Evidence‐Based Medicine: The Design and Interpretation of Noninferiority Clinical Trials in Veterinary Medicine

Noninferiority trials are clinical studies designed to demonstrate that an investigational drug is at least as effective as an established treatment within a predetermined margin. They are conducted, in part, because of ethical concerns of administering a placebo to veterinary patients when an established effective treatment exists. The use of noninferiority trial designs has become more common in veterinary medicine with the increasing number of established veterinary therapeutics and the desire to eliminate potential pain or distress in a placebo‐controlled study. Selecting the appropriate active control and an a priori noninferiority margin between the investigational and active control drug are unique and critical design factors for noninferiority studies. Without reliable historical knowledge of the disease response in the absence of treatment and of the response to the selected active control drug, proper design and interpretation of a noninferiority trial is not possible. Despite the appeal of conducting noninferiority trials to eliminate ethical concerns of placebo‐controlled studies, there are real limitations and possible ethical conundrums associated with noninferiority trials. The consequences of incorrect study conclusions because of poor noninferiority trial design need careful attention. Alternative trial designs to typical noninferiority studies exist, but these too have limitations and must also be carefully considered.     

Advances in total joint replacement in small animals

Total joint replacement is now considered a routine surgical option for small animals with advanced joint disease. This review highlights the current state of the art in total hip, elbow and knee replacement in small animals, as well as the potential for application to other joints such as the intervertebral discs, hock and shoulder. Advances in cementless fixation, the use of less-invasive, bone-sparing procedures, and the development of custom implant options for revision surgery and oncologic reconstruction will be discussed. As growth in this field continues, it is important that future developments in total joint replacement are built on a foundation of detailed laboratory evaluation of new implant designs and surgical techniques, objective assessment of clinical outcomes, and a comprehensive, coordinated retrieval program that evaluates both well-functioning and failed joint replacements.     

Practical considerations for ventilating calf barns in winter.

The use of air sampling devices to measure the concentrations of airborne bacteria in clinical investigations and research trials in calf barns has indicated that traditional systems of ventilation are problematic in cold weather. Individual pen designs should have two solid sides, but the front and rear should be as open as possible. Thermal stress should be managed by providing deep bedding and not by enclosing the pen. Air hygiene can be improved by reducing stocking density and using supplemental positive-pressure ventilation systems to deliver small amounts of air to each pen. Implementation of these recommendations can produce calf barns that seem to equal calf hutches in minimizing disease and provide better working conditions for the caregivers.     

Experimental design for integrated research projects to estimate genetic and numerous treatment effects

There is a trend towards integrated research, where experimenters aim to make the best possible use of available resources, and individuals or institutions pool their expertise, make use of common resources and collaborate towards a common set of scientific goals. This allows a larger number of factors to be investigated, enabling the most influential or important ones to be identified as well as providing information on how the different factors interact or fit together.The issues involved in generating complex multi-factor designs are described and discussed, using as examples the entire series of experiments in the Australian Beef Cattle CRC and a simpler experiment to estimate genetic marker effects. An algorithm to generate suitable designs is presented. For the genetic marker experiment, the resultant designs were up to 10% more efficient than less sophisticated designs. In the case of the Beef Cattle CRC, achieving the same accuracy of estimating treatment and sire effects without sophisticated designs would have required 5–10% more animals, at a cost of $150,000–300,000 for purchase, transport and feeding of animals. If all additional costs of experimentation were included, the total savings from use of efficient designs were estimated to lie between $0.5 and $1 million.     

Variance decomposition in the estimation of genetic variance with selected data.

Estimation of genetic variance in populations under selection involves assumptions on base animals. Base animals are often considered unselected and it also has been proposed to treat selected base animals as fixed. The consequences of assumptions on base animals in the estimation of genetic variance in selected populations are not fully understood. Variance decompositions are introduced for simple designs to quantify the differences between models that treat base animals in different ways. Independent contrasts were constructed and REML estimates of variance components were compared for different designs and selection rules. The method shows how selection is accounted for in a complete model and why estimation of variance components can become biased when base animals are treated as fixed.     

Controlled but pragmatic investigations of interventions for behavioural disturbances in dogs and cats

Veterinarians are often directly involved in clinical studies or requested for information to help interpret their results. Therefore, it is reasonable to examine the reservoir of study methods. This article transfers methodological considerations from clinical research into veterinary medicine. The study question determines the appropriate study method. Recently a ten-step procedure was suggested for selection of appropriate study designs in humans. Based on this approach, a pragmatic study design was adapted to the conditions prevailing in interventional studies in dogs and cats with disturbed behaviour. The different concepts for clinical studies are introduced. Whether or not the design and the evaluation of pragmatic studies in dogs and cats with disturbed behaviour has been maintained and the prerequisites have thereby been fulfilled so that the obtained results are suitable to be applied under everyday conditions can be tested in eight steps. Using the pragmatic design the superiority of complex interventions can be investigated. The results of pragmatic studies help to substantiate a value judgement, i. e., the recommendation or rejection of a specific therapeutic intervention for a defined disease entity in a specific therapeutic setting. The goal of pragmatic studies is to obtain results appropriate for use in everyday situations. In conclusion, the suggested procedure is useful for the selection of the appropriate study designs for specific questions. This procedure is also suitable to test whether the conclusions of published study results coincide with the chosen methods.     

Sampling genotype configurations in a large complex pedigree

Many genetic problems can be solved by Monte Carlo method. This often requires sampling genotype configurations over pedigree. Current available samplers are inefficient for large animal pedigrees. A new sampler suitable for large complex pedigrees has been developed and evaluated. The sampler uses simple and iterative peeling algorithms alternately. The sampler was compared to two other samplers on hypothetical pedigree of 79 individuals and recessive disease. The behaviour of the sampler was evaluated in four experimental designs on real bovine pedigree of 907,903 animals. The application of the sampler was also exemplified in identical by descent study.     

Bayesian estimation of variance partition coefficients adjusted for imperfect test sensitivity and specificity

The variance partition coefficient (VPC) measures the clustering of infection/disease among individuals with a specific covariate pattern. Covariate-pattern-specific VPCs provide insight to the groups of individuals that exhibit great heterogeneity and should be targeted for intervention. VPCs should be taken into consideration when planning study designs, modeling data and estimating sample sizes. We present a Bayesian discrete mixed model for the estimation of covariate-pattern-specific VPCs when measurement of the infection/disease is based on an imperfect test. The utility of the presented model is demonstrated with three applications. In all cases, imperfect tests biased VPC estimates towards the null but corrected estimates could be obtained by modeling the sensitivity and specificity of the test procedure with beta distributions. The comparison of adjusted VPCs between the intercept only and the fitted models with higher level covariates explained the portion of heterogeneity in the data that was accounted for by the covariates.     

A review of dicrocoeliosis of ruminants including recent advances in the diagnosis and treatment

Despite its widespread presence among grazing ruminants, dicrocoeliosis, also known as "small liver fluke" disease, is poorly known and often underestimated by researchers and practitioners in many countries. This is primarily due to the multiple parasitic infections which affect ruminant livestock and mask the pathology of dicrocoeliosis, to the difficulties in diagnosing it with coprological techniques and, finally, to the few effective drugs found. Furthermore, the biological cycle of Dicrocoelium, which requires a snail and an ant as intermediate hosts, and the high number of ecological and epidemiological variables affecting the disease make it difficult to set up experimental designs to study dicrocoeliosis. In the past 50 years, many aspects of this disease have been broadly investigated (aetiology, life cycle, diffusion, epidemiology, pathogenesis and immunology) but its diagnosis and treatment still remain moot issues. Dicrocoeliosis often remains clinically undetected and its diagnosis is mostly based on adult dicrocoelia recovered in the liver post mortem or on egg detected at coprological examination. The prophylaxis of the small liver fluke has been difficult and unsatisfactory to date due to the complexity of its biological life cycle and epidemiology. Many anti-helminthic drugs are practically ineffective against dicrocoeliosis if used at the dosage recommended against other gastrointestinal helminths and lungworms. The most important aspects of the aetiology, biological cycle, spread, epidemiology and pathogenesis of dicrocoeliosis are reviewed and the recent advances in the diagnosis and treatment are focused on.     

A simulation-based study comparing a traditional and an alternative design for studies of experimentally induced intestinal diseases in pigs

A traditional design for studies of pathogenesis of experimentally induced diseases of the large intestine in individual pigs involves euthanasia and necropsy at scheduled times. An alternative design has been developed enabling the sequential in vivo monitoring of events in the intestine of the individual pig before and during disease. The alternative design is based on repeated endoscopy and biopsy sampling to monitor the course of the disease. The purpose of this study was to compare the statistical properties of the two designs. One hundred replicates of a computer-generated simulation of data were performed. In the simulation, data were generated for individual pigs according to the alternative design, i.e. four sequential observations on the outcome of interest, and the 'traditional' design was created by retaining only one observation per pig. Data from the alternative design were analysed with a generalized linear mixed model, accounting for the clustering of observations within animals and the time-dependence between observations within animals, i.e. observations within animals are interdependent and follow a pattern over time. Data from the 'traditional' design were analysed with an ordinary linear model. The results showed that the alternative design was much more efficient than the traditional and that only one-fourth as many pigs were required to achieve the same precision in estimated effects.     

INVITED REVIEW: STUDY DESIGN CONSIDERATIONS FOR CLINICAL RESEARCH IN VETERINARY RADIOLOGY AND RADIATION ONCOLOGY

High quality clinical research is essential for advancing knowledge in the areas of veterinary radiology and radiation oncology. Types of clinical research studies may include experimental studies, method‐comparison studies, and patient‐based studies. Experimental studies explore issues relative to pathophysiology, patient safety, and treatment efficacy. Method‐comparison studies evaluate agreement between techniques or between observers. Patient‐based studies investigate naturally acquired disease and focus on questions asked in clinical practice that relate to individuals or populations (e.g., risk, accuracy, or prognosis). Careful preplanning and study design are essential in order to achieve valid results. A key point to planning studies is ensuring that the design is tailored to the study objectives. Good design includes a comprehensive literature review, asking suitable questions, selecting the proper sample population, collecting the appropriate data, performing the correct statistical analyses, and drawing conclusions supported by the available evidence. Most study designs are classified by whether they are experimental or observational, longitudinal or cross‐sectional, and prospective or retrospective. Additional features (e.g., controlled, randomized, or blinded) may be described that address bias. Two related challenging aspects of study design are defining an important research question and selecting an appropriate sample population. The sample population should represent the target population as much as possible. Furthermore, when comparing groups, it is important that the groups are as alike to each other as possible except for the variables of interest. Medical images are well suited for clinical research because imaging signs are categorical or numerical variables that might be predictors or outcomes of diseases or treatments.     

Behavioral needs of the transition cow and considerations for special needs facility design.

Improved building designs come from a better understanding of the behavioral needs of the dairy cow. The costs to provide for these needs in the facility must be offset by improved milk production, health, and longevity. Research is still required to more fully understand the health implications of many building design considerations and their impact on disease. Perhaps the most important end result of an improved environment for the transition cow, however, is an improvement in animal well-being. Better buildings that accommodate the behavioral needs of cows present "win-win" situations where dairy cattle thrive and work is more enjoyable. This results in an improved image for the industry, greater consumer confidence in the quality and safety of the final food product, and a prosperous dairy industry.     

鸡群对病毒灭活疫苗免疫后抗体动态检测的不同实验设计和统计学分析的比较研究

面对普遍发生的产蛋下降,许多种鸡场或商品代蛋鸡场往往不得不在产蛋期甚至产蛋高峰期用灭活疫苗强化免疫.但过度频繁的强化免疫,常给产蛋鸡的生产性能带来严重的不良影响.合理的免疫程序应是在抗体滴度刚开始下降时即给以强化免疫.本研究试图比较不同统计学方法对抗体动态分析的影响,以期为免疫程序的设计提供参考.分别按不同的统计学设计和分析方法,在一个大型种鸡场的产蛋期对新城疫病毒、H9-和H5-禽流感病毒抗体滴度动态变化做了比较研究.对免疫后抗体动态变化的动态研究表明,用成对数据比较的t-test比按成组数据平均数的t-test更能准确可靠地确定免疫后抗体滴度的真实下降.在开产前用选择的最好的疫苗免疫后,高滴度的抗体水平可稳定地维持21周.在此期间对3种病毒的疫苗均未做任何强化免疫,直到21周后真正开始下降时才实施加强免疫.按此免疫程序,该种鸡群的产蛋率达到了历史最高水平.成对数据比较的t-test既可避免由于技术误差或系统误差产生某时段抗体下降的假象,也可避免由于个体差异而掩盖了幅度不大但确是真实的抗体滴度下降,可以为制定最合理的免疫程序提供科学的数据.     

Total hip arthroplasty for treatment of canine hip dysplasia.

Cemented THA is an established procedure for treating arthropathies of the hip in large, healthy mature dogs. CHD with secondary osteoarthritis is the most common indication. Although comparative studies between THA and excision arthroplasty have not been reported, our experience with both procedures and studies in the literature support THA as the best available treatment for crippling degenerative joint disease secondary to CHD in large, mature dogs. Negative aspects of the procedure include high cost, the potential of significant complications, and the sophisticated surgical technique required. Improvements in patient selection, design of implants, surgical technique, and postoperative care have decreased the incidence of complications and improved the success rate to over 90%. However, concerns in human applications about cement disease and the desire to increase the functional life of the prosthesis have renewed interest in cementless systems. Advantages of cementless THA are fixation via bony ingrowth and avoidance of problems associated with PMMA. Important technical aspects of cementless THA relate to optimal fit and fill so that the prosthesis is stable, bone ingrowth is promoted, and weight-bearing forces are transferred to the proximal femur physiologically. Problems recognized with cementless systems include fissure fracture, bone resorption, and excessive motion between the implant and bone. Currently, investigations are being conducted to develop prosthetic materials that more closely match the stiffness of bone, stem designs that provide optimal fit and maintain normal strain patterns in the bone, and coating materials that promote permanent fixation by bone ingrowth. Early clinical results in humans and research results in dogs have been encouraging but have not withstood the test of time. Results of cementless techniques must be compared to the standards set by cemented THA over the last 20 years in humans and the last 10 years in dogs.     

Evaluation and interpretation of results for three cross-over designs.

In cross-over designs, individual sequences of treatments are applied to the animals. Within such designs it is possible that every treatment could modify the effect of the subsequent treatment applied to the same animal. We compared three cross-over designs each with three treatments, three periods, and two blocks. This comparison was done with respect to the variance of the estimations of the effects and its biases caused by the interactions between the treatment and the carry over effect of the foregoing treatment. Moreover, different methods of estimating variance components and calculating the degrees of freedom were compared by means of simulation. If the animal variance component is small, then the bias of the REML estimator of the variance components is greater than one of the widespread ANOVA-estimator called 'TYPE3'. But nevertheless, the mean squared error of this estimation is smaller in the case of REML in comparison to ANOVA. Therefore, the REML method should be preferred. For calculating the degrees of freedom, the Kenward-Roger method should be used. After applying this method, the true significance level is almost equal to its required value, but if the Satterthwaite method is used, the true significance level will be too high. If the interaction (treatment x carry over) is ignored in the model although it exists, the standard error of the treatment effect estimation is too great, and, therefore, the true significance level is too small. The methods which have been evaluated are available in the SAS-procedure MIXED (SAS Institute, 1999a). To assist the investigation of cross-over designs by using this software, we developed programs for data management and data analysis. These programs are available from the first author.     

Pre-harvest interventions to reduce the shedding of E. coli O157 in the faeces of weaned domestic ruminants: a systematic review

Our objective was to use formal systematic review methods to evaluate the efficacy of interventions to reduce faecal shedding of Escherichia coli O157 in post-weaned ruminants by increasing animal resistance. The methodology consisted of an extensive search to identify all potentially relevant research, screening of titles and abstracts for relevance to the research question, quality assessment of relevant research, extraction of data from research of sufficient quality, and qualitative summarization of results. The interventions evaluated included probiotics, vaccination, antimicrobials, sodium chlorate, bacteriophages and other feed additives. There was evidence of efficacy for the probiotic combination Lactobacillus acidophilus NP51 (NPC 747) and Propionibacterium freudenreichii and for sodium chlorate in feed or water. The effectiveness of vaccination varied among studies and among vaccine protocols and there was no consistent evidence to suggest that antibiotic use was associated with a decrease in faecal shedding of E. coli O157, or that current industry uses of antimicrobials were associated with increased faecal shedding. There were an insufficient number of studies available to address the effectiveness of bacteriophages and several other feed additives. In general, few of the primary studies evaluated the interventions under commercial housing conditions with a natural disease challenge, there were inconsistencies in the results among study designs and in some cases among studies within study designs, and a relatively large proportion of publications were excluded based on quality assessment criteria. Few studies reported on associations between the proposed intervention and production parameters, such as average daily gain and feed: gain ratio. While the results suggest that some interventions may be efficacious, there are knowledge gaps in our understanding of the efficacy of pre-harvest interventions to increase animal resistance to E. coli O157 that require further targeted research.     

Assessing statistical precision, power, and robustness of alternative experimental designs for two color microarray platforms based on mixed effects models

Recommendations on experimental designs for two color microarray systems have been generally conflicting as they pertain to the general choice between reference and non-reference loop designs. This conflict may currently exist because many previously published assessments may not have effectively connected design layout with the level of biological relative to technical replication. We reassess various reference and non-reference designs for statistical efficiency in terms of standard errors of mean differences, power of test, and robustness using recently developed mixed model software tools. In minimally replicated cases (n = 2), it appears that the reference design outperforms the classical loop design whereby a sample from each animal is used for only one particular array hybridization. Alternatively, the reference design was consistently inferior to those connected loop designs in which a sample from each animal is used in two different hybridizations. Nevertheless, the gap in power between these two designs diminished as the biological to residual variance ratio increased. The statistical efficiency of a single large classical loop design for the comparison of many treatments was demonstrated to be highly sensitive to missing arrays relative to a common reference design (n = 2). However, the use of two loops within an interwoven loop design was shown to be substantially more robust to missing arrays and statistically more efficient relative to a common reference design. Furthermore, the use of more than one loop leads to less disparity in precision and power comparisons between any two treatments.     

Utility, limitations, and promise of proteomics in animal science

Proteomics experiments have the ability to simultaneously identify and quantify thousands of proteins in one experiment. The use of this technology in veterinary/animal science is still in its infancy, yet it holds significant promise as a method for advancing veterinary/animal science research. Examples of current experimental designs and capabilities of proteomic technology and basic principles of mass spectrometry are discussed. In addition, challenges and limitations of proteomics are presented, stressing those that are unique to veterinary/animal sciences.     

Quantitative trait locus mapping based on selective DNA pooling

Concepts of a simple method to map quantitative trait loci (QTL) based on selective DNA pooling in half-sib family, backcross, and F₂ designs were developed. It is shown that the position of a QTL can be estimated from differences in allele frequencies for two flanking markers between individuals with high and low phenotypes and does not depend on the phenotypic means of the selected groups. An estimate of the QTL effect was obtained by relating group differences in phenotypic means to differences in QTL frequencies, which can be estimated from the QTL position and marker allele frequencies. Simulation of a half-sib family and a F₂ family of 2000 individuals showed that the method gives close to unbiased results when power is high. Biases increased when measurement errors on marker allele frequencies increased and when the effect of the QTL was small. Similarities of QTL mapping based on selective DNA pooling data and on individual genotyping data are discussed, as are opportunities to extend the selective DNA pooling method to the use of multiple markers and multiple half-sib family designs. This study shows that the use of selective DNA pooling can be extended from the detection of marker associations to the mapping of QTL. Selective DNA pooling can greatly reduce the number of genotypings required.