Weekly OHDSI Digest - 6Aug2018

MauraBeaton · August 6, 2018, 6:27pm

RESEARCH OPPORTUNITIES

Generalizability Network Research Study
Columbia University is looking for collaborators to run their Generalizability, Applicability, and Replicability of RCTs Study. To learn more about the study, check out @aaveritt’s presentation from the July 10th community call: https://drive.google.com/file/d/1acWhR5FCsTMWL7hxmYCgVlUt6r1bBFm1/view?usp=sharing
The study protocol is posted here: OHDSI/StudyProtocolSandbox/Generalizability
To participate in the study, please contact let @aaveritt know by replying to her forum post:
Generalizability, Applicability, and Replicability of RCTs: A Study

OHDSI MEETINGS THIS WEEK

OHDSI Community Call - Tuesday at 12pm ET
https://meetings.webex.com/collabs/#/meetings/detail?uuid=M59X2V1U61WC9ASID2Z5N3UT95-D1JL&rnd=811649.986822111212121
US TOLL: +1-415-655-0001
Meeting Number: 199 982 907

Population-Level Estimation (Eastern hemisphere) Workgroup Call - Wednesday at 3pm Hong Kong time
https://meetings.webex.com/collabs/meetings/join?uuid=M6WE9AOKFETH2VEFPVCZWWBIT0-D1JL

Architecture Working Group - Thursday at 10am ET
Webex: https://jjconferencing.webex.com/mw3100/mywebex/default.do?service=1&siteurl=jjconferencing&nomenu=true&main_url=%2Fmc3100%2Fe.do%3Fsiteurl%3Djjconferencing%26AT%3DMI%26EventID%3D610982452%26UID%3D501476547%26Host%3DQUhTSwAAAAQu4P1o9qm71JJ1Zj4-uvZbjQttsCinu71JCRxBAHAXnzjjRAiTspTzU9ojLmjMF4CcTBWw4zn1dqYPTWu5vJ9_0%26FrameSet%3D2%26MTID%3Dm3e1ceeca56f1e94c9fcf1ae98c10e02e

GIS Working Group Meeting - Next Monday (August 13th) at 10am ET

Meeting Number: 735 317 239
Password: gaia

ANNOUNCEMENTS

2018 Collaborator Showcase - DEADLINE TO SUBMIT ABSTRACTS TOMORROW AT 5PM
Time is running out to submit abstracts for the 2018 collaborator showcase. Collaborators are encouraged to submit abstracts for present posters, software demonstration or oral presentations during the collaborator showcase which will take place during the main symposium on Friday, October 12th. More details are available here: https://www.ohdsi.org/collaborator-showcase/

2018 OHDSI Symposium - REGISTER NOW
Registration is open for the 2018 OHDSI Symposium which will take place Friday, October 12th. You can register here: https://www.ohdsi.org/symposium-registration-2/

2018 OHDSI Symposium - TUTORIAL REGISTRATION OPEN
Registration is now open for tutorial sessions at this year’s OHDSI symposium. Intro tutorials will take place on Thursday, October 11th. Advanced tutorials will take place Saturday, October 13th. More information about tutorials is available here:
https://www.ohdsi.org/tutorial-workshops/
https://www.ohdsi.org/tutorial-registration-2/

Intro tutorials are being offered free of cost, however registration does not guarantee a seat in the tutorial. When you register, you will be placed on the tutorial wait-list. The final participant list will be determined by the tutorial faculty.

Advanced tutorials also offer the free wait-list registration. In addition, we are also offering a limited number of paid tickets ($318.17) which will guarantee your seat in the tutorial.

Call for Community Feedback
We’re developing official guidelines on how to start an OHDSI working group and OHDSI chapter and are eager for some community feedback. Guidelines can be found here:
Guidelines for starting an OHDSI chapter
Guidelines for starting an OHDSI working group

Nothing has really happened until it has been recorded.

Virginia Woolf

COMMUNITY PUBLICATIONS

De-identification of free-text descriptions of suspected adverse drug reactions using
deep learning

Empirical confidence interval calibration for population-level effect estimation studies in observational healthcare data.

ncbi.nlm.nih.gov

Empirical confidence interval calibration for population-level effect estimation studies in observational healthcare data.

MJ Schuemie, G Hripcsak, PB Ryan, D Madigan and MA Suchard, Proceedings of the National Academy of Sciences of the United States of America, 2018 13 03

Observational healthcare data, such as electronic health records and administrative claims, offer potential to estimate effects of medical products at scale. Observational studies have often been found to be nonreproducible, however, generating conflicting results even when using the same database to answer the same question. One source of discrepancies is error, both random caused by sampling variability and systematic (for example, because of confounding, selection bias, and measurement error). Only random error is typically quantified but converges to zero as databases become larger, whereas systematic error persists independent from sample size and therefore, increases in relative importance. Negative controls are exposure-outcome pairs, where one believes no causal effect exists; they can be used to detect multiple sources of systematic error, but interpreting their results is not always straightforward. Previously, we have shown that an empirical null distribution can be derived from a sample of negative controls and used to calibrate P values, accounting for both random and systematic error. Here, we extend this work to calibration of confidence intervals (CIs). CIs require positive controls, which we synthesize by modifying negative controls. We show that our CI calibration restores nominal characteristics, such as 95% coverage of the true effect size by the 95% CI. We furthermore show that CI calibration reduces disagreement in replications of two pairs of conflicting observational studies: one related to dabigatran, warfarin, and gastrointestinal bleeding and one related to selective serotonin reuptake inhibitors and upper gastrointestinal bleeding. We recommend CI calibration to improve reproducibility of observational studies.