Compatibility

Compatibility

DarwinShinyModules is compatible with other modules or shiny code. One module that you might have noticed in the documentation is the OhdsiModule. The OhdsiModule is a module that wraps around the xViewer and xServer functions. This enables you to easily integrate already existing modules from OhdsiShinyModules into your shiny apps built with DarwinShinyModules.

The base setup to do this would look something like this:

# Create connection details to the results database
connectionDetails <- DatabseConnector::createConnectionDetails(
  dbms = "sqlite",
  server = "./results.sqlite"
)

# Create a ConnectionHandler object
connectionHandler <- ResultModelManager::ConnectionHandler$new(
  connectionDetails = connectionDetails
)

# Make our module
estimation <- OhdsiModule$new(
  connectionHandler = connectionHandler,
  viewerFun = estimationViewer,
  serverFun = estimationServer,
  resultDatabaseSettings = createDefaultResultDatabaseSettings()
)

We can then either preview the module

DarwinShinyModules::preview(estimation)

Launch a shiny app using the appStructure

appStructure <- list(
  Estimation = estimation
)

DarwinShinyModules::launchDarwinDashboardApp(appStructure)

Or integrate it in a bespoke shiny app

server <- function(input, output, session) {
  # Do bespoke stuff ...
  estimation$server(input, output, session)
}

ui <- shiny::fluidPage(
  # Bespoke UI ...
  estimation$UI()
)

shiny::shinyApp(ui, server)

Example using CohortMethod

A full example using CohortMethod to run a comparative effectiveness analysis of Celexocib vs Diclofenac, with outcome GiBleed is shown below.

Run the analysis

library(DatabaseConnector)
library(Eunomia)
library(CohortMethod)

connectionDetails <- Eunomia::getEunomiaConnectionDetails()
Eunomia::createCohorts(
  connectionDetails = connectionDetails,
  cdmDatabaseSchema = "main",
  cohortDatabaseSchema = "main",
  cohortTable = "cohort"
)

covarSettings <- FeatureExtraction::createDefaultCovariateSettings(
  excludedCovariateConceptIds = c(
    # Celexocib
    1118084,
    # Diclofenac
    1124300
  ),
  addDescendantsToExclude = TRUE
)

tcos <- list(CohortMethod::createTargetComparatorOutcomes(
  targetId = 1,
  comparatorId = 2,
  outcomes = list(CohortMethod::createOutcome(
    outcomeId = 3,
    outcomeOfInterest = TRUE,
    trueEffectSize = NA
  ))
))

getDbCmDataArgs <- CohortMethod::createGetDbCohortMethodDataArgs(
  studyStartDate = "",
  studyEndDate = "",
  covariateSettings = covarSettings
)

studyPopArgs <- CohortMethod::createCreateStudyPopulationArgs(
  firstExposureOnly = FALSE,
  restrictToCommonPeriod = FALSE,
  removeDuplicateSubjects = "keep all",
  removeSubjectsWithPriorOutcome = TRUE,
  minDaysAtRisk = 1,
  startAnchor = "cohort start",
  riskWindowStart = 0,
  endAnchor = "cohort end",
  riskWindowEnd = 30
)

psArgs <- CohortMethod::createCreatePsArgs()

matchPsArgs <- CohortMethod::createMatchOnPsArgs(caliper = 0.2)

outcomeModelArgs <- CohortMethod::createFitOutcomeModelArgs(modelType = "cox")

analysis <- CohortMethod::createCmAnalysis(
  analysisId = 1,
  description = "Celecoxib vs Diclofenac",
  getDbCohortMethodDataArgs = getDbCmDataArgs,
  createStudyPopArgs = studyPopArgs,
  createPsArgs = psArgs,
  matchOnPsArgs = matchPsArgs,
  computeCovariateBalanceArgs = CohortMethod::createComputeCovariateBalanceArgs(),
  computeSharedCovariateBalanceArgs = CohortMethod::createComputeCovariateBalanceArgs(),
  fitOutcomeModelArgs = outcomeModelArgs
)

outputFolder <- file.path(tempdir(), "cm-test-output")

refTable <- CohortMethod::runCmAnalyses(
  connectionDetails = connectionDetails,
  cdmDatabaseSchema = "main",
  exposureDatabaseSchema = "main",
  exposureTable = "cohort",
  outcomeDatabaseSchema = "main",
  outcomeTable = "cohort",
  outputFolder = outputFolder,
  cmAnalysisList = list(analysis),
  targetComparatorOutcomesList = tcos
)

Export results

exportFolder <- file.path(tempdir(), "cm-test-export")

CohortMethod::exportToCsv(
  outputFolder = outputFolder,
  exportFolder = exportFolder,
  databaseId = "Eunomia"
)

Upload results to database

cg_cohort <- data.frame(
  cohortId = c(1, 2, 3),
  cohortName = c("celexocib", "diclofenac", "GiBleed"),
  sql = "",
  json = ""
)

resDb <- file.path(tempdir(), "cm-test-res.sqlite")

resConnectionDetails <- DatabaseConnector::createConnectionDetails(
  dbms = "sqlite",
  server = resDb
)

CohortMethod::uploadExportedResults(
  connectionDetails = resConnectionDetails,
  databaseSchema = "main",
  append = FALSE,
  exportFolder = exportFolder,
  cohorts = cg_cohort
)

Make a shiny module using DarwinShinyModules and OhdsiShinyModules

library(DarwinShinyModules)

connectionHandler <- ResultModelManager::ConnectionHandler$new(connectionDetails = resConnectionDetails)

estimation <- DarwinShinyModules::OhdsiModule$new(
  connectionHandler = connectionHandler,
  viewerFun = OhdsiShinyModules::estimationViewer,
  serverFun = OhdsiShinyModules::estimationServer
)

appStructure <- list(
  Comparative_Effectiveness = estimation
)

# Default themed
DarwinShinyModules::launchShinydashboardApp(appStructure, title = "celecoxib vs diclofenac")

# DARWIN themed
DarwinShinyModules::launchDarwinDashboardApp(appStructure, title = "celecoxib vs diclofenac")