The task of a research idea from design through execution through publication can seem a daunting challenge, much like rolling a boulder up a hill. That task is all the more challenging when researchers try to go it alone, as each step requires a distinct set of skills. Observational study design requires epidemiologic understanding and statistical methodological expertise. Implementing a study design requires statistical programming ability. Interpreting and reporting results requires domain knowledge of the clinical problem.
But when you are part of the OHDSI community, you never have to go it alone. And as a team effort, what seems a arguous task can become an efficient and effective process.
We are seeking good research questions that you want to take from idea to publication. The OHDSI community will provide support through every step of the process, working with you to design an appropriate protocol, implement a network analysis package, execute across OHDSI data partners, and prepare a manuscript for publication. Our goal is to see your project through in 1 month or less, utilizing the open-source tools and process that OHDSI has established.
Our first SOS Challenge will be a population-level effect estimation study, using the comparative cohort design. Using the OHDSI framework, research questions can be expressed in the following form:
To compare the risk of [insert outcome here] between [insert target cohort here] and [insert comparator cohort here], we will estimate the population-level effect of exposure on the [insert relative risk metric] of the outcome during [insert time-at-risk here].
Example questions from our comparative cohort analysis tutorial:
-
To compare the risk of incident ischemic stroke, observed in inpatient setting between dabigatran new users with prior atrial fibrillation and warfarin new users with prior atrial fibrillation, we will estimate the population-level effect of exposure on the hazards of the outcome during the period from 1 days from cohort start date to 0 days from cohort end date.
-
To compare the risk of upper gastrointestinal complication (UGIC) events between celecoxib new users and diclofenac new users, we will estimate the population-level effect of exposure on the rate of the outcome during the period from 0 days from cohort start date to 0 days from cohort end date.
We are particularly interested in supporting junior researchers looking to conduct their first network analysis, but all OHDSI collaborators are welcome and encouraged to submit their ideas. If you are interested in participating in the SOS Challenge, please complete this form: SOS Challenge Form, where you’ll be asked to provide your research question and a statement about why you think your question is clinically important and should be answered by the OHDSI network. If your research idea is selected, then in exchange for the support, you agree to share your experience on the weekly OHDSI community calls so others can learn alongside you as the process unfolds.
We look forward to collaborating with you on this SOS challenge.
One must imagine Sisyphus happy.
- Albert Camus