Package 'CohortSymmetry'

Intersecting the index and marker cohorts prior to calculating Sequence Symmetry Ratios

Description

Join two tables in the CDM (one for index and the other for marker cohorts) into a new table in the cdm taking into account the maximum time interval between events. Index and marker cohorts should be instantiated in advance by the user.

Usage

generateSequenceCohortSet(
  cdm,
  indexTable,
  markerTable,
  name,
  indexId = NULL,
  markerId = NULL,
  cohortDateRange = as.Date(c(NA, NA)),
  daysPriorObservation = 0,
  washoutWindow = 0,
  indexMarkerGap = Inf,
  combinationWindow = c(0, 365),
  movingAverageRestriction = 548
)

Arguments

cdm	A CDM reference.
indexTable	A table in the CDM that the index cohorts should come from.
markerTable	A table in the CDM that the marker cohorts should come from.
name	The name within the cdm that the output is called. Default is joined_cohorts.
indexId	Cohort definition IDs in indexTable to be considered for the analysis. Change to NULL if all indices are wished to be included.
markerId	Cohort definition IDs in markerTable to be considered for the analysis. Change to NULL if all markers are wished to be included.
cohortDateRange	Two dates indicating study period and the sequences that the user wants to restrict to.
daysPriorObservation	The minimum amount of prior observation required on both the index and marker cohorts per person.
washoutWindow	A washout window to be applied on both the index cohort event and marker cohort.
indexMarkerGap	The maximum allowable gap between the end of the first episode and the start of the second episode in a sequence/combination.
combinationWindow	A constrain to be placed on the gap between two initiations. Default c(0,365), meaning the gap should be larger than 0 but less than or equal to 365.
movingAverageRestriction	The moving window when calculating nSR, default is 548.

Value

A table within the cdm reference.

Examples

library(CohortSymmetry)
cdm <- mockCohortSymmetry()
cdm <- generateSequenceCohortSet(
  cdm = cdm,
  name = "joined_cohorts",
  indexTable = "cohort_1",
  markerTable = "cohort_2"
)
 cdm$joined_cohorts
 CDMConnector::cdmDisconnect(cdm = cdm)

---

Creates mock cdm object for testing

Description

Creates a mock cdm with two default synthetic cohorts, one is the index cohort and the other one is the marker cohort. However the users could specify them should they wish.

Usage

mockCohortSymmetry(
  seed = 1,
  indexCohort = NULL,
  markerCohort = NULL,
  con = DBI::dbConnect(duckdb::duckdb(), ":memory:"),
  schema = "main"
)

Arguments

seed	The seed to be inputted.
indexCohort	The tibble of your index cohort. Default is NULL, which means the default indexCohort is being used.
markerCohort	The tibble of your marker cohort. Default is NULL, which means the default markerCohort is being used.
con	Connection detail.
schema	Name of your write schema.

Value

A mock cdm object contains your index and marker cohort

Examples

library(CohortSymmetry)
cdm <- mockCohortSymmetry()
cdm
CDMConnector::cdmDisconnect(cdm = cdm)

---

A plot for the sequence ratios.

Description

It provides a ggplot of the sequence ratios of index and marker cohorts.

Usage

plotSequenceRatios(
  result,
  onlyaSR = FALSE,
  plotTitle = NULL,
  labs = c("SR", "Drug Pairs"),
  colours = c("red", "blue")
)

Arguments

result	Table output from summariseSequenceRatios.
onlyaSR	If the only SR to be plotted is the adjusted SR.
plotTitle	Title of the plot, if NULL no title will be plotted.
labs	Axis labels for the plot.
colours	Colours for both parts of the plot, pre- and post- time 0.

Value

A plot for the sequence ratios of index and marker cohorts.

Examples

library(CohortSymmetry)

cdm <- mockCohortSymmetry()
cdm <- generateSequenceCohortSet(cdm = cdm,
                                 indexTable = "cohort_1",
                                 markerTable = "cohort_2",
                                 name = "joined_cohort")
sequence_ratio <- summariseSequenceRatios(cohort = cdm$joined_cohort,
                                          minCellCount = 0)
plotSequenceRatios(result = sequence_ratio)
CDMConnector::cdmDisconnect(cdm = cdm)

---

A plot for the temporal symmetry of cohorts.

Description

It provides a ggplot of the temporal symmetry of two or more cohorts.

Usage

plotTemporalSymmetry(
  result,
  plotTitle = NULL,
  labs = c("Time (months)", "Individuals (N)"),
  xlim = c(-12, 12),
  colours = c("blue", "red"),
  scales = "free"
)

Arguments

result	Table output from summariseTemporalSymmetry.
plotTitle	Title of the plot, if NULL no title will be plotted.
labs	Axis labels for the plot.
xlim	Limits for the x axis of the plot.
colours	Colours for both parts of the plot, pre- and post- time 0.
scales	Whether to set free y scales for the facet wrap when there are multiple plots (i.e. each plot has its own scaled y axis) or set them equal for all. Only accepts "free" for the former and "fixed" for the latter.

Value

A plot for the temporal symmetry of cohorts.

Examples

library(CohortSymmetry)
cdm <- mockCohortSymmetry()
cdm <- generateSequenceCohortSet(cdm = cdm,
                                 indexTable = "cohort_1",
                                 markerTable = "cohort_2",
                                 name = "joined_cohort")
temporal_symmetry <- summariseTemporalSymmetry(cohort = cdm$joined_cohort,
                                               minCellCount = 0)
plotTemporalSymmetry(result = temporal_symmetry)
CDMConnector::cdmDisconnect(cdm = cdm)

---

Sequence ratio calculations

Description

Using generateSequenceCohortSet to obtain sequence ratios for the desired outcomes.

Usage

summariseSequenceRatios(
  cohort,
  cohortId = NULL,
  confidenceInterval = 95,
  minCellCount = 5
)

Arguments

cohort	A cohort table in the cdm.
cohortId	The Ids in the cohort that are to be included in the analyses.
confidenceInterval	Default is 95, indicating the central 95% confidence interval.
minCellCount	The minimum number of events to reported, below which results will be obscured. If 0, all results will be reported.

Value

A local table with all the analyses.

Examples

library(CohortSymmetry)
cdm <- mockCohortSymmetry()
cdm <- generateSequenceCohortSet(cdm = cdm,
                                 name = "joined_cohorts",
                                 indexTable = "cohort_1",
                                 markerTable = "cohort_2")
pssa_result <- summariseSequenceRatios(cohort = cdm$joined_cohorts, minCellCount = 0)
pssa_result
CDMConnector::cdmDisconnect(cdm)

---

Summarise temporal symmetry

Description

Using generateSequenceCohortSet to obtain temporal symmetry (aggregated counts) of two cohorts.

Usage

summariseTemporalSymmetry(
  cohort,
  cohortId = NULL,
  timescale = "month",
  minCellCount = 5
)

Arguments

cohort	A cohort table in the cdm.
cohortId	The Ids in the cohort that are to be included in the analyses.
timescale	Timescale for the x axis of the plot (month, day, year).
minCellCount	The minimum number of events to reported, below which results will be obscured. If 0, all results will be reported.

Value

An aggregated table with difference in time (marker - index) and the relevant counts.

Examples

library(CohortSymmetry)
cdm <- mockCohortSymmetry()
cdm <- generateSequenceCohortSet(cdm = cdm,
                                 name = "joined_cohorts",
                                 indexTable = "cohort_1",
                                 markerTable = "cohort_2")
temporal_symmetry <- summariseTemporalSymmetry(cohort = cdm$joined_cohorts)
CDMConnector::cdmDisconnect(cdm)

---

A formatted visualization of sequence_symmetry objects.

Description

It provides a formatted table with the contents of the summariseSequenceRatios output.

Usage

tableSequenceRatios(
  result,
  type = "gt",
  estimateNameFormat = c(`N (%)` = "<count> (<percentage> %)", `SR (CI)` =
    "<point_estimate> (<lower_CI> - <upper_CI>)"),
  style = "default",
  studyPopulation = TRUE,
  cdmName = TRUE,
  .options = NULL
)

Arguments

result	A sequence_symmetry object.
type	Type of desired formatted table, possibilities: "gt", "flextable", "tibble".
estimateNameFormat	The columns that the user wishes to see for the formatted table, by default it would display both the counts and sequence ratios.
style	Named list that specifies how to style the different parts of a gt table or flextable. See visOmopResults package for more information on how to define a style. Alternatively, use "default" to get visOmopResults style, or NULL for gt/flextable default styling.
studyPopulation	whether to report the study population.
cdmName	whether to report database names.
.options	named list with additional formatting options. tableSequenceRatiosOptions() shows allowed arguments and their default values.

Value

A formatted version of the sequence_symmetry object.

Examples

library(CohortSymmetry)
cdm <- mockCohortSymmetry()
cdm <- generateSequenceCohortSet(cdm = cdm,
                                 indexTable = "cohort_1",
                                 markerTable = "cohort_2",
                                 name = "joined_cohort")
res <- summariseSequenceRatios(cohort = cdm$joined_cohort)
gtResult <- tableSequenceRatios(res)
CDMConnector::cdmDisconnect(cdm = cdm)

---

A formatted visualization of sequence_ratios objects.

Description

It provides a list of allowed inputs for .option argument in tableSequenceRatios and their given default value.

Usage

tableSequenceRatiosOptions()

Value

The default .options named list.

Examples

{
 tableSequenceRatiosOptions()
}

---