rates of early mortality when compared with hospitals that did not enroll patients in trials.

Our findings confirm and extend the only previous study that has looked carefully for evidence of a trial participation effect. In that study, Du Bois and colleagues11 examined the treatments and outcomes of 476 patients with ovarian cancer treated at 165 German hospitals. Compared with hospitals that did not participate in trials, they reported that participating hospitals had a 77% increase (P = .04) in guideline-concordant chemotherapy and a 28% reduction (P = .001) in long-term mortality.11

Although rarely studied empirically, a hospital trial participation effect has been invoked in the past as one of the reasons that patients treated at certain types of hospitals (ie, high-volume, research-intensive, academic centers with residency and fellowship training programs) seem to have better outcomes.7,8,1113 It has been speculated that trial investigators are prominent opinion leaders and physician advocates who gain early clinical experience with new drugs and devices and have many opportunities to interact with colleagues and exchange ideas regularly,15 that research personnel who are well trained and highly educated might develop additional capacity for training their colleagues outside the trial,7,8 and that the trial protocols for standard treatment might be so useful and effective that they are integrated into the delivery of usual care after the trial finishes.1113

Alternately, it is plausible that our findings could be a result of confounding by difficult to capture or unmeasured institutional characteristics that are tightly linked with a hospital's proclivity or ability to participate in trials. It could be that hospitals that foster and reward a culture of high-quality cardiac care (ie, large, high-volume, financially stable academic centers with cardiology training programs and capacity for revascularization) also attract and recruit the physicians who are most inclined to adopt new evidence; who are predisposed to implementing treatment protocols, decision support systems, and other interventions known to improve quality of care; who insist on thorough documentation of eligibility and contraindications data for each patient; and who willingly respond to credible audit and feedback. To the extent that we could, we adjusted for institutional characteristics that might capture some of these attributes (eg, size, caseload, location, revascularization capacity, teaching status, and cardiologist supervised care), controlled for a host of patient-level sociodemographic and clinical variables, accounted for the potential lack of statistical independence of individual patients treated within the same hospital, and still demonstrated an independent association between hospital trial participation and outcomes. Nevertheless, the potential for confounding does exist, and we acknowledge that we did not survey all 494 hospitals to ascertain the level of penetration of various quality improvement interventions nor did we use qualitative methods to describe institutional culture.1 This is certainly a worthwhile line of inquiry for future research, as would be determining if the trial participation effect is present for other conditions.

Some additional limitations need to be considered when interpreting our results. First, we did not have detailed hospital-level information regarding other possible mediators of a trial participation effect. For example, we do not have data regarding the amount of trial funding or additional research personnel brought in by investigators. Second, although we know that patients were enrolled in trials related to acute coronary syndromes, we do not know the specific trials in which they were involved. Third, we examined only hospitals that were volunteers in a quality improvement initiative, in and of itself a form of trial participation. CRUSADE sites were likely more highly motivated and interested in conducting research and improving quality of care than the average US hospital and might not be considered representative. This form of bias should lead to an underestimate in the magnitude of any trial participation effect (ie, hospitals that neither participate in trials nor quality improvement efforts might have even lower rates of guideline adherence and higher mortality).

In conclusion, patients treated at hospitals that participate in clinical trials seem to receive better quality of care and seem to have significantly better outcomes than patients treated at hospitals that do not participate in trials—at least in the setting of acute coronary syndrome. From the patient's perspective, even if our results are somewhat confounded, it still seems sensible to seek care at hospitals that participate in clinical trials. For policy makers and physicians, our findings should assuage some of the concerns related to the possible opportunity costs and potential downsides of participating in the clinical research enterprise.

Correspondence: Sumit R. Majumdar, MD, MPH, Department of Medicine, University of Alberta, 2E3.07 Walter Mackenzie Health Sciences Centre, University of Alberta Hospital, 8440-112th St, Edmonton, AB T6G 2B7, Canada (me2.majumdar@ualberta.ca).

Accepted for Publication: September 27, 2007.

Author Contributions:Study concept and design: Majumdar, Roe, Gibler, and Armstrong. Acquisition of data: Roe, Peterson, Chen, and Armstrong. Analysis and interpretation of data: Majumdar, Roe, Peterson, Chen, and Armstrong. Drafting of the manuscript: Majumdar and Armstrong. Critical revision of the manuscript for important intellectual content: Majumdar, Roe, Peterson, Chen, Gibler, and Armstrong. Statistical analysis: Majumdar, Peterson, and Chen. Obtained funding: Roe, Peterson, and Gibler. Administrative, technical, and material support: Roe, Peterson, Gibler, and Armstrong. Study supervision: Peterson and Armstrong.

Financial Disclosure: None reported.

Funding/Support: This study was supported by CRUSADE, a National Quality Improvement Initiative of the Duke Clinical Research Institute, which was funded by Schering Plough Corporation, Bristol Myers Squibb/Sanofi Aventis Pharmaceuticals Partnership, and Millenium Pharmaceuticals. Dr Majumdar receives salary support awards from the Alberta Heritage Foundation for Medical Research (Health Scholar) and the Canadian Institutes of Health Research (New Investigator).

Role of the Sponsor: The funding bodies had no role in the design and conduct of the study; in the collection, analysis, and interpretation of the data; or in the preparation, review, or approval of the manuscript.

Bradley  EHHolmboe  ESMattera  JARoumanis  SARadford  MJKrumholz  HM A qualitative study of increasing β-blocker use after myocardial infarction: why do some hospitals succeed? JAMA 2001;285 (20) 2604- 2611
PubMed Link to Article
Romano  PS Improving the quality of hospital care in America. N Engl J Med 2005;353 (3) 302- 304
PubMed Link to Article
O’Brien  SMPeterson  ED Identifying high-quality hospitals. Arch Intern Med 2007;167 (13) 1342- 1344
PubMed Link to Article
Peterson  EDRoe  MTMulgund  J  et al.  Association between hospital process performance and outcomes among patients with acute coronary syndromes. JAMA 2006;295 (16) 1912- 1920
PubMed Link to Article
Jha  AKOrav  EJLi  ZEpstein  AM The inverse relationship between mortality rates and performance in the Hospital Quality Alliance measures. Health Aff (Millwood) 2007;26 (4) 1104- 1110
PubMed Link to Article
Patel  MRChen  AYRoe  MT  et al.  A comparison of acute coronary syndrome care at academic and nonacademic hospitals. Am J Med 2007;120 (1) 40- 46
PubMed Link to Article
Prescott  RJCounsell  CEGillespie  WJ  et al.  Factors that limit the quality, number and progress of randomised controlled trials. Health Technol Assess 1999;3 (20) 1- 143
PubMed
Braunholtz  DAEdwards  SJLLilford  RJ Are randomized clinical trials good for us (in the short term)? evidence for a “trial effect.” J Clin Epidemiol 2001;54 (3) 217- 224
PubMed Link to Article
Tognoni  GFranzosi  MGGarattini  SMaggioni  A The case of GISSI in changing the attitudes of Italian cardiologists. Stat Med 1990;9 (1-2) 17- 27
PubMed Link to Article
Kandzari  DERoe  MTChen  AY  et al.  Influence of clinical trial enrollment on the quality of care and outcomes for patients with non–ST-segment elevation acute coronary syndromes. Am Heart J 2005;149 (3) 474- 481
PubMed Link to Article
Du Bois  ARochon  JLamparter  CPfisterer  J Pattern of care and impact of participation in clinical studies on the outcome in ovarian cancer. Int J Gynecol Cancer 2005;15 (2) 183- 191
PubMed Link to Article
Stiller  CA Centralised treatment, entry to trials and survival. Br J Cancer 1994;70 (2) 352- 362
PubMed Link to Article
Janni  WKiechle  MSommer  H  et al. ADEBAR Study Group, Study participation improves treatment strategies and individual patient care in participating centers. Anticancer Res 2006;26 ((5B)) 3661- 3668
PubMed
Zeger  SLLiang  KY Longitudinal data analysis for discrete and continuous outcomes. Biometrics 1986;42 (1) 121- 130
PubMed Link to Article
Majumdar  SRChang  WArmstrong  PW Do the investigative sites that take part in a positive clinical trial translate that evidence into practice? Am J Med 2002;113 (2) 140- 145
PubMed Link to Article

Figures

Place holder to copy figure label and caption
Figure 1.

Adherence to evidence-based guideline recommendations at 494 CRUSADE (Can Rapid Risk Stratification of Unstable Angina Patients Suppress Adverse Outcomes With Early Implementation of the American College of Cardiology/American Heart Association Guidelines) hospitals, according to tertiles of hospital-level clinical trial participation. Tertiles of hospital-level trial participation include no participation tertile (0% of patients enrolled in trials) vs low trial participation tertile (1.0% enrolled [interquartile range, 0.5%-1.4%]) vs high trial participation tertile (4.9% enrolled [interquartile range, 3.5%-9.7%]). P = .36 for the difference between the no- and low-enrollment groups, P = .01 for the difference between the no- and high-enrollment groups, and P = .03 for the difference between the low- and high-enrollment groups. P values are weighted by the number of patients at each site and adjusted for all hospital-level variables presented in Table 1 (see the “Analysis” subsection of the “Methods” section).

Graphic Jump Location
Place holder to copy figure label and caption
Figure 2.

Short-term mortality in 174 062 patients with acute coronary syndrome treated at 494 CRUSADE (Can Rapid Risk Stratification of Unstable Angina Patients Suppress Adverse Outcomes With Early Implementation of the American College of Cardiology/American Heart Association Guidelines) hospitals, according to tertiles of hospital-level clinical trial participation. Tertiles of hospital-level trial participation include no participation tertile (0% of patients enrolled in trials) vs low trial participation (1.0% enrolled [interquartile range, 0.5%-1.4%]) vs high trial participation tertile (4.9% enrolled [interquartile range, 3.5%-9.7%]). P = .04 for the difference between the no- and low-enrollment groups, P = .003 for the difference between the no- and high-enrollment groups, and P = .06 for the difference between the low- and high-enrollment groups. P values are adjusted for all hospital- and patient-level variables presented in Table 1 and Table 2, respectively (see the “Analysis” subsection of the “Methods” section).

Graphic Jump Location

Tables

Table Graphic Jump LocationTable 1. Characteristics of 494 CRUSADE Hospitals by Tertiles of Clinical Trial Participation
Table Graphic Jump LocationTable 2. Characteristics of 174062 CRUSADE Patients by Tertiles of Clinical Trial Participation
Table Graphic Jump LocationTable 3. Measures of Processes of Care According to Tertiles of Clinical Trial Participation

References

Bradley  EHHolmboe  ESMattera  JARoumanis  SARadford  MJKrumholz  HM A qualitative study of increasing β-blocker use after myocardial infarction: why do some hospitals succeed? JAMA 2001;285 (20) 2604- 2611
PubMed Link to Article
Romano  PS Improving the quality of hospital care in America. N Engl J Med 2005;353 (3) 302- 304
PubMed Link to Article
O’Brien  SMPeterson  ED Identifying high-quality hospitals. Arch Intern Med 2007;167 (13) 1342- 1344
PubMed Link to Article
Peterson  EDRoe  MTMulgund  J  et al.  Association between hospital process performance and outcomes among patients with acute coronary syndromes. JAMA 2006;295 (16) 1912- 1920
PubMed Link to Article
Jha  AKOrav  EJLi  ZEpstein  AM The inverse relationship between mortality rates and performance in the Hospital Quality Alliance measures. Health Aff (Millwood) 2007;26 (4) 1104- 1110
PubMed Link to Article
Patel  MRChen  AYRoe  MT  et al.  A comparison of acute coronary syndrome care at academic and nonacademic hospitals. Am J Med 2007;120 (1) 40- 46
PubMed Link to Article
Prescott  RJCounsell  CEGillespie  WJ  et al.  Factors that limit the quality, number and progress of randomised controlled trials. Health Technol Assess 1999;3 (20) 1- 143
PubMed
Braunholtz  DAEdwards  SJLLilford  RJ Are randomized clinical trials good for us (in the short term)? evidence for a “trial effect.” J Clin Epidemiol 2001;54 (3) 217- 224
PubMed Link to Article
Tognoni  GFranzosi  MGGarattini  SMaggioni  A The case of GISSI in changing the attitudes of Italian cardiologists. Stat Med 1990;9 (1-2) 17- 27
PubMed Link to Article
Kandzari  DERoe  MTChen  AY  et al.  Influence of clinical trial enrollment on the quality of care and outcomes for patients with non–ST-segment elevation acute coronary syndromes. Am Heart J 2005;149 (3) 474- 481
PubMed Link to Article
Du Bois  ARochon  JLamparter  CPfisterer  J Pattern of care and impact of participation in clinical studies on the outcome in ovarian cancer. Int J Gynecol Cancer 2005;15 (2) 183- 191
PubMed Link to Article
Stiller  CA Centralised treatment, entry to trials and survival. Br J Cancer 1994;70 (2) 352- 362
PubMed Link to Article
Janni  WKiechle  MSommer  H  et al. ADEBAR Study Group, Study participation improves treatment strategies and individual patient care in participating centers. Anticancer Res 2006;26 ((5B)) 3661- 3668
PubMed
Zeger  SLLiang  KY Longitudinal data analysis for discrete and continuous outcomes. Biometrics 1986;42 (1) 121- 130
PubMed Link to Article
Majumdar  SRChang  WArmstrong  PW Do the investigative sites that take part in a positive clinical trial translate that evidence into practice? Am J Med 2002;113 (2) 140- 145
PubMed Link to Article

Correspondence

CME
Meets CME requirements for:
Browse CME for all U.S. States
Accreditation Information
The American Medical Association is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The AMA designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 CreditTM per course. Physicians should claim only the credit commensurate with the extent of their participation in the activity. Physicians who complete the CME course and score at least 80% correct on the quiz are eligible for AMA PRA Category 1 CreditTM.
Note: You must get at least of the answers correct to pass this quiz.
You have not filled in all the answers to complete this quiz
The following questions were not answered:
Sorry, you have unsuccessfully completed this CME quiz with a score of
The following questions were not answered correctly:
Commitment to Change (optional):
Indicate what change(s) you will implement in your practice, if any, based on this CME course.
Your quiz results:
The filled radio buttons indicate your responses. The preferred responses are highlighted
For CME Course: A Proposed Model for Initial Assessment and Management of Acute Heart Failure Syndromes
Indicate what changes(s) you will implement in your practice, if any, based on this CME course.
Submit a Comment

Multimedia

Some tools below are only available to our subscribers or users with an online account.

Web of Science® Times Cited: 44

Related Content

Customize your page view by dragging & repositioning the boxes below.

Articles Related By Topic
Related Collections
PubMed Articles