A Video Analysis of Intra- and Inter-professional Leadership Behaviours Within “The Burns Suite”: Identifying Key Leadership Models

Objective Leadership is particularly important in complex highly interprofessional health care contexts involving a number of staff, some from the same specialty (intraprofessional), and others from different specialties (interprofessional). The authors recently published the concept of “The Burns Suite” (TBS) as a novel simulation tool to deliver interprofessional and teamwork training. It is unclear which leadership behaviors are the most important in an interprofessional burns resuscitation scenario, and whether they can be modeled on to current leadership theory. The purpose of this study was to perform a comprehensive video analysis of leadership behaviors within TBS. Methods A total of 3 burns resuscitation simulations within TBS were recorded. The video analysis was grounded-theory inspired. Using predefined criteria, actions/interactions deemed as leadership behaviors were identified. Using an inductive iterative process, 8 main leadership behaviors were identified. Cohen’s κ coefficient was used to measure inter-rater agreement and calculated as κ = 0.7 (substantial agreement). Each video was watched 4 times, focusing on 1 of the 4 team members per viewing (senior surgeon, senior nurse, trainee surgeon, and trainee nurse). The frequency and types of leadership behavior of each of the 4 team members were recorded. Statistical significance to assess any differences was assessed using analysis of variance, whereby a p < 0.05 was taken to be significant. Leadership behaviors were triangulated with verbal cues and actions from the videos. Results All 3 scenarios were successfully completed. The mean scenario length was 22 minutes. A total of 362 leadership behaviors were recorded from the 12 participants. The most evident leadership behaviors of all team members were adhering to guidelines (which effectively equates to following Advanced Trauma and Life Support/Emergency Management of Severe Burns resuscitation guidelines and hence “maintaining standards”), followed by making decisions. Although in terms of total frequency the senior surgeon engaged in more leadership behaviors compared with the entire team, statistically there was no significant difference between all 4 members within the 8 leadership categories. This analysis highlights that “distributed leadership” was predominant, whereby leadership was “distributed” or “shared” among team members. The leadership behaviors within TBS also seemed to fall in line with the “direction, alignment, and commitment” ontology. Conclusions Effective leadership is essential for successful functioning of work teams and accomplishment of task goals. As the resuscitation of a patient with major burns is a dynamic event, team leaders require flexibility in their leadership behaviors to effectively adapt to changing situations. Understanding leadership behaviors of different team members within an authentic simulation can identify important behaviors required to optimize nontechnical skills in a major resuscitation. Furthermore, attempting to map these behaviors on to leadership models can help further our understanding of leadership theory. Collectively this can aid the development of refined simulation scenarios for team members, and can be extrapolated into other areas of simulation-based team training and interprofessional education.

Evaluating statistical and machine learning methods to predict risk of in-hospital child mortality in Uganda

Thesis (Master's)--University of Washington, 2016-08

Modelling and estimating models of physician labour supply and productivity

Ma thèse s’intéresse aux politiques de santé conçues pour encourager l’offre de services de santé. L’accessibilité aux services de santé est un problème majeur qui mine le système de santé de la plupart des pays industrialisés. Au Québec, le temps médian d’attente entre une recommandation du médecin généraliste et un rendez-vous avec un médecin spécialiste était de 7,3 semaines en 2012, contre 2,9 semaines en 1993, et ceci malgré l’augmentation du nombre de médecins sur cette même période. Pour les décideurs politiques observant l’augmentation du temps d’attente pour des soins de santé, il est important de comprendre la structure de l’offre de travail des médecins et comment celle-ci affecte l’offre des services de santé. Dans ce contexte, je considère deux principales politiques. En premier lieu, j’estime comment les médecins réagissent aux incitatifs monétaires et j’utilise les paramètres estimés pour examiner comment les politiques de compensation peuvent être utilisées pour déterminer l’offre de services de santé de court terme. En second lieu, j’examine comment la productivité des médecins est affectée par leur expérience, à travers le mécanisme du "learning-by-doing", et j’utilise les paramètres estimés pour trouver le nombre de médecins inexpérimentés que l’on doit recruter pour remplacer un médecin expérimenté qui va à la retraite afin de garder l’offre des services de santé constant. Ma thèse développe et applique des méthodes économique et statistique afin de mesurer la réaction des médecins face aux incitatifs monétaires et estimer leur profil de productivité (en mesurant la variation de la productivité des médecins tout le long de leur carrière) en utilisant à la fois des données de panel sur les médecins québécois, provenant d’enquêtes et de l’administration. Les données contiennent des informations sur l’offre de travail de chaque médecin, les différents types de services offerts ainsi que leurs prix. Ces données couvrent une période pendant laquelle le gouvernement du Québec a changé les prix relatifs des services de santé. J’ai utilisé une approche basée sur la modélisation pour développer et estimer un modèle structurel d’offre de travail en permettant au médecin d’être multitâche. Dans mon modèle les médecins choisissent le nombre d’heures travaillées ainsi que l’allocation de ces heures à travers les différents services offerts, de plus les prix des services leurs sont imposés par le gouvernement. Le modèle génère une équation de revenu qui dépend des heures travaillées et d’un indice de prix représentant le rendement marginal des heures travaillées lorsque celles-ci sont allouées de façon optimale à travers les différents services. L’indice de prix dépend des prix des services offerts et des paramètres de la technologie de production des services qui déterminent comment les médecins réagissent aux changements des prix relatifs. J’ai appliqué le modèle aux données de panel sur la rémunération des médecins au Québec fusionnées à celles sur l’utilisation du temps de ces mêmes médecins. J’utilise le modèle pour examiner deux dimensions de l’offre des services de santé. En premierlieu, j’analyse l’utilisation des incitatifs monétaires pour amener les médecins à modifier leur production des différents services. Bien que les études antérieures ont souvent cherché à comparer le comportement des médecins à travers les différents systèmes de compensation,il y a relativement peu d’informations sur comment les médecins réagissent aux changementsdes prix des services de santé. Des débats actuels dans les milieux de politiques de santé au Canada se sont intéressés à l’importance des effets de revenu dans la détermination de la réponse des médecins face à l’augmentation des prix des services de santé. Mon travail contribue à alimenter ce débat en identifiant et en estimant les effets de substitution et de revenu résultant des changements des prix relatifs des services de santé. En second lieu, j’analyse comment l’expérience affecte la productivité des médecins. Cela a une importante implication sur le recrutement des médecins afin de satisfaire la demande croissante due à une population vieillissante, en particulier lorsque les médecins les plus expérimentés (les plus productifs) vont à la retraite. Dans le premier essai, j’ai estimé la fonction de revenu conditionnellement aux heures travaillées, en utilisant la méthode des variables instrumentales afin de contrôler pour une éventuelle endogeneité des heures travaillées. Comme instruments j’ai utilisé les variables indicatrices des âges des médecins, le taux marginal de taxation, le rendement sur le marché boursier, le carré et le cube de ce rendement. Je montre que cela donne la borne inférieure de l’élasticité-prix direct, permettant ainsi de tester si les médecins réagissent aux incitatifs monétaires. Les résultats montrent que les bornes inférieures des élasticités-prix de l’offre de services sont significativement positives, suggérant que les médecins répondent aux incitatifs. Un changement des prix relatifs conduit les médecins à allouer plus d’heures de travail au service dont le prix a augmenté. Dans le deuxième essai, j’estime le modèle en entier, de façon inconditionnelle aux heures travaillées, en analysant les variations des heures travaillées par les médecins, le volume des services offerts et le revenu des médecins. Pour ce faire, j’ai utilisé l’estimateur de la méthode des moments simulés. Les résultats montrent que les élasticités-prix direct de substitution sont élevées et significativement positives, représentant une tendance des médecins à accroitre le volume du service dont le prix a connu la plus forte augmentation. Les élasticitésprix croisées de substitution sont également élevées mais négatives. Par ailleurs, il existe un effet de revenu associé à l’augmentation des tarifs. J’ai utilisé les paramètres estimés du modèle structurel pour simuler une hausse générale de prix des services de 32%. Les résultats montrent que les médecins devraient réduire le nombre total d’heures travaillées (élasticité moyenne de -0,02) ainsi que les heures cliniques travaillées (élasticité moyenne de -0.07). Ils devraient aussi réduire le volume de services offerts (élasticité moyenne de -0.05). Troisièmement, j’ai exploité le lien naturel existant entre le revenu d’un médecin payé à l’acte et sa productivité afin d’établir le profil de productivité des médecins. Pour ce faire, j’ai modifié la spécification du modèle pour prendre en compte la relation entre la productivité d’un médecin et son expérience. J’estime l’équation de revenu en utilisant des données de panel asymétrique et en corrigeant le caractère non-aléatoire des observations manquantes à l’aide d’un modèle de sélection. Les résultats suggèrent que le profil de productivité est une fonction croissante et concave de l’expérience. Par ailleurs, ce profil est robuste à l’utilisation de l’expérience effective (la quantité de service produit) comme variable de contrôle et aussi à la suppression d’hypothèse paramétrique. De plus, si l’expérience du médecin augmente d’une année, il augmente la production de services de 1003 dollar CAN. J’ai utilisé les paramètres estimés du modèle pour calculer le ratio de remplacement : le nombre de médecins inexpérimentés qu’il faut pour remplacer un médecin expérimenté. Ce ratio de remplacement est de 1,2.

Essays on Machine Learning and Hedonic Models

Thesis (Ph.D.)--University of Washington, 2016-08

Probabilistic Models for Scalable Knowledge Graph Construction

In the past decade, systems that extract information from millions of Internet documents have become commonplace. Knowledge graphs -- structured knowledge bases that describe entities, their attributes and the relationships between them -- are a powerful tool for understanding and organizing this vast amount of information. However, a significant obstacle to knowledge graph construction is the unreliability of the extracted information, due to noise and ambiguity in the underlying data or errors made by the extraction system and the complexity of reasoning about the dependencies between these noisy extractions. My dissertation addresses these challenges by exploiting the interdependencies between facts to improve the quality of the knowledge graph in a scalable framework. I introduce a new approach called knowledge graph identification (KGI), which resolves the entities, attributes and relationships in the knowledge graph by incorporating uncertain extractions from multiple sources, entity co-references, and ontological constraints. I define a probability distribution over possible knowledge graphs and infer the most probable knowledge graph using a combination of probabilistic and logical reasoning. Such probabilistic models are frequently dismissed due to scalability concerns, but my implementation of KGI maintains tractable performance on large problems through the use of hinge-loss Markov random fields, which have a convex inference objective. This allows the inference of large knowledge graphs using 4M facts and 20M ground constraints in 2 hours. To further scale the solution, I develop a distributed approach to the KGI problem which runs in parallel across multiple machines, reducing inference time by 90%. Finally, I extend my model to the streaming setting, where a knowledge graph is continuously updated by incorporating newly extracted facts. I devise a general approach for approximately updating inference in convex probabilistic models, and quantify the approximation error by defining and bounding inference regret for online models. Together, my work retains the attractive features of probabilistic models while providing the scalability necessary for large-scale knowledge graph construction. These models have been applied on a number of real-world knowledge graph projects, including the NELL project at Carnegie Mellon and the Google Knowledge Graph.

The renormalization group via statistical inference

In physics, one attempts to infer the rules governing a system given only the results of imperfect measurements. Hence, microscopic theories may be effectively indistinguishable experimentally. We develop an operationally motivated procedure to identify the corresponding equivalence classes of states, and argue that the renormalization group (RG) arises from the inherent ambiguities associated with the classes: one encounters flow parameters as, e.g., a regulator, a scale, or a measure of precision, which specify representatives in a given equivalence class. This provides a unifying framework and reveals the role played by information in renormalization. We validate this idea by showing that it justifies the use of low-momenta n-point functions as statistically relevant observables around a Gaussian hypothesis. These results enable the calculation of distinguishability in quantum field theory. Our methods also provide a way to extend renormalization techniques to effective models which are not based on the usual quantum-field formalism, and elucidates the relationships between various type of RG.

Mixture of time-dependent growth models with an application to blue swimmer crab length-frequency data

Understanding how aquatic species grow is fundamental in fisheries because stock assessment often relies on growth dependent statistical models. Length-frequency-based methods become important when more applicable data for growth model estimation are either not available or very expensive. In this article, we develop a new framework for growth estimation from length-frequency data using a generalized von Bertalanffy growth model (VBGM) framework that allows for time-dependent covariates to be incorporated. A finite mixture of normal distributions is used to model the length-frequency cohorts of each month with the means constrained to follow a VBGM. The variances of the finite mixture components are constrained to be a function of mean length, reducing the number of parameters and allowing for an estimate of the variance at any length. To optimize the likelihood, we use a minorization–maximization (MM) algorithm with a Nelder–Mead sub-step. This work was motivated by the decline in catches of the blue swimmer crab (BSC) (Portunus armatus) off the east coast of Queensland, Australia. We test the method with a simulation study and then apply it to the BSC fishery data.

Statistical modeling of protein lysate array data

The protein lysate array is an emerging technology for quantifying the protein concentration ratios in multiple biological samples. It is gaining popularity, and has the potential to answer questions about post-translational modifications and protein pathway relationships. Statistical inference for a parametric quantification procedure has been inadequately addressed in the literature, mainly due to two challenges: the increasing dimension of the parameter space and the need to account for dependence in the data. Each chapter of this thesis addresses one of these issues. In Chapter 1, an introduction to the protein lysate array quantification is presented, followed by the motivations and goals for this thesis work. In Chapter 2, we develop a multi-step procedure for the Sigmoidal models, ensuring consistent estimation of the concentration level with full asymptotic efficiency. The results obtained in this chapter justify inferential procedures based on large-sample approximations. Simulation studies and real data analysis are used to illustrate the performance of the proposed method in finite-samples. The multi-step procedure is simpler in both theory and computation than the single-step least squares method that has been used in current practice. In Chapter 3, we introduce a new model to account for the dependence structure of the errors by a nonlinear mixed effects model. We consider a method to approximate the maximum likelihood estimator of all the parameters. Using the simulation studies on various error structures, we show that for data with non-i.i.d. errors the proposed method leads to more accurate estimates and better confidence intervals than the existing single-step least squares method.

Development of error correction techniques for nitrate-N load estimation models

Excess nutrient loads carried by streams and rivers are a great concern for environmental resource managers. In agricultural regions, excess loads are transported downstream to receiving water bodies, potentially causing algal blooms, which could lead to numerous ecological problems. To better understand nutrient load transport, and to develop appropriate water management plans, it is important to have accurate estimates of annual nutrient loads. This study used a Monte Carlo sub-sampling method and error-corrected statistical models to estimate annual nitrate-N loads from two watersheds in central Illinois. The performance of three load estimation methods (the seven-parameter log-linear model, the ratio estimator, and the flow-weighted averaging estimator) applied at one-, two-, four-, six-, and eight-week sampling frequencies were compared. Five error correction techniques; the existing composite method, and four new error correction techniques developed in this study; were applied to each combination of sampling frequency and load estimation method. On average, the most accurate error reduction technique, (proportional rectangular) resulted in 15% and 30% more accurate load estimates when compared to the most accurate uncorrected load estimation method (ratio estimator) for the two watersheds. Using error correction methods, it is possible to design more cost-effective monitoring plans by achieving the same load estimation accuracy with fewer observations. Finally, the optimum combinations of monitoring threshold and sampling frequency that minimizes the number of samples required to achieve specified levels of accuracy in load estimation were determined. For one- to three-weeks sampling frequencies, combined threshold/fixed-interval monitoring approaches produced the best outcomes, while fixed-interval-only approaches produced the most accurate results for four- to eight-weeks sampling frequencies.

THE PHYSICS OF IDEAS: INFERRING THE MECHANICS OF OPINION FORMATION FROM MACROSCOPIC STATISTICAL PATTERNS

In a microscopic setting, humans behave in rich and unexpected ways. In a macroscopic setting, however, distinctive patterns of group behavior emerge, leading statistical physicists to search for an underlying mechanism. The aim of this dissertation is to analyze the macroscopic patterns of competing ideas in order to discern the mechanics of how group opinions form at the microscopic level. First, we explore the competition of answers in online Q&A (question and answer) boards. We find that a simple individual-level model can capture important features of user behavior, especially as the number of answers to a question grows. Our model further suggests that the wisdom of crowds may be constrained by information overload, in which users are unable to thoroughly evaluate each answer and therefore tend to use heuristics to pick what they believe is the best answer. Next, we explore models of opinion spread among voters to explain observed universal statistical patterns such as rescaled vote distributions and logarithmic vote correlations. We introduce a simple model that can explain both properties, as well as why it takes so long for large groups to reach consensus. An important feature of the model that facilitates agreement with data is that individuals become more stubborn (unwilling to change their opinion) over time. Finally, we explore potential underlying mechanisms for opinion formation in juries, by comparing data to various types of models. We find that different null hypotheses in which jurors do not interact when reaching a decision are in strong disagreement with data compared to a simple interaction model. These findings provide conceptual and mechanistic support for previous work that has found mutual influence can play a large role in group decisions. In addition, by matching our models to data, we are able to infer the time scales over which individuals change their opinions for different jury contexts. We find that these values increase as a function of the trial time, suggesting that jurors and judicial panels exhibit a kind of stubbornness similar to what we include in our model of voting behavior.

Analyzing change in medication use - statistical approaches

The objective of this study was to gain an understanding of the effects of population heterogeneity, missing data, and causal relationships on parameter estimates from statistical models when analyzing change in medication use. From a public health perspective, two timely topics were addressed: the use and effects of statins in populations in primary prevention of cardiovascular disease and polypharmacy in older population. Growth mixture models were applied to characterize the accumulation of cardiovascular and diabetes medications among apparently healthy population of statin initiators. The causal effect of statin adherence on the incidence of acute cardiovascular events was estimated using marginal structural models in comparison with discrete-time hazards models. The impact of missing data on the growth estimates of evolution of polypharmacy was examined comparing statistical models under different assumptions for missing data mechanism. The data came from Finnish administrative registers and from the population-based Geriatric Multidisciplinary Strategy for the Good Care of the Elderly study conducted in Kuopio, Finland, during 2004–07. Five distinct patterns of accumulating medications emerged among the population of apparently healthy statin initiators during two years after statin initiation. Proper accounting for time-varying dependencies between adherence to statins and confounders using marginal structural models produced comparable estimation results with those from a discrete-time hazards model. Missing data mechanism was shown to be a key component when estimating the evolution of polypharmacy among older persons. In conclusion, population heterogeneity, missing data and causal relationships are important aspects in longitudinal studies that associate with the study question and should be critically assessed when performing statistical analyses. Analyses should be supplemented with sensitivity analyses towards model assumptions.

Hand detection on images based on deformable part models and additional features

Hand detection on images has important applications on person activities recognition. This thesis focuses on PASCAL Visual Object Classes (VOC) system for hand detection. VOC has become a popular system for object detection, based on twenty common objects, and has been released with a successful deformable parts model in VOC2007. A hand detection on an image is made when the system gets a bounding box which overlaps with at least 50% of any ground truth bounding box for a hand on the image. The initial average precision of this detector is around 0.215 compared with a state-of-art of 0.104; however, color and frequency features for detected bounding boxes contain important information for re-scoring, and the average precision can be improved to 0.218 with these features. Results show that these features help on getting higher precision for low recall, even though the average precision is similar.

Statistical forecasting of the Australian macadamia crop

Three types of forecasts of the total Australian production of macadamia nuts (t nut-in-shell) have been produced early each year since 2001. The first is a long-term forecast, based on the expected production from the tree census data held by the Australian Macadamia Society, suitably scaled up for missing data and assumed new plantings each year. These long-term forecasts range out to 10 years in the future, and form a basis for industry and market planning. Secondly, a statistical adjustment (termed the climate-adjusted forecast) is made annually for the coming crop. As the name suggests, climatic influences are the dominant factors in this adjustment process, however, other terms such as bienniality of bearing, prices and orchard aging are also incorporated. Thirdly, industry personnel are surveyed early each year, with their estimates integrated into a growers and pest-scouts forecast. Initially conducted on a 'whole-country' basis, these models are now constructed separately for the six main production regions of Australia, with these being combined for national totals. Ensembles or suites of step-forward regression models using biologically-relevant variables have been the major statistical method adopted, however, developing methodologies such as nearest-neighbour techniques, general additive models and random forests are continually being evaluated in parallel. The overall error rates average 14% for the climate forecasts, and 12% for the growers' forecasts. These compare with 7.8% for USDA almond forecasts (based on extensive early-crop sampling) and 6.8% for coconut forecasts in Sri Lanka. However, our somewhatdisappointing results were mainly due to a series of poor crops attributed to human reasons, which have now been factored into the models. Notably, the 2012 and 2013 forecasts averaged 7.8 and 4.9% errors, respectively. Future models should also show continuing improvement, as more data-years become available.

Stochastic simulation of time-series models combined with geostatistics to predict water-table scenarios in a Guarani Aquifer System outcrop area, Brazil

Stochastic methods based on time-series modeling combined with geostatistics can be useful tools to describe the variability of water-table levels in time and space and to account for uncertainty. Monitoring water-level networks can give information about the dynamic of the aquifer domain in both dimensions. Time-series modeling is an elegant way to treat monitoring data without the complexity of physical mechanistic models. Time-series model predictions can be interpolated spatially, with the spatial differences in water-table dynamics determined by the spatial variation in the system properties and the temporal variation driven by the dynamics of the inputs into the system. An integration of stochastic methods is presented, based on time-series modeling and geostatistics as a framework to predict water levels for decision making in groundwater management and land-use planning. The methodology is applied in a case study in a Guarani Aquifer System (GAS) outcrop area located in the southeastern part of Brazil. Communication of results in a clear and understandable form, via simulated scenarios, is discussed as an alternative, when translating scientific knowledge into applications of stochastic hydrogeology in large aquifers with limited monitoring network coverage like the GAS.

Local influence analysis in general spatial models