## Fall 2024 Courses

**PBHLTH C240B 001 - Biostatistical Methods: Survival Analysis and Causality**

**Course Instructor:** Mark van der Laan**Course Catalog Description:** Analysis of survival time data using parametric and non-parametric models, hypothesis testing, and methods for analyzing censored (partially observed) data with covariates. Topics include marginal estimation of a survival function, estimation of a generalized multivariate linear regression model (allowing missing covariates and/or outcomes), estimation of a multiplicative intensity model (such as Cox proportional hazards model) and estimation of causal parameters assuming marginal structural models. General theory for developing locally efficient estimators of the parameters of interest in censored data models. Computing techniques, numerical methods, simulation and general implementation of biostatistical analysis techniques with emphasis on data applications.

**PBHLTH 243A 001 - Targeted Learning**

**Course Instructor:** Mark van der Laan**Course Catalog Description:** PH 243A teaches students to construct efficient estimators & obtain robust inference for parameters that utilize data-adaptive estimation strategies (i.e., machine learning). Students perform hands-on implementation of novel estimators using high-dimensional data structures, providing students with a toolbox for analyzing complex longitudinal, observational & randomized control trial data. Students learn & apply the core principles of the Targeted Learning methodology, which generalizes machine learning to any estimand of interest; obtains an optimal estimator of the given estimand, grounded in theory; integrates state-of-the-art ensemble machine learning techniques; & provides formal statistical inference in confidence intervals & testing.

**PBHLTH C242C 001 - Longitudinal Data Analysis**

**Course Instructor:** Alan Hubbard**Course Catalog Description:** Course covers statistical issues surrounding estimation of effects using data on units followed through time. Course emphasizes a regression model approach for estimating associations of disease incidence modeling, continuous outcome data/linear models & longitudinal extensions to nonlinear models forms (e.g., logistic). Course emphasizes complexities that repeated measures has on the estimation process & opportunities it provides if data is modeled appropriately. Most time is spent on 2 approaches: mixed models based upon explicit (latent variable) maximum likelihood estimation of the sources of the dependence, versus empirical estimating equation approaches (generalized estimating equations). Primary focus is from the analysis side.

**PBHLTH W250 001 - Epidemiologic Methods I**

**Course Instructor:** Zachary Butzin-Dozier**Course Catalog Description: **This introductory graduate course presents principles & methods of epidemiology, including descriptive & analytic approaches to assessing the distributions of health, disease & injury in the population & factors influencing those distributions. Emphasis is on developing an understanding of concepts, rather than quantitative methods. Basic calculations are involved. Course consists of readings, critical review of epidemiology papers, brief video lectures to explain key concepts, quizzes & exams that allow students to practice applying epidemiological concepts. Students without prior coursework in epidemiology will acquire the core competencies in epidemiology expected of all MPH graduates. Course shares the same content & learning of PH 250A

**PBHLTH W250B 001 - Epidemiologic Methods II**

**Course Instructor:** Andrew N Mertens**Course Catalog Description:** This course is an intermediate level course in epidemiology. It replaces previously approved and delivered courses PHW250F and PHW250G. Topics include causal inference; measurement of disease rates; inferential reasoning; research study designs, ecologic, case-control, cohort, intervention trials, meta-analytic designs; potential sources of bias, confounding, effect modification in research design are explored in depth; topics in clinical epidemiology, likelihood ratios, receiver operator curves, the sensitivity, specificity, predictive value of a test; brief introduction to logistic regression and survival analysis. Topics are covered at a advanced level than PH250A or PHW250. Readings from this course provide a firm foundation for PH250C.

## Courses Taught Previously by CTML Faculty and Staff

**Spring 2024**

Course | Course Instructor | Course Catalog Description |
---|---|---|

PBHLTH C240A Introduction to Modern Biostatistical Theory and Practice |
Alan E Hubbard and Mark van der Laan |
Course covers major topics in general statistical theory, with a focus on statistical methods in epidemiology. The course provides a broad theoretical framework for understanding the properties of commonly-used and more advanced methods. Emphasis is on estimation in nonparametric models in the context of contingency tables, regression (e.g., linear, logistic), density estimation and more. Topics include maximum likelihood and loss-based estimation, asymptotic linearity/normality, the delta method, bootstrapping, machine learning, targeted maximum likelihood estimation. Comprehension of broad concepts is the main goal, but practical implementation in R is also emphasized. Basic knowledge of probability/statistics and calculus are assume |

PBHLTH W250 Epidemiologic Methods I |
John M Colford Jr., Andrew N Mertens, Zachary Butzin-Dozier |
This introductory graduate course presents principles & methods of epidemiology, including descriptive & analytic approaches to assessing the distributions of health, disease & injury in the population & factors influencing those distributions. Emphasis is on developing an understanding of concepts, rather than quantitative methods. Basic calculations are involved. Course consists of readings, critical review of epidemiology papers, brief video lectures to explain key concepts, quizzes & exams that allow students to practice applying epidemiological concepts. Students without prior coursework in epidemiology will acquire the core competencies in epidemiology expected of all MPH graduates. Course shares the same content & learning of PH 250A |

PBHLTH 241 Intermediate Biostatistics for Public Health |
Alejandro Schuler |
In this course, students will study biostatistical concepts and modeling relevant to the design and analysis of multifactor population-based cohort and case-control studies, including matching. Key topics include: measures of association, causal inference, confounding interaction, with modeling focusing on logistic regression. |

PBHLTH W252A Introduction to Causal Inference for Public Health Professionals |
Laura Balzer and Nerissa Nance |
With the ongoing “data explosion”, methods to delineate causation from correlation are perhaps more pressing now than ever. This course will introduce a general framework for Causal Inference in Public Health: 1) clear statement of the research question, 2) definition of the causal model and effect of interest, 3) assessment of identifiability, 4) choice and implementation of estimators including parametric and non-parametric methods, and 5) appropriate interpretation of findings. The statistical methods include G-computation, inverse probability weighting (IPW), and targeted minimum loss-based estimation (TMLE) with machine learning. |

**Fall 2023**

Course | Course Instructor | Course Catalog Description |
---|---|---|

PHC242C: Longitudinal Data Analysis |
Alan E Hubbard |
The course covers the statistical issues surrounding estimation of effects using data on subjects followed through time. The course emphasizes a regression model approach and discusses disease incidence modeling and both continuous outcome data/linear models and longitudinal extensions to nonlinear models (e.g., logistic and Poisson). The primary focus is from the analysis side, but mathematical intuition behind the procedures will also be discussed. The statistical/mathematical material includes some survival analysis, linear models, logistic and Poisson regression, and matrix algebra for statistics. The course will conclude with an introduction to recently developed causal regression techniques (e.g., marginal structural models). Time permitting, serially correlated data on ecological units will also be discussed. |

PH 142: Introduction to Probability and Statistics in Biology and Public Health |
Alan E Hubbard |
PH 142 is an introduction to statistics and data science, primarily for MPH and undergraduate public health majors, and others interested in public health majors. The course material focuses on the biomedical applications of basic data summarization using the statistical programming language: R, classical problems in probability/statistical distributions (Normal, binomial, Poisson), and statistical inference techniques. |

**Fall 2021**

Course | Course Instructor | Course Catalog Description |
---|---|---|

PH252E: Advanced Topics in Causal Inference |
Alan E Hubbard |
The course will be conducted as a seminar with readings and discussions on a range of more advanced topics. We will cover case-control designs; longitudinal causal models, identifiability and estimation; direct and indirect effects; dynamic regimes (individualized treatment rules); approaches for diagnosing and responding to violations in the positivity assumption. Additional topics may include stochastic interventions, community-based interventions, and Collaborative-TMLE. There will also be some guest lectures and presentations from current students and faculty members. |

PH243A: Targeted Learning |
Mark van der Laan |
PH 243A teaches students to construct efficient estimators & obtain robust inference for parameters that utilize data-adaptive estimation strategies (i.e., machine learning). Students perform hands-on implementation of novel estimators using high-dimensional data structures, providing students with a toolbox for analyzing complex longitudinal, observational & randomized control trial data. Students learn & apply the core principles of the Targeted Learning methodology, which generalizes machine learning to any estimand of interest; obtains an optimal estimator of the given estimand, grounded in theory; integrates state-of-the-art ensemble machine learning techniques; & provides formal statistical inference in confidence intervals & testing. |

This work by CTML Faculty and/or Staff is licensed under CC BY 4.0