Course description

Block 1 | Block 2 | Block 3 | Block 4 | Block 5 | Block 6 | Stata 1 | Stata 2

Block 1

Principles of Biostatistics - M. Pagano (Harvard School of Public Health)

Introduces the fundamental principles of statistics applied to biomedicine. The topics to be covered include: descriptive statistics, measures of central tendency, probability, diagnostic testing, population and sample, comparison of proportions. At the end of the course, students will be able to understand the descriptive statistical methodologies which are used in clinical and epidemiological studies and to utilize the estimates obtained from suitably selected samples, in order to draw statistical inferences.

Linear Regression for Medical Research - R. Bellocco (University of Milano-Bicocca and Karolinska Institutet)

The course introduces students to the practice and application of regression modeling. Through the use of Stata®, students will learn how to fit a regression, estimate, and test regression coefficients. Particular emphasis will be placed on the interpretation of the regression coefficients of continuous and categorical predictors. Analysis of variance models and their correspondence with regression models will also be covered together with procedures and issues in model selection, including confounding and interaction. Model building, goodness of fit, residual analysis, and appropriate regression diagnostics will be discussed.

Causal Inference in Epidemiology - A. Sjölander (Karolinska Institutet)

Causal inference from observational data is a key task of biostatistics and of allied sciences such as sociology, econometrics, behavioral sciences, demography, economics, health services research, etc. These disciplines share a methodological framework for causal inference that has been developed over the last decades. This course presents this unifying causal theory and shows how biostatistical concepts and methods can be understood within this general framework. The course emphasizes conceptualization but also introduces statistical models and methods for causal effect estimation. Specifically, this course strives to a) formally define causal concepts such as causal effect and confounding using potential outcomes and counterfactuals, b) identify the conditions required to estimate causal effects using Directed Acyclic Graphs (DAGs), and c) introduce analytical methods that, under those conditions, provide estimates that can be endowed with a causal interpretation. Examples of such methods are regression adjustment, standardization and inverse probability weighting.

Block 2

Principles of Epidemiology - E. Mostofsky (Harvard School of Public Health)

This course provides an introduction to the skills needed by public health professionals and clinicians to critically interpret the epidemiologic literature. It will provide participants with the basic principles and practical experience needed to develop these skills. This will be accomplished by covering the basic principles and methods of the design, conduct and interpretation of epidemiologic studies, including descriptive studies, observational analytic studies (case-control and cohort), and randomized clinical trials. In addition, the course will address the calculation and interpretation of measures of disease frequency and association; the assessment of association versus causation in the interpretation of study results; and an introduction to issues related to the evaluation of chance, bias, confounding, and effect modification. Lectures will be complemented by seminars devoted to case studies, exercises, or critiques of relevant examples of epidemiologic studies.

Logistic Regression for Medical Research - D. Wypij (Harvard School of Public Health)

This course introduces students to the practice and application of logistic regression modeling for binary outcomes. Students will fit, evaluate, and interpret binary data models arising from epidemiological studies, clinical trials, or other application areas. Topics include assessment of confounding and effect modification, use of indicator variables, model building methods, goodness-of-fit assessment, presentation of logistic regression models for reports and publications, and an introduction to conditional and ordinal logistic regression. Data sets from the medical and public health literature will be used as case studies to be analyzed using the Stata® statistical package.

Survival Analysis - P. Dickman (Karolinska Institutet)

The course introduces statistical methods for survival analysis, that is, the analysis of studies where the outcome is a time-to-event. We will study methods for estimating patient survival (life table and Kaplan-Meier methods), comparing survival between patient subgroups (log-rank test), and modelling survival (Poisson regression, Cox regression, and flexible parametric regression). Emphasis will be placed on describing how epidemiological cohort studies can be analysed in the framework of survival analysis. We will study the concept of 'time' as a potential confounder or effect modifier and discuss various approaches to defining 'time' (e.g., time since entry, attained age, calendar time). The course will emphasise basic concepts of statistical modelling in epidemiology, such as controlling for confounding and assessing effect modification. The course will consist of lectures, classroom exercises, and computing exercises (using Stata®).

Block 3

Effectiveness Research with Longitudinal Healthcare Databases - S. Schneeweiss & E. Patorno (Harvard School of Public Health) *

Large longitudinal healthcare databases have become important tools for studying the utilization patterns and clinical effectiveness of medical products and interventions in a wide variety of care settings. This course will prepare students to identify and use longitudinal databases for their own research. Strengths and limitations of large longitudinal healthcare databases that are commonly used for research will be considered. Special attention will be devoted to nationally representative databases that are critical for comparative effectiveness research. The course focuses on analytic principles and their application to database research. Participants will learn through lectures by experienced faculty and by evaluating published database studies. In computer labs they will learn to implement a database study comparing two medical products in a large healthcare claims database. The project will be conducted using the Aetion platform with an intuitive user interface that does not require any programming skills. The course requires a working understanding of epidemiologic study designs and typical analysis strategies. The target audience consists of researchers working in academia, medical product industry, health plans, government institutions, regulatory agencies, who have access to large longitudinal healthcare databases. They may use such data to evaluate the effectiveness of medical interventions and care patterns, to understand the comparative effectiveness and safety of medical products (drugs, devices), to test the impact of coverage policy changes, or to monitor the outcome of risk-sharing arrangements.

* This course can also be taken in its second part only, starting from Thursday (please contact us for details) You may also register for this course through the Eu2P program. This enables you to be awarded with a Eu2P academic certificate diploma bearing 3 ECTS credits.

Block 4

Research Methods in Health: Biostatistics - M. Bonetti (Bocconi University)

This course is designed to provide the student with an understanding of the foundations of biostatistics and of the various statistical techniques that have been developed to answer research questions in the health sciences. Students will be introduced to methods for the comparison of outcome between two groups (t-test and non parametric tests), as well as the extension to the comparison of outcome across several groups (ANOVA); methods for the study of association between two continuous variables (correlation and linear regression); the analysis of contingency tables; the study of survival (time-to-event) data. The afternoon sessions are devoted to discussion and learning to use Stata® to implement materials covered in the morning lectures.

Mediation Analysis - L. Valeri (Harvard School of Public Health & McLean Hospital)

The course will cover some of the recent developments in causal mediation analysis and provide practical tools to implement these techniques. Mediation analysis concerns assessing the mechanisms and pathways by which causal effects operate. The course will cover the relationship between traditional methods for mediation in epidemiology and the social sciences and new methods in causal inference. For dichotomous, continuous, and time-to-event outcomes, discussion will be given as to when the standard approaches to mediation analysis are valid. Using ideas from causal inference and natural direct and indirect effects, alternative mediation analysis techniques will be described when the standard approaches will not work. The no-confounding assumptions needed for these techniques will be described. Stata and R macros to implement these techniques will be covered and distributed to course participants. The use and implementation of sensitivity analysis techniques to assess the how sensitive conclusions are to violations of assumptions will be covered. Discussion will be given to how such mediation analysis approaches can be extended to settings in which data come from a case-control study design. The methods will be illustrated by various applications. The course will employ a combination of lecture, discussion, and software demonstration. Slides will be used to present material in lecture form. Extensive printed notes will be available for students. A wide variety of examples from epidemiology and the social sciences will be used to illustrate the techniques and approaches. Ample time will be given for discussion and questions. A variety of software packages will be discussed. Students will have worked exercises that they can complete on their own.

Longitudinal Data Analysis - G. Fitzmaurice (Harvard School of Public Health)

This course focuses on methods for analyzing longitudinal and repeated measures data. The defining feature of longitudinal studies is that measurements of the same individuals are taken repeatedly through time, thereby allowing the direct study of change over time. This type of study design encompasses epidemiological follow-up studies as well as clinical trials. The course covers many well-established methods for the analysis of longitudinal data when the response variable is continuous. Methods for discrete response variables (e.g., repeated binary responses and counts) are introduced, but not emphasized. An introductory course in biostatistics and a good background in linear regression analysis are prerequisites for this course.

Block 5

Research Methods in Health: Epidemiology - M. Mittleman (Harvard School of Public Health)

This course will explore in greater depth the fundamental epidemiologic concepts introduced in Principles of Epidemiology (Week 1). The course will be taught with an emphasis on causal inference in epidemiologic research. Topics will mainly focus on chronic disease epidemiology, with a special emphasis on practical study design. Epidemiologic examples from major chronic diseases/conditions (e.g. heart disease and cancer) will be discussed. Students will revisit the issues of confounding, selection bias, effect modification, and generalizability in the context of these topics. Lectures will be augmented by workshops to illustrate practical examples in the epidemiologic literature. The material covered in Principles of Epidemiology will be assumed of the students entering this course.

Competing Risks for Survival Analysis - N.P. Jewell (Berkeley University)

An understanding of competing risk methodology allows an epidemiologist to analyze all-cause mortality (or similar outcome) focusing on factors associated with a specific cause of death (e.g. cardiovascular death) while accommodating that other causes may be related the cause of interest (and, for example may preclude observation of a cardiovascular death by occurring earlier). The appropriate statistical methods extend concepts commonly used to study time to event data including the Kaplan-Meier estimator and proportional hazards models. Descriptive methods focus on the cumulative incidence function; two regression models will be covered and contrasted--both for the cause-specific hazard function and the cumulative incidence function. Multiple examples will be presented and programs to implement the ideas within Stata will also be discussed. The focus of the class will be on interpretation rather than theory. Learning objectives include recognizing competing risk data structures and using data to estimate key parameters along with their correct interpretation.

Flexible Modeling of Quantitative Predictors - N. Orsini (Karolinska Institutet)

The aim of this course is to introduce participants to different ways of modeling the association between quantitative predictors (exposures, confounders, calendar time) and different kinds of outcome data (continuous, binary, counts) in common epidemiological studies of individual data (case-control, cohort) or aggregated data (interrupted time-series, quantitative review, pooling project). The emphasis is on the interpretation and presentation of the findings either in a tabular or graphical form suitable for publication. Specific topics include dose-response analysis using regression splines, assess longitudinal effects of interventions, and plot adjusted measures of association (mean differences, percentile differences, odds ratios, rate ratios, pooled relative risks) as function of the quantitative predictor based on multivariable models (linear, logistic, Poisson, Cox, meta-analysis). As motivating and instructive examples both lectures and exercises will be based on real studies published in top scientific journals. Stata software will be used throughtout the course. Familiarity with basic principles of statistics (p-value, confidence intervals) and regression models is recommended.

Block 6

Monitoring and Evaluation of Public Health Programs: Systems Approaches and Techniques - M. Pagano & E. Savoia (Harvard School of Public Health)

This course introduces methods and tools necessary for monitoring and evaluating public health programs during routine public health activities as well as during large scale emergencies and public health crisis. The course will use the case based teaching method developed by the Harvard Business School and examples from the ebola outbreak, recent water crisis and other types of events to describe how evaluation methods can be used to inform public health decision making. The range of topics includes: evaluation planning, survey development and validation techniques, assessment of modern and rapid testing methods; an overview of various methodologies and designs for estimating coverage and changes for a region; methods for evaluating sub-regional performance (i.e. the health districts of a region); and comprehensive monitoring and evaluation approaches that allow for both local and regional assessment. Emphasis will be on the practical aspects of design, analysis and presentation. Students will use a public health systems approach to the evaluation of the programs and discuss as a group the consequences of the decisions they make on the implementation and evaluation of specific public health programs.

Stata 1

Basics of Stata® - B. Pongiglione (Institute of Education, University College London)

This course is designed to introduce students to the basics of Stata. It will focus on the minimum set of commands everyone should know to organize their own work. Specific topics include data-management, data-reporting, graphics and basic use of do-files. By the end of this one-day course, the student should be capable of using Stata independently.

Meta-analysis using Stata® - R. D'Amico (University of Modena and Reggio Emilia)

The aim of this course is to provide an overview of methods to perform meta-analysis. We will cover the following topics: data preparation and input, fixed and random-effect models, forrest plots (publication quality graph), heterogeneity across studies, publications bias, sensitivity analysis, meta-regression models and dose-response meta-analysis.

Analysis of prospective studies with Stata® - L. Hong (Bocconi University)

This course is designed to introduce student to the analysis of cohort studies, managing person-times, estimating counts and incidence rate ratios of both fixed and time-varying exposures and fitting count regression models. By the end of the course, the student will be familiar these epidemiogical techniques using Stata.

Stata 2

Basics of Stata® - F. Gallo (CPO Piemonte - S.C. Epidemiologia, Screening e Registro Tumori)

This course is designed to introduce students to the basics of Stata. It will focus on the minimum set of commands everyone should know to organize their own work. Specific topics include data-management, data-reporting, graphics and basic use of do-files. By the end of this one-day course, the student should be capable of using Stata independently.

Tables for epidemiologists using Stata® - A. Discacciati (Karolinska Institutet)

This course is designed to introduce students to basic Stata commands useful in epidemiological research: descriptive statistics to estimate the incidence of a binary response and to characterize the demographic information supplied by study participants; statistical tests to identify univariate predictors associated with the binary response; graph the incidence of a binary response as a function of a predictor; and table of standardized means and proportions.

Multiple Imputation using Stata® - G. DiTanna (Queen Mary University of London)

The course introduces the basics of multiple imputations, in particular imputation by chained equations. Students should have a background in regression methods prior to taking this course.

Data Visualization with Stata® - G. Capelli (University of Cassino and Southern Lazio)

The course introduces students to the logic and the strategies for visualizing data in Stata. Among the topics, the course will explore the issues in the choice of the most appropriate graphic (distributional, compositional or correlational) for different data and aims, and tips and tricks to prepare data for different graphical schemes. In particular, the power and flexibility of multiple "layers" in twoway Stata panels will be exploited. By the end of this one-day course, students will be able to produce Stata Graphs, and export them to JPG, TIFF or PDF formats for further applications.