Post Doctoral Medical Fellow - Applied Biostatistics, Clinical Trial Simulation

Description

The gBDS / Med Data/AI team at Boehringer Ingelheim is seeking a Postdoctoral Research Fellow to help advance the development of robust clinical trial simulation tools that improve the probability of success by integrating multimodal datasets. This applied research and implementation role focuses on translating rigorous statistical methodology from research into confirmatory, pivotal clinical studies.

It is an exceptional opportunity for candidates wanting to bridge academia and industry from a clinical statistical perspective, working at the intersection of methodological rigor and real‑world drug development impact.

As an employee of Boehringer Ingelheim, you will actively contribute to the discovery, development, and delivery of our products to patients and customers. Our global presence and collaborative environment support meaningful, mobility‑friendly, and work‑life‑balanced work.

You will work with considerable autonomy, taking ownership of projects with scientific guidance available from supervisors and the broader team. Core responsibilities include:

• Integrate multimodal data sources across trial operations and clinical outcomes to parameterize and calibrate simulations (internal trial data + operational metrics + external benchmarks where appropriate).
• Evaluate and validate clinical trial simulation frameworks that quantify how operational and clinical factors (e.g., recruitment pace, dropout, missing data, site heterogeneity, nonadherence, endpoint variability, protocol deviations) impact probability of success.
• Build simulation models incorporating relevant data sources (clinical trial, real‑world evidence) for scenario‑based planning and integration via simulation.
• Build scenario engines for trial execution planning (e.g., “what‑if” analyses on site mix, monitoring intensity, recruitment strategies, data cleaning rules, visit schedules, or enrichment approaches) and translate results into actionable decision support.
• Implement and compare statistical methods relevant to trial robustness under execution realities, including:
  • missing data mechanisms and sensitivity analyses (MAR/MNAR)
  • Bayesian models (meta‑analytic, hierarchical, etc.)
  • intercurrent events and estimand‑aligned strategies
  • site‑level/random effects and cluster heterogeneity
  • treatment effect attenuation due to nonadherence
  • informative dropout and time‑varying covariates
• Conduct simulation studies to evaluate design choices (sample size, randomization schemes, interim decision rules, adaptive elements, endpoint definitions) and predict performance metrics such as power, type I error, bias, estimation bias, predictive probabilities, and other operating characteristics.
• Create reusable, reproducible simulation pipelines (R/Python) for simulation, diagnostics, and reporting—version‑controlled, well‑tested, and well‑documented for long‑term sustainability.
• Communicate results clearly to cross‑functional stakeholders via manuscripts, conference abstracts, internal whitepapers, and presentations.

• Ph.D. in Biostatistics or Statistics (or closely related quantitative discipline) from an accredited institution, awarded prior to start date.
• Strong foundation in clinical trial methodology and inference (GLMs, mixed models, survival analysis, longitudinal models, causal/estimand thinking, multiplicity basics, Bayesian experience).
• Demonstrated ability to program in R and/or Python for statistical modeling and simulation; experience with reproducible workflows (Git, unit testing, code review, literate reporting such as Quarto/RMarkdown/Jupyter) preferred.
• Familiarity with Monte Carlo simulation, power/operating characteristic evaluation, and principled model checking/validation.
• Interest or experience working with multimodal trial data including operational metrics (site performance, enrollment curves, deviations) and clinical outcomes data; real‑world or observational health data experience is a plus.
• Ability to work in a highly collaborative environment and communicate statistical results to non‑statistical stakeholders with clarity and pragmatism.
• Demonstrated capacity to work independently, manage…