Sep
25
12:00pm12:00pm

TC-CFAR Seminar: Dr. Robert L. Murphy

Dr. Robert L. Murphy will be presenting for the TC-CFAR Seminar Series

Location: 
Northwestern University - Chicago Campus
Wieboldt Hall, Room 431
340 E. Superior Ave.

Dr. Robert L. Murphy is the Director of the Center for Global Health at Northwestern University where he holds his primary academic appointment as the John P. Phair Professor of Medicine. He also holds an appointment as Professeur Associé de Recherche at the Pierre et Marie Curie Université-Paris in France. Dr. Murphy’s primary research and clinical interest is in viral infections. His research includes drug development of new antiretroviral drugs and vaccines for HIV and viral hepatitis and the scale-up of therapy for AIDS, tuberculosis and malaria in sub-Saharan Africa.

Dr. Murphy is Special Advisor to the President’s Emergency Plan for AIDS Relief (PEPFAR) program in Nigeria, sponsored by the Harvard School of Public Health, where he has overseen the set up of 42 clinics that currently treat over 75,000 patients with HIV/AIDS. He also consults on NIH-supported antiretroviral education projects in Senegal and is Pricipal Investigator for Northwestern’s NIH/Fogarty International AIDS Training Grant based in Nigeria and Mali and the Northwestern Fogarty Frameworks grant. International activities and interests include assisting in the establishment of an AIDS Clinic in rural southern Kenya funded by the African Village Clinics Foundation of Chicago.

Dr. Murphy is the Principal Investigator for the National Institutes for Allergy and Infectious Diseases (NIAID) Adult AIDS Clinical Trials Group (ACTG) at Northwestern. Within ACTG he has held numerous leadership positions including membership on the Scientific Agenda Steering Committee and Adult Executive Committee, the governing body of the group. He is a member of multiple medical societies and sits on the boards of several non-profit organizations including the Drucker Family Charitable Trust, the International AIDS Education Project, Objectif Recherche Vaccin SIDA and the Midwest AIDS Foundation, of which he is the founder.

Professor Murphy has been with Northwestern University since 1978. After receiving his AB from Boston University, he attended the Loyola Stritch School of Medicine and later completed his internship, residency, and fellowship in infectious diseases at the McGaw Medical Center, Northwestern University. Professor Murphy has remained with Northwestern, becoming professor of medicine in 1999. He presently holds the distinguished post of John P. Phair Professor of Medicine.

View Event →
Oct
2
9:00am 9:00am

TC-CFAR Symposium: “Bridging the Disciplines to Understand HIV Transmission”

Symposium Overview

Join us for the Annual Third Coast CFAR Symposium: Bridging the Disciplines to Understand HIV Transmission. Taking a multi-disciplinary approach, the goal of the Third Coast CFAR Symposium is to integrate cellular, molecular, clinical, and behavioral perspectives into three sessions that examine HIV transmission in the context of the host, the virus, and intervention.

These sessions will include invited experts from outside Chicago as well as short talks from Third Coast CFAR investigators. Two poster sessions will showcase the breadth of HIV/AIDS research being conducted across Chicago as well as highlighting services provided by the Third Coast CFAR.

Register for Symposium

Call for Abstracts

Symposium Agenda at a Glance

Detailed schedule to be posted in September

Continental Breakfast & Poster Set-up

Session I: Focus on Host
Paul Volberding, MD, University of California San Francisco
Petronela Ancuta, PhD, University of Montreal
Maximo Brito, MD, MPH, University of Illinois-Chicago

Short talks

Poster Session I

Lunch

Session II: Focus on Virus
Ron Swanstrom, PhD, University of North Carolina
Jonathan Carlson, PhD, Microsoft

Short talks

Poster Session II

Session III: Focus on Interventions
Steven Safren, PhD, University of Miami
Adam Carrico, PhD, University of Miami

Short talks

Reception

-------------------

Administrative Contacts

Kamara Fant
kamara.fant@northwestern.edu
312-503-4641

Fern Murdoch
f-murdoch@northwestern.edu
 312-503-4624

The Third Coast CFAR Symposium Organizing Committee

Thomas Hope, PhD                                    
Cell and Molecular Biology
Northwestern University

Judith Moskowitz, PhD
Medical Social Sciences
Northwestern University

Jessica Ridgway, MD
Medicine/Infectious Diseases
University of Chicago

Symposium Information

The Annual Third Coast CFAR Symposium, Bridging the Disciplines to Understand HIV Transmission, will take place at the Prentice Women’s Hospital 3rd Floor Conference Center October 2, 2017.

Third Coast Center for AIDS Research
625 N Michigan Ave.
Suite 1400
Chicago, IL 60611

Prentice Women’s Hospital 3rd Floor Conference Center
250 E. Superior St.
Chicago, IL 60611

 

View Event →
Oct
3
12:00pm12:00pm

PSMG: HIV Series - Anna Satcher-Johnson & Rick Song

Estimating HIV incidence, prevalence, and undiagnosed infection in the United States

Anna Satcher-Johnson, M.P.H. & Rick Song, Ph.D.
Centers for Disease Control and Prevention

ABSTRACT:

The burden of HIV infection and health outcomes for people living with HIV varies across the United States. New methods allow for estimating national and state-level HIV incidence, prevalence, and undiagnosed infections using surveillance data and CD4 values. 

Methods: HIV surveillance data reported to the Centers for Disease Control and Prevention and the first CD4 value after diagnosis were used to estimate the distribution of delay from infection to diagnosis based on a well-characterized CD4 depletion model. This distribution was used to estimate HIV incidence, prevalence, and undiagnosed infections during 2010–2014. Estimated annual percentage changes were calculated to assess trends. 

During 2010–2014, HIV incidence decreased 10.3% (EAPC = -3.1%) and the percentage of undiagnosed infection decreased from 17.1% to 15.0% (EAPC = -3.3%) in the United States; HIV prevalence increased 9.1% (EAPC = 2.2%). In 2014, Southern states accounted for 50% of both new HIV infections and undiagnosed infections. HIV incidence and undiagnosed infection decreased in the United States during 2010–2014; however, outcomes varied by state and region. Progress in national HIV prevention is encouraging but intensified efforts for testing and treatment are needed in the South and states with high percentages of undiagnosed infection.

View Event →
Oct
10
12:00pm12:00pm

PSMG: HIV Series - Steven Goodreau

Sources of racial disparities in HIV prevalence in men who have sex with men in Atlanta, GA, USA: A modeling study

Steven Goodreau, Ph.D.
University of Washington, Seattle

ABSTRACT:

Black men who have sex men (MSM) in the US have a substantially higher prevalence of infection than White MSM, and many proximal and distal explanations have been offered to account for pieces of this disparity.

We created a simulation model to assess the strength of existing hypotheses and data. We built a dynamic, stochastic, agent-based network model of Black and White MSM aged 18–39 years in Atlanta, that incorporated race-specific individual and dyadic-level prevention and risk behaviors, network attributes, and care patterns. We estimated parameters from two Atlanta-based studies in this population (n=1117), supplemented by other published work. We modeled the ability for racial assortativity to generate or sustain disparities in the prevalence of HIV infection, alone or in conjunction with scenarios of observed racial patterns in behavioral, care, and susceptibility parameters. 

Race-assortative mixing alone could not sustain a pre-existing disparity in prevalence of HIV between Black and White MSM. Differences in care cascade, stigma-related behaviors, and CCR5 genotype each contributed substantially to the disparity (explaining 10.0%, 12.7%, and 19.1% of the disparity, respectively), but nearly half (44.5%) could not be explained by the factors investigated. A scenario assessing race-specific reporting differences in risk behavior was the only one to yield a prevalence in black MSM (44.1%) similar to that observed (43.4%). Racial assortativity is an inadequate explanation for observed disparities. Work to close the gap in the care cascade by race is imperative, as are efforts to increase serodiscussion and strengthen relationships among Black MSM particularly. Further work is urgently needed to identify other sources of, and pathways for, this disparity, to integrate concomitant epidemics into models, and to understand reasons for racial differences in behavioral reporting.

View Event →
Oct
17
12:00pm12:00pm

PSMG: HIV Series - David Holtgrave & Cathy Maulsby

Cost and cost-utility analysis of RiC: A national Retention in Care program in real-world settings

David Holtgrave, Ph.D. & Cathy Maulsby, Ph.D.
Johns Hopkins University

ABSTRACT:
Persons diagnosed with HIV but not retained in HIV medical care accounted for the majority of HIV transmissions in 2009 in the United States (U.S.). There is an urgent need to implement and disseminate HIV retention in care programs; however little is known about the costs associated with implementing retention in care programs. We assessed the costs and cost-utility for six Retention in Care (RiC) programs using standard methods recommended by the U.S. Panel on Cost-effectiveness in Health and Medicine.  Program costs from the societal perspective ranged from $47,919 to $423,913 per year or $594 to $2,753 per participant. The programs averted between 0.23-1.65 HIV infections per year.  QALYs gained ranged from 1.51-11.00.  Using a threshold of $163,889 USD, all of the programs were cost-effective and four were cost-saving. Across a range of program models, retention in care interventions were cost effective (and the majority were cost saving), suggesting that retention in care programs are a judicious use of resources.

View Event →
Oct
23
12:00pm12:00pm

TC-CFAR Seminar: Dr. Robert Garofalo

Dr. Robert Garofalo will be presenting for the Third Coast CFAR Seminar Series

Dr. Garofalo’s primary clinical and research activities relate to the care of marginalized youth populations including HIV+ and LGBT young people. Research is largely HIV prevention in nature, mostly targeting either young men who have sex with men or transgender individuals. A new area of research and clinical interest involves working with young gender variant children and families.

View Event →
Nov
7
12:00pm12:00pm

PSMG: HIV Series - Kathryn Risher

Challenges in the evaluation of interventions to improve engagement along the HIV care continuum in the United States

Kathryn Risher, Ph.D, M.H.S
London School of Hygiene & Tropical Medicine

ABSTRACT:
In the United States (US), a high proportion of individuals living with HIV remain unlinked to care, disengaged from care, or incompletely adherent to antiretroviral therapy (ART). We conducted a systematic review of interventions to improve linkage to care, retention in care, re-engagement among those disengaged from care, and adherence to ART in the US. We find that the bulk of evidence (117/152 included studies) addresses adherence interventions, while a very small minority address linkage or reengagement interventions (7/152 and 4/152, respectively). There was tremendous heterogeneity in measures used to evaluate interventions. We found that most (59%) of studies report significantly improved outcomes, but the effect size was variable across studies and populations. The presentation will additionally include recommendations to address the challenges identified by the systematic review.

View Event →
Nov
13
12:00pm12:00pm

TC-CFAR Seminar: Nanette Benbow

Nanette Benbow will be presenting for the TC-CFAR Speaker Series. 

 Nanette Benbow, M.A.S, (she/her) is a research assistant professor in the Department of Psychiatry and Behavioral Sciences at the Northwestern University Feinberg School of Medicine; Director of the Third Coast Center for AIDS Research (TC-CFAR) End HIV Scientific Working Group; and member of the Center for Prevention Implementation Methodology. Her research interests include the use of epidemiologic and network modeling to improve HIV prevention and care continua; health equity for Latinos and sexual and gender minorities; and developing academic-public health partnerships in implementing evidence-based interventions that reduce HIV incidence in local settings. 

Before joining the Northwestern Faculty in 2016, Ms. Benbow was deputy commissioner of the STI/HIV Services Bureau of the Chicago Department of Public Health (CDPH) where she oversaw the development and implementation of prevention and care interventions.

View Event →
Dec
12
12:00pm12:00pm

PSMG: HIV Series - Cyprian Wejnert

CDC's National HIV Behavioral Surveillance (NHBS) system: Methods and impact on HIV prevention among key populations

Cyprian Wejnert, Ph.D.
Centers for Disease Control and Prevention

ABSTRACT:
The National HIV Behavioral Surveillance (NHBS) system was designed to monitor risk factors for HIV infection and HIV prevalence among individuals at increased risk for HIV infection, that is, sexually active men who have sex with men who attend venues, persons who recently injected drugs, and heterosexuals of low socioeconomic status living in urban areas. These groups were selected as priorities for behavioral surveillance because they represent the major HIV transmission routes and the populations with the highest HIV burden. Accurate data on HIV risk and testing behaviors in these populations are critical for understanding trends in HIV infections and planning and evaluating effective HIV prevention activities. This presentation will provide an overview of NHBS and the methods it employs (respondent-driven sampling and venue-based sampling), and will highlight some of the impact findings from NHBS have had on HIV prevention.

 

View Event →
Jan
23
12:00pm12:00pm

PSMG: Implementation Methods Series - Shri Narayanan

Deriving multimodal behavioral informatics for health applications

Shri Narayanan, Ph.D.
University of Southern California

ABSTRACT:
The convergence of sensing, communication and computing technologies is allowing capture and access to data, in diverse forms and modalities, in ways that were unimaginable even a few years ago.  These include data that afford the analysis and interpretation of multimodal cues of verbal and non-verbal human behavior to facilitate human behavioral research and its translational applications in healthcare.  These data not only carry crucial information about a person’s intent, identity and trait but also underlying attitudes, emotions and other mental state constructs. Automatically capturing these cues, although vastly challenging, offers the promise of not just efficient data processing but in creating tools for discovery that enable hitherto unimagined scientific insights, and means for supporting diagnostics and interventions. Recent computational approaches that have leveraged judicious use of both data and knowledge have yielded significant advances in this regard, for example in deriving rich, context-aware information from multimodal signal sources including human speech, language, and videos of behavior. These are even complemented and integrated with data about human brain and body physiology.   This talk will focus on some of the advances and challenges in gathering such data and creating algorithms for machine processing of such cues.  It will highlight some of our ongoing efforts in Behavioral Signal Processing (BSP)—technology and algorithms for quantitatively and objectively understanding typical, atypical and distressed human behavior—with a specific focus on communicative, affective and social behavior. The talk will illustrate Behavioral Informatics applications of these techniques that contribute to quantifying higher-level, often subjectively described, human behavior in a domain-sensitive fashion. Examples will be drawn from mental health and well being realms such as Autism Spectrum Disorders, Couple therapy, Depression and Addiction counseling. 

View Event →

Sep
19
12:00pm12:00pm

PSMG: HIV Series - Carl Latkin

Social network approaches to HIV Prevention and Care

Carl Latkin, Ph.D.
Johns Hopkins University

ABSTRACT:
The presentation will outline how social network approach can be utilized to reach hidden populations. Embedded in social network dynamics are social influence process that can be capitalized on to promote and sustain behavior change. The presentation will briefly provide guidance on tailoring social network inventories and then focus on developing and implementing RCT of network interventions and  measuring outcomes and controlling for contamination.

View Event →
Sep
11
12:00pm12:00pm

TC-CFAR Seminar: Dr. Warner C. Greene

Warner C. Greene, MD, PhD
Director, Senior Investigator, and Nick and Sue Hellmann Distinguished Processor, Translational Medicine at Gladstone Institute of Virology and Immunology
Professor of Medicine, Microbiology and of Immunology, UCSF
Member, Institute of Medicine of the National Academies
Fellow, American Academy for the Advancement of Science
Co-Director, UCSF-GIVI Center for AIDS Research
Councilor and President, Association of American Physicians.

Presenting: Death By Friendly Fire During HIV Infection
Location: 

Northwestern University - Chicago Campus
Wieboldt Hall, Room 431
340 E. Superior Ave.

 

Dr. Greene earned a bachelor’s degree at Stanford University and an MD/PhD at Washington University School of Medicine. He took his internship and residency training in Medicine at the Massachusetts General Hospital at Harvard. After serving as a Senior Investigator at the National Cancer Institute and a Professor of Medicine and Howard Hughes Investigator at Duke University Medical Center, Dr. Greene accepted his current position as the Founding Director of the Gladstone Institute of Virology and Immunology in 1991. The ongoing research in Dr. Greene’s laboratory focuses on the molecular mechanisms underlying HIV pathogenesis, latency, and transmission. He is the author of more than 366 scientific papers and has been recognized as one of the 100 Most Cited Scientists in the world. In 2007, Dr. Greene expanded his work to include global health in sub-Saharan Africa in his service as president and executive chairman of the Accordia Global Health Foundation. Accordia established the Infectious Diseases Institute at Makerere University in Uganda, which has trained thousands of African health care workers, is caring for 30,000 HIV-infected patients, and has brought health care to nearly 500,000 people living in remote rural regions of Uganda.

 

View Event →
Sep
7
12:00pm12:00pm

ISGMH Lecture: Joy Messinger at ISGMH's Current Issues in LGBTQ Health Lecture Series

When: Thursday, September 7th from 12:00-1:30 pm

Location: Stonewall Conference Room, 625 N. Michigan Suite 1400, Chicago, IL 60611

Lecture Title: “Our Survival Depends on Each Other: The Urgency of Intersectionality to Support the Health, Wellness, and Healing of LGBTQ Communities”

Click here for more lecture details and to RSVP.

If you are unable to join in person, we invite you to attend remotely using BlueJeans. Lunch will be served!

ISGMH’s Current Issues in LGBTQ Health Lecture Series focuses on highlighting important work being done in the field of LGBTQ health. Each lecture showcases the work of a different speaker or speakers. All of our lectures are open to the public to attend (as space allows) and available via livestream. Unless otherwise stated, our lectures are held in the Stonewall Conference Room at 625 N. Michigan Avenue, Suite 1400, Chicago, IL 60611.

 

 

View Event →
Jul
26
9:00am 9:00am

Conference: 1st North American Social Networks (NASN) Conference 2017

Workshop on Agent-Based Models in Social Network Analysis Using NetLogo
2 sessions, morning & afternoon: 9:00am until 4:00 pm on July 26, 2017
Attendance limited to 16 people

Instructors:
Dr. Wouter Vermeer and Gabriella Anton of Northwestern University

Session 1:
1) Introduction to ABM and Netlogo [1.5 hour]
2) Practical Assignments (part 1) [1.5 hour]
(Participants are encouraged to bring their own research questions and data)

Session 2:
1) Centralized evaluation of practical assignments (part 1) [0.5 hour]
2) Network extension [1 hour]
3) Practical Assignment 2 [1.5 hour]
(Participants are encouraged to bring their own research questions and data)

Description:
NetLogo (http://ccl.northwestern.edu/netlogo/) is a free, open source, modeling environment for simulating natural and social phenomena, authored by Prof. Wilensky, and it is currently the most cited agent-based modeling language in social sciences. NetLogo is designed to be a low threshold, high sealing programming environment that affords use among both young (or novice) coders and experts alike. Ever since its conception in 1999, the Center for Connected Learning and Computer Based Modeling (CCL) has continuously developed new language features and extensions to NetLogo. Among them is the Network-extension, which provides powerful Network Science capabilities in NetLogo.
NetLogo models, and agent-based models in general, are well suited for studying complex systems over time, and for executing scenario analyses. Modelers can give instructions to hundreds or thousands of "agents", all operating independently. This makes it possible to explore the 13 connection between the micro-level behavior of individuals and the macro level patterns that emerge from their interaction. With the addition of the Network extensions, NetLogo is particularly well suited for incorporating the networks that underlie systems in the analysis of such patterns, and for exploring the local rules that allow certain network structures to emerge. In the workshop, we provide an introduction on the use of NetLogo. We adopt a practical, hands on workshop approach, in which approximately half of the time will be spent with participants exploring and programming in NetLogo. For this workshop participants are encouraged to provide their own research questions (and/or data). We will help participants get started on their own NetLogo models, specifically focusing on the use of the Network extension. We emphasize providing support to new and current users for effectively integrating NetLogo in their research.  The workshop will consist of two 3 hours sessions. To allow for sufficient room for interactions we impose a maximum number 16 participants to each session.

Link to Conference Information: http://insna.org/nasn2017/

View Event →
Jun
29
2:00pm 2:00pm

Symposium: The State of LGBTQ Youth Health and Wellbeing: Strengthening Schools and Families to Build Resilience

Be sure to RSVP here: bit.ly/2qIonGX

We are excited to invite you to our second Annual Symposium entitled “The State of LGBTQ Youth Health and Wellbeing: Strengthening Schools and Families to Build Resilience.” Organized by ISGMH, the Center for Prevention Implementation MethodologyAdvocates for Youth, and the AIDS Foundation of Chicago, the symposium will be held on June 29th from 2:00-4:30 with a reception to follow. This event is supported by the Illinois Safe Schools Alliance.

This event will be live streamed. If you wish to attend this event remotely, please visit https://bluejeans.com/579239263  

Keynote SpeakerDr. David Purcell, JD, Ph.D., Deputy Director for Behavioral and Social Science, Center for Disease Control and Prevention

Panel Presenters:

Dr. Guillermo (Willy) Prado, Leonard M. Miller Professor of Public Health Sciences, Director of the Division of Prevention Science and Community Health at the Miller School of Medicine, and Dean of the Graduate School, University of Miami

Dr. Dorothy Espelage, Ph.D., Professor of Psychology, University of Florida.

Dr. Brian Mustanski, Ph.D., Director of the Institute for Sexual and Gender Minority Health and Wellbeing, Professor of Medical Social Sciences, Psychiatry and Behavioral Sciences and Weinberg College of Arts and Sciences, Northwestern University

Youth leaders from the Illinois Caucus for Adolescent Health

Remarks:

Dr. Karen Parker, Ph.D., M.S.W., Director of the Sexual & Gender Minority Research Office, National Institutes of Health

Debra Hauser, President, Advocates for Youth

View Event →
Jun
23
9:00am 9:00am

Meeting: Center for Translational and Implementation Research (CTRIS) Meeting @ National Heart, Lung, and Blood Institute (NHLBI)

Meeting Topic:
Eliminating Cardiovascular Disparities through Community Engaged Research

Methods and Milestones Session:
Session Chairs are Dr. Hendricks Brown & Dr. JD Smith of Ce-PIM

For more information about the Center for Translational Science & Implementation Research and the National Heart, Lung, and Blood Institute, click here: https://www.nhlbi.nih.gov/about/org/ctris

 

View Event →
May
23
12:00pm12:00pm

Chen-Pin Wang: Assessing causal inference and disparity in the latent variable prediction framework

Chen-Pin Wang, Ph.D.
University of Texas-San Antonio
05/23/2017

Latent Growth Mixture Modeling (LGMM) is a useful statistical tool to characterize the heterogeneity of the longitudinal development of a prognostic variable using the so-called latent classes. Recently an advanced statistical learning methodology (Jo 2016) was developed to validate the scientific utility of the latent classes regarding the prediction of a target outcome of interest. This presentation focuses on deriving causal inference and health disparity in this prediction model framework. The proposed method involves LGMM analysis of the prognostic variable, validating the prediction of the latent classes for the distal (future) outcome of interest, and then incorporating the inverse propensity score weighting technique to deriving causal relationship between the prognostics classes with the distal outcome and the associated health disparity. I will demonstrate the proposed method using a longitudinal epidemiology study of patients with type 2 diabetes that aimed at assessing the prediction of glycemic control for cardiovascular diseases related hospitalization and the racial disparity in this relationship.

View Event →
May
16
12:00pm12:00pm

Booil Jo: Statistical learning with latent prediction targets

Booil Jo, Ph.D.
Stanford University
05/16/2017

In predictive modeling, often aided by machine learning methods, much effort is concentrated in identifying good predictors. However, the same level of rigor is often absent in improving the outcome side of models. In this study, we focus on this rather neglected aspect of model development and demonstrate the use of longitudinal information as a way of improving the outcome side of predictive models. This involves optimally characterizing individuals’ outcome status, classifying them, and validating the formulated prediction targets. None of these tasks are straightforward, which may explain why longitudinal prediction targets are not commonly used in practice despite their compelling benefits. As a practical way of improving this situation, we explore a semi-supervised learning approach based on growth mixture modeling (GMM), a method of identifying latent subpopulations that manifest heterogeneous outcome trajectories. In the proposed approach, we utilize the benefits of the conventional use of GMM for the purpose of generating potential candidate models based on empirical model fitting, which can be viewed as unsupervised learning. We then evaluate candidate GMM models on the basis of a direct measure of success; how well the trajectory types are predicted by clinically and demographically relevant baseline features, which can be viewed as supervised learning. We examine the proposed approach focusing on a particular utility of latent trajectory classes, as outcomes that can be used as valid prediction targets in clinical prognostic models. Our approach is illustrated using data from the Longitudinal Assessment of Manic Symptoms study.

Suggested readings:

Jo B, Findling RL, Wang C-P, Hastie JT & the LAMS group (2017). Targeted use of growth mixture modeling: A learning perspective. Statistics in Medicine, 36, 671-686.

Jo B, Findling RL, Hastie JT, Youngstrom EA, Wang C-P & the LAMS group (in press). Construction of longitudinal prediction targets using semi-supervised learning. Statistical Methods in Medical Research.

View Event →
May
2
12:00pm12:00pm

Teppei Yamamoto: Introduction to causal mediation analysis using R

Teppei Yamamoto, Ph.D.
MIT
05/02/2017

Researchers often seek to study not only whether a treatment has a causal effect on an outcome but also how and why such a causal relationship comes about. Causal mediation analysis is a popular method to analyze causal mechanisms using experimental or observational data. In this webinar, we provide an overview of the theoretical framework underpinning the mediation methods and discuss assumptions that play a key role for valid inference about causal mechanisms. We also cover practical issues in using the framework for social, behavioral and medical science applications. A particular focus will be on the R package mediation and how to use it in various applied setting.

View Event →
Apr
25
12:00pm12:00pm

Naihua Duan: Personalized biostatistics, small data, and N-of-1 trials

Naihua Duan, Ph.D.
Columbia University
04/25/2017

As personalized medicine is emerging as a promising paradigm to improve clinical decision-making, to customize clinical decisions for individual patients, to accommodate the unique needs and preferences for each specific patient, there is a growing need for biostatistical methods to be developed and deployed to serve the needs for this emerging paradigm. As an example, the on-going NINR-funded PREEMPT Study, http://www.ucdmc.ucdavis.edu/chpr/preempt/, has developed a smartphone app that allows chronic pain patients and clinicians to run personalized experiments (n-of-1 trials), comparing two different pain treatments, to help patients and their clinicians to choose the most appropriate pain treatment for each individual patient. Such personalized biostatistical toolkits can be utilized by frontline clinicians and their patients to address the specific clinical questions confronted by each individual patient, to enable the specification and execution of the personalized trial protocol, to facilitate the collection of outcome and process data, to analyze and interpret the data acquired, and to produce reports to the end users to help them with evidence-based decision making. This paradigm exemplifies the potential for “Small Data” (as opposed to “Big Data”) to be deployed in clinical applications for the benefits of today’s patients as the primary aim for the evidence-based investigations. Importantly, personalized biostatistics based on Small Data provides strong incentives for end-users to participate in the evidence-based investigations, as they are targeted to benefit directly from such investigations, instead of the traditional clinical research that aims primarily to benefit future patients. There is a promising potential this merging paradigm will play an important role in the future of health and related domains, to supplement the traditional top-down model for research with a bottom-up model for quality improvement.

View Event →
Apr
18
12:00pm12:00pm

Min Lu: Estimating individual treatment effect in observational data using random forest methods

Min Lu, Doctoral student
University of Miami
04/18/2017

Estimation of individual treatment effect in observational data is complicated due to the challenges of confounding and selection bias. A useful inferential framework to address this is the counterfactual model which takes the hypothetical stance of asking what if an individual had received both treatments. Making use of random forests (RF) within the counterfactual framework we estimate individual treatment effects by directly modeling the response. We find accurate estimation of individual treatment effects is possible even in complex heterogeneous settings but that the type of RF approach plays an important role in accuracy. Methods designed to be adaptive to confounding, when used in parallel with out-of-sample estimation, do best. One method found to be especially promising is counterfactual synthetic forests. We illustrate this new methodology by applying it to a large comparative effectiveness trial, Project Aware, in order to explore the role drug use plays in sexual risk. The analysis reveals important connections between risky behavior, drug usage, and sexual risk.

View Event →
Apr
11
12:00pm12:00pm

Tihomir Asparouhov: Dynamic structural equation modeling of intensive longitudinal data using Mplus version 8 (Part 4)

Tihomir Asparouhov
04/11/2017

*** For additional Mplus resources, please visit https://www.statmodel.com/ ***

Mplus Version 8 features new methods for analyzing intensive longitudinal data such as that obtained with ecological momentary assessments, experience sampling methods, daily diary methods, and ambulatory assessments. Typically, such data have a large number of time points, T = 20-150, and may come from a single case (e.g., person or dyad), or multiple cases.In this first of a series of three talks, we will begin by explaining how intensive longitudinal data differs from other forms of data, such as cross-sectional and traditional panel data. We will argue that a particularly valuable feature of intensive longitudinal data is that they allows us to study the dynamics of processes over time. This can be done by using time series analysis, a class of statistical techniques that is very popular in fields like econometrics. However, time series analysis is typically restricted to single case data. Dynamic structural equation modeling (DSEM) as it is now developed and implemented in Mplus version 8 allows for the analysis of time series data from a single case, but also allows for multilevel extensions of these models, such that quantitative differences in the dynamics of different cases can be accounted for by the inclusion of random effects. We will present two empirical applications to illustrate the modeling opportunities that DSEM offers.

Part 2 on DSEM in Mplus Version 8 builds on Ellen Hamaker’s Part 1 introduction and describes further DSEM analysis possibilities. Advantages of the Bayesian analysis used for DSEM over ML are briefly discussed. To put the new modeling features in a familiar context, the shortcomings of conventional growth modeling approaches using wide and long format for intensive longitudinal data applications are contrasted with the flexibility of multilevel time series modeling for these data. A series of increasingly more general models are explored using data from a smoking cessation study with N=200 and T=100 (five random measurement occasions per day for 20 days). The starting point is a two-level model with random effects as in Hamaker’s presentation. Cross-classified time series analysis is then introduced where subject and time are crossed resulting in more flexible modeling than possible with two-level analysis in that the random effects are allowed to vary across not only subjects but also time. The cross-classified analysis is used to identify which random effects vary across time and to get an idea of time trends. Trend analysis is then added to the two-level time series modeling in a general time series version of the conventional two-level, long format approach. The advantages of the cross-classified analysis as compared to Time-Varying Effect Modeling (TVEM) are discussed. Throughout the presentation, new time series plots are shown. The modeling part of the talk concludes with a brief overview of two-level and cross-classified DSEM factor analysis with random measurement intercepts and random loadings.

In closing, new non-time series features in Mplus Version 8 are mentioned, including two-level modeling with random variances; two-level random autocorrelation modeling for short longitudinal data; standardization for two-level models with random slopes and random variances; random slopes for covariates with missing data; new within/between scatter plots and histograms for two-level models, including sample and model-estimated cluster-specific means and variances; new Posterior Predictive P-values for BSEM; and output in HTML format.

View Event →
Apr
4
12:00pm12:00pm

Tyler VanderWeele: Conceptual foundations for selecting optimal subgroups for treatment

Tyler VanderWeele
Harvard University
04/03/2017

What data are relevant when making a treatment decision for me? What replications are relevant for quantifying the uncertainty of this personalized decision? What does “relevant” even mean here? The multi-resolution (MR) perspective from the wavelets literature provides a convenient theoretical framework for contemplating such questions. Within the MR framework, signal and noise are two sides of the same coin: variation. They differ only in the resolution of that variation—a threshold, the primary resolution, divides them. We use observed variations at or below the primary resolution (signal) to estimate a model and those above the primary resolution (noise) to estimate our uncertainty. The higher the primary resolution, the more relevant our model is for predicting a personalized response. The search for the appropriate primary resolution is a quest for an age old bias-variance trade-off: estimating more precisely a less relevant treatment decision versus estimating less precisely a more relevant one. However, the MR setup crystallizes how the tradeoff depends on three objects: (i) the estimand which is independent of any statistical model, (ii) a model which links the estimand to the data, and (iii) the estimator of the model. This trivial, yet often overlooked distinction, between estimand, model, and estimator, supplies surprising new ways to improve mean squared error. The MR framework also permits a conceptual journey into the counterfactual world as the resolution level approaches infinite, where “me” becomes unique and hence can only be given a single treatment, necessitating the potential outcome setup. A real-life Simpson’s paradox involving two kidney stone treatments will be used to illustrate these points and engage the audience.

View Event →
Apr
4
12:00pm12:00pm

Xiao-Li Meng: Building a statistical theory for individualized treatments: A mulit-resolution perspective

Xiao-Li Meng, Ph.D.
Harvard University
04/04/2017

What data are relevant when making a treatment decision for me? What replications are relevant for quantifying the uncertainty of this personalized decision? What does “relevant” even mean here? The multi-resolution (MR) perspective from the wavelets literature provides a convenient theoretical framework for contemplating such questions. Within the MR framework, signal and noise are two sides of the same coin: variation. They differ only in the resolution of that variation—a threshold, the primary resolution, divides them. We use observed variations at or below the primary resolution (signal) to estimate a model and those above the primary resolution (noise) to estimate our uncertainty. The higher the primary resolution, the more relevant our model is for predicting a personalized response. The search for the appropriate primary resolution is a quest for an age old bias-variance trade-off: estimating more precisely a less relevant treatment decision versus estimating less precisely a more relevant one. However, the MR setup crystallizes how the tradeoff depends on three objects: (i) the estimand which is independent of any statistical model, (ii) a model which links the estimand to the data, and (iii) the estimator of the model. This trivial, yet often overlooked distinction, between estimand, model, and estimator, supplies surprising new ways to improve mean squared error. The MR framework also permits a conceptual journey into the counterfactual world as the resolution level approaches infinite, where “me” becomes unique and hence can only be given a single treatment, necessitating the potential outcome setup. A real-life Simpson’s paradox involving two kidney stone treatments will be used to illustrate these points and engage the audience.

View Event →
Mar
28
12:00pm12:00pm

Bengt Muthén: Dynamic structural equation modeling of intensive longitudinal data using Mplus version 8 (Part 3)

Bengt Muthén, Ph.D.
UCLA
03/28/2017

*** For additional Mplus resources, please visit https://www.statmodel.com/ ***

Mplus Version 8 features new methods for analyzing intensive longitudinal data such as that obtained with ecological momentary assessments, experience sampling methods, daily diary methods, and ambulatory assessments. Typically, such data have a large number of time points, T = 20-150, and may come from a single case (e.g., person or dyad), or multiple cases.In this first of a series of three talks, we will begin by explaining how intensive longitudinal data differs from other forms of data, such as cross-sectional and traditional panel data. We will argue that a particularly valuable feature of intensive longitudinal data is that they allows us to study the dynamics of processes over time. This can be done by using time series analysis, a class of statistical techniques that is very popular in fields like econometrics. However, time series analysis is typically restricted to single case data.Dynamic structural equation modeling (DSEM) as it is now developed and implemented in Mplus version 8 allows for the analysis of time series data from a single case, but also allows for multilevel extensions of these models, such that quantitative differences in the dynamics of different cases can be accounted for by the inclusion of random effects. We will present two empirical applications to illustrate the modeling opportunities that DSEM offers.

Part 2 on DSEM in Mplus Version 8 builds on Ellen Hamaker’s Part 1 introduction and describes further DSEM analysis possibilities. Advantages of the Bayesian analysis used for DSEM over ML are briefly discussed. To put the new modeling features in a familiar context, the shortcomings of conventional growth modeling approaches using wide and long format for intensive longitudinal data applications are contrasted with the flexibility of multilevel time series modeling for these data. A series of increasingly more general models are explored using data from a smoking cessation study with N=200 and T=100 (five random measurement occasions per day for 20 days). The starting point is a two-level model with random effects as in Hamaker’s presentation. Cross-classified time series analysis is then introduced where subject and time are crossed resulting in more flexible modeling than possible with two-level analysis in that the random effects are allowed to vary across not only subjects but also time. The cross-classified analysis is used to identify which random effects vary across time and to get an idea of time trends. Trend analysis is then added to the two-level time series modeling in a general time series version of the conventional two-level, long format approach. The advantages of the cross-classified analysis as compared to Time-Varying Effect Modeling (TVEM) are discussed. Throughout the presentation, new time series plots are shown. The modeling part of the talk concludes with a brief overview of two-level and cross-classified DSEM factor analysis with random measurement intercepts and random loadings.

In closing, new non-time series features in Mplus Version 8 are mentioned, including two-level modeling with random variances; two-level random autocorrelation modeling for short longitudinal data; standardization for two-level models with random slopes and random variances; random slopes for covariates with missing data; new within/between scatter plots and histograms for two-level models, including sample and model-estimated cluster-specific means and variances; new Posterior Predictive P-values for BSEM; and output in HTML format.

View Event →
Mar
21
12:00pm12:00pm

Bengt Muthén: Dynamic structural equation modeling of intensive longitudinal data using Mplus version 8 (Part 2)

Bengt Muthén
UCLA
03/21/2017

*** For additional Mplus resources, please visit https://www.statmodel.com/ ***

Mplus Version 8 features new methods for analyzing intensive longitudinal data such as that obtained with ecological momentary assessments, experience sampling methods, daily diary methods, and ambulatory assessments. Typically, such data have a large number of time points, T = 20-150, and may come from a single case (e.g., person or dyad), or multiple cases.In this first of a series of three talks, we will begin by explaining how intensive longitudinal data differs from other forms of data, such as cross-sectional and traditional panel data. We will argue that a particularly valuable feature of intensive longitudinal data is that they allows us to study the dynamics of processes over time. This can be done by using time series analysis, a class of statistical techniques that is very popular in fields like econometrics. However, time series analysis is typically restricted to single case data.Dynamic structural equation modeling (DSEM) as it is now developed and implemented in Mplus version 8 allows for the analysis of time series data from a single case, but also allows for multilevel extensions of these models, such that quantitative differences in the dynamics of different cases can be accounted for by the inclusion of random effects. We will present two empirical applications to illustrate the modeling opportunities that DSEM offers.

Part 2 on DSEM in Mplus Version 8 builds on Ellen Hamaker’s Part 1 introduction and describes further DSEM analysis possibilities. Advantages of the Bayesian analysis used for DSEM over ML are briefly discussed. To put the new modeling features in a familiar context, the shortcomings of conventional growth modeling approaches using wide and long format for intensive longitudinal data applications are contrasted with the flexibility of multilevel time series modeling for these data. A series of increasingly more general models are explored using data from a smoking cessation study with N=200 and T=100 (five random measurement occasions per day for 20 days). The starting point is a two-level model with random effects as in Hamaker’s presentation. Cross-classified time series analysis is then introduced where subject and time are crossed resulting in more flexible modeling than possible with two-level analysis in that the random effects are allowed to vary across not only subjects but also time. The cross-classified analysis is used to identify which random effects vary across time and to get an idea of time trends. Trend analysis is then added to the two-level time series modeling in a general time series version of the conventional two-level, long format approach. The advantages of the cross-classified analysis as compared to Time-Varying Effect Modeling (TVEM) are discussed. Throughout the presentation, new time series plots are shown. The modeling part of the talk concludes with a brief overview of two-level and cross-classified DSEM factor analysis with random measurement intercepts and random loadings.

In closing, new non-time series features in Mplus Version 8 are mentioned, including two-level modeling with random variances; two-level random autocorrelation modeling for short longitudinal data; standardization for two-level models with random slopes and random variances; random slopes for covariates with missing data; new within/between scatter plots and histograms for two-level models, including sample and model-estimated cluster-specific means and variances; new Posterior Predictive P-values for BSEM; and output in HTML format.

View Event →
Mar
14
12:00pm12:00pm

Ellen Hamaker: Dynamic structural equation modeling of intensive longitudinal data using Mplus version 8 (Part 1)

Ellen Hamaker, Ph.D.
Universiteit Utrecht
03/14/2017

*** For additional Mplus resources, please visit https://www.statmodel.com/ ***

Mplus Version 8 features new methods for analyzing intensive longitudinal data such as that obtained with ecological momentary assessments, experience sampling methods, daily diary methods, and ambulatory assessments. Typically, such data have a large number of time points, T = 20-150, and may come from a single case (e.g., person or dyad), or multiple cases.In this first of a series of three talks, we will begin by explaining how intensive longitudinal data differs from other forms of data, such as cross-sectional and traditional panel data. We will argue that a particularly valuable feature of intensive longitudinal data is that they allows us to study the dynamics of processes over time. This can be done by using time series analysis, a class of statistical techniques that is very popular in fields like econometrics. However, time series analysis is typically restricted to single case data. Dynamic structural equation modeling (DSEM) as it is now developed and implemented in Mplus version 8 allows for the analysis of time series data from a single case, but also allows for multilevel extensions of these models, such that quantitative differences in the dynamics of different cases can be accounted for by the inclusion of random effects. We will present two empirical applications to illustrate the modeling opportunities that DSEM offers.

Part 2 on DSEM in Mplus Version 8 builds on Ellen Hamaker’s Part 1 introduction and describes further DSEM analysis possibilities. Advantages of the Bayesian analysis used for DSEM over ML are briefly discussed. To put the new modeling features in a familiar context, the shortcomings of conventional growth modeling approaches using wide and long format for intensive longitudinal data applications are contrasted with the flexibility of multilevel time series modeling for these data. A series of increasingly more general models are explored using data from a smoking cessation study with N=200 and T=100 (five random measurement occasions per day for 20 days). The starting point is a two-level model with random effects as in Hamaker’s presentation. Cross-classified time series analysis is then introduced where subject and time are crossed resulting in more flexible modeling than possible with two-level analysis in that the random effects are allowed to vary across not only subjects but also time. The cross-classified analysis is used to identify which random effects vary across time and to get an idea of time trends. Trend analysis is then added to the two-level time series modeling in a general time series version of the conventional two-level, long format approach. The advantages of the cross-classified analysis as compared to Time-Varying Effect Modeling (TVEM) are discussed. Throughout the presentation, new time series plots are shown. The modeling part of the talk concludes with a brief overview of two-level and cross-classified DSEM factor analysis with random measurement intercepts and random loadings.

In closing, new non-time series features in Mplus Version 8 are mentioned, including two-level modeling with random variances; two-level random autocorrelation modeling for short longitudinal data; standardization for two-level models with random slopes and random variances; random slopes for covariates with missing data; new within/between scatter plots and histograms for two-level models, including sample and model-estimated cluster-specific means and variances; new Posterior Predictive P-values for BSEM; and output in HTML format.

View Event →
Feb
28
12:00pm12:00pm

Otto Koppius: Prediction vs. explanation in statistical model building

Otto Koppius, Ph.D.
Erasmus University
02/28/2017

Predictive modeling, where one tries to predict for instance the outcome of a particular process or the occurrence of a certain event, is common in many research areas. In medical research, predictive models (often called prognostic models) are for instance used to predict patient outcomes or the effect of certain treatments. A common practice is using the best explanatory statistical model for prognostic purposes, i.e. the model with significant coefficients for the independent variables. While common, this is incorrect for prognostic purposes, as the best explanatory model is almost always different from the best prognostic model for a number of reasons, which I will describe in this talk. Furthermore, the process of building a predictive model is fundamentally different from building an explanatory model, as differences occur in every step of the modeling process. Moreover, predictive models have additional roles to play alongside explanatory models in theory building and theory testing, such as new theory generation, measurement development, comparison of competing theories, improvement of the conceptual structure of existing models, relevance assessment, and assessment of the predictability of empirical phenomena. I will illustrate the differences between the modeling approaches with examples from the literature on adoption of new health technologies and from diffusion over networks.

View Event →
Feb
7
12:00pm12:00pm

Wouter Vermeer: The impact of the propagation mechanism on system wide dynamics

Wouter Vermeer, Ph.D.
Northwestern University
02/07/2017

Often systems exhibit behavior which is difficult to predict and steer, as interactions on the micro level (between actors within the system) results in propagation of behavior which can cause unforeseen dynamics. Understanding the effects of propagation is crucial in order to understand how the state and behavior of a system will change and what the effect of interventions will be. The structure of interactions, the network structure, has been identified in literature as a prime driver of propagation. By focusing on the network structure, however, the impact of the mechanism by which propagation takes place has been pushed to the background. In this presentation I argue that the mechanism plays a crucial role in determining how propagation dynamics scale, and is thus critical when effectively intervening in a system. While various mechanisms exist in literature there is little coherence and overlap in their use, therefore I put forward a generic Agent-Based framework which allows for capturing the propagation mechanism in more detail and in a structured way. I show that using such a framework not only results in a more detailed and methodologically stronger model of propagation, but also that it is a prerequisite for effective interventions into propagation.

View Event →