OhdsiReportGenerator

Description

A package for extracting analyses results and creating reports.

Author(s)

Maintainer: Jenna Reps jreps@its.jnj.com

Authors:

• Anthony Sena sena@ohdsi.org

See Also

Useful links:

• https://ohdsi.github.io/OhdsiReportGenerator/

• https://github.com/OHDSI/OhdsiReportGenerator

• Report bugs at https://github.com/OHDSI/OhdsiReportGenerator/issues

addTarColumn

Description

Finds the four TAR columns and creates a new column called tar that pastes the columns into a nice string

Usage

addTarColumn(data)

Arguments

Details

Create a friendly single tar column

Value

The data data.frame object with the tar column added if seperate TAR columns are found

See Also

Other helper: formatBinaryCovariateName(), getExampleConnectionDetails(), kableDark(), printReactable(), removeSpaces()

Examples

addTarColumn(data.frame(
tarStartWith = 'cohort start',
tarStartOffset = 1,
tarEndWith = 'cohort start',
tarEndOffset = 0
))

formatBinaryCovariateName

Description

Removes the long part of the covariate name to make it friendly

Usage

formatBinaryCovariateName(data)

Arguments

Details

Makes the covariateName more friendly and shorter

Value

The data data.frame object with the ovariateName column changed to be more friendly

See Also

Other helper: addTarColumn(), getExampleConnectionDetails(), kableDark(), printReactable(), removeSpaces()

Examples

formatBinaryCovariateName(data.frame(
covariateName = c("fdfgfgf: dgdgff","made up test")
))

generateFullReport

Description

Generates a full report from a Strategus analysis

Usage

generateFullReport(
  server,
  username,
  password,
  dbms,
  resultsSchema = NULL,
  targetId = 1,
  outcomeIds = 3,
  comparatorIds = 2,
  indicationIds = "",
  cohortNames = c("target name", "outcome name", "comp name"),
  cohortIds = c(1, 3, 2),
  includeCI = TRUE,
  includeCharacterization = TRUE,
  includeCohortMethod = TRUE,
  includeSccs = TRUE,
  includePrediction = TRUE,
  webAPI = NULL,
  authMethod = NULL,
  webApiUsername = NULL,
  webApiPassword = NULL,
  outputLocation,
  outputName = paste0("full_report_", gsub(":", "_", gsub(" ", "_",
    as.character(date()))), ".html"),
  intermediateDir = tempdir(),
  pathToDriver = Sys.getenv("DATABASECONNECTOR_JAR_FOLDER")
)

Arguments

Details

Specify the connection details to the result database and the schema name to generate the full report.

Value

An html document containing the full results for the target, comparators, indications and outcomes specified.

See Also

Other Reporting: generatePresentation(), generatePresentationMultiple()

generatePresentation

Description

Generates a presentation from a Strategus result

Usage

generatePresentation(
  server,
  username,
  password,
  dbms,
  resultsSchema = NULL,
  dbDetails = NULL,
  lead = "add name",
  team = "name 1 name 2",
  trigger = "A signal was found in spontaneous reports",
  safetyQuestion = "",
  objective = "",
  topline1 =
    "Very brief executive summary. You can copy-paste language from the conclusion.",
  topline2 =
    "If an estimation was requested but not feasible, this should be mentioned here.",
  topline3 =
    "If no estimation study was requested, this high-level summary might be skipped.",
  date = as.character(Sys.Date()),
  targetId = 1,
  outcomeIds = 3,
  cohortNames = c("target name", "outcome name"),
  cohortIds = c(1, 3),
  covariateIds = NULL,
  details = list(studyPeriod = "All Time", restrictions = "Age - None"),
  evaluationText = "",
  includeCI = TRUE,
  includeCharacterization = TRUE,
  includeCM = TRUE,
  includeSCCS = TRUE,
  includePLP = TRUE,
  outputLocation,
  outputName = paste0("presentation_", gsub(":", "_", gsub(" ", "_",
    as.character(date()))), ".html"),
  intermediateDir = tempdir(),
  pathToDriver = Sys.getenv("DATABASECONNECTOR_JAR_FOLDER")
)

Arguments

Details

Specify the connection details to the result database and the schema name to generate a presentation.

Value

An named R list with the elements 'standard' and 'source'

See Also

Other Reporting: generateFullReport(), generatePresentationMultiple()

generatePresentationMultiple

Description

Generates a presentation from a Strategus result

Usage

generatePresentationMultiple(
  server,
  username,
  password,
  dbms,
  resultsSchema = NULL,
  targetId = 1,
  targetName = "target cohort",
  cmSubsetId = 2,
  sccsSubsetId = NULL,
  indicationName = NULL,
  outcomeIds = 3,
  outcomeNames = "outcome cohort",
  comparatorIds = c(2, 4),
  comparatorNames = c("comparator cohort 1", "comparator cohort 2"),
  covariateIds = NULL,
  details = list(studyPeriod = "All Time", restrictions = "Age - None"),
  title = "ASSURE 001 ...",
  lead = "add name",
  date = Sys.Date(),
  backgroundText = "",
  evaluationText = "",
  outputLocation,
  outputName = paste0("presentation_", gsub(":", "_", gsub(" ", "_",
    as.character(date()))), ".html"),
  intermediateDir = tempdir()
)

Arguments

Details

Specify the connection details to the result database and the schema name to generate a presentation.

Value

An named R list with the elements 'standard' and 'source'

See Also

Other Reporting: generateFullReport(), generatePresentation()

A function to extract case series characterization results

Description

A function to extract case series characterization results

Usage

getBinaryCaseSeries(
  connectionHandler,
  schema,
  cTablePrefix = "c_",
  cgTablePrefix = "cg_",
  databaseTable = "database_meta_data",
  targetId = NULL,
  outcomeId = NULL
)

Arguments

Details

Specify the connectionHandler, the schema and the target/outcome cohort IDs

Value

A data.frame with the characterization case series results

See Also

Other Characterization: getBinaryRiskFactors(), getCaseBinaryFeatures(), getCaseContinuousFeatures(), getCaseCounts(), getCharacterizationDemographics(), getContinuousCaseSeries(), getContinuousRiskFactors(), getDechallengeRechallenge(), getIncidenceRates(), getTargetBinaryFeatures(), getTargetContinuousFeatures(), getTargetCounts(), getTimeToEvent(), plotAgeDistributions(), plotSexDistributions()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

cs <- getBinaryCaseSeries(
  connectionHandler = connectionHandler, 
  schema = 'main',
  targetId = 1, 
  outcomeId = 3
)

A function to extract non-case and case binary characterization results

Description

A function to extract non-case and case binary characterization results

Usage

getBinaryRiskFactors(
  connectionHandler,
  schema,
  cTablePrefix = "c_",
  cgTablePrefix = "cg_",
  databaseTable = "database_meta_data",
  targetId = NULL,
  outcomeId = NULL,
  analysisIds = c(3)
)

Arguments

Details

Specify the connectionHandler, the schema and the target/outcome cohort IDs

Value

A data.frame with the characterization results for the cases and non-cases

See Also

Other Characterization: getBinaryCaseSeries(), getCaseBinaryFeatures(), getCaseContinuousFeatures(), getCaseCounts(), getCharacterizationDemographics(), getContinuousCaseSeries(), getContinuousRiskFactors(), getDechallengeRechallenge(), getIncidenceRates(), getTargetBinaryFeatures(), getTargetContinuousFeatures(), getTargetCounts(), getTimeToEvent(), plotAgeDistributions(), plotSexDistributions()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

rf <- getBinaryRiskFactors(
  connectionHandler = connectionHandler, 
  schema = 'main',
  targetId = 1, 
  outcomeId = 3
)

Extract the cohort method results

Description

This function extracts the single database cohort method estimates for results that can be unblinded and have a calibrated RR

Usage

getCMEstimation(
  connectionHandler,
  schema,
  cmTablePrefix = "cm_",
  cgTablePrefix = "cg_",
  databaseTable = "database_meta_data",
  targetIds = NULL,
  outcomeIds = NULL,
  comparatorIds = NULL
)

Arguments

Details

Specify the connectionHandler, the schema and the target/comparator/outcome cohort IDs

Value

Returns a data.frame with the columns:

• databaseName the name of the database

• analysisId the analysis design unique identifier

• description the analysis design description

• targetName the target cohort name

• targetId the target cohort unique identifier

• comparatorName the comparator cohort name

• comparatorId the comparator cohort unique identifier

• outcomeName the outcome name

• outcomeId the outcome unique identifier

• calibratedRr the calibrated relative risk

• calibratedRrCi95Lb the calibrated relative risk 95 percent confidence interval lower bound

• calibratedRrCi95Ub the calibrated relative risk 95 percent confidence interval upper bound

• calibratedP the two sided calibrated p value

• calibratedOneSidedP the one sided calibrated p value

• calibratedLogRr the calibrated relative risk logged

• calibratedSeLogRr the standard error of the calibrated relative risk logged

• targetSubjects the number of people in the target cohort

• comparatorSubjects the number of people in the comparator cohort

• targetDays the total number of days at risk across the target cohort people

• comparatorDays the total number of days at risk across the comparator cohort people

• targetOutcomes the total number of outcomes occuring during the time at risk for the target cohort people

• comparatorOutcomes the total number of outcomes occuring during the time at risk for the comparator cohort people

• targetEstimator ...

See Also

Other Estimation: getCmDiagnosticsData(), getCmMetaEstimation(), getSccsDiagnosticsData(), getSccsEstimation(), getSccsMetaEstimation(), plotCmEstimates(), plotSccsEstimates()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

cmEst <- getCMEstimation(
  connectionHandler = connectionHandler, 
  schema = 'main',
  targetIds = 1,
  outcomeIds = 3
)

Extract aggregate statistics of binary feature analysis IDs of interest for cases

Description

This function extracts the feature extraction results for cases corresponding to specified target and outcome cohorts.

Usage

getCaseBinaryFeatures(
  connectionHandler,
  schema,
  cTablePrefix = "c_",
  cgTablePrefix = "cg_",
  databaseTable = "database_meta_data",
  targetIds = NULL,
  outcomeIds = NULL,
  analysisIds = c(3)
)

Arguments

Details

Specify the connectionHandler, the schema and the target/outcome cohort IDs

Value

Returns a data.frame with the columns:

• databaseName the name of the database

• targetName the target cohort name

• targetId the target cohort unique identifier

• outcomeName the outcome name

• outcomeId the outcome unique identifier

• minPriorObservation the minimum required observation days prior to index for an entry

• outcomeWashoutDays patients with the outcome occurring within this number of days prior to index are excluded (NA means no exclusion)

• riskWindowStart the number of days ofset the start anchor that is the start of the time-at-risk

• startAnchor the start anchor is either the target cohort start or cohort end date

• riskWindowEnd the number of days ofset the end anchor that is the end of the time-at-risk

• endAnchor the end anchor is either the target cohort start or cohort end date

• covariateId the id of the feature

• covariateName the name of the feature

• sumValue the number of cases who have the feature value of 1

• averageValue the mean feature value

See Also

Other Characterization: getBinaryCaseSeries(), getBinaryRiskFactors(), getCaseContinuousFeatures(), getCaseCounts(), getCharacterizationDemographics(), getContinuousCaseSeries(), getContinuousRiskFactors(), getDechallengeRechallenge(), getIncidenceRates(), getTargetBinaryFeatures(), getTargetContinuousFeatures(), getTargetCounts(), getTimeToEvent(), plotAgeDistributions(), plotSexDistributions()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

cbf <- getCaseBinaryFeatures(
connectionHandler = connectionHandler, 
schema = 'main'
)

Extract aggregate statistics of continuous feature analysis IDs of interest for targets

Description

This function extracts the continuous feature extraction results for cases corresponding to specified target and outcome cohorts.

Usage

getCaseContinuousFeatures(
  connectionHandler,
  schema,
  cTablePrefix = "c_",
  cgTablePrefix = "cg_",
  databaseTable = "database_meta_data",
  targetIds = NULL,
  outcomeIds = NULL,
  analysisIds = NULL
)

Arguments

Details

Specify the connectionHandler, the schema and the target/outcome cohort IDs

Value

Returns a data.frame with the columns:

• databaseName the name of the database

• targetName the target cohort name

• targetId the target cohort unique identifier

• outcomeName the outcome name

• outcomeId the outcome unique identifier

• minPriorObservation the minimum required observation days prior to index for an entry

• outcomeWashoutDays patients with the outcome occurring within this number of days prior to index are excluded (NA means no exclusion)

• riskWindowStart the time at risk start point

• riskWindowEnd the time at risk end point

• startAnchor the time at risk start point offset

• endAnchor the time at risk end point offset

• covariateName the name of the feature

• covariateId the id of the feature

• countValue the number of cases who have the feature

• minValue the minimum value observed for the feature

• maxValue the maximum value observed for the feature

• averageValue the mean value observed for the feature

• standardDeviation the standard deviation of the value observed for the feature

• medianValue the median value observed for the feature

• p10Value the 10th percentile of the value observed for the feature

• p25Value the 25th percentile of the value observed for the feature

• p75Value the 75th percentile of the value observed for the feature

• p90Value the 90th percentile of the value observed for the feature

See Also

Other Characterization: getBinaryCaseSeries(), getBinaryRiskFactors(), getCaseBinaryFeatures(), getCaseCounts(), getCharacterizationDemographics(), getContinuousCaseSeries(), getContinuousRiskFactors(), getDechallengeRechallenge(), getIncidenceRates(), getTargetBinaryFeatures(), getTargetContinuousFeatures(), getTargetCounts(), getTimeToEvent(), plotAgeDistributions(), plotSexDistributions()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

ccf <- getCaseContinuousFeatures(
connectionHandler = connectionHandler, 
schema = 'main'
)

Extract the outcome cohort counts result

Description

This function extracts outcome cohort counts across databases in the results for specified target and outcome cohorts.

Usage

getCaseCounts(
  connectionHandler,
  schema,
  cTablePrefix = "c_",
  cgTablePrefix = "cg_",
  databaseTable = "database_meta_data",
  targetIds = NULL,
  outcomeIds = NULL
)

Arguments

Details

Specify the connectionHandler, the schema and the target/outcome cohort IDs

Value

Returns a data.frame with the columns:

• databaseName the name of the database

• targetName the target cohort name

• targetId the target cohort unique identifier

• outcomeName the outcome name

• outcomeId the outcome unique identifier

• rowCount the number of entries in the cohort

• personCount the number of people in the cohort

• minPriorObservation the minimum required observation days prior to index for an entry

• outcomeWashoutDays patients with the outcome occurring within this number of days prior to index are excluded (NA means no exclusion)

• riskWindowStart the number of days ofset the start anchor that is the start of the time-at-risk

• startAnchor the start anchor is either the target cohort start or cohort end date

• riskWindowEnd the number of days ofset the end anchor that is the end of the time-at-risk

• endAnchor the end anchor is either the target cohort start or cohort end date

See Also

Other Characterization: getBinaryCaseSeries(), getBinaryRiskFactors(), getCaseBinaryFeatures(), getCaseContinuousFeatures(), getCharacterizationDemographics(), getContinuousCaseSeries(), getContinuousRiskFactors(), getDechallengeRechallenge(), getIncidenceRates(), getTargetBinaryFeatures(), getTargetContinuousFeatures(), getTargetCounts(), getTimeToEvent(), plotAgeDistributions(), plotSexDistributions()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

cc <- getCaseCounts(
connectionHandler = connectionHandler, 
schema = 'main'
)

Extract the binary age groups for the cases and targets

Description

This function extracts the age group feature extraction results for cases and targets corresponding to specified target and outcome cohorts.

Usage

getCharacterizationDemographics(
  connectionHandler,
  schema,
  cTablePrefix = "c_",
  cgTablePrefix = "cg_",
  databaseTable = "database_meta_data",
  targetId = NULL,
  outcomeId = NULL,
  type = "age"
)

Arguments

Details

Specify the connectionHandler, the schema and the target/outcome cohort IDs

Value

Returns a data.frame with the columns:

• databaseName the name of the database

• targetName the target cohort name

• targetId the target cohort unique identifier

• outcomeName the outcome name

• outcomeId the outcome unique identifier

• minPriorObservation the minimum required observation days prior to index for an entry

• outcomeWashoutDays patients with the outcome occurring within this number of days prior to index are excluded (NA means no exclusion)

• riskWindowStart the number of days ofset the start anchor that is the start of the time-at-risk

• startAnchor the start anchor is either the target cohort start or cohort end date

• riskWindowEnd the number of days ofset the end anchor that is the end of the time-at-risk

• endAnchor the end anchor is either the target cohort start or cohort end date

• covariateName the name of the feature

• sumValue the number of cases who have the feature value of 1

• averageValue the mean feature value

See Also

Other Characterization: getBinaryCaseSeries(), getBinaryRiskFactors(), getCaseBinaryFeatures(), getCaseContinuousFeatures(), getCaseCounts(), getContinuousCaseSeries(), getContinuousRiskFactors(), getDechallengeRechallenge(), getIncidenceRates(), getTargetBinaryFeatures(), getTargetContinuousFeatures(), getTargetCounts(), getTimeToEvent(), plotAgeDistributions(), plotSexDistributions()

Examples

# example code

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

ageData <- getCharacterizationDemographics(
connectionHandler = connectionHandler, 
schema = 'main'
)

Extract the cohort method diagostic results

Description

This function extracts the cohort method diagnostics that examine whether the analyses were sufficiently powered and checks for different types of bias.

Usage

getCmDiagnosticsData(
  connectionHandler,
  schema,
  cmTablePrefix = "cm_",
  cgTablePrefix = "cg_",
  databaseTable = "database_meta_data",
  targetIds = NULL,
  outcomeIds = NULL,
  comparatorIds = NULL
)

Arguments

Details

Specify the connectionHandler, the schema and the target/comparator/outcome cohort IDs

Value

Returns a data.frame with the columns:

• databaseName the name of the database

• analysisId the analysis unique identifier

• description a description of the analysis

• targetName the target cohort name

• targetId the target cohort unique identifier

• comparatorName the comparator cohort name

• comparatorId the comparator cohort unique identifier

• outcomeName the outcome name

• outcomeId the outcome cohort unique identifier

• maxSdm max allowed standardized difference of means when comparing the target to the comparator after PS adjustment for the ballance diagnostic diagnostic to pass.

• sharedMaxSdm max allowed standardized difference of means when comparing the target to the comparator after PS adjustment for the ballance diagnostic diagnostic to pass.

• equipoise the bounds on the preference score to determine whether a subject is in equipoise.

• mdrr the maximum passable minimum detectable relative risk (mdrr) value. If the mdrr is greater than this the diagnostics will fail.

• attritionFraction (depreciated) the minmum attrition before the diagnostics fails.

• ease The expected absolute systematic error (ease) measures residual bias.

• balanceDiagnostic whether the balance diagnostic passed or failed.

• sharedBalanceDiagnostic whether the shared balance diagnostic passed or failed.

• equipoiseDiagnostic whether the equipose diagnostic passed or failed.

• mdrrDiagnostic whether the mdrr (power) diagnostic passed or failed.

• attritionDiagnostic (depreciated) whether the attrition diagnostic passed or failed.

• easeDiagnostic whether the ease diagnostic passed or failed.

• unblindForEvidenceSynthesis whether the results can be unblinded for the meta analysis.

• unblind whether the results can be unblinded.

• summaryValue summary of diagnostics results. FAIL, PASS or number of warnings.

See Also

Other Estimation: getCMEstimation(), getCmMetaEstimation(), getSccsDiagnosticsData(), getSccsEstimation(), getSccsMetaEstimation(), plotCmEstimates(), plotSccsEstimates()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

cmDiag <- getCmDiagnosticsData(
  connectionHandler = connectionHandler, 
  schema = 'main',
  targetIds = 1,
  outcomeIds = 3
)

Extract the cohort method meta analysis results

Description

This function extracts any meta analysis estimation results for cohort method.

Usage

getCmMetaEstimation(
  connectionHandler,
  schema,
  cmTablePrefix = "cm_",
  cgTablePrefix = "cg_",
  esTablePrefix = "es_",
  targetIds = NULL,
  outcomeIds = NULL,
  comparatorIds = NULL
)

Arguments

Details

Specify the connectionHandler, the schema and the target/comparator/outcome cohort IDs

Value

Returns a data.frame with the columns:

• databaseName the name of the database

• analysisId the analysis unique identifier

• description a description of the analysis

• targetName the target cohort name

• targetId the target cohort unique identifier

• comparatorName the comparator cohort name

• comparatorId the comparator cohort unique identifier

• outcomeName the outcome name

• outcomeId the outcome cohort unique identifier

• calibratedRr the calibrated relative risk

• calibratedRrCi95Lb the calibrated relative risk 95 percent confidence interval lower bound

• calibratedRrCi95Ub the calibrated relative risk 95 percent confidence interval upper bound

• calibratedP the two sided calibrated p value

• calibratedOneSidedP the one sided calibrated p value

• calibratedLogRr the calibrated relative risk logged

• calibratedSeLogRr the standard error of the calibrated relative risk logged

• targetSubjects the number of people in the target cohort across included database

• comparatorSubjects the number of people in the comparator cohort across included database

• targetDays the total number of days at risk across the target cohort people across included database

• comparatorDays the total number of days at risk across the comparator cohort people across included database

• targetOutcomes the total number of outcomes occuring during the time at risk for the target cohort people across included database

• comparatorOutcomes the total number of outcomes occuring during the time at risk for the comparator cohort people across included database

• nDatabases the number of databases included

See Also

Other Estimation: getCMEstimation(), getCmDiagnosticsData(), getSccsDiagnosticsData(), getSccsEstimation(), getSccsMetaEstimation(), plotCmEstimates(), plotSccsEstimates()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

cmMeta <- getCmMetaEstimation(
  connectionHandler = connectionHandler, 
  schema = 'main',
  targetIds = 1,
  outcomeIds = 3
)

Extract the cohort definition details

Description

This function extracts all cohort definitions for the targets of interest.

Usage

getCohortDefinitions(
  connectionHandler,
  schema,
  cgTablePrefix = "cg_",
  targetIds = NULL
)

Arguments

Details

Specify the connectionHandler, the schema and the target cohort IDs

Value

Returns a data.frame with the cohort details

See Also

Other Cohorts: getCohortSubsetDefinitions(), processCohorts()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

cohortDef <- getCohortDefinitions(
  connectionHandler = connectionHandler, 
  schema = 'main'
)

Extract the cohort subset definition details

Description

This function extracts all cohort subset definitions for the subsets of interest.

Usage

getCohortSubsetDefinitions(
  connectionHandler,
  schema,
  cgTablePrefix = "cg_",
  subsetIds = NULL
)

Arguments

Details

Specify the connectionHandler, the schema and the subset IDs

Value

Returns a data.frame with the cohort subset details

See Also

Other Cohorts: getCohortDefinitions(), processCohorts()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

subsetDef <- getCohortSubsetDefinitions(
  connectionHandler = connectionHandler, 
  schema = 'main'
)

A function to extract case series continuous feature characterization results

Description

A function to extract case series continuous feature characterization results

Usage

getContinuousCaseSeries(
  connectionHandler,
  schema,
  cTablePrefix = "c_",
  cgTablePrefix = "cg_",
  databaseTable = "database_meta_data",
  targetId = NULL,
  outcomeId = NULL
)

Arguments

Details

Specify the connectionHandler, the schema and the target/outcome cohort IDs

Value

A data.frame with the characterization case series results

See Also

Other Characterization: getBinaryCaseSeries(), getBinaryRiskFactors(), getCaseBinaryFeatures(), getCaseContinuousFeatures(), getCaseCounts(), getCharacterizationDemographics(), getContinuousRiskFactors(), getDechallengeRechallenge(), getIncidenceRates(), getTargetBinaryFeatures(), getTargetContinuousFeatures(), getTargetCounts(), getTimeToEvent(), plotAgeDistributions(), plotSexDistributions()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

cs <- getContinuousCaseSeries(
  connectionHandler = connectionHandler, 
  schema = 'main',
  targetId = 1, 
  outcomeId = 3
)

A function to extract non-case and case continuous characterization results

Description

A function to extract non-case and case continuous characterization results

Usage

getContinuousRiskFactors(
  connectionHandler,
  schema,
  cTablePrefix = "c_",
  cgTablePrefix = "cg_",
  databaseTable = "database_meta_data",
  targetId = NULL,
  outcomeId = NULL,
  analysisIds = NULL
)

Arguments

Details

Specify the connectionHandler, the schema and the target/outcome cohort IDs

Value

A data.frame with the characterization results for the cases and non-cases

See Also

Other Characterization: getBinaryCaseSeries(), getBinaryRiskFactors(), getCaseBinaryFeatures(), getCaseContinuousFeatures(), getCaseCounts(), getCharacterizationDemographics(), getContinuousCaseSeries(), getDechallengeRechallenge(), getIncidenceRates(), getTargetBinaryFeatures(), getTargetContinuousFeatures(), getTargetCounts(), getTimeToEvent(), plotAgeDistributions(), plotSexDistributions()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

rf <- getContinuousRiskFactors(
  connectionHandler = connectionHandler, 
  schema = 'main',
  targetId = 1, 
  outcomeId = 3
)

Extract the dechallenge rechallenge results

Description

This function extracts all dechallenge rechallenge results across databases for specified target and outcome cohorts.

Usage

getDechallengeRechallenge(
  connectionHandler,
  schema,
  cTablePrefix = "c_",
  cgTablePrefix = "cg_",
  databaseTable = "database_meta_data",
  targetIds = NULL,
  outcomeIds = NULL
)

Arguments

Details

Specify the connectionHandler, the schema and the target/outcome cohort IDs

Value

Returns a data.frame with the columns:

• databaseName the name of the database

• targetName the target cohort name

• targetId the target cohort unique identifier

• outcomeName the outcome name

• outcomeId the outcome unique identifier

• dechallengeStopInterval An integer specifying the how much time to add to the cohort_end when determining whether the event starts during cohort and ends after

• dechallengeEvaluationWindow A period of time evaluated for outcome recurrence after discontinuation of exposure, among patients with challenge outcomes

• numExposureEras Distinct number of exposure events (i.e. drug eras) in a given target cohort

• numPersonsExposed Distinct number of people exposed in target cohort. A person must have at least 1 day exposure to be included

• numCases Distinct number of persons in outcome cohort. A person must have at least 1 day of observation time to be included

• dechallengeAttempt Distinct count of people with observable time after discontinuation of the exposure era during which the challenge outcome occurred

• dechallengeFail Among people with challenge outcomes, the distinct number of people with outcomes during dechallengeEvaluationWindow

• dechallengeSuccess Among people with challenge outcomes, the distinct number of people without outcomes during the dechallengeEvaluationWindow

• rechallengeAttempt Number of people with a new exposure era after the occurrence of an outcome during a prior exposure era

• rechallengeFail Number of people with a new exposure era during which an outcome occurred, after the occurrence of an outcome during a prior exposure era

• rechallengeSuccess Number of people with a new exposure era during which an outcome did not occur, after the occurrence of an outcome during a prior exposure era

• pctDechallengeAttempt Percent of people with observable time after discontinuation of the exposure era during which the challenge outcome occurred

• pctDechallengeFail Among people with challenge outcomes, the percent of people without outcomes during the dechallengeEvaluationWindow

• pctDechallengeSuccess Among people with challenge outcomes, the percent of people with outcomes during dechallengeEvaluationWindow

• pctRechallengeAttempt Percent of people with a new exposure era after the occurrence of an outcome during a prior exposure era

• pctRechallengeFail Percent of people with a new exposure era during which an outcome did not occur, after the occurrence of an outcome during a prior exposure era

• pctRechallengeSuccess Percent of people with a new exposure era during which an outcome occurred, after the occurrence of an outcome during a prior exposure era

See Also

Other Characterization: getBinaryCaseSeries(), getBinaryRiskFactors(), getCaseBinaryFeatures(), getCaseContinuousFeatures(), getCaseCounts(), getCharacterizationDemographics(), getContinuousCaseSeries(), getContinuousRiskFactors(), getIncidenceRates(), getTargetBinaryFeatures(), getTargetContinuousFeatures(), getTargetCounts(), getTimeToEvent(), plotAgeDistributions(), plotSexDistributions()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

dcrc <- getDechallengeRechallenge(
connectionHandler = connectionHandler, 
schema = 'main'
)

create a connection detail for an example OHDSI results database

Description

This returns an object of class 'ConnectionDetails' that lets you connect via 'DatabaseConnector::connect()' to the example result database.

Usage

getExampleConnectionDetails(exdir = tempdir())

Arguments

Details

Finds the location of the example result database in the package and calls 'DatabaseConnector::createConnectionDetails' to create a 'ConnectionDetails' object for connecting to the database.

Value

An object of class 'ConnectionDetails' with the details to connect to the example OHDSI result database

See Also

Other helper: addTarColumn(), formatBinaryCovariateName(), kableDark(), printReactable(), removeSpaces()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

Extract the cohort incidence result

Description

This function extracts all incidence rates across databases in the results for specified target and outcome cohorts.

Usage

getIncidenceRates(
  connectionHandler,
  schema,
  ciTablePrefix = "ci_",
  cgTablePrefix = "cg_",
  databaseTable = "database_meta_data",
  targetIds = NULL,
  outcomeIds = NULL
)

Arguments

Details

Specify the connectionHandler, the schema and the target/outcome cohort IDs

Value

Returns a data.frame with the columns:

• databaseName the name of the database

• targetName the target cohort name

• targetId the target cohort unique identifier

• outcomeName the outcome name

• outcomeId the outcome unique identifier

• cleanWindow clean windown around outcome

• subgroupName name for the result subgroup

• ageGroupName name for the result age group

• genderName name for the result gender group

• startYear name for the result start year

• tarStartWith time at risk start reference

• tarStartOffset time at risk start offset from reference

• tarEndWith time at risk end reference

• tarEndOffset time at risk end offset from reference

• personsAtRiskPe persons at risk per event

• personsAtRisk persons at risk

• personDaysPe person days per event

• personDays person days

• personOutcomesPe person outcome per event

• personOutcomes persons outcome

• outcomesPe number of outcome per event

• outcomes number of outcome

• incidenceProportionP100p incidence proportion per 100 persons

• incidenceRateP100py incidence rate per 100 person years

See Also

Other Characterization: getBinaryCaseSeries(), getBinaryRiskFactors(), getCaseBinaryFeatures(), getCaseContinuousFeatures(), getCaseCounts(), getCharacterizationDemographics(), getContinuousCaseSeries(), getContinuousRiskFactors(), getDechallengeRechallenge(), getTargetBinaryFeatures(), getTargetContinuousFeatures(), getTargetCounts(), getTimeToEvent(), plotAgeDistributions(), plotSexDistributions()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

ir <- getIncidenceRates(
connectionHandler = connectionHandler, 
schema = 'main'
)

Extract a complete set of cohorts used in the prediction results

Description

This function extracts the target and outcome cohorts used to develop any model in the results

Usage

getPredictionCohorts(
  connectionHandler,
  schema,
  plpTablePrefix = "plp_",
  cgTablePrefix = "cg_"
)

Arguments

Details

Specify the connectionHandler, the resultDatabaseSettings and any targetIds or outcomeIds to restrict models to

Value

Returns a data.frame with the columns:

• cohortId the cohort definition ID

• cohortName the name of the cohort

• type whether the cohort was used as a target or outcome cohort

See Also

Other Prediction: getPredictionDiagnosticTable(), getPredictionDiagnostics(), getPredictionHyperParamSearch(), getPredictionIntercept(), getPredictionModelDesigns(), getPredictionPerformanceTable(), getPredictionPerformances(), getPredictionTopPredictors()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

predCohorts <- getPredictionCohorts(
  connectionHandler = connectionHandler, 
  schema = 'main'
)

Extract specific diagnostic table

Description

This function extracts the specified diagnostic table

Usage

getPredictionDiagnosticTable(
  connectionHandler,
  schema,
  plpTablePrefix = "plp_",
  table = "diagnostic_participants",
  diagnosticId = NULL
)

Arguments

Details

Specify the connectionHandler, the resultDatabaseSettings, the table of interest and (optionally) a diagnosticId to filter to

Value

Returns a data.frame with the specified table

See Also

Other Prediction: getPredictionCohorts(), getPredictionDiagnostics(), getPredictionHyperParamSearch(), getPredictionIntercept(), getPredictionModelDesigns(), getPredictionPerformanceTable(), getPredictionPerformances(), getPredictionTopPredictors()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

diagPred <- getPredictionDiagnosticTable(
  connectionHandler = connectionHandler, 
  schema = 'main',
  table = 'diagnostic_predictors'
)

Extract the model design diagnostics for a specific development database

Description

This function extracts the PROBAST diagnostics

Usage

getPredictionDiagnostics(
  connectionHandler,
  schema,
  plpTablePrefix = "plp_",
  cgTablePrefix = "cg_",
  databaseTable = "database_meta_data",
  databaseTablePrefix = "",
  modelDesignId = NULL,
  threshold1_2 = 0.9
)

Arguments

Details

Specify the connectionHandler, the resultDatabaseSettings and (optionally) a modelDesignId and threshold1_2 a threshold value to use for the PROBAST 1.2

Value

Returns a data.frame with the columns:

• modelDesignId the unique identifier for the model design

• diagnosticId the unique identifier for diagnostic result

• developmentDatabaseName the name for the database used to develop the model

• developmentTargetName the name for the development target population

• developmentOutcomeName the name for the development outcome

• probast1_1 Were appropriate data sources used, e.g., cohort, RCT, or nested case-control study data?

• probast1_2 Were all inclusions and exclusions of paticipants appropriate?

• probast2_1 Were predictors defined and assessed in a similar way for all participants?

• probast2_2 Were predictors assessments made without knowledge of outcome data?

• probast2_3 All all predictors available at the time the model is intended to be used?

• probast3_4 Was the outcome defined and determined in a similar way for all participants?

• probast3_6 Was the time interval between predictor assessment and outcome determination appropriate?

• probast4_1 Were there a reasonable number of participants with the outcome?

See Also

Other Prediction: getPredictionCohorts(), getPredictionDiagnosticTable(), getPredictionHyperParamSearch(), getPredictionIntercept(), getPredictionModelDesigns(), getPredictionPerformanceTable(), getPredictionPerformances(), getPredictionTopPredictors()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

diag <- getPredictionDiagnostics(
  connectionHandler = connectionHandler, 
  schema = 'main'
)

Extract hyper parameters details

Description

This function extracts the hyper parameters details

Usage

getPredictionHyperParamSearch(
  connectionHandler,
  schema,
  plpTablePrefix = "plp_",
  modelDesignId = NULL,
  databaseId = NULL
)

Arguments

Details

Specify the connectionHandler, the resultDatabaseSettings, the modelDesignId and the databaseId

Value

Returns a data.frame with the columns:

• metric the hyperparameter optimization metric

• fold the fold in cross validation

• value the metric value for the fold with the specified hyperparameter combination

plus columns for all the hyperparameters and their values

See Also

Other Prediction: getPredictionCohorts(), getPredictionDiagnosticTable(), getPredictionDiagnostics(), getPredictionIntercept(), getPredictionModelDesigns(), getPredictionPerformanceTable(), getPredictionPerformances(), getPredictionTopPredictors()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

hyperParams <- getPredictionHyperParamSearch(
  connectionHandler = connectionHandler, 
  schema = 'main'
)

Extract model interception (for logistic regression)

Description

This function extracts the interception value

Usage

getPredictionIntercept(
  connectionHandler,
  schema,
  plpTablePrefix = "plp_",
  modelDesignId = NULL,
  databaseId = NULL
)

Arguments

Details

Specify the connectionHandler, the resultDatabaseSettings, the modelDesignId and the databaseId

Value

Returns a single value corresponding to the model intercept or NULL if not a logistic regression model

See Also

Other Prediction: getPredictionCohorts(), getPredictionDiagnosticTable(), getPredictionDiagnostics(), getPredictionHyperParamSearch(), getPredictionModelDesigns(), getPredictionPerformanceTable(), getPredictionPerformances(), getPredictionTopPredictors()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

intercepts <- getPredictionIntercept(
  connectionHandler = connectionHandler, 
  schema = 'main'
)

Extract the model designs and aggregate performances for the prediction results

Description

This function extracts the model design settings and min/max/mean AUROC values of the models developed using the model design across databases

Usage

getPredictionModelDesigns(
  connectionHandler,
  schema,
  plpTablePrefix = "plp_",
  cgTablePrefix = "cg_",
  targetIds = NULL,
  outcomeIds = NULL
)

Arguments

Details

Specify the connectionHandler, the resultDatabaseSettings and (optionally) any targetIds or outcomeIds to restrict model designs to

Value

Returns a data.frame with the columns:

• modelDesignId a unique identifier in the database for the model design

• modelType the type of classifier or surival model

• developmentTargetId a unique identifier for the development target ID

• developmentTargetName the name of the development target cohort

• developmentTargetJson the json of the target cohort

• developmentOutcomeId a unique identifier for the development outcome ID

• developmentOutcomeName the name of the development outcome cohort

• timeAtRisk the time at risk string

• developmentOutcomeJson the json of the outcome cohort

• covariateSettingsJson the covariate settings json

• populationSettingsJson the population settings json

• tidyCovariatesSettingsJson the tidy covariate settings json

• plpDataSettingsJson the plp data extraction settings json

• featureEngineeringSettingsJson the feature engineering settings json

• splitSettingsJson the split settings json

• sampleSettingsJson the sample settings json

• minAuroc the min AUROC value of models developed using the model design across databases

• meanAuroc the mean AUROC value of models developed using the model design across databases

• maxAuroc the max AUROC value of models developed using the model design across databases

• noDiagnosticDatabases the number of databases where the model design diagnostics were generated

• noDevelopmentDatabases the number of databases where the model design was used to develop models

• noValidationDatabases the number of databases where the models developed using the model design was externally validated

See Also

Other Prediction: getPredictionCohorts(), getPredictionDiagnosticTable(), getPredictionDiagnostics(), getPredictionHyperParamSearch(), getPredictionIntercept(), getPredictionPerformanceTable(), getPredictionPerformances(), getPredictionTopPredictors()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

modDesign <- getPredictionModelDesigns(
  connectionHandler = connectionHandler, 
  schema = 'main'
)

Extract specific results table

Description

This function extracts the specified table

Usage

getPredictionPerformanceTable(
  connectionHandler,
  schema,
  plpTablePrefix = "plp_",
  table = "attrition",
  performanceId = NULL
)

Arguments

Details

Specify the connectionHandler, the resultDatabaseSettings, the table of interest and (optionally) a performanceId to filter to

Value

Returns a data.frame with the specified table

See Also

Other Prediction: getPredictionCohorts(), getPredictionDiagnosticTable(), getPredictionDiagnostics(), getPredictionHyperParamSearch(), getPredictionIntercept(), getPredictionModelDesigns(), getPredictionPerformances(), getPredictionTopPredictors()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

attrition <- getPredictionPerformanceTable(
  connectionHandler = connectionHandler, 
  schema = 'main',
  table = 'attrition'
)

Extract the model performances

Description

This function extracts the model performances

Usage

getPredictionPerformances(
  connectionHandler,
  schema,
  plpTablePrefix = "plp_",
  cgTablePrefix = "cg_",
  databaseTable = "database_meta_data",
  databaseTablePrefix = "",
  modelDesignId = NULL,
  developmentDatabaseId = NULL
)

Arguments

Details

Specify the connectionHandler, the resultDatabaseSettings and (optionally) a modelDesignId and/or developmentDatabaseId to restrict models to

Value

Returns a data.frame with the columns:

• performanceId the unique identifier for the performance

• modelDesignId the unique identifier for the model design

• developmentDatabaseId the unique identifier for the database used to develop the model

• validationDatabaseId the unique identifier for the database used to validate the model

• developmentTargetId the unique cohort id for the development target population

• developmentTargetName the name for the development target population

• developmentOutcomeId the unique cohort id for the development outcome

• developmentOutcomeName the name for the development outcome

• developmentDatabase the name for the database used to develop the model

• validationDatabase the name for the database used to validate the model

• validationTargetName the name for the validation target population

• validationOutcomeName the name for the validation outcome

• timeStamp the date/time when the analysis occurred

• auroc the test/validation AUROC value for the model

• auroc95lb the test/validation lower bound of the 95 percent CI AUROC value for the model

• auroc95ub the test/validation upper bound of the 95 percent CI AUROC value for the model

• calibrationInLarge the test/validation calibration in the large value for the model

• eStatistic the test/validation calibration e-statistic value for the model

• brierScore the test/validation brier value for the model

• auprc the test/validation discrimination AUPRC value for the model

• populationSize the test/validation population size used to develop the model

• outcomeCount the test/validation outcome count used to develop the model

• evalPercent the percentage of the development data used as the test set

• outcomePercent the outcome percent in the development data

• validationTimeAtRisk time at risk for the validation

• predictionResultType development or validation

See Also

Other Prediction: getPredictionCohorts(), getPredictionDiagnosticTable(), getPredictionDiagnostics(), getPredictionHyperParamSearch(), getPredictionIntercept(), getPredictionModelDesigns(), getPredictionPerformanceTable(), getPredictionTopPredictors()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

perf <- getPredictionPerformances(
  connectionHandler = connectionHandler, 
  schema = 'main'
)

Extract the top N predictors per model

Description

This function extracts the top N predictors per model from the prediction results tables

Usage

getPredictionTopPredictors(
  connectionHandler,
  schema,
  plpTablePrefix = "plp_",
  cgTablePrefix = "cg_",
  targetIds = NULL,
  outcomeIds = NULL,
  numberPredictors = 100
)

Arguments

Details

Specify the connectionHandler, the resultDatabaseSettings and (optionally) any targetIds or outcomeIds to restrict models to

Value

Returns a data.frame with the columns:

• databaseName the name of the database the model was developed on

• tarStartDay the time-at-risk start day

• tarStartAnchor whether the time-at-risk start is relative to cohort start or end

• tarEndDay the time-at-risk end day

• tarEndAnchor whether the time-at-risk end is relative to cohort start or end

• performanceId a unique identifier for the performance

• covariateId the FeatureExtraction covariate identifier

• covariateName the name of the covariate

• conceptId the covariates corresponding concept or 0

• covariateValue the feature importance or coefficient value

• covariateCount how many people had the covariate

• covariateMean the fraction of the target population with the covariate

• covariateStDev the standard deviation

• withNoOutcomeCovariateCount the number of the target population without the outcome with the covariate

• withNoOutcomeCovariateMean the fraction of the target population without the outcome with the covariate

• withNoOutcomeCovariateStDev the covariate standard deviation of the target population without the outcome

• withOutcomeCovariateCount the number of the target population with the outcome with the covariate

• withOutcomeCovariateMean the fraction of the target population with the outcome with the covariate

• withOutcomeCovariateStDev the covariate standard deviation of the target population with the outcome

• standardizedMeanDiff the standardized mean difference comparing the target population with outcome and without the outcome

• rn the row number showing the covariate rank

See Also

Other Prediction: getPredictionCohorts(), getPredictionDiagnosticTable(), getPredictionDiagnostics(), getPredictionHyperParamSearch(), getPredictionIntercept(), getPredictionModelDesigns(), getPredictionPerformanceTable(), getPredictionPerformances()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

topPreds <- getPredictionTopPredictors(
  connectionHandler = connectionHandler, 
  schema = 'main',
  targetIds = 1,
  outcomeIds = 3
)

Extract the self controlled case series (sccs) diagostic results

Description

This function extracts the sccs diagnostics that examine whether the analyses were sufficiently powered and checks for different types of bias.

Usage

getSccsDiagnosticsData(
  connectionHandler,
  schema,
  sccsTablePrefix = "sccs_",
  cgTablePrefix = "cg_",
  databaseTable = "database_meta_data",
  targetIds = NULL,
  outcomeIds = NULL
)

Arguments

Details

Specify the connectionHandler, the schema and the target/outcome cohort IDs

Value

Returns a data.frame with the columns:

• databaseName the database name

• analysisId the analysis unique identifier

• description an analysis description

• targetName the target name

• targetId the target cohort id

• outcomeName the outcome name

• outcomeId the outcome cohort id

• indicationName the indication name

• indicatonId the indication cohort id

• covariateName whether main or secondary analysis

• mdrr the maximum passable minimum detectable relative risk (mdrr) value. If the mdrr is greater than this the diagnostics will fail.

• ease The expected absolute systematic error (ease) measures residual bias.

• timeTrendP The p for whether the mean monthly ratio between observed and expected is no greater than 1.25.

• preExposureP One-sided p-value for whether the rate before expore is higher than after, against the null of no difference.

• mdrrDiagnostic whether the mdrr (power) diagnostic passed or failed.

• easeDiagnostic whether the ease diagnostic passed or failed.

• timeTrendDiagnostic Pass / warning / fail / not evaluated classification of the time trend (unstalbe months) diagnostic.

• preExposureDiagnostic Pass / warning / fail / not evaluated classification of the time trend (unstalbe months) diagnostic.

• unblind whether the results can be unblinded.

• unblindForEvidenceSynthesis whether the results can be unblinded for the meta analysis.

• summaryValue summary of diagnostics results. FAIL, PASS or number of warnings.

See Also

Other Estimation: getCMEstimation(), getCmDiagnosticsData(), getCmMetaEstimation(), getSccsEstimation(), getSccsMetaEstimation(), plotCmEstimates(), plotSccsEstimates()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

sccsDiag <- getSccsDiagnosticsData(
  connectionHandler = connectionHandler, 
  schema = 'main',
  targetIds = 1,
  outcomeIds = 3
)

Extract the self controlled case series (sccs) results

Description

This function extracts the single database sccs estimates

Usage

getSccsEstimation(
  connectionHandler,
  schema,
  sccsTablePrefix = "sccs_",
  cgTablePrefix = "cg_",
  databaseTable = "database_meta_data",
  targetIds = NULL,
  outcomeIds = NULL
)

Arguments

Details

Specify the connectionHandler, the schema and the target/outcome cohort IDs

Value

Returns a data.frame with the columns:

• databaseName the database name

• analysisId the analysis unique identifier

• description an analysis description

• targetName the target name

• targetId the target cohort id

• outcomeName the outcome name

• outcomeId the outcome cohort id

• indicationName the indication name

• indicatonId the indication cohort id

• covariateName whether main or secondary analysis

• outcomeSubjects The number of subjects with at least one outcome.

• outcomeEvents The number of outcome events.

• outcomeObservationPeriods The number of observation periods containing at least one outcome.

• covariateSubjects The number of subjects having the covariate.

• covariateDays The total covariate time in days.

• covariateEras The number of continuous eras of the covariate.

• covariateOutcomes The number of outcomes observed during the covariate time.

• observedDays The number of days subjects were observed.

• rr the relative risk

• ci95Lb the lower bound of the 95 percent confidence interval for the relative risk

• ci95Ub the upper bound of the 95 percent confidence interval for the relative risk

• p the p-value for the relative risk

• logRr the log of the relative risk

• seLogRr the standard error or the log of the relative risk

• calibratedRr the calibrated relative risk

• calibratedCi95Lb the lower bound of the 95 percent confidence interval for the calibrated relative risk

• calibratedCi95Ub the upper bound of the 95 percent confidence interval for the calibrated relative risk

• calibratedP the calibrated p-value

• calibratedOneSidedP the calibrated one sided p-value

• calibratedLogRr the calibrated log of the relative risk

• calibratedSeLogRr the calibrated log of the relative risk standard error

• llr The log of the likelihood ratio (of the MLE vs the null hypothesis of no effect).

• mdrr The minimum detectable relative risk.

• unblind Whether the results can be unblinded

See Also

Other Estimation: getCMEstimation(), getCmDiagnosticsData(), getCmMetaEstimation(), getSccsDiagnosticsData(), getSccsMetaEstimation(), plotCmEstimates(), plotSccsEstimates()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

sccsEst <- getSccsEstimation(
  connectionHandler = connectionHandler, 
  schema = 'main',
  targetIds = 1,
  outcomeIds = 3
)

Extract the self controlled case series (sccs) meta analysis results

Description

This function extracts any meta analysis estimation results for sccs.

Usage

getSccsMetaEstimation(
  connectionHandler,
  schema,
  sccsTablePrefix = "sccs_",
  cgTablePrefix = "cg_",
  esTablePrefix = "es_",
  targetIds = NULL,
  outcomeIds = NULL
)

Arguments

Details

Specify the connectionHandler, the schema and the targetoutcome cohort IDs

Value

Returns a data.frame with the columns:

#'

• databaseName the database name

• analysisId the analysis unique identifier

• description an analysis description

• targetName the target name

• targetId the target cohort id

• outcomeName the outcome name

• outcomeId the outcome cohort id

• indicationName the indicationname

• indicationId the indication cohort id

• covariateName whether main or secondary analysis

• outcomeSubjects The number of subjects with at least one outcome.

• outcomeEvents The number of outcome events.

• outcomeObservationPeriods The number of observation periods containing at least one outcome.

• covariateSubjects The number of subjects having the covariate.

• covariateDays The total covariate time in days.

• covariateEras The number of continuous eras of the covariate.

• covariateOutcomes The number of outcomes observed during the covariate time.

• observedDays The number of days subjects were observed.

• calibratedRr the calibrated relative risk

• calibratedCi95Lb the lower bound of the 95 percent confidence interval for the calibrated relative risk

• calibratedCi95Ub the upper bound of the 95 percent confidence interval for the calibrated relative risk

• calibratedP the calibrated p-value

• calibratedOneSidedP the calibrated one sided p-value

• calibratedLogRr the calibrated log of the relative risk

• calibratedSeLogRr the calibrated log of the relative risk standard error

• nDatabases The number of databases included in the estimate.

See Also

Other Estimation: getCMEstimation(), getCmDiagnosticsData(), getCmMetaEstimation(), getSccsDiagnosticsData(), getSccsEstimation(), plotCmEstimates(), plotSccsEstimates()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

sccsMeta <- getSccsMetaEstimation(
  connectionHandler = connectionHandler, 
  schema = 'main',
  targetIds = 1,
  outcomeIds = 3
)

Extract aggregate statistics of binary feature analysis IDs of interest for targets

Description

This function extracts the feature extraction results for targets corresponding to specified target and outcome cohorts.

Usage

getTargetBinaryFeatures(
  connectionHandler,
  schema,
  cTablePrefix = "c_",
  cgTablePrefix = "cg_",
  databaseTable = "database_meta_data",
  targetIds = NULL,
  outcomeIds = NULL,
  analysisIds = c(3)
)

Arguments

Details

Specify the connectionHandler, the schema and the target/outcome cohort IDs

Value

Returns a data.frame with the columns:

• databaseName the name of the database

• targetName the target cohort name

• targetId the target cohort unique identifier

• outcomeName the outcome name

• outcomeId the outcome unique identifier

• minPriorObservation the minimum required observation days prior to index for an entry

• outcomeWashoutDays patients with the outcome occurring within this number of days prior to index are excluded (NA means no exclusion)

• covariateId the id of the feature

• covariateName the name of the feature

• sumValue the number of target patients who have the feature value of 1 (minus those excluded due to having the outcome prior)

• rawSum the number of target patients who have the feature value of 1 (ignoring exclusions)

• rawAverage the fraction of target patients who have the feature value of 1 (ignoring exclusions)

See Also

Other Characterization: getBinaryCaseSeries(), getBinaryRiskFactors(), getCaseBinaryFeatures(), getCaseContinuousFeatures(), getCaseCounts(), getCharacterizationDemographics(), getContinuousCaseSeries(), getContinuousRiskFactors(), getDechallengeRechallenge(), getIncidenceRates(), getTargetContinuousFeatures(), getTargetCounts(), getTimeToEvent(), plotAgeDistributions(), plotSexDistributions()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

tbf <- getTargetBinaryFeatures (
connectionHandler = connectionHandler, 
schema = 'main'
)

Extract aggregate statistics of continuous feature analysis IDs of interest for targets

Description

This function extracts the continuous feature extraction results for targets corresponding to specified target cohorts.

Usage

getTargetContinuousFeatures(
  connectionHandler,
  schema,
  cTablePrefix = "c_",
  cgTablePrefix = "cg_",
  databaseTable = "database_meta_data",
  targetIds = NULL,
  analysisIds = NULL
)

Arguments

Details

Specify the connectionHandler, the schema and the target/outcome cohort IDs

Value

Returns a data.frame with the columns:

• databaseName the name of the database

• targetName the target cohort name

• targetId the target cohort unique identifier

• minPriorObservation the minimum required observation days prior to index for an entry

• covariateName the name of the feature

• covariateId the id of the feature

• countValue the number of cases who have the feature

• minValue the minimum value observed for the feature

• maxValue the maximum value observed for the feature

• averageValue the mean value observed for the feature

• standardDeviation the standard deviation of the value observed for the feature

• medianValue the median value observed for the feature

• p10Value the 10th percentile of the value observed for the feature

• p25Value the 25th percentile of the value observed for the feature

• p75Value the 75th percentile of the value observed for the feature

• p90Value the 90th percentile of the value observed for the feature

See Also

Other Characterization: getBinaryCaseSeries(), getBinaryRiskFactors(), getCaseBinaryFeatures(), getCaseContinuousFeatures(), getCaseCounts(), getCharacterizationDemographics(), getContinuousCaseSeries(), getContinuousRiskFactors(), getDechallengeRechallenge(), getIncidenceRates(), getTargetBinaryFeatures(), getTargetCounts(), getTimeToEvent(), plotAgeDistributions(), plotSexDistributions()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

tcf <- getTargetContinuousFeatures(
connectionHandler = connectionHandler, 
schema = 'main'
)

Extract the target cohort counts result

Description

This function extracts target cohort counts across databases in the results for specified target and outcome cohorts.

Usage

getTargetCounts(
  connectionHandler,
  schema,
  cTablePrefix = "c_",
  cgTablePrefix = "cg_",
  databaseTable = "database_meta_data",
  targetIds = NULL,
  outcomeIds = NULL
)

Arguments

Details

Specify the connectionHandler, the schema and the target/outcome cohort IDs

Value

Returns a data.frame with the columns:

• databaseName the name of the database

• targetName the target cohort name

• targetId the target cohort unique identifier

• outcomeName the outcome name

• outcomeId the outcome unique identifier

• rowCount the number of entries in the cohort

• personCount the number of people in the cohort

• minPriorObservation the minimum required observation days prior to index for an entry

• outcomeWashoutDays patients with the outcome occurring within this number of days prior to index are excluded (NA means no exclusion)

See Also

Other Characterization: getBinaryCaseSeries(), getBinaryRiskFactors(), getCaseBinaryFeatures(), getCaseContinuousFeatures(), getCaseCounts(), getCharacterizationDemographics(), getContinuousCaseSeries(), getContinuousRiskFactors(), getDechallengeRechallenge(), getIncidenceRates(), getTargetBinaryFeatures(), getTargetContinuousFeatures(), getTimeToEvent(), plotAgeDistributions(), plotSexDistributions()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

tc <- getTargetCounts(
connectionHandler = connectionHandler, 
schema = 'main'
)

Extract the time to event result

Description

This function extracts all time to event results across databases for specified target and outcome cohorts.

Usage

getTimeToEvent(
  connectionHandler,
  schema,
  cTablePrefix = "c_",
  cgTablePrefix = "cg_",
  databaseTable = "database_meta_data",
  targetIds = NULL,
  outcomeIds = NULL
)

Arguments

Details

Specify the connectionHandler, the schema and the target/outcome cohort IDs

Value

Returns a data.frame with the columns:

• databaseName the name of the database

• targetName the target cohort name

• targetId the target cohort unique identifier

• outcomeName the outcome name

• outcomeId the outcome unique identifier

• outcomeType Whether the outcome is the first or subsequent

• targetOutcomeType The interval that the outcome occurs

• timeToEvent The number of days from index

• numEvents The number of target cohort entries

• timeScale The correspondin time-scale

See Also

Other Characterization: getBinaryCaseSeries(), getBinaryRiskFactors(), getCaseBinaryFeatures(), getCaseContinuousFeatures(), getCaseCounts(), getCharacterizationDemographics(), getContinuousCaseSeries(), getContinuousRiskFactors(), getDechallengeRechallenge(), getIncidenceRates(), getTargetBinaryFeatures(), getTargetContinuousFeatures(), getTargetCounts(), plotAgeDistributions(), plotSexDistributions()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

tte <- getTimeToEvent(
connectionHandler = connectionHandler, 
schema = 'main'
)
 

output a nicely formatted html table

Description

This returns a html table with the input data

Usage

kableDark(data, caption = NULL, position = NULL)

Arguments

Details

Input the data that you want to be shown via a dark html table

Value

An object of class 'knitr_kable' that will show the data via a nice html table

See Also

Other helper: addTarColumn(), formatBinaryCovariateName(), getExampleConnectionDetails(), printReactable(), removeSpaces()

Examples

kableDark(
data = data.frame(a=1,b=4), 
caption = 'A made up table to demonstrate this function',
position = 'h'
)

Plots the age distributions using the binary age groups

Description

Creates bar charts for the target and case age groups.

Usage

plotAgeDistributions(
  ageData,
  riskWindowStart = "1",
  riskWindowEnd = "365",
  startAnchor = "cohort start",
  endAnchor = "cohort start"
)

Arguments

Details

Input the data returned from 'getCharacterizationDemographics(type = 'age')' and the time-at-risk

Value

Returns a ggplot with the distributions

See Also

Other Characterization: getBinaryCaseSeries(), getBinaryRiskFactors(), getCaseBinaryFeatures(), getCaseContinuousFeatures(), getCaseCounts(), getCharacterizationDemographics(), getContinuousCaseSeries(), getContinuousRiskFactors(), getDechallengeRechallenge(), getIncidenceRates(), getTargetBinaryFeatures(), getTargetContinuousFeatures(), getTargetCounts(), getTimeToEvent(), plotSexDistributions()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

ageData <- getCharacterizationDemographics(
connectionHandler = connectionHandler, 
schema = 'main',
targetId = 1, 
outcomeId = 3, 
type = 'age'
)

plotAgeDistributions(ageData = ageData)

Plots the cohort method results for one analysis

Description

Creates nice cohort method plots

Usage

plotCmEstimates(
  cmData,
  cmMeta = NULL,
  targetName,
  comparatorName,
  selectedAnalysisId
)

Arguments

Details

Input the cohort method data

Value

Returns a ggplot with the estimates

See Also

Other Estimation: getCMEstimation(), getCmDiagnosticsData(), getCmMetaEstimation(), getSccsDiagnosticsData(), getSccsEstimation(), getSccsMetaEstimation(), plotSccsEstimates()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

cmEst <- getCMEstimation(
  connectionHandler = connectionHandler, 
  schema = 'main',
  targetIds = 1,
  outcomeIds = 3
)
plotCmEstimates(
  cmData = cmEst, 
  cmMeta = NULL, 
  targetName = 'target', 
  comparatorName = 'comp', 
  selectedAnalysisId = 1
)

Plots the self controlled case series results for one analysis

Description

Creates nice self controlled case series plots

Usage

plotSccsEstimates(sccsData, sccsMeta = NULL, targetName, selectedAnalysisId)

Arguments

Details

Input the self controlled case series data

Value

Returns a ggplot with the estimates

See Also

Other Estimation: getCMEstimation(), getCmDiagnosticsData(), getCmMetaEstimation(), getSccsDiagnosticsData(), getSccsEstimation(), getSccsMetaEstimation(), plotCmEstimates()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

sccsEst <- getSccsEstimation(
  connectionHandler = connectionHandler, 
  schema = 'main',
  targetIds = 1,
  outcomeIds = 3
)
plotSccsEstimates(
  sccsData = sccsEst, 
  sccsMeta = NULL, 
  targetName = 'target', 
  selectedAnalysisId = 1
)

Plots the sex distributions using the sex features

Description

Creates bar charts for the target and case sex.

Usage

plotSexDistributions(
  sexData,
  riskWindowStart = "1",
  riskWindowEnd = "365",
  startAnchor = "cohort start",
  endAnchor = "cohort start"
)

Arguments

Details

Input the data returned from 'getCharacterizationDemographics(type = 'sex')' and the time-at-risk

Value

Returns a ggplot with the distributions

See Also

Other Characterization: getBinaryCaseSeries(), getBinaryRiskFactors(), getCaseBinaryFeatures(), getCaseContinuousFeatures(), getCaseCounts(), getCharacterizationDemographics(), getContinuousCaseSeries(), getContinuousRiskFactors(), getDechallengeRechallenge(), getIncidenceRates(), getTargetBinaryFeatures(), getTargetContinuousFeatures(), getTargetCounts(), getTimeToEvent(), plotAgeDistributions()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

sexData <- getCharacterizationDemographics(
  connectionHandler = connectionHandler, 
  schema = 'main',
  targetId = 1, 
  outcomeId = 3, 
  type = 'sex'
)
plotSexDistributions(sexData = sexData)

prints a reactable in a quarto document

Description

This function lets you print a reactable in a quarto document

Usage

printReactable(
  data,
  columns = NULL,
  groupBy = NULL,
  defaultPageSize = 20,
  highlight = TRUE,
  striped = TRUE,
  searchable = TRUE,
  filterable = TRUE
)

Arguments

Details

Input the values for reactable::reactable

Value

Nothing but the html code for the table is printed (to be used in a quarto document)

See Also

Other helper: addTarColumn(), formatBinaryCovariateName(), getExampleConnectionDetails(), kableDark(), removeSpaces()

Examples

printReactable(
data = data.frame(a=1,b=4)
)

Extract the cohort parents and children cohorts (cohorts derieved from the parent cohort)

Description

This function lets you split the cohort data.frame into the parents and the children per parent.

Usage

processCohorts(cohort)

Arguments

Details

Finds the parent cohorts and children cohorts

Value

Returns a list containing parents: a named vector of all the parent cohorts and cohortList: a list the same length as the parent vector with the first element containing all the children of the first parent cohort, the second element containing the children of the second parent, etc.

See Also

Other Cohorts: getCohortDefinitions(), getCohortSubsetDefinitions()

Examples

conDet <- getExampleConnectionDetails()

connectionHandler <- ResultModelManager::ConnectionHandler$new(conDet)

cohortDef <- getCohortDefinitions(
  connectionHandler = connectionHandler, 
  schema = 'main'
)

parents <- processCohorts(cohortDef)

removeSpaces

Description

Removes spaces and replaces with under scroll

Usage

removeSpaces(x)

Arguments

Details

Removes spaces and replaces with under scroll

Value

A string without spaces

See Also

Other helper: addTarColumn(), formatBinaryCovariateName(), getExampleConnectionDetails(), kableDark(), printReactable()

Examples

removeSpaces(' made up.   string')