Summarise cohort entries

1.1 Introduction

In this example we’re going to summarise the characteristics of individuals with an ankle sprain, ankle fracture, forearm fracture, or a hip fracture using the Eunomia synthetic data.

We’ll begin by creating our study cohorts.

library(CDMConnector)
library(CohortCharacteristics)
library(dplyr)
library(ggplot2)

con <- DBI::dbConnect(duckdb::duckdb(),
  dbdir = CDMConnector::eunomia_dir()
)
cdm <- CDMConnector::cdm_from_con(con,
  cdm_schem = "main",
  write_schema = "main",
  cdm_name = "Eunomia"
)

cdm <- generateConceptCohortSet(
  cdm = cdm,
  name = "injuries",
  conceptSet = list(
    "ankle_sprain" = 81151,
    "ankle_fracture" = 4059173,
    "forearm_fracture" = 4278672,
    "hip_fracture" = 4230399
  ),
  end = "event_end_date",
  limit = "all"
)

1.2 Summarising cohort counts

We can first quickly summarise and present the overall counts of our cohorts.

cohort_counts <- summariseCohortCount(cdm[["injuries"]])
tableCohortCount(cohort_counts)

CDM name	Variable name	Estimate name	Cohort name
			Ankle sprain	Ankle fracture	Hip fracture	Forearm fracture
Eunomia	Number records	N	1,915	464	138	569
	Number subjects	N	1,357	427	132	510

Moreover, we can also easily stratify these counts. For example, here we add age groups and then stratify our counts by t We can summarise the overall counts of our cohorts.

cdm[["injuries"]] <- cdm[["injuries"]] |>
  PatientProfiles::addAge(ageGroup = list(
    c(0, 3),
    c(4, 17),
    c(18, Inf)
  )) |>
  compute(temporary = FALSE, name = "injuries")

cohort_counts <- summariseCohortCount(cdm[["injuries"]], strata = "age_group")
tableCohortCount(cohort_counts)

CDM name	Age group	Variable name	Estimate name	Cohort name
				Ankle sprain	Ankle fracture	Forearm fracture	Hip fracture
Eunomia	Overall	Number records	N	1,915	464	569	138
	0 to 3	Number records	N	202	49	51	7
	18 or above	Number records	N	1,047	213	268	88
	4 to 17	Number records	N	666	202	250	43
	Overall	Number subjects	N	1,357	427	510	132
	0 to 3	Number subjects	N	196	49	51	7
	18 or above	Number subjects	N	847	204	249	83
	4 to 17	Number subjects	N	597	195	239	43

We can also apply minimum cell count suppression to our cohort counts. In this case we will obscure any counts below 10.

cohort_counts <- suppress(cohort_counts, minCellCount = 10)
tableCohortCount(cohort_counts)

CDM name	Age group	Variable name	Estimate name	Cohort name
				Ankle sprain	Ankle fracture	Forearm fracture	Hip fracture
Eunomia	Overall	Number records	N	1,915	464	569	138
	0 to 3	Number records	N	202	49	51	<10
	18 or above	Number records	N	1,047	213	268	88
	4 to 17	Number records	N	666	202	250	43
	Overall	Number subjects	N	1,357	427	510	132
	0 to 3	Number subjects	N	196	49	51	<10
	18 or above	Number subjects	N	847	204	249	83
	4 to 17	Number subjects	N	597	195	239	43

1.3 Summarising cohort attrition

Say we specify two inclusion criteria. First, we keep only cohort entries after the year 2000. Second, we keep only cohort entries for those aged 18 or older. We can easily create plots summarising our cohort attrition.

cdm <- generateConceptCohortSet(
  cdm = cdm,
  name = "ankle_sprain",
  conceptSet = list("ankle_sprain" = 81151),
  end = "event_end_date",
  limit = "all"
)

cdm[["ankle_sprain"]] <- cdm[["ankle_sprain"]] |>
  filter(year(cohort_start_date) >= 2000) |>
  recordCohortAttrition("Restrict to cohort_start_date >= 2000") |>
  compute(temporary = FALSE, name = "ankle_sprain")

attrition_summary <- summariseCohortAttrition(cdm[["ankle_sprain"]])

plotCohortAttrition(attrition_summary)

cdm[["ankle_sprain"]] <- cdm[["ankle_sprain"]] |>
  PatientProfiles::addAge() |>
  filter(age >= 18) |>
  compute(temporary = FALSE, name = "ankle_sprain") |>
  recordCohortAttrition("Restrict to age >= 18")

attrition_summary <- summariseCohortAttrition(cdm[["ankle_sprain"]])

plotCohortAttrition(attrition_summary, cohortId = 1)

We could, of course, have applied these requirements the other way around.

cdm <- generateConceptCohortSet(
  cdm = cdm,
  name = "ankle_sprain",
  conceptSet = list("ankle_sprain" = 81151),
  end = "event_end_date",
  limit = "all"
)

cdm[["ankle_sprain"]] <- cdm[["ankle_sprain"]] |>
  PatientProfiles::addAge() |>
  filter(age >= 18) |>
  compute(temporary = FALSE, name = "ankle_sprain") |>
  recordCohortAttrition("Restrict to age >= 18")

cdm[["ankle_sprain"]] <- cdm[["ankle_sprain"]] |>
  filter(year(cohort_start_date) >= 2000) |>
  recordCohortAttrition("Restrict to cohort_start_date >= 2000") |>
  compute(temporary = FALSE, name = "ankle_sprain")


attrition_summary <- summariseCohortAttrition(cdm[["ankle_sprain"]])

plotCohortAttrition(attrition_summary, cohortId = 1)

As well as plotting cohort attrition, we can also create a table of our results.

tableCohortAttrition(attrition_summary)

	CDM name
	Eunomia
Reason	Variable
	Number records	Number subjects	Excluded records	Excluded subjects
ankle_sprain
Initial qualifying events	1915	1357	0	0
Restrict to age >= 18	1047	847	868	510
Restrict to cohort_start_date >= 2000	454	420	593	427