User Tools

Site Tools


research:seawas

This is an old revision of the document!


Investigating Birth Month-Disease Risk Relationships Across the OHDSI Network

Objective: Derive data on birth month-disease risk relationships from large healthcare organizations/hospitals and healthcare facilities across the United States of America and throughout the world. Integrative analysis will be performed using established climate models, and knowledge of environmental factors that vary by season and climate.

Rationale: Many studies on birth month-disease relationships exist in the epidemiology literature. However, these studies are plagued by publication and disease selection biases. Therefore, we developed a high-throughput algorithm that uses Electronic Health Record data recorded during the clinical encounter and uncovers diseases with statistically significant birth month dependencies. These relationships are thought to be due to environmental exposures that occurring during development (i.e., either prenatal or perinatal effects). We call our algorithm SeaWAS: Season-Wide Association Study. However it is a “phenome-wide” approach whereby it investigates relationships between all diseases recorded in the EHR and birth month (assuming certain prevalence thresholds are met). Our algorithm runs in R and uses the OMOP CDM. Therefore, we are interested in implementing a new study that uses the OHDSI research network to run our analyses at multiple sites with different environmental exposures. This should enable us to tease out the relationship between the environmental factor and its contribution to increased or decreased disease risk.

Project Lead(s): Mary Regina Boland, George Hripcsak, Nicholas P Tatonetti

Coordinating Institution(s): Columbia University

Additional Participants (currently seeking additional collaborators): Patrick Ryan, Vojtech Huser, Adler Perotte, Rae Woong Park, Lisa Schilling, Nigam Shah, Marc Suchard, Jack Li, Usman Iqbal, Alex Nguyen

Several others have indicated interest and available data including Peter Rijnbeek, and Charles Bailey in April's Forum Discussion. Any interested collaborators please contact me (Mary Regina Boland) at mb3402@columbia.edu and I will be happy to add you to the collaborator list.

Participating Institution(s): Columbia University, Ajou University School of Medicine, Stanford University, University of Colorado, University of California, Los Angeles, Children's Hospital of Philadelphia (April Discussion on Forum), Erasmus University Medical Center (April Discussion on Forum), Taipei Medical University (TMU)

Full Protocol: OHDSI Climate/Birth Month Disease Risk Protocol

Initial Proposal Date: 5/16/2015

Launch Date: 8/18/2015

Complete Launch Date (after IRB approvals from various sites): Target 9/18/2015

Receive Results from Analysis Date: Target 10/15/2015

Study Closure Date: <fill out once finalized>

Results Submission: Email or SFTP

Customizable OHDSI-Formatted IRB

Requirements

CDM: V4 or V5 (both versions fully tested)

Table Accessed: CONDITION_OCCURRENCE, CONDITION_ERA, CONCEPT, PERSON

Database Dialects: SQL Server, Postgres, Oracle

Software: SQL and R

Code

Discussion

Datasets Run

  • CUMC/NYP
research/seawas.1443483601.txt.gz · Last modified: 2015/09/28 23:40 by mary_regina_boland