Evaluation of Complication Rates after Coronary Artery Bypass Surgery using Administrative Data

Our objectives were (1) to determine if studying hospital complication rates after coronary artery bypass graft (CABG) surgery provides information not available when only mortality is studied, and (2) to reexplore the utility of ICD-9-CM administrative data for CABG outcomes assessment. Using data from Massachusetts, we identified CABG cohorts from 1990 and 1992 to respectively develop and validate multivariate risk adjustment models predicting in-hospital mortality and complications. The resulting models had good discrimination and calibration. In 1992, adjusted hospital complication rates ranged widely from 13.0% to 57.6%, while mortality rates ranged from 1.4% to 6.1 %. Hospitals with high complication rates tended to have high mortality (r = 0.74, P = 0.006), but 2 of the 12 hospitals studied ranked quite differently when judged by complications rather than mortality. We conclude that (1) complications after CABG occur frequently and may provide information about hospital quality beyond that obtained from hospital mortality rates, and that (2) administrative data continue to be a promising resource for outcomes research.

Capturing diagnosis-timing in ICD-coded hospital data: recommendations from the WHO ICD-11 topic advisory group on quality and safety

PURPOSE

To develop a consensus opinion regarding capturing diagnosis-timing in coded hospital data.

METHODS

As part of the World Health Organization International Classification of Diseases-11th Revision initiative, the Quality and Safety Topic Advisory Group is charged with enhancing the capture of quality and patient safety information in morbidity data sets. One such feature is a diagnosis-timing flag. The Group has undertaken a narrative literature review, scanned national experiences focusing on countries currently using timing flags, and held a series of meetings to derive formal recommendations regarding diagnosis-timing reporting.

RESULTS

The completeness of diagnosis-timing reporting continues to improve with experience and use; studies indicate that it enhances risk-adjustment and may have a substantial impact on hospital performance estimates, especially for conditions/procedures that involve acutely ill patients. However, studies suggest that its reliability varies, is better for surgical than medical patients (kappa in hip fracture patients of 0.7-1.0 versus kappa in pneumonia of 0.2-0.6) and is dependent on coder training and setting. It may allow simpler and more precise specification of quality indicators.

CONCLUSIONS

As the evidence indicates that a diagnosis-timing flag improves the ability of routinely collected, coded hospital data to support outcomes research and the development of quality and safety indicators, the Group recommends that a classification of 'arising after admission' (yes/no), with permitted designations of 'unknown or clinically undetermined', will facilitate coding while providing flexibility when there is uncertainty. Clear coding standards and guidelines with ongoing coder education will be necessary to ensure reliability of the diagnosis-timing flag.

International variation in the definition of 'main condition' in ICD-coded health data

Hospital-based medical records are abstracted to create International Classification of Disease (ICD) coded discharge health data in many countries. The 'main condition' is not defined in a consistent manner internationally. Some countries employ a 'reason for admission' rule as the basis for the main condition, while other countries employ a 'resource use' rule. A few countries have recently transitioned from one of these approaches to the other. The definition of 'main condition' in such ICD data matters when it is used to define a disease cohort to assign diagnosis-related groups and to perform risk adjustment. We propose a method of harmonizing the international definition to enable researchers and international organizations using ICD-coded health data to aggregate or compare hospital care and outcomes across countries in a consistent manner. Inter-observer reliability of alternative harmonization approaches should be evaluated before finalizing the definition and adopting it worldwide.

Clinical and socio-demographic factors associated with diabetic ketoacidosis hospitalization in adults with Type 1 diabetes

AIM

To identify the clinical and socio-demographic factors associated with hospitalization for diabetic ketoacidosis in adults with Type 1 diabetes.

METHODS

We combined clinical and administrative health data from a large urban diabetes clinic to perform a data linkage study. We identified adults (aged ≥ 18 years old) with Type 1 diabetes and linked to hospital discharge abstracts to assess for diabetic ketoacidosis hospitalization. The study period was between 1 January 2004 and 31 December 2009, with all individuals living in the same geographic area. Multivariate logistic regression was used to identify potential predictors of diabetic ketoacidosis hospitalization.

RESULTS

We identified 255 individuals with a diabetic ketoacidosis hospitalization and 1739 without a diabetic ketoacidosis hospitalization. Mean (standard deviation) age was 40.0 (15.8) years, 51.7% were men and mean duration of diabetes was 17.8 (12.9) years. Diabetic ketoacidosis hospitalization was associated with shorter duration of diabetes (odds ratio 0.96 per year; 95% confidence interval 0.95-0.98), gastroparesis (odds ratio 4.13; 95% confidence interval 1.82-9.35), psychiatric diagnosis (odds ratio 1.98; 95% confidence interval 1.22-3.19), and higher HbA1c (odds ratio 1.25 per 1% increase; 95% confidence interval 1.16-1.35).

CONCLUSIONS

This study identifies specific clinical factors associated with diabetic ketoacidosis hospitalization in adults with Type 1 diabetes. This information can help to inform the detection of high-risk patients, for whom special attention and interventions may be warranted to prevent diabetic ketoacidosis.

Understanding coronary artery disease patients' decisions regarding the use of chelation therapy for coronary artery disease: descriptive decision modeling

BACKGROUND

A considerable number of patients receive chelation therapy to treat their coronary artery disease. However, there is no current empirical evidence to support its use.

AIM

To better understand patient's decision-making processes regarding the use of chelation therapy as a treatment for coronary artery disease.

METHODS

Based on qualitative interviews with 32 coronary artery disease patients, a taxonomy of decision-related issues, hierarchical decision-model, and survey based on the model were developed. The model was then pilot tested with another group of 30 patients and revised accordingly. The final model was tested with another group of 167 patients (27 current users, 72 previous users, and 68 never users of chelation therapy). The primary examination of the model was to determine the degree to which it successfully identified people who fell within each behavioral group. This was done by dividing the total number of successes by the total number of cases on all paths (or questions in the questionnaire).

RESULTS

The most important elements in the decision to use or not use chelation therapy were: previous experience with or learning about chelation therapy, openness to alternative treatments, satisfaction with current level of (traditional) care, physician opinion regarding chelation therapy, costs associated with chelation therapy, perceived access to chelation therapy provider, current state of health (good or bad), and wanting to do 'all one can' for heart health. When tested, the ability of the model to predict the appropriate outcome was nearly 93%. The most salient junctures in the model that led participants to different behavioral outcomes were: considering using non-traditional treatments; perceptions regarding potential risks and benefits; cost; and believing that using chelation therapy was 'doing all that they can' to help their heart health.

CONCLUSIONS

Descriptive decision-modeling is a useful method to depict cardiac patients' decision-making concerning the use of chelation therapy. It can also assist healthcare providers and policy makers in directing interventions and policy aimed at enhancing the use of evidence-based therapies for cardiac patients.

Audit of the consultation process on general internal medicine services

OBJECTIVE

To determine the proportion of consultations requested by general internal medicine services that communicate key components of the consultation process to medical subspecialists.

DESIGN

Retrospective chart review by two researchers, using a standardised chart abstraction instrument (93.1% agreement, kappa 0.85).

SETTING

Calgary, Alberta, Canada.

SAMPLE

A random sample of medical consultations was selected from those generated on two medical teaching units (MTUs) from 2003 to 2004.

MEASUREMENTS

The primary measure of interest was whether a "clear clinical question" was posed to the subspecialist, a binary variable.

RESULTS

Two hundred consultations were sampled from the 2885 subspecialty consultations. Of the selected consultations, 94.0% (188/200) were available for review. A clear clinical question was posed in 69.7% (131/188) of consultations (CI 0.63 to 0.74). In a secondary analysis involving a larger sample permitting comparison across subspecialties, 95.1% (368/387) of the consultations, representative of the subspecialties, were available for review. An MTU member contacted the subspecialist for 74.2% of consultations. If a consultation was urgent, a member of the MTU contacted the subspecialist in 81.0% of consultations. Of these urgent consultations, 63.3% had a clear clinical question.

CONCLUSION

More than one in four consultations does not contain a clear clinical question, illustrating suboptimal communication between physicians. Innovative strategies that provide a sustainable solution for overcoming barriers to communication could have a significant impact on quality of care.

Recurrent cerebral ischemia in medically treated patent foramen ovale: a meta-analysis

BACKGROUND

Among patients with a patent foramen ovale (PFO) and a prior cryptogenic ischemic stroke or TIA, the absolute and relative risk of recurrent events is unclear.

METHODS

We conducted a systematic review and meta-analysis of clinical studies in any language published up to February 2008. We included studies reporting original data on recurrent cerebrovascular events in patients with prior cryptogenic stroke or TIA and PFO. Two authors independently extracted data and evaluated study quality.

RESULTS

We identified 15 eligible studies, four with a non-PFO comparison group. In these four studies, the pooled relative risk (RR) for recurrent ischemic stroke or TIA in patients with vs without a PFO was 1.1 (95% confidence interval [CI] 0.8 to 1.5). For ischemic stroke, the pooled RR was 0.8 (95% CI 0.5 to 1.3). We tabulated the absolute rate of recurrent events in all 15 studies. The pooled absolute rate of recurrent ischemic stroke or TIA in patients with PFO was 4.0 events per 100 person-years (95% CI 3.0 to 5.1) while the rate of recurrent ischemic stroke was 1.6 events per 100 person-years (95% CI 1.1 to 2.1).

CONCLUSIONS

In medically treated patients with prior cryptogenic stroke, while the absolute rate of recurrent events is variable, available evidence does not support an increased relative risk of recurrent ischemic events in those with vs without a patent foramen ovale. Patent foramen ovale closure in these patients cannot be recommended until the results of ongoing clinical trials are reported.

Predicting patient complaints in hospital settings

BACKGROUND

The prediction of patient complaints is not clearly understood. This is important in so far as patient complaints have been shown to correlate with other adverse outcomes of interest in acute care facilities.

OBJECTIVES

To evaluate the complexity of the patient case and patient safety culture as predictors of patient complaints.

DESIGN

A matched case-control analysis of data from patients filing complaints (cases) and matched patients who did not file complaints (controls) in 2005. Staff surveys were used to measure the Patient Safety Culture on individual units.

SETTING

45 inpatient acute care units from four general hospitals in a large metropolitan centre in western Canada.

SAMPLE

586 patients registering complaints in 2005.

METHOD

The primary outcome was patient complaints (number and type). Predictors included unit-level measures of patient safety culture based on a survey and patient admission characteristics (including age, gender, treatment unit, primary diagnosis, case resource intensity).

RESULTS

The probability of a patient complaint was positively associated with cases of higher complexity (beta = 0.145, p = 0.032; odds ratio = 1.16; CI 0.994 to 1.344). The culture of patient safety within hospital units was not related to the probability of complaints within a given unit.

CONCLUSIONS

Patient complaints are associated with higher clinical complexity. However, the confidence interval around the odds ratio for this association just crosses 1.0 and is thus not "significant" in a traditional framework of dichotomously judging statistical significance at the 95% confidence level. The lack of association with a unit's safety culture, meanwhile, implies that the non-modifiable clinical complexity factor is a more important determinant of patient complaints.

Derivation and validation of a clinical index for prediction of rapid progression of kidney dysfunction

BACKGROUND

Chronic kidney disease is common among the elderly, and these patients are at risk of progressive kidney dysfunction.

AIM

To develop an index to predict rapid progression of kidney dysfunction.

DESIGN

Community-based cohort divided into derivation (n = 6789) and validation (n = 3395) subsets.

METHODS

We identified 10 184 subjects aged >/=66 years from computerized laboratory data. Prescription drug data was used to define disease categories and medication exposure, and an index for predicting rapid progression of kidney dysfunction (> or =25% decline in glomerular filtration rate over a 2-year period) was obtained from a logistic regression model in the derivation cohort. The risk score for each subject was calculated by summing the component variables together, which were subsequently categorized into five risk classes.

RESULTS

Five predictors of rapid progression were identified: age >75 years, cardiac disease, diabetes mellitus, gout, and use of anti-emetic medications. Rates of rapid progression for risk classes I through V were 8.6%, 10.9%, 13.9%, 15.6%, and 24.1%, respectively, for the derivation cohort, and 8.4%, 11.6%, 15.5%, 17.3%, 21.9%, respectively, for the validation cohort. The risk index distinguished between low and high risk of rapid progression, with a 2.5-fold greater risk for the highest, compared to the lowest, risk decile.

DISCUSSION

Readily available clinical data can be used to identify most elderly at risk of rapid progression of kidney dysfunction. This simple index could help clinicians to identify patients at risk, and implement strategies to slow the progression of kidney dysfunction.

Progression of kidney dysfunction in the community-dwelling elderly

Despite the high prevalence of chronic kidney disease among the elderly, few studies have described their loss of kidney function. We sought to determine the progression of kidney dysfunction among a community-based cohort of elderly subjects. The cohort included 10 184 subjects 66 years of age or older, who had one or more outpatient serum creatinine measurements during each of two time periods: 1 July to 31 December 2001 and 1 July to 31 December 2003. A mixed effects model, including covariates for age, gender, diabetes mellitus, and comorbidity, was used to determine the rate of decline in estimated glomerular filtration rate (eGFR, in ml/min/1.73 m2) per year over a median follow-up of 2.0 years. Subjects with diabetes mellitus had the greatest decline in eGFR of 2.1 (95% CI 1.8-2.5) and 2.7 (95% CI 2.3-3.1) ml/min/1.73 m2 per year in women and men, respectively. The rate of decline for women and men without diabetes mellitus was 0.8 (95% CI 0.6-1.0) and 1.4 (95% CI 1.2-1.6) ml/min/1.73 m2 per year. Subjects with a study mean eGFR<30 ml/min/1.73 m2, both those with and without diabetes mellitus, experienced the greatest decline in eGFR. In conclusion, we found that the majority of elderly subjects have no or minimal progression of kidney disease over 2 years. Strategies aimed at slowing progression of kidney disease should consider underlying risk factors for progression and the negligible loss of kidney function that occurs in the majority of older adults.

Simultaneous carotid endarterectomy and coronary artery bypass surgery in Canada

Stroke neurologists are commonly asked to review patients who require coronary artery bypass grafting (CABG) but who also have comorbid severe carotid stenosis; such patients may be offered simultaneous carotid endarterectomy (CEA). In Canada, 0.51% of CABG procedures were combined CEA-CABG. The adjusted stroke and death rate was 2.67-fold greater in the combined CEA-CABG group compared to CABG alone. Randomized trials of the combined procedure are needed.

Quality of life in patients treated with hemodialysis or peritoneal dialysis: What are the important determinants?

BACKGROUND

Patients with end-stage renal disease (ESRD) have significant impairments in health-related quality of life (HRQOL). In part, this is due to the intrusiveness of the treatment (hemodialysis or peritoneal dialysis) that is required. It is unclear whether hemodialysis or peritoneal dialysis is associated with a higher HRQOL.

METHODS

192 prevalent patients who self-selected treatment with hemodialysis (either in-center, satellite or home/self-care hemodialysis) or peritoneal dialysis were studied to determine whether treatment with hemodialysis or peritoneal dialysis is associated with a higher HRQOL. Demographic, laboratory and clinical information (including the presence of comorbid conditions using the Charlson comorbidity index) was assessed at baseline. The outcome of interest was HRQOL, which was measured using the Kidney Disease Quality of Life-Short Form (KDQOL-SF), the Short-Form 36 (SF-36) and the EuroQol EQ-5D at baseline and after 6 and 12 months of follow-up.

RESULTS

There was no significant difference in HRQOL scores for the SF-36, the EQ-5D and for 9 of 11 KDQOL dimensions for patients treated with hemodialysis or peritoneal dialysis at baseline. As expected, HRQOL was significantly lower for patients who had more comorbid disease, required assistance with their daily care, and for patients with less than a grade 12 education. After controlling for the effect of other important variables, HRQOL (as measured by the EQ-5D visual analog or index scores) did not differ between hemodialysis and peritoneal dialysis patients. HRQOL was stable over time, both for patients who started on hemodialysis or peritoneal dialysis.

CONCLUSIONS

There is no significant difference in HRQOL for prevalent ESRD patients treated with hemodialysis or peritoneal dialysis. It will be important to determine if this finding holds true for incident patients treated with hemodialysis or peritoneal dialysis.

Use of chelation therapy after coronary angiography

PURPOSE

Among patients who had undergone coronary angiography, we sought to determine the proportion of chelation therapy users, their sociodemographic and clinical characteristics, and the association of chelation therapy with subsequent revascularization.

METHODS

We studied all patients who underwent coronary angiography in the province of Alberta, Canada, during 1995 and 1996. The cohort was followed for up to 6 years to determine subsequent revascularization status. Use of chelation therapy was determined by a mailed survey 1 year after angiography.

RESULTS

Among the 5854 patients who responded to the mail survey (70% response rate), 210 (3.6%) reported current use of chelation therapy and 252 (4.3%) reported past use. Current use of chelation therapy was associated with extensive coronary artery disease (adjusted odds ratio [OR] = 3.3; 95% confidence interval [CI]: 1.9 to 5.7 for 3-vessel disease; and OR = 2.7; 95% CI: 1.2 to 6.0 for left main disease, as compared with those with normal anatomy) and the absence of diabetes (OR = 0.6; 95% CI: 0.4 to 0.9). Current users were less likely to have undergone percutaneous transluminal coronary angioplasty (OR = 0.7; 95% CI: 0.5 to 0.9) and coronary artery bypass graft (CABG) surgery (OR = 0.3; 95% CI: 0.2 to 0.5) in the first year after angiography, but were as likely as nonusers of chelation therapy to have undergone CABG surgery in the subsequent 3- to 5-year period (adjusted hazard ratio [HR] = 1.1; 95% CI: 0.7 to 1.9). Past use of chelation therapy was associated with a history of CABG surgery before coronary angiography (OR = 1.6; 95% CI: 1.1 to 2.3) and extensive coronary artery disease. Past users were also more likely to have undergone CABG surgery in the follow-up period (HR = 1.7; 95% CI: 1.1 to 2.6).

CONCLUSIONS

About 8% of patients who underwent cardiac catheterization for coronary artery disease were using or had previously tried chelation therapy. Users may have foregone revascularization in favor of this less invasive yet unproven treatment, with some users subsequently undergoing conventional treatment after chelation. Alternatively, some patients may have turned to chelation as a "last resort" after having been judged unsuitable for revascularization.

Extended out-of-hospital low-molecular-weight heparin prophylaxis against deep venous thrombosis in patients after elective hip arthroplasty: a systematic review

PURPOSE

Evidence-based medicine guidelines based on venographic end points recommend in-hospital prophylaxis with low-molecular-weight heparin (LMWH) in patients having elective hip surgery. Emerging data suggest that out-of-hospital use may offer additional protection; however, uncertainty remains about the risk-benefit ratio. To provide clinicians with a practical pathway for translating clinical research into practice, we systematically reviewed trials comparing extended out-of-hospital LMWH prophylaxis versus placebo.

DATA SOURCES

Studies were identified by 1) searching PubMed, MEDLINE, and the Cochrane Library Database for reports published from January 1976 to May 2001; 2) reviewing references from retrieved articles; 3) scanning abstracts from conference proceedings; and 4) contacting pharmaceutical companies and investigators of the original reports.

STUDY SELECTION

Randomized, controlled trials comparing extended out-of-hospital prophylaxis with LMWH versus placebo in patients having elective hip arthroplasty.

DATA EXTRACTION

Two reviewers extracted data independently. Reviewers evaluated study quality by using a validated four-item instrument.

DATA SYNTHESIS

Six of seven original articles met the defined inclusion criteria. The included studies were double-blind trials that used proper randomization procedures. Compared with placebo, extended out-of-hospital prophylaxis decreased the frequency of all episodes of deep venous thrombosis (placebo rate, 150 of 666 patients [22.5%]; relative risk, 0.41 [95% CI, 0.32 to 0.54; P < 0.001]), proximal venous thrombosis (placebo rate, 76 of 678 patients [11.2%]; relative risk, 0.31 [CI, 0.20 to 0.47; P < 0.001]), and symptomatic venous thromboembolism (placebo rate, 36 of 862 patients [4.2%]; relative risk, 0.36 [CI, 0.20 to 0.67; P = 0.001]). Major bleeding was rare, occurring in only one patient in the placebo group.

CONCLUSIONS

Extended LMWH prophylaxis showed consistent effectiveness and safety in the trials (regardless of study variations in clinical practice and length of hospital stay) for venographic deep venous thrombosis and symptomatic venous thromboembolism. The aggregate findings support the need for extended out-of-hospital prophylaxis in patients undergoing hip arthroplasty surgery.

Geographic variation in the rate of carotid endarterectomy in Canada

Background and Purpose-- Carotid endarterectomy (CEA) is an important method of stroke prevention, but its usage in Canada is not well known. The indications for CEA have been well informed by the recent clinical trials, but the impact of this information on the rate and regional variation in the rate of CEA is unknown. This study sought to determine the rate and the regional variation in the rate of CEA in Canada, its provinces, and census divisions for 1994-1997.

METHODS

Discharge data from all hospitals in Canada except Quebec were obtained from the Canadian Institute for Health Information for 1994-1997 and were searched for CEA by residential site. Rates and variations in rates were calculated.

RESULTS

The national age- and sex-adjusted rate per 100 000 people of CEA for those aged >/=40 years rose from 31.7 in 1994 to 40.5 in 1997. Provincial rates in 1997 varied from a low of 25.7 in Saskatchewan to high of 82.8 in Prince Edward Island. The census division rates varied even more, from a low of 0 in several divisions to a high of 179.

CONCLUSIONS

The recent slight increase in CEA rates may reflect the release of new efficacy results for CEA, especially for asymptomatic carotid stenosis, but the rates are still far below US levels. The marked regional variation in rates may reflect differing views on the appropriateness of indications such as asymptomatic carotid stenosis for CEA and the inconsistency of published clinical practice guidelines.

Comparison of 2 methods for calculating adjusted survival curves from proportional hazards models

CONTEXT

Adjusted survival curves are often presented in medical research articles. The most commonly used method for calculating such curves is the mean of covariates method, in which average values of covariates are entered into a proportional hazards regression equation. Use of this method is widespread despite published concerns regarding the validity of resulting curves.

OBJECTIVE

To compare the mean of covariates method to the less widely used corrected group prognosis method in an analysis evaluating survival in patients with and without diabetes. In the latter method, a survival curve is calculated for each level of covariates, after which an average survival curve is calculated as a weighted average of the survival curves for each level of covariates.

DESIGN, SETTING, AND PATIENTS

Analysis of cohort study data from 11 468 Alberta residents undergoing cardiac catheterization between January 1, 1995, and December 31, 1996.

MAIN OUTCOME MEASURES

Crude and risk-adjusted survival for up to 3 years after cardiac catheterization in patients with vs without diabetes, analyzed by the mean of covariates method vs the corrected group prognosis method.

RESULTS

According to the mean of covariates method, adjusted survival at 1044 days was 94.1% and 94.9% for patients with and without diabetes, respectively, with misleading adjusted survival curves that fell above the unadjusted curves. With the corrected group prognosis method, the corresponding survival values were 91.3% and 92.4%, with curves that fell more appropriately between the unadjusted curves.

CONCLUSIONS

Misleading adjusted survival curves resulted from using the mean of covariates method of analysis for our data. We recommend using the corrected group prognosis method for calculating risk-adjusted curves.

Timing of initial administration of low-molecular-weight heparin prophylaxis against deep vein thrombosis in patients following elective hip arthroplasty: a systematic review

BACKGROUND

Perioperative and postoperative venous thrombosis are common in patients undergoing elective hip surgery. Prophylactic regimens include subcutaneous low-molecular-weight heparin 12 hours or more before or after surgery and oral anticoagulants. Recent clinical trials suggest that low-molecular-weight heparin initiated in closer proximity to surgery is more effective than the present clinical practice. We performed a systematic review of the literature to assess the efficacy and safety of low-molecular-weight heparin administered at different times in relation to surgery vs oral anticoagulant prophylaxis.

METHODS

Reviewers (A.F.M. and S.M.M.) identified studies by searching MEDLINE, reviewing references from retrieved articles, scanning abstracts from conference proceedings, and contacting investigators and pharmaceutical companies. Randomized trials comparing low-molecular-weight heparin administered at different times relative to surgery with oral anticoagulants in patients undergoing elective hip arthroplasty, evaluated using contrast phlebography, were selected. Two reviewers (A.F.M. and S.M.M.) extracted data independently.

RESULTS

The literature review identified 4 randomized trials meeting predefined inclusion criteria. The results indicate that low-molecular-weight heparin initiated in close proximity to surgery resulted in absolute risk reductions of 11% to 13% for deep vein thrombosis, corresponding to relative risk reductions of 43% to 55% compared with oral anticoagulants. Low-molecular-weight heparin initiated 12 hours before surgery or 12 to 24 hours postoperatively was not more effective than oral anticoagulants. Low-molecular-weight heparin initiated postoperatively in close proximity to surgery at half the usual dose was not associated with a clinically or statistically significant increase in major bleeding rates (P =.16).

CONCLUSIONS

The timing of initiating low-molecular-weight heparin significantly influences antithrombotic effectiveness. The practice of delayed initiation of low-molecular-weight heparin prophylaxis results in suboptimal antithrombotic effectiveness without a substantive safety advantage.

New methods for estimating pretest probability in the diagnosis of pulmonary embolism

The clinical assessment of the probability of pulmonary embolism is a key step in proposed diagnostic strategies for pulmonary embolism, because the interpretation of noninvasive test results is conditional on the pretest probability derived from the presence or absence of clinical factors. The past year has brought important progress in the general area of clinical prediction of pulmonary embolism with the publication of two new simple clinical prediction rules. Each of the prediction rules includes a total of seven clinical variables that, when combined, allow for the categorization of patients into categories of low, intermediate, or high pretest probability of pulmonary embolism. Although these clinical prediction rules are perhaps only slightly better than the estimates of experienced clinicians, they provide an explicit method for estimating the probability of PE as an adjunct to diagnostic testing. Further validation work is now needed to assess how well these new prediction rules perform in settings other than the derivation sites.

Clinical assessment of clonidine in the treatment of new-onset rapid atrial fibrillation: a prospective, randomized clinical trial

PURPOSE

The role of digoxin and verapamil in the control of ventricular response in rapid atrial fibrillation is well established. This study investigates how clonidine compares with these standard therapies in rate control for new-onset rapid atrial fibrillation. We set out to test the hypothesis that clonidine effectively reduces heart rate in patients with new-onset rapid atrial fibrillation.

SUBJECTS AND METHODS

Forty patients were seen in the emergency department with new-onset (< or =24 hours' duration), stable, rapid atrial fibrillation. Eligible patients were randomized to receive either clonidine, digoxin, or verapamil. Changes in heart rate and blood pressure over 6 hours, as well as frequency of conversion to sinus rhythm were recorded and analyzed.

RESULTS

The mean reduction in heart rate over 6 hours was 44.4 beats/min (95% confidence interval [CI] 28.4-60.4 beats/min) in the clonidine group, 52.1 beats/min (95% CI 40.8-63.4 beats/min) in the digoxin group, and 41.8 beats/min (95% CI 22.5-61.0 beats/min) in the verapamil group. Analysis of variance of the heart rate changes in the 3 groups after 6 hours was not significant (P =.55). At 6 hours, 7 of 12 clonidine patients, 8 of 15 digoxin patients, and 7 of 13 verapamil patients remained in atrial fibrillation (P =.962 on chi(2)).

CONCLUSION

Clonidine controls ventricular rate in new-onset atrial fibrillation with an efficacy comparable to that of standard agents.

The Southern Alberta Renal Program database: a prototype for patient management and research initiatives

The Southern Alberta Renal Program (SARP) database was developed to respond to an urgent need for local information on clinical outcomes, laboratory information, and health care costs, and to enable our local renal program to monitor the implementation of established clinical practice guidelines. The database captures detailed demographic, clinical, and laboratory information and is unique by also capturing comorbidity, health-related quality of life and costing information for patients with end-stage renal disease (ESRD) in southern Alberta, storing the information in one common database. By collecting information on patient comorbidity, health outcomes and costs, the SARP database has enabled many quality assurance initiatives as well as research opportunities for projects involving patients with ESRD. Due to the availability of links with other available local clinical and administrative databases, information is collected with a minimal need for manual data entry. This type of database is a method by which health programs could improve the quality of patient care. Programs caring for patients with chronic medical conditions such as ESRD should examine how computer databases could assist in clinical care and improve the efficiency with which that care is delivered to their patients.

Validation of three myocardial jeopardy scores in a population-based cardiac catheterization cohort

BACKGROUND

The Jeopardy Score from Duke University and the Myocardial Jeopardy Index from the Bypass Angioplasty Revascularization Investigation (BARI) have been validated but never applied to a large unselected cohort. We assessed the prognostic value of these existing jeopardy scores, along with that of a new Lesion Score developed for the Alberta Provincial Project for Outcome Assessment in Coronary Heart Disease (APPROACH), a clinical data collection initiative capturing all patients undergoing cardiac catheterization in the province of Alberta.

METHODS

The predictive value of these three scores were compared in a cohort of >20,000 patients (9922 treated medically, 6334 treated with percutaneous intervention, and 3811 treated with bypass surgery). Scores were considered individually in logistic regression models for their ability to predict outcome and then added to models containing sociodemographic data, comorbidities, ejection fraction, indication for procedure, and descriptors of coronary anatomy.

RESULTS

All scores were found to be predictive of 1-year mortality, especially when patients are treated medically or with percutaneous intervention. In these patients, the APPROACH Lesion Score performed slightly better than the other jeopardy scores. The Duke Jeopardy Score was most predictive in those patients undergoing coronary bypass surgery.

CONCLUSIONS

Myocardial jeopardy scores provide independent prognostic information for patients with ischemic heart disease, especially if those patients are treated medically or with percutaneous intervention. These scores represent potentially valuable tools in cardiovascular outcome studies. The APPROACH Lesion Score may perform slightly better than previously developed jeopardy scores.

Risk adjustment using administrative data: impact of a diagnosis-type indicator

OBJECTIVES

To determine the frequency with which commonly coded clinical variables are complications, as opposed to baseline comorbidities, and to compare the results of 2 risk-adjusted outcome analyses for coronary artery bypass graft surgery for which we either (a) ignored, or (b) used the available "diagnosis-type indicator."

DESIGN

Analysis of existing administrative data.

SETTING

Twenty-three Canadian hospitals.

PATIENTS

A total of 50,357 coronary artery bypass graft surgery cases.

MEASUREMENTS AND MAIN RESULTS

Among 21 clinical variables whose definitions involve the diagnosis-type indicator, 14 were predominantly (> or =97%) baseline risk factors when present. Seven variables were often complication diagnoses: renal disease (when present, 13% coded as complications), recent myocardial infarction (15%), peptic ulcer disease (15%), congestive heart failure (17%), cerebrovascular disease (26%), hemiplegia (34%), and severe liver disease (35%). The results of risk adjustment analyses predicting in-hospital mortality differed when the diagnosis-type indicator was either used or ignored, and as a result, adjusted hospital mortality rates and rankings changed, often dramatically, with rankings increasing for 10 hospitals, decreasing for 9 hospitals, and remaining the same for only 4 hospitals.

CONCLUSIONS

The results of analyses performed using the diagnosis-type indicator in Canadian administrative data differ considerably from analyses that ignore the indicator. The widespread introduction of such an indicator should be considered in other countries, because risk-adjustment analyses performed without a diagnosis-type indicator may yield misleading results.

Long-term survival in 11,661 patients with multivessel coronary artery disease in the era of stenting: a report from the Alberta Provincial Project for Outcome Assessment in Coronary Heart Disease (APPROACH) Investigators

BACKGROUND

Studies of survival of patients with multivessel coronary artery disease (MVD) in the prestent era suggested that outcomes after coronary artery bypass surgery (CABG) are similar to those after percutaneous coronary intervention (PCI) in subsets of coronary severity. The purpose of this study of the Alberta Provincial Project for Outcome Assessment in Coronary Heart Disease (APPROACH) was to examine the association between treatment and survival up to 5 years in patients with MVD enrolled from 1995 through 1998.

METHODS AND RESULTS

Data on patient characteristics were obtained at the time of the initial coronary angiography. Survival was determined through data linkage to the provincial Bureau of Vital Statistics. Risk-adjusted hazard ratios were calculated to compare different treatments. In the 11,661 patients with MVD, CABG was the initial therapy in 3782, PCI in 3540, and medical therapy in 4339. Cumulative 5-year survival was 91.4% with CABG, 91.9% with PCI, and 82.9% with medical therapy (P <.001). Hazard ratios were CABG: medical 0.53 (95% confidence interval [CI] 0.46-0.71), PCI: medical 0.65 (95% CI 0.56-0.74), and CABG: PCI 0.81 (95% CI 0.68-0.96). Analysis across coronary severity groups revealed a benefit of CABG compared with PCI only in the group with severe left main CAD: 0.30 (95% CI 0.17-0.54).

CONCLUSIONS

In a multicenter clinical setting, MVD patients treated with revascularization have significantly higher 5-year survival rate than do those treated medically. Risk-adjusted comparison reveals PCI treatment to be associated with long-term survival similar to treatment with CABG in all coronary severity subgroups except the group with severe left main coronary artery disease. Patient selection factors are likely to be contributing to these findings.

Reviewing the reviewers: the quality of reporting in three secondary journals

BACKGROUND

Secondary journals such as ACP Journal Club (ACP), Journal Watch (JW) and Internal Medicine Alert (IMA) have enormous potential to help clinicians remain up to date with medical knowledge. However, for clinicians to evaluate the validity and applicability of new findings, they need information on the study design, methodology and results.

METHODS

Beginning with the first issue in March 1997, we selected 50 consecutive summaries of studies addressing therapy or prevention and internal medicine content from each of the ACP, JW and IMA. We evaluated the summaries for completeness of reporting key aspects of study design, methodology and results.

RESULTS

All of the summaries in ACP reported study design, as compared with 72% of the summaries in JW and IMA (p < 0.001). In summaries of randomized controlled trials the 3 secondary journals were similar in reporting concealment of patient allocation (none reported this), blinding status of participants (ACP 62%, JW 70% and IMA 70% [p = 0.7]), blinding status of health care providers (ACP 12%, JW 4% and IMA 4% [p = 0.4]) and blinding status of judicial assessors of outcomes (ACP 4%, JW 4% and IMA 0% [p = 0.4]). ACP was the only one to report whether investigators conducted an intention-to-treat analysis (in 38% of summaries [p < 0.001]), and it was more likely than the other 2 journals to report the precision of the treatment effect (as a p value or 95% confidence interval) (ACP 100%, JW 0% and IMA 55% [p < 0.001]).

INTERPRETATION

Although ACP provided more information on study design, methodology and results, all 3 secondary journals often omitted important information. More complete reporting is necessary for secondary journals to fulfill their potential to help clinicians evaluate the medical literature.

A case study of hospital closure and centralization of coronary revascularization procedures

BACKGROUND

Despite nation-wide efforts to reduce health care costs through hospital closures and centralization of services, little is known about the impact of such actions. We conducted this study to determine the effect of a hospital closure in Calgary and the resultant centralization of coronary revascularization procedures from 2 facilities to a single location.

METHODS

Administrative data were used to identify patients who underwent coronary artery bypass grafting (CABG), including those who had combined CABG and valve procedures, and patients who underwent percutaneous transluminal angioplasty (PTCA) in the Calgary Regional Health Authority from July 1994 to March 1998. This period represents the 21 months preceding and the 24 months following the March 1996 hospital closure. Measures, including mean number of discharges, length of hospital stay, burden of comorbidity and in-hospital death rates, were compared before and after the hospital closure for CABG and PTCA patients. Multivariate analyses were used to derive risk-adjustment models to control for sociodemographic variables and comorbidity.

RESULTS

The number of patients undergoing CABG was higher in the year following than in the year preceding the hospital closure (51.6 per 100,000 before v. 67.3 per 100,000 after the closure); the same was true for the number of patients undergoing PTCA (129.8 v. 143.6 per 100,000). The burden of comorbidity was significantly higher after than before the closure, both for CABG patients (comorbidity index 1.3 before v. 1.5 after closure, p < 0.001) and for PTCA patients (comorbidity index 1.0 before v. 1.1 after, p = 0.04). After adjustment for comorbidity, the mean length of hospital stay was significantly lower after than before the closure for CABG patients (by 1.3 days) and for PTCA patients (by 1.0 days). The adjusted rates of death were slightly lower after than before the closure in the CABG group. The adjusted rates of death or CABG in the PTCA group did not differ significantly between the 2 periods.

INTERPRETATION

Hospital closure and the centralization of coronary revascularization procedures in Calgary was associated with increased population rates of procedures being performed, on sicker patients, with shorter hospital stays, and, for CABG patients, a trend toward more favourable short-term outcomes. Our findings indicate that controversial changes to the structure of the health care system can occur without loss of efficiency and reduction in quality of care.

Physician interpretations and textbook definitions of blinding terminology in randomized controlled trials

CONTEXT

When clinicians assess the validity of randomized controlled trials (RCTs), they commonly evaluate the blinding status of individuals in the RCT. The terminology authors often use to convey blinding status (single, double, and triple blinding) may be open to various interpretations.

OBJECTIVE

To determine physician interpretations and textbook definitions of RCT blinding terms.

DESIGN AND SETTING

Observational study undertaken at 3 Canadian university tertiary care centers between February and May 1999.

PARTICIPANTS

Ninety-one internal medicine physicians who responded to a survey.

MAIN OUTCOME MEASURES

Respondents identified which of the following groups they thought were blinded in single-, double-, and triple-blinded RCTs: participants, health care providers, data collectors, judicial assessors of outcomes, data analysts, and personnel who write the article. Definitions from 25 systematically identified textbooks published since 1990 providing definitions for single, double, or triple blinding.

RESULTS

Physician respondents identified 10, 17, and 15 unique interpretations of single, double, and triple blinding, respectively, and textbooks provided 5, 9, and 7 different definitions of each. The frequencies of the most common physician interpretation and textbook definition were 75% (95% confidence interval [CI], 65%-83%) and 74% (95% CI, 52%-90%) for single blinding, 38% (95% CI, 28%-49%) and 43% (95% CI, 24%-63%) for double blinding, and 18% (95% CI, 10%-28%) and 14% (95% CI, 0%-58%) for triple blinding, respectively.

CONCLUSIONS

Our study suggests that both physicians and textbooks vary greatly in their interpretations and definitions of single, double, and triple blinding. Explicit statements about the blinding status of specific groups involved in RCTs should replace the current ambiguous terminology.

Poor long-term survival after coronary angiography in patients with renal insufficiency

Cardiovascular disease is common among dialysis patients, but much less is known regarding non-dialysis-dependent renal insufficiency (NDDRI) and its association with cardiac disease. We undertook a study to assess the impact of renal insufficiency on survival post-coronary angiography by comparing three groups of patients: dialysis-dependent patients, patients with NDDRI (creatinine > 2.3 mg/dL), and a reference group with creatinine levels less than 2.3 mg/dL and not on dialysis therapy. We used a prospective cohort that consisted of all patients undergoing coronary angiography in Alberta, Canada, from January 1, 1995, to December 31, 1997. Of the 16,989 patients, 196 patients (1.2%) were on dialysis therapy, 262 patients (1.5%) had NDDRI, and 16,531 patients (97.3%) formed the reference group. Mortality rates 1 year after angiography were 30.2% for patients with NDDRI, 15.8% for dialysis patients, and 4.1% for the reference group. Compared with the reference group, crude 4-year survival was significantly worse for dialysis patients and those with NDDRI, with hazard ratios of 4.05 (95% confidence interval, 3.02 to 5.42) and 7.32 (95% confidence interval, 5.97 to 8.97), respectively. Even after adjusting for clinical risk factors, survival remained worse for dialysis patients and those with NDDRI, with hazard ratios of 2.59 (95% confidence interval, 1.92 to 3.49) and 2.51 (95% confidence interval, 2.02 to 3.12), respectively. We conclude that renal insufficiency, both dialysis dependent and non-dialysis dependent, is an independent risk factor for increased mortality and poor long-term survival among patients undergoing coronary angiography.

Physicians' attitudes towards prevention: importance of intervention-specific barriers and physicians' health habits

BACKGROUND

Several studies have explored physicians' attitudes towards prevention and barriers to the delivery of preventive health interventions. However, the relative importance of these previously identified barriers, both in general terms and in the context of a number of specific preventive interventions, has not been identified. Certain barriers may only pertain to a subset of preventive interventions.

OBJECTIVES

We aimed to determine the relative importance of identified barriers to preventive interventions and to explore the association between physicians' characteristics and their attitudes towards prevention.

METHODS

We conducted a cross-sectional survey of 496 of the 686 (72.3% response rate) generalist physicians from three Swiss cantons through a questionnaire asking physicians to rate the general importance of eight preventive health strategies and the relative importance of seven commonly cited barriers in relation to each specific preventive health strategy.

RESULTS

The proportion of physicians rating each preventive intervention as being important varied from 76% for colorectal cancer screening to 100% for blood pressure control. Lack of time and lack of patient interest were generally considered to be important barriers by 41% and 44% of physicians, respectively, but the importance of these two barriers tended to be specifically higher for counselling-based interventions. Lack of training was most notably a barrier to counselling about alcohol and nutrition. Four characteristics of physicians predicted negative attitudes toward alcohol and smoking counselling: consumption of more than three alcoholic drinks per day [odds ratio (OR) = 8.4], sedentary lifestyle (OR = 3.4), lack of national certification (OR = 2.2) and lack of awareness of their own blood pressure (OR = 2.0).

CONCLUSIONS

The relative importance of specific barriers varies across preventive interventions. This points to a need for tailored practice interventions targeting the specific barriers that impede a given preventive service. The negative influence of physicians' own health behaviours indicates a need for associated population-based interventions that reduce the prevalence of high-risk behaviours in the population as a whole.

Prognostic significance of diabetes as a predictor of survival after cardiac catheterization. APPROACH Investigators. Alberta Provincial Program for Outcome Assessment in Coronary Heart Disease

PURPOSE

Diabetes is a recognized risk factor for the development of cardiac disease, but its importance as a prognostic factor among patients with known cardiovascular disease is less clear. We evaluated survival in patients with and without diabetes who underwent cardiac catheterization for presumed coronary artery disease.

SUBJECTS AND METHODS

We analyzed data from a prospective cohort study that captures detailed clinical information and longitudinal outcomes for all patients who undergo cardiac catheterization in Alberta, Canada. We studied 11,468 patients, 1959 (17%) of whom had diabetes. Logistic regression was used to model predictors of 1-year mortality, and proportional hazards analysis was used to model predictors of survival up to 3 years after cardiac catheterization.

RESULTS

One-year mortality was 7.6% for patients with diabetes versus 4.1% for those without diabetes (odds ratio = 1.9, 95% confidence interval [CI]: 1.6 to 2.3). After adjusting for other characteristics of the patients, including comorbid conditions, previous cardiac history, coronary anatomy, and renal function, the odds ratio for 1-year mortality was 1.1 (95% CI: 0.8 to 1.3). Similarly, the adjusted hazard ratio for longer term mortality was 1. 2 (95% CI: 1.0 to 1.4, mean follow-up of 702 days).

CONCLUSIONS

These results suggest that there is little or no independent association between diabetes and mortality for up to 3 years after cardiac catheterization. Estimates of short- to intermediate-term prognosis for diabetic patients with coronary artery disease should be based on the presence of other prognostic factors associated with diabetes.

Overview of the Alberta Provincial Project for Outcome Assessment in Coronary Heart Disease. On behalf of the APPROACH investigators

The Alberta Provincial Project for Outcome Assessment in Coronary Heart Disease (APPROACH) is an ongoing prospective data collection initiative that began in January 1995. The cohort for the initiative is all patients undergoing cardiac catheterization in Alberta. Patients are followed longitudinally for the determination of short and long term clinical, economic and quality of life outcomes. The project is producing valuable information on the processes and outcomes of cardiac care in Alberta, and is now being implemented in British Columbia as well. This paper provides an overview of APPROACH with specific attention to the project's general objectives, salient features, database structure and technical specifications. Examples of applied research projects based on APPROACH data are also provided.

A Canadian comparison of data sources for coronary artery bypass surgery outcome "report cards"

BACKGROUND

Prior comparisons of administrative versus clinical data for creating coronary artery bypass graft (CABG) surgery outcome "report cards" are all from the United States and yield inconsistent conclusions regarding the validity of administrative data report cards. In this study, we compared 2 CABG surgery outcome report cards for Ontario, Canada: one derived from clinical data from the Cardiac Care Network of Ontario and one derived from administrative data from the Canadian Institute for Health Information.

METHODS

Data from 4 fiscal years, 1992-93 through 1995-96, were used. The Canadian Institute for Health Information report card was derived from administrative data only. The Cardiac Care Network report card drew on prospectively collected clinical information that included variables such as left ventricular ejection fraction but also required linkages to the Canadian Institute for Health Information data for ascertainment of selected comorbidities and in-hospital mortality rates. Logistic regression models were used to calculate risk-adjusted death rates for each of the 9 hospitals performing CABG surgery in Ontario.

RESULTS

The risk-adjusted death rates were quite similar between data sources for 7 of the 9 hospitals. For 2 hospitals, rather large absolute differences in adjusted death rates of 0.58% and 0.64% were seen between report cards. There was a strong correlation between data sources for risk-adjusted hospital death rates (intraclass correlation coefficient = 0.927, P <.001) and for rankings of adjusted hospital death rates (Spearman correlation coefficient = 0.828, P =.02).

CONCLUSION

These results from Ontario, Canada, reveal general similarities between administrative and clinical data report cards for CABG surgery. However, clinical data are likely needed if individual hospitals are to be publicly scrutinized in outcome report cards.

Physicians' recommendations for patients who undergo noncardiac surgery

OBJECTIVE

To investigate how consulting physicians attempt to modify perioperative cardiac risk for patients who undergo noncardiac surgery by comparing the preoperative cardiac recommendations of consulting physicians in 2 university centres.

DESIGN

Retrospective cross-sectional analysis.

SETTING

Five hospitals affiliated with 2 Canadian universities.

PATIENTS

Three hundred and eight preoperative consultations were evaluated in 297 patients who were 40 years of age or older and scheduled for noncardiac surgery.

OUTCOME MEASURES

Cardiac drug recommendations at the preoperative consultation [corrected]; overall recommendations and practice variation between the 2 centres.

RESULTS

The greatest changes in drug management suggested by consultants were the initiation of nitrates in 13% of the patients and a decrease in acetylsalicylic acid administration from 27% to 17%. Centre A physicians recommended adding an angiotensin-converting enzyme inhibitor 11% of the time, whereas centre B physicians recommended such an inhibitor in only 1% of the patients (p = 0.001). In patients taking acetylsalicylic acid at the preoperative consultation, Centres A and B physicians recommended withholding the drug 47% and 22% of the time, respectively (p = 0.03). These differences persisted between the 2 centres after controlling for physician estimates of risk.

CONCLUSIONS

Consultants frequently recommended perioperative changes in the use of cardiac medications, and there were differences in practice patterns between the 2 centres. These differences may be affecting patient outcomes and highlight the need for randomized clinical trials to determine the impact of perioperative drug administration on bleeding, myocardial infarction and death.

Dealing with missing data in observational health care outcome analyses

Observational outcome analyses appear frequently in the health research literature. For such analyses, clinical registries are preferred to administrative databases. Missing data are a common problem in any clinical registry, and pose a threat to the validity of observational outcomes analyses. Faced with missing data in a new clinical registry, we compared three possible responses: exclude cases with missing data; assume that the missing data indicated absence of risk; or merge the clinical database with an existing administrative database. The predictive model derived using the merged data showed a higher C statistic (C = 0.770), better model goodness-of-fit as measured in a decile-of-risk analysis, the largest gradient of risk across deciles (46.3), and the largest decrease in deviance (-2 log likelihood = 406.2). The superior performance of the enhanced data model supports the use of this "enhancement" methodology and bears consideration when researchers are faced with nonrandom missing data.

The impact of leaving against medical advice on hospital resource utilization

OBJECTIVE

To assess the effect of hospital discharge against medical advice (AMA) on the interpretation of charges and length of stay attributable to alcoholism.

DESIGN

Retrospective cohort. Three analytic strategies assessed the effect of having an alcohol-related diagnosis (ARD) on risk-adjusted utilization in multivariate regressions. Strategy 1 did not adjust for leaving AMA, strategy 2 adjusted for leaving AMA, and strategy 3 restricted the sample by excluding AMA discharges.

SETTING

Acute care hospitals.

PATIENTS

We studied 23,198 pneumonia hospitalizations in a statewide administrative database.

MEASUREMENTS AND MAIN RESULTS

Among these admissions, 3.6% had an ARD, and 1.2% left AMA. In strategy 1 an ARD accounted for a $1,293 increase in risk-adjusted charges for a hospitalization compared with cases without an ARD ( p =.012). ARD-attributable increases of $1,659 ( p =.002) and $1,664 ( p =. 002) in strategies 2 and 3 respectively, represent significant 28% and 29% increases compared with strategy 1. Similarly, using strategy 1 an ARD accounted for a 0.6-day increase in risk-adjusted length of stay over cases without an ARD ( p =.188). An increase of 1 day was seen using both strategies 2 and 3 ( p =.044 and p =.027, respectively), representing significant 67% increases attributable to ARDs compared with strategy 1.

CONCLUSIONS

Discharge AMA affects the interpretation of the relation between alcoholism and utilization. The ARD-attributable utilization was greater when analyses adjusted for or excluded AMA cases. Not accounting for leaving AMA resulted in an underestimation of the impact of alcoholism on resource utilization.

Why do patients with atrial fibrillation not receive warfarin?

Atrial fibrillation (AF) is a growing public health problem associated with significant morbidity and mortality. Numerous randomized controlled trials of warfarin have conclusively demonstrated that long-term anticoagulation therapy can reduce the risk for stroke by approximately 68% per year in patients with nonvalvular AF, and even more in patients with valvular AF. However, available data show that of those patients with AF and no contraindication to warfarin therapy, only 15% to 44% are prescribed warfarin. Our literature review has identified patient-, physician-, and health care system-related barriers to warfarin prescription. However, the relative importance of these specific barriers remains unknown. Further work is needed to understand the discrepancy between the randomized controlled trial evidence and clinical practice patterns.

Successful teaching in evidence-based medicine

OBJECTIVES

Several published articles have described the importance of exposing medical trainees to the 'new paradigm' of evidence-based medicine (EBM). Recognizing this, we sought to develop and objectively evaluate a mini-course in EBM for third-year medical students.

DESIGN

We developed a mini-course consisting of four sessions in which students learn to derive sequentially focused questions, search MEDLINE, review articles critically and apply information from the literature to specific clinical questions. To evaluate the teaching intervention, we performed a controlled educational study. Students at the intervention site (n=34) attended the EBM mini-course, while students at the control site (n=26) received more 'traditional' didactic teaching on various clinical topics. Intervention and control students were surveyed immediately before and after the mini-course to assess changes in reading and literature searching skills, as well as a tendency to use the literature to answer clinical questions.

SETTING

Boston University School of Medicine.

SUBJECTS

Third-year medical students.

RESULTS

The intervention was associated with significant changes in students' self-assessed skills and attitudes. MEDLINE and critical appraisal skills increased significantly in the intervention group relative to the control group (significance of between group differences: P=0.002 for MEDLINE and P=0.0002 for critical appraisal), as did students' tendency to use MEDLINE and original research articles to solve clinical problems (significance of between group differences: P=0.002 and P=0.0008, respectively).

CONCLUSIONS

We conclude that this brief teaching intervention in EBM has had a positive impact on student skills and attitudes at our medical school. We believe that the key elements of this intervention are (1) active student involvement, (2), clinical relevance of exercises and (3) integrated teaching targeting each of the component skills of EBM.