Search

The FAIR (findable, accessible, interoperable and reusable) principles of scientific data management and stewardship are aimed at facilitating data reuse at scale by both humans and machines. Research and development (R&D) in the pharmaceutical industry is becoming increasingly data driven, but managing its data assets according to FAIR principles remains a costly and challenging endeavour. To date, little scientific evidence has been gathered about how FAIR is currently implemented in practice, what its associated costs and benefits are and how decisions are made about the FAIRification of existing datasets in pharmaceutical R&D. This thesis sets out to illuminate such issues, adding to the literature by documenting another critical aspect of FAIR - the decision-making process. To this end, semi-structured interviews were conducted with pharmaceutical professionals to examine their current practices in- depth and establish a conceptual model for the FAIRification decisions. Pharmaceutical industrial requirements for the design of a framework that aids decision making regarding FAIRification were identified. On the basis of the results, a decision-making framework called FAIR-Decide was developed using a novel method that involved the application of business analysis techniques (cost-benefit and multi-criteria analyses) in assessing estimated costs and expected benefits. To validate the framework, a FAIR-Decide tool was created and evaluated through focus group discussions of two scenarios (industry and non-industry) as a means of ascertaining the suitability of the tool for its intended work environment. The findings have significant implications for pharmaceutical R&D professionals engaged in driving FAIR implementation and for external parties who seek to better understand existing practices and challenges.

Current practice for sample size computations in clinical trials is largely based on frequentist or classical methods. These methods have the drawback of requiring a point estimate of the variance of treatment effect and are based on arbitrary settings of type I and II errors. They also do not directly address the question of achieving the best balance between the costs of the trial and the possible benefits by using a new medical treatment, and fail to consider the important fact that the number of users depends on evidence for improvement compared with the current treatment. A novel Bayesian approach, Behavioral Bayes (or BeBay for short) (Gittins and Pezeshk, 2000a,b, 2002a,b; Pezeshk, 2003), assumes that the number of patients switching to the new treatment depends on the strength of the evidence which is provided by clinical trials, and takes a value between zero and the number of potential patients in the country. The better a new treatment, the more patients switch to it and the more the resulting benefit. The model defines the optimal sample size to be the sample size that maximises the expected net benefit resulting from a clinical trial. Gittins and Pezeshk use a simple form of benefit function for paired comparisons between two medical treatments and assume that the variance of the efficacy is known. The research in this thesis generalises these original conditions by introducing a logistic benefit function to take account of differences in efficacy and safety between two drugs. The model is also extended to the more general cases of unpaired comparisons and unknown variance. The expected net benefit defined by Gittins and Pezeshk is based on the efficacy of the new drug only. It does not consider the incidence of adverse reactions and their effect on patients’ preferences. Here we include the costs of treating adverse reactions and calculate the total benefit in terms of how much the new drug can reduce societal expenditure. We describe how our model may be used for the design of phase III clinical trials, cluster randomised clinical trials and bridging studies. This is done in some detail and using illustrative examples based on published studies. For phase III trials we allow the possibility of unequal treatment group sizes, which often occur in practice. Bridging studies are those carried out to extend the range of applicability of an established drug, for example to new ethnic groups. Throughout the objective of our procedures is to optimise the costbenefit in terms of national health-care. BeBay is the leading methodology for determining sample sizes on this basis. It explicitly takes account of the roles of three decision makers, namely patients and doctors, pharmaceutical companies and the health authority.

Abstract: More than 40% of working-age Canadians live with low literacy skills. Furthermore, there is no foreseeable decline in this striking percentage, with nearly 20% of Canadians ages 16-19 demonstrating low literacy skills. To address this literacy deficit, numerous interventions are available that address the fundamental building blocks of literacy. Interventions often vary in terms of costs (i.e., materials, training for teachers, time), and choosing the most effective interventions from the abundance of options can be a difficult algorithm to navigate. Further, choosing cost-effective interventions can benefit students while efficiently using limited resource funds. My aims in the present study were to develop a novel process for analyzing the cost-benefit of academic interventions, and to apply this process to fluency (i.e., word reading speed) interventions. The cost-benefit process compares interventions by studying the effect size benefit per unit of cost. In applying this process to fluency interventions, I reviewed 614 publications, and was able to successfully compare two interventions (i.e., QuickReads, Helping Early Literacy with Practice Strategies; HELPS). The HELPS intervention was seven times more cost-effective than QuickReads, which suggests that HELPS may increase fluency (i.e., word reading speed) gains more quickly and use fewer resources than QuickReads. The cost-benefit methodology developed in the present study provided useful comparisons and is potentially adaptable for other academic interventions.

Challenges exist in breeding horticultural fruit crops in the Rosaceae family including self-incompatibility, high heterozygosity, and lengthy juvenile phases. Breeders are constantly exploring new methods of overcoming these challenges with marker-assisted selection (MAS) being a prominent example in current programs. MAS has been identified as a method for improving crops by countering some of these challenges. A list of criteria has been developed to gauge the effectiveness of MAS in a program. For MAS to be effective: the trait should be simply inherited, the trait should be expressed in mature plants, and monetary and genetic resources must exist for implementation (Luby and Shaw, 2001). The juvenile phase in some of these crops is lengthy and require multiple years to achieve fruit production. MAS allows the screening of younger plants in the program and earlier removal before considerable labor cost investments. However, breeders have difficulty incorporating MAS and other genetic technologies due to a lack of DNA test availability testing facilities, uncertainties calculating cost of utilization, and a lack of monetary investment (Ru et al., 2015). This research focuses on removing some of these barriers by developing a budget tool to identify cost-efficient DNA-informed breeding applications across three breeding operations in rosaceous breeding programs: apple, peach, and strawberry. The budget tool incorporates aspects of economics (both fixed and variable costs) and plant breeding (breeding records and program details). This tool will allow breeders to more quickly and efficiently judge how implementation of DNA technologies affects costs and benefits to their program.

To achieve more cost efficient transportation infrastructure within the state of Arkansas, a series of cost-benefit analyses were performed. The analyses, specifically developed for load resistance factor design (LRFD) designed drilled shaft foundations, were designed to compare the fiscal impacts of the drilling and sampling, in-field and laboratory testing, full-scale load testing, and concrete mixing techniques utilized by the University of Arkansas (UofA), the Arkansas Highway and Transportation Department (AHTD), and the Missouri Department of Transportation (MODOT). The methodologies were compared at three test site locations within the state of Arkansas: Siloam Springs, AR, Turrell, AR, and Monticello, AR. These sites were selected to represent the predominant geological deposits within the state. Three drilled shaft foundations (two four-foot diameter drilled shafts and one six-foot diameter drilled shaft) were each constructed at the Siloam Springs Arkansas Test Site (SSATS) and the Turrell Arkansas Test Site (TATS). A drilled shaft database was developed containing soil properties, and predicted and measured results of the full-scale load tests performed upon each drilled shaft. Cost-benefit analyses for each methodology were compared by means of predicted and measured static axial capacity. Values of predicted axial capacity were generated for each methodology using the Bridge Software Institute FB-Deep, Ensoft SHAFTv2012, and Microsoft Excel® spreadsheet programs. Based on the results of the full-scale load testing, the FB-Deep program utilizing the data from the UofA sampling and testing method was selected as the best methodology for predicting the axial capacity values for drilled shaft foundations in the state of Arkansas. Results of the cost-benefit analyses indicated a potential savings of $262,800 (32 percent) for drilled shaft foundations in rock. A potential savings of $323,800 for performing full-scale drilled shaft load tests in rock were attained. Unit cost per ton of resistance values of $24.11 and $82.70, and $75.47 and $141.57 were determined for the UofA and AHTD sampling and testing methods at the SSATS and TATS, respectively. Drilled shaft foundations tested in liquefiable soil were concluded to cost $137,500 (8.7%) more than driven pile foundations, but provided the benefit of additional lateral resistance.

The kinetic energy in river streams, tidal currents, or other artificial water channels has been used as a feasible source of renewable power through different conversion systems. Thus, hydrokinetic energy conversion systems are attracting worldwide interest as another form of distributed alternative energy. Because these systems are still in early stages of development, the basic approaches need significant research. The main challenges are not only to have efficient systems, but also to convert energy more economically so that the cost-benefit analysis drives the growth of this alternative energy form. One way to view this analysis is in terms of the energy conversion efficiency per unit cost. This study presents a detailed assessment of a prototype hydrokinetic energy system along with power output costs. This experimental study was performed using commercial low-cost blades of 20 in diameter inside a tank with water flow speed up to 1.3 m/s. The work was divided into two stages: (a) a fixed-pitch blade configuration, using a radial permanent magnet generator (PMG), and (b) the same hydrokinetic turbine, with a variable-pitch blade and an axial-flux PMG. The results indicate that even though the efficiency of a simple blade configuration is not high, the power coefficient is in the range of other, more complicated designs/prototypes. Additionally, the low manufacturing and operation costs of this system offer an option for low-cost distributed power applications.

This thesis presents a new approach to understanding social transmission among nonhuman animals. Drawing on basic principles of behavioral ecology, I suggest that animals are likely to test new behaviors when the payoffs for doing so exceed the payoffs for engaging in familiar behaviors. If true, expected utility theory and Bayesian decision making models can help us predict how confident of success an individual must be before trying a new behavior. When the probability of a positive payoff exceeds the predicted threshold, the individual should attempt the new behavior. Otherwise, the individual should refrain from experimentation. Animals can assess the probability of a positive payoff by attending to social and environmental cues. Reliable cues provide animals with the required level of confidence. Animals should heed reliable cues, and they should depend exclusively on social cues when they are the only reliable cues available. I investigate the implications of this cue reliability approach for our understanding of decisions about (1) sampling new foods, (2) responding to potential predators, and (3) responding to unfamiliar conspecifics. I suggest that once we understand the expected payoffs associated with such decisions, we can use this information to test for the existence of traditions among wild populations. I also offer a formal analysis of social cue reliability and present a new model of the evolution of teaching that derives from the work of Hamilton (1964) and Trivers (1974). Although previous research provides some support for the cue reliability approach, more studies are needed to test its quantitative predications. I report the results of three experiments that examine how animals respond to reliable and unreliable cues. As predicted, the results indicate that animals discriminate between reliable and unreliable cues. The results also suggest that animals depend more on social cues when nonsocial cues are unreliable.