Senior Scientist - Data Strategy & RWE Architect

You are an “out-of-the box” thinker, innovative, collaborative, and passionate about making MCED accessible globally. To this end, you will operate as an expert strategic thought leader and serve as the end-to-end owner of real world data to real world evidence strategy and architecture, focusing on data access/capture, linkage/interoperability, processing, bridging data acquisition and analysis, iteration with stakeholders and presentation of results in a high quality and timely fashion.

In this exciting role you are an individual contributor joining our Real World Evidence team in Medical Affairs, with a dotted line into our Biostatistics and Clinical Data Management Department working within GRAIL’s Clinical Development organization.

This role can either be located in Menlo Park, California as a hybrid role with 2 days a week onsite or it can be 100% remote

Responsibilities

• Develop harmonized data models/ data dictionaries that guide the collection, processing, linkage, integration and analyses of healthcare RWD from disparate sources (EHRs, claims, cancer & death registries, PROs, etc.).
• Evaluate new and existing data sources for fitness & feasibility to unlock new opportunities.
• Ensure data quality, privacy and security through pre-defined data quality review documentation, and pre-established data governance policies/procedures.
• Explore development of new approaches to RWD capture through GRAIL-sponsored studies, registries, clinical surveillance and/or other RWE activities.
• Collaborate with cross functional stakeholders to develop and implement shared data processing strategies, state-of-the-art statistical methods and machine learning algorithms (e.g. NLP) to enable RWE generation, interpretation & visualization from global datasets, in an interpretable, reliable fashion.
• Collaborate with Biostatistics group in developing statistical methods used in a RWE study setting and causal inference for RWE (e.g., propensity score matching/weighting/subclassification, external control arms, count data regressions, survival analysis).
• Remain up-to-date on the latest advancements in observational research methods and their applications; healthcare data sources;
  Findable, Accessible, Interoperable, and Reusable (FAIR) Data Principles; data privacy; etc.
• As a member of a cross-functional team, GRAIL’s Principal Scientist, RWE will build and maintain cross-functional relationships with Medical Affairs, Clinical Development, Regulatory, and business teams; and collaborate with the team effectively to ensure the implementation of rigorous methods and delivery of scientifically sound results, in support of regulatory submissions, publications and/or other uses.
• Drive the evaluation of technical strengths/weaknesses, work to minimize the limitations of RWD, and influence the decision-making process.
• Support the development of appropriate sections within study protocols, statistical analysis plans, data management plans, data quality review documentation, etc.
• Collaborate with external partners (e.g., key opinion leaders, academic & community health systems and networks, CROs, vendors, etc.) on the design and execution of RWD/E studies.
• As needed, support management of data acquisition and/or analysis vendors and external partners.
• Contribute to scientific publications, conference presentations, and internal knowledge sharing activities to disseminate research findings and promote scientific excellence within the organization.
• Mentor junior members of the team.

Preferred Qualifications

• Ideal candidate will have a Ph.D. with 12+ years relevant experience or M.S. with 15+ years relevant experience in data science, applied mathematics, biostatistics, epidemiology or similar. Experience in Oncology or diagnostic devices is strongly preferred.
• Expertise in R or Python is required. Direct experience in SQL is preferred.
• Solid understanding of statistics and machine learning techniques; and integrating ML/NLP and other automated technologies into data processing design.
• Deep knowledge of the journey of real world data→ real world evidence, such as challenges in data access, de-identification, missingness, data…