How do academic health centers value and encourage clinical research?

To investigate whether there is a misalignment of the perceived values of and incentives for clinical research within U.S. academic health centers (AHCs), in 1999 the authors surveyed medical school deans, academic administrators, department chairs, and faculty members at 80 AHCs that are the members of the University HealthSystem Consortium, a not-for-profit consortium of AHCs. A total of 358 faculty from 58% of the institutions surveyed responded, with a mean of 3.76 responses/institution. There was general agreement that clinical research offers AHCs a considerable spectrum of benefits, including prestige, recruitment and retention of faculty, criteria for promotion of faculty, and financial support. Investigator-initiated research and government-funded research ranked highest in terms of their desirability compared with industry-sponsored and contract research. This preference was agreed upon across all categories of respondents and types of research (translational, clinical tests, and outcomes). Significant differences existed between the perceptions of deans/AHC administrators, who stated that they were increasing their emphasis on clinical investigation in the areas of research space (56% of responders), administrative support (81%), and patient recruitment (61%) and the perceptions of their departmental chairs/faculties in the same areas (34%, 52%, and 40%, respectively; p <.05). Faculty opinions documented few new investments in the actual infrastructure dedicated to clinical research. The authors conclude that their findings, which they consider reasonably representative, strongly suggest that the value of clinical research to AHCs is well understood. Their findings also identify important opportunities for AHCs to provide a wider range of incentives for the conduct of clinical research.

Using a multihospital systems framework to evaluate and establish drug use policy.

PURPOSE: In order to develop rational drug purchasing and use policy for a class of pharmaceuticals used in a consortium system of 14 university based hospitals, the antiemetic use patterns of inpatients receiving cancer chemotherapy were evaluated to assess the comparative effectiveness of granisetron, ondansetron, and conventional antiemetics. PATIENTS AND METHODS: A prospective, observational study was conducted in 14 academic health centers linked under research and purchasing consortium arrangements from October to December 1994. The use of antiemetics was evaluated in hospitalized patients receiving cancer chemotherapy agents with a known propensity for causing, alone or in combination, varying degrees of nausea or vomiting. Clinical outcomes measured were the impact of chemotherapy administration on the functional status of patients, and the occurrence of post-treatment vomiting. RESULTS: The most often prescribed cancer chemotherapy regimens consisted of cisplatin, paclitaxel, etoposide and cyclophosphamide, and the most often prescribed antiemetics were the 5-hydroxytryptamine subtype-3 antagonists (5-HT3 antagonists, granisetron and ondansetron), dexamethasone and lorazepam. Of the 439 patients studied, 329 (75%) reported no episodes of emesis. Of the patients receiving highly emetogenic chemotherapy, those receiving 5-HT3 antagonists experienced better overall outcomes (as measured by functional health status and the absence of vomiting) than patients receiving conventional (non-5-HT3 antagonist) antiemetics. In contrast, patients receiving chemotherapy associated with moderate or low emetogenicity experienced similar outcomes, regardless of the antiemetic regimen selected. No statistical difference was seen between granisetron and ondansetron in achieving positive patient outcomes. CONCLUSION: The study results suggest that 5-HT3 antagonists are associated with better clinical outcomes than other antiemetics in patients receiving highly emetogenic chemotherapy. Less costly conventional antiemetic therapy (or, in some cases, no antiemetic therapy) provide comparable outcomes in patients receiving chemotherapy associated with moderate or low emetogenic potential. Granisetron and ondansetron were found to be clinically comparable.

A randomized, crossover comparison of warfarin products in the treatment of chronic atrial fibrillation.

OBJECTIVE: To compare the dosing requirements and international normalized ratios (INRs) associated with two bioequivalent crystalline warfarin sodium products in patients with chronic atrial fibrillation. METHODS: A multicenter, single-blind (prescriber), randomized, crossover evaluation of Apothecon warfarin and DuPont warfarin (Coumadin) was conducted in consenting adults with chronic or paroxysmal atrial fibrillation who had been receiving DuPont warfarin chronically for the prevention of thromboembolism. Patients were randomly assigned to initially either continue DuPont warfarin or receive Apothecon warfarin for four weeks, with weekly evaluation of dosage and INR changes, safety, and efficacy. Subsequently, patients crossed over and received the other product for four weeks. RESULTS: There were 113 patients randomized to receive study treatment. Neither the propensity for a dosage change or an INR change nor the magnitude of a dosage change or INR change appeared related to a particular warfarin product (NS for each variable after each study period). After four weeks of treatment, the same number of patients (n = 7) experienced a > or = 20% change in warfarin dosage from the respective baseline with each product. The number of patients with INRs outside the desired protocol range after four weeks of treatment was similar for both groups (< 1.8, n = 9 for both products, or > 3.2, n = 9 for DuPont, n = 10 for Apothecon). No major hemorrhagic or thromboemoblic events occurred. CONCLUSIONS: The results of this study show that Apothecon warfarin and DuPont warfarin provide equivalent anticoagulation in patients with chronic or paroxysmal atrial fibrillation.

A multicenter drug use surveillance of intravenous immunoglobulin utilization in US academic health centers.

BACKGROUND: The role of intravenous immunoglobulin (IVIG) in treating a variety of diseases is controversial and under active investigation for at least two reasons: first, a severe shortage of IVIG products exists in the US; second, numerous off-label (not specified in the Food and Drug Administration [FDA]-approved label) uses for IVIG have been, and continue to be, described in the literature. However, most off-label uses are not supported by evidence from properly designed clinical trials. OBJECTIVE: To assess inpatient use of IVIG in a sample of US academic health centers and to compare it with published evidence-based model guidelines for IVIG use. METHODS: Data on the use of IVIG and subsequent clinical outcomes in 251 patients were collected prospectively from 12 institutions. Recommendations from consensus guidelines were used to categorize patients who received IVIG into one of four groups: labeled uses; off-label, recommended; off-label, recommended as alternative; and off-label, not recommended. Outcomes were scored according to guideline criteria. RESULTS: One hundred seven patients (43%) received IVIG for indications contained in the FDA-approved product label, 130 patients (52%) received IVIG for off-label indications, and 14 (5%) received undefined treatment. Among all patients administered IVIG, 31 (12%) were treated for off-label recommended reasons; 64 (26%) received off-label recommended as alternative therapy; and 35 (14%) received off-label not recommended therapy, as defined by model guidelines. Outcomes were not significantly different between the groups. CONCLUSIONS: Our findings suggest that IVIG continues to be used to treat a wide variety of conditions not specified in the product label. Furthermore, a substantial proportion of the reported off-label uses are not recommended according to evidence-based guidelines.

Clinical and economic outcomes of coronary angioplasty alone or in combination with stents in academic health centers: a retrospective database analysis.

OBJECTIVES: Indications for the use of stents are evolving and their optimal place in therapy remains to be defined. The objective was to compare the 1-year clinical and economic outcomes of percutaneous transluminal coronary agioplasty (PTCA) with and without the use of stents. METHODS: This retrospective database analysis was conducted using data from 60 US academic medical centers in 1996 and 1997. Outcomes of interest included in-hospital mortality (both on sentinel and readmission visits), readmission rates, revascularization procedures, length of stay, and the cost of sentinel and readmission hospitalization. RESULTS: A total of 27,020 patients either did (n = 13,254) or did not (n = 13,766) receive coronary stents in conjunction with PTCA. The median cost of hospitalization for the stent group was about dollar 1,409 higher (mean, dollar 1,200) than the no-stent group and the length of stay was similar (4.3 days vs. 4.5 days, respectively, P = .2628). Mortality rates on readmission (0.9% stent vs. 0.8% no-stent, P = NS) did not differ. However, patients in the stent group had better mortality rates compared to the no-stent group during the sentinel visit (1.7% vs. 2.7%, P < .001). Stent use was not associated with a significantly lower risk of a revascularization procedure on readmission to the same institution (OR 0.95, 95% CI 0.87-1.04, P = .28). CONCLUSIONS: Stents were associated with a 1% decrease of inpatient mortality during the sentinel visit without impacting length of stay or readmission rates. This advantage was associated with a dollar 1,409 higher median cost of hospitalization in the stent group.

Cost-offset analysis of aprotinin in high-risk coronary artery bypass.

Aprotinin, a naturally occurring protease inhibitor derived from bovine lung, is used prophylactically to minimize the amount of perioperative blood loss in patients undergoing coronary artery bypass graft surgery who are at high risk for excessive bleeding. A retrospective multicenter evaluation of aprotinin use was performed in high-risk coronary artery bypass graft patients treated either with aprotinin or according to usual-care to assess (1) differences in demographic and medical history characteristics, and (2) clinical and economic outcomes associated with their care. This study suggests that in many cases, the cost of aprotinin is offset by reductions in overall cost. Additional study is required to better understand this potential. In other cases, however, a more conservative approach to aprotinin use appears to be warranted.

Quality of care for patients hospitalized with heart failure at academic medical centers.

BACKGROUND: The purpose of this study was to determine the standard of care provided by academic medical centers for the management of congestive heart failure (CHF). METHODS AND RESULTS: The standard of care was estimated by assessing adherence to the treatment guidelines published by the US Agency for Health Care Policy and Research among 522 patients hospitalized at 7 university hospitals with a diagnosis of CHF. Data were abstracted by retrospective chart review. Of the 522 patients analyzed, 435 (83%) had a left ventricular ejection fraction (LVEF) measured or documented. Among these patients, 192 were considered "ideal" candidates for angiotensin-converting enzyme (ACE) inhibitor therapy (ie, with systolic dysfunction [LVEF <40%] and no contraindications to ACE inhibitors). In this cohort of "ideal" candidates, 138 (72%) were receiving ACE inhibitors at hospital discharge, including 60 (44%) who were prescribed doses recommended in large clinical trials. Compliance with patient education guidelines was assessed in all 487 patients who were alive at the time of discharge. Of these patients, 365 (75%) received dietary counseling, 404 (83%) were educated about exercise, 54 (11%) were instructed to follow daily weights, and 468 (96%) were counseled regarding medication compliance. Among the 87 smokers who were alive at time of discharge, 8 (9%) had documented advice to quit smoking. CONCLUSIONS: This study indicates that academic medical centers performed fairly well on the assessment of LVEF, the prescription of ACE inhibitors at discharge, and on education regarding diet, exercise, and compliance with medications. However, the results suggest opportunities for improvement in ACE inhibitor dosing and patient education regarding the importance of monitoring daily weights and smoking cessation.

Economic impact of digoxin toxicity.

The costs of digoxin toxicity to the US healthcare system have not been previously reported. Therefore, the 1994 database of US University Health-System Consortium (UHC) was searched for cases of digoxin toxicity using the International Classification of Diseases (9th edition) [ICD-9] codes. In addition, the medical records of 17 patients admitted to the University of Illinois Hospital from September 1994 to July 1995 with a diagnosis of digoxin toxicity were also reviewed. Of the 17 patients, 14 were admitted with a primary diagnosis of digoxin toxicity. Causes of digoxin toxicity were worsening renal function (6 patients), excessive dosage prescribed (4 patients), excessive dosage self-administered (2 patients), multiple prescriptions (2 patients), accidental ingestion (1 patient), drug-drug interaction (1 patient) and unknown (1 patient). Digoxin toxicity could have been prevented in 9 (53%) of the 17 patients. The mean length of stay in the hospital as a result of digoxin toxicity was 3.3 +/- 1.2 days. The mean laboratory cost associated with digoxin toxicity was $US275.54 +/- $US106.57 and the mean hospital bed cost was $US3781.92 +/- $US2572.22. The mean overall cost associated with digoxin toxicity was $US4087.05 +/- $US2659.76. There was a significant correlation between the total cost associated with digoxin toxicity and the serum digoxin concentration on admission (r = 0.73, p < 0.01). From the UHC database, a total of 836 cases of digoxin toxicity in 56 hospitals were identified. This represented the occurrence of digoxin toxicity in 0.07% of all patients admitted to these US academic hospitals. Digoxin toxicity results in considerable costs to the healthcare system. Most cases can be considered readily preventable with proper patient counselling and education.

Clinical characteristics and management of acute stroke in patients with atrial fibrillation admitted to US university hospitals.

UNLABELLED: The optimal evaluation and management of patients with atrial fibrillation who suffer an acute ischemic stroke remains controversial. METHODS: Medical records of 171 consecutive patients with atrial fibrillation and acute stroke at six U.S. university hospitals were reviewed. Data collected included the use of antithrombotic therapy, brain and cardiac imaging, bleeding complications, stroke risk factors, and contraindications to anticoagulation. RESULTS: Mean age was 75.4 years. Cardiovascular risk factors associated with increased stroke risk were present in 87%; 35% had at least one contraindication to anticoagulation. Half of the patients with stroke risk factors and no contraindications to anticoagulation were not receiving any antithrombotic therapy at the time of admission. Of the 22 patients who were treated with warfarin, and had INR values on admission, 16 had levels of < 2.0; only six had INR values between 2.0 and 3.0. Transthoracic echocardiography was performed in 107 patients (63%); intracardiac thrombi were visualized in only 5%. Initial brain imaging revealed hemorrhagic transformation in nine. Heparin was used in 93 patients (54%), usually within 48 hours of stroke onset. Patients who received delayed heparin typically did not have repeat brain imaging prior to starting heparin. One patient had a delayed symptomatic cerebral hemorrhage. Of the survivors, 47% were discharged and treated with warfarin (or warfarin plus aspirin), 28% with ASA, 7% with other antithrombotic therapies, and 18% with no antithrombotic therapy. CONCLUSION: Antithrombotic therapy was underutilized and inadequately monitored in atrial fibrillation patients prior to stroke onset. After hospital admission, a wide range of diagnostic and management strategies, which often did not follow current recommendations, were employed.

Anticoagulant therapy monitoring with international normalized ratio at US academic health centers.

OBJECTIVE: To assess the extent of incorporation of international normalized ratio (INR) reporting in US academic hospitals. DESIGN: Survey of academic hospital clinical laboratories in January 1995. SETTING/PARTICIPANTS: Fifty-eight academic hospital clinical laboratories at institutions that are members of the University Health System Consortium. MAIN OUTCOME MEASURES: The methods for monitoring oral anticoagulant therapy at the surveyed laboratories were determined. The extent of reporting of prothrombin time (PT), PT ratio, INR, and INR therapeutic range was determined. RESULTS: All 58 of the responding hospital clinical laboratories reported INR in patients receiving oral anticoagulation. The median length of time that hospitals had been reporting INR was 24 months (range 3-108). A majority of hospitals continued to report PT values (95%) and PT reference ranges (93%) in addition to INR. Therapeutic INR ranges were reported by only 25 of the laboratories (43%). Of those that report INR ranges, many follow the published recommendations by the American College of Chest Physicians and the Food and Drug Administration. A majority of the hospitals (79%) do not confirm the accuracy of the international sensitivity index (ISI) for their own analyzers. CONCLUSIONS: Contrary to previous reports, academic hospital clinical laboratories have now adopted the more accurate system of reporting INR values in addition to PT values in patients receiving oral anticoagulation. However, better reporting of INR ranges, use of more sensitive thromboplastins, and confirmation of the accuracy of the ISI for local analyses would further improve the monitoring of oral anticoagulation.

Status of antithrombotic therapy for patients with atrial fibrillation in university hospitals.

BACKGROUND: The risk of stroke in patients with atrial fibrillation can be significantly reduced with antithrombotic therapy. Despite this, many physicians remain hesitant to prescribe warfarin sodium or aspirin therapy for patients with atrial fibrillation. OBJECTIVE: To assess the use of antithrombotic therapy in patients with atrial fibrillation at 6 academic hospitals in the United States. METHODS: Records were reviewed from consecutive hospital admissions of 309 patients with atrial fibrillation at 6 members of the University Health System Consortium, Oak Brook, III, which is a member driven alliance of 70 academic health centers in the United States. Risk factors for stroke, contraindications to anticoagulant therapy, and use of antithrombotic therapy at admission and discharge were recorded. RESULTS: The mean age of patients was 71.6 years, 54% had chronic, 22% paroxysmal, and 24% new-onset atrial fibrillation. Eighty-two percent of the patients had cardiovascular risk factors that have been associated with increased risk of stroke. At least 1 relative contraindication to anticoagulant therapy was present in 44%. At the time of admission. 32% of the patients with previously diagnosed atrial fibrillation (n = 235) were receiving warfarin (or warfarin plus aspirin), 31% were receiving aspirin alone, and 36% were receiving no antithrombotic therapy. At discharge (n = 230), 41% of these patients were taking warfarin (or warfarin plus aspirin) and 36% were taking aspirin. Forty-four percent of the patients with risk factors for stroke and no contraindications to anticoagulation (n = 134) were discharged on a regimen of warfarin (or warfarin plus aspirin), 34% were discharged on a regimen of aspirin, and 22% received no antithrombotic therapy. CONCLUSIONS: About half of the patients with atrial fibrillation admitted to these academic hospitals had clinical risk factors that are associated with increased risk of stroke and no contraindications to anticoagulation. Antithrombotic therapy was underused in these patients.

Pharmacologic profile of survivors of acute myocardial infarction at United States academic hospitals.

Optimal drug therapy for patients with acute myocardial infarction (AMI) is well described in the medical literature. However, data on the actual pharmacologic management of patients surviving AMI at academic hospitals is unavailable. The purpose of this study was to document treatment profiles in 500 patients surviving AMI at 12 academic hospitals in the United States. These profiles were compared with established guidelines and were evaluated for trends. Overall, thrombolytics (streptokinase > or = tissue-type plasminogen activator) were administered in 29% of the patients, with a greater proportion of patients receiving beta-blockers than calcium channel antagonists in the initial 72 hours (61% vs 40%; p < 0.005) and at discharge (51% vs 35%; p < 0.005). Further, women were less likely than men to receive thrombolytic therapy (odds ratio [OR] = 0.61; confidence interval [CI], 0.54 to 0.69) or beta-blocker therapy within the first 72 hours (OR = 0.61; CI, 0.55 to 0.67) or at hospital discharge (OR = 0.53; CI, 0.48 to 0.58). Overall, improvements could still be made in the number of patients who receive thrombolytic and acute and chronic beta-blocker therapies after AMI, particularly in women. Changes in treatment profiles may be a reflection of the publication of large clinical trials.

Albumin and nonprotein colloid solution use in US academic health centers.

BACKGROUND: Crystalloids, nonprotein colloids (NPCs), and albumin are used for many indications. The use of the least costly agent in situations where these products are clinically interchangeable can reduce health care costs. OBJECTIVES: To characterize the prescribing of albumin and NPC. To evaluate the appropriateness and cost implications of their use. METHODS: An observational study conducted in 15 academic health centers from April 11 through May 6, 1994, to assess the appropriateness of albumin and NPC use, based on "model" consensus-derived indication guidelines. RESULTS: A total of 969 case report forms were evaluated. Albumin and NPCs were administered in 83% and 17% of the cases, respectively. Albumin and NPCs were administered mostly in the intensive care (50%) or operating room (31%) settings. The most common prescribers of these products were surgeons (45%) and anesthesiologists (20%). In 87% of cases, albumin or NPC was administered to reach a defined end point (eg, to achieve a target physiological state or to resolve a pathophysiological condition). Only one albumin recipient experienced an adverse event; no adverse events were noted with NPC administration. Approximately $203,000 was spent on albumin and NPC therapy for the 969 cases; $49,702 (24%) was spent on appropriate administrations, $124,939 (62%) on inappropriate administrations, and $28,014 (14%) on unevaluated indications. CONCLUSIONS: Evaluated against model guidelines, most of the albumin and NPC use in the study was found to be inappropriate. The need for institutions to define and implement guidelines that focus on the cost-efficient use of these agents is recommended in an increasingly cost-conscious health care environment.