OHDSI Home | Forums | Wiki | Github

OHDSI global survey of data recency to support Covid-19 research...please respond!

(Patrick Ryan) #1


I have received several questions, including from our friends at the EMA and FDA, about the recency of databases within the OHDSI network. The basic thing on everyone’s mind: how long will we have to wait until data partners begin to capture Covid-19 positive cases and associated outcomes? During our OHDSI study-a-thon this week, we will be focused on answering research questions that can be addressed through retrospective analysis of existing data, but we’ll also be designing analysis packages to be prepared to execute studies as Covid-19 data accumulate.

If you have access to a patient-level observational dataset, please complete the following survey: https://forms.gle/cyz3ZXHRsDozNUREA.

The survey asks for basic information about you and the database you have access to:

  • database name
  • geography (country, region)
  • data type (e.g. claims, EHR, registries)
  • does it capture outpatient care?
  • does it capture inpatient hospital admission?

then it asks a couple questions about the current dataset you currently have access to:

  • Date when data was last prepared for analysis
  • Date when the data starts (e.g. first observation)
  • Date of the most recent data (.e.g. last observation)
  • OMOP CDM version
  • OMOP vocabulary version

and a few questions about your future access to data:

  • What is your refresh cycle of the source data (ex. daily, weekly, monthly, annually)?
  • What is the expected lag between the next refresh data and the data of the latest observation?
  • When will the next data refresh be completed?
  • What will be the month/year of the last observation in the next refresh?

Your timely response is greatly appreciated. This information will be used to inform our OHDSI study planning activities and will be shared with public health officials on request, but does not represent any expectation or requirement for participation in any future activities.

(Nigel Hughes) #2

Hi Patrick

You may want to post this request also via ProMED mail, a global mail server for infectious diseases and outbreaks, run by the International Society for Infectious Diseases: https://promedmail.org/coronavirus/

They can be contacted via: info@isid.org

They service thousands of specialists in 32 countries and if they post a request it will more than likely be seen by some who may have useful clinical data.



(Christian Reich) #3


Just put the IQVIA data in. Encourage everybody to do the same. This is the network effect. The more people declare their participation the more will declare their participation, which will increase our chance to create meaningful results on the Covid-19.

(Patrick Ryan) #4

Thanks @nigehughes, yes, to reinforce, if anyone knows of others with access to observational data which could inform the current situation, please pass along this survey and encourage them to respond. The more complete the responses we collect, the more informed our public health agencies can be about when to expect real-world evidence to contribute to their decision-making.

(Vojtech Huser) #5

I think the most consent-ready and infrastructure ready are AllOfUs sites and use of AllOfUs umbrella.

I reached out to Josh Denny (and Eric Dishman) about 10 days ago asking about similar questions you posted in the first post and how many COVID-19 cases can the core AllOfUs team see as of now. In his reply, he indicated that they are thinking hard of ways to use AllOfUs.

I predicted prospective OHDSI COVID-19 focus several days ago at our steering committee call. With respect to my other comment about emergency and research - South Korea (per economist article) created a special provision to access EHRs. I did not find any other details other than this sentence from this Economist article (published March 21):

South Korea changed the law to allow the state to gain access to medical records and share them without a warrant.

It would be interesting to see more details on this from Korean researchers here.
link https://www.economist.com/leaders/2020/03/19/paying-to-stop-the-pandemic

(Luca Foschini) #6

This a great initiative!

I’ve listed a dataset collected at Evidation as part of participatory flu surveillance program. It consists of a weekly 26-item questionnaire around ILI symptoms and treatment that has been answered at least once by 340k+ Americans so far this year. We get 10k+ new reports every day, many of which are from recurring responders.

The data is consented to be shared with academic research centers, but it is NOT in OMOP CDM format. If anyone is interested in helping us map questionnaire responses to OMOP CDM (if at all possible), or have other ideas to collaborate, including pursuing more prospective data collection, please feel free to reach out: luca@evidation.com



Is there an OHDSI ambassador who wants to contact „ The collaborative international ICU registry for critically ill COVID-19 patients“ to ask for their participation? https://sites.google.com/view/risc-19-icu/main-page

(Christophe Lambert) #8

I just requested access, but it says it could take a few days to process. It may require depositing intensive care unit data, which could be a challenge to get authorization here.