In Proceedings of the Wosp 07 Sixth International Workshop on Software and Performance February 5 8 2007 Buenos Aires Argentina Acm Press New York Us, 2007
... We will now re-strict our analysis of application service provision to systems of SLAs govern... more ... We will now re-strict our analysis of application service provision to systems of SLAs governing the timeliness of service provision ... We therefore define events corresponding to the completion of each of the events in the scenario E = {x, y, z, w} in relation to our actions such that ...
Standard models for the analysis of repeated measurements assume a common response profile for al... more Standard models for the analysis of repeated measurements assume a common response profile for all experimental units within a treatment group. However, in many applications this under-represents the nature of the response. There may be several distinct modes of response within a group (for example, responders versus non-responders to a given treatment), or there may be a set of distinct response profiles which are common to all the treatment groups. In these situations the effect of treatment can be characterized both by the shape of the fitted profiles and by estimating the proportion of cases who exhibit each particular response profile. This paper describes how such experiments may be analysed through the introduction of a latent variable into the standard model. Maximum likelihood estimation is straight-forward using the EM algorithm. Model choice requires some care, but good-fitting models can be identified via inspection of residuals and the use of empirical semi-variogram plots. Once the number of distinct profiles has been determined, treatment effects can be investigated using likelihood-ratio statistics. The approach is illustrated with a re-analysis of a dataset first described by Grizzle and Allen.
1. Am J Cardiol. 2002 Dec 15;90(12):1374-7. Comparison of outcomes of patients with myocardial in... more 1. Am J Cardiol. 2002 Dec 15;90(12):1374-7. Comparison of outcomes of patients with myocardial infarction when living alone versus those not living alone. O'Shea JC, Wilcox RG, Skene AM, Stebbins AL, Granger CB, Armstrong ...
To investigate the characteristics of published trials in order to establish the origin of the di... more To investigate the characteristics of published trials in order to establish the origin of the differing results obtained in trials of platelet inhibitors after peripheral bypass procedures. Analysis of the information from 11 randomised, controlled trials of platelet inhibitors after peripheral bypass procedures published up until 1999 and involving 2302 patients undergoing peripheral bypass operations, 1250 of whom were treated with platelet inhibitors. There is a significant treatment benefit of platelet inhibitors on meta-analysis of the trials, but a significant heterogeneity amongst the individual trial results. The proportion of patients in a trial with prosthetic grafts was a significant factor in explaining the heterogeneity. Proportion of prosthetic grafts was associated with sample size and with the proportion of grafts above the knee, but these were not found to make an independent contribution to the heterogeneity observed. The platelet inhibitor regimen used, the severity of ischaemic symptoms and the proportion of smokers included were also not found to be important. The improvement of graft patency by aspirin and related platelet inhibitors in clinical trials in peripheral bypass procedures can be attributed to an effect on patients with prosthetic grafts. There is little evidence that these agents prevent occlusion of vein grafts. The conclusion of an earlier meta-analysis that antiplatelet agents should be used for all bypasses is not supported.
To assess the medium to long term outcome of patients ineligible for thrombolysis compared to tho... more To assess the medium to long term outcome of patients ineligible for thrombolysis compared to those enrolled in a clinical trial of thrombolysis and patients receiving non-trial thrombolysis. Cohort study based on the Nottingham heart attack register. Two district general hospitals serving a defined urban/rural population. All patients admitted with a confirmed acute myocardial infarction during 1992 categorised as either participants of a thrombolytic trial (group A, n = 140), receiving non-trial thrombolysis (group B, n = 329), or deemed ineligible for lytic treatment (group C, n = 431). Background characteristics, inhospital treatment, patterns of follow up, referrals to cardiologists, revascularisation rates, and short and long term survival. Clinical trial recruits were younger by almost 10 years, were less likely to have a previous history of myocardial infarction, and more likely to be in Killip class 1 on admission than those ineligible for thrombolysis. Cardiology follow up...
To assess the impact of myocardial infarction on quality of life in four year survivors compared ... more To assess the impact of myocardial infarction on quality of life in four year survivors compared to data from "community norms", and to determine factors associated with a poor quality of life. Cohort study based on the Nottingham heart attack register. Two district general hospitals serving a defined urban/rural population. All patients admitted with acute myocardial infarction during 1992 and alive at a median of four years. Short form 36 (SF 36) domain and overall scores. Of 900 patients with an acute myocardial infarction in 1992, there were 476 patients alive and capable of responding to a questionnaire in 1997. The response rate was 424 (89. 1%). Compared to age and sex adjusted normative data, patients aged under 65 years exhibited impairment in all eight domains, the largest differences being in physical functioning (mean difference 20 points), role physical (mean difference 23 points), and general health (mean difference 19 points). In patients over 65 years mean ...
Five observers each examined 20 jaundiced patients, recording clinical signs and symptoms on a fo... more Five observers each examined 20 jaundiced patients, recording clinical signs and symptoms on a form which also gave the definitions used for the study. The balanced design of the study allowed examination for order effects, but none were found, except for a tendency for agreement on indicants with more than two categories to improve as the study progressed. Chance agreement was corrected by the use of kappa statistics which showed that 80% of the indicants showed agreement significantly greater than expected by chance. Certain indicants (dark urine, variability of jaundice, abdominal pain, character of liver edge and presence of spleen) showed no evidence of significant agreement, even though the indicants were frequently observed in both states - present or absent. The percentage of correct clinical diagnoses reached by the observers (without biochemical or any other information) varied between 65% and 84%. The consensus diagnosis was correct in 80% of cases. Agreement was higher i...
Background—In patients with chronic heart failure (CHF), a -blocker is generally added to a regim... more Background—In patients with chronic heart failure (CHF), a -blocker is generally added to a regimen containing an angiotensin-converting-enzyme (ACE) inhibitor. It is unknown whether -blockade as initial therapy may be as useful. Methods and Results—We randomized 1010 patients with mild to moderate CHF and left ventricular ejection fraction 35%, who were not receiving ACE inhibitor, -blocker, or angiotensin receptor blocker
Proceedings of the 6th international workshop on Software and performance - WOSP '07, 2007
... We will now re-strict our analysis of application service provision to systems of SLAs govern... more ... We will now re-strict our analysis of application service provision to systems of SLAs governing the timeliness of service provision ... We therefore define events corresponding to the completion of each of the events in the scenario E = {x, y, z, w} in relation to our actions such that ...
Resumen: Discussion on the papers by Savage, I. Richard, On not being rational and by Kadane, Jos... more Resumen: Discussion on the papers by Savage, I. Richard, On not being rational and by Kadane, Joseph B. and Sedransj, Nell, Toward a more ethical clinical trial, both of them part of a round table on Personal and inter-personal ethics held in the First International ...
A three-stage hierarchical model is proposed for two treatment, binary response studies conducted... more A three-stage hierarchical model is proposed for two treatment, binary response studies conducted in a number of centres. The approach adopted is Bayesian. Marginal densities for second stage parameters are shown to provide useful summaries both of comparative efficacy and of the heterogeneity of treatment effects across centres. Sensitivity studies of model assumptions are illustrated.
Simulation has become practical as an everyday tool given the wide availability of high performan... more Simulation has become practical as an everyday tool given the wide availability of high performance workstations. We argue that simulation can play an important role in determining the number of patients required in a survival study, particularly if the assumption of proportional hazards does not hold, or if the study design is complex. The argument is illustrated by two examples. The first considers the design of a post myocardial infarction survival trial in which the hazard ratio is not constant. The second provides sample size estimates for a multicentre heart failure study in which both the treatment effect and the control mortality rate vary across centres.
In proportional hazards survival studies, power depends on the observed number of deaths, d*. For... more In proportional hazards survival studies, power depends on the observed number of deaths, d*. For a given choice of survival, loss, and patient entry distributions, sample sizes can be determined by equating d* to the expected number of deaths. Approximating the survival and loss distributions with piecewise exponential distributions, and patient entry with a piecewise linear distribution, significantly reduces the computational overhead, and the expected number of deaths can be evaluated routinely. The merits of this approach are illustrated by a clinical trial of chemotherapy for large bowel cancer.
... Figure 1 is a display of the marginal posterior densities of these terms, estimated from the ... more ... Figure 1 is a display of the marginal posterior densities of these terms, estimated from the sam-ple together with the corresponding normal approxima-tions. It can be seen that the normal approximations are slightly light tailed in comparison. ... 284 Forster and Skene 2.6 2.4 2.2 ...
In Proceedings of the Wosp 07 Sixth International Workshop on Software and Performance February 5 8 2007 Buenos Aires Argentina Acm Press New York Us, 2007
... We will now re-strict our analysis of application service provision to systems of SLAs govern... more ... We will now re-strict our analysis of application service provision to systems of SLAs governing the timeliness of service provision ... We therefore define events corresponding to the completion of each of the events in the scenario E = {x, y, z, w} in relation to our actions such that ...
Standard models for the analysis of repeated measurements assume a common response profile for al... more Standard models for the analysis of repeated measurements assume a common response profile for all experimental units within a treatment group. However, in many applications this under-represents the nature of the response. There may be several distinct modes of response within a group (for example, responders versus non-responders to a given treatment), or there may be a set of distinct response profiles which are common to all the treatment groups. In these situations the effect of treatment can be characterized both by the shape of the fitted profiles and by estimating the proportion of cases who exhibit each particular response profile. This paper describes how such experiments may be analysed through the introduction of a latent variable into the standard model. Maximum likelihood estimation is straight-forward using the EM algorithm. Model choice requires some care, but good-fitting models can be identified via inspection of residuals and the use of empirical semi-variogram plots. Once the number of distinct profiles has been determined, treatment effects can be investigated using likelihood-ratio statistics. The approach is illustrated with a re-analysis of a dataset first described by Grizzle and Allen.
1. Am J Cardiol. 2002 Dec 15;90(12):1374-7. Comparison of outcomes of patients with myocardial in... more 1. Am J Cardiol. 2002 Dec 15;90(12):1374-7. Comparison of outcomes of patients with myocardial infarction when living alone versus those not living alone. O'Shea JC, Wilcox RG, Skene AM, Stebbins AL, Granger CB, Armstrong ...
To investigate the characteristics of published trials in order to establish the origin of the di... more To investigate the characteristics of published trials in order to establish the origin of the differing results obtained in trials of platelet inhibitors after peripheral bypass procedures. Analysis of the information from 11 randomised, controlled trials of platelet inhibitors after peripheral bypass procedures published up until 1999 and involving 2302 patients undergoing peripheral bypass operations, 1250 of whom were treated with platelet inhibitors. There is a significant treatment benefit of platelet inhibitors on meta-analysis of the trials, but a significant heterogeneity amongst the individual trial results. The proportion of patients in a trial with prosthetic grafts was a significant factor in explaining the heterogeneity. Proportion of prosthetic grafts was associated with sample size and with the proportion of grafts above the knee, but these were not found to make an independent contribution to the heterogeneity observed. The platelet inhibitor regimen used, the severity of ischaemic symptoms and the proportion of smokers included were also not found to be important. The improvement of graft patency by aspirin and related platelet inhibitors in clinical trials in peripheral bypass procedures can be attributed to an effect on patients with prosthetic grafts. There is little evidence that these agents prevent occlusion of vein grafts. The conclusion of an earlier meta-analysis that antiplatelet agents should be used for all bypasses is not supported.
To assess the medium to long term outcome of patients ineligible for thrombolysis compared to tho... more To assess the medium to long term outcome of patients ineligible for thrombolysis compared to those enrolled in a clinical trial of thrombolysis and patients receiving non-trial thrombolysis. Cohort study based on the Nottingham heart attack register. Two district general hospitals serving a defined urban/rural population. All patients admitted with a confirmed acute myocardial infarction during 1992 categorised as either participants of a thrombolytic trial (group A, n = 140), receiving non-trial thrombolysis (group B, n = 329), or deemed ineligible for lytic treatment (group C, n = 431). Background characteristics, inhospital treatment, patterns of follow up, referrals to cardiologists, revascularisation rates, and short and long term survival. Clinical trial recruits were younger by almost 10 years, were less likely to have a previous history of myocardial infarction, and more likely to be in Killip class 1 on admission than those ineligible for thrombolysis. Cardiology follow up...
To assess the impact of myocardial infarction on quality of life in four year survivors compared ... more To assess the impact of myocardial infarction on quality of life in four year survivors compared to data from "community norms", and to determine factors associated with a poor quality of life. Cohort study based on the Nottingham heart attack register. Two district general hospitals serving a defined urban/rural population. All patients admitted with acute myocardial infarction during 1992 and alive at a median of four years. Short form 36 (SF 36) domain and overall scores. Of 900 patients with an acute myocardial infarction in 1992, there were 476 patients alive and capable of responding to a questionnaire in 1997. The response rate was 424 (89. 1%). Compared to age and sex adjusted normative data, patients aged under 65 years exhibited impairment in all eight domains, the largest differences being in physical functioning (mean difference 20 points), role physical (mean difference 23 points), and general health (mean difference 19 points). In patients over 65 years mean ...
Five observers each examined 20 jaundiced patients, recording clinical signs and symptoms on a fo... more Five observers each examined 20 jaundiced patients, recording clinical signs and symptoms on a form which also gave the definitions used for the study. The balanced design of the study allowed examination for order effects, but none were found, except for a tendency for agreement on indicants with more than two categories to improve as the study progressed. Chance agreement was corrected by the use of kappa statistics which showed that 80% of the indicants showed agreement significantly greater than expected by chance. Certain indicants (dark urine, variability of jaundice, abdominal pain, character of liver edge and presence of spleen) showed no evidence of significant agreement, even though the indicants were frequently observed in both states - present or absent. The percentage of correct clinical diagnoses reached by the observers (without biochemical or any other information) varied between 65% and 84%. The consensus diagnosis was correct in 80% of cases. Agreement was higher i...
Background—In patients with chronic heart failure (CHF), a -blocker is generally added to a regim... more Background—In patients with chronic heart failure (CHF), a -blocker is generally added to a regimen containing an angiotensin-converting-enzyme (ACE) inhibitor. It is unknown whether -blockade as initial therapy may be as useful. Methods and Results—We randomized 1010 patients with mild to moderate CHF and left ventricular ejection fraction 35%, who were not receiving ACE inhibitor, -blocker, or angiotensin receptor blocker
Proceedings of the 6th international workshop on Software and performance - WOSP '07, 2007
... We will now re-strict our analysis of application service provision to systems of SLAs govern... more ... We will now re-strict our analysis of application service provision to systems of SLAs governing the timeliness of service provision ... We therefore define events corresponding to the completion of each of the events in the scenario E = {x, y, z, w} in relation to our actions such that ...
Resumen: Discussion on the papers by Savage, I. Richard, On not being rational and by Kadane, Jos... more Resumen: Discussion on the papers by Savage, I. Richard, On not being rational and by Kadane, Joseph B. and Sedransj, Nell, Toward a more ethical clinical trial, both of them part of a round table on Personal and inter-personal ethics held in the First International ...
A three-stage hierarchical model is proposed for two treatment, binary response studies conducted... more A three-stage hierarchical model is proposed for two treatment, binary response studies conducted in a number of centres. The approach adopted is Bayesian. Marginal densities for second stage parameters are shown to provide useful summaries both of comparative efficacy and of the heterogeneity of treatment effects across centres. Sensitivity studies of model assumptions are illustrated.
Simulation has become practical as an everyday tool given the wide availability of high performan... more Simulation has become practical as an everyday tool given the wide availability of high performance workstations. We argue that simulation can play an important role in determining the number of patients required in a survival study, particularly if the assumption of proportional hazards does not hold, or if the study design is complex. The argument is illustrated by two examples. The first considers the design of a post myocardial infarction survival trial in which the hazard ratio is not constant. The second provides sample size estimates for a multicentre heart failure study in which both the treatment effect and the control mortality rate vary across centres.
In proportional hazards survival studies, power depends on the observed number of deaths, d*. For... more In proportional hazards survival studies, power depends on the observed number of deaths, d*. For a given choice of survival, loss, and patient entry distributions, sample sizes can be determined by equating d* to the expected number of deaths. Approximating the survival and loss distributions with piecewise exponential distributions, and patient entry with a piecewise linear distribution, significantly reduces the computational overhead, and the expected number of deaths can be evaluated routinely. The merits of this approach are illustrated by a clinical trial of chemotherapy for large bowel cancer.
... Figure 1 is a display of the marginal posterior densities of these terms, estimated from the ... more ... Figure 1 is a display of the marginal posterior densities of these terms, estimated from the sam-ple together with the corresponding normal approxima-tions. It can be seen that the normal approximations are slightly light tailed in comparison. ... 284 Forster and Skene 2.6 2.4 2.2 ...
Uploads