C. David Naylor

Incorporating variations in the quality of individual randomized trials into meta-analysis.

Meta-analysis is a method of synthesizing evidence from multiple sources. It has been increasingly applied to combine results from randomized trials of therapeutic strategies. Unfortunately there is often variation in the quality of the trials that are included in meta-analyses, limiting the value of combining the results in an overview. This variation in quality can lead to both bias and reduction in precision of the estimate of the therapys effectiveness. There are a number of methods for quantifying the quality of trials including the detailed Chalmers system and simple scales. The nature of the relationship between these quality scores and the true estimate of effectiveness is unknown at this time. We discuss four methods of incorporating quality into meta-analysis: threshold score as inclusion/exclusion criterion, use of quality score as a weight in statistical pooling, visual plot of effect size against quality score and sequential combination of trial results based on quality score. The last method permits an examination of the relation between quality and both bias and precision on the pooled estimates. We conclude that while it is possible to incorporate the effect of variation of quality of individual trials into overviews, this issue requires more study.

Multicenter Validation of a Risk Index for Mortality, Intensive Care Unit Stay, and Overall Hospital Length of Stay After Cardiac Surgery

BACKGROUND A multicenter population-based study was conducted to develop and validate a risk index for mortality, intensive care unit (ICU) length of stay, and postoperative length of stay after cardiac surgery. METHODS AND RESULTS Data were collected from 13,098 patients undergoing cardiac surgery between April 1, 1991, and March 31, 1993, at all nine adult cardiac surgery institutions in Ontario, Canada. A six-variable risk index (age, sex, left ventricular function, type of surgery, urgency of surgery, and repeat operation) was developed using logistic regression analysis to predict in-hospital mortality, ICU stay in days, and postoperative stay in days after cardiac surgery in a derivation set of 6213 patients who had cardiac surgery during fiscal year 1991 (April 1, 1991, to March 31, 1992). The index predicted mortality, prolonged ICU stay (> or = 6 days), and prolonged postoperative length of stay (> or = 17 days) after cardiac surgery with areas under the receiver-operating characteristic (ROC) curve of 0.75, 0.66, and 0.69, respectively, in an independent validation set of 6885 patients who had cardiac surgery during fiscal year 1992 (April 1, 1992, to March 31, 1993). Increasing risk scores were associated with greater mortality rates and longer ICU and postoperative stays at all nine institutions. CONCLUSIONS Mortality, ICU length of stay, and postoperative length of stay after cardiac surgery can be predicted using a simple six-variable risk index. The index has potential application as a risk stratification tool for comparing patient outcomes and resource use among different hospitals and surgeons.

Characteristics and mortality outcomes of thrombolysis trial participants and nonparticipants: a population-based comparison.

OBJECTIVES This study was done to compare characteristics and outcomes of patients with acute myocardial infarction participating in two thrombolysis trials with those of nontrial patients at study hospitals and external hospitals. BACKGROUND Preferential recruitment of lower risk patients into randomized trials of thrombolysis has been suggested by earlier studies. However, to date there has not been a definitive population-based comparison of characteristics and outcomes for thrombolysis trial participants and nonparticipants. METHODS Population-based data on hospital admissions and mortality from acute myocardial infarction for all hospitals in Ontario from 1989 to 1992 were linked to data on trial participants in two distinct thrombolysis studies (GUSTO I and LATE). Included were 1,304 patients entered into GUSTO, 12,657 nonparticipants at GUSTO hospitals, 249 patients entered into LATE, 5,997 nonparticipants at LATE hospitals and 12,299 patients at external hospitals. The main outcomes were differences in age, gender, comorbidity scores, coronary revascularization and survival to hospital discharge. RESULTS Patients in both GUSTO and LATE were significantly more likely to be <70 years old (odds ratio [OR] 2.8 and 3.2, respectively), to be male (OR 2.0 and 2.1, respectively), to have low comorbidity scores (OR 2.0 and 2.3, respectively) and, for GUSTO alone, to undergo coronary revascularization (OR 2.4). Nontrial patients were similar between trial hospitals and external hospitals. In-hospital mortality rates for GUSTO and LATE patients were lower (6.9% and 6.6%, respectively) than for nonparticipants at study hospitals (16.8% and 19.7%, respectively; p<0.001 for both comparisons). Survival to hospital discharge remained higher among GUSTO (OR 1.9) and LATE patients (OR 2.0) than nonparticipants at study hospitals even after adjustment for age, gender, revascularization and comorbidity scores. CONCLUSIONS Compared with nontrial patients, thrombolysis trial participants are younger, more often male, undergo more revascularization and have less comorbid disease. Even after adjustment for these factors, participants have a survival advantage over nonparticipants that is larger than expected from thrombolysis alone. These findings are not attributable to inferior care or skewed populations at hospitals that did not join these major trials. Further study of these selection biases may guide future trial design and deepen our understanding of why thrombolytics have been underused for high risk patients in routine practice.

Parenteral Nutrition With Branched-Chain Amino Acids in Hepatic Encephalopathy

Meta-analytic methods were applied to review clinical trials published in English that have assessed the efficacy of parenteral nutrition for cirrhotic patients with acute hepatic encephalopathy. Modified amino acid solutions containing increased amounts of branched-chain amino acids were used as part of the treatment regimen in all studies. Pooled analysis of five randomized controlled studies showed a highly significant improvement in mental recovery (p less than 0.001) from high-grade encephalopathy over follow-up times varying from 5 to 14 days. The significance level of the treatment effect did not change when the analysis was repeated using alternative methods of counting and attributing events in these trials. Sharp differences in direction of treatment effect precluded pooling case fatality data. Two studies reported an increased risk of death in the treatment group. Two others showed a clear benefit from administration of parenteral nutrition: the aggregate relative risk reduction was 0.59 (95% confidence interval: 0.23-0.80, p = 0.002). Addition of unpublished data from a third positive study increased the relative risk reduction to 0.82 (p less than 0.0001), and the most conservative interpretation of the published data still yielded a significant reduction in mortality (p = 0.023). However, given the uncertainty about effects on mortality and short follow-up times in all studies, a confirmatory randomized controlled trial with longer follow-up periods is warranted.

Ready-made, recalibrated, or Remodeled? Issues in the use of risk indexes for assessing mortality after coronary artery bypass graft surgery.

BACKGROUND Risk indexes for operative mortality after cardiac surgery are used for comparative profiling of surgeons or centers. We examined whether clinicians and managers should use an existing index without modification, recalibrate it for their populations, or derive a new model altogether. METHODS AND RESULTS Drawing on 7491 consecutive patients who underwent isolated CABG at 2 Toronto teaching hospitals between 1993 and 1996, we compared 3 strategies: (1) using a ready-made model originally derived and validated in our jurisdiction; (2) recalibrating the ready-made model to better fit the population; and (3) deriving a new model with additional risk factors. We assessed statistical accuracy, ie, area under a receiver-operator characteristic curve (ROC); precision, ie, statistical goodness-of-fit; and actual impact on both risk-adjusted operative mortalities (RAOM) and performance rankings for 14 surgeons. The new model was slightly more accurate than the ready-made model (ROC, 0.78 versus 0.76; P<0.05), albeit not different from the recalibrated model (ROC, 0.77). The ready-made model showed poor fit between the predicted and observed results (P<0.001), leading to significant underestimation of RAOM (1.6+/-0. 2%) compared with the other strategies (2.5+/-0.2%; P=0.048). Remodeling also changed the performance rankings among half the surgeons with higher RAOM. CONCLUSIONS Poorly calibrated risk algorithms can bias the calculation of RAOM and alter the results of surgeon-specific profiles. Any existing index used for risk assessment in cardiac surgery should be episodically recalibrated or compared with new models derived from local subjects to ensure that its performance remains optimal.

Fifteen-Year Trends in Risk Severity and Operative Mortality in Elderly Patients Undergoing Coronary Artery Bypass Graft Surgery

BACKGROUND Trends in risk-severity and operative mortality (OM) were examined in 3330 consecutive patients aged 70 years and older who underwent isolated coronary artery bypass graft surgery (CABG) between 1982 and 1996. METHODS AND RESULTS The proportion of elderly patients rose significantly over time (P<.001). Crude OM among the elderly was 7.2% in 1982 to 1986, fell to 4.4% in 1987 to 1991, but did not improve thereafter. Logistic regression analysis of OM was used to construct relative risk groups (low, medium, or high). The prevalence of high-risk elderly patients rose significantly over time (P=.001) from 16.2% in 1982 to 1986 to 19.5% in 1987 to 1991 and 26.9% in 1992 to 1996. OM in high-risk patients fell significantly (P=.044) from 17.2% in 1982 to 1986 to 9.1% in 1987 to 1991 and was 8.9% in 1992 to 1996. Contemporary independent predictors of OM among elderly patients were poor ventricular function (LV grade 2 to 3, odds ratio [OR], 2.6; 95% confidence interval [CI], 1.3 to 5.2; and LV grade 4, OR, 10.7; 95% CI, 4.4 to 26); previous CABG (OR, 3.7; 95% CI, 2.0 to 7.0), female sex (OR, 1.8; 95% CI, 1.1 to 2.8), peripheral vascular disease (OR, 1.8; 95% CI, 1.1 to 2.8), and diabetes (OR, 1.7; 95% CI, 1.1 to 2.7). Previous angioplasty was protective (OR, 0.3; 95% CI, 0.1 to 0.9). CONCLUSIONS OM in elderly patients has declined significantly in recent years despite an increase in the prevalence and severity of their risk factors. A careful weighing of risk, rather than advanced age alone, should determine who is offered surgical revascularization. In this regard, poor ventricular function and repeat CABG continue to have the greatest impact on OM in elderly patients.

Socioeconomic Status and Mortality after Acute Myocardial Infarction

Context Many studies show worse cardiovascular outcomes among poor persons than among affluent persons, but the factors that mediate this relationship are unknown. Contribution In this prospective study of 3407 Canadians who were hospitalized for myocardial infarction, 7.1% of the patients in the high-income group died within 2 years compared with 15.3% of the patients in the low-income group. Adjustment for age, preexisting cardiovascular disease, and risk factors greatly attenuated the relationship between mortality rates and income. Adjustment for other factors had little effect. Implications A history of 1 or more cardiovascular events and worse cardiovascular risk factors may explain why poor people have worse outcomes than affluent people after myocardial infarction. The Editors For many decades (1, 2) and across multiple nations (3-6), differences in socioeconomic status have been consistently associated with variations in cardiovascular disease and mortality rates (6, 7). This wealthhealth gradient (8) is independent of the socioeconomic indicator used (9), persists even after such cardiovascular events as acute myocardial infarction (MI) (10), and has been observed in countries with publicly funded universal health care (10-13). The causes of these incomeoutcome gradients are debatable (14-17). Poorer patients are more likely to smoke or have diabetes and hypertension, all of which lead to accelerated atherosclerosis and higher subsequent mortality rates (18, 19). However, incomeoutcome gradients persist after adjustment for cardiovascular events and traditional cardiac risk factors. These residual effects of income or education have led to speculation about differences in behaviors after MI, psychosocial stressors, and variations in access to medical care (10, 20-24). Although the mechanisms whereby psychosocial factors affect cardiovascular health are still incompletely delineated, ordinary risk factors (such as cigarette use) and health service intensity are potentially modifiable among the poor and those with less education. Therefore, our study focused on delineating the extent to which the association between socioeconomic factors and increased mortality rates can be explained by traditional risk factors and variations in service use. We hypothesized that cardiovascular risk factors remain the central intermediary pathway by which socioeconomic status is linked to increased mortality rates. We tested this hypothesis by using a cohort of patients who were hospitalized after an acute MI. By evaluating medium- term all-cause mortality in this sample, we increased the likelihood that death would be the result of a vascular event and reduced the risk for confounding by other causes of death (25). We aggregated traditional risk factors with previous vascular disease to obtain a powerful proxy for cumulative atherosclerotic burden, thereby enabling us to focus on assessing the incremental prognostic effect of socioeconomic status. Methods Data Source We obtained data from the Socio-Economic and Acute Myocardial Infarction (SESAMI) study, a prospective observational study of patients who were hospitalized because of acute MI throughout Ontario, Canada (19, 24). Of these data, we included a 13-item patient-completed questionnaire that addressed risk factors for atherosclerosis and socioeconomic status. By using encrypted health card numbers, we linked survey data to administrative databases for additional clinical information. We tracked each patients hospitalization history by using computerized abstracts that were assembled by the Canadian Institute for Health Information from 1 April 1988 to the date of the patients admission for the index event. We ascertained the number and types of cardiac procedures performed during and following the index admission by using the institutes data and physician billing claims from the Ontario Health Insurance Plan databases (10, 24). Procedure use reported in administrative databases was compared with patient self-reports; agreement levels ranged from 74% (coronary angiography) to 98% (coronary artery bypass surgery) and were similar across socioeconomic strata (24). We calculated patient deaths by acquiring data from the Ontario Registered Persons Data Base. Study Sample The SESAMI investigators recruited English-speaking patients who were admitted through the emergency departments in 53 of 57 large-volume (defined as having 100 or more patient admissions for MI per year) Ontario hospitals between 1 December 1999 and 26 February 2003. Trained nurses identified eligible patients through chart surveillance while patients were hospitalized in coronary or intensive care units. The diagnosis of MI was confirmed if at least 2 of 3 criteria were met: presence of symptoms, abnormal electrocardiographic findings, or elevated serum levels of cardiac enzymes. Of all eligible patients, 96% had acute MI confirmed by chart audits (19). We excluded patients younger than 19 years of age or older than 101 years of age, those lacking a valid health card number issued by the province of Ontario, and those who were transferred to the recruiting hospital. To be eligible, patients were required to complete a self-administered baseline survey at study entry. This requirement rendered ineligible those dying within 24 hours, those who had very severe illness (for example, patients receiving mechanical ventilation), those who had language barriers, or those undergoing early discharge or transfer (19, 24). Among 4668 consecutive eligible patients approached for consent, 3504 agreed to participate in baseline surveys and subsequent data linkage. For purposes of this study, 97 patients could not be linked to administrative data because of invalid or inaccurately documented health card numbers; 3407 patients remained available for participation. Socioeconomic Status, Ethnicity, and Demographic Factors We assessed self-reported household annual income (from all sources) as a 7-level categorical variable ranging from less than

Impact of Patient and Target-Vessel Characteristics on Arterial and Venous Bypass Graft Patency Insight From a Randomized Trial

15000 to greater than

Atenolol Use and Clinical Outcomes After Thrombolysis for Acute Myocardial Infarction: The GUSTO-I Experience

80000 Canadian. Self-reported educational attainment was analyzed as a 5-level categorical variable ranging from an incomplete high school education to a university degree. Income and education levels were self-reported by 92% and 98% of participants, respectively. To mitigate the confounding effect of retirement from the labor force, the cohort was stratified into persons younger than 65 years of age and those 65 years of age and older (12, 26). To ensure similar sample sizes across socioeconomic subgroups, we reaggregated income categories into a 3-level ordinal variable for each age group. For the younger age group, income categories were less than

Assessing the outcomes of coronary artery bypass graft surgery : How many risk factors are enough?

30000,