Our mission
The mission of the Biostatistics and Data Management Unit is to strengthen pediatric research and patient care by providing guidance and expertise in biostatistics, statistical programming, and clinical data management. We support members of the Pediatric Department at Michigan Medicine in the conduct of high quality, innovative research projects while also providing opportunities for career development in research.
What we offer
Our group includes PhD and Masters-level biostatisticians with expertise in quantitative and qualitative research methods. Additionally, our group can work with you to clean, organize, simplify, and code existing study data and create analysis ready datasets. Our services are available to all levels of investigators, from residents and fellows, to junior faculty, and well-established investigators. We are your statistical co-investigators and co-authors. We also collaborate with other University of Michigan units. Team members are independent researchers working on their own projects.
We are able to assist with the following:
- Study design, protocol review and development
- Sample size justification
- Randomization
- Review of data collection instruments
- General statistical consulting
- Dataset organization, linking, and transformation
- Statistical methods selection and implementation
- Statistical analysis and interpretation
- Preparation of statistical graphics and tabular displays of data
- Independent statistical reviews for manuscripts
- Writing statistical methods and results sections for manuscripts, abstracts, or presentations
- Grant proposal development
Who we are
- Richard Eikstadt, BSEE, Senior Data Architect
Mr. Eikstadt earned his Bachelor of Science in Electrical Engineering at Ohio University in 1994. He has worked as a programmer at UM since 1998, solely authored the UMHS kidney paired donation software program (leading to nearly 100 kidney transplants) and currently leads a nephrology research data coordinating center at the University. He has contributed to numerous publications involving nephrology and transplantation. 2019 is his first year with the Woodson program and he looks forward to supporting the pediatric research community at Michigan Medicine.
- Dr. Niko Kaciroti, PhD, Research Scientist
Dr. Kaciroti’s research focuses on using Bayesian modeling techniques for analyzing longitudinal data from randomized clinical trials with missing data as well as for modeling nonlinear and dynamic models in a multilevel setting. His applied research is related to the effect of iron deficiency on brain, behavior and development; obesity; managing chronic disease; hypertension and cardiovascular diseases; and emotion regulation as complex systems in preschoolers.
- Harlan McCaffery, MS, Statistician Intermediate
Mr. McCaffery earned his MS in Biostatistics with concentration in Statistical Methods and Practice from Northwestern University in 2018. He has expertise in statistical programming; data visualization; statistical consulting and study design; and the application of advanced statistical methods, including mixed effects modeling, nonparametric regression modeling, and survival data analysis. His applied research is on infant growth and development; obesity and the interaction of child/parent behaviors; and social stress in food-insecure adults.
- Julie Sturza, MPH, Statistician Lead
Ms. Sturza earned her MPH at the University of Michigan School of Public Health in 2007. She has worked as a statistician with multidisciplinary teams of researchers at UM since 2011, and has coauthored dozens of manuscripts on a range of pediatric health topics. Prior to her work at UM, she was an analyst at the US Environmental Protection Agency in DC. She has served as a consultant with the Woodson Program since 2015 and finds great joy in supporting the work of the passionate pediatric professionals at Michigan Medicine.
- Yujie Wang, MS, Statistician Intermediate
Ms. Wang earned her MS in Applied Statistics with concentration in human development and education from Teachers College, Columbia University in 2017. She has expertise in statistical programming, statistical modeling, data visualization and study design. Prior to her work at UM, she was an analyst at the National Center for Restructuring Education, Schools, and Teaching (NCREST) at Teachers College and provided consultation to the Michigan Department of Education (MDE) regarding to the quantitative research on the effectiveness of Early Middle College (EMC) programs in Michigan.
How it works
Collaborations range from brief consultation appointments to months-long comprehensive partnerships. Each investigator is provided a maximum of 20 free hours of statistical services per fiscal year as well as 20 free hours of data management per fiscal year. If an investigator wishes to utilize additional consulting services beyond the 20 hours of either stats or data services, there are options to do so.
The best time to involve our team in your project is during the initial phases of study design. Early involvement of a biostatistician and/or data architect can ensure that your research questions are well-defined, can improve your understanding of currently available health system data, services, and systems, and will help assure that your study design will allow for the testing of appropriate hypotheses. We recognize that this early involvement is not always possible, and are able to work with you at any stage of your project. However, involving our team too late in the process could mean that you are unable to meet your intended study goals or answer your intended research questions.
When we contribute to extramural grant applications, we are included as Co-Investigators on the applications, typically at a minimum of 10% effort per year. When we contribute to analyses, we are included as co-authors on manuscripts, presentations, or posters. To begin the consultation process, please submit the following FORM You will be contacted within a week of submission.
Timeframe guidance
Our team will do their best to complete work on your project as quickly as possible, but it should be understood that they routinely balance a large number of competing projects at any given time. Here are some basic time approximations for some common collaboration types:
- Study design: 2 to 4 weeks
- Data set gap analysis assessment: 1 to 2 weeks
- Data intake consultation: 1 to 3 weeks
- Data linking, cleaning, coding: 1 to 3 weeks
- Statistical analysis: 2 to 8 weeks
- Abstract: 2 to 4 weeks
- Grant preparation: 1 to 3 months
- Manuscript: 3 to 5 weeks after analysis is complete
- Protocol review: 1 to 3 weeks
In all cases, providing our team with well-defined research objectives/questions and clean datasets with accurate data dictionaries (in cases when data is already collected) will ensure that your project is completed as quickly as possible.
Questions may be directed to: [email protected]
* NOTE: This department initiative is funded by the Charles Woodson Fund for Clinical Research. Recipients/awardees must acknowledge support from the Charles Woodson Fund for Clinical Research in resulting publications, abstracts and presentations.