Multi-level modelling, the ecologic fallacy, and hybrid study designs.
Ontology highlight
ABSTRACT:
SUBMITTER: Wakefield J
PROVIDER: S-EPMC2663723 | biostudies-literature | 2009 Apr
REPOSITORIES: biostudies-literature
Publications
Multi-level modelling, the ecologic fallacy, and hybrid study designs.
International journal of epidemiology 20090401 2
Similar Datasets
Project description:Ecological studies that make use of data on groups of individuals, rather than on the individuals themselves, are subject to numerous biases that cannot be resolved without some individual-level data. In the context of a rare outcome, the hybrid design for ecological inference efficiently combines group-level data with individual-level case-control data. Unfortunately, except in relatively simple settings, use of the design in practice is limited since evaluation of the hybrid likelihood is computationally prohibitively expensive. In this article we first propose and develop an alternative representation of the hybrid likelihood. Second, based on this new representation, a series of approximations are proposed that drastically reduce computational burden. A comprehensive simulation shows that, in a broad range of scenarios, estimators based on the approximate hybrid likelihood exhibit the same operating characteristics as the exact hybrid likelihood, without any penalty in terms of increased bias or reduced efficiency. Third, in settings where the approximations may not hold, a pragmatic estimation and inference strategy is developed that uses the approximate form for some likelihood contributions and the exact form for others. The strategy gives researchers the ability to balance computational tractability with accuracy in their own settings. Finally, as a by-product of the development, we provide the first explicit characterization of the hybrid aggregate data design which combines data from an aggregate data study (Prentice and Sheppard, 1995, Biometrika 82, 113-125) with case-control samples. The methods are illustrated using data from North Carolina on births between 2007 and 2009.
| S-EPMC4445683 | biostudies-literature
Project description:Longitudinal models have become increasingly popular in recent years because of their power to test theoretically derived hypotheses by modeling within-person processes with repeated measures. Growth models constitute a flexible framework for modeling a range of complex trajectories across time in outcomes of interest, including non-linearities and time-varying covariates. However, these models can be expanded to include the effects of multiple growth processes at once on a single outcome. Here, I outline such an extension, showing how multiple growth processes can be modeled as a specific case of the general ability to include time-varying covariates in growth models. I show that this extension of growth models cannot be accomplished by statistical models alone, and that study design plays a crucial role in allowing for proper parameter recovery. I demonstrate these principles through simulations to mimic important theoretical conditions where modeling the effects of multiple growth processes can address developmental theory including, disaggregating the effects of age and practice or treatment in repeated assessments and modeling age- and puberty-related effects during adolescence. I compare how these models behave in two common longitudinal designs, cohort and accelerated, and how planned missingness in observations is key to parameter recovery. I conclude with directions for future substantive research using the method outlined here.
| S-EPMC8363832 | biostudies-literature
Project description:BackgroundBayesian adaptive designs can be more efficient than traditional methods for multi-arm randomised controlled trials. The aim of this work was to demonstrate how Bayesian adaptive designs can be constructed for multi-arm phase III clinical trials and assess potential benefits that these designs offer.MethodsWe constructed several alternative Bayesian adaptive designs for the Collaborative Ankle Support Trial (CAST), which was a randomised controlled trial that compared four treatments for severe ankle sprain. These designs incorporated response adaptive randomisation (RAR), arm dropping, and early stopping for efficacy or futility. We studied the operating characteristics of the Bayesian designs via simulation. We then virtually re-executed the trial by implementing the Bayesian adaptive designs using patient data sampled from the CAST study to demonstrate the practical applicability of the designs.ResultsWe constructed five Bayesian adaptive designs, each of which had high power and recruited fewer patients on average than the original designs target sample size. The virtual executions showed that most of the Bayesian designs would have led to trials that declared superiority of one of the interventions over the control. Bayesian adaptive designs with RAR or arm dropping were more likely to allocate patients to better performing arms at each interim analysis. Similar estimates and conclusions were obtained from the Bayesian adaptive designs as from the original trial.ConclusionsUsing CAST as an example, this case study shows how Bayesian adaptive designs can be constructed for phase III multi-arm trials using clinically relevant decision criteria. These designs demonstrated that they can potentially generate earlier results and allocate more patients to better performing arms. We recommend the wider use of Bayesian adaptive approaches in phase III clinical trials.Trial registrationCAST study registration ISRCTN, ISRCTN37807450. Retrospectively registered on 25 April 2003.
| S-EPMC6961269 | biostudies-literature
Project description:BackgroundThe Dabaoshan mine in the southeast of Guangdong Province, China, is at high risk of multi-metal pollutant discharge into a local river (Hengshihe) and the surrounding area. Following approximately 30 years of exposure to these metals, little is known regarding the subsequent health effects and risks for the local residents. In our present study, we have estimated the relationships between long-term environmental exposure to multiple heavy metals and the risk of cancer mortality in a Chinese population in the vicinity of Dabaoshan.MethodsAn ecologic study was performed. Between 2000-2007, a total population of 194,131 lived in the nine agricultural villages that surround the Hengshihe area. Heavy metals concentrations were determined in local environmental samples (water and crops) and whole blood taken from 1152 local residents of both a high-exposure area (HEA) and a low-exposure area (LEA). We calculated the rate ratio and standardized mortality ratios based on age- and gender-specific cancer mortality rates for the different reference populations (based on district, county and province). Simple, multiple linear and ridge regression models were used to evaluate the associations between exposure to multiple heavy metals and cancer mortality in the nine villages, after adjustment for age and sex.ResultsThe geometric mean blood levels of cadmium and lead were measured at 24.10 μg/L and 38.91 μg/dL for subjects (n = 563) in the HEA and 1.87 μg/L and 4.46 μg/dL for subjects (n = 589) from the LEA, respectively (P < 0.001). The rate of mortality from all cancers in the HEA was substantially elevated in comparison with the corresponding mortality rate in the LEA for men (rate ratio = 2.13; 95% confidence intervals = 1.63 - 2.77) and women (2.83; 1.91 - 4.19); rates were also significantly elevated compared with the rate when compared to the entire Wengyuan County area, or the provincial reference population. In addition, mortality rates were significantly increased for stomach, lung and esophageal cancer in the HEA in comparison with the corresponding rates in the LEA, in Wengyuan County and the provincial reference population for men, women and both combined. Further analysis showed that there were significantly positive correlations between exposure to cadmium and lead and the risk of all-cancers and stomach cancer mortality among women and both sexes, whilst zinc exposure showed no association with the risk of site-specific cancer mortality in the nine villages evaluated.ConclusionsThe findings of this study reveal probable associations between long-term environmental exposure to both cadmium and lead and an increased risk of mortality from all cancer, as well as from stomach, esophageal and lung-cancers.
| S-EPMC3112132 | biostudies-literature
Project description:This paper is concerned with predicting the progressive damage and failure of multi-layered hybrid textile composites subjected to uniaxial tensile loading, using a novel two-scale computational mechanics framework. These composites include three-dimensional woven textile composites (3DWTCs) with glass, carbon and Kevlar fibre tows. Progressive damage and failure of 3DWTCs at different length scales are captured in the present model by using a macroscale finite-element (FE) analysis at the representative unit cell (RUC) level, while a closed-form micromechanics analysis is implemented simultaneously at the subscale level using material properties of the constituents (fibre and matrix) as input. The N-layers concentric cylinder (NCYL) model (Zhang and Waas 2014 Acta Mech. 225, 1391-1417; Patel et al. submitted Acta Mech.) to compute local stress, srain and displacement fields in the fibre and matrix is used at the subscale. The 2-CYL fibre-matrix concentric cylinder model is extended to fibre and (N-1) matrix layers, keeping the volume fraction constant, and hence is called the NCYL model where the matrix damage can be captured locally within each discrete layer of the matrix volume. The influence of matrix microdamage at the subscale causes progressive degradation of fibre tow stiffness and matrix stiffness at the macroscale. The global RUC stiffness matrix remains positive definite, until the strain softening response resulting from different failure modes (such as fibre tow breakage, tow splitting in the transverse direction due to matrix cracking inside tow and surrounding matrix tensile failure outside of fibre tows) are initiated. At this stage, the macroscopic post-peak softening response is modelled using the mesh objective smeared crack approach (Rots et al. 1985 HERON 30, 1-48; Heinrich and Waas 2012 53rd AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics and Materials Conference, Honolulu, HI, 23-26 April 2012 AIAA 2012-1537). Manufacturing-induced geometric imperfections are included in the simulation, where the FE mesh of the unit cell is generated directly from micro-computed tomography (MCT) real data using a code Simpleware Results from multi-scale analysis for both an idealized perfect geometry and one that includes geometric imperfections are compared with experimental results (Pankow et al. 2012 53rd AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics and Materials Conference, Honolulu, HI, 23-26 April 2012 AIAA 2012-1572). This article is part of the themed issue 'Multiscale modelling of the structural integrity of composite materials'.
| S-EPMC4901255 | biostudies-other
Project description:BackgroundAdvances in sequencing technologies have enabled the characterization of multiple microbial and host genomes, opening new frontiers of knowledge while kindling novel applications and research perspectives. Among these is the investigation of the viral communities residing in the human body and their impact on health and disease. To this end, the study of samples from multiple tissues is critical, yet, the complexity of such analysis calls for a dedicated pipeline. We provide an automatic and efficient pipeline for identification, assembly, and analysis of viral genomes that combines the DNA sequence data from multiple organs. TRACESPipe relies on cooperation among 3 modalities: compression-based prediction, sequence alignment, and de novo assembly. The pipeline is ultra-fast and provides, additionally, secure transmission and storage of sensitive data.FindingsTRACESPipe performed outstandingly when tested on synthetic and ex vivo datasets, identifying and reconstructing all the viral genomes, including those with high levels of single-nucleotide polymorphisms. It also detected minimal levels of genomic variation between different organs.ConclusionsTRACESPipe's unique ability to simultaneously process and analyze samples from different sources enables the evaluation of within-host variability. This opens up the possibility to investigate viral tissue tropism, evolution, fitness, and disease associations. Moreover, additional features such as DNA damage estimation and mitochondrial DNA reconstruction and analysis, as well as exogenous-source controls, expand the utility of this pipeline to other fields such as forensics and ancient DNA studies. TRACESPipe is released under GPLv3 and is available for free download at https://github.com/viromelab/tracespipe.
| S-EPMC7439602 | biostudies-literature
Project description:Advances in mobile and wireless technologies offer tremendous opportunities for extending the reach and impact of psychological interventions and for adapting interventions to the unique and changing needs of individuals. However, insufficient engagement remains a critical barrier to the effectiveness of digital interventions. Human delivery of interventions (e.g., by clinical staff) can be more engaging but potentially more expensive and burdensome. Hence, the integration of digital and human-delivered components is critical to building effective and scalable psychological interventions. Existing experimental designs can be used to answer questions either about human-delivered components that are typically sequenced and adapted at relatively slow timescales (e.g., monthly) or about digital components that are typically sequenced and adapted at much faster timescales (e.g., daily). However, these methodologies do not accommodate sequencing and adaptation of components at multiple timescales and hence cannot be used to empirically inform the joint sequencing and adaptation of human-delivered and digital components. Here, we introduce the hybrid experimental design (HED)-a new experimental approach that can be used to answer scientific questions about building psychological interventions in which human-delivered and digital components are integrated and adapted at multiple timescales. We describe the key characteristics of HEDs (i.e., what they are), explain their scientific rationale (i.e., why they are needed), and provide guidelines for their design and corresponding data analysis (i.e., how can data arising from HEDs be used to inform effective and scalable psychological interventions).
| S-EPMC10024531 | biostudies-literature
Project description:Quantum dots are ideally suited for color conversion in light emitting diodes owing to their spectral tunability, high conversion efficiency and narrow emission bands. These properties are particularly important for display backlights; the highly saturated colors generated by quantum dots justify their higher production cost. Here, we demonstrate the benefits of a hybrid remote phosphor approach that combines a green-emitting europium-doped phosphor with red-emitting CdSe/CdS core/shell quantum dots. Different stacking geometries, including mixed and separate layers of both materials, are studied at the macroscopic and microscopic levels to identify the configuration that achieves maximum device efficiency while minimizing material usage. The influence of reabsorption, optical outcoupling and refractive index-matching between the layers is evaluated in detail with respect to device efficiency and cost. From the findings of this study, general guidelines are derived to optimize both the cost and efficiency of CdSe/CdS and other (potentially cadmium-free) quantum dot systems. When reabsorption of the green and/or red emission is significant compared to the absorption strength for the blue emission of the pumping light emitting diode, the hybrid remote phosphor approach becomes beneficial.
| S-EPMC6062237 | biostudies-literature
Project description:Supersaturated designs (SSDs) are often used to reduce the number of experimental runs in screening experiments with a large number of factors. As more factors are used in the study, the search for an optimal SSD becomes increasingly challenging because of the large number of feasible selection of factor level settings. This paper tackles this discrete optimization problem via an algorithm based on swarm intelligence. Using the commonly used E(s2) criterion as an illustrative example, we propose an algorithm to find E(s2)-optimal SSDs by showing that they attain the theoretical lower bounds in Bulutoglu and Cheng (2004) and Bulutoglu (2007). We show that our algorithm consistently produces SSDs that are at least as efficient as those from the traditional CP exchange method in terms of computational effort, frequency of finding the E(s2)-optimal SSD and also has good potential for finding D3-, D4- and D5-optimal SSDs.
| S-EPMC4835032 | biostudies-literature
Project description:The visibility of natural greenness is associated with several health benefits along multiple pathways, including stress recovery and attention restoration mechanisms. However, existing methodologies are inadequate for capturing eye-level greenness visibility exposure at high spatial resolutions for observers located on the ground. As a response, we developed an innovative methodological approach to model and map eye-level greenness visibility exposure for 5 m interval locations within a large study area. We used multi-source spatial data and applied viewshed analysis in conjunction with a distance decay model to compute a novel Viewshed Greenness Visibility Index (VGVI) at more than 86 million observer locations. We compared our eye-level visibility exposure map with traditional top-down greenness exposure metrics such as Normalised Differential Vegetation Index (NDVI) and a Street view based Green View Index (SGVI). Furthermore, we compared greenness visibility at street-only locations with total neighbourhood greenness visibility. We found strong to moderate correlations (r = 0.65-0.42, p < 0.05) between greenness visibility and mean NDVI, with a decreasing trend in correlation strength at increasing buffer distances from observer locations. Our findings suggest that top-down and eye-level measurements of greenness are two distinct metrics for assessing greenness exposure. Additionally, VGVI showed a strong correlation (r = 0.481, p < 0.01) with SGVI. Although the new VGVI has good agreement with existing street view based measures, we found that street-only greenness visibility values are not wholly representative of total neighbourhood visibility due to the under-representation of visible greenness in locations such as backyards and community parks. Our new methodology overcomes such underestimations, is easily transferable, and offers a computationally efficient approach to assessing eye-level greenness exposure.
| S-EPMC7562921 | biostudies-literature