Fellows are nominated by a Harvard-MIT Center for Regulatory Science faculty member, and the application is submitted to the Fellowship Committee for consideration. The application packet consists of the candidate’s CV, a written statement from the candidate, and a project description jointly developed between the candidate and mentor(s).

For more information, contact the fellowship director Dr. Florence Bourgeois at

Available Fellowship Opportunities

Drug-drug interaction detection in real-world claims data and healthcare records

The Harvard Program in Therapeutic Science (HiTS) invites applications for a fellowship position at the Harvard-MIT Center for Regulatory Science. The fellow will work with a multidisciplinary team to develop and evaluate novel computational methods to support automated detection of drug-drug interactions in large and disparate healthcare datasets.

For this position, the Fellow will work with faculty at Harvard in the Department of Biomedical Informatics The fellow will be based in the Avillach Lab, an interdisciplinary research group focused on the development of cutting edge techniques for the integration of multiple heterogeneous biological and clinical cohorts with the overarching goal of supporting individualized data-driven healthcare.

Project Details
The application of healthcare claims data and electronic health records for use in pharmacoepidemiological studies is booming. Drug–drug interactions can be identified by mining such databases, including adverse drug reactions (ADRs) resulting from the prescription of a combination of drugs (CADRs), considered a special class of drug-drug interactions. While many ADRs can be anticipated and hence avoided, CADRs are highly complex and challenging to identify and, as a consequence, have not been well studied. CADRs may represent as many as 30 % of unexpected ADRs and their elucidation is of particular interest to the FDA. The fellowship project will develop new signal detection algorithms using data mining techniques in real-world healthcare data derived from claims and electronic healthcare records. The project will aim to identify potential pharmacovigilance signals concerning drug-drug interactions across multiple diseases and syndromes.

Research activities include, but are not limited to:

  • Extract and analyze information from a nationwide claims database including 67 million subjects
  • Extract and analyze information from healthcare records in an academic pediatric hospital including 1.8 million patients
  • Develop and test methods to detect adverse drug reactions from billing codes, drug prescriptions and lab value results
  • Disseminate findings through national and international conferences and in peer-reviewed scientific publications

Basic Qualifications

  • PhD in computer science, engineering or a related field
  • Expertise in SQL query language
  • Expertise in R or Python programing languages

Application Procedure and Requirements
Applications will be accepted until the position is filled. Please include the following when applying:

  • Curriculum vitae
  • Cover letter and a 2-page description of relevant experience as a single PDF
  • DOI or PMCID of up to three relevant publications

Selected candidates will be asked to provide letters of reference.

Position Description
Host Institution: Harvard Medical School
Appointing Department: Harvard Program in Therapeutic Science
Research Laboratory: Paul Avillach
Location: Boston, MA
Category: Scientist
Other responsibilities: no administrative or teaching obligations
Duration of fellowship: 2 years

Harvard Medical School is an Equal Opportunity/Affirmative Action Employer.
Women and minorities are especially encouraged to apply.

For additional information contact
Paul Avillach, MD, PhD
Assistant Professor
Department of Biomedical Informatics | Harvard Medical School
Children’s Hospital Informatics Program | Boston Children’s Hospital