Rob Cavanaugh


I’m a research data analyst at the Observational Health Data Sciences and Informatics Center (OHDSI) at Northeastern University’s Roux Institute in Portland, Maine. My research converges at the intersection of rehabilitation health services research, population health, and data science, primarily using insurance claims and electronic health record data to examine disparities in healthcare outcomes. In parallel, my research aims to identify dose-response relationships and individual differences in treatment outcomes for treatments across the scope of practice for speech-language pathology.

My clinical experiences as a speech-language pathologist in neurorehabilitation serve as the foundation for this work. Outcomes and access to care after neurological injury or diagnosis of neurodegenerative condition vary widely, and this phenomenon is most visible and frustrating to clinicians in the trenches. These experiences motivated my PhD work at the University of Pittsburgh examining real-world dosage, as well as mechanisms and ingredients of discourse-level outcomes in anomia treatment for post-stroke aphasia.

I’m also a statistical co-investigator on several ongoing studies and mentor of undergraduate, masters, and doctoral students in quantitative methods and data analytics. My quantitative skillset includes hierarchical/mixed-effects models, Bayesian methods, item-response theory, and emerging skills in causal inference frameworks. I’m passionate about teaching and mentoring researchers at all levels and students in statistics, open science, and thoughtful use of quantitative methods to achieve their goals. I believe that every student can become a confident methodologist through active and experiential learning using data they are passionate about. I also believe that a strong conceptual foundation in statistical basics is critical for the next generation of health and rehabilitation science students to successfully evaluate and implement cutting edge evidence-based practice.

I have developed multiple web-applications, websites, and R packages for science dissemination and implementation using R, Javascript, and Python. These include computer adaptive testing platforms, the allofus R package, and a repository for statistics training resources for allied health researchers. I also have expertise in Natural Language Processing methods via NIH-NRSA funded training in computational linguistics and discourse analysis of disordered language. If you’ve made it this far, I encourage you to look through this website which includes summaries and links to these projects.

Outside of my day job, you can find me and my wife Amanda exploring the outdoors with our two dogs, Murphy and Willa.

Rob Cavanaugh


I’m a research data analyst at the Observational Health Data Sciences and Informatics Center (OHDSI) at Northeastern University’s Roux Institute in Portland, Maine. My research converges at the intersection of rehabilitation health services research, population health, and data science, primarily using insurance claims and electronic health record data to examine disparities in healthcare outcomes. In parallel, my research aims to identify dose-response relationships and individual differences in treatment outcomes for treatments across the scope of practice for speech-language pathology.

My clinical experiences as a speech-language pathologist in neurorehabilitation serve as the foundation for this work. Outcomes and access to care after neurological injury or diagnosis of neurodegenerative condition vary widely, and this phenomenon is most visible and frustrating to clinicians in the trenches. These experiences motivated my PhD work at the University of Pittsburgh examining real-world dosage, as well as mechanisms and ingredients of discourse-level outcomes in anomia treatment for post-stroke aphasia.

I’m also a statistical co-investigator on several ongoing studies and mentor of undergraduate, masters, and doctoral students in quantitative methods and data analytics. My quantitative skillset includes hierarchical/mixed-effects models, Bayesian methods, item-response theory, and emerging skills in causal inference frameworks. I’m passionate about teaching and mentoring researchers at all levels and students in statistics, open science, and thoughtful use of quantitative methods to achieve their goals. I believe that every student can become a confident methodologist through active and experiential learning using data they are passionate about. I also believe that a strong conceptual foundation in statistical basics is critical for the next generation of health and rehabilitation science students to successfully evaluate and implement cutting edge evidence-based practice.

I have developed multiple web-applications, websites, and R packages for science dissemination and implementation using R, Javascript, and Python. These include computer adaptive testing platforms, the allofus R package, and a repository for statistics training resources for allied health researchers. I also have expertise in Natural Language Processing methods via NIH-NRSA funded training in computational linguistics and discourse analysis of disordered language. If you’ve made it this far, I encourage you to look through this website which includes summaries and links to these projects.

Outside of my day job, you can find me and my wife Amanda exploring the outdoors with our two dogs, Murphy and Willa.