Skip to content

ohdsi-studies/MyositisNetworkStudy

Repository files navigation

OHDSI Network Study: Dermatomyositis Phenotype Evaluation

This is an OHDSI study being coordinated by the Johns Hopkins Myositis Center. Please see our post on the OHDSI forum for more information.

The Coordinating site is currently in the process of obtaining IRB approval through the Johns Hopkins Medicine IRB. The JHU Application is IRB00373992 and the JHU PI is Dr. Christopher Mecoli, MD. Participating data partners should ensure that ethics board or IRB approval has been obtained as appropriate.

To participate in this network study:

  1. Review the study protocol found in the documents directory and ensure that your site meets the inclusion criteria.
  2. Contact the study team for a submission link.
  3. Follow these instructions for setting up your R environment, including RTools, devtools, renv, and Java.
  4. Clone the repository git clone https://github.com/ohdsi-studies/MyositisNetworkStudy.git or using the RStudio GUI.
  5. Open your study package in RStudio. Use renv::restore() to install dependencies. Tip: On windows UNC paths, Sys.setenv(R_INSTALL_STAGED = FALSE) my be useful if encountering permission denied errors while installing dependencies using renv.
  6. In RStudio, select 'Build' then 'Load all' to build the package and load it into memory using devtools.
  7. Modify the 'env setup' section of the extra/CodeToRun.R script to connect to your CDM and set metadata for the execution. A file zip file with the results and meta data will be created
  8. Execute the modified CodeToRun.R script
  9. Submit the generated zip file to the coordinating study team using the OneDrive link provided.
Element Description
Study Title Creation and Evaluation of Dermatomyositis Phenotypes Across Different Database Sources
Study Status Study Status: Started
Next Steps Generate the 8 phenotypes on multiple external OMOP CDMs (at least 1 claims and 1 EHR based CDMs) and run Phevaluator on external CDMs
Research Question How do dermatomyositis phenotypes perform across different data sources (EHR, claims, registries) that may have differences in how the OMOP ETL was performed?
Uncertainty We remain uncertain as to how our phenotypes will perform across other database types (claims, other EHR, etc).
Study type Clinical Application.
Tags OHDSI Myositis
Study lead Dr. Christopher Mecoli, MD
Study lead forums tag Christopher_Mecoli
Study team See protocol
Phenotype Development Created and tested 8 myositis phenotypes with Phevaluator using Johns Hopkins OMOP CDM
Phenotype Evaluation Currently under development process. This is the purpose of the study
Cohort Definitions Myositis Phenotype Cohort IDs are defined in the cohorts directory. These can also be viewed on the OHDSI Atlas demo instance; ID: 1781804,1788567,1787425,1788503,1789031,1789032,1788875,1789289
Cohort Diagnostics Currently under development process. (A hyperlink to the R Shiny app where the cohort diagnostics results can be viewed.)
Analysis Specifications Develop a OHDSI protocol to run a multi-centric patient level prediction study on complication comparisons among different drug use for dermatomyositis patients.
HADES Packages DatabaseConnecter, CohortDiagnostics, Phevaluator
Study Sites Pending, but likely to include Stanford, Columbia, and other datasources accessed through J&J
Results explorer N/A
Study start date Dec 11, 2023
Study end date Not completed
Protocol See documents directory
Publications N/A

Study Progress

Study Attribute Value
IRB materials sufficient for review Yes
Cohort definition(s) available Yes
Data partner recruitment status Ready
Deadline for adding new data partners N/A
Protocol building team recruitment status ready
Deadline for adding new protocol building team members N/A
Manuscript preparation team recruitment status Ready
Deadline for adding new manuscript preparation team members N/A

About

OHDSI Network Study lead by John Hopkins Myositis Group

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 4

  •  
  •  
  •  
  •  

Languages