Search

In oncology trials, control group patients often switch onto the experimental treatment during follow-up, usually after disease progression. In this case, an intention-to-treat analysis will not address the policy question of interest – that of whether the new treatment represents an effective and cost-effective use of health care resources, compared to the standard treatment. Rank preserving structural failure time models (RPSFTM),\ud inverse probability of censoring weights (IPCW) and two-stage estimation (TSE) have often been used to adjust for switching to inform treatment reimbursement policy decisions. TSE has been applied using a simple approach (TSEsimp), assuming no time-dependent confounding between the time of disease progression and the time of switch. This is problematic if there is a delay between progression and switch. In this paper we introduce TSEgest, which uses structural nested models and g-estimation to account for time-dependent confounding, and\ud compare it to TSEsimp, RPSFTM and IPCW. We simulated scenarios where control group patients could switch onto the experimental treatment with and without time-dependent confounding being present. We varied switching proportions, treatment effects and censoring proportions. We assessed adjustment methods according to their estimation of control group restricted mean survival times that would have been observed in the absence of switching. All methods performed well in scenarios with no time-dependent confounding. TSEgest and RPSFTM continued to perform well in scenarios with time-dependent confounding, but TSEsimp resulted in substantial bias. IPCW also performed well in scenarios with time-dependent confounding, except when inverse probability weights were high in relation to the size of the group being subjected to weighting, which occurred when there was a combination of modest sample size and high switching proportions. TSEgest represents a useful addition to the collection of methods that may be used to adjust for treatment switching in trials in order to address policy-relevant questions.

Treatment switching often has a crucial impact on estimates of effectiveness and cost-effectiveness of new oncology treatments. Rank preserving structural failure time models (RPSFTM) and two-stage estimation (TSE) methods estimate ‘counterfactual’ (i.e. had there been no switching) survival times and incorporate re-censoring to guard against informative censoring in the counterfactual dataset. However, re-censoring causes a loss of longer term survival information which is problematic when estimates of long-term survival effects are required, as is often the case for health technology assessment decision making. We present a simulation study designed to investigate applications of the RPSFTM and TSE with and without re-censoring, to determine whether re-censoring should always be recommended within adjustment analyses. We investigate a context where switching is from the control group onto the experimental treatment in scenarios with varying switch proportions, treatment effect sizes and time-dependencies, disease severity and switcher prognosis. Methods were assessed according to their estimation of control group restricted mean survival (that would be observed in the absence of switching) at the end of the simulated trial follow-up. We found that RPSFTM and TSE analyses which incorporated re-censoring usually produced negative bias (i.e. under-estimating control group restricted mean survival and therefore over-estimating the treatment effect). RPSFTM and TSE analyses that did not incorporate re-censoring consistently produced positive bias (i.e. under-estimating the treatment effect) which was often smaller in magnitude than the bias associated with the re-censored analyses. We believe that analyses should be conducted with and without re-censoring, as this may provide decision makers with useful information on where the true treatment effect is likely to lie. Analyses that incorporate re-censoring should not always represent the default approach when the objective is to estimate long-term survival times and treatment effects on long-term survival.

Background Meta-analysis of individual participant time-to-event data from multiple prospective epidemiological studies enables detailed investigation of exposure-risk relationships, but involves a number of analytical challenges. Methods This article describes statistical approaches adopted in the Emerging Risk Factors Collaboration, in which primary data from more than 1 million participants in more than 100 prospective studies have been collated to enable detailed analyses of various risk markers in relation to incident cardiovascular disease outcomes. Results Analyses have been principally based on Cox proportional hazards regression models stratified by sex, undertaken in each study separately. Estimates of exposure-risk relationships, initially unadjusted and then adjusted for several confounders, have been combined over studies using meta-analysis. Methods for assessing the shape of exposure-risk associations and the proportional hazards assumption have been developed. Estimates of interactions have also been combined using meta-analysis, keeping separate within-and between-study information. Regression dilution bias caused by measurement error and within-person variation in exposures and confounders has been addressed through the analysis of repeat measurements to estimate corrected regression coefficients. These methods are exemplified by analysis of plasma fibrinogen and risk of coronary heart disease, and Stata code is made available. Conclusion Increasing numbers of meta-analyses of individual participant data from observational data are being conducted to enhance the statistical power and detail of epidemiological studies. The statistical methods developed here can be used to address the needs of such analyses. © The Author 2010; all rights reserved.

Introduction: Reliable estimates of health-related behaviours, such as levels of alcohol consumption in the population, are required to formulate and evaluate policies. National surveys provide such data; validity depends on generalisability, but this is threatened by declining response levels. Attempts to address bias arising from non-response are typically limited to survey weights based on sociodemographic characteristics, which do not capture differential health and related behaviours within categories. This project aims to explore and address non-response bias in health surveys with a focus on alcohol consumption.\ud \ud Methods and analysis: The Scottish Health Surveys (SHeS) aim to provide estimates representative of the Scottish population living in private households. Survey data of consenting participants (92% of the achieved sample) have been record-linked to routine hospital admission (Scottish Morbidity Records (SMR)) and mortality (from National Records of Scotland (NRS)) data for surveys conducted in 1995, 1998, 2003, 2008, 2009 and 2010 (total adult sample size around 40 000), with maximum follow-up of 16 years. Also available are census information and SMR/NRS data for the general population. Comparisons of alcohol-related mortality and hospital admission rates in the linked SHeS-SMR/NRS with those in the general population will be made. Survey data will be augmented by quantification of differences to refine alcohol consumption estimates through the application of multiple imputation or inverse probability weighting. The resulting corrected estimates of population alcohol consumption will enable superior policy evaluation. An advanced weighting procedure will be developed for wider use.\ud \ud Ethics and dissemination: Ethics approval for SHeS has been given by the National Health Service (NHS) Multi-Centre Research Ethics Committee and use of linked data has been approved by the Privacy Advisory Committee to the Board of NHS National Services Scotland and Registrar General. Funding has been granted by the MRC. The outputs will include four or five public health and statistical methodological international journal and conference papers.

The Regents Park and Tower Environmental Experiment (REPARTEE) comprised two campaigns in London in October 2006 and October/November 2007. The experiment design involved measurements at a heavily trafficked roadside site, two urban background sites and an elevated site at 160-190 m above ground on the BT Tower, supplemented in the second campaign by Doppler lidar measurements of atmospheric vertical structure. A wide range of measurements of airborne particle physical metrics and chemical composition were made as well as measurements of a considerable range of gas phase species and the fluxes of both particulate and gas phase substances. Significant findings include (a) demonstration of the evaporation of traffic-generated nanoparticles during both horizontal and vertical atmospheric transport; (b) generation of a large base of information on the fluxes of nanoparticles, accumulation mode particles and specific chemical components of the aerosol and a range of gas phase species, as well as the elucidation of key processes and comparison with emissions inventories; (c) quantification of vertical gradients in selected aerosol and trace gas species which has demonstrated the important role of regional transport in influencing concentrations of sulphate, nitrate and secondary organic compounds within the atmosphere of London; (d) generation of new data on the atmospheric structure and turbulence above London, including the estimation of mixed layer depths; (e) provision of new data on trace gas dispersion in the urban atmosphere through the release of purposeful tracers; (f) the determination of spatial differences in aerosol particle size distributions and their interpretation in terms of sources and physico-chemical transformations; (g) studies of the nocturnal oxidation of nitrogen oxides and of the diurnal behaviour of nitrate aerosol in the urban atmosphere, and (h) new information on the chemical composition and source apportionment of particulate matter size fractions in the atmosphere of London derived both from bulk chemical analysis and aerosol mass spectrometry with two instrument types. © 2011 Author(s).

Context Circulating concentration of lipoprotein(a) (Lp[a]), a large glycoprotein attached to a low-density lipoprotein–like particle, may be associated with risk of coronary heart disease (CHD) and stroke. Objective To assess the relationship of Lp(a) concentration with risk of major vascular and nonvascular outcomes. Study Selection Long-term prospective studies that recorded Lp(a) concentration and subsequent major vascular morbidity and/or cause-specific mortality published between January 1970 and March 2009 were identified through electronic searches of MEDLINE and other databases, manual searches of reference lists, and discussion with collaborators. Data Extraction Individual records were provided for each of 126 634 participants in 36 prospective studies. During 1.3 million person-years of follow-up, 22 076 firstever fatal or nonfatal vascular disease outcomes or nonvascular deaths were recorded, including 9336 CHD outcomes, 1903 ischemic strokes, 338 hemorrhagic strokes, 751 unclassified strokes, 1091 other vascular deaths, 8114 nonvascular deaths, and 242 deaths of unknown cause. Within-study regression analyses were adjusted for within-person variation and combined using meta-analysis. Analyses excluded participants with known preexisting CHD or stroke at baseline. Data Synthesis Lipoprotein(a) concentration was weakly correlated with several conventional vascular risk factors and it was highly consistent within individuals over several years. Associations of Lp(a) with CHD risk were broadly continuous in shape. In the 24 cohort studies, the rates of CHD in the top and bottom thirds of baseline Lp(a) distributions, respectively, were 5.6 (95% confidence interval [CI], 5.4-5.9) per 1000 personyears and 4.4 (95% CI, 4.2-4.6) per 1000 person-years. The risk ratio for CHD, adjusted for age and sex only, was 1.16 (95% CI, 1.11-1.22) per 3.5-fold higher usual Lp(a) concentration (ie, per 1 SD), and it was 1.13 (95% CI, 1.09-1.18) following further adjustment for lipids and other conventional risk factors. The corresponding adjusted risk ratios were 1.10 (95% CI, 1.02-1.18) for ischemic stroke, 1.01 (95% CI, 0.98-1.05) for the aggregate of nonvascular mortality, 1.00 (95% CI, 0.97-1.04) for cancer deaths, and 1.00 (95% CI, 0.95-1.06) for nonvascular deaths other than cancer. Conclusion Under a wide range of circumstances, there are continuous, independent, and modest associations of Lp(a) concentration with risk of CHD and stroke that appear exclusive to vascular outcomes. ©2009 American Medical Association. All rights reserved.

Background: There is currently no validated method for antepartum prediction of the risk of failed vaginal birth after cesarean section and no information on the relationship between the risk of emergency cesarean delivery and the risk of uterine rupture.\ud \ud Methods and Findings: We linked a national maternity hospital discharge database and a national registry of perinatal deaths. We studied 23,286 women with one prior cesarean delivery who attempted vaginal birth at or after 40-wk gestation. The population was randomly split into model development and validation groups. The factors associated with emergency cesarean section were maternal age (adjusted odds ratio [OR] = 1.22 per 5-y increase, 95% confidence interval [CI]: 1.16 to 1.28), maternal height (adjusted OR = 0.75 per 5-cm increase, 95% CI: 0.73 to 0.78), male fetus (adjusted OR = 1.18, 95% CI: 1.08 to 1.29), no previous vaginal birth (adjusted OR = 5.08, 95% CI: 4.52 to 5.72), prostaglandin induction of labor (adjusted OR = 1.42, 95% CI: 1.26 to 1.60), and birth at 41-wk (adjusted OR = 1.30, 95% CI: 1.18 to 1.42) or 42-wk (adjusted OR = 1.38, 95% CI: 1.17 to 1.62) gestation compared with 40-wk. In the validation group, 36% of the women had a low predicted risk of caesarean section (40%); 10.9% and 47.7% of these women, respectively, actually had deliveries by caesarean section. The predicted risk of caesarean section was also associated with the risk of all uterine rupture (OR for a 5% increase in predicted risk = 1.22, 95% CI: 1.14 to 1.31) and uterine rupture associated with perinatal death (OR for a 5% increase in predicted risk = 1.32, 95% CI: 1.02 to 1.73). The observed incidence of uterine rupture was 2.0 per 1,000 among women at low risk of cesarean section and 9.1 per 1,000 among those at high risk (relative risk = 4.5, 95% CI: 2.6 to 8.1). We present the model in a simple-to-use format.\ud \ud Conclusions: We present, to our knowledge, the first validated model for antepartum prediction of the risk of failed vaginal birth after prior cesarean section. Women at increased risk of emergency caesarean section are also at increased risk of uterine rupture, including catastrophic rupture leading to perinatal death.