Biostatistics Dissertation Defenses


Events that appear on this page are in the Sponsoring Department of "Biostatistics" with "Dissertation Defenses" as its Category.

To add or edit events on this page, go to

Upcoming Biostatistice Dissertation Defenses

Thu 7/23/2020 9:00AM - 11:00AM
Biostatistics Dissertation Defense
Marie Tuft-Statistical Learning for the Spectral Analysis of Time Series Data Biostatistics Dissertation Defense
Marie Tuft-Statistical Learning for the Spectral Analysis of Time Series Data
Thu 7/23/2020 9:00AM - 11:00AM
** Online/Virtual Event **

Marie Tuft of the Department of Biostatistics defends her dissertation on "Statistical Learning for the Spectral Analysis of Time Series Data". 

** Online/Virtual Event **


Biostatistics Dissertation Defense

Christopher Keener - Power and Sample Size Determination for Stepped Wedge Cluster Randomized Trials

Wednesday 7/18 10:00AM - 12:00PM
7139 Public Health, Peterson Seminar Room

Christopher Keener of the Department of Biostatistics defends his dissertation on "Power and Sample Size Determination for Stepped Wedge Cluster Randomized Trials".

Graduate faculty of the University and all other interested parties are invited to attend


A stepped wedge trial is a type of cluster randomized trial with unidirectional crossover from control to intervention. In this study, we classified stepped wedge trial designs according to subject recruitment and outcome exposure. Based upon those criteria, we proposed three types of classification, that is, fixed cohort (baseline recruitment with longitudinal exposure), expanding cohort (continuous recruitment with longitudinal exposure), and cross-sectional (continuous recruitment with cross-sectional exposure). For each design type, we proposed a corresponding model for estimating treatment effect. We conducted Monte Carlo simulations to study the impact of design and analytic assumptions on the sample size and power determination. These assumptions include homogeneous or heterogeneous temporal effects between clusters, fixed or time-varying treatment effect, modeling temporal trend through or not through step effects, and choice of correlation structure.  

To investigate how these assumptions were made in the published trials, we conducted a systematic review of 300 stepped wedge trials published up to 2017. From the review we found that more than one fourth of these trials did not make it clear in their reports about the type, the assumptions, or models in estimating treatment effect and sample size calculations. The majority of the trials did not mention the methods for handling missing data. This suggests the need for developing standards of reporting stepped wedge trials like CONSORT for randomized trials or STROBE for observational studies. 

PUBLIC HEALTH SIGNIFICANCE: Stepped wedge trials are popular for evaluating community-based interventions in public health. This research has focused on three areas of improving the design of a stepped wedge trial: classification of key design aspects, power and sample size determination, and modeling method for estimation and inference the effect of an intervention. Sample size determination is important to ensure that the trial is adequately powered. Model misspecification and incorrect analytic assumption both can lead to inflated Type I error rate or an underpowered trial. Our systematic review found that many stepped wedge trials failed to define key aspects and assumptions of their designs when publishing. Thus, use of our classification of stepped wedge trials will improve technical communication on trials commonly used for public health research. 


Last Updated On Tuesday, October 9, 2018 by Valenti, Renee Nerozzi
Created On Tuesday, June 12, 2018

Questions and Submissions

The schedule of dissertation defenses in the Department of Biostatistics is maintained by...

Renee Valenti


Share the details for your defense to ensure it is displayed here as well as on the department calendar, Pitt Public Health's calendar, Weekly Update, and LCD screens.

Defenses by department