Interventions to control nosocomial infections : study designs and statistical issues

There is a wide range of potential study designs for intervention studies to decrease nosocomial infections in hospitals. The analysis is complex due to competing events, clustering, multiple timescales and time-dependent period and intervention variables. This review considers the popular pre-post quasi-experimental design and compares it with randomized designs. Randomization can be done in several ways: randomization of the cluster [intensive care unit (ICU) or hospital] in a parallel design; randomization of the sequence in a cross-over design; and randomization of the time of intervention in a stepped-wedge design. We introduce each design in the context of nosocomial infections and discuss the designs with respect to the following key points: bias, control for nonintervention factors, and generalizability. Statistical issues are discussed. A pre-post-intervention design is often the only choice that will be informative for a retrospective analysis of an outbreak setting. It can be seen as a pilot study with further, more rigorous designs needed to establish causality. To yield internally valid results, randomization is needed. Generally, the first choice in terms of the internal validity should be a parallel cluster randomized trial. However, generalizability might be stronger in a stepped-wedge design because a wider range of ICU clinicians may be convinced to participate, especially if there are pilot studies with promising results. For analysis, the use of extended competing risk models is recommended.

Wood properties and knot occlusion of plantation grown Eucalyptus dunnii and Corymbia citriodora subsp. variegata in a pruning and thinning experiment.

A 2 × 2 factorial combination of thinned or unthinned, and pruned or unpruned 11-year-old Eucalyptus dunnii (DWG) and 12-year-old Corymbia citriodora subsp. variegata (CCV) was destructively sampled to provide 60 trees in total per species. Two 1.4 m long billets were cut from each tree and were rotary veneered in a spindleless lathe down to a 45 mm diameter core to expose knots which were classified as either alive, partially occluded or fully occluded. Non-destructive evaluation of a wider range of thinning treatments available in these trials was undertaken with Pilodyn and Fakopp tools. Disc samples were also taken for basic density and modulus of elasticity. Differences between treatments for all wood property assessments were generally small and not significantly different.Thinning and pruning had little effect on the stem diameter growth required to achieve occlusion, therefore occlusion would be more rapid after thinning due to more rapid stem diameter growth. The difference between the treatments of greatest management interest, thinned and pruned (T&P) and unthinned and unpruned (UT&UP) were small. The production of higher value clear wood produced after all knots had occluded, measured as the average stem diameter growth over occlusion of the three outermost knots, was approximately 2 centimetres diameter. Two of the treatments can be ruled out as viable management alternatives: (i) the effect of thinning without pruning (T&UP) is clear, leading to a large inner core of stem wood containing knots (large knotty core diameter) and (ii) pruning without thinning (UT&P) results in a small knotty core diameter, however the tree and therefore log diameters are also small.

Trapped ion magnetic resonance: concepts and designs

A novel spectroscopy of trapped ions is proposed which will bring single-ion detection sensitivity to the observation of magnetic resonance spectra. The approaches developed here are aimed at resolving one of the fundamental problems of molecular spectroscopy, the apparent incompatibility in existing techniques between high information content (and therefore good species discrimination) and high sensitivity. Methods for studying both electron spin resonance (ESR) and nuclear magnetic resonance (NMR) are designed. They assume established methods for trapping ions in high magnetic field and observing the trapping frequencies with high resolution (<1 Hz) and sensitivity (single ion) by electrical means. The introduction of a magnetic bottle field gradient couples the spin and spatial motions together and leads to a small spin-dependent force on the ion, which has been exploited by Dehmelt to observe directly the perturbation of the ground-state electron's axial frequency by its spin magnetic moment.

A series of fundamental innovations is described m order to extend magnetic resonance to the higher masses of molecular ions (100 amu = 2x 10^5 electron masses) and smaller magnetic moments (nuclear moments = 10^(-3) of the electron moment). First, it is demonstrated how time-domain trapping frequency observations before and after magnetic resonance can be used to make cooling of the particle to its ground state unnecessary. Second, adiabatic cycling of the magnetic bottle off between detection periods is shown to be practical and to allow high-resolution magnetic resonance to be encoded pointwise as the presence or absence of trapping frequency shifts. Third, methods of inducing spindependent work on the ion orbits with magnetic field gradients and Larmor frequency irradiation are proposed which greatly amplify the attainable shifts in trapping frequency.

The dissertation explores the basic concepts behind ion trapping, adopting a variety of classical, semiclassical, numerical, and quantum mechanical approaches to derive spin-dependent effects, design experimental sequences, and corroborate results from one approach with those from another. The first proposal presented builds on Dehmelt's experiment by combining a "before and after" detection sequence with novel signal processing to reveal ESR spectra. A more powerful technique for ESR is then designed which uses axially synchronized spin transitions to perform spin-dependent work in the presence of a magnetic bottle, which also converts axial amplitude changes into cyclotron frequency shifts. A third use of the magnetic bottle is to selectively trap ions with small initial kinetic energy. A dechirping algorithm corrects for undesired frequency shifts associated with damping by the measurement process.

The most general approach presented is spin-locked internally resonant ion cyclotron excitation, a true continuous Stern-Gerlach effect. A magnetic field gradient modulated at both the Larmor and cyclotron frequencies is devised which leads to cyclotron acceleration proportional to the transverse magnetic moment of a coherent state of the particle and radiation field. A preferred method of using this to observe NMR as an axial frequency shift is described in detail. In the course of this derivation, a new quantum mechanical description of ion cyclotron resonance is presented which is easily combined with spin degrees of freedom to provide a full description of the proposals.

Practical, technical, and experimental issues surrounding the feasibility of the proposals are addressed throughout the dissertation. Numerical ion trajectory simulations and analytical models are used to predict the effectiveness of the new designs as well as their sensitivity and resolution. These checks on the methods proposed provide convincing evidence of their promise in extending the wealth of magnetic resonance information to the study of collisionless ions via single-ion spectroscopy.

A range-wide experiment to investigate nutrient and soil moisture interactions in loblolly pine plantations

© 2015 by the authors.The future climate of the southeastern USA is predicted to be warmer, drier and more variable in rainfall, which may increase drought frequency and intensity. Loblolly pine (Pinus taeda) is the most important commercial tree species in the world and is planted on ~11 million ha within its native range in the southeastern USA. A regional study was installed to evaluate effects of decreased rainfall and nutrient additions on loblolly pine plantation productivity and physiology. Four locations were established to capture the range-wide variability of soil and climate. Treatments were initiated in 2012 and consisted of a factorial combination of throughfall reduction (approximate 30% reduction) and fertilization (complete suite of nutrients). Tree and stand growth were measured at each site. Results after two growing seasons indicate a positive but variable response of fertilization on stand volume increment at all four sites and a negative effect of throughfall reduction at two sites. Data will be used to produce robust process model parameterizations useful for simulating loblolly pine growth and function under future, novel climate and management scenarios. The resulting improved models will provide support for developing management strategies to increase pine plantation productivity and carbon sequestration under a changing climate.

Factores explicativos de las actitudes hacia el castigo juvenil. Evidencias de un diseño factorial

El rol desempeñado por la opinión pública en el desarrollo de la política criminal actual justifica el incremento de investigaciones destinadas a evaluar las actitudes de los ciudadanos hacia el castigo. No obstante, los avances en este ámbito han sido limitados debido a la utilización de rudimentarios instrumentos de medida. Por ello, el presente trabajo tiene como propósito explorar el efecto que generan en la opinión ciudadana ciertas variables referidas al hecho delictivo y al infractor, precisando su contribución relativa y la interacción existente entre ellas. Para satisfacer este objetivo se recurrió a un diseño factorial de la encuesta, creando una población de 256 casos-escenario fruto de la combinación de cuatro factores: la edad del joven, su historial delictivo, el grado de implicación en el hecho y el tipo de delito cometido. Los mismos fueron distribuidos en grupos de ocho casos ordenados aleatoriamente y fueron suministrados a 32 sujetos. Posteriormente se aplicaron análisis de regresión logística binaria. Los resultados obtenidos revelan que la naturaleza violenta de los hechos, la implicación activa de los jóvenes y el historial delictivo son predictores importantes de las condenas punitivas. Sin embargo la edad, una variable fundamental en la configuración de la justicia juvenil, no resulta significativa. De este modo, el trabajo muestra el potencial explicativo de este conjunto de factores y debate sus implicaciones teóricas y metodológicas para la investigación futura en este terreno.

Arbuscular mycorrhizal fungal hyphae reduce soil erosion by surface water flow in a greenhouse experiment

The role of arbuscular mycorrhizal fungi (AMF) in resisting surface flow soil erosion has never been tested experimentally. We set up a full factorial greenhouse experiment using Achillea millefolium with treatments consisting of addition of AMF inoculum and non-microbial filtrate, non-AMF inoculum and microbial filtrate, AMF inoculum and microbial filtrate, and non-AMF inoculum and non-microbial filtrate (control) which were subjected to a constant shear stress in the form of surface water flow to quantify the soil detachment rate through time. We found that soil loss can be explained by the combined effect of roots and AMF extraradical hyphae and we could disentangle the unique effect of AMF hyphal length, which significantly reduced soil loss, highlighting their potential importance in riparian systems.

Three-level main-effects designs exploiting prior information about model uncertainty

To explore the projection efficiency of a design, Tsai, et al [2000. Projective three-level main effects designs robust to model uncertainty. Biometrika 87, 467-475] introduced the Q criterion to compare three-level main-effects designs for quantitative factors that allow the consideration of interactions in addition to main effects. In this paper, we extend their method and focus on the case in which experimenters have some prior knowledge, in advance of running the experiment, about the probabilities of effects being non-negligible. A criterion which incorporates experimenters' prior beliefs about the importance of each effect is introduced to compare orthogonal, or nearly orthogonal, main effects designs with robustness to interactions as a secondary consideration. We show that this criterion, exploiting prior information about model uncertainty, can lead to more appropriate designs reflecting experimenters' prior beliefs. (c) 2006 Elsevier B.V. All rights reserved.

Efeito da Proteína Bruta e de Diferentes Balanços Eletrolíticos das Dietas Sobre o Desempenho de Frangos no Período Inicial

Foram realizados dois experimentos utilizando-se 520 pintos machos de um dia para avaliar o efeito da proteína bruta (PB) e do balanço eletrolítico (Na + K- Cl) sobre o desempenho de frangos de corte no período inicial. O Experimento I (1 a 7 dias de idade das aves) foi realizado em baterias, utilizando-se 160 pintos machos Cobb, no Experimento II (1 a 21 dias) foram utilizados 360 pintos machos Aviam Farms que foram alojados em um galpão experimental dividido em box. Nos dois experimentos as aves foram aquecidas com lâmpadas infravermelhas e receberam água e ração à vontade. O delineamento foi inteiramente casualizado, em um esquema fatorial 2 x 2 (PB x balanço eletrolítico - BE), com 5 repetições e 8 aves por unidade experimental no Experimento I e fatorial 2 x 3 (PB x BE), com 3 repetições e 20 aves por unidade experimental no Experimento II. Os níveis de K foram mantidos constantes, oscilando-se o Na e o Cl para obter os balanços eletrolíticos desejados. Os parâmetro de desempenho (consumo de ração, ganho de peso e conversão alimentar) foram analisados no final de cada fase experimental. em ambos os experimentos não houve interação entre PB e BE. Os níveis de 21,0 e 23,5% de PB não afetaram o desempenho das aves. em dietas pré-iniciais e iniciais o melhor desempenho foi atingido com 260 mEq/kg.

Estudo Comparativo de gráaficos de probabilidade normal para Anáalise de experimentos fatoriais não replicados

Two-level factorial designs are widely used in industrial experimentation. However, many factors in such a design require a large number of runs to perform the experiment, and too many replications of the treatments may not be feasible, considering limitations of resources and of time, making it expensive. In these cases, unreplicated designs are used. But, with only one replicate, there is no internal estimate of experimental error to make judgments about the significance of the observed efects. One of the possible solutions for this problem is to use normal plots or half-normal plots of the efects. Many experimenters use the normal plot, while others prefer the half-normal plot and, often, for both cases, without justification. The controversy about the use of these two graphical techniques motivates this work, once there is no register of formal procedure or statistical test that indicates \which one is best". The choice between the two plots seems to be a subjective issue. The central objective of this master's thesis is, then, to perform an experimental comparative study of the normal plot and half-normal plot in the context of the analysis of the 2k unreplicated factorial experiments. This study involves the construction of simulated scenarios, in which the graphics performance to detect significant efects and to identify outliers is evaluated in order to verify the following questions: Can be a plot better than other? In which situations? What kind of information does a plot increase to the analysis of the experiment that might complement those provided by the other plot? What are the restrictions on the use of graphics? Herewith, this work intends to confront these two techniques; to examine them simultaneously in order to identify similarities, diferences or relationships that contribute to the construction of a theoretical reference to justify or to aid in the experimenter's decision about which of the two graphical techniques to use and the reason for this use. The simulation results show that the half-normal plot is better to assist in the judgement of the efects, while the normal plot is recommended to detect outliers in the data

Multistratum response surface designs

Response surface designs are usually described as if the treatments have been completely randomized to the experimental units. However, in practice there is often a structure to the units, implying the need for blocking. If, in addition, some factors are more difficult to vary between units than others, a multistratum structure arises naturally. We present a general strategy for constructing response surface designs in multistratum unit structures. Designs are constructed stratum by stratum, starting in the highest stratum. In each stratum a prespecified treatment set for the factors applied in that stratum is arranged to be nearly orthogonal to the units in the higher strata, allowing-for all the effects that have to be estimated. Three examples are given to show the applicability of the method and are also used to check the relationship of the final design to the choice of treatment set. Finally, some practical considerations in randomization are discussed.

Row-column response surface designs

Factorial experiments are widely used in industry to investigate the effects of process factors on quality response variables. Many food processes, for example, are not only subject to variation between days, but also between different times of the day. Removing this variation using blocking factors leads to row-column designs. In this paper, an algorithm is described for constructing factorial row-column designs when the factors are quantitative, and the data are to be analysed by fitting a polynomial model. The row-column designs are constructed using an iterative interchange search, where interchanges that result in an improvement in the weighted mean of the efficiency factors corresponding to the parameters of interest are accepted. Some examples illustrating the performance of the algorithm are given.

Difference variance dispersion graphs for comparing response surface designs with applications in food technology

Variance dispersion graphs have become a popular tool in aiding the choice of a response surface design. Often differences in response from some particular point, such as the expected position of the optimum or standard operating conditions, are more important than the response itself. We describe two examples from food technology. In the first, an experiment was conducted to find the levels of three factors which optimized the yield of valuable products enzymatically synthesized from sugars and to discover how the yield changed as the levels of the factors were changed from the optimum. In the second example, an experiment was conducted on a mixing process for pastry dough to discover how three factors affected a number of properties of the pastry, with a view to using these factors to control the process. We introduce the difference variance dispersion graph (DVDG) to help in the choice of a design in these circumstances. The DVDG for blocked designs is developed and the examples are used to show how the DVDG can be used in practice. In both examples a design was chosen by using the DVDG, as well as other properties, and the experiments were conducted and produced results that were useful to the experimenters. In both cases the conclusions were drawn partly by comparing responses at different points on the response surface.

Optimum design of experiments for statistical inference

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Evaluation of reusable cardboard box designs: Biomechanical and perceptual aspects

Reusable cardboard boxes can be ergonomically designed for internal transportation of dry products in industrial settings. In this study we compared the effects of handling a regular commercial box and two cardboard prototypes on upper limb postures through the evaluation of movements, myoelectrical activity, perceived grip acceptability and capacity for reuse. The ergonomic designs provided a more acceptable grip, safer wrist and elbow movements and lower wrist extensors and biceps activity. Biomechanical disadvantages were observed only for one of the prototypes when handling to high surface. The prototypes were durable and suitable for extensive reuse (more than 2000 handlings) in internal industrial transportation. Despite being slightly more expensive than regular cardboard, the prototypes showed good cost-benefit considering their high durability. Relevance to industry: Cardboard boxes can be efficiently redesigned for allowing safer upper limb movements and lower muscle workload in manual materials handling. New designs can also be extensively reused for internal industrial transportation with good cost-benefit. © 2012 Elsevier B.V.

Dissemination of Caribbean census microdata to researchers: Including an experiment in the anonymization of microdata for Grenada and Trinidad and Tobago

Caribbean census microdata are not easily accessible to researchers. Although there are well-established and commonly used procedures technical, administrative and legal which are used to disseminate anonymized census microdata to researchers, they have not been widely used in the Caribbean. The small size of Caribbean countries makes anonymization relatively more difficult and standard methods are not always directly applicable. This study reviews commonly used methods of disseminating census microdata and considers their applicability to the Caribbean. It demonstrates the application of statistical disclosure control methods using the census datasets of Grenada and Trinidad and Tobago and considers various possible designs of microdata release file in terms of disclosure risk and utility to researchers. It then considers how various forms of microdata dissemination: public use files, licensed use files, remote data access and secure data laboratories could be used to disseminate census microdata. It concludes that there is scope for a substantial expansion of access to Caribbean census microdata and that through collaboration with international organisations and data archives, this can be achieved with relatively little burden on statistical offices.