compute the IC statistics — calculateStatisticsIc • IcTemporalPatternDiscovery

Computes the IC statistics.

calculateStatisticsIc(
  ictpdData,
  multipleControlPeriods = "110",
  multipleRiskPeriods = "10000",
  shrinkage = 0.5,
  icPercentile = 0.025,
  metric = "IC025"
)

Arguments

ictpdData: An object containing the counts, as created using the getDbIctpdData function.
multipleControlPeriods: Defines the control periods to use where 100 means the control period defined by controlPeriodStart/End, 010 means the period -30 to -1 day before prescription and 001 means the control period on the day of prescription
multipleRiskPeriods: Defines the risk periods to use 10000 is 1-30 days, 01000 is 1 to 360 days, 00100 is 31 to 90 days, 00010 is 91 to 180 and 00001 is 721 to 1080 days after prescription default is '10000'
shrinkage: Shrinkage used in IRR calculations, required >0 to deal with 0 case counts, but larger number means more shrinkage. default is 0.5
icPercentile: The lower bound of the credibility interval for the IC values (IClow). default is 0.025,
metric: Defines wether the output will contain the point estimate or the lower bound. Available input is 'IC and 'IC025' default is 'IC025'

Value

An object of type ictpdResults containing the results.

Examples

if (FALSE) {
library(SelfControlledCohort)

connectionDetails <- createConnectionDetails(dbms = "postgresql",
                                             user = "joe",
                                             password = "secret",
                                             server = "myserver")
exposureOutcomePairs <- data.frame(outcomeId = c(196794, 196794, 312648),
                                   exposurId = c(1501700, 1545958, 1551803))
ictpdData <- getDbIctpdData(connectionDetails,
                            cdmDatabaseSchema = "cdm_schema.dbo",
                            exposureOutcomePairs = exposureOutcomePairs)
ictpdResults <- calculateStatisticsIC(ictpdData)
ictpdResults
}