CohortCharacteristics

Characterise cohorts

2025-07-06

CohortCharacteristics

Context

This package aims to standardise and provide the tools to conduct Characterisation studies as of the Darwin-EU Catalogue of Standard Analytics.

Package overview

Functions

Workflow

We have three types of functions:

  • summarise: these functions produce an standardised output to summarise a cohort. This standard output is called summarised_result.

  • plot: these functions produce plots (currently, only ggplot, but working to implement plotly) from a summarised_result object.

  • table: these functions produce tables (gt and flextable) from a summarised_result object.

result <- summariseXXX(...)
tableXXX(result)
plotXXX(result)

flowchart LR
  A[summarise function ] --> B[Plot function ]
  A --> C[Table function ]

Loadd needed packages

Create GiBleed cdm_reference object

datasetName <- "GiBleed"
requireEunomia(datasetName = datasetName)
con <- dbConnect(drv = duckdb(dbdir = eunomiaDir(datasetName = datasetName)))
Creating CDM database /home/runner/work/darwinTutorial2025/darwinTutorial2025/mock_datasets//GiBleed_5.3.zip
cdm <- cdmFromCon(con = con, cdmSchema = "main", writeSchema = "main", cdmName = datasetName)
cdm

── # OMOP CDM reference (duckdb) of GiBleed ────────────────────────────────────────────────────────────────────────────
• omop tables: person, observation_period, visit_occurrence, visit_detail, condition_occurrence, drug_exposure,
procedure_occurrence, device_exposure, measurement, observation, death, note, note_nlp, specimen, fact_relationship,
location, care_site, provider, payer_plan_period, cost, drug_era, dose_era, condition_era, metadata, cdm_source,
concept, vocabulary, domain, concept_class, concept_relationship, relationship, concept_synonym, concept_ancestor,
source_to_concept_map, drug_strength
• cohort tables: -
• achilles tables: -
• other tables: -

Let’ create a acetaminophen cohort

Let’s create a cohort that is all records of acetaminophen with the following inclusion criteria:

  • Collapse records separated by 30 days or less in the same episode.
  • No prior record of acetaminophen in the last 180 days.
  • At least 180 days of prior observation.
  • Index date between 1990 and 2019

Let’ create a acetaminophen cohort

codelist <- getDrugIngredientCodes(cdm = cdm, name = "acetaminophen", nameStyle = "{concept_name}")
cdm$acetaminophen <- conceptCohort(
  cdm = cdm, 
  conceptSet = codelist, 
  name = "acetaminophen",
  exit = "event_end_date"
) |>
  collapseCohorts(gap = 30) |>
  requireCohortIntersect(
    targetCohortTable = "acetaminophen", 
    window = c(-180, -1), 
    intersections = 0
  ) |>
  requirePriorObservation(minPriorObservation = 180) |>
  requireInDateRange(dateRange = as.Date(c("1990-01-01", "2019-12-31")))

summariseCohortAttrition

summariseCohortAttrition extracts the data from attrition(cdm$dus_cohort) in a standard format:

result <- summariseCohortAttrition(cohort = cdm$acetaminophen)
glimpse(x = result)
Rows: 44
Columns: 13
$ result_id        <int> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, …
$ cdm_name         <chr> "GiBleed", "GiBleed", "GiBleed", "GiBleed", "GiBleed", "GiBleed", "GiBleed", "GiBleed", "GiBl…
$ group_name       <chr> "cohort_name", "cohort_name", "cohort_name", "cohort_name", "cohort_name", "cohort_name", "co…
$ group_level      <chr> "acetaminophen", "acetaminophen", "acetaminophen", "acetaminophen", "acetaminophen", "acetami…
$ strata_name      <chr> "reason", "reason", "reason", "reason", "reason", "reason", "reason", "reason", "reason", "re…
$ strata_level     <chr> "Initial qualifying events", "Initial qualifying events", "Initial qualifying events", "Initi…
$ variable_name    <chr> "number_records", "number_subjects", "excluded_records", "excluded_subjects", "number_records…
$ variable_level   <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, N…
$ estimate_name    <chr> "count", "count", "count", "count", "count", "count", "count", "count", "count", "count", "co…
$ estimate_type    <chr> "integer", "integer", "integer", "integer", "integer", "integer", "integer", "integer", "inte…
$ estimate_value   <chr> "14205", "2679", "0", "0", "14205", "2679", "0", "0", "14205", "2679", "0", "0", "14205", "26…
$ additional_name  <chr> "reason_id", "reason_id", "reason_id", "reason_id", "reason_id", "reason_id", "reason_id", "r…
$ additional_level <chr> "1", "1", "1", "1", "2", "2", "2", "2", "3", "3", "3", "3", "4", "4", "4", "4", "5", "5", "5"…

summariseCohortAttrition

settings(result) |>
  glimpse()
Rows: 1
Columns: 13
$ result_id             <int> 1
$ result_type           <chr> "summarise_cohort_attrition"
$ package_name          <chr> "CohortCharacteristics"
$ package_version       <chr> "1.0.0"
$ group                 <chr> "cohort_name"
$ strata                <chr> "reason"
$ additional            <chr> "reason_id"
$ min_cell_count        <chr> "0"
$ cdm_version           <chr> "5.3"
$ cohort_definition_id  <chr> "1"
$ min_prior_observation <chr> "180"
$ table_name            <chr> "acetaminophen"
$ vocabulary_version    <chr> "v5.0 18-JAN-19"

brief mention of summaried_result

object consist on the main table that we have seen + a table of settings.

  • bind Join multiple results objects.

  • settings Read the settings of the result object.

  • suppress Suppress the results to counts less than a certain number.

  • exportSummarisedResult Export the result object to a csv file.

  • importSummarisedResult Import a result object from a csv file.

tableCohortAttrition

tableCohortAttrition(result = result)
Reason
Variable name
number_records number_subjects excluded_records excluded_subjects
GiBleed; acetaminophen
Initial qualifying events 14,205 2,679 0 0
Record start <= record end 14,205 2,679 0 0
Record in observation 14,205 2,679 0 0
Non-missing sex 14,205 2,679 0 0
Non-missing year of birth 14,205 2,679 0 0
Merge overlapping records 13,908 2,679 297 0
Collapse cohort with a gap of 30 days. 13,860 2,679 48 0
Not in cohort acetaminophen between -180 & -1 days relative to cohort_start_date 13,506 2,679 354 0
Prior observation requirement: 180 days 13,361 2,677 145 2
cohort_start_date after 1990-01-01 5,401 2,340 7,960 337
cohort_start_date before 2019-12-31 5,401 2,340 0 0

Export the gt table

gt tables can easily be exported:

tab <- tableCohortAttrition(result = result)
gtsave(data = tab, filename = here("attrition.docx")) # pdf/html/png also allowed

flextable is also supported

tableCohortAttrition(result = result, type = "flextable")

Reason

Variable name

number_records

number_subjects

excluded_records

excluded_subjects

GiBleed; acetaminophen

Initial qualifying events

14,205

2,679

0

0

Record start <= record end

14,205

2,679

0

0

Record in observation

14,205

2,679

0

0

Non-missing sex

14,205

2,679

0

0

Non-missing year of birth

14,205

2,679

0

0

Merge overlapping records

13,908

2,679

297

0

Collapse cohort with a gap of 30 days.

13,860

2,679

48

0

Not in cohort acetaminophen between -180 & -1 days relative to cohort_start_date

13,506

2,679

354

0

Prior observation requirement: 180 days

13,361

2,677

145

2

cohort_start_date after 1990-01-01

5,401

2,340

7,960

337

cohort_start_date before 2019-12-31

5,401

2,340

0

0

plotCohortAttrition

plotCohortAttrition(result = result)

Your turn

Can you visualise the attrition of a simvastatin cohort?

  • Any record of simvastatin
  • Collapse records separated by 90 days or less in the same episode.
  • First record ever.
  • At least 365 days of prior observation.
  • Index date between 1990 and 2019

Can you suppress counts under 10?

Result

Solution

💡 Click to see solution 
# create codelist
codelist <- getDrugIngredientCodes(cdm = cdm, name = "simvastatin")

# create cohort
cdm$simvastatin_cohort <- conceptCohort(
  cdm = cdm,
  conceptSet = codelist, 
  name = "simvastatin_cohort", 
  exit = "event_end_date"
) |>
  collapseCohorts(gap = 90) |>
  requireIsFirstEntry() |>
  requirePriorObservation(minPriorObservation = 365) |>
  requireInDateRange(dateRange = as.Date(c("1990-01-01", "2019-12-31")))

# summarise attrition
result <- cdm$simvastatin_cohort |>
  summariseCohortAttrition() |>
  suppress(minCellCount = 10)

# plot attrition
plotCohortAttrition(result)

summariseCharacteristics

summariseCharacteristics() summarises the demographics of the cohort by default, but you can use the intersect arguments to evaluate:

  • Number of visits
  • Presence of medicines or conditions
  • …

You need to identify the medications and/or conditions of interest as a codelist or cohort beforehand.

Let’s create some codelists

medications <- getDrugIngredientCodes(
  cdm = cdm, 
  name = c("warfarin", "morphine", "verapamil", "atorvastatin", "nitroglycerin"), 
  nameStyle = "{concept_name}"
)
conditions <- list(
  "fever" = getCandidateCodes(cdm = cdm, keywords = "fever")$concept_id,
  "cough" = getCandidateCodes(cdm = cdm, keywords = "cough")$concept_id,
  "cardiac_arrest" = getCandidateCodes(cdm = cdm, keywords = "cardiac arrest")$concept_id,
  "myocardial_infarction" = getCandidateCodes(cdm = cdm, keywords = "myocardial infarction")$concept_id,
  "headache" = getCandidateCodes(cdm = cdm, keywords = "headache")$concept_id
)

summariseCharacteristics

result <- cdm$acetaminophen |>
  addSex() |>
  summariseCharacteristics(
    strata = list("sex"),
    demographics = TRUE,
    ageGroup = list(c(0, 19), c(20, 39), c(40, 59), c(60, 79), c(80, Inf)),
    tableIntersectCount = list(
      "Number visits prior year" = list(tableName = "visit_occurrence", window = c(-365, -1))
    ),
    conceptIntersectFlag = list(
      "Conditions any time prior" = list(conceptSet = conditions, window = c(-Inf, -1)),
      "Medications prior year" = list(conceptSet = medications, window = c(-365, -1))
    )
  )
result |>
  glimpse()
Rows: 239
Columns: 13
$ result_id        <int> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, …
$ cdm_name         <chr> "GiBleed", "GiBleed", "GiBleed", "GiBleed", "GiBleed", "GiBleed", "GiBleed", "GiBleed", "GiBl…
$ group_name       <chr> "cohort_name", "cohort_name", "cohort_name", "cohort_name", "cohort_name", "cohort_name", "co…
$ group_level      <chr> "acetaminophen", "acetaminophen", "acetaminophen", "acetaminophen", "acetaminophen", "acetami…
$ strata_name      <chr> "overall", "overall", "overall", "overall", "overall", "overall", "overall", "overall", "over…
$ strata_level     <chr> "overall", "overall", "overall", "overall", "overall", "overall", "overall", "overall", "over…
$ variable_name    <chr> "Number records", "Number subjects", "Cohort start date", "Cohort start date", "Cohort start …
$ variable_level   <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, "0 to 19", "0 to …
$ estimate_name    <chr> "count", "count", "min", "q25", "median", "q75", "max", "min", "q25", "median", "q75", "max",…
$ estimate_type    <chr> "integer", "integer", "date", "date", "date", "date", "date", "date", "date", "date", "date",…
$ estimate_value   <chr> "5401", "2340", "1990-01-01", "1996-04-28", "2004-01-07", "2011-10-21", "2019-06-20", "1990-0…
$ additional_name  <chr> "overall", "overall", "overall", "overall", "overall", "overall", "overall", "overall", "over…
$ additional_level <chr> "overall", "overall", "overall", "overall", "overall", "overall", "overall", "overall", "over…

tableCharacteristics

tableCharacteristics(result = result)
CDM name
GiBleed
Sex Variable name Variable level Estimate name
Cohort name
acetaminophen
overall Number records - N 5,401
Number subjects - N 2,340
Cohort start date - Median [Q25 - Q75] 2004-01-07 [1996-04-28 - 2011-10-21]
Range 1990-01-01 to 2019-06-20
Cohort end date - Median [Q25 - Q75] 2004-01-21 [1996-05-12 - 2011-11-07]
Range 1990-01-10 to 2019-06-24
Age - Median [Q25 - Q75] 43 [32 - 55]
Mean (SD) 44.58 (17.93)
Range 5 to 108
Age group 0 to 19 N (%) 366 (6.78%)
20 to 39 N (%) 1,909 (35.35%)
40 to 59 N (%) 2,117 (39.20%)
60 to 79 N (%) 784 (14.52%)
80 or above N (%) 225 (4.17%)
Sex Female N (%) 2,687 (49.75%)
Male N (%) 2,714 (50.25%)
Prior observation - Median [Q25 - Q75] 15,837 [11,937 - 20,419]
Mean (SD) 16,464.35 (6,548.63)
Range 1,851 to 39,792
Future observation - Median [Q25 - Q75] 5,193 [2,385 - 8,100]
Mean (SD) 5,213.44 (3,203.25)
Range 0 to 10,720
Days in cohort - Median [Q25 - Q75] 15 [8 - 15]
Mean (SD) 16.95 (15.69)
Range 1 to 128
Number visits prior year - Median [Q25 - Q75] 0.00 [0.00 - 0.00]
Mean (SD) 0.01 (0.08)
Range 0.00 to 1.00
Conditions any time prior Cardiac arrest N (%) 210 (3.89%)
Myocardial infarction N (%) 17 (0.31%)
Cough N (%) 0 (0.00%)
Fever N (%) 0 (0.00%)
Headache N (%) 0 (0.00%)
Medications prior year Nitroglycerin N (%) 104 (1.93%)
Warfarin N (%) 8 (0.15%)
Verapamil N (%) 8 (0.15%)
Morphine N (%) 6 (0.11%)
Atorvastatin N (%) 0 (0.00%)
Female Number records - N 2,687
Number subjects - N 1,168
Cohort start date - Median [Q25 - Q75] 2004-02-01 [1996-08-10 - 2011-07-11]
Range 1990-01-02 to 2019-06-19
Cohort end date - Median [Q25 - Q75] 2004-02-19 [1996-08-20 - 2011-07-19]
Range 1990-01-16 to 2019-06-24
Age - Median [Q25 - Q75] 42 [32 - 55]
Mean (SD) 44.27 (18.08)
Range 5 to 108
Age group 0 to 19 N (%) 181 (6.74%)
20 to 39 N (%) 980 (36.47%)
40 to 59 N (%) 1,047 (38.97%)
60 to 79 N (%) 358 (13.32%)
80 or above N (%) 121 (4.50%)
Sex Female N (%) 2,687 (100.00%)
Prior observation - Median [Q25 - Q75] 15,632 [11,772 - 20,152]
Mean (SD) 16,353.62 (6,603.54)
Range 1,851 to 39,792
Future observation - Median [Q25 - Q75] 5,235 [2,602 - 8,028]
Mean (SD) 5,268.14 (3,141.99)
Range 0 to 10,713
Days in cohort - Median [Q25 - Q75] 15 [8 - 15]
Mean (SD) 17.28 (16.24)
Range 1 to 97
Number visits prior year - Median [Q25 - Q75] 0.00 [0.00 - 0.00]
Mean (SD) 0.01 (0.08)
Range 0.00 to 1.00
Conditions any time prior Cardiac arrest N (%) 120 (4.47%)
Myocardial infarction N (%) 1 (0.04%)
Cough N (%) 0 (0.00%)
Fever N (%) 0 (0.00%)
Headache N (%) 0 (0.00%)
Medications prior year Nitroglycerin N (%) 21 (0.78%)
Warfarin N (%) 2 (0.07%)
Verapamil N (%) 2 (0.07%)
Morphine N (%) 4 (0.15%)
Atorvastatin N (%) 0 (0.00%)
Male Number records - N 2,714
Number subjects - N 1,172
Cohort start date - Median [Q25 - Q75] 2003-12-10 [1996-03-11 - 2012-02-06]
Range 1990-01-01 to 2019-06-20
Cohort end date - Median [Q25 - Q75] 2003-12-26 [1996-03-27 - 2012-02-17]
Range 1990-01-10 to 2019-06-20
Age - Median [Q25 - Q75] 44 [32 - 56]
Mean (SD) 44.88 (17.78)
Range 5 to 105
Age group 0 to 19 N (%) 185 (6.82%)
20 to 39 N (%) 929 (34.23%)
40 to 59 N (%) 1,070 (39.43%)
60 to 79 N (%) 426 (15.70%)
80 or above N (%) 104 (3.83%)
Sex Male N (%) 2,714 (100.00%)
Prior observation - Median [Q25 - Q75] 16,107 [12,010 - 20,689]
Mean (SD) 16,573.97 (6,493.16)
Range 2,070 to 38,529
Future observation - Median [Q25 - Q75] 5,101 [2,223 - 8,132]
Mean (SD) 5,159.29 (3,262.43)
Range 0 to 10,720
Days in cohort - Median [Q25 - Q75] 15 [8 - 15]
Mean (SD) 16.63 (15.12)
Range 1 to 128
Number visits prior year - Median [Q25 - Q75] 0.00 [0.00 - 0.00]
Mean (SD) 0.00 (0.07)
Range 0.00 to 1.00
Conditions any time prior Cardiac arrest N (%) 90 (3.32%)
Myocardial infarction N (%) 16 (0.59%)
Cough N (%) 0 (0.00%)
Fever N (%) 0 (0.00%)
Headache N (%) 0 (0.00%)
Medications prior year Nitroglycerin N (%) 83 (3.06%)
Warfarin N (%) 6 (0.22%)
Verapamil N (%) 6 (0.22%)
Morphine N (%) 2 (0.07%)
Atorvastatin N (%) 0 (0.00%)

tableCharacteristics

tableCharacteristics(
  result = result,
  header = c("cdm_name", "sex"),
  groupColumn = c("cohort_name")
)
CDM name
GiBleed
Variable name Variable level Estimate name
Sex
overall Female Male
acetaminophen
Number records - N 5,401 2,687 2,714
Number subjects - N 2,340 1,168 1,172
Cohort start date - Median [Q25 - Q75] 2004-01-07 [1996-04-28 - 2011-10-21] 2004-02-01 [1996-08-10 - 2011-07-11] 2003-12-10 [1996-03-11 - 2012-02-06]
Range 1990-01-01 to 2019-06-20 1990-01-02 to 2019-06-19 1990-01-01 to 2019-06-20
Cohort end date - Median [Q25 - Q75] 2004-01-21 [1996-05-12 - 2011-11-07] 2004-02-19 [1996-08-20 - 2011-07-19] 2003-12-26 [1996-03-27 - 2012-02-17]
Range 1990-01-10 to 2019-06-24 1990-01-16 to 2019-06-24 1990-01-10 to 2019-06-20
Age - Median [Q25 - Q75] 43 [32 - 55] 42 [32 - 55] 44 [32 - 56]
Mean (SD) 44.58 (17.93) 44.27 (18.08) 44.88 (17.78)
Range 5 to 108 5 to 108 5 to 105
Age group 0 to 19 N (%) 366 (6.78%) 181 (6.74%) 185 (6.82%)
20 to 39 N (%) 1,909 (35.35%) 980 (36.47%) 929 (34.23%)
40 to 59 N (%) 2,117 (39.20%) 1,047 (38.97%) 1,070 (39.43%)
60 to 79 N (%) 784 (14.52%) 358 (13.32%) 426 (15.70%)
80 or above N (%) 225 (4.17%) 121 (4.50%) 104 (3.83%)
Sex Female N (%) 2,687 (49.75%) 2,687 (100.00%) -
Male N (%) 2,714 (50.25%) - 2,714 (100.00%)
Prior observation - Median [Q25 - Q75] 15,837 [11,937 - 20,419] 15,632 [11,772 - 20,152] 16,107 [12,010 - 20,689]
Mean (SD) 16,464.35 (6,548.63) 16,353.62 (6,603.54) 16,573.97 (6,493.16)
Range 1,851 to 39,792 1,851 to 39,792 2,070 to 38,529
Future observation - Median [Q25 - Q75] 5,193 [2,385 - 8,100] 5,235 [2,602 - 8,028] 5,101 [2,223 - 8,132]
Mean (SD) 5,213.44 (3,203.25) 5,268.14 (3,141.99) 5,159.29 (3,262.43)
Range 0 to 10,720 0 to 10,713 0 to 10,720
Days in cohort - Median [Q25 - Q75] 15 [8 - 15] 15 [8 - 15] 15 [8 - 15]
Mean (SD) 16.95 (15.69) 17.28 (16.24) 16.63 (15.12)
Range 1 to 128 1 to 97 1 to 128
Number visits prior year - Median [Q25 - Q75] 0.00 [0.00 - 0.00] 0.00 [0.00 - 0.00] 0.00 [0.00 - 0.00]
Mean (SD) 0.01 (0.08) 0.01 (0.08) 0.00 (0.07)
Range 0.00 to 1.00 0.00 to 1.00 0.00 to 1.00
Conditions any time prior Cardiac arrest N (%) 210 (3.89%) 120 (4.47%) 90 (3.32%)
Myocardial infarction N (%) 17 (0.31%) 1 (0.04%) 16 (0.59%)
Cough N (%) 0 (0.00%) 0 (0.00%) 0 (0.00%)
Fever N (%) 0 (0.00%) 0 (0.00%) 0 (0.00%)
Headache N (%) 0 (0.00%) 0 (0.00%) 0 (0.00%)
Medications prior year Nitroglycerin N (%) 104 (1.93%) 21 (0.78%) 83 (3.06%)
Warfarin N (%) 8 (0.15%) 2 (0.07%) 6 (0.22%)
Verapamil N (%) 8 (0.15%) 2 (0.07%) 6 (0.22%)
Morphine N (%) 6 (0.11%) 4 (0.15%) 2 (0.07%)
Atorvastatin N (%) 0 (0.00%) 0 (0.00%) 0 (0.00%)

tableCharacteristics

result |>
  filter(!estimate_name %in% c("min", "max", "mean", "sd")) |>
  tableCharacteristics(
    header = c("cdm_name", "sex"),
    groupColumn = c("cohort_name")
  )
CDM name Variable name Variable level Estimate name
Sex
overall Female Male
acetaminophen
GiBleed Number records - N 5,401 2,687 2,714
Number subjects - N 2,340 1,168 1,172
Cohort start date - Median [Q25 - Q75] 2004-01-07 [1996-04-28 - 2011-10-21] 2004-02-01 [1996-08-10 - 2011-07-11] 2003-12-10 [1996-03-11 - 2012-02-06]
Cohort end date - Median [Q25 - Q75] 2004-01-21 [1996-05-12 - 2011-11-07] 2004-02-19 [1996-08-20 - 2011-07-19] 2003-12-26 [1996-03-27 - 2012-02-17]
Age - Median [Q25 - Q75] 43 [32 - 55] 42 [32 - 55] 44 [32 - 56]
Age group 0 to 19 N (%) 366 (6.78%) 181 (6.74%) 185 (6.82%)
20 to 39 N (%) 1,909 (35.35%) 980 (36.47%) 929 (34.23%)
40 to 59 N (%) 2,117 (39.20%) 1,047 (38.97%) 1,070 (39.43%)
60 to 79 N (%) 784 (14.52%) 358 (13.32%) 426 (15.70%)
80 or above N (%) 225 (4.17%) 121 (4.50%) 104 (3.83%)
Sex Female N (%) 2,687 (49.75%) 2,687 (100.00%) -
Male N (%) 2,714 (50.25%) - 2,714 (100.00%)
Prior observation - Median [Q25 - Q75] 15,837 [11,937 - 20,419] 15,632 [11,772 - 20,152] 16,107 [12,010 - 20,689]
Future observation - Median [Q25 - Q75] 5,193 [2,385 - 8,100] 5,235 [2,602 - 8,028] 5,101 [2,223 - 8,132]
Days in cohort - Median [Q25 - Q75] 15 [8 - 15] 15 [8 - 15] 15 [8 - 15]
Number visits prior year - Median [Q25 - Q75] 0.00 [0.00 - 0.00] 0.00 [0.00 - 0.00] 0.00 [0.00 - 0.00]
Conditions any time prior Cardiac arrest N (%) 210 (3.89%) 120 (4.47%) 90 (3.32%)
Myocardial infarction N (%) 17 (0.31%) 1 (0.04%) 16 (0.59%)
Cough N (%) 0 (0.00%) 0 (0.00%) 0 (0.00%)
Fever N (%) 0 (0.00%) 0 (0.00%) 0 (0.00%)
Headache N (%) 0 (0.00%) 0 (0.00%) 0 (0.00%)
Medications prior year Nitroglycerin N (%) 104 (1.93%) 21 (0.78%) 83 (3.06%)
Warfarin N (%) 8 (0.15%) 2 (0.07%) 6 (0.22%)
Verapamil N (%) 8 (0.15%) 2 (0.07%) 6 (0.22%)
Morphine N (%) 6 (0.11%) 4 (0.15%) 2 (0.07%)
Atorvastatin N (%) 0 (0.00%) 0 (0.00%) 0 (0.00%)

Your turn

Characterise the ‘simvastatin’ cohort with:

  • Demographics
  • Two age groups: 0-49; >=50
  • Number of records in condition_occurrence on the index date.
  • Using same drugs than for acetaminophen which ones where used in the prior year including index date
  • Using same conditions than for acetaminophen which ones where used any time prior including index date

Suppress counts under 10

Visualise it in a nice table

Can you stratify the result by age group? (PatientProfiles::addAge() allows you to add age_group column to any table).

Result

Variable name Variable level Estimate name
Age group
overall 0 to 49 50 or above
36567_simvastatin
Number records - N 182 61 121
Number subjects - N 182 61 121
Cohort start date - Median [Q25 - Q75] 2002-12-21 [1992-07-12 - 2011-12-20] 1995-06-24 [1992-03-06 - 2003-02-25] 2007-02-25 [1993-01-23 - 2014-03-16]
Cohort end date - Median [Q25 - Q75] 2002-12-21 [1992-07-12 - 2011-12-20] 1995-06-24 [1992-03-06 - 2003-02-25] 2007-02-25 [1993-01-23 - 2014-03-16]
Age - Median [Q25 - Q75] 57 [42 - 72] 37 [33 - 42] 68 [57 - 77]
Age group 0 to 49 N (%) 61 (33.52%) 61 (100.00%) -
50 or above N (%) 121 (66.48%) - 121 (100.00%)
Sex Female N (%) 47 (25.82%) 19 (31.15%) 28 (23.14%)
Male N (%) 135 (74.18%) 42 (68.85%) 93 (76.86%)
Prior observation - Median [Q25 - Q75] 21,136 [15,351 - 26,352] 13,594 [12,138 - 15,351] 24,955 [21,140 - 28,280]
Future observation - Median [Q25 - Q75] 3,238 [1,503 - 6,678] 6,391 [3,466 - 9,072] 2,275 [1,292 - 4,824]
Days in cohort - Median [Q25 - Q75] 1 [1 - 1] 1 [1 - 1] 1 [1 - 1]
Number conditions on index - Median [Q25 - Q75] 1.00 [0.00 - 1.00] 1.00 [0.00 - 1.00] 1.00 [1.00 - 1.00]
Conditions any time prior Myocardial infarction N (%) <10 <10 <10
Cardiac arrest N (%) <10 <10 <10
Cough N (%) 0 (0.00%) 0 (0.00%) 0 (0.00%)
Fever N (%) 0 (0.00%) 0 (0.00%) 0 (0.00%)
Headache N (%) 0 (0.00%) 0 (0.00%) 0 (0.00%)
Medications prior year Nitroglycerin N (%) 155 (85.16%) 44 (72.13%) 111 (91.74%)
Atorvastatin N (%) 0 (0.00%) 0 (0.00%) 0 (0.00%)
Morphine N (%) 0 (0.00%) 0 (0.00%) 0 (0.00%)
Verapamil N (%) 0 (0.00%) 0 (0.00%) 0 (0.00%)
Warfarin N (%) 0 (0.00%) 0 (0.00%) 0 (0.00%)

Solution

💡 Click to see solution 
result <- cdm$simvastatin_cohort |>
  PatientProfiles::addAge(ageGroup = list(c(0, 49), c(50, Inf))) |>
  summariseCharacteristics(
    strata = list("age_group"),
    demographics = TRUE,
    ageGroup = list(c(0, 49), c(50, Inf)),
    tableIntersectCount = list(
      "Number conditions on index" = list(tableName = "condition_occurrence", window = c(0, 0))
    ),
    cohortIntersectFlag = list(
      "Conditions any time prior" = list(targetCohortTable = "conditions", window = c(-Inf, 0)),
      "Medications prior year" = list(targetCohortTable = "medications", window = c(-365, 0))
    )
  )
result |>
  suppress(minCellCount = 10) |>
  dplyr::filter(!estimate_name %in% c("min", "max", "mean", "sd")) |>
  tableCharacteristics(
    header = "age_group", 
    groupColumn = "cohort_name",
    hide = c("table", "window", "value", "cdm_name")
  )

summariseLargeScaleCharacteristics

Large scale characterisation is a data driven approach to characterise data and find which are the most frequently concepts in the desired windows of observation.

result <- cdm$acetaminophen |>
  summariseLargeScaleCharacteristics(
    strata = list(),
    window = list(c(-365, -1), c(0, 0), c(1, 365)), 
    eventInWindow = c("condition_occurrence", "observation"), 
    episodeInWindow =  "drug_exposure", 
    minimumFrequency = 0
  )

tableTopLargeScaleCharacteristics

result |>
  filterSettings(table_name == "drug_exposure") |>
  tableTopLargeScaleCharacteristics(topConcepts = 10)
Top
Window
-365 to -1 0 to 0 1 to 365
1 tetanus and diphtheria toxoids, adsorbed, preservative free, for adult use (40213227)
513 (9.5%)		 Acetaminophen 325 MG Oral Tablet (1127433)
4249 (78.7%)		 Acetaminophen 325 MG Oral Tablet (1127433)
4296 (79.8%)
2 hepatitis A vaccine, adult dosage (40213296)
208 (3.9%)		 Acetaminophen 21.7 MG/ML / Dextromethorphan Hydrobromide 1 MG/ML / doxylamine succinate 0.417 MG/ML Oral Solution (40229134)
874 (16.2%)		 Acetaminophen 21.7 MG/ML / Dextromethorphan Hydrobromide 1 MG/ML / doxylamine succinate 0.417 MG/ML Oral Solution (40229134)
908 (16.9%)
3 Acetaminophen 325 MG Oral Tablet (1127433)
172 (3.2%)		 Acetaminophen 325 MG / Hydrocodone Bitartrate 7.5 MG Oral Tablet (40162522)
153 (2.8%)		 tetanus and diphtheria toxoids, adsorbed, preservative free, for adult use (40213227)
473 (8.8%)
4 zoster vaccine, live (40213260)
145 (2.7%)		 Acetaminophen 325 MG / Oxycodone Hydrochloride 5 MG Oral Tablet (40231925)
118 (2.2%)		 hepatitis A vaccine, adult dosage (40213296)
244 (4.5%)
5 Amoxicillin 250 MG / Clavulanate 125 MG Oral Tablet (1713671)
123 (2.3%)		 Nitroglycerin 0.4 MG/ACTUAT Mucosal Spray (19126352)
98 (1.8%)		 Ibuprofen 200 MG Oral Tablet (19078461)
174 (3.2%)
6 Ibuprofen 200 MG Oral Tablet (19078461)
109 (2.0%)		 Ibuprofen 200 MG Oral Tablet (19078461)
95 (1.8%)		 hepatitis B vaccine, adult dosage (40213306)
164 (3.0%)
7 Nitroglycerin 0.4 MG/ACTUAT Mucosal Spray (19126352)
104 (1.9%)		 clopidogrel 75 MG Oral Tablet (19075601)
86 (1.6%)		 Amoxicillin 250 MG / Clavulanate 125 MG Oral Tablet (1713671)
152 (2.8%)
8 clopidogrel 75 MG Oral Tablet (19075601)
101 (1.9%)		 Amoxicillin 500 MG Oral Tablet (19073188)
83 (1.5%)		 Naproxen sodium 220 MG Oral Tablet (1115171)
120 (2.2%)
9 Naproxen sodium 220 MG Oral Tablet (1115171)
90 (1.7%)		 Naproxen sodium 220 MG Oral Tablet (1115171)
74 (1.4%)		 zoster vaccine, live (40213260)
111 (2.1%)
10 hepatitis B vaccine, adult dosage (40213306)
75 (1.4%)		 Acetaminophen 750 MG / Hydrocodone Bitartrate 7.5 MG Oral Tablet (19133768)
61 (1.1%)		 Amoxicillin 500 MG Oral Tablet (19073188)
97 (1.8%)

tableLargeScaleCharacteristics

tableLargeScaleCharacteristics

tableLargeScaleCharacteristics(result = result, compareBy = "variable_level")

tableLargeScaleCharacteristics

tableLargeScaleCharacteristics(result = result, compareBy = "variable_level", smdReference = "-365 to -1")

plotLargeScaleCharacteristics

plotComparedLargeScaleCharacteristics

plotComparedLargeScaleCharacteristics(
  result = result, 
  colour = "variable_level", 
  facet = "table_name"
)

plotComparedLargeScaleCharacteristics

plotComparedLargeScaleCharacteristics(
  result = result, 
  colour = "variable_level", 
  facet = "table_name"
) |>
  ggplotly()

CohortCharacteristics

👉 Packages website
👉 CRAN link
👉 Manual

📧 marti.catalasabate@ndorms.ox.ac.uk