Online Program
Wed, Oct 7 |
||
Conference Registration |
Wed, Oct 7, 7:30 AM - 6:30 PM
L'Apogee 17 |
|
|
||
Workshop 1: Bias Adjustment in Comparative Real World Data Research |
Wed, Oct 7, 8:00 AM - 10:00 AM
Grand Ballroom |
|
Instructor(s): Douglas E Faries, Eli Lilly & Company; Guido W Imbens, Stanford University, Graduate School of Business | ||
Real world data is a growing source of information for medical and policy decision making. However, researchers tasked with performing and interpreting comparative analyses based on real world data face many analytical challenges, including selection bias adjustment, potential coding / measurement bias, switching between medications, and unmeasured confounding.
Existing guidance documents (ISPOR, GRACE, PCORI) provide excellent high level direction, but fail to provide sufficient detail for quality implementation of comparative real world analysis. In addition, recent years have seen an explosion of new approaches offering potential improvements in bias control and capabilities for addressing multiple cohorts and longitudinal analyses involving treatment switching. This workshop will address some of these challenges by starting with the basics of bias adjusted analyses for comparative real world analyses, including propensity score matching and subclassification, then build to cover newer advances in bias control (e.g. multi-cohort propensity scoring, entropy balancing) as well as longitudinal bias control when medications change over time. Download Handouts |
||
Workshop 2: Data Confidentiality – Past, Present and Future |
Wed, Oct 7, 8:00 AM - 10:00 AM
Renaissance Salon |
|
Instructor(s): Ofer Harel, University of Connecticut | ||
Data is essential for most types of research. In particular, economic, medical, educational, health policy and health services research requires increasing amounts of data. At the same time, there is an increasing concern regarding the confidentiality of the subjects whom the data is collected for. Often times it is unethical to release private information to the public, and for certain medical and educational data, it is illegal. Somehow, a balance must be struck between the release of data for research purposes and the risk of disclosing private information. In this tutorial we will introduce this problem and some past and current solutions, and emphasize the future direction of this important topic. Download Handouts |
||
CANCELLED Workshop 3: Methods Incubator for Junior Researchers |
Wed, Oct 7, 8:00 AM - 12:15 PM
|
|
Instructor(s): Laura Hatfield, Harvard Medical School; Sherri Rose, Harvard Medical School | ||
Conferences present a golden opportunity to meet with our colleagues across the world, but, too often, we talk at people rather than working with them. Thus, the Methods Incubator concept! This new type of interactive half-day workshop will target independent, early-career researchers (post-PhD) looking to form new and productive collaborations within the field of health policy statistics.
Anyone interested in attending the workshop will submit a short (1-2 paragraphs) description of an original approach to the problem for a particular team and submit this along with a CV to the leader of the team they wish to join by June 30. Team leaders will review and invite 6-8 team members to register for the workshop by early August. These participants will develop a 1-page "pitch" that describes the approach further, preferably with some preliminary data analysis. At least 1 week prior to the meeting, they will send these to the team leader, who will circulate to the whole team.
Format: All participants should bring laptops. On the day of the workshop, each team member will briefly pitch their approach to their team. Teams will divide into 2-3 project groups based on the most promising ideas and spend 2-3 hours working to develop those ideas. Project groups will develop concepts, outline papers, and possibly work on the data. At the end of the workshop, each project group will present their ideas briefly to the whole group for feedback and critique. We hope that these project groups will continue to collaborate after the workshop as co-authors on manuscripts.
Application Instructions: http://ow.ly/LxCQF |
||
Workshop 4: Predictive Modeling on Data with Severe Class Imbalance: Applications on Electronic Health Records |
Wed, Oct 7, 10:15 AM - 12:15 PM
Grand Ballroom |
|
Instructor(s): Birol Emir, Pfizer Inc and Columbia University; Max Kuhn, Pfizer Inc | ||
Healthcare records are used more and more often for making health care decisions and policies. Particularly, Electronic Health Care (EHR) data are collected by either specialized private companies such as Humedica (US) and Cegedim THIN (UK) or publicly available such as Behavioral Risk Factor Surveillance System (BRFSS), and Health and Retirement Survey (HRS). EHR data are useful in understanding insights in patient management. As data has become more readily available, companies and institutions desire to harness this information for predictive purposes. Prediction of undiagnosed fibromyalgia (FM) patients, for example, seeks to uncover relationship between predictors such as demographics, healthcare resources and FM. In many cases, the event of interest is observed with relatively small frequencies, leading to a class imbalance that can confound modelers.
This workshop discussed ways to mitigate the effects of severe class imbalances. The course outline is:
• Description of the problem with class imbalances (with illustrative data)
• A short refresher on predictive models, parameter tuning and resampling
• A description of tree-based classification models (single models and ensembles)
• Sampling methods for combating class imbalances
• Cost-Sensitive learning methods.
Participants should have some experience with classification models (e.g. logistic regression, linear discriminant analysis, etc.). Although software is not explicitly described to solve the class imbalance issue, class participants will receive a copy of the illustrative data as well as R code to reproduce all of the analyses shown in the workshop.
|
||
Workshop 5: Getting Practical: How to Fit Bayesian Models for Health Policy Evaluation |
Wed, Oct 7, 10:15 AM - 12:15 PM
Renaissance Salon |
|
Instructor(s): Mariel McKenzie Finucane, Mathematica Policy Research | ||
This workshop will provide a hands-on introduction to Bayesian health policy evaluation. Bayesian approaches to health policy analyses have long been theoretically appealing because (a) they can increase power by borrowing strength across related data sources, (b) they include a natural penalty on model complexity that can obviate the need to worry about multiple comparisons, and (c) they produce inference that can be summarized using intuitive probability statements, which provide more information and flexibility to stakeholders than the ‘thumbs-up thumbs-down’ inference of traditional evaluation methods. Until recently though, it was not computationally feasible to fit the complex Bayesian models needed to answer real-world policy questions. Even simple models could present substantial computational hurdles. An important recent advance – Stan, a free, open-source, probabilistic programming language – makes complex Bayesian modeling not only feasible, but fast, elegant, and newcomer-friendly. We will introduce attendees to Stan, using an evaluation of an Affordable Care Act initiative as a real-world motivating example. Download Handouts |
||
Workshop 6: Patient-Reported Outcomes: Measurement, Implementation and Interpretation |
Wed, Oct 7, 1:30 PM - 3:30 PM
Grand Ballroom |
|
Instructor(s): Andrew Bushmakin, Pfizer Inc; Joseph C Cappelleri, Pfizer Inc | ||
This workshop provides an exposition on health measurement scales – specifically, on patient-reported outcomes based on the instructors’ co-authored book. Patient-reported outcomes (PROs) is an umbrella term that encompasses a whole host of subjective outcomes such as pain, fatigue, depression, aspects of well-being (e.g., physical, functional, psychological), treatment satisfaction, health-related quality of life, and physical symptoms such as nausea and dizziness. PROs are often relevant in studying a variety of conditions—including pain, erectile dysfunction, fatigue, migraine, mental functioning, physical functioning, and depression—that cannot be assessed adequately without a patient’s evaluation and whose key questions require patient’s input on the impact of a disease or a treatment. To be useful to patients and other decision makers (e.g., health policy and clinical researchers, pharmaceutical companies, regulatory agencies, reimbursement authorities), who are stakeholders in medical care, a PRO must undergo a validation process to confirm that it measures what it is intended to measure reliably and accurately. Some key elements in the development of a patient-reported outcome measure will be noted in this workshop. The core topics of validity and reliability of a PRO measure will be discussed. Exploratory factor analysis and confirmatory factor analysis will be discussed as techniques to understand the underlying structure of a PRO measure. Other topics may be considered if time permits. Illustrations will be provided through real-life examples and simulated examples. Download Handouts |
||
Workshop 7: The Medical Expenditure Panel Survey (MEPS): A National Data Resource to Inform Health Policy |
Wed, Oct 7, 1:30 PM - 3:30 PM
Renaissance Salon |
|
Instructor(s): Jeffrey Rhoades, Agency for Healthcare Research and Quality | ||
The purpose of this Workshop is to facilitate the use of the Medical Expenditure Panel Survey Household Component (MEPS HC) public use data files by the health services research community. To meet this objective, participants are provided with a general overview of the MEPS, a description of available data files, information about on-line data tools, and some examples of the type of research projects the MEPS data can support. Major changes have taken place in the Nation's health care delivery system over the last decade. The most notable is the recent passage of the Affordable Care Act. Also consider the rapid expansion of managed care arrangements such as health maintenance organizations, preferred provider organizations, and other provider networks that seek to minimize increases in health care costs. The MEPS is a vital national data resource designed to continually provide health service researchers, policymakers, health care administrators, businesses, and others with timely, comprehensive information about health care use and costs in the United States. Newly released MEPS public use files provide analysts with opportunities to create unique analytic files for policy relevant analysis in the field of health services research, such as access to care and health disparities. In order to capture the unparalleled scope and detail of the MEPS HC, analysts need to understand the complexities of MEPS data files and data file linkages. This workshop will provide the knowledge necessary to formulate research plans utilizing the various MEPS HC files and linkage capabilities. Download Handouts |
||
Workshop 8: Causal Inference in Healthcare Studies with Multiple Treatments |
Wed, Oct 7, 1:30 PM - 3:30 PM
Garden Room |
|
Instructor(s): Roee Gutman, Brown University; Michael J Lopez, Skidmore College | ||
One of the main objectives of Causal Inference for Patient-Centered Outcomes Research (PCOR) is to identify which among multiple interventions affect an outcome the most. However, most of the work on causal inference methodologies is centered on methodologies for binary intervention, mainly using propensity scores (PS). Most of the literature on PS estimation and its use focuses on comparing only two treatments. This shortcoming leads researchers to dichotomize continuous, ordinal and categorical treatments, or to abandon PS methods entirely. Dichotomizing the non-binary treatment limits causal claims to the effects of the artificially dichotomized treatment, and it may suffer from loss of power, residual confounding, and possible bias in estimates.
Methods for estimating causal effects when comparing multiple treatments attempt to reduce potential bias by balancing covariates across treatment groups or weighting. In some cases, these methods are supplemented with regression adjustments to reduce the bias more. These methods may suffer from possible deficiencies such as non-transitivity of the estimated effects (e.g. treatment 1 may display better outcomes than treatment 2, and treatment 2 may display better than treatment 3, but treatment 1 will have worse outcome than treatment 3), non-sufficient reduction in bias due to leftover covariate imbalances, or extreme weights that can result in erratic causal estimates, an issue which becomes more likely as the number of treatments increases and the treatment assignment probabilities decrease.
Throughout, the workshop will include implementation of the methods using statistical software and illustrate the methods with several health-related examples. The workshop is intended for health services researchers and statisticians who are interested in estimating causal inferences with more than one treatment, and it is designed for a broad health services research audience with moderate levels of statistical understanding. Download Handouts |
||
Workshop 9: Evaluating the Impact of Unmeasured Confounding in Comparative Observational Research |
Wed, Oct 7, 3:45 PM - 5:45 PM
Grand Ballroom |
|
Instructor(s): Daniel Beavers, Wake Forest School of Medicine; Xiang Zhang, Eli Lilly and Company | ||
The use of retrospective observational research as a tool for medical decision making, particularly with data from healthcare claims databases and electronic medical records, has been growing in recent years. Comparative effectiveness studies based on observational data can assess how drugs of interest perform in real world settings and provide important information to assist in treatment recommendations in practice. However, the use of such observational data for comparative effectiveness is challenged by selection bias and potential for unmeasured confounding. This is especially problematic for analysis using insurance claim databases, which are not collected for research purposes and key clinical measures are often not available for analysis.
The quantitative assessment of the potential influence of unmeasured confounders in observational data analysis is rare, despite the reliance of the validity of any cohort comparison on the “no unmeasured confounders” assumption. In fact, much research simply mentions this assumption as a limitation without any assessment of its likely validity or impact if it is not valid. Recently, multiple approaches useful for sensitivity assessment for unmeasured confounding have emerged, with the particular method and performance depending on the amount of information available on the unmeasured confounders. For instance, Bayesian twin regression modeling has been shown to be a flexible tool that can incorporate information on potential confounders obtained from the literature or other sources (Faries et. al, 2013). Recent research (Stamey et al, 2014) provides guidance on the performance of such techniques, demonstrating situations where such sensitivity analysis can recover a substantial portion of the missing information. Given the recent methodological advances, quantitative assessment of the potential impact of unmeasured confounding should be an important part of sensitivity analyses when comparative research based on real world data is utilized for medical decision making.
The main objectives of this workshop are to demonstrate the importance of addressing unmeasured confounding in comparative effectiveness analyses from real world data and to equip the participants with a framework and tools to select and implement appropriate sensitivity analyses, leading to higher quality and more interpretable observational research. Download Handouts |
||
Workshop 10: Social Network Analysis with Applications to Medicine and Health Policy |
Wed, Oct 7, 3:45 PM - 5:45 PM
Renaissance Salon |
|
Instructor(s): A. James O'Malley, Geisel School of Medicine at Dartmouth College | ||
Social network analysis (SNA) is an emerging and very broad area of research involving multiple disciplines. Because many questions about the social organization of medicine and health policy involve interdependencies among individuals that may be depicted by networks of relationships, SNA has tremendous potential to advance research and practice in these fields. Although social network studies have been pursued for some time in social science disciplines, where numerous descriptive methods for analyzing them have been proposed, interest among statistician and other quantitative researchers has only blossomed in the era of modern computing. However, interest in the analysis of social network data has recently rapidly grown among both statisticians, who have developed more elaborate models and methods for analyzing network data, and applied researchers interested in utilizing this powerful new area. I will review fundamentals of and recent innovations in SNA using networks from past work in medicine and health care policy for illustration. The workshop will consist of three main parts: (i) Introduction to the forms of network data, basic network statistics, and common descriptive measures; (ii) Models or analyses in which networks are fundamental to the construction of explanatory variables, including estimation methods that seek to distinguish social influence (e.g., peer effects) from other social phenomena such as homophily; (iii) Relational models in which the network itself is a multivariate dependent variable. Due to the potential complex (causal) dependencies and correlation structures among observations in network data, complexities in estimating both types of models arise. These will be discussed along with strategies for utilizing currently available software. Download Handouts |
||
Workshop 11: Estimating the Causal Effects of Time-Varying Treatments using State-of-the-Art Methods |
Wed, Oct 7, 3:45 PM - 5:45 PM
Garden Room |
|
Instructor(s): Lane Burgette, RAND Corporation | ||
This course will provide attendees with hands-on experience and guidance for advance issues in weight estimation for causal inference problems. Estimation of causal effects is a primary activity of many health policy research studies. Examples include testing which state-level health policies have the biggest impact on health outcomes or whether repeated exposure to a particular substance abuse treatment program might cumulate in effectiveness over time. Controlled experiments are the gold standard for estimating such effects. However, experiments are often infeasible, forcing analysts to rely on observational data in which treatment assignments are out of the control of the researchers. This short course will provide guidelines for using state-of-the-art methods to estimate weights for drawing causal inference in two settings: (1) settings with more than 2 treatment (or exposure) groups of interest and (2) settings where treatments of interest vary over time. The course will provide a brief introduction to causal modeling using the potential outcomes framework and the use of propensity score weighting and inverse probability of treatment weights to estimate causal effects from observational data with more than 2 treatments or time-varying treatments. It will also present step-by-step guidelines on how to estimate and perform diagnostic checks of the estimated weights for testing the relative effectiveness of two or more interventions and the cumulative effects of time-varying interventions. Attendees will gain hands-on experience estimating each type of weight using boosted models and covariate balancing propensity scores in R, SAS and Stata; evaluating the quality of those weights; and using them to estimate intervention effects. Attendees should be familiar with linear and logistic regression. Download Handouts |
||
Interactive Poster Session I & Welcome Reception |
Wed, Oct 7, 6:00 PM - 7:30 PM
L'Apogee 17 |
|
Organizer(s): Recai M. Yucel, State University of New York at Albany; Kelly H. Zou, Pfizer Inc | ||
|
||
Combining Item Response Theory with Multiple Imputation to Crosswalk Between Health Assessment Questionnaire
|
||
Marginalized Two-Part Models for Generalized Gamma Family of Distributions
|
||
Large, Sparse Optimal Matching with Refined Covariate Balance in an Observational Study of the Health Outcomes Produced by New Surgeons
|
||
WITHDRAWN: Semiparametric Functional Response Model with Incompletely Observed Functional Data
|
||
Comparison of One-Part Models and a Two-Part Marginalized Model for the Analysis of Health Care Expenditures
|
||
WITHDRAWN: Sparse Principal Component Analysis in Linear Regression Survival Model
|
||
WITHDRAWN: Estimation and Inference for Optimal Treatment Regimes Under Constraints
|
||
Composite Endpoints in the Semi-Competing Risks Setting
|
||
A Bayesian Hierarchical Summary Receiver Operating Characteristic Model for Network Meta-Analysis of Diagnostic Tests
|
||
Logistic Ridge Regression for Domain Importance Assessment in Patient-Reported Outcomes Studies
|
||
Using MCMC for Estimating Precision of Estimates of Complicated Functions of Parameters When Modeling Categorical Data
|
||
The Public Health Impact of Air Quality Regulations Through Change in Ambient PM2.5
|
||
Utilizing External Validation Data with Bayesian Data Augmentation and Variable Selection to Adjust for Confounding
|
||
Incorporating Non-Randomized Data with Randomized Clinical Trials Using Commensurate Priors
|
||
Regression Analysis of Diagnostic Accuracy of Biomarkers with a Continuous Gold Standard
|
||
A Bayesian Credible Subgroups Approach to Identifying Patient Subgroups with Positive Treatment Effects
|
||
Are Providers Ready to Move to a Web Response Option Model for Medical Records Submission? Results from a Qualitative Assessment of Providers from the National Immunization Survey
|
||
Basic Versus Advanced Life Support Ambulances for Out-of-Hospital Medical Emergencies
|
||
Overview of the Development of the U.S. National Cancer Institute's Patient-Reported Outcomes Version of the Common Terminology Criteria for Adverse Events (PRO-CTCAE)
|
||
The Impact of Misclassification of Obesity Coding in Administrative Claims Data on Negative Health Outcomes
|
||
Optimal Diagnostic Testing Strategies via Influence Diagrams
|
||
Pathoscope 2.0: Statistical and Computational Methods for Accurate Characterization of Microbes in Sequencing Samples
|
||
Risk Factors Associated with Nocturnal Hypoglycemia Events in Insulin-Treated Type 2 Diabetes Mellitus Patients: An Integrated Clinical Trial Database Analysis
|
||
Controlling for Unobserved Spatially Correlated Confounders in Observational Studies
|
||
A Network Analysis of Variation in Evidence-based ICD Implantation
|
||
Thu, Oct 8 |
||
Conference Registration |
Thu, Oct 8, 7:00 AM - 5:00 PM
L'Apogee 17 |
|
|
||
Interactive Poster Session II |
Thu, Oct 8, 7:30 AM - 8:30 AM
L'Apogee 17 |
|
Organizer(s): Recai M. Yucel, State University of New York at Albany; Kelly H. Zou, Pfizer Inc | ||
|
||
Methods for Responder Analysis of Patient-Reported Outcomes
|
||
Comparison of Binomial and Multinomial Network Meta-Analysis Model by Simulation
|
||
Malaria Prevention: Are High-Risk Households in Kenya Receiving Treatments?
|
||
Analysis and Comparison of Air Pollutants in Wuhan, China
|
||
Treatment Effects of Opioids Versus NSAIDs Prescribed from the Emergency Department Following Motor Vehicle Crash: A Propensity Matched Analysis
|
||
Functional Status, Social Support, and Wealth Affect Readmissions for Pneumonia and Heart Failure Beyond Standard Medicare Risk Adjustors
|
||
An Alarm System for Flu Outbreaks Using Google Flu Trend Data
|
||
Challenges and Opportunities in Incorporating Spatial Information in Analyses of Periodontal Data
|
||
Simulation Study to Investigate the Impact of Pre-Surgical Cost Analysis on Estimating Bariatric Surgery Cost-Effectiveness in the VA System
|
||
OpenMeta-Analyst: Open-Source, Cross-Platform Software for Advanced Meta-Analysis
|
||
Health Care Utilization Patterns and Costs of Care: A Longitudinal Analysis of the Implications of the Mental Health Parity Act for Those with Comorbid Diabetes and Depression Diagnoses
|
||
Meta-Analysis of Predictive Test Accuracies
|
||
Impact of a Return-to-Work Program on the Economic Burden of Injury in Malaysia
|
||
Generalized Gamma Distribution for Bayesian Profiling of Facility-Level Variation in 30-Day, Risk-Adjusted Costs Following Percutaneous Coronary Intervention (PCI): The VA’s CART Program
|
||
Propensity Score Weighting Methods for a Continuous Treatment in a Multilevel Data Setting
|
||
Components of Cost Variation in a Veterans Pain Population
|
||
Robust Estimation of Causal Effects for Comparative Effectiveness Research
|
||
Comparison of Readily Available Causal Mediation Methods for Evaluating Policy Interventions
|
||
Baseline Factors Associated with the Glycemic Response to Treatment with Once Weekly Dulaglutide in Patients with Type 2 Diabetes
|
||
Hypothesis Testing in Multiple Health Care Databases for Scientific Insight Generation
|
||
Mortality Comparison of Endovascular Versus Open Repair for Abdominal Aortic Aneurysm Using Instrumental Variables
|
||
Comparison of Multivariate Matching Methods That Select a Subset of Treatment and Control Observations
|
||
Propensity Model Evaluation of the Effects of Online Social Network Participation on Promoting Smoking Cessation
|
||
Clinicians’ Preference for Clinical Laboratories Using Logistic Regression Model
|
||
Welcome, Keynote #1, and Plenary Address |
Thu, Oct 8, 8:30 AM - 10:00 AM
Grand Ballroom |
|
Chair(s): Recai M. Yucel, State University of New York at Albany; Kelly H. Zou, Pfizer Inc | ||
|
||
Session 12 (Invited) Linking Records from Different Sources in Healthcare Applications |
Thu, Oct 8, 10:30 AM - 12:15 PM
Grand Ballroom |
|
Organizer(s): Roee Gutman, Brown University | ||
Chair(s): Michael J Lopez, Skidmore College | ||
|
||
10:35 AM |
Error Adjustments for File-Linking Methods Using Encrypted Unique Client Identifier (eUCI) with Application to Recently Released HIV+ Prisoners
|
|
11:00 AM |
Statistical Analysis and Modeling for Errors in Record Linkage
|
|
11:25 AM |
Causes and Consequences of Data Linkage Errors: False and Missed Matches Following Linkage of Hospital Data
|
|
11:50 AM |
Validity of Deterministic and Probabilistic Record Linkage Using Multiple Indirect Personal Identifiers: Linking a Large Registry to Claims Data
|
|
Session 13 (Invited) Advances in combining information and meta-analysis |
Thu, Oct 8, 10:30 AM - 12:15 PM
Garden Room |
|
Organizer(s): Thomas Trikalinos, Brown University; Min-ge Xie, Rutgers University | ||
Chair(s): Thomas Trikalinos, Brown University | ||
|
||
10:35 AM |
A Bayesian Nonparametric Meta-Analysis Model
|
|
11:00 AM |
Large-Scale Comparison of Univariate and Multivariate Meta-Analysis for Categorical Outcomes
|
|
11:25 AM |
Exact Inference in Meta-Analysis via Exact Confidence Distributions
|
|
Discussant(s): Min-ge Xie, Rutgers University |
||
Session 14 (Topic-Contributed) Statistics and Payment Reform: Toward Better Value in Health Care |
Thu, Oct 8, 10:30 AM - 12:15 PM
Renaissance Salon |
|
Organizer(s): Frank B. Yoon, Mathematica Policy Research | ||
Chair(s): Frank B. Yoon, Mathematica Policy Research | ||
|
||
10:35 AM |
Health Plan Type Variations in Spells of Health Care Treatment
|
|
10:55 AM |
Macro-Level Composite Measures: Their Value in a Pay-for-Performance Program
|
|
11:15 AM |
Post-Adjustment of Payments by Patient and Provider Categories to Reduce Payment Disparities Under Pay-for-Performance Schemes
|
|
Discussant(s): Eugene Rich, Mathematica Policy Research |
||
Session 15 (Contributed) Outcomes Research |
Thu, Oct 8, 10:30 AM - 12:15 PM
Salon 2 |
|
Chair(s): Layla Parast, RAND Corporation | ||
|
||
10:30 AM |
Intensive Efforts Can Drive Health Care Survey Response Rates Over 50%
|
|
10:45 AM |
Analysis of Overlapping Symptoms in Co-Occurring Conditions
|
|
11:00 AM |
Trajectories of Medical Costs Among Patients with Chronic Diseases from Health Plan and Provider-Delivered Care Management Programs: Four Years Follow-Up Results Using Bayesian Adaptive Splines
|
|
11:15 AM |
Provider Profiling for Quality of End-of-Life Care
|
|
11:30 AM |
Using Propensity Score Methods to Combine Multiple Data Sources to Compare the Effectiveness of Transcranial Magnetic Stimulation (TMS) with Antidepressant Drug Therapy for Depression
|
|
11:45 AM |
Measuring the Effects of Time-Varying Medication Adherence on Health Outcomes
|
|
12:00 PM |
Inference for Identifying Outlying Health Care Providers
|
|
Session 16 (Invited) Tools for Policy: Bayesian Assessments to Support Decision Makers |
Thu, Oct 8, 1:30 PM - 3:15 PM
Grand Ballroom |
|
Organizer(s): Mariel McKenzie Finucane, Mathematica Policy Research | ||
Chair(s): Jose Alvir, Pfizer Inc. | ||
|
||
1:35 PM |
Evolving Needs of Policymakers for Drawing Inferences from Evaluation Results
|
|
2:00 PM |
A Hierarchical Bayesian Evaluation of Health System Change Using Administrative Data
|
|
2:25 PM |
Causal Inference Methods for Evaluating Air Quality Control Policies
|
|
Discussant(s): Sharon-Lise Normand, Harvard Medical School, Health Care Policy |
||
Session 17 (Invited) Unmeasured Confounding and Sensitivity Analysis for Observational Data |
Thu, Oct 8, 1:30 PM - 3:15 PM
Garden Room |
|
Organizer(s): Joseph W Hogan, Brown University | ||
Chair(s): Hana Lee, Brown University School of Public Health | ||
|
||
1:35 PM |
Bayesian Nonparametrics, Informative Priors, and Causal Inference
|
|
2:05 PM |
Calibrating Sensitivity Analyses to Observed Covariates in Observational Studies
|
|
2:35 PM |
Representing Unmeasured Confounding in Causal Models for Observational Data
|
|
Session 18 (Topic-Contributed) Network meta-analysis for comparative effectiveness research |
Thu, Oct 8, 1:30 PM - 3:15 PM
Renaissance Salon |
|
Organizer(s): Hwanhee Hong, Johns Hopkins Bloomberg School of Public Health | ||
Chair(s): Elizabeth A. Stuart, Johns Hopkins Bloomberg School of Public Health | ||
|
||
1:35 PM |
Bayesian Network Meta-Analysis for Estimating Drug Class Effects, with Applications to Primary Open Angle Glaucoma
|
|
1:55 PM |
Efficient Network Meta-Analysis: A Confidence Distribution Approach
|
|
2:15 PM |
Network Meta-Analysis Modeling for Diagnostic Accuracy Studies
|
|
2:35 PM |
||
Session 19 (Contributed) Advanced Statistical Inferences For Complex Problems |
Thu, Oct 8, 1:30 PM - 3:15 PM
Salon 2 |
|
Chair(s): Jun Su, Novartis | ||
|
||
1:35 PM |
||
1:50 PM |
Dietary Patterns and Determinants of Mercury and Omega-3 Exposure in Pregnant Women Living in the Seychelles
|
|
2:05 PM |
WITHDRAWN: Identification of Cell Subgroups Related to Driver Mutations in Cancer
|
|
2:20 PM |
Multivariate Network Meta-Analysis for Predicting Treatment Effect from Surrogate Endpoints
|
|
2:35 PM |
The Utilization and Role of Elderly Imaging: Analysis Using Electronic Health Record Data
|
|
2:50 PM |
||
Session 20 (Invited) Beyond propensity scores: causal inference with evidence synthesis, machine learning, and response surface techniques |
Thu, Oct 8, 3:30 PM - 5:15 PM
Grand Ballroom |
|
Organizer(s): Laura Hatfield, Harvard Medical School | ||
Chair(s): Juned Siddique, Northwestern University | ||
|
||
3:35 PM |
Evidence synthesis using observational and randomized studies: empirical results, statistical methods, and practical recommendations
|
|
4:00 PM |
Competing Strategies for Estimating Causal Response Surfaces Using Bayesian Nonparametric Models
|
|
4:25 PM |
A Machine Learning Framework to Prevent 'Gaming' in Plan Payment Risk Adjustment
|
|
4:50 PM |
Bayesian Approaches for Network Meta-Analysis of Randomized and Nonrandomized Clinical Trial Data
|
|
Session 21 (Invited) Improving Medical Decision Making in the Era of Personalized Medicine |
Thu, Oct 8, 3:30 PM - 5:15 PM
Garden Room |
|
Organizer(s): Yuanjia Wang, Columbia University | ||
Chair(s): Christopher Schmid, Brown University | ||
|
||
3:35 PM |
A Cure-Rate Model for Estimating the Optimal Dynamic Treatment Sequence Following Bone Marrow Transplantation
|
|
4:00 PM |
Targeted Learning of the Optimal Dynamic Treatment and Statistical Inference for the Mean Outcome Under the Optimal Dynamic Treatment
|
|
4:25 PM |
Estimating Individualized Dosage Rules Using Outcome-Weighted Learning
|
|
4:50 PM |
Sequential Multiple Assignment Randomization Trials with EnRichment (SMARTer) Design
|
|
Session 22 (Topic-Contributed) Statistical Challenges in Conducting Veteran Health Policy Studies |
Thu, Oct 8, 3:30 PM - 5:15 PM
Renaissance Salon |
|
Organizer(s): Maren Olsen, Duke University | ||
Chair(s): Maren Olsen, Duke University | ||
|
||
3:35 PM |
A Statistician’s View from the Evaluation Desk: The VA Caregiver Support Program Partnered Evaluation Center
|
|
3:55 PM |
Practical Challenges of Evaluating the Veterans Health Administration (VHA) Lung Cancer Screening Clinical Demonstration Project
|
|
Discussant(s): James Burgess, Jr., VA Boston Healthcare System; Matthew Maciejewski, Duke University Medical Center |
||
Session 23 (Contributed) Regulatory Topics on Policy Government and Health Policy |
Thu, Oct 8, 3:30 PM - 5:15 PM
Salon 2 |
|
Chair(s): Corwin Zigler, Harvard School of Public Health | ||
|
||
3:35 PM |
A Quantitative Evaluation of Gonorrhea and Chlamydia (STD) in Washington DC (2000–2013)
|
|
3:50 PM |
||
4:05 PM |
Stronger Nudges: Improving Health Care Quality by Nudging Medicare Beneficiaries Using Targeted Quality Information
|
|
4:20 PM |
Superfood or Not? DEA Makes the Call
|
|
4:35 PM |
Cost and Quality of Health Care Provider Organizations in California: A Comparison of Value Metrics
|
|
4:50 PM |
Optimizing Surveillance of Low-Risk Prostate Cancer
|
|
5:05 PM |
||
Social Event: Student Travel Awards and Performance by The Imposteriors (on own) |
Thu, Oct 8, 7:00 PM - 10:00 PM
Offsite - The Rosendale |
|
|
||
Fri, Oct 9 |
||
Conference Registration |
Fri, Oct 9, 7:30 AM - 11:30 AM
L'Apogee 17 |
|
|
||
HPSS Awards and Keynote #2 Address |
Fri, Oct 9, 8:00 AM - 9:30 AM
Grand Ballroom |
|
Chair(s): Recai M. Yucel, State University of New York at Albany; Kelly H. Zou, Pfizer Inc | ||
|
||
Session 24 (Invited) Joint models for the analysis of longitudinal health-related outcomes |
Fri, Oct 9, 9:45 AM - 11:30 AM
Grand Ballroom |
|
Organizer(s): Juned Siddique, Northwestern University | ||
Chair(s): Ofer Harel, University of Connecticut | ||
|
||
9:50 AM |
Joint Modeling of Mean and Variance in Longitudinal Data: Shared Random Effects vs. Latent Class Models
|
|
10:15 AM |
Tailoring Treatment Information Using Personal Characteristics and Health Outcome Preferences
|
|
10:40 AM |
Joint Modeling the Frequency and Duration of Physical Activity Using Data from a Lifestyle Intervention Trial
|
|
11:05 AM |
A Shared Parameter Location Scale Item Response Theory (IRT) Model for Repeated Ordinal Questionnaire Data
|
|
Session 25 (Invited) Advances in non-experimental causal inference methods for patient-centered health services and health policy research |
Fri, Oct 9, 9:45 AM - 11:30 AM
Garden Room |
|
Organizer(s): Elizabeth Stuart, Johns Hopkins Bloomberg School of Public Health | ||
Chair(s): Elizabeth Stuart, Johns Hopkins Bloomberg School of Public Health | ||
|
||
9:50 AM |
Methods for Comparative Effectiveness Analyses in a High-Dimensional Covariate Space with Few Events
|
|
10:15 AM |
Evaluating Observational Data Analyses: Confounding Control and Treatment Effect Heterogeneity
|
|
10:40 AM |
Selecting Optimal Observational Methods for Comparative Effectiveness Research
|
|
Discussant(s): Emily Evans, PCORI |
||
Session 26 (Topic-Contributed) Bayesian Analysis of Heterogeneous Treatment Effect in Patient-Centered Outcomes Research |
Fri, Oct 9, 9:45 AM - 11:30 AM
Renaissance Salon |
|
Organizer(s): Ravi Varadhan, Department of Oncology, Johns Hopkins University | ||
Chair(s): Roee Gutman, Brown University | ||
|
||
9:50 AM |
Bayesian HTE Analysis for PCOR
|
|
10:10 AM |
BEANZ: A Web-Based Software for Bayesian Analysis of Heterogeneous Treatment Effect in Patient-Centered Outcomes Research
|
|
10:30 AM |
The Choice of Treatment Effect Scale for the Analysis of Heterogeneity of Treatment Effect (HTE) and Qualitative Interactions
|
|
10:50 AM |
Guidance for Bayesian Analyses of Heterogeneity of Treatment Effect
|
|
11:10 AM |
||
Session 27 (Contributed) Recent Advances in Modeling and Analysis |
Fri, Oct 9, 9:45 AM - 11:30 AM
Salon 2 |
|
Chair(s): Haekyung Jeon-Slaughter, The University of Texas Southwestern Medical Center | ||
|
||
9:50 AM |
Joint GEEs for Multivariate Correlated Data with Incomplete Binary Outcomes
|
|
10:05 AM |
A Spatiotemporal Quantile Regression Model for Emergency Department Expenditures
|
|
10:20 AM |
Multidimensional Time Model for Probability Cumulative Function
|
|
10:35 AM |
A Look at Quantile Regression
|
|
10:50 AM |
Weighting Methods for Assessing Mediation Effect Variation in Multi-Site Trials
|
|
11:05 AM |
Derivation and Validation of Clinical Phenotypes for Asthma: A Systematic Review
|
|
11:20 AM |
||
Session 28 (Invited) Accuracy, Equity, Incentives: Improving Analytic Methods for Quality Measures |
Fri, Oct 9, 12:30 PM - 2:15 PM
Grand Ballroom |
|
Organizer(s): Therese A Stukel, Institute for Clinical Evaluative Sciences (ICES)/ University of Toronto | ||
Chair(s): Therese A Stukel, Institute for Clinical Evaluative Sciences (ICES)/ University of Toronto | ||
|
||
12:35 PM |
Enhancing the Hospital Compare Mortality Model Using a Bayesian Framework
|
|
12:55 PM |
Should We Adjust Quality Measures for Patient Socioeconomic Status?
|
|
1:15 PM |
Setting the Research Agenda for Quality Performance Analyses
|
|
1:35 PM |
Maximizing the Effectiveness of Public Reporting to Improve Health Care Quality
|
|
1:55 PM |
||
Session 29 (Invited) Novel Statistical Methods for Large Medical Databases |
Fri, Oct 9, 12:30 PM - 2:15 PM
Garden Room |
|
Organizer(s): Sherri Rose, Harvard Medical School | ||
Chair(s): Sherri Rose, Harvard Medical School | ||
|
||
12:35 PM |
Machine learning methods for risk prediction with censored electronic health data
|
|
1:00 PM |
Nonstandard Measurements in Electronic Medical Records
|
|
1:25 PM |
Big Data Approaches for Health Policy: Characterizing the Diffusion of New Medical Technologies
|
|
1:50 PM |
Bayesian Methods for Comparative Effectiveness Research
|
|