An EM-based Semi-Parametric Mixture Model Approach to the Regression Analysis of Competing-Risks Data

We consider a mixture model approach to the regression analysis of competing-risks data. Attention is focused on inference concerning the effects of factors on both the probability of occurrence and the hazard rate conditional on each of the failure types. These two quantities are specified in the mixture model using the logistic model and the proportional hazards model, respectively. We propose a semi-parametric mixture method to estimate the logistic and regression coefficients jointly, whereby the component-baseline hazard functions are completely unspecified. Estimation is based on maximum likelihood on the basis of the full likelihood, implemented via an expectation-conditional maximization (ECM) algorithm. Simulation studies are performed to compare the performance of the proposed semi-parametric method with a fully parametric mixture approach. The results show that when the component-baseline hazard is monotonic increasing, the semi-parametric and fully parametric mixture approaches are comparable for mildly and moderately censored samples. When the component-baseline hazard is not monotonic increasing, the semi-parametric method consistently provides less biased estimates than a fully parametric approach and is comparable in efficiency in the estimation of the parameters for all levels of censoring. The methods are illustrated using a real data set of prostate cancer patients treated with different dosages of the drug diethylstilbestrol. Copyright (C) 2003 John Wiley Sons, Ltd.

Arrhythmias and mortality after myocardial infarction in diabetic patients - Relationship to diabetes treatment

OBJECTIVE- To assess the relationship between clinical course after acute myocardial infarction (AMI) and diabetes treatment. RESEARCH DESIGN AND METHODS- Retrospective analysis of data from all patients aged 25-64 years admitted to hospitals in Perth, Australia, between 1985 and 1993 with AMI diagnosed according to the International Classification of Diseases (9th revision) criteria was conducted. Short- (28-day) and long-term survival and complications in diabetic and nondiabetic patients were compared. For diabetic patients, 28-day survival, dysrhythmias, heart block, and pulmonary edema were treated as outcomes, and factors related to each were assessed using multiple logistic regression. Diabetes treatment was added to the model to assess its significance. Long-term survival was compared by means of a Cox proportional hazards model. RESULTS- Of 5,715 patients, 745 (12.9%) were diabetic. Mortality at 28 days was 12.0 and 28.1% for nondiabetic and diabetic patients, respectively (P < 0.001); there were no significant drug effects in the diabetic group. Ventricular fibrillation in diabetic patients taking glibenclamide (11.8%) was similar to that of nondiabetic patients (11.0%) but was lower than that for those patients taking either gliclazide (18.0%; 0.1 > P > 0.05) or insulin (22.8%; P < 0.05). There were no other treatment-related differences in acute complications. Long-term survival in diabetic patients was reduced in those taking digitalis and/or diuretics but type of diabetes treatment at discharge had no significant association with outcome. CONCLUSlONS- These results do not suggest that ischemic heart disease should influence the choice of diabetes treatment regimen in general or of sulfonylurea drug in particular.

A Randomized Trial of Sodium Fluoride (60 mg) ± Estrogen in Postmenopausal Osteoporotic Vertebral Fractures

Postmenopausal Caucasian women aged less than 80 years (n = 99) with one or more atraumatic vertebral fracture and no hip fractures, were treated by cyclical administration of enteric coated sodium fluoride (NaF) or no NaF for 27 months, with precautions to prevent excessive stimulation of bone turnover. In the first study 65 women, unexposed to estrogen (-E study), age 70.8 +/- 0.8 years (mean SEM) were all treated with calcium (Ca) 1.0-1.2 g daily and ergocalciferol (D) 0.25 mg per 25 kg once weekly and were randomly assigned to cyclical NaF (6 months on. 3 months off, initial dose 60 mg/day; group F CaD, n = 34) or no NaF (group CaD, n = 3 1). In the second study 34 patients. age 65.5 +/- 1.2 years, on hormone replacement therapy (E) at baseline, had this standardized, and were all treated with Ca and D and similarly randomized (FE CaD, n = 17, E CaD, n = 17) (+E study). The patients were stratified according to E status and subsequently assigned randomly to NaF. Seventy-five patients completed the trial. Both groups treated with NaF showed an increase in lumbar spinal density (by DXA) above baseline by 27 months: FE CaD + 16.2% and F CaD +9.3% (both p = 0.0001). In neither group CaD nor E CaD did lumbar spinal density increase. Peripheral bone loss occurred at most sites in the F CaD group at 27 months: tibia/fibula shaft -7.3% (p = 0.005); femoral shaft -7.1% (p = 0.004); distal forearm -4.0% (p = 0.004); total hip -4.1% (p = 0. 003); and femoral neck -3.5% (p = 0.006). No significant loss occurred in group FE CaD. Differences between the two NaF groups were greatest at the total hip at 27 months but were not significant [p < 0.05; in view of the multiple bone mineral density (BMD) sites, an alpha of 0.01 was employed to denote significance in BMD changes throughout this paper]. Using Cox's proportional hazards model, in the -E study there were significantly more patients with first fresh vertebral fractures in those treated with NaF than in those not so treated (RR = 24.2, p = 0.008, 95% CI 2.3-255). Patients developing first fresh fractures in the first 9 months were markedly different between groups: -23% of F CaD, 0 of CaD, 29% of FE CaD and 0 of E CaD. The incidence of incomplete (stress) fractures was similar in the two NaF-treated groups. Complete nonvertebral fractures did not occur in the two +E groups, there were no differences between groups F CaD and CaD. Baseline BMD (spine and femoral neck) was related to incident vertebral fractures in the control groups (no NaF), but not in the two NaF groups. Our results and a literature review indicate that fluoride salts. if used, should be at low dosage, with pretreatment and co-treatment with a bone resorption inhibitor.

Optimal patch-leaving behaviour: a case study using the parasitoid Cotesia rubecular

1. Parasitoids are predicted to spend longer in patches with more hosts, but previous work on Cotesia rubecula (Marshall) has not upheld this prediction, Tests of theoretical predictions may be affected by the definition of patch leaving behaviour, which is often ambiguous. 2. In this study whole plants were considered as patches and assumed that wasps move within patches by means of walking or flying. Within-patch and between-patch flights were distinguished based on flight distance. The quality of this classification was tested statistically by examination of log-survivor curves of flight times. 3. Wasps remained longer in patches with higher host densities, which is consistent with predictions of the marginal value theorem (Charnov 1976). tinder the assumption that each flight indicates a patch departure, there is no relationship between host density and leaving tendency. 4. Oviposition influences the patch leaving behaviour of wasps in a count down fashion (Driessen et al. 1995), as predicted by an optimal foraging model (Tenhumberg, Keller & Possingham 2001). 5. Wasps spend significantly longer in the first patch encountered following release, resulting in an increased rate of superparasitism.

Randomized, controlled trial of topical exit-site application of honey (Medihoney) versus Mupirocin for the prevention of catheter-associated infections in hemodialysis patients

The clinical usefulness of hemodialysis catheters is limited by increased infectious morbidity and mortality. Topical antiseptic agents, such as mupirocin, are effective at reducing this risk but have been reported to select for antibiotic-resistant strains. The aim of the present study was to determine the efficacy and the safety of exit-site application of a standardized antibacterial honey versus mupirocin in preventing catheter-associated infections. A randomized, controlled trial was performed comparing the effect of thrice-weekly exit-site application of Medihoney versus mupirocin on infection rates in patients who were receiving hemodialysis via tunneled, cuffed central venous catheters. A total of 101 patients were enrolled. The incidences of catheter-associated bacteremias in honey-treated (n = 51) and mupirocin-treated (n = 50) patients were comparable (0.97 versus 0.85 episodes per 1000 catheter-days, respectively; NS). On Cox proportional hazards model analysis, the use of honey was not significantly associated with bacteremia-free survival (unadjusted hazard ratio, 0.94; 95% confidence interval, 0.27 to 3.24; P = 0.92). No exit-site infections occurred. During the study period, 2% of staphylococcal isolates within the hospital were mupirocin resistant. Thrice-weekly application of standardized antibacterial honey to hemodialysis catheter exit sites was safe, cheap, and effective and resulted in a comparable rate of catheter-associated infection to that obtained with mupirocin (although the study was not adequately powered to assess therapeutic equivalence). The effectiveness of honey against antibiotic-resistant microorganisms and its low likelihood of selecting for further resistant strains suggest that this agent may represent a satisfactory alternative means of chemoprophylaxis in patients with central venous catheters.

Platelet nitric oxide responsiveness - A novel prognostic marker in acute coronary syndromes

Objectives - Nitric oxide (NO) is critically important in the regulation of vascular tone and the inhibition of platelet aggregation. We have shown previously that patients with acute coronary syndromes (ACS) or stable angina pectoris have impaired platelet responses to NO donors when compared with normal subjects. We tested the hypotheses that platelet hyporesponsiveness to NO is a predictor of (1) cardiovascular readmission and/or death and (2) all-cause mortality in patients with ACS (unstable angina pectoris or non-Q-wave myocardial infarction). Methods and Results - Patients (n = 51) with ACS had evaluation of platelet aggregation within 24 hours of coronary care unit admission using impedance aggregometry. Patients were categorized as having normal (>= 32% inhibition of ADP-induced aggregation with the NO donor sodium nitroprusside; 10 mu mol/L; n = 18) or impaired (>= 32% inhibition of ADP-induced aggregation; n = 33) NO responses. We then compared the incidence of cardiovascular readmission and death during a median of 7 years of follow-up in these 2 groups. Using a Cox proportional hazards model adjusting for age, sex, index event, postdischarge medical treatment, revascularization status, left ventricular systolic dysfunction, concurrent disease states, and cardiac risk factors, impaired NO responsiveness was associated with an increased risk of the combination of cardiovascular readmission and/or death (relative risk, 2.7; 95% CI, 1.03 to 7.10; P = 0.041) and all-cause mortality (relative risk, 6.3; 95% CI, 1.09 to 36.7; P = 0.033). Conclusions - Impaired platelet NO responsiveness is a novel, independent predictor of increased mortality and cardiovascular morbidity in patients with high-risk ACS.

Prognostic value of intraventricular dyssynchrony according to clinical stage of left ventricular impairment

Intraventricular dyssynchrony has prognostic implications in patients who have severe functional limitation and decreased ejection fraction. Patients with less advanced cardiac disease often exhibit intraventricular dyssynchrony, but there is little available information about its prognostic relevance in such patients. We investigated the prognostic effect of intraventricular dyssynchrony on outcome in 318 patients with known or suspected coronary artery disease who were classified according to the presence or absence of left ventricular dysfunction and heart failure symptoms. Mortality was considered the primary end point over a median follow-up of 56 months, and a Cox proportional hazards model was used for survival analysis. Despite a low prevalence (8%) of left bundle branch block, there was a high prevalence of intraventricular dyssynchrony even in patients without symptomatic heart failure. The magnitude of intraventricular dyssynchrony correlated poorly with QRS duration (r = 0.25),end-systolic volume index (r = 0.27), and number of scar segments (r = 0.25). There,were 58 deaths during follow-up. Ventricular volume, ischemic burden, and magnitude of intraventricular dyssynchrony predicted outcome, but magnitude of intraventricular dyssynchrony was an independent predictor of survival only in patients with asymptomatic left ventricular dysfunction. In conclusion, patients with known or suspected coronary artery disease have a high prevalence of intraventricular dyssynchrony. Although ventricular volume, ischemic burden, and intraventricular dyssynchrony are potentially important prognostic markers, the relative importance of intraventricular dyssynchrony changes with the clinical setting and, may be greatest-in patients with preclinical disease. (c) 2006 Elsevier Inc. All rights reserved.

Long-term survival of women with breast cancer in New South Wales

Several long-term studies of breast cancer survival have shown continued excess mortality from breast cancer up to 20-40 years following treatment. The purpose of this report was to investigate temporal trends in long-term survival from breast cancer in all New South Wales (NSW) women. Breast cancer cases incident in 1972-1996 (54,228) were derived from the NSW Central Cancer Registry a population-based registry which began in 1972. All cases of breast cancer not known to be dead were matched against death records. The expected survival for NSW women was derived from published annual life tables. Relative survival analysis compared the survival of cancer cases with the age, sex and period matched mortality of the total population. Cases were considered alive at the end of 1996, except when known to be dead. Proportional hazards regression was employed to model survival on age, period and degree of spread at diagnosis. Survival at 5, 10, 15, 20 and 25 years of follow-up was 76 per cent, 65 per cent, 60 per cent, 57 per cent and 56 per cent. The annual hazard rate for excess mortality was 4.3 per cent in year 1, maximal at 6.5 per cent in year 3, declining to 4.7 per cent in year 5, 2.7 per cent in year 10, 1.4 per cent in year 15, 1.0 per cent for years 16-20, and 0.4 per cent for years 20-25 of follow-up. Relative survival was highest in 40-49 year-olds. Cases diagnosed most recently (1992-1996) had the highest survival, compared with cases diagnosed in previous periods. Five-year survival improved over time, especially from the late 1980s for women in the screening age group (50-69 years). Survival was highest for those with localised cancer at diagnosis: 88.4 per cent, 79.1 per cent, 74.6 per cent, 72.7 per cent and 72.8 per cent at 5, 10, 15, 20 and 25 years follow-up (excluding those aged greater than or equal to 70 years). There was no significant difference between the survival of the breast cancer cases and the general population at 20-25 years follow-up. Degree of spread was less predictive of survival 5-20 years after diagnosis, compared with 0-5 years after diagnosis, and was not significant at 20-25 years of follow-up. Relative survival from breast cancer in NSW women continues to decrease to 25 years after diagnosis, but there is little excess mortality after 15 years follow-up, especially for those with localised cancer at diagnosis, and the minimal excess mortality at 20-25 years of follow-up is not statistically significant. (C) 2002 Elsevier Science Ltd. All rights reserved.

Modelling the Distribution of Ischaemic Stroke-Specific Survival Time Using an EM-based Mixture Approach with Random Effects Adjustment

A two-component survival mixture model is proposed to analyse a set of ischaemic stroke-specific mortality data. The survival experience of stroke patients after index stroke may be described by a subpopulation of patients in the acute condition and another subpopulation of patients in the chronic phase. To adjust for the inherent correlation of observations due to random hospital effects, a mixture model of two survival functions with random effects is formulated. Assuming a Weibull hazard in both components, an EM algorithm is developed for the estimation of fixed effect parameters and variance components. A simulation study is conducted to assess the performance of the two-component survival mixture model estimators. Simulation results confirm the applicability of the proposed model in a small sample setting. Copyright (C) 2004 John Wiley Sons, Ltd.

Clinical and economic impact of exercise electrocardiography and exercise echocardiography in clinical practice

Background Patients with known or suspected coronary disease are often investigated to facilitate risk assessment. We sought to examine the cost-effectiveness of strategies based on exercise echocardiography and exercise electrocardiography. Methods and results We studied 7656 patients undergoing exercise testing; of whom half underwent exercise echocardiography. Risk was defined with the Duke treadmill score for those undergoing exercise electrocardiography alone, and by the extent of ischaemia by exercise echocardiography. Cox proportional hazards models, risk adjusted for pretest likelihood of coronary artery disease, were used to estimate time to cardiac death or myocardial infarction. Costs (including diagnostic and revascularisation procedures, hospitalisations, and events) were calculated, inflation-corrected to year 2000 using Medicare trust fund rates and discounted at a rate of 5%. A decision model was employed to assess the marginal cost effectiveness (cost/life year saved) of exercise echo compared with exercise electrocardiography. Exercise echocardiography identified more patients as low-risk (51% vs 24%, p<0.001), and fewer as intermediate- (27% vs 51%, p<0.001) and high-risk (22% vs 4%); survival was greater in low- and intermediate- risk and less in high-risk patients. Although initial procedural costs and revascularisation costs (in intermediate- high risk patients) were greater, exercise echocardiography was associated with a greater incremental life expectancy (0.2 years) and a lower use of additional diagnostic procedures when compared with exercise electrocardiography (especially in lower risk patients). Using decision analysis, exercise echocardiography (Euro 2615/life year saved) was more cost effective than exercise electrocardiography. Conclusion Exercise echocardiography may enhance cost-effectiveness for the detection and management of at risk patients with known or suspected coronary disease. (C) 2003 Published by Elsevier Science Ltd on behalf of The European Society of Cardiology.

Analysis of Melanoma Onset: Assessing Familial Aggregation by Using Estimating Equations and Fitting Variance Components via Bayesian Random Effects Models

We investigate whether relative contributions of genetic and shared environmental factors are associated with an increased risk in melanoma. Data from the Queensland Familial Melanoma Project comprising 15,907 subjects arising from 1912 families were analyzed to estimate the additive genetic, common and unique environmental contributions to variation in the age at onset of melanoma. Two complementary approaches for analyzing correlated time-to-onset family data were considered: the generalized estimating equations (GEE) method in which one can estimate relationship-specific dependence simultaneously with regression coefficients that describe the average population response to changing covariates; and a subject-specific Bayesian mixed model in which heterogeneity in regression parameters is explicitly modeled and the different components of variation may be estimated directly. The proportional hazards and Weibull models were utilized, as both produce natural frameworks for estimating relative risks while adjusting for simultaneous effects of other covariates. A simple Markov Chain Monte Carlo method for covariate imputation of missing data was used and the actual implementation of the Bayesian model was based on Gibbs sampling using the free ware package BUGS. In addition, we also used a Bayesian model to investigate the relative contribution of genetic and environmental effects on the expression of naevi and freckles, which are known risk factors for melanoma.

Repeated occurrence of basal cell carcinoma of the skin and multifailure survival analysis: Follow-up data from the nambour skin cancer prevention trial

The aim of this study was to apply multifailure survival methods to analyze time to multiple occurrences of basal cell carcinoma (BCC). Data from 4.5 years of follow-up in a randomized controlled trial, the Nambour Skin Cancer Prevention Trial (1992-1996), to evaluate skin cancer prevention were used to assess the influence of sunscreen application on the time to first BCC and the time to subsequent BCCs. Three different approaches of time to ordered multiple events were applied and compared: the Andersen-Gill, Wei-Lin-Weissfeld, and Prentice-Williams-Peterson models. Robust variance estimation approaches were used for all multifailure survival models. Sunscreen treatment was not associated with time to first occurrence of a BCC (hazard ratio = 1.04, 95% confidence interval: 0.79, 1.45). Time to subsequent BCC tumors using the Andersen-Gill model resulted in a lower estimated hazard among the daily sunscreen application group, although statistical significance was not reached (hazard ratio = 0.82, 95% confidence interval: 0.59, 1.15). Similarly, both the Wei-Lin-Weissfeld marginal-hazards and the Prentice-Williams-Peterson gap-time models revealed trends toward a lower risk of subsequent BCC tumors among the sunscreen intervention group. These results demonstrate the importance of conducting multiple-event analysis for recurring events, as risk factors for a single event may differ from those where repeated events are considered.

Predictors of death and vascular events in the elderly - The Perth Community Stroke Study

Background and Purpose-The goal of the present study was to identify risk factors for vascular disease in the elderly. Methods-We conducted a prospective study of control subjects from a population-based study of stroke in Perth, Western Australia, that was completed in 1989 to 1990 and used record linkage and a survey of survivors to identify deaths and nonfatal vascular events. Data validated through reference to medical records were analyzed with the use of Cox proportional hazards models. Results-Follow-up for the 931 subjects was 88% complete. By June 24, 1994, 198 (24%) of the subjects had died (96 from vascular disease), and there had been 45 nonfatal strokes or myocardial infarctions. The hazard ratio for diabetes exceeded 2.0 for all end points, whereas the consumption of meat >4 times weekly was associated with a reduction in risk of less than or equal to 30%. In most models, female sex and consumption of alcohol were associated with reduced risks, whereas previous myocardial infarction was linked to an increase in risk. Conclusions-There are only limited associations between lifestyle and major vascular illness in old age. Effective health promotion activities in early and middle life may be the key to a longer and healthier old age.

Icodextrin as salvage therapy in peritoneal dialysis patients with refractory fluid overload

Background Icodextrin is a high molecular weight, starch-derived glucose polymer, which is capable of inducing sustained ultrafiltration over prolonged (12–16 hour) peritoneal dialysis (PD) dwells. The aim of this study was to evaluate the ability of icodextrin to alleviate refractory, symptomatic fluid overload and prolong technique survival in PD patients. Methods A prospective, open-label, pre-test/post-test study was conducted in 17 PD patients (8 females/9 males, mean age 56.8 ± 2.9 years) who were on the verge of being transferred to haemodialysis because of symptomatic fluid retention that was refractory to fluid restriction, loop diuretic therapy, hypertonic glucose exchanges and dwell time optimisation. One icodextrin exchange (2.5 L 7.5%, 12-hour dwell) was substituted for a long-dwell glucose exchange each day. Results Icodextrin significantly increased peritoneal ultrafiltration (885 ± 210 ml to 1454 ± 215 ml, p < 0.05) and reduced mean arterial pressure (106 ± 4 to 96 ± 4 mmHg, p < 0.05), but did not affect weight, plasma albumin concentration, haemoglobin levels or dialysate:plasma creatinine ratio. Diabetic patients (n = 12) also experienced improved glycaemic control (haemoglobin Alc decreased from 8.9 ± 0.7% to 7.9 ± 0.7%, p < 0.05). Overall PD technique survival was prolonged by a mean of 11.6 months (95% CI 6.0–17.3 months). On multivariate Cox proportional hazards analysis, extension of technique survival by icodextrin was only significantly predicted by baseline net daily peritoneal ultrafiltration (adjusted HR 2.52, 95% CI 1.13–5.62, p < 0.05). Conclusions Icodextrin significantly improved peritoneal ultrafiltration and extended technique survival in PD patients with symptomatic fluid overload, especially those who had substantially impaired peritoneal ultrafiltration.

Acute symptoms, not rectally administered sucralfate, predict for late radiation proctitis: Longer term follow-up of a phase III trial - trans-Tasman radiation oncology group

Image : To assess the potential for sucralfate administered rectally to reduce the risk of late rectal morbidity in patients undergoing nonconformal radiotherapy (RT) for carcinoma of the prostate and to study the variables potentially contributing to late rectal morbidity and particularly to explore the relationship between acute and late toxicity. Image : Eighty-six patients with localized prostate carcinoma were randomized in a double-blind, placebo-controlled study to a daily enema of 3 g of sucralfate in a 15-mL suspension or the same suspension without sucralfate. The enema began the first day of RT and was continued for 2 weeks after treatment completion. The primary end point of the study was acute Radiation Therapy Oncology Group (RTOG)/European Organization for Research and Treatment of Cancer (EORTC) toxicity; however, the patients were followed for an additional 5 years on a 6-month basis. The evaluation included late RTOG/EORTC toxicity and a patient self-assessment questionnaire. Image : With a median follow-up of 5 years, the Kaplan-Meier probability of late Grade 2 RTOG/EORTC toxicity was 12% (95% confidence interval [CI] 2–22%) for placebo and 5% (95% CI 0–12%) for sucralfate (p = 0.26). The probability of late rectal bleeding was 59% (95% CI 45–73%) for placebo and 54% (95% CI 40–68%) for sucralfate. No statistically significant difference was found between the treatment arms for the peak incidence of any of the other patient self-assessment variables. Cox proportional hazards modeling indicated acute RTOG/EORTC toxicity of Grade 2 or greater was associated with a hazard ratio of 2.74 (95% CI 1.31–5.73) for the development of late toxicity of Grade 1 or greater. Substituting the patient self-assessment variables for acute RTOG/EORTC toxicity revealed that rectal pain of a moderate or severe grade during RT was the best predictor of the subsequent development of late toxicity, with a hazard ratio of 3.44 (95% CI 1.68–7). Image : The results of this study do not support the use of sucralfate administered rectally as a method for reducing the late toxicity of nonconformal RT for prostate cancer. There appears to be an association between the development of acute and subsequent late toxicity, although the nature of this association remains to be determined