Pretty Darn Good Control: When are Approximate Solutions Better than Approximate Models

Existing methods for optimal control struggle to deal with the complexity commonly encountered in real-world systems, including dimensionality, process error, model bias and data heterogeneity. Instead of tackling these system complexities directly, researchers have typically sought to simplify models to fit optimal control methods. But when is the optimal solution to an approximate, stylized model better than an approximate solution to a more accurate model? While this question has largely gone unanswered owing to the difficulty of finding even approximate solutions for complex models, recent algorithmic and computational advances in deep reinforcement learning (DRL) might finally allow us to address these questions. DRL methods have to date been applied primarily in the context of games or robotic mechanics, which operate under precisely known rules. Here, we demonstrate the ability for DRL algorithms using deep neural networks to successfully approximate solutions (the "policy function" or control rule) in a non-linear three-variable model for a fishery without knowing or ever attempting to infer a model for the process itself. We find that the reinforcement learning agent discovers a policy that outperforms both constant escapement and constant mortality policies-the standard family of policies considered in fishery management. This DRL policy has the shape of a constant escapement policy whose escapement values depend on the stock sizes of other species in the model.

Toward a Standard of Medical Care: Why Medical Professionals Can Refuse to Prescribe Puberty Blockers

That a standard of medical care must outline services that benefit the patient is relatively uncontroversial. However, one must determine how the practices outlined in a medical standard of care should benefit the patient. I will argue that practices outlined in a standard of medical care must not detract from the patient's well-functioning and that clinicians can refuse to provide services that do. This paper, therefore, will advance the following two claims: (1) a standard of medical care must not cause dysfunction, and (2) if a physician is medically rational to not provide some service which fails to meet the above condition (i.e. fails to be a standard of medical care), then she may refuse to do so. I then apply my thesis to the prescription of puberty blockers to children with gender dysphoria.

Decompressive Craniectomy After Traumatic Brain Injury: Incorporating Patient Preferences into Decision-Making

BACKGROUND

Decompressive craniectomy (DC) is highly effective in relieving intracranial hypertension; however, patient selection, intracranial pressure threshold, timing, and long-term functional outcomes are all subject to controversy. Recently, recommendations were made to update the Brain Trauma Foundation guidelines in regards to the use of DC based on the DECRA (Decompressive Craniectomy in Patients with Severe Traumatic Brain Injury) and RESCUEicp (Trial of Decompressive Craniectomy for Traumatic Intracranial Hypertension) clinical trials. Neither the updated recommendations, nor the aforementioned trials, provide a method in incorporating individualized patient or surrogate decision-maker preferences into decision making.

METHODS

In this manuscript, we aimed to redress the gap of not incorporating patient preferences in such value-laden decision making as in the case of DC for refractory post-traumatic intracranial hypertension. We proposed a decision aid based on principles of Decision Theory, and specifically of Expected Utility Theory.

RESULTS

We showed that 1) early secondary DC as studied in DECRA, and based on the 1-year outcome data, is associated with decreased expected utility for all possible preference rankings of outcomes; and 2) recommending a late secondary DC versus tier-3 medical therapy, as studied in RESCUEicp, should be informed by individualized patient preference rankings of outcomes as elicited via shared decision-making.

CONCLUSIONS

The 1-year outcomes from DECRA and RESCUEicp have served as the basis for updated guidelines. However, unaided interpretation of trial data may not be adequate for individualized decision-making; we suggest that the latter can be significantly supported by decision aids such as the one described here and based on expected utility theory.

Converging global crises are forcing the rapid adoption of disruptive changes in drug discovery

Spiralling research costs combined with urgent pressures from the Coronavirus 2019 (COVID-19) pandemic and the consequences of climate disruption are forcing changes in drug discovery. Increasing the predictive power of in vitro human assays and using them earlier in discovery would refocus resources on more successful research strategies and reduce animal studies. Increasing laboratory automation enables effective social distancing for researchers, while allowing integrated data capture from remote laboratory networks. Such disruptive changes would not only enable more cost-effective drug discovery, but could also reduce the overall carbon footprint of discovering new drugs.

How hospitals can improve their public quality metrics: a decision-theoretic model

The public reporting of hospitals' quality of care is providing additional motivation for hospitals to deliver high-quality patient care. Hospital Compare, a consumer-oriented website by the Centers for Medicare and Medicaid Services (CMS), provides patients with detailed quality of care data on most US hospitals. Given that many quality metrics are the aggregate result of physicians' individual clinical decisions, the question arises if and how hospitals could influence their physicians so that their decisions positively contribute to hospitals' quality goals. In this paper, we develop a decision-theoretic model to explore how three different hospital interventions-incentivization, training, and nudging-may affect physicians' decisions. We focus our analysis on Outpatient Measure 14 (OP-14), which is an imaging quality metric that reports the percentage of outpatients with a brain computed tomography (CT) scan, who also received a same-day sinus CT scan. In most cases, same-day brain and sinus CT scans are considered unnecessary, and high utilizing hospitals aim to reduce their OP-14 metric. Our model captures the physicians' imaging decision process accounting for medical and behavioral factors, in particular the uncertainty in clinical assessment and a physician's diagnostic ability. Our analysis shows how hospital interventions of incentivization, training, and nudging affect physician decisions and consequently OP-14. This decision-theoretic model provides a foundation to develop insights for policy makers on the multi-level effects of their policy decisions.

Certainty of evidence and intervention's benefits and harms are key determinants of guidelines' recommendations

OBJECTIVE

Many factors are postulated to affect guidelines developments. We set out to identify the key determinants.

STUDY DESIGN AND SETTING

a) Web-based survey of 12 panels of 153 "voting" members who issued 2941 recommendations; b) qualitative analysis of 13 panels of 311 attendees (panel members, systematic review teams and observers).

RESULTS

Compared with "no recommendations", when intervention's benefit outweigh harms (BH-balance), probability of issuing strong recommendations in favor of intervention was 0.22 (95%CI: 0.08 to 0.36) when certainty of evidence (CoE) was very low; 0.5 (95%CI:0.36 to 0.63) when low; 0.74 (95%CI 0.61 to 0.87) when moderate and 0.85 (95%CI:0.71 to 1.00) when high. No other postulated factor significantly affected recommendations. The findings are consistent with a J- curve model when recommendations are issued in favor but not against an intervention. Panelists often changed their judgments as a result of the meeting discussion (67% for CoE to 92% for balance between benefits and harms). The panels spent over 50% of their time debating CoE; the chairs and co-chairs dominated discussion.

CONCLUSIONS

CoE and BH-balance are key determinants of recommendations in favor of an intervention. Chairs and co-chairs dominate discussion. Panelists often change their judgments as a result of panel deliberation.

A rational route to transformative decisions

According to Paul (Transformative experience, 1st edn, Oxford University Press, 2014), transformative experiences pose a challenge to decision theory since their value cannot be anticipated. Building on Pettigrew's (in: Lambert, Schwenkler (eds) Becoming someone new: essays on transformative experience, choice, and change, Oxford University Press, pp 100-121, 2020) redescription, this paper presents a new approach to how and when transformative decisions can nevertheless be made rationally. Thanks to fundamental higher-order facts that apply to any kind of experience, an agent always at least knows the general shape of the utility space. This in combination with the knowledge about the non-transformative alternative in the choice set can enable rational decision-making despite the presence of a transformative experience. For example, this paper's approach provides novel arguments for why gender transition (cf. McKinnon in Res Philosophica 92(2):419-440, 2015) or staying childfree (cf. Barnes in Philos Phenomenol Res 91(3):775-786, 2015) can be rational.

Evaluating Biosignatures for Life Detection

Conceptual frameworks are developed for evaluating the ability of different biosignatures to provide evidence for the presence of life in planned missions or observational studies. The focus is on intrinsic characteristics of biosignatures in space environments rather than on their detection, which depends on technology. Evaluation procedures are drawn from extensive studies in decision theory on related problems in business, engineering, medical fields, and the social arena. Three approaches are particularly useful. Two of them, Signal Detection Theory and Bayesian hypothesis testing, are based on probabilities. The third approach is based on utility theory. In all the frameworks, knowledge about a subject matter has to be translated into probabilities and/or utilities in a multistep process called elicitation. We present the first attempt to cover all steps, from acquiring knowledge about biosignatures to assigning probabilities or utilities to global quantities, such as false positives and false negatives. Since elicitation involves human judgment that is always prone to perceptual and cognitive biases, the relevant biases are discussed and illustrated in examples. We further discuss at which stage of elicitation human judgment should be involved to ensure the most reliable outcomes. An example, how evaluating biosignatures might be implemented, is given in the Supplementary Information.

UNDERSTANDING DOCTOR DECISION MAKING: THE CASE OF DEPRESSION TREATMENT

Treatment for depression is complex, requiring decisions that may involve trade-offs between exploiting treatments with the highest expected value and experimenting with treatments with higher possible payoffs. Using patient claims data, we show that among skilled doctors, using a broader portfolio of drugs predicts better patient outcomes, except in cases where doctors' decisions violate loose professional guidelines. We introduce a behavioral model of decision making guided by our empirical observations. The model's novel feature is that the trade-off between exploitation and experimentation depends on the doctor's diagnostic skill. The model predicts that higher diagnostic skill leads to greater diversity in drug choice and better matching of drugs to patients even among doctors with the same initial beliefs regarding drug effectiveness. Consistent with the finding that guideline violations predict poorer patient outcomes, simulations of the model suggest that increasing the number of possible drug choices can lower performance.

Bayesian Methods in Regulatory Science

Regulatory science comprises the tools, standards, and approaches that regulators use to assess safety, efficacy, quality, and performance of drugs and medical devices. A major focus of regulatory science is the design and analysis of clinical trials. Clinical trials are an essential part of clinical research programs that aim to improve therapies and reduce the burden of disease. These clinical experiments help us learn about what works clinically and what does not work. The results of clinical trials support therapeutic and policy decisions. When designing clinical trials, investigators make many decisions regarding various aspects of how they will carry out the study, such as the primary objective of the study, primary and secondary endpoints, methods of analysis, sample size, etc. This paper provides a brief review of the clinical development of new treatments and argues for the use of Bayesian methods and decision theory in clinical research.

Value of Information: Sensitivity Analysis and Research Design in Bayesian Evidence Synthesis

Suppose we have a Bayesian model that combines evidence from several different sources. We want to know which model parameters most affect the estimate or decision from the model, or which of the parameter uncertainties drive the decision uncertainty. Furthermore, we want to prioritize what further data should be collected. These questions can be addressed by Value of Information (VoI) analysis, in which we estimate expected reductions in loss from learning specific parameters or collecting data of a given design. We describe the theory and practice of VoI for Bayesian evidence synthesis, using and extending ideas from health economics, computer modeling and Bayesian design. The methods are general to a range of decision problems including point estimation and choices between discrete actions. We apply them to a model for estimating prevalence of HIV infection, combining indirect information from surveys, registers, and expert beliefs. This analysis shows which parameters contribute most of the uncertainty about each prevalence estimate, and the expected improvements in precision from specific amounts of additional data. These benefits can be traded with the costs of sampling to determine an optimal sample size. Supplementary materials for this article, including a standardized description of the materials available for reproducing the work, are available as an online supplement.

Risk-Taking Behavior in Attention Deficit/Hyperactivity Disorder (ADHD): a Review of Potential Underlying Mechanisms and of Interventions

PURPOSE OF REVIEW

Attention deficit/hyperactivity disorder (ADHD) is associated with several forms of risk-taking behavior (RTB). This paper aims to examine the scope of ADHD-related RTB, to highlight potential underlying mechanisms of this association, and to review initial evidence for interventions aimed to treat ADHD-related RTB.

RECENT FINDINGS

Multiple lines of evidence indicate that ADHD is associated with real-life RTB across several domains (e.g., reckless driving, substance use, and unprotected sex), which is corroborated by evidence on laboratory risk-taking tasks. Several individual differences, some of them informed by decision theory, e.g., comorbid disorders, parental monitoring, and perceived enlarged benefits of RTB, may explain the link between ADHD and RTB. A number of studies showed that interventions designed for ADHD may reduce RTB. ADHD is linked to RTB across several domains. Decision theory may serve as a conceptual framework for understanding the underlying mechanisms, and thus may inform future research.

Structured decision-making drives guidelines panels' recommendations "for" but not "against" health interventions

BACKGROUND AND OBJECTIVES

The determinants of guideline panels' recommendations remain uncertain. The objective of this study was to investigate factors considered by members of 8 panels convened by the American Society of Hematology (ASH) to develop guidelines using Grading of Recommendations Assessment, Development, and Evaluation (GRADE) system.

STUDY DESIGN AND SETTING

Web-based survey of the participants in the ASH guidelines panels.

ANALYSIS

two-level hierarchical, random-effect, multivariable regression analysis to explore the relation between GRADE and non-GRADE factors and strength of recommendations (SOR).

RESULTS

In the primary analysis, certainty in evidence [OR = 1.83; (95CI% 1.45-2.31)], balance of benefits and harms [OR = 1.49 (95CI% 1.30-1.69)] and variability in patients' values and preferences [OR = 1.47 (95CI% 1.15-1.88)] proved the strongest predictors of SOR. In a secondary analysis, certainty of evidence was associated with a strong recommendation [OR = 3.60 (95% CI 2.16-6.00)] when panel members recommended "for" interventions but not when they made recommendations "against" interventions [OR = 0.98 (95%CI: 0.57-1.8)] consistent with "yes" bias. Agreement between individual members and the group in rating SOR varied (kappa ranged from -0.01 to 0.64).

CONCLUSION

GRADE's conceptual framework proved, in general, to be highly associated with SOR. Failure of certainty of evidence to be associated with SOR against an intervention, suggest the need for improvements in the process.

Understanding health decision making: An exploration of homophily

The phenomenon of homophily first was described in Lazarsfeld and Merton's classic 1954 friendship analysis as a tendency for friendships to form between those who are alike in some respect. Although theories of decision making address a host of factors that affect the process, the influence of individuals with homophilic ties remains unaccounted for and unexplained. The purpose of this paper is to review theories relevant to decision making and describe what is known about the relationship between homophily and health care decision making. Further, we provide new evidence suggesting the influence of homophily on decision making in results from a randomized, multi-center clinical trial of American men with localized prostate cancer. A diverse sample of 293 men with a new diagnosis of localized prostate cancer reported relevant personal factors influencing the care management decision before randomization to a decision aid or usual care, between 2013 and 2015. Among these personal factors were the level of influence or importance ascribed to various individuals at the time of the treatment decision. One month later, participants reported how prepared they were for decision making. 123 men (42%) reported friends and/or coworkers as information sources, of which 65 (53%) indicated that friends and/or coworkers influenced the care decision. Men who reported friends/coworkers as information sources had significantly higher one-month preparation scores. Our review of decision making theories and practical applicability suggests the influence of homophilic relationships manifests in health care decision making. Faced with a list of options to manage health conditions, decision makers turn to known individuals in their environments, particularly those individuals with whom the decision maker can identify. Clinicians may solicit information from patients about influential others and explain how that support fits into the health decision at hand without dishonoring the importance of the homophilic relationship.

Evaluation of Two Approaches to Defining Extinction Risk under the U.S. Endangered Species Act

The predominant definition of extinction risk in conservation biology involves evaluating the cumulative distribution function (CDF) of extinction time at a particular point (the "time horizon"). Using the principles of decision theory, this article develops an alternative definition of extinction risk as the expected loss (EL) to society resulting from eventual extinction of a species. Distinct roles are identified for time preference and risk aversion. Ranges of tentative values for the parameters of the two approaches are proposed, and the performances of the two approaches are compared and contrasted for a small set of real-world species with published extinction time distributions and a large set of hypothetical extinction time distributions. Potential issues with each approach are evaluated, and the EL approach is recommended as the better of the two. The CDF approach suffers from the fact that extinctions that occur at any time before the specified time horizon are weighted equally, while extinctions that occur beyond the specified time horizon receive no weight at all. It also suffers from the fact that the time horizon does not correspond to any natural phenomenon, and so is impossible to specify nonarbitrarily; yet the results can depend critically on the specified value. In contrast, the EL approach has the advantage of weighting extinction time continuously, with no artificial time horizon, and the parameters of the approach (the rates of time preference and risk aversion) do correspond to natural phenomena, and so can be specified nonarbitrarily.

Use of multi-criteria decision analysis to identify potentially dangerous glacial lakes

Glacial Lake Outburst Floods (GLOFs) represent a significant threat in deglaciating environments, necessitating the development of GLOF hazard and risk assessment procedures. Here, we outline a Multi-Criteria Decision Analysis (MCDA) approach that can be used to rapidly identify potentially dangerous lakes in regions without existing tailored GLOF risk assessments, where a range of glacial lake types exist, and where field data are sparse or non-existent. Our MCDA model (1) is desk-based and uses freely and widely available data inputs and software, and (2) allows the relative risk posed by a range of glacial lake types to be assessed simultaneously within any region. A review of the factors that influence GLOF risk, combined with the strict rules of criteria selection inherent to MCDA, has allowed us to identify 13 exhaustive, non-redundant, and consistent risk criteria. We use our MCDA model to assess the risk of 16 extant glacial lakes and 6 lakes that have already generated GLOFs, and found that our results agree well with previous studies. For the first time in GLOF risk assessment, we employed sensitivity analyses to test the strength of our model results and assumptions, and to identify lakes that are sensitive to the criteria and risk thresholds used. A key benefit of the MCDA method is that sensitivity analyses are readily undertaken. Overall, these sensitivity analyses lend support to our model, although we suggest that further work is required to determine the relative importance of assessment criteria, and the thresholds that determine the level of risk for each criterion. As a case study, the tested method was then applied to 25 potentially dangerous lakes in the Bolivian Andes, where GLOF risk is poorly understood; 3 lakes are found to pose 'medium' or 'high' risk, and require further detailed investigation.

Value of information methods to design a clinical trial in a small population to optimise a health economic utility function

Background

Most confirmatory randomised controlled clinical trials (RCTs) are designed with specified power, usually 80% or 90%, for a hypothesis test conducted at a given significance level, usually 2.5% for a one-sided test. Approval of the experimental treatment by regulatory agencies is then based on the result of such a significance test with other information to balance the risk of adverse events against the benefit of the treatment to future patients. In the setting of a rare disease, recruiting sufficient patients to achieve conventional error rates for clinically reasonable effect sizes may be infeasible, suggesting that the decision-making process should reflect the size of the target population.

Methods

We considered the use of a decision-theoretic value of information (VOI) method to obtain the optimal sample size and significance level for confirmatory RCTs in a range of settings. We assume the decision maker represents society. For simplicity we assume the primary endpoint to be normally distributed with unknown mean following some normal prior distribution representing information on the anticipated effectiveness of the therapy available before the trial. The method is illustrated by an application in an RCT in haemophilia A. We explicitly specify the utility in terms of improvement in primary outcome and compare this with the costs of treating patients, both financial and in terms of potential harm, during the trial and in the future.

Results

The optimal sample size for the clinical trial decreases as the size of the population decreases. For non-zero cost of treating future patients, either monetary or in terms of potential harmful effects, stronger evidence is required for approval as the population size increases, though this is not the case if the costs of treating future patients are ignored.

Conclusions

Decision-theoretic VOI methods offer a flexible approach with both type I error rate and power (or equivalently trial sample size) depending on the size of the future population for whom the treatment under investigation is intended. This might be particularly suitable for small populations when there is considerable information about the patient population.

Electronic supplementary material

The online version of this article (10.1186/s12874-018-0475-0) contains supplementary material, which is available to authorized users.

Determination of the optimal sample size for a clinical trial accounting for the population size

The problem of choosing a sample size for a clinical trial is a very common one. In some settings, such as rare diseases or other small populations, the large sample sizes usually associated with the standard frequentist approach may be infeasible, suggesting that the sample size chosen should reflect the size of the population under consideration. Incorporation of the population size is possible in a decision-theoretic approach either explicitly by assuming that the population size is fixed and known, or implicitly through geometric discounting of the gain from future patients reflecting the expected population size. This paper develops such approaches. Building on previous work, an asymptotic expression is derived for the sample size for single and two-arm clinical trials in the general case of a clinical trial with a primary endpoint with a distribution of one parameter exponential family form that optimizes a utility function that quantifies the cost and gain per patient as a continuous function of this parameter. It is shown that as the size of the population, N, or expected size, N∗ in the case of geometric discounting, becomes large, the optimal trial size is O(N1/2) or O(N∗1/2). The sample size obtained from the asymptotic expression is also compared with the exact optimal sample size in examples with responses with Bernoulli and Poisson distributions, showing that the asymptotic approximations can also be reasonable in relatively small sample sizes.

Multiple Criteria Decision Analysis (MCDA) for evaluating new medicines in Health Technology Assessment and beyond: The Advance Value Framework

Escalating drug prices have catalysed the generation of numerous "value frameworks" with the aim of informing payers, clinicians and patients on the assessment and appraisal process of new medicines for the purpose of coverage and treatment selection decisions. Although this is an important step towards a more inclusive Value Based Assessment (VBA) approach, aspects of these frameworks are based on weak methodologies and could potentially result in misleading recommendations or decisions. In this paper, a Multiple Criteria Decision Analysis (MCDA) methodological process, based on Multi Attribute Value Theory (MAVT), is adopted for building a multi-criteria evaluation model. A five-stage model-building process is followed, using a top-down "value-focused thinking" approach, involving literature reviews and expert consultations. A generic value tree is structured capturing decision-makers' concerns for assessing the value of new medicines in the context of Health Technology Assessment (HTA) and in alignment with decision theory. The resulting value tree (Advance Value Tree) consists of three levels of criteria (top level criteria clusters, mid-level criteria, bottom level sub-criteria or attributes) relating to five key domains that can be explicitly measured and assessed: (a) burden of disease, (b) therapeutic impact, (c) safety profile (d) innovation level and (e) socioeconomic impact. A number of MAVT modelling techniques are introduced for operationalising (i.e. estimating) the model, for scoring the alternative treatment options, assigning relative weights of importance to the criteria, and combining scores and weights. Overall, the combination of these MCDA modelling techniques for the elicitation and construction of value preferences across the generic value tree provides a new value framework (Advance Value Framework) enabling the comprehensive measurement of value in a structured and transparent way. Given its flexibility to meet diverse requirements and become readily adaptable across different settings, the Advance Value Framework could be offered as a decision-support tool for evaluators and payers to aid coverage and reimbursement of new medicines.

Defender-Attacker Decision Tree Analysis to Combat Terrorism

We propose a methodology, called defender-attacker decision tree analysis, to evaluate defensive actions against terrorist attacks in a dynamic and hostile environment. Like most game-theoretic formulations of this problem, we assume that the defenders act rationally by maximizing their expected utility or minimizing their expected costs. However, we do not assume that attackers maximize their expected utilities. Instead, we encode the defender's limited knowledge about the attacker's motivations and capabilities as a conditional probability distribution over the attacker's decisions. We apply this methodology to the problem of defending against possible terrorist attacks on commercial airplanes, using one of three weapons: infrared-guided MANPADS (man-portable air defense systems), laser-guided MANPADS, or visually targeted RPGs (rocket propelled grenades). We also evaluate three countermeasures against these weapons: DIRCMs (directional infrared countermeasures), perimeter control around the airport, and hardening airplanes. The model includes deterrence effects, the effectiveness of the countermeasures, and the substitution of weapons and targets once a specific countermeasure is selected. It also includes a second stage of defensive decisions after an attack occurs. Key findings are: (1) due to the high cost of the countermeasures, not implementing countermeasures is the preferred defensive alternative for a large range of parameters; (2) if the probability of an attack and the associated consequences are large, a combination of DIRCMs and ground perimeter control are preferred over any single countermeasure.

A general framework for updating belief distributions

We propose a framework for general Bayesian inference. We argue that a valid update of a prior belief distribution to a posterior can be made for parameters which are connected to observations through a loss function rather than the traditional likelihood function, which is recovered as a special case. Modern application areas make it increasingly challenging for Bayesians to attempt to model the true data-generating mechanism. For instance, when the object of interest is low dimensional, such as a mean or median, it is cumbersome to have to achieve this via a complete model for the whole data distribution. More importantly, there are settings where the parameter of interest does not directly index a family of density functions and thus the Bayesian approach to learning about such parameters is currently regarded as problematic. Our framework uses loss functions to connect information in the data to functionals of interest. The updating of beliefs then follows from a decision theoretic approach involving cumulative loss functions. Importantly, the procedure coincides with Bayesian updating when a true likelihood is known yet provides coherent subjective inference in much more general settings. Connections to other inference frameworks are highlighted.

Gain-of-Function Research: Ethical Analysis

Gain-of-function (GOF) research involves experimentation that aims or is expected to (and/or, perhaps, actually does) increase the transmissibility and/or virulence of pathogens. Such research, when conducted by responsible scientists, usually aims to improve understanding of disease causing agents, their interaction with human hosts, and/or their potential to cause pandemics. The ultimate objective of such research is to better inform public health and preparedness efforts and/or development of medical countermeasures. Despite these important potential benefits, GOF research (GOFR) can pose risks regarding biosecurity and biosafety. In 2014 the administration of US President Barack Obama called for a "pause" on funding (and relevant research with existing US Government funding) of GOF experiments involving influenza, SARS, and MERS viruses in particular. With announcement of this pause, the US Government launched a "deliberative process" regarding risks and benefits of GOFR to inform future funding decisions-and the US National Science Advisory Board for Biosecurity (NSABB) was tasked with making recommendations to the US Government on this matter. As part of this deliberative process the National Institutes of Health commissioned this Ethical Analysis White Paper, requesting that it provide (1) review and summary of ethical literature on GOFR, (2) identification and analysis of existing ethical and decision-making frameworks relevant to (i) the evaluation of risks and benefits of GOFR, (ii) decision-making about the conduct of GOF studies, and (iii) the development of US policy regarding GOFR (especially with respect to funding of GOFR), and (3) development of an ethical and decision-making framework that may be considered by NSABB when analyzing information provided by GOFR risk-benefit assessment, and when crafting its final recommendations (especially regarding policy decisions about funding of GOFR in particular). The ethical and decision-making framework ultimately developed is based on the idea that there are numerous ethically relevant dimensions upon which any given case of GOFR can fare better or worse (as opposed to there being necessary conditions that are either satisfied or not satisfied, where all must be satisfied in order for a given case of GOFR to be considered ethically acceptable): research imperative, proportionality, minimization of risks, manageability of risks, justice, good governance (i.e., democracy), evidence, and international outlook and engagement. Rather than drawing a sharp bright line between GOFR studies that are ethically acceptable and those that are ethically unacceptable, this framework is designed to indicate where any given study would fall on an ethical spectrum-where imaginable cases of GOFR might range from those that are most ethically acceptable (perhaps even ethically praiseworthy or ethically obligatory), at one end of the spectrum, to those that are most ethically problematic or unacceptable (and thus should not be funded, or conducted), at the other. The aim should be that any GOFR pursued (and/or funded) should be as far as possible towards the former end of the spectrum.

Low-template DNA: A single DNA analysis or two replicates?

This study investigates the following two questions: (1) Should the DNA analyst concentrate the DNA extract into a single amplification or should he/she split it up to do two replicates? (2) Given the electropherogram obtained from a first analysis, is it worthwhile for the DNA analyst to invest in obtaining a second replicate? A decision-theoretic approach addresses these questions by quantitatively expressing the expected net gain (ENG) of each DNA analysis of interest. The results indicate that two replicates generally have a greater ENG than a single DNA analysis for DNA quantities capable of producing two replicates having an average allelic peak height as low as 43rfu. This supports the position that two replicates increase the information content with regard to a single analysis.

A decision-theoretic phase I-II design for ordinal outcomes in two cycles

This paper is motivated by a phase I-II clinical trial of a targeted agent for advanced solid tumors. We study a stylized version of this trial with the goal to determine optimal actions in each of two cycles of therapy. A design is presented that generalizes the decision-theoretic two-cycle design of Lee and others (2015. Bayesian dose-finding in two treatment cycles based on the joint utility of efficacy and toxicity. Journal of the American Statistical Association, to appear) to accommodate ordinal outcomes. Backward induction is used to jointly optimize the actions taken for each patient in each of the two cycles, with the second action accounting for the patient's cycle 1 dose and outcomes. A simulation study shows that simpler designs obtained by dichotomizing the ordinal outcomes either perform very similarly to the proposed design, or have much worse performance in some scenarios. We also compare the proposed design with the simpler approaches of optimizing the doses in each cycle separately, or ignoring the distinction between cycles 1 and 2.

On the Bayesness, minimaxity and admissibility of point estimators of allelic frequencies

In this paper, decision theory was used to derive Bayes and minimax decision rules to estimate allelic frequencies and to explore their admissibility. Decision rules with uniformly smallest risk usually do not exist and one approach to solve this problem is to use the Bayes principle and the minimax principle to find decision rules satisfying some general optimality criterion based on their risk functions. Two cases were considered, the simpler case of biallelic loci and the more complex case of multiallelic loci. For each locus, the sampling model was a multinomial distribution and the prior was a Beta (biallelic case) or a Dirichlet (multiallelic case) distribution. Three loss functions were considered: squared error loss (SEL), Kulback-Leibler loss (KLL) and quadratic error loss (QEL). Bayes estimators were derived under these three loss functions and were subsequently used to find minimax estimators using results from decision theory. The Bayes estimators obtained from SEL and KLL turned out to be the same. Under certain conditions, the Bayes estimator derived from QEL led to an admissible minimax estimator (which was also equal to the maximum likelihood estimator). The SEL also allowed finding admissible minimax estimators. Some estimators had uniformly smaller variance than the MLE and under suitable conditions the remaining estimators also satisfied this property. In addition to their statistical properties, the estimators derived here allow variation in allelic frequencies, which is closer to the reality of finite populations exposed to evolutionary forces.

Bayesian designs and the control of frequentist characteristics: a practical solution

Frequentist concepts, such as the control of the type I error or the false discovery rate, are well established in the medical literature and often required by regulators. Most Bayesian designs are defined without explicit considerations of frequentist characteristics. Once the Bayesian design is structured, statisticians use simulations and adjust tuning parameters to comply with a set of targeted operating characteristics. These adjustments affect the use of prior information and utility functions. Here we consider a Bayesian decision theoretic approach for experimental designs with explicit frequentist requisites. We define optimal designs under a set of constraints required by a regulator. Our approach combines the use of interpretable utility functions with frequentist criteria, and selects an optimal design that satisfies a set of required operating characteristics. We illustrate the approach using a group-sequential multi-arm Phase II trial and a bridging trial.

Biodiversity, conservation biology, and rational choice

This paper critically discusses two areas of Sahotra Sarkar's recent work in environmental philosophy: biodiversity and conservation biology and roles for decision theory in incorporating values explicitly in the environmental policy process. I argue that Sarkar's emphasis on the practices of conservation biologists, and especially the role of social and cultural values in the choice of biodiversity constituents, restricts his conception of biodiversity to particular practical conservation contexts. I argue that life scientists have many reasons to measure many types of diversity, and that biodiversity metrics could be value-free. I argue that Sarkar's emphasis on the limitations of normative decision theory is in tension with his statement that decision theory can "put science and ethics together." I also challenge his claim that multi-criteria decision tools lacking axiomatic foundations in preference and utility theory are "without a rational basis," by presenting a case of a simple "outranking" multi-criteria decision rule that can violate a basic normative requirement of preferences (transitivity) and ask whether there may nevertheless be contexts in which such a procedure might assist decision makers.

Environmental philosophy: response to critics

The following piece is a response to the critiques from Frank, Garson, and Odenbaugh. The issues at stake are: the definition of biodiversity and its normativity, historical fidelity in ecological restoration, naturalism in environmental ethics, and the role of decision theory. The normativity of the concept of biodiversity in conservation biology is defended. Historical fidelity is criticized as an operative goal for ecological restoration. It is pointed out that the analysis requires only minimal assumptions about ethics. Decision theory is presented as a tool, not a domain-limiting necessary requirement for environmental philosophy.

A decision-theoretical alternative to testing many hypotheses

Testing a large number of hypotheses is a key problem in the analysis of microarray experiments and in other studies in which many elementary experiments are conducted, and the exceptions among them, for which a particular hypothesis does not hold, have to be identified. A class of approaches to this problem is based on controlling the false discovery rate, even though failure to discover should also be considered. We develop a decision-theoretical approach in which errors (misclassifications) of the two kinds are associated with uneven losses, and the total expected loss in the collection of the classifications (decisions made or options selected) is minimized.

The effective distribution system for the concentration of patients to extra-large hospitals

PURPOSE

In Korean society, extra-large hospitals are congested with the majority of patients. Because of the congestions, the urgent patients need to wait anywhere from as short as a month to around three months. These concentrations of the patients on the extra-large hospitals causes not only the economic problem in terms of loss of opportunity cost and resources of other medium and large hospitals but also the fear and the consequential stress of the patients and the families of the patients who are waiting for the surgeries. The phenomenon of the concentrations derived due to the insufficient information to the medical consumers. If the information on medical treatment services such as surgery schedule is provided before the selection of hospital, we expect that the selection of hospital for the patients and their family will differ, resulting in redistribution of concentration phenomenon. In this paper, we propose and verify the effective distribution system for the concentration on the extra-large hospitals.

METHODS

Web simulation survey was conducted. A total 100 respondents were divided into 4 groups of 25 respondents and the different information was provided to each group.

RESULTS

Through multiple comparisons among groups, only group which was provided with both information about 'the difference of surgical results' and 'the waiting time for surgery', had difference in significance.

CONCLUSION

By providing two sets of information to patients, reckless selection of extra-large hospitals can be spread to more appropriate hospitals and therefore achieve effective distribution of the population concentration on extra-large hospital.