Job Description

Job Summary
The Department of Surgery at the University of Michigan is committed to creating a positive, substantive impact on human health, through collaboration among clinicians, scientists, medical and business professionals to achieve innovative science and meaningful discovery.

The individual in this position will work under the direction of Dr. Ryan Howard. Dr. Howard is a minimally invasive surgeon and health services researcher at Michigan Medicine. His research focuses on studying the quality and safety of surgical care. To do this, he uses large healthcare datasets to analyze trends in care delivery, the costs of care, and patient outcomes. Some of his specific areas of interest include long-term patient outcomes after hernia and bariatric surgery, the comparative effectiveness of medical and surgical treatment of obesity, and postoperative opioid prescribing.

The statistician in this position will be integral to executing research projects by performing organization and maintenance of large administrative datasets, statistical programming (i.e., coding), data analysis, and summarization and interpretation of the results. The individual in this position will also help determine the appropriate analytic methodologies for specific projects. In this role, this individual will support and work alongside Dr. Howard, his research collaborators, and various trainees working on these research projects.

The statistician in this position should be comfortable cleaning large datasets and creating cohort files. This position requires management of very large data sets including Medicare, Medicaid, MarketScan, Truven, regional insurance claims, and patient registries.

Responsibilities*
The Statistician will work closely with the research team to:

understand each research project and propose appropriate design and methodology;
write, test, and implement analytic code using Stata, R, Python, or SAS to clean, manage, merge, and analyze data;
develop statistical analyses for presentation and discussion with the principal investigator and others;
contribute to study reports or manuscripts by preparing tables, graphs and other data displays
create and maintain technical documentation;
contribute to the development of methods to evaluate progress of data collection on new projects;
participate as a team member in discussions on analysis and improvement of data collection, quality of data analyses, programming and documentation;
other duties as assigned
Required Qualifications*
Degrees in related field with strong computing and quantitative skills will be considered
Bachelors degree in related field, masters degree strongly preferred
3-5 years experience using statistical analysis software (e.g. Stata, R, Python, SAS)
Possesses and able to apply a broad knowledge of principles, practices, and procedures of health services research to the completion of difficult assignments
Must be able to program, manage data, and conduct statistical analysis
Desired Qualifications*
Masters degree in Statistics, biostatistics, Public Health, or related field
Experience with coding in at least two of the following; Stata (preferred), Python, R, or SAS
Experience performing multivariable and multilevel regression modeling (experience performing difference-in-differences and instrumental variables analysis is a plus)
Experience supporting, managing, and analyzing major research projects using large administrative datasets such as Medicare, Medicaid, and commercial insurance is preferred
Experience with database management and complex data structures and linkages between data sources including understanding medical procedure codes
Excellent interpersonal and written communication skills
Experience working with teams and collaborating
Working knowledge of epidemiologic concepts and methodology
Strong programming knowledge using a high-level programming language
Ability to draft statistical tests in accordance with project designs and to solve analytic problems, including descriptive statistics, linear regression, logistic regression, survival analysis, mixed models with repeated measures, multi-level modeling