measurementCohort() creates cohorts based on patient records from the
measurement or observation tables. It extends the function conceptCohort()
by allowing users to specify measurement values associated with those records.
This function supports both concept-based and value-based filtering:
Either
valueAsConceptorvalueAsNumbermust be provided.If one of them is specified (not NULL), only records that satisfy the other filter will be included.
If both are provided, records that meet either filter will be included.
Usage
measurementCohort(
cdm,
conceptSet,
name,
valueAsConcept = NULL,
valueAsNumber = NULL,
table = NULL,
useRecordsBeforeObservation = FALSE
)Arguments
- cdm
A cdm reference.
- conceptSet
A conceptSet, which can either be a codelist or a conceptSetExpression.
- name
Name of the new cohort table created in the cdm object.
- valueAsConcept
A named list defining cohorts based on measurement values as concept IDs. Each element name defines the name of cohort to create, and its value is a vector of concept IDs used to filter measurements by
value_as_concept_id.If NULL, all records will be included regardless ofvalue_as_concept_id.For instance, to create two bmi cohorts from a bmi
conceptSetwe can do the following:valueAsConcept = list(high_bmi = c(4328749, 35819253), low_bmi = c(4267416, 45881666))See more examples in the function examples.
- valueAsNumber
A named list defining cohorts based on numeric measurement ranges. Each list element should contain one or more numeric vectors of length two, specifying the allowed range(s) for the measurement value. If the numeric vector is named, the name should correspond to the
unit_concept_idthat will be used for that range.For example, the following creates a cohort named "low_weight" based on measurements recorded in kilograms (unit_concept_id = 9529) and stones (unit_concept_id = 21498905):
valueAsNumber = list("low_weight" = list("9529" = c(30, 40), "21498905" = c(4.7, 6.3)))See the examples below for how to define multiple cohorts based on different measurement filters.
- table
Character vector specifying which OMOP tables to use. Accepts "measurement", "observation", or both.
- useRecordsBeforeObservation
Logical. If FALSE, only records within the observation period are used. If TRUE, records prior to the observation period are included, with their cohort start date set to the start of the next observation period.
Examples
# \donttest{
library(CohortConstructor)
library(omock)
library(dplyr)
cdm <- mockVocabularyTables(concept = tibble(
concept_id = c(4326744, 4298393, 45770407, 8876, 4124457),
concept_name = c("Blood pressure", "Systemic blood pressure",
"Baseline blood pressure", "millimeter mercury column",
"Normal range"),
domain_id = "Measurement",
vocabulary_id = c("SNOMED", "SNOMED", "SNOMED", "UCUM", "SNOMED"),
standard_concept = "S",
concept_class_id = c("Observable Entity", "Observable Entity",
"Observable Entity", "Unit", "Qualifier Value"),
concept_code = NA_character_,
valid_start_date = as.Date(NA_character_),
valid_end_date = as.Date(NA_character_),
invalid_reason = NA_character_
)) |>
mockCdmFromTables(tables = list(
measurement = tibble(
measurement_id = 1:4L,
person_id = c(1L, 1L, 2L, 3L),
measurement_concept_id = c(4326744L, 4298393L, 4298393L, 45770407L),
measurement_date = as.Date(c("2000-07-01", "2000-12-11", "2002-09-08", "2015-02-19")),
measurement_type_concept_id = 0L,
value_as_number = c(100, 125, NA, NA),
value_as_concept_id = c(0L, 0L, 0L, 4124457L),
unit_concept_id = c(8876L, 8876L, 0L, 0L)
)
))
#> Warning: ! 1 casted column in concept as do not match expected column type:
#> • `concept_id` from numeric to integer
# create one cohort of blood pressure measurements indicating normal levels
cdm$cohort <- measurementCohort(
cdm = cdm,
name = "cohort",
conceptSet = list("blood_pressure" = c(4326744, 4298393, 45770407)),
valueAsConcept = list("normal_blood_preassure" = c(4124457)),
valueAsNumber = list("normal_blood_preassure" = list("8876" = c(70, 120))),
useRecordsBeforeObservation = FALSE
)
#> Warning: ! `codelist` casted to integers.
#> ℹ Subsetting table measurement using 3 concepts with domain: measurement.
#> ℹ Combining tables.
#> ℹ Applying measurement requirements.
#> ℹ Getting records in observation.
#> ℹ Creating cohort attributes.
#> ✔ Cohort cohort created.
cdm$cohort
#> # A tibble: 2 × 4
#> cohort_definition_id subject_id cohort_start_date cohort_end_date
#> <int> <int> <date> <date>
#> 1 1 1 2000-07-01 2000-07-01
#> 2 1 3 2015-02-19 2015-02-19
# create two cohorts of blood preassure measurements, one with results
# indicating normal blood pressure and the other inidcating high
cdm$cohort2 <- measurementCohort(
cdm = cdm,
name = "cohort2",
conceptSet = list("blood_pressure" = c(4326744, 4298393, 45770407)),
valueAsConcept = list(
"normal_blood_pressure" = 4124457,
"high_blood_pressure" = 4328749
),
valueAsNumber = list(
"normal_blood_pressure" = list("8876" = c(70, 120)),
"high_blood_pressure" = list("8876" = c(121, 200))
),
useRecordsBeforeObservation = TRUE
)
#> Warning: ! `codelist` casted to integers.
#> ℹ Subsetting table measurement using 3 concepts with domain: measurement.
#> ℹ Combining tables.
#> ℹ Applying measurement requirements.
#> ℹ Getting records in observation.
#> Warning: There were 3 warnings in `dplyr::mutate()`.
#> The first warning was:
#> ℹ In argument: `trim_record = &...`.
#> ℹ In group 1: `cohort_definition_id = 1`, `subject_id = 1`, `cohort_start_date
#> = 2000-12-11`, `cohort_end_date = 2000-12-11`.
#> Caused by warning in `min()`:
#> ! no non-missing arguments to min; returning Inf
#> ℹ Run `dplyr::last_dplyr_warnings()` to see the 2 remaining warnings.
#> ℹ Creating cohort attributes.
#> ✔ Cohort cohort2 created.
cdm$cohort2 |> settings()
#> # A tibble: 2 × 4
#> cohort_definition_id cohort_name cdm_version vocabulary_version
#> <int> <chr> <chr> <chr>
#> 1 1 high_blood_pressure 5.3 mock
#> 2 2 normal_blood_pressure 5.3 mock
# }