Biostatistician 2 - 70759
Description The Department of Cardiothoracic Surgery at Stanford University is conducting a search for a full-time Biostatistician to support clinical research activities of the Departments Faculty. This Faculty Statistician will work with Department Faculty on a number of initiatives that are in need of statistical support and leadership.
The candidate will:
- Lead the statistical analysis for a variety of studies.Clinical efforts will include maintenance, manipulation, and analysis of clinical research datasets that will include both single-institution in-house patient populations, cohorts of patients assembled through multi-institution collaboration, and large-scale administrative and registry databases.
Supervise the conduct of the Cardiovascular Surgery clinical research database, including integrity checks and developing guidelines and rules for database error identification.
Examples of the large-scale databases that will be utilized include but are not limited to the Surveillance, Epidemiology, and End Results (SEER) program, the SEER-Medicare linked database, the Nationwide Inpatient Sample (NIS), the National Cancer Database (NCDB), the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP), and CMS MEDPAR.
Utilize conventional statistical methods including logistic regression analysis, proportional and non-proportional hazard modeling, propensity score analysis, and actual (or observed cumulative frequency) survival analyses to supplement the typical non-parametric actuarial methods for the evaluation of:
- Short-term perioperative outcomes of morbidity and mortality following cardiovascular and thoracic surgical treatment, and identification of important factors related to these outcomes.
Incorporate novel statistical methods and cost analysis into the above studies as indicated, including via interfacing and collaboration with other statistical and modeling experts both within Stanford University and at external institutions.
- Long-term survival after treatment of cardiovascular and thoracic surgical diseases.
Develop and apply novel biostatistical methods to clinical investigations. This includes: 1.) Parametric modeling to decompose multiple time-varying hazards using a multiphase hazard model in the time domain, incorporating the effects of competing risks; 2.) Computer-intensive machine learning (or bagging) methods with bootstrap aggregation; random survival forest (RSF) analysis (nonparametric statistical ensemble method that utilizes all variable data without advance knowledge of the relationship [linear, nonlinear] of a variable over time or whether interactions exist); 3.) Nelsons cumulative event function to obtain nonparametric estimates; and, 4.) Multiple imputation using a Markov Chain Monte Carlo technique to impute missing values (SAS PROC MI).
- Long-term survival and freedom from adverse events after treatment of cardiovascular and thoracic surgical diseases.
Take a major role in applications for external funding, through support in developing, writing, and reviewing the statistical components of grants.
Participate in strategy sessions for designing and implementing studies, including proposals for analysis of external databases such as those maintained by the Society of Thoracic Surgeons for adult cardiac surgery patients, congenital cardiac surgery patients, and general thoracic surgery patients.
Upon completion of analyses, actively participate and collaborate with the cardiothoracic surgical faculty in preparation of both presentations at major thoracic surgical meetings and publications in major clinical and scientific journals based on the results.
- A Masters or PhD degree in statistics, biostatistics, epidemiology, or a related field.
- Three years or more of productive experience as a collaborating statistician on a variety of clinical study - designs, preferably in the cardiovascular realm, with previous publications in the field.
- Knowledge and experience in SAS/SPSS/R/STATA or equivalent software/programming environments.
- Knowledge of acquiring and maintaining publically available databases.
- Highly self-motivated individual, enthusiastic about scientific discovery and able to collaborate closely and effectively with other members of a research team.
- Dedication to teaching of clinical residents and fellows and the laboratory post-doctoral research fellows.
- Excellent communication skills (verbal, written, and presentation).
- Professional knowledge of biomedical research and new biostatistical methods
Job: Information Analytics
Location: School of Medicine
Job Code: 5522