SFOepi seminars and workshops

2023

Causal inference with counting processes: Graphical identifiability rules and efficient estimation.

Speaker: Kjetil Røysland 

December 13, 10.15-11.15, Petrén, Nobels väg 12 B 

No registration needed.

Abstract: We consider continuous-time survival or more general event-history settings, where our aim is to graphically represent causal structures allowing us to answer queries such as when a causal parameter is identified from observational data. This causal parameter is formalised as the effect on an outcome event of a (possibly hypothetical) intervention on the intensity of a treatment process, i.e. a stochastic intervention. To establish identifiability, i.e. whether valid inference about the interventional situation can be drawn from typical observational (non-experimental) data, we propose novel graphical rules indicating whether the observed information is sufficient to obtain the desired causal effect by suitable re-weighting. This requires a different type of graph than in discrete time. We formally define causal semantics for the corresponding dynamic graphs that represent local independence models for multivariate counting processes. We propose a new method to  calculate efficient influence functions for various  causal parameters in these models, and then construct corresponding efficient estimators based on techniques from stochastic differential equations.

This is joint work with Vanessa Didelez and Pål C. Ryalen.

Dismantling the complexity of the evidence for healthcare interventions using network meta-analysis

Speaker: Anna Chaimani

Time: Nov 28 at 10:00 in Rockefeller lecture hall
 

STRATOS guidance for the selection of variables and functional forms – outstanding issues and the multivariable fractional polynomial approach

Speaker: Willi Sauerbrei, senior statistician and professor in medical biometry at the Center for Medical Biometry and Medical Informatics, Medical Center - University of Freiburg, Germany

Time: Nov 7 at 10:00 in lecture hall Peter Reichard, Biomedicum 

Survival analysis of cancer screening programmes

Speaker: Maja Pohar Perme

Time: Oct 31 at 13:30 in lecture hall Ragnar Granit, Biomedicum

IPTW plus adjusted outcome models does not always equal doubly robust

Speaker: Erin Gabriel, Associate Professor in Biostatistics, University of Copenhagen

Time: Oct 26 at 10:15 in Rockefeller lecture hall

2022

Statistical Methods Week

Time: June 7-10, 9:00-16:30

The strategic research area in epidemiology and biostatistics and the statistical methods node of the national network for register-based research invite all interested in statistical methods for the analysis of data to attend the seminars.The goal is for researchers in all disciplines to interact, discuss new statistical methods, and discover the wealth of methods that are being developed. The tutorials are intended for biomedical researchers or practitioners with research in or using mathematics, biostatistics, and computational statistics. Link to the full program and information on how to sign up here.  

Seminars:

• Dr. Jason Liang - "Prognostic accuracy measures for survival models"
• Dr. Brice Ozenne -  "Toward a unified framework for analyzing repeated measurements with a continuous outcome"
• Professor Hongwei Zhao - "An improved survival estimator for censored medical costs using kernel methods"
• Professor Ken Rice - "Fixing fixed-effects meta-analysis: some theoretical and practical advances"
• Professor Andrea Rotnitzky - "Towards deriving graphical rules for efficient estimation in causal graphical models"
• Professor Karla Diaz-Ordaz  - "Causal machine learning for heterogeneous treatment effects"
• Dr. Tim Morris - "Nonparametric bootstrapping for standard errors and confidence intervals: silver bullet or fool’s gold?"
• Dr. Michael Sachs - "Recent advances in regression modeling of censored time-to-event outcomes using pseudo-observations"

Tutorials:                                                                                          

• "Prognostic accuracy measures for survival with Covid-19 example" in R with Dr. Liang.
• "Analysis of repeated measurements with mixed models using the R package LMMstar" with Dr. Ozenne and Dr. Forman
• "Analysis of medical costs with censored data" In Stata with Professor Zhao.
• "Introduction to meta-analysis" in R with Professor Rice.
• "Introduction to causal machine learning" in R with Professor Diaz-Ordaz.
• "Nonparametric bootstrap" in Stata with Dr. Morris. 
• "Using the eventglm R package for regression modeling of censored time-to-event outcomes" with Dr. Sachs. 

Workshop in Advanced Quantitative Genetics Using Twin and Family Data

Instructors: Ralf Kuja-Halkola, PhD, Ass. Professor, MEB, KI, Mark Taylor, Principal researcher, MEB, Cen Chen, PhD student, MEB.

Time: June 7-9 in Campus Solna

SFOepi workshop: The prevention of multimorbidity: from primary to quaternary

Speakers: Anna Head, Martin O’Flaherty, Davide Vetrano, Amaia Calderón-Larrañaga, Caroline Wachtler, Katharina Schmidt Mende

Time: from 9:00, May 18 until 12:30, May 19 in  lecture hall Karolina Hall, Widerströmska huset and on Zoom

2021

World-wide trends and inequalities in cancer survival and their application to public health policy

Speakers: Michel Coleman, BA BM BCh MSc FFPH, Professor of Epidemiology and Vital Statistics, Head of the Cancer Survival Group, London School of Hygiene & Tropical Medicine and Claudia Allemani, MSc PhD FHEA MFPH, Professor of Global Public Health, London School of Hygiene & Tropical Medicine

Time: Nov 25 at 9:00 on zoom

Development of a multiparametric MR classifier for prostate cancer

Speaker: Joseph Koopmeiners, Mayo Professor and Head, Division of Biostatistics, School of Public Health, University of Minnesota

Time: Nov 19 at 16:00 on zoom

Causal effects conditional on post-treatment events

Speaker: Mats J. Stensrud, MD PhD, Assistant professor of statistics at Swiss Federal Institute of Technology Lausanne

Time: Oct 13 at 09:30 on zoom

Sfo Epi Seminar in Biostatistics: Distributed Privacy-Preserving Estimation of Time-to-Event Models

Speaker: Dr. C Jason Liang PhD, Biostatistics Research Branch, NIAID, NIH, USA

Time: September 17 at 15:00 on zoom

Reconsideration of the Kaplan-Meier Estimator:  Censoring and Time-varying Covariates

Time: June 11 at 15:00 on zoom
Speaker: Professor Rebecca Betensky, Chair of the Department of Biostatistics, School of Global Public Health at New York University

Using EHR to Extend Understanding of Treatment Effectiveness Beyond Clinical Trials

Speakers: Professor Rebecca Hubbard PhD, Division of Biostatistics, Perelman School of Medicine, University of Pennsylvania and Dr. Ronac Mamtani MD, Assistant Professor, Division of Hematology/Oncology, Abramson Cancer Center, Perelman School of Medicine, University of Pennsylvania 

Time: May 10 at 15:30 on zoom

New Approaches for Inference on Optimal Treatment Regimes

Speaker: Professor Lan Wang, Department of Management Science, University of Miami

Time: 16-04-2021 at 15:00 on zoom

Panel on Artificial Intelligence in Epidemiology

Panelists: 

• Dr. Mattias Rantalainen, Senior Lecturer of Epidemiology at the Department of Medical Epidemiology Karolinska Institutet.
• Dr. Sema Sgaier, Co-Founder and Executive Director of Surgo Foundation, Adjunct Assistant Professor of Global Health, Harvard T.H. Chan School of Public Health, and  Adjunct Assistant Professor of Global Health University of Washington. 
• Dr. Andrea Ganna, FIMM-EMBL group leader, Institute for Molecular Medicine, University of Helsinki, Finland and Instructor, Harvard Medical School.
• Dr. Christian Guttman, Vice president, global head of Artificial Intelligence and Chief Artificial Intelligence and Data officer at TietoEVRY, adjunct associate professor at the University of New South Wales, Australia, and Adjunct researcher at Department of Learning, Informatics, Management, and Ethics and the Karolinska Institutet.
• Chair: Dr. Elizabeth Arkema, Clinical Epidemiology Unit, Department of Medicine, Karolinska Institutet.

Details: February 12 at 15:00 on zoom

Target Trial Emulation to Study the Effects of Treatments and Vaccines During Pregnancy

Speaker: Sonia Hernández-Díaz, Professor of Epidemiology, Department of Epidemiology Harvard T.C. Chan school of public health

Time: February 5 at 15:00 on zoom

Impacts of Biostatistics on Advancing Human Health in the 20th Century and Beyond: My Personal View 

Speaker: President Kung-Yee Liang, Institute of Population Health Sciences, National Health Research Institutes, Taiwan, R.O.C., Professor of Biostatistics, Johns Hopkins University, Department of Biostatistics 

Time: January 29, at 9:00 am zoom

2020

Statistical models for the natural history of breast cancer: likelihood and likelihood-free estimation

Speaker: Professor Marco Bonetti, Bocconi University, Milan, Italy 

Time:  November 6 at 11:00 on zoom

A reevaluation of dementia incidence trends in the Framingham Heart Study cohort

Speaker: Dr Nadine Binder, Senior researcher, Universitätsklinikum Freiburg, Institut für Digitalisierung in der Medizin

Time: October 16 at 11:00 on zoom

Pairwise composite likelihood for mixed models under complex sampling 

Speaker: Professor Thomas Lumley, Chair in Biostatistics at The University of Auckland, New Zealand

Time: September 25 at 9:00 on zoom

Machine Learning for Causal Inference: review of recent developments

Speaker: Dr. Karla Diaz-Ordaz, associate professor of biostatistics, the London School of Hygiene and Tropical Medicine

Time:  August 28 at 11:00 on zoom

SFOepi Epidemiology and Biostatistics Day 2020 (cancelled due to travel restrictions)

Time: May 8, 2020, 09:00-18:00

Location: Aula Medica

Hot Topics in Epidemiology and Biostatistics  - a full-day conference with confirmed Keynote Speakers Professor Sonia Hernandez-Diaz and President of Taiwan’s National Health Research Institutes, Professor Kung-Yee Liang 

2019

Epi-Biostat Expo 2019: Sparking collaborations

Time: Nov 26, 09:00-15:00

Location: Biomedicum, Karolinska Institutet, Solnavägen 9

Speakers: Petter Ljungman, Elisabeth Arkama, Davide Vetrano, Erin Gabriel, Gaetano Marrone

Causal mediation with longitudinal mediator and survival outcome

Time: Nov 15, at 11:00

Location: Strix, von Eulers väg 4, Solna

Professor Vanessa Didelez, Leibniz Institute for Prevention Research and Epidemiology - BIPS, Bremen, Germany

Host: Erin Gabriel

When is a statistical method fit for use?

Time: Nov 8, at 11:00

Location: Strix, von Eulers väg 4, Solna

Dr Tim Morris, MRC Clinical Trials Unit at UCL
Host: Erin Gabriel

Career Day for Epidemiologists 2019

Time: May 9, at 11.30
Location: Biomedicum1, Biomedicum, Solna
Host: Camilla Wiklund

A Bayesian adaptive comparative effectiveness trial - an example in Status Epilepticus - CANCELLED

Time: May 3, at 9.30
Location: Strix, von Eulers väg 4, Solna
Speakers: Dr. Jason Connor
Host: Erin Gabriel

Introduction to Mendelian Randomization

Time: May 2 at 9.00-11.45
Location: Biomedicum 1, Biomedicum, Solna
Speakers: Arvid Sjölander and Sara Hägg
Host: Elisabeth Dahlqwist and Arvid Sjölander

Pleiotropy robust Mendelian Randomization Workshop

Time: May 3 at 9.00-17.30
Location: Biomedicum 1, Biomedicum, Solna
Speakers: Jack Bowden and Wes Spiller from the University of Bristol
Host: Elisabeth Dahlqwist and Arvid Sjölander

Polygenic risks in cardiometabolic diseases and breast and prostate cancers and their impact on behaviour

Time: April 5, at 11.00
Location: Strix von Eulers väg 4, Solna
Speakers: Professor Samuli Ripatti
Host: Erin Gabriel

Flexible parametric survival models in epidemiology: standard survival, competing risks and multistate mode

Time: April 3-5
Location: To be confirmed (Campus Nord, Solna)
Speakers: Dr Michael Crowther and Professor Paul Lambert
Host: Professor Paul Lambert

Towards Personalized Computer Simulation of Breast Cancer Treatment

Time: March 1, at 11.00
Location: Strix von Eulers väg 4, Solna
Speakers: Professor Arnoldo Frigessi
Host: Erin Gabriel

Course on Competing Risks and Multistate Models

Time: Februari 5-7, 9.00-17.00
Location: To be confirmed (Campus Nord, Solna)
Speakers: Dr. Ronald Geskus and Professor Hein Putter
Host: Sandra Eloranta, KEP, MedS
Registration: https://goo.gl/forms/LR6tuiZtV7eRJPdp1

2018

SfoEpi Research Seminar: Psychological stress, health, and potential biological mechanisms

Time: December 11, 9.30-16.00
Location: Biomedicum 3 Lecture Hall, Karolinska Institutet, Solna
Speakers: Professors Meena Kumari, Nina Alexander, Hugo Westerlund, Asya Rolls, Erica Sloan and Dr. Robert Miller
Moderators: Professors Mats Lekander and Unnur Valdimarsdóttir
Host: Fang Fang and Kelli Lehto, MEB, Karolinska Institutet

Climate change, Air Pollution and Health: The use of big data in epidemiology for causal inference and implications for affecting health policy

Time: November 29, 13.00-16.00
Location: Biomedicum 1 Lecture Hall, Karolinska Institutet, Solna
Speakers: Professor Joel Schwartz and Associate Professor Gregory Wellenius
Hosts: Petter Ljungman and Erik Melén, Karolinska Institutet

SFO-Epi workshop: Multimorbidity research at the cross-roads: developing the evidence for clinical practice and health policy

Time: May 21, 8.30-16.30
Location: Nobel Forum, Nobels väg 1, Karolinska Institutet, Solna
Speakers: Mary Tinetti, Martin Roland, Luigi Ferrucci, Stewart Mercer, Cynthia Boyd, Jose M. Valderas and Alessandra Marengoni
Host: Amaia Calderón-Larrañaga, Aging Research Center, Karolinska Institutet
Further information: http://www.multimorbidity2018-stockholm.se/

SFO-Epi workshop: How to prepare a Cochrane systematic review: Practical and methodological issues

Time: April 12, 08.30-16.00
Location: Andreas Vesalius, Berzelius väg 3, plan 2, Karolinska Institutet, Solna
Abstract: The scope of the Worskhop is to provide an introduction to the understanding of systematic reviews and meta-analyses following the Cochrane methodology, with both theoretical and practical prospective. In addition, to inspire and support researchers to become authors of systematic reviews and meta-analyses, presenting available resources and tools and national and international level.
Further information and registration: http://sweden.cochrane.org/how-prepare-cochrane-systematic-review-practical-and-methodological-issues

SFO-Epi workshop: Time series methods in health research

Time: April 12, 9.00-12.00
Location: Lecture hall Atrium, Nobels väg 12B, Karolinska Institutet, Solna
Register for the course using the following link: https://docs.google.com/forms/
Host: Petter Ljungman (e-mail: petter.ljungman@ki.se), Unit of Environmental Epidemiology, Institute of Environmental Medicine, Karolinska Institutet
The session will involve a mix of mini-lectures and mini-practicals on the various topics covered in the in workshop, including illustrative examples and real-data analyses. The session will cover:

• Introduction to study design and statistical models for time series analysis using R
• The use of interrupted time series as a quasi-experimental design: applications for the evaluation of public health interventions
• Time series models for the analysis of short-term effects: applications for investigating health risks of environmental factors
• More sophisticated techniques: two-stage designs and the use of distributed lag models
• Individual-level analyses with the novel case time series design: applications in clinical and pharmaco-epidemiological studies

Invited speakers: 
- Antonio Gasparrini, PhD, is an Associate Professor of Biostatistics and Epidemiology at the London School of Hygiene and Tropical Medicine (LSHTM). He is a leading expert in the development of study designs and statistical methods for time series analysis applied for environmental studies or public health evaluation. Antonio is the author of the R packages dlnm and mvmeta, and of several tutorials on the use of R for epidemiological analyses. He has taught R and time series analysis in MSc courses at LSHTM for several years.
- Ana Maria Vicedo-Cabrera, PhD, is an Assistant Professor in Environmental Epidemiology and Statistics at LSHTM, with special interests in Climate Change research. She is an experienced epidemiologist in time series analysis, mainly on studies on air pollution and temperature-related mortality. Ana has developed and applied routines in R for state-of-the-art modelling approaches in environmental epidemiological analyses.
- Francesco Sera, MSc, is a Research Fellow in Biostatistics and Epidemiology at the London School of Hygiene and Tropical Medicine (LSHTM). His research activity focuses on modelling data from observational and experimental biomedical studies. Currently, his interest lies on time series models in environmental epidemiology and in the development on multivariate meta-analytical statistical methods and software. Francesco is an experienced R user, programmer, and teacher.

SFO-Epi seminar: N-of-1 Trials for Making Personalized Treatment Decisions

Time: April 4- at 13:00-14:00
Location: Andreas Vesalius, Berzelius väg 3, plan 2, Karolinska Institutet, Solna
Speaker: Christopher Schmid; Professor of Biostatistics; Co-Director, Center for Evidence Synthesis in Health; Brown University School of Public Health
Please register here: http://bit.ly/2DqgxF8
Abstract: N-of-1 trials hold great promise for enabling patients to create personalized protocols to decide on medical treatments. Fundamentally, they are single-participant multiple-crossover studies for determining the relative comparative effectiveness of two or more treatments for one individual. An individual selects treatments and outcomes of interest, carries out the trial, and then makes a final treatment decision with or without a clinician based on results of the trial. Established in a clinical environment, an N-of-1 practice provides data on multiple trials from different patients. Such data can be combined using meta-analytic techniques to inform both individual and population treatment effects. When patients undertake trials with different treatments, the data form a treatment network and suggest use of network meta-analysis methods. This talk will discuss ongoing and completed clinical research projects using N-of-1 trials for chronic pain, atrial fibrillation, inflammatory bowel disease, fibromyalgia and attention deficit hyperactivity disorder. It will describe design and analytic challenges as well as unique aspects deriving from use of the N-of-1 design for personalized decision making such as defining treatments, presenting results, and assessing model assumptions and the pros and cons of combining information from different patients in order to provide a better estimate of each individual’s effect than from his or her own data alone.Register-based Randomized Clinical Trials – New Possibilities for Clinical Research

Time: February 7 at 15:00-18.00
Location: Leksellsalen, Eugeniahemmet, Karolinska University Hospital, Solna
Invited speakers: 
- Stefan James, Uppsala University: RRCT studies in cardiology using Swedeheart
- Torsten Olbers, University of Gothenburg: Obesity surgery RRCTs
- Martin Neovius, Karolinska Institutet: Economic outcomes in RA using Register Enriched RCTs
Host: Olof Stephansson

Past, present and future of cardiovascular disease prevention: applications of multi-state life tables

Time: January 30 at 15:00-16.00
Location: Samuelssonsalen, Karolinska Institutet
Invited speaker: Professor Oscar Franco, Department of Epidemiology, ERASMUS Medical University, Rotterdam, the Netherlands
Host: Karin Leander
Link to SfoEpi event

2017

Stress research seminar and networking event

Time: December 11 at 12:00-17.30
Location: Klara Strand, Klarabergsviadukten 90, Stockholm
Invited speakers:
- Prof. Unnur Valdimarsdóttir, University of Iceland, Iceland
- Prof. Mats Lekander, Stockholm Stress Center / Karolinska Institutet, Sweden
- Assoc. Prof. Jiong Li, Aarhus University, Denmark
- Assoc. Prof. Katja Fall, Örebro University, Sweden
Hosts: Fang Fang, Kelli Lehto, Mina Rydell, Bronwyn Brew

Lunch seminar at IMM: “What are they doing at MEB?”

Time: Monday January 30 at 12.00-13.00
Location: Bergendorff lecture hall, IMM (level 2)
Invited speakers:
- Catarina Almqvist; Twin studies and new methods to detect causality
- Kamila Czene; Update on breast cancer prognosis and key determinants
- Jonas Ludvigsson; Celiac disease prevention: state-of the-art approaches 2017 and beyond
- Weimin Ye; Understanding gastro-intestinal cancer development using molecular studies
Hosts: Sofia Carlsson, Rickard Ljung, Erik Melén, Karin Modig

2016

Advanced survival models for correlated data

Time: October 24-25 at 9.00-12.00 and 13.00-16.00
Location: Karolina. Tomtebodavägen 18A, Widerströmska huset
Invited Lecturer: Virginie Rondeau, INSERM (Bordeaux, France).
Aims:
1. Standard frailty models for recurrent and clustered data (models, estimation, dynamic prediction and illustrations)
2. Extensions with nested and additive frailty models (models, estimation and illustrations)
3. Extensions with joint frailty models for recurrent events and a longitudinal biomarker (models, estimation, dynamic prediction and illustrations)
4. Extensions with multivariate frailty models (models, estimation and illustrations)
Abstract: Simple shared frailty models have been largely developed and applied for recurrent or clustered survival data in the literature. However extensions of frailty models are less common in publications and are not well developed in classical softwares. We are aiming at filling this gap by considering extension of frailty models (as additive frailty models, nested frailty models or joint frailty models) and by presenting an implementation of these models using also the R package frailtypack. A particular interest will be devoted to joint frailty models in order to analysis jointly recurrent events such as cancer relapses and a dependent terminal event (death or lost to follow-up). Prediction tools associated with this package will be presented also.
The first part of this course will introduce general frailty models, the estimation methods and the research questions they may address. The second part of this course will be dedicated to the joint frailty models with illustration on real data. The estimation and the predictive dynamic tools that can be derived from them will be exposed, with methods to evaluate their performance.
Emphasis is given, via examples on real data, of the ability of extended frailty models to describe a very broad range of practical situations. Each concept will be illustrated through implementation of these models using the R package frailtypack.
Host: Nicola Orsini

2015

Current Status Data: Epidemiology, Avalanches and Screening

Time: December 10 at 14:00-15:00
Location: Lecture Hall Atrium, Nobels Väg 12A, Solna Campus
Speaker: Nicholas P. Jewell, Professor of Biostatistics & Statistics,University of California, Berkeley
Abstract: More than 800 people died from avalanches in Europe and North America over the six winters from 2003--2009. Avalanche survival curves describe the probability of survival as a function of burial time. These curves provide the basis for international recommendations for rescue and resuscitation and for the design of safety and rescue devices. However, estimation of such curves is complicated by the fact that the time of death is unknown for individuals who are not alive when rescued. Such data is formally referred to as current status data. I will review current status data analysis techniques, statistical ideas that have broad application to a variety of other data structures that arise naturally in epidemiology, demography, and economics. I will briefly discuss some new estimation problems related to current status measurements associated with grouped screening.
Hosts: Rino Bellocco (MEB), Nicola Orsini (IMM)

Consensus formation from observation of complex systems with limited intervention: Why statistics needs to be absorbed into epistemology

Time: Thursday, April 2 at 13.00-14.00
Location: Samuelssonsalen, Tomtebodavägen 6, Karolinska Institutet
Speaker: Sander Greenland,
Department of Epidemiology and Department of Statistics, University of California, Los Angeles, California
Hosted by: Nicola Orsini, Institute of Environmental Medicine, Karolinska Institutet.
Aim: Experienced epidemiologists recognize that current statistical formalisms are inadequate for inference in risk assessment and health policy; astute social scientists recognize analogously that those formalism are inadequate for inference in economic and social policy. The inference problems in these fields are examples of consensus formation from observation of complex systems with limited intervention (CFOCLI). CFOCLI is a conceptually hard problem being addressed only slowly the statistics profession, in part because addressing it requires an epistemology both far more broad and more detailed than found in conventional statistical theory. There are alternative inference systems under development in computer science which claim to address at least some of the more severe limits of current statistical formalisms; their ability to address CFOCLI is intriguing but as yet far from demonstrated. Some crucial elements for CFOCLI absent from most formalisms include coalescing inferences from multiple agents that each have severe cognitive biases. These biases may be viewed as priors that down-weight large regions of the parameter space to the point of no influence. Shared biases develop and can lead to disastrous loss when all agents severely down-weight a region that contains reality in its interior. Examples include the history of diet and health recommendations by academically-based societies, as well as the collapse of hedge funds run by econometric theories. Such examples underscore the need to incorporate concepts from antifragility as well as information and cognitive sciences into the core of CFOCLI.

2014

Three new SfoEpi awardees present their projects

Location: Lecture hall Wargentin, Nobels väg 12A, Karolinska Institutet
Time: December 15 at 12.00-13.00
Aim: To introduce our three new SfoEpi aawardees and their projects

PhD course 2790: How to conduct systematic reviews and meta-analyses (3hp)

Organizer: Nele Brusselaers, post-doc (Dept. Molecular Medicine and Surgery)
Background:This course falls within the second SFO theme“Molecular, genetic and clinicalepidemiology”.The interactive lectures and seminars will be given by the 3 lecturers mentioned above who are allepidemiologists, coming from 3 different epidemiology units (and departments) within KarolinskaInstitutet. One afternoon will be organised by the library of Karolinska Institutet (Klas Moberg, CarlGornitzki), to learn practical skills on how to perform a proper systematic literature search.
Aim: This epidemiology course aims to provide students with the basic skills and knowledge to performsystematic reviews and meta-analyses. The exams mainly consists of writing a study protocol for anown systematic review/meta-analyses, including a detailed description of the systematic search of the literature.
Planned schedule: 1-5 December: interactive lectures and seminars; Wednesday 10 December: Feedback seminar on study protocol; Wednesday 17 December: Examination

Epidemiological designs in a statistical framework (short course)

Organizer: Anna Johansson (biostatistician/phd student), Cecilia Lundholm (head of the applied biostatistics group at MEB), Caroline Weibull (biostatistician/phd student), Therese Andersson (post-doc)
Background: The lecturer will be Professor Esa Läärä from the Department of Mathematical Sciences at Oulu University, Finland (http://stat.oulu.fi/laara/). He has agreed to deliver this one-week course, where epidemiological concepts will be derived and discussed from a general methodological view combining statistics with epidemiology. Professor Läärä gave a similar, much appreciated course at our department in 2011 and has since developed the material further. Furthermore, Professor Läärä teaches at the highly-regarded course “Statistical Practice in Epidemiology using R” in Tartu.
Aim: We aim to host a one-week advanced course on epidemiology theory aimed at participants with a strong epidemiological and statistical background working at Karolinska Institutet. The scientific aims are:
- To give students a deep understanding of epidemiological theory and concepts
- To discuss around implications of epidemiological designs for different study questions
- To encourage interaction between epidemiology and statistics
Schedule: 3-7 November: The planned course will consist of lectures and interactive sessions, combined with practical work and group discussions among participants. The course covers the main types of epidemiologic designs and discusses their setup from a statistical sampling perspective. It builds on the distinctions concerning the concepts of study population, study base, and sampling strategy. Study populations can be closed or open, and the study base may be cross-sectional or longitudinal. Measurement of risk factor data can be based on full census of the study base, or on some outcome-selective sampling strategy, such as traditional case-control study, density sampling including nested case-control design, and case-cohort sampling. The properties of various designs are compared in terms of their statistical validity and efficiency in estimating the target parameters as well as their practicality in concrete research settings.
Target Group: The course is aimed at participants from any department or research group at Karolinska Institutet, with a strong background in epidemiological methods and biostatistics, e.g. persons with either a degree in statistics or who otherwise have a good knowledge of statistics within epidemiology (e.g., have completed biostat 3, and preferably also advanced doctoral courses such as a causal inference course and longitudinal data analysis). We have informed potential participants at MEB about the planned course, and have had expressions of interest from students and staff at MEB, but we expect there to be interest from other epidemiology departments at KI, such as IMM, KEP and PHS, as well. If the high level of interest should result in the course being over-subscribed, we will prioritise the applications based on the participant’s previous training, area of work and the visiting lecturer’s advice concerning level.

Neuroepidemiology Workshop 16-17 July

Organizer: : Fang Fang, M.D. Ph.D., Assistant Professor, MEB
Background: Neuroepidemiology is a relatively young member of the entire epidemiological research field at KI. Although it has been under vivid development during the last years at MEB for example, a strengthened collaboration within KI and with research entities outside KI is largely needed. To fulfill this aim, in 2012, we organized the first Neuroepidemiology workshop including participants from MEB, IMM and Department of Clinical Neurosciences. This workshop resulted in several important collaborations afterwards. Furthermore, already in 2008, we initiated a collaborative PhD program on ALS between MEB and NIEHS where the PhD student spent one-year research sojourn as a predoctoral fellow at NIEHS. The student was jointly supported by KI and NIEHS. During the following years, this initiative was supported by the KI-NIH joint PhD program in Neuroscience and we have to date enrolled another 3 PhD students (2 in ALS and 1 in PD) of the same training track. Many collaborative works have been carried out through these years, leading to ~20 publications. Now we aim to hold another Neuroepidemiology workshop, aiming to further the development of Neuroepidemiology research at KI
Scientific Aim: To organize a workshop on Neuroepidemiology, focusing on amyotrophic lateral sclerosis (ALS) and Parkinson’s disease (PD), and to foster collaboration between different epidemiology units at KI as well as between KI and the Epidemiology Branch, National Institute of Environmental Health Sciences, National Institutes of Health (NIEHS/NIH), on epidemiological research of neurodegenerative diseases
Target group: Researchers interested in Neuroepidemiology from MEB, IMM, KEP, Department of Public Health Sciences, Department of Clinical Neurosciences, etc.

Advanced Regression Methods for Large Register Studies

Organizer: Marie Reilly
Background: This will be a hands-on course emphasizing both general concepts and application of the methods (using Stata). The instructor is David V. Glidden, from University of California at San Francisco, co-author of the text book Regression Methods in Biostatistics, Vittinghoff, et al. Second Edition, which will be used as the course text.
Scientific Aims: The scientific aims are to enhance the methodological tools in use for register-based research to enable researchers to conduct meaningful analyses of health and disease over the life course using real observational data with recurrent events, competing risks, missingness and confounding. A second aim is to present a range of epidemiological designs for sampling from a cohort for a more efficient study, which is of relevance for pursuing genetic and molecular studies within register cohorts.
Target Group: The target group is participants from any department or research group in Karolinska who work with data from national, regional and quality registers of health and disease. Collaborating epidemiologists and biostatisticians will be encouraged to attend together and to apply the methods to their current research work.
Schedule: 14-16 April

Two 2.5-days advanced courses in epidemiology

Aim: To provide in-depth knowledge and understanding of modern epidemiological methods.
Organizers: Nicola Orsini, Anita Berglund (IMM)
Target Group: Any epidemiologists, biostatisticians (doctoral students, postdocs, junior and senior researchers) at KI.
Time: 2-4 April

Course 1: Time-varying exposures and confounders
Teacher: Professor of Epidemiology, Miguel Hernan, Harvard School of Public Health
Aim: To provide an in depth investigation of statistical methods for drawing causal inferences from observational studies with time-varying exposures.
Background: Epidemiologic concepts such as time-varying exposures/confounders and selection bias, intermediate variables, overall effects and direct effects are formally defined within the context of a counterfactual causal model. Methods for the analysis of the causal effects of time-varying exposures in the presence of time-varying covariates that are simultaneously confounders and intermediate variables are emphasized. These methods include g-estimation of structural nested models, inverse probability weighting of marginal structural models, and the g-formula.

Course 2: Sensitivitey and Bayesian analysis for biases
Teacher: Professor of Epidemiology and Professor of Statistics, Sander Greenland, UCLA
Aim: to explain and illustrate how to perform a sensitivity and Bayesian analysis for common biases in epidemiological research.
Background: In the analysis of observational studies different types of biases or errors (selection, participation, measurement, confounding) arise from the fact that is often not possible or feasible to observe exactly one would like to observe. Only few papers, however, investigate the role of bias on the observed findings. Bias analysis is about quantifying the direction and magnitude of potential sources bias on the results.

Post graduate course in molecular epidemiology at KI

Time: 5-9 May
Organizer: Bruna Gigante
Background: Molecular Epidemiology is a growing field of research in epidemiology where basic concepts and methods of molecular biology are applied to understand the determinants of a disease using an epidemiological approach. The aim of this application is to get support from the SFO to organize a course in Molecular Epidemiology for doctoral students during the autumn term 2013. Such a course is given regularly at the Imperial College in London does not exist at Karolinska Institutet and is indeed of importance for the next generation of epidemiologists to approach critically these novel methods, understand their potential and limitations and how to use them to answer their research questions.
Aims: 
- to give the students an overview of the different –omics (transcriptomics, metabolomics and proteomics) platforms available
- to discuss the differences among the different –omics approaches
- to analyse how to interpreter the results in the different study designs
- ethical issue related to occasional findings

Practical GWAS analysis course

Time: 6-12 November
Organizer: Boel Brynedal, IMM
Aim:The course follows a classical GWAS analysis of a complex trait, from receiving the data to processed and illustrated association results. During the course the student will use multiple different tools and computer languages. Each step of the analysis will be discussed and assessed within the group to reach critical understanding.
Background: The latest decade of years have brought a genomic revolution, in large mediated through the ever decreasing cost, and increasing feasibility and range of genomic assays. The problem is no longer the production of data, but the handling and analysis. Very few individuals have the biological understanding as well as the technical expertise in programming to perform the analytical tasks needed. This course is intended for students with an epidemiological/biological/biostatistical background who are or will perform analysis of genome wide association study (GWAS) data. The focus is on understanding and hands on knowledge, and experience, of each step of a standard GWAS analysis. This course is unique at KI, and will enable large-scale projects in genetic epidemiology. This course will be given for the first time this autumn within the PhD program for epidemiology. During January a pilot course was conducted for 5 students, which was given a very good evaluation.

2013

Lifespan approach to health and disease

Organizer: Dr Marita Södergren and Centre of Family medicine at NVS
Aim: The aims of the visit are to stimulate collaboration between researchers and academic staff at the Departments of NVS and of Public Health Sciences (PHS), and initiate and foster collaboration between Karolinska Institutet and Deakin University. The fields of family medicine and social medicine are concerned with peoples health and disease in interaction with surrounding society. The work involved in conducting education or research within the field of nutrition and physical activity at NVS/PHS encompasses epidemiology, healthcare research, health promotion and preventive work, focusing on risk groups and risk factors in their social and gender-related aspects. Benefits with collaboration are likely to be largest when it involves partners from more divergent scientific backgrounds, like in this visit. An international collaboration is a source of stimulation, creativity, intellectual companionship, and a cross-fertilisation of ideas which may in turn generate new insights or perspectives. Thus, collaboration ensures a more effective transfer of knowledge, skills and techniques.

Epidemiologic research on inflammatory diseases

Organizer: Erik Melén, IMM, Julia Simard and Johan Askling, KEP
Aim: The aim of this full-day workshop is therefore to bring together active epidemiology researchers from these different fields to present their work and thoughts on the study of inflammation and inflammatory diseases through an epidemiologic lens, and thereby to foster important discussions and knowledge transfer. We hope to increase collaboration and awareness of methodologies and to provide a platform to further support translational research.
Brief background: Inflammation underlies a myriad of diseases and is the subject of scientific investigations in countless areas. Whether the target is the skin, the airway, the vascular system, the brain, the gut, or the joint, chronic inflammatory processes are pivotal. Karolinska Institutet has a strong track record of translational research, including epidemiology, in the field of chronic inflammatory diseases. Yet, despite a highly translational setting for epidemiology within each inflammatory field (respiratory, gastrointestinal, rheumatic diseases, et cetera), there is little cross-fertilization between the epidemiology research groups across these fields, which are further physically dispersed all over Karolinska Institutet. In fact, we are convinced that many, like us, lack an overview of ongoing activities, upcoming methodologies, and emerging data here at KI.

Prediction models in medicine

Organizers: Hatef Darabi, Andrea Ganna, Marie Reilly
Aim: With this application we aim to host a short course focused on the development, evaluation and implementation of prediction models in medicine.
Background: Prediction models are used in clinical practice to guide the decision-making regarding further intervention strategies for preventing diseases or adverse treatment responses. These models use several risk factors or marker of diseases to calculate the individual’s chances of illness. Predictors may range from subject characteristics (e.g. age and sex), history and physical examinations to biological or genetic markers. With the development of new technologies and opportunities for large-scale analysis of different omics profiles, sophisticated modelling to develop a truly individualised risk assessment is increasingly important. Together with the ability to build efficient prediction models, biostatisticians and epidemiologists should be able to make an accurate evaluation of the usefulness of a prediction model in clinical practice

Registry-based research in Sweden - past, present and future

Organizer: Rickard Ljung, MMK, KI
Aim: To deepen the understanding of opportunities and limitations of registry-based research.
Background: The excellent opportunities for registry-based research in Sweden can improve health, but this ’goldmine of data’ is underused and there are also methodological limitations and funding issues requiring attention.

Two-day course in competing risks

Organizers: Sandra Eloranta, MEB, Therese Andersson, MEB and Karin Ekström Smedby KEP
Aim: Conceptual understanding of central topics for competing risks analyses.
Short Background: Competing risks is an increasingly recognized problem in follow-up studies. However, the existing literature is often very technical and sometimes even contradictory, which has led to a lot of confusion on this topic. At the same time, many studies investigating prognosis and survival do not recognize potential implications of competing risks neither in their analyses or interpretation of results. We propose a two-day seminar/course with the aim to clarify the definition of competing risks and to help identify study designs, settings and research questions where competing risks need to be addressed versus when competing risks do not need to be taken into account, and a shorter session on methods to deal with competing risks. The overarching goal is to increase the analytical quality in this field at KI. The proposed activity mainly supports SFO theme II Molecular, genetic and clinical epidemiology, but also supports other themes to the extent that study questions involve follow-up analyses. The course will be organized in collaboration between two epi units, namely MEB and KEP. Involvement of other epi centers/departments at KI will be ensured by reserving a defined number of course slots for each center (IMM, MEB, KEP, IARC, other) until up to two weeks before the start of the course.

Psychiatric genetic epidemiology: Process and progress

Organizer: Sarah Bergen, MEB
Aims: 
- Allow leading experts in conducting, analyzing, and overseeing psychiatric genetic epidemiology research from the Medical Epidemiology and Biostatistics (MEB) department the Theme Center for Inflammatory Diseases, and the Broad Institute to share their knowledge and experience at the forefront of this field with the Swedish research community.
- Maintain and strengthen ties between MEB and the Theme Center for Inflammatory Diseases at Karolinska and the Broad Institute and foster communication between researchers at these institutes.
- Facilitate new collaborations between these departments and institutes to enhance academic excellence in the field of psychiatric genetic epidemiology at Karolinska Institutet.
Background: Large, international collaborations have formed to unravel the genetics of many complex genetic diseases and traits including psychiatric disorders. The Broad Institute of MIT and Harvard has been a world leader in the genotyping and analysis for many of these studies and has close ties to the Karolinska Institute. Talks explaining the formation and management of very large-scale international projects, the data and analyses used for them, the recent explosion of breakthroughs in this field, and the future of psychiatric genetic epidemiology will all be given by experts on these topics.

Data management and graphics using STATA

Organizers: Debora Rizzuto, Giola Santoni and Weili Xu, Aging Research Center (ARC), NVS, Karolinska Institutet
Aim: This workshop aims to discuss key issues on data management and preparation for data analysis, and to introduce methods on making simple and complicate graphs as well as better Log storage using Stata based on the practical issues raised by researchers.
Background: ARC regularly organizes statistical workshops, where PhD students, postdocs and researchers congregate to review key issues on statistics in their data analysis and management, and discuss these issues with the experts in statistics. As Stata has been commonly used, this workshop focuses on practical issues about data cleaning and conversions between formats, preparation for analysis and making graphs using Stata. During this activity, participants may share experiences and views on data quality, collection, and analysis. The Statisticians may provide a review and comments on methodologies and concepts of statistics/indicators. Such workshop is an important platform to ensure proper methods and interpretation that are applied in data analysis. Although many participants participated in many statistical courses previously, practical issues are often encountered and their knowledge on statistics needs to be updated. This workshop also promotes cooperation and collaboration among different epidemiological units within KI.

Basic reproductive epidemiology

Organizers: with Dr Anastasia Nyman Iliadou, MEB, Prof Sven Cnattingius, and Dr Olof Stefansson, MedS
Aim: The aim of the course is to first lay the foundations - the basic principles of reproductive physiology, demography, infectious diseases, and genetics as they apply to human reproduction. In parallel it will deal with the endpoints of reproductive epidemiology - a spectrum ranging from infertility and fetal loss to birth defects and the delayed effects of fetal exposures. The course will end with discussions of unsolved problems, suggesting possible research projects for a new generation of epidemiologists.
Background: There is no regular course in reproductive epidemiology at Karolinska Institutet today. Nevertheless reproductive epidemiology is a large scientific field in which several departments are involved both epidemiological and clinical, but also involving basic science. Hnce the need for a course in reproductive epidemiology is warranted.

Bayesian methods and bias analysis in epidemiologic research

Organizer: Nicola Orsini, IMM
Teacher: Professor Sander Greenland, who is considered a leading authority on quantitative methods and statistical theory in epidemiology.
Aim: Bayesian methods and bias analysis have become common in advanced training and research in epidemiology. Unfortunately there is no elementary training available at KI on these specific topics. Therefore, aim of this short course is to provide the knowledge and understanding of these modern epidemiological methods.
Background: It is unlikely that an epidemiologist collects and analyzes its own data without any prior knowledge on the topic. Nevertheless, standard analysis (logistic, Cox) disregards the scientific evidence accumulated in the field and focus on the data at hand. A Bayesian approach is about incorporating (e.g. averaging) prior knowledge or expert judgments with study-specific results. In the analysis of observational studies different types of biases or errors (selection, participation, measurement, confounding) arise from the fact that is often not possible or feasible to observe exactly one would like to observe. Only few papers, however, investigate the role of bias on the observed findings. Bias analysis is about quantifying the direction and magnitude of potential sources bias on the results.

2012

Activities aiming to increase scientific collaboration between epidemiology units at Karolinska Institutet

Organizer: Yang Cao, Ph.D., Assistant Professor, Unit of Biostatistics, Division of Epidemiology, IMM
Aim: To hold a full-day workshop on research and data resources of Shanghai Municipal Center for Disease Control and Prevention (SCDC) and Shanghai Institutes of Preventive Medicine (SIPM), and to foster collaboration between KI epidemiology units and between KI and SCDC/SIPM.
Brief background: Shanghai is the largest city in China and in the world. The 2010 census put Shanghai's total population at 23,019,148, which is more than double of Swedish national population. The diversity and complexity of disease model in Shanghai provides a unique opportunity for global epidemiologists to conduct research on cause and mechanism of diseases. SCDC is the first and one of the largest provincial CDCs in China, which was based on the example of the US CDC and marked a significant step forward in improving public health in China. SCDC owns abundant and high quality data on birth and death, cancers, chronic diseases and communicable diseases and has a well-known reputation for scientific works. On the international level, SCDC collaborates with WHO, World Bank, UNICEF, EU, US CDC, US NIH, the Department of Health Services of California, and University of Toronto. Now, it is looking for research collaborators in Nordic countries.

Applied statistical workshop

Organizer: Weili Xu, Neda Agahi and Anna-Karin Welmer, Aging Research Center (ARC), NVS, Karolinska Institutet
Aim: This workshop aims to discuss key issues and review updated statistical methods for data analysis on quantitative measurements, and to solve practical problems that researchers commonly encountered in the data analysis.
Brief background: ARC regularly organizes statistical workshops, where PhD students, postdocs and researchers congregate to review key issues on statistics in their data analysis, and discuss these issues with the experts in statistics. This workshop focuses on a selected set of estimation issues that we generally encounter for quantitative measurements in data analysis. During this activity, participants may share experiences and views on data quality, collection, and analysis. The experts may provide a review and comments on new statistical tools, methodologies and concepts of statistics/indicators. Such workshop is an important platform to ensure proper methods and interpretation that are applied in data analysis. Although many participants participated in many statistical courses previously, practical issues are often encountered and their knowledge on statistics needs to be updated. This workshop also promotes cooperation and collaboration among different epidemiological units within KI.

Early life seminar with invited speakers

Organizers: Catarina Almqvist Malmros and Anastasia Nyman Iliadou on behalf of the Strategic Research Program (SFO) in Epidemiology
Aim: To visualize current knowledge on the effect of conception, IVF and early perinatal factors on subsequent health
Background: The importance of early life effects such as IVF, intrauterine growth restriction, preterm birth as well as growth patterns in childhood has been assessed in contemporary cohorts. Fetal growth has been found to have a profound effect on subsequent diseases across the lifespan in a plethora of studies. Further, differing growth patterns have also been associated to later adult diseases. IVF is related to low birth weight and preterm birth with an unknown etiology. Epigenetics has been suggested as a possible mechanism. Epigenetic programming may operate through manipulations of the gametes and embryos in early stages or through maternal characteristics during pregnancy. It has also been shown to be related to mode of delivery.