White Papers

Important:

These original OMOP resources are no longer supported. You can find the latest developments by visiting the OHDSI website.


The purpose of OMOP white papers is to inform you on the processes and decisions or the way that we choose to go for a particular problem or question during the OMOP research. It is not to demonstrate best practice, but to give OMOP readers a single place to be informed while the research is in progress. From time-to-time updates and new papers will be posted.

If you have any questions or feedback on the white papers, please contact OMOP.

OMOP White Papers

Review of Observational Analysis Methods - Feb 2009
An important component of the OMOP effort is to develop a central repository of potential methods and their characteristics to facilitate the structure and development of protocol concepts. An initial list was developed by soliciting contributions and supporting publications from a workgroup, informal literature reviews, and informal reviews of presentations at relevant meetings. These findings are reported with the Review of Observational Analysis Methods and Methods Matrix.

Review of Observational Analysis Methods

Methods Matrix
................................................................................................................................................................................................................................

Establishing a Drug Era Persistence Window for Active Surveillance - January 2010
The construct "Drug eras" are time periods of continuous drug exposure used in active drug safety surveillance. This paper discusses how OMOP went about to define the optimal window of persistence using two scenarios, a 0-day to a 30-day tolerance window of persistence.


................................................................................................................................................................................................................................

Establishing a Condition Era Persistence Window for Active Surveillance - January 2010
The construct of ‘condition era’ tables is a means to systematically apply consistent rules for all medical conditions to infer distinct episodes of care from available information, such as diagnosis codes and problem lists. One important decision to make when applying condition era logic to an observational database (administrative claims or electronic health record) is to define the persistence window, or the period of time when two related conditions would be considered part of the same episode of care. This analysis evaluates the impact of the 0-day and 30-day persistence window on the construction of condition eras across the central databases within the OMOP data community.


................................................................................................................................................................................................................................

Defining a Reference Set for Evaluating the Performance of Active Surveillance Methods - January 2010
The OMOP research requires a reference set to provide a pre-defined benchmark to base comparisons of methods against. In this context, the reference set is a list of test cases of drug-condition pairs classified dichotomously as ‘true associations’ or ‘negative controls’. Methods will be executed against these test cases, and the estimates derived from the methods will be compared to the benchmark. This paper describes the OMOP reference set used for each analysis program and the rationale that was used to develop them.

................................................................................................................................................................................................................................

Selecting Comparators in Active Surveillance Analyses - March 2010
The OMOP research requires a set of comparator drugs. The desire is to construct a comparator which is sufficiently similar to the target group that effects observed may be plausibly attributable to the exposure of interest. Analysts take great pains to define study inclusion / exclusion criteria when extracting the data and apply additional analysis techniques (such as matching, stratification, and multivariate modeling) to minimize effects of confounding and other biases. Such design decisions are pivotal to the final result, but can be resource intensive and require subjective expert opinion. This white paper proposes ways to develop and evaluate alternative approaches to comparator selection.

................................................................................................................................................................................................................................

Applying Natural Language Processing to Extract Codify Adverse Drug Reaction in Medication Labels - August 2010
Authors: Jeffrey Friedlin, D.O and Jon Duke, M.D.
Institutions: Indiana University School of Medicine and Regenstrief Institute Indianapolis, Indiana 46202

Medical informaticists developed an automated method of identifying, extracting and codifying adverse drug reaction data contained in the Structured Product Labels for drugs to be studied as part of the OMOP project.

................................................................................................................................................................................................................................

OMOP Distributed Partner Research Reports - February 2011
Authors: OMOP Distributed Research Partners

The study reports were produced by each OMOP Distributed Research Partner to document and discuss site specific aspects of the OMOP research and findings. Each report also includes discussion regarding the performance of methods and lessons learned. Click on the Distributed Research Partner to download their study report.

1. Department of Veterans Affairs Center for Medication Safety / Outcomes Research, Pharmacy Benefits Management Services

2. Humana, Inc.

3. Partners Healthcare System

4. Regenstrief Institute / Indiana University

5. SDI Health

................................................................................................................................................................................................................................

Exploration of Four Outcomes: Outcomes and labeling information, in conjunction with other evidence - May 2011
Authors: Jeffrey Friedlin, D.O and Jon Duke, M.D.
Institutions: Indiana University School of Medicine and Regenstrief Institute Indianapolis, Indiana 46202