Improved Regression Estimation of a Multivariate Relationship with Population Data on the Bivariate Relationship

Regression coefficients specify the partial effect of a regressor on the dependent variable. Sometimes the bivariate or limited multivariate relationship of that regressor variable with the dependent variable is known from population-level data. We show here that such population- level data can be used to reduce variance and bias about estimates of those regression coefficients from sample survey data. The method of constrained MLE is used to achieve these improvements. Its statistical properties are first described. The method constrains the weighted sum of all the covariate-specific associations (partial effects) of the regressors on the dependent variable to equal the overall association of one or more regressors, where the latter is known exactly from the population data. We refer to those regressors whose bivariate or limited multivariate relationships with the dependent variable are constrained by population data as being ‘‘directly constrained.’’ Our study investigates the improvements in the estimation of directly constrained variables as well as the improvements in the estimation of other regressor variables that may be correlated with the directly constrained variables, and thus ‘‘indirectly constrained’’ by the population data. The example application is to the marital fertility of black versus white women. The difference between white and black women’s rates of marital fertility, available from population-level data, gives the overall association of race with fertility. We show that the constrained MLE technique both provides a far more powerful statistical test of the partial effect of being black and purges the test of a bias that would otherwise distort the estimated magnitude of this effect. We find only trivial reductions, however, in the standard errors of the parameters for indirectly constrained regressors.

Collaborative Change: Environmental Jolt, Network (Re)Design and Firm Performance

Motivated by the need to understand which are the underlying forces that trigger network evolution, we develop a multilevel theoretical and empirically testable model to examine the relationship between changes in the external environment and network change. We refer to network change as the dissolution or replacement of an interorganizational tie, adding also the case of the formation of new ties with new or preexisting partners. Previous research has paid scant attention to the organizational consequences of quantum change enveloping entire industries in favor of an emphasis on continuous change. To highlight radical change we introduce the concept of environmental jolt. The September 11 terrorist attacks provide us with a natural experiment to test our hypotheses on the antecedents and the consequences of network change. Since network change can be explained at multiple levels, we incorporate firm-level variables as moderators. The empirical setting is the global airline industry, which can be regarded as a constantly changing network of alliances. The study reveals that firms react to environmental jolts by forming homophilous ties and transitive triads as opposed to the non jolt periods. Moreover, we find that, all else being equal, firms that adopt a brokerage posture will have positive returns. However, we find that in the face of an environmental jolt brokerage relates negatively to firm performance. Furthermore, we find that the negative relationship between brokerage and performance during an environmental jolt is more significant for larger firms. Our findings suggest that jolts are an important predictor of network change, that they significantly affect operational returns and should be thus incorporated in studies of network dynamics.

Erythropoietin or Darbepoetin for patients with cancer--meta-analysis based on individual patient data

BACKGROUND: Erythropoiesis-stimulating agents (ESAs) reduce anemia in cancer patients and may improve quality of life, but there are concerns that ESAs might increase mortality. OBJECTIVES: Our objectives were to examine the effect of ESAs and identify factors that modify the effects of ESAs on overall survival, progression free survival, thromboembolic and cardiovascular events as well as need for transfusions and other important safety and efficacy outcomes in cancer patients. SEARCH STRATEGY: We searched the Cochrane Library, Medline, Embase and conference proceedings for eligible trials. Manufacturers of ESAs were contacted to identify additional trials. SELECTION CRITERIA: We included randomized controlled trials comparing epoetin or darbepoetin plus red blood cell transfusions (as necessary) versus red blood cell transfusions (as necessary) alone, to prevent or treat anemia in adult or pediatric cancer patients with or without concurrent antineoplastic therapy. DATA COLLECTION AND ANALYSIS: We performed a meta-analysis of randomized controlled trials comparing epoetin alpha, epoetin beta or darbepoetin alpha plus red blood cell transfusions versus transfusion alone, for prophylaxis or therapy of anemia while or after receiving anti-cancer treatment. Patient-level data were obtained and analyzed by independent statisticians at two academic departments, using fixed-effects and random-effects meta-analysis. Analyses were according to the intention-to-treat principle. Primary endpoints were on study mortality and overall survival during the longest available follow-up, regardless of anticancer treatment, and in patients receiving chemotherapy. Tests for interactions were used to identify differences in effects of ESAs on mortality across pre-specified subgroups. The present review reports only the results for the primary endpoint. MAIN RESULTS: A total of 13933 cancer patients from 53 trials were analyzed, 1530 patients died on-study and 4993 overall. ESAs increased on study mortality (combined hazard ratio [cHR] 1.17; 95% CI 1.06-1.30) and worsened overall survival (cHR 1.06; 95% CI 1.00-1.12), with little heterogeneity between trials (I(2) 0%, p=0.87 and I(2) 7.1%, p=0.33, respectively). Thirty-eight trials enrolled 10441 patients receiving chemotherapy. The cHR for on study mortality was 1.10 (95% CI 0.98-1.24) and 1.04; 95% CI 0.97-1.11) for overall survival. There was little evidence for a difference between trials of patients receiving different cancer treatments (P for interaction=0.42). AUTHORS' CONCLUSIONS: ESA treatment in cancer patients increased on study mortality and worsened overall survival. For patients undergoing chemotherapy the increase was less pronounced, but an adverse effect could not be excluded.

Assessing the Relevance of Anomie Theory for Explaining Spatial Variation in Lethal Criminal Violence: An Aggregate-Level Analysis of Homicide within the United States

One of the most influential statements in the anomie theory tradition has been Merton’s argument that the volume of instrumental property crime should be higher where there is a greater imbalance between the degree of commitment to monetary success goals and the degree of commitment to legitimate means of pursing such goals. Contemporary anomie theories stimulated by Merton’s perspective, most notably Messner and Rosenfeld’s institutional anomie theory, have expanded the scope conditions by emphasizing lethal criminal violence as an outcome to which anomie theory is highly relevant, and virtually all contemporary empirical studies have focused on applying the perspective to explaining spatial variation in homicide rates. In the present paper, we argue that current explications of Merton’s theory and IAT have not adequately conveyed the relevance of the core features of the anomie perspective to lethal violence. We propose an expanded anomie model in which an unbalanced pecuniary value system – the core causal variable in Merton’s theory and IAT – translates into higher levels of homicide primarily in indirect ways by increasing levels of firearm prevalence, drug market activity, and property crime, and by enhancing the degree to which these factors stimulate lethal outcomes. Using aggregate-level data collected during the mid-to-late 1970s for a sample of relatively large social aggregates within the U.S., we find a significant effect on homicide rates of an interaction term reflecting high levels of commitment to monetary success goals and low levels of commitment to legitimate means. Virtually all of this effect is accounted for by higher levels of property crime and drug market activity that occur in areas with an unbalanced pecuniary value system. Our analysis also reveals that property crime is more apt to lead to homicide under conditions of high levels of structural disadvantage. These and other findings underscore the potential value of elaborating the anomie perspective to explicitly account for lethal violence.

Long-term outcomes of biodegradable polymer versus durable polymer drug-eluting stents in patients with diabetes a pooled analysis of individual patient data from 3 randomized trials

BACKGROUND There is ongoing debate on the optimal drug-eluting stent (DES) in diabetic patients with coronary artery disease. Biodegradable polymer drug-eluting stents (BP-DES) may potentially improve clinical outcomes in these high-risk patients. We sought to compare long-term outcomes in patients with diabetes treated with biodegradable polymer DES vs. durable polymer sirolimus-eluting stents (SES). METHODS We pooled individual patient-level data from 3 randomized clinical trials (ISAR-TEST 3, ISAR-TEST 4 and LEADERS) comparing biodegradable polymer DES with durable polymer SES. Clinical outcomes out to 4years were assessed. The primary end point was the composite of cardiac death, myocardial infarction and target-lesion revascularization. Secondary end points were target lesion revascularization and definite or probable stent thrombosis. RESULTS Of 1094 patients with diabetes included in the present analysis, 657 received biodegradable polymer DES and 437 durable polymer SES. At 4years, the incidence of the primary end point was similar with BP-DES versus SES (hazard ratio=0.95, 95% CI=0.74-1.21, P=0.67). Target lesion revascularization was also comparable between the groups (hazard ratio=0.89, 95% CI=0.65-1.22, P=0.47). Definite or probable stent thrombosis was significantly reduced among patients treated with BP-DES (hazard ratio=0.52, 95% CI=0.28-0.96, P=0.04), a difference driven by significantly lower stent thrombosis rates with BP-DES between 1 and 4years (hazard ratio=0.15, 95% CI=0.03-0.70, P=0.02). CONCLUSIONS In patients with diabetes, biodegradable polymer DES, compared to durable polymer SES, were associated with comparable overall clinical outcomes during follow-up to 4years. Rates of stent thrombosis were significantly lower with BP-DES.

Safety and efficacy of drug-eluting stents in women: a patient-level pooled analysis of randomised trials

BACKGROUND The safety and efficacy of drug-eluting stents (DES) in the treatment of coronary artery disease have been assessed in several randomised trials. However, none of these trials were powered to assess the safety and efficacy of DES in women because only a small proportion of recruited participants were women. We therefore investigated the safety and efficacy of DES in female patients during long-term follow-up. METHODS We pooled patient-level data for female participants from 26 randomised trials of DES and analysed outcomes according to stent type (bare-metal stents, early-generation DES, and newer-generation DES). The primary safety endpoint was a composite of death or myocardial infarction. The secondary safety endpoint was definite or probable stent thrombosis. The primary efficacy endpoint was target-lesion revascularisation. Analysis was by intention to treat. FINDINGS Of 43,904 patients recruited in 26 trials of DES, 11,557 (26·3%) were women (mean age 67·1 years [SD 10·6]). 1108 (9·6%) women received bare-metal stents, 4171 (36·1%) early-generation DES, and 6278 (54·3%) newer-generation DES. At 3 years, estimated cumulative incidence of the composite of death or myocardial infarction occurred in 132 (12·8%) women in the bare-metal stent group, 421 (10·9%) in the early-generation DES group, and 496 (9·2%) in the newer-generation DES group (p=0·001). Definite or probable stent thrombosis occurred in 13 (1·3%), 79 (2·1%), and 66 (1·1%) women in the bare-metal stent, early-generation DES, and newer-generation DES groups, respectively (p=0·01). The use of DES was associated with a significant reduction in the 3 year rates of target-lesion revascularisation (197 [18·6%] women in the bare-metal stent group, 294 [7·8%] in the early-generation DES group, and 330 [6·3%] in the newer-generation DES group, p<0·0001). Results did not change after adjustment for baseline characteristics in the multivariable analysis. INTERPRETATION The use of DES in women is more effective and safe than is use of bare-metal stents during long-term follow-up. Newer-generation DES are associated with an improved safety profile compared with early-generation DES, and should therefore be thought of as the standard of care for percutaneous coronary revascularisation in women. FUNDING Women in Innovation Initiative of the Society of Cardiovascular Angiography and Interventions.

Long-term outcomes of biodegradable versus durable polymer drug-eluting stents in patients with acute ST-segment elevation myocardial infarction: a pooled analysis of individual patient data from three randomised trials

Aims: Arterial plaque rupture and thrombus characterise ST-elevation myocardial infarction (STEMI) and may aggravate delayed arterial healing following durable polymer drug-eluting stent (DP-DES) implantation. Biodegradable polymer (BP) may improve biocompatibility. We compared long-term outcomes in STEMI patients receiving BP-DES vs. durable polymer sirolimus-eluting stents (DP-SES). Methods and results: We pooled individual patient-level data from three randomised clinical trials (ISAR-TEST-3, ISAR-TEST-4 and LEADERS) comparing outcomes from BP-DES with DP-SES at four years. The primary endpoint (MACE) comprised cardiac death, MI, or target lesion revascularisation (TLR). Secondary endpoints were TLR, cardiac death or MI, and definite or probable stent thrombosis. Of 497 patients with STEMI, 291 received BP-DES and 206 DP-SES. At four years, MACE was significantly reduced following treatment with BP-DES (hazard ratio [HR] 0.59, 95% CI: 0.39-0.90; p=0.01) driven by reduced TLR (HR 0.54, 95% CI: 0.30-0.98; p=0.04). Trends towards reduction were seen for cardiac death or MI (HR 0.63, 95% CI: 0.37-1.05; p=0.07) and definite or probable stent thrombosis (3.6% vs. 7.1%; HR 0.49, 95% CI: 0.22-1.11; p=0.09). Conclusions: In STEMI, BP-DES demonstrated superior clinical outcomes to DP-SES at four years. Trends towards reduced cardiac death or myocardial infarction and reduced stent thrombosis require corroboration in specifically powered trials.

Prognostic implications of coronary calcification in patients with obstructive coronary artery disease treated by percutaneous coronary intervention: a patient-level pooled analysis of 7 contemporary stent trials

OBJECTIVE To investigate the long-term prognostic implications of coronary calcification in patients undergoing percutaneous coronary intervention for obstructive coronary artery disease. METHODS Patient-level data from 6296 patients enrolled in seven clinical drug-eluting stents trials were analysed to identify in angiographic images the presence of severe coronary calcification by an independent academic research organisation (Cardialysis, Rotterdam, The Netherlands). Clinical outcomes at 3-years follow-up including all-cause mortality, death-myocardial infarction (MI), and the composite end-point of all-cause death-MI-any revascularisation were compared between patients with and without severe calcification. RESULTS Severe calcification was detected in 20% of the studied population. Patients with severe lesion calcification were less likely to have undergone complete revascularisation (48% vs 55.6%, p<0.001) and had an increased mortality compared with those without severely calcified arteries (10.8% vs 4.4%, p<0.001). The event rate was also high in patients with severely calcified lesions for the combined end-point death-MI (22.9% vs 10.9%; p<0.001) and death-MI- any revascularisation (31.8% vs 22.4%; p<0.001). On multivariate Cox regression analysis, including the Syntax score, the presence of severe coronary calcification was an independent predictor of poor prognosis (HR: 1.33 95% CI 1.00 to 1.77, p=0.047 for death; 1.23, 95% CI 1.02 to 1.49, p=0.031 for death-MI, and 1.18, 95% CI 1.01 to 1.39, p=0.042 for death-MI- any revascularisation), but it was not associated with an increased risk of stent thrombosis. CONCLUSIONS Patients with severely calcified lesions have worse clinical outcomes compared to those without severe coronary calcification. Severe coronary calcification appears as an independent predictor of worse prognosis, and should be considered as a marker of advanced atherosclerosis.

Comparative Effectiveness and Safety of New-Generation Versus Early-Generation Drug-Eluting Stents According to Complexity of Coronary Artery Disease: A Patient-Level Pooled Analysis of 6,081 Patients.

OBJECTIVES The purpose of this study was to compare the 2-year safety and effectiveness of new- versus early-generation drug-eluting stents (DES) according to the severity of coronary artery disease (CAD) as assessed by the SYNTAX (Synergy between Percutaneous Coronary Intervention with Taxus and Cardiac Surgery) score. BACKGROUND New-generation DES are considered the standard-of-care in patients with CAD undergoing percutaneous coronary intervention. However, there are few data investigating the effects of new- over early-generation DES according to the anatomic complexity of CAD. METHODS Patient-level data from 4 contemporary, all-comers trials were pooled. The primary device-oriented clinical endpoint was the composite of cardiac death, myocardial infarction, or ischemia-driven target-lesion revascularization (TLR). The principal effectiveness and safety endpoints were TLR and definite stent thrombosis (ST), respectively. Adjusted hazard ratios (HRs) with 95% confidence intervals (CIs) were calculated at 2 years for overall comparisons, as well as stratified for patients with lower (SYNTAX score ≤11) and higher complexity (SYNTAX score >11). RESULTS A total of 6,081 patients were included in the study. New-generation DES (n = 4,554) compared with early-generation DES (n = 1,527) reduced the primary endpoint (HR: 0.75 [95% CI: 0.63 to 0.89]; p = 0.001) without interaction (p = 0.219) between patients with lower (HR: 0.86 [95% CI: 0.64 to 1.16]; p = 0.322) versus higher CAD complexity (HR: 0.68 [95% CI: 0.54 to 0.85]; p = 0.001). In patients with SYNTAX score >11, new-generation DES significantly reduced TLR (HR: 0.36 [95% CI: 0.26 to 0.51]; p < 0.001) and definite ST (HR: 0.28 [95% CI: 0.15 to 0.55]; p < 0.001) to a greater extent than in the low-complexity group (TLR pint = 0.059; ST pint = 0.013). New-generation DES decreased the risk of cardiac mortality in patients with SYNTAX score >11 (HR: 0.45 [95% CI: 0.27 to 0.76]; p = 0.003) but not in patients with SYNTAX score ≤11 (pint = 0.042). CONCLUSIONS New-generation DES improve clinical outcomes compared with early-generation DES, with a greater safety and effectiveness in patients with SYNTAX score >11.

Anatomic characteristics and clinical implications of angiographic coronary thrombus: insights from a patient-level pooled analysis of SYNTAX, RESOLUTE, and LEADERS Trials.

BACKGROUND The distribution of thrombus-containing lesions (TCLs) in an all-comer population admitted with a heterogeneous clinical presentation (stable, ustable angina, or an acute coronary syndrome) and treated with percutaneous coronary intervention is yet unclear, and the long-term prognostic implications are still disputed. This study sought to assess the distribution and prognostic implications of coronary thrombus, detected by coronary angiography, in a population recruited in all-comer percutaneous coronary intervention trials. METHODS AND RESULTS Patient-level data from 3 contemporary coronary stent trials were pooled by an independent academic research organization (Cardialysis, Rotterdam, the Netherlands). Clinical outcomes in terms of major adverse cardiac events (major adverse cardiac events, a composite of death, myocardial infarction, and repeat revascularization), death, myocardial infarction, and repeated revascularization were compared between patients with and without angiographic TCL. Preprocedural TCL was present in 257 patients (5.8%) and absent in 4193 (94.2%) patients. At 3-year follow-up, there was no difference for major adverse cardiac events (25.3 versus 25.4%; P=0.683); all-cause death (7.4 versus 6.8%; P=0.683); myocardial infarction (5.8 versus 6.0%; P=0.962), and any revascularizations (17.5 versus 17.7%; P=0.822) between patients with and without TCL. The comparison of outcomes in groups weighing the jeopardized myocardial by TCL also did not show a significant difference. TCL were seen more often in the first 2 segments of the right (43.6%) and left anterior descending (36.8%) coronary arteries. The association of TCL and bifurcation lesions was present in 40.1% of the prespecified segments. CONCLUSIONS TCL involved mainly the proximal coronary segments and did not have any effect on clinical outcomes. A more detailed thrombus burden quantification is required to investigate its prognostic implications. CLINICAL TRIAL REGISTRATION URL: http://www.clinicaltrials.gov. Unique identifiers: NCT00114972, NCT01443104, NCT00617084.

Effect of Chronic Kidney Disease in Women Undergoing Percutaneous Coronary Intervention With Drug-Eluting Stents: A Patient-Level Pooled Analysis of Randomized Controlled Trials.

OBJECTIVES This study sought to evaluate: 1) the effect of impaired renal function on long-term clinical outcomes in women undergoing percutaneous coronary intervention (PCI) with drug-eluting stent (DES); and 2) the safety and efficacy of new-generation compared with early-generation DES in women with chronic kidney disease (CKD). BACKGROUND The prevalence and effect of CKD in women undergoing PCI with DES is unclear. METHODS We pooled patient-level data for women enrolled in 26 randomized trials. The study population was categorized by creatinine clearance (CrCl) <45 ml/min, 45 to 59 ml/min, and ≥60 ml/min. The primary endpoint was the 3-year rate of major adverse cardiovascular events (MACE). Participants for whom baseline creatinine was missing were excluded from the analysis. RESULTS Of 4,217 women included in the pooled cohort treated with DES and for whom serum creatinine was available, 603 (14%) had a CrCl <45 ml/min, 811 (19%) had a CrCl 45 to 59 ml/min, and 2,803 (66%) had a CrCl ≥60 ml/min. A significant stepwise gradient in risk for MACE was observed with worsening renal function (26.6% vs. 15.8% vs. 12.9%; p < 0.01). Following multivariable adjustment, CrCl <45 ml/min was independently associated with a higher risk of MACE (adjusted hazard ratio: 1.56; 95% confidence interval: 1.23 to 1.98) and all-cause mortality (adjusted hazard ratio: 2.67; 95% confidence interval: 1.85 to 3.85). Compared with older-generation DES, the use of newer-generation DES was associated with a reduction in the risk of cardiac death, myocardial infarction, or stent thrombosis in women with CKD. The effect of new-generation DES on outcomes was uniform, between women with or without CKD, without evidence of interaction. CONCLUSIONS Among women undergoing PCI with DES, CKD is a common comorbidity associated with a strong and independent risk for MACE that is durable over 3 years. The benefits of newer-generation DES are uniform in women with or without CKD.

Safety and Efficacy of New-Generation Drug-Eluting Stents in Women at High Risk for Atherothrombosis: From the Women in Innovation and Drug-Eluting Stents Collaborative Patient-Level Pooled Analysis

BACKGROUND The safety and efficacy of new-generation drug-eluting stents (DES) in women with multiple atherothrombotic risk (ATR) factors is unclear. METHODS AND RESULTS We pooled patient-level data for women enrolled in 26 randomized trials. Study population was categorized based on the presence or absence of high ATR, which was defined as having history of diabetes mellitus, prior percutaneous or surgical coronary revascularization, or prior myocardial infarction. The primary end point was major adverse cardiovascular events defined as a composite of all-cause mortality, myocardial infarction, or target lesion revascularization at 3 years of follow-up. Out of 10 449 women included in the pooled database, 5333 (51%) were at high ATR. Compared with women not at high ATR, those at high ATR had significantly higher risk of major adverse cardiovascular events (15.8% versus 10.6%; adjusted hazard ratio: 1.53; 95% confidence interval: 1.34-1.75; P=0.006) and all-cause mortality. In high-ATR risk women, the use of new-generation DES was associated with significantly lower risk of 3-year major adverse cardiovascular events (adjusted hazard ratio: 0.69; 95% confidence interval: 0.52-0.92) compared with early-generation DES. The benefit of new-generation DES on major adverse cardiovascular events was uniform between high-ATR and non-high-ATR women, without evidence of interaction (Pinteraction=0.14). At landmark analysis, in high-ATR women, stent thrombosis rates were comparable between DES generations in the first year, whereas between 1 and 3 years, stent thrombosis risk was lower with new-generation devices. CONCLUSIONS Use of new-generation DES even in women at high ATR is associated with a benefit consistent over 3 years of follow-up and a substantial improvement in very-late thrombotic safety.

A COMPARISON OF NEONATAL MORTALITY BETWEEN HOSPITALS WITH LEVEL III NEWBORN INTENSIVE CARE UNITS AND ALL OTHER HOUSTON HOSPITALS

A number of medical and social developments have had an impact on the neonatal mortality over the past ten to 15 years in the United States. The purpose of this study was to examine one of these developments, Newborn Intensive Care Units (NICUs), and evaluate their impact on neonatal mortality in Houston, Texas.^ This study was unique in that it used as its data base matched birth and infant death records from two periods of time: 1958-1960 (before NICUs) and 1974-1976 (after NICUs). The neonatal mortality of single, live infants born to Houston resident mothers was compared for two groups: infants born in hospitals which developed NICUs and infants born in all other Houston hospitals. Neonatal mortality comparisons were made using the following birth-characteristic variables: birthweight, gestation, race, sex, maternal age, legitimacy, birth order and prenatal care.^ The results of the study showed that hospitals which developed NICUs had a higher percentage of their population with high risk characteristics. In spite of this, they had lower neonatal mortality rates in two categories: (1) white 3.5-5.5 pounds birthweight infants, (2) low birthweight infants whose mothers received no prenatal care. Black 3.5-5.5 pounds birthweight infants did equally well in either hospital group. While the differences between the two hospital groups for these categories were not statistically significant at the p < 0.05 level, data from the 1958-1960 period substantiate that a marked change occurred in the 3.5-5.5 pounds birthweight category for those infants born in hospitals which developed NICUs. Early data were not available for prenatal care. These findings support the conclusion that, in Houston, NICUs had some impact on neonatal mortality among moderately underweight infants. ^