tests/testthat/test-summariseCodeUse.R
In oxford-pharmacoepi/CodelistGenerator: Identify Relevant Clinical Codes and Evaluate Their Use
test_that("summarise code use - eunomia", {
skip_on_cran()
con <- DBI::dbConnect(duckdb::duckdb(), dbdir = CDMConnector::eunomiaDir())
cdm <- CDMConnector::cdmFromCon(con, cdmSchema = "main", writeSchema = "main")
acetiminophen <- c(1125315, 1127433, 40229134,
40231925, 40162522, 19133768, 1127078)
poliovirus_vaccine <- c(40213160)
cs <- list(acetiminophen = acetiminophen,
poliovirus_vaccine = poliovirus_vaccine)
startNames <- CDMConnector::listSourceTables(cdm)
results <- summariseCodeUse(cs,
cdm = cdm,
byYear = TRUE,
bySex = TRUE,
ageGroup = list(c(0,17),
c(18,65),
c(66, 100)))
endNames <- CDMConnector::listSourceTables(cdm)
expect_true(length(setdiff(endNames, startNames)) == 0)
expect_no_error(results_no_by_concept <- summariseCodeUse(cs,
cdm = cdm,
byYear = TRUE,
bySex = TRUE,
byConcept = FALSE))
expect_true(all(results_no_by_concept |> dplyr::pull("group_level") |> unique() == c("acetiminophen","poliovirus_vaccine")))
expect_true(all(results_no_by_concept |>
dplyr::filter(group_level == "acetiminophen", strata_level == "overall") |>
dplyr::pull("estimate_value") == c("14205", "2679")))
# min cell counts:
expect_true(
all(
omopgenerics::suppress(results) |>
dplyr::filter(
variable_name == "overall",
strata_level == "1909",
group_level == "acetiminophen"
) |>
dplyr::pull("estimate_value") == "-"
))
# check is a summarised result
expect_true("summarised_result" %in% class(results))
expect_equal(omopgenerics::resultColumns(),
colnames(results))
# overall record count
expect_true(results |>
dplyr::filter(group_name == "codelist_name" &
strata_name == "overall" &
strata_level == "overall" &
group_level == "acetiminophen" &
estimate_name == "record_count",
variable_name == "overall") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$drug_exposure |>
dplyr::filter(drug_concept_id %in% acetiminophen) |>
dplyr::tally() |>
dplyr::pull("n"))
# overall person count
expect_true(results |>
dplyr::filter(group_name == "codelist_name" &
strata_name == "overall" &
strata_level == "overall" &
group_level == "acetiminophen" &
estimate_name == "person_count",
variable_name == "overall") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$drug_exposure |>
dplyr::filter(drug_concept_id %in% acetiminophen) |>
dplyr::select("person_id") |>
dplyr::distinct() |>
dplyr::tally() |>
dplyr::pull("n"))
# by year
# overall record count
expect_true(results |>
dplyr::filter(group_name == "codelist_name" &
strata_name == "year" &
strata_level == "2008" &
group_level == "acetiminophen" &
estimate_name == "record_count",
variable_name == "overall") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$drug_exposure |>
dplyr::filter(drug_concept_id %in% acetiminophen) |>
dplyr::filter(year(drug_exposure_start_date) == 2008) |>
dplyr::tally() |>
dplyr::pull("n"))
# overall person count
expect_true(results |>
dplyr::filter(group_name == "codelist_name" &
strata_name == "year" &
strata_level == "2008" &
group_level == "acetiminophen" &
estimate_name == "person_count",
variable_name == "overall") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$drug_exposure |>
dplyr::filter(drug_concept_id %in% acetiminophen) |>
dplyr::filter(year(drug_exposure_start_date) == 2008) |>
dplyr::select("person_id") |>
dplyr::distinct() |>
dplyr::tally() |>
dplyr::pull("n"))
# by age group and sex
# overall record count
expect_true(results |>
dplyr::filter(group_name == "codelist_name" &
strata_name == "sex" &
strata_level == "Male" &
group_level == "acetiminophen" &
estimate_name == "record_count",
variable_name == "overall") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$drug_exposure |>
dplyr::filter(drug_concept_id %in% acetiminophen) |>
PatientProfiles::addSex() |>
dplyr::filter(sex == "Male") |>
dplyr::tally() |>
dplyr::pull("n"))
expect_true(results |>
dplyr::filter(group_name == "codelist_name" &
strata_name == "age_group &&& sex" &
strata_level == "18 to 65 &&& Male" &
group_level == "acetiminophen" &
estimate_name == "record_count",
variable_name == "overall") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$drug_exposure |>
dplyr::filter(drug_concept_id %in% acetiminophen) |>
PatientProfiles::addAge(indexDate = "drug_exposure_start_date") |>
PatientProfiles::addSex() |>
dplyr::filter(sex == "Male" &
age >= "18" &
age <= "65") |>
dplyr::tally() |>
dplyr::pull("n"))
# overall person count
expect_true(results |>
dplyr::filter(group_name == "codelist_name" &
strata_name == "age_group &&& sex" &
strata_level == "18 to 65 &&& Male" &
group_level == "acetiminophen" &
estimate_name == "person_count",
variable_name == "overall") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$drug_exposure |>
dplyr::filter(drug_concept_id %in% acetiminophen) |>
PatientProfiles::addAge(indexDate = "drug_exposure_start_date") |>
PatientProfiles::addSex() |>
dplyr::filter(sex == "Male" &
age >= "18" &
age <= "65") |>
dplyr::select("person_id") |>
dplyr::distinct() |>
dplyr::tally() |>
dplyr::pull("n"))
# Check date range
results <- summariseCodeUse(cs,
cdm = cdm,
byYear = TRUE,
bySex = TRUE,
ageGroup = list(c(0,17),
c(18,65),
c(66, 100)),
dateRange = as.Date(c("2010-01-01","2015-01-01")))
expect_equal(results |>
omopgenerics::settings() |>
dplyr::select("date_range_start", "date_range_end") |>
as.character(),
c("2010-01-01","2015-01-01"))
# overall record count
expect_true(results |>
dplyr::filter(group_name == "codelist_name" &
strata_name == "overall" &
strata_level == "overall" &
group_level == "acetiminophen" &
estimate_name == "record_count",
variable_name == "overall") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$drug_exposure |>
dplyr::filter(drug_exposure_start_date >= as.Date("2010-01-01"),
drug_exposure_start_date <= as.Date("2015-01-01")) |>
dplyr::filter(drug_concept_id %in% acetiminophen) |>
dplyr::tally() |>
dplyr::pull("n"))
# overall person count
expect_true(results |>
dplyr::filter(group_name == "codelist_name" &
strata_name == "overall" &
strata_level == "overall" &
group_level == "acetiminophen" &
estimate_name == "person_count",
variable_name == "overall") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$drug_exposure |>
dplyr::filter(drug_concept_id %in% acetiminophen) |>
dplyr::filter(drug_exposure_start_date >= as.Date("2010-01-01"),
drug_exposure_start_date <= as.Date("2014-12-31")) |>
dplyr::select("person_id") |>
dplyr::distinct() |>
dplyr::tally() |>
dplyr::pull("n"))
# by year
expect_true(
results |>
dplyr::filter(strata_name == "year") |>
dplyr::pull("strata_level") |>
unique() |>
as.numeric() |> max() == 2014)
expect_true(
results |>
dplyr::filter(strata_name == "year") |>
dplyr::pull("strata_level") |>
unique() |>
as.numeric() |> min() == 2010)
results <- summariseCodeUse(list("acetiminophen" = acetiminophen),
cdm = cdm, countBy = "person",
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL)
expect_true(nrow(results |>
dplyr::filter(estimate_name == "person_count")) > 0)
expect_true(nrow(results |>
dplyr::filter(estimate_name == "record_count")) == 0)
results <- summariseCodeUse(list("acetiminophen" = acetiminophen),
cdm = cdm, countBy = "record",
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL)
expect_true(nrow(results |>
dplyr::filter(estimate_name == "person_count")) == 0)
expect_true(nrow(results |>
dplyr::filter(estimate_name == "record_count")) > 0)
# domains covered
# condition
expect_true(nrow(summariseCodeUse(list(cs= c(4112343)),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))>1)
# visit
expect_true(nrow(summariseCodeUse(list(cs= c(9201)),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))>1)
# drug
expect_true(nrow(summariseCodeUse(list(cs= c(40213160)),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))>1)
# measurement
expect_true(nrow(summariseCodeUse(list(cs= c(3006322)),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))>1)
# procedure and condition
expect_true(nrow(summariseCodeUse(list(cs= c(4107731,4112343)),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))>1)
# no records
expect_message(results <- summariseCodeUse(list(cs= c(999999)),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))
expect_true(nrow(results) == 0)
# expected errors
expect_error(summariseCodeUse("not a concept",
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))
expect_error(summariseCodeUse("123",
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))
expect_error(summariseCodeUse(list("123"), # not named
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))
expect_error(summariseCodeUse(list(a = 123),
cdm = "not a cdm",
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))
expect_error(summariseCodeUse(list(a = 123),
cdm = cdm,
byYear = "Maybe",
bySex = FALSE,
ageGroup = NULL))
expect_error(summariseCodeUse(list(a = 123),
cdm = cdm,
byYear = FALSE,
bySex = "Maybe",
ageGroup = NULL))
expect_error(summariseCodeUse(list(a = 123),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = 25))
expect_error(summariseCodeUse(list(a = 123),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = list(c(18,17))))
expect_error(summariseCodeUse(list(a = 123),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = list(c(0,17),
c(15,20))))
expect_error(summariseCodeUse(list(a = 123),
cdm = cdm,
dateRange = c("a","b")))
CDMConnector::cdmDisconnect(cdm)
})
test_that("summarise cohort code use - eunomia", {
skip_on_cran()
con <- DBI::dbConnect(duckdb::duckdb(), dbdir = CDMConnector::eunomiaDir())
cdm <- CDMConnector::cdmFromCon(cdmName = "cdm", con, cdmSchema = "main", writeSchema = "main")
pharyngitis <- c(4112343)
cdm <- CDMConnector::generateConceptCohortSet(cdm = cdm,
conceptSet = list(pharyngitis = pharyngitis),
name = "pharyngitis",
end = "observation_period_end_date",
overwrite = TRUE)
# any
results_all <- summariseCodeUse(list(cs = 4134304),
cdm = cdm)
results_cohort <- summariseCohortCodeUse(list(cs = 4134304),
cdm = cdm,
cohortTable = "pharyngitis",
timing = "any")
expect_no_error(summariseCohortCodeUse(list(cs = 4134304),
cdm = cdm,
cohortTable = "pharyngitis",
timing = "any",
byConcept = FALSE))
expect_true(inherits(results_cohort, "summarised_result"))
expect_true(all(c("result_id", "result_type", "package_name", "package_version", "timing") %in%
colnames(omopgenerics::settings(results_cohort))))
expect_true(results_cohort |>
dplyr::filter(variable_name == "overall" &
strata_name == "overall" &
strata_level == "overall" &
estimate_name == "person_count") |>
dplyr::pull("estimate_value") |>
as.numeric() <
results_all |>
dplyr::filter(variable_name == "overall" &
strata_name == "overall" &
strata_level == "overall" &
estimate_name == "person_count") |>
dplyr::pull("estimate_value") |>
as.numeric())
# at entry - everyone in the cohort should have the code
results_cohort <- summariseCohortCodeUse(list(pharyngitis = pharyngitis),
cdm = cdm,
cohortTable = "pharyngitis",
timing = "entry")
results_cohort |>
dplyr::filter(variable_name == "overall" &
strata_name == "overall" &
strata_level == "overall" &
estimate_name == "person_count") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
CDMConnector::cohortCount(cdm$pharyngitis) |>
dplyr::pull("number_subjects")
# 260139
# on index
index_260139 <- cdm$pharyngitis |>
dplyr::left_join(cdm$condition_occurrence,
by=c("subject_id"="person_id")) |>
dplyr::filter(condition_start_date == cohort_start_date) |>
dplyr::filter(condition_concept_id == 260139) |>
dplyr::select("subject_id") |>
dplyr::distinct() |>
dplyr::count() |>
dplyr::pull()
results_cohort_260139 <- summariseCohortCodeUse(list(cs = 260139),
cdm = cdm,
cohortTable = "pharyngitis",
timing = "entry")
expect_equal(results_cohort_260139 |>
dplyr::filter(variable_name == "overall" &
strata_name == "overall" &
strata_level == "overall" &
estimate_name == "person_count") |>
dplyr::pull("estimate_value") |>
as.numeric(), index_260139)
# 260139 or 19133873 or 1127433
# on index
index_260139_19133873_1127433 <- dplyr::union_all(
cdm$pharyngitis |>
dplyr::left_join(cdm$condition_occurrence,
by=c("subject_id"="person_id")) |>
dplyr::filter(condition_start_date == cohort_start_date) |>
dplyr::filter(condition_concept_id == 260139) |>
dplyr::select("subject_id"),
cdm$pharyngitis |>
dplyr::left_join(cdm$drug_exposure,
by=c("subject_id"="person_id")) |>
dplyr::filter(drug_exposure_start_date == cohort_start_date) |>
dplyr::filter(drug_concept_id %in% c(19133873,1127433)) |>
dplyr::select("subject_id")) |>
dplyr::count() |>
dplyr::pull()
results_cohort_260139_19133873_1127433<- summariseCohortCodeUse(list(cs = c(260139,19133873,1127433)),
cdm = cdm,
cohortTable = "pharyngitis",
timing = "entry")
expect_equal(results_cohort_260139_19133873_1127433 |>
dplyr::filter(variable_name == "overall" &
strata_name == "overall" &
strata_level == "overall" &
estimate_name == "record_count") |>
dplyr::pull("estimate_value") |>
as.numeric(),
index_260139_19133873_1127433)
expect_equal(results_cohort_260139_19133873_1127433 |>
dplyr::filter(stringr::str_detect(variable_name, "Acute bronchitis")) |>
dplyr::filter(strata_name == "overall" &
strata_level == "overall" &
estimate_name == "person_count") |>
dplyr::pull("estimate_value") |>
as.numeric(),
index_260139)
# multiple cohorts
cdm <- CDMConnector::generateConceptCohortSet(cdm = cdm,
conceptSet = list(a = 260139,
b = 1127433 ),
name = "cohorts",
end = "observation_period_end_date",
overwrite = TRUE)
results_cohort_mult <- summariseCohortCodeUse(list(cs = c(260139,19133873,1127433)),
cdm = cdm,
cohortTable = "cohorts",
timing = "entry")
expect_true(nrow(results_cohort_mult |>
dplyr::filter(stringr::str_detect(variable_name, "Acute bronchitis")) |>
dplyr::filter(strata_name == "overall" &
strata_level == "overall" &
estimate_name == "person_count")) == 2)
expect_equal(c("a", "b"), results_cohort_mult |>
dplyr::filter(stringr::str_detect(variable_name, "Acute bronchitis")) |>
dplyr::filter(strata_name == "overall" &
strata_level == "overall" &
estimate_name == "person_count") |>
visOmopResults::splitGroup() |>
dplyr::pull("cohort_name"))
# empty cohort - no results
cdm$pharyngitis <- cdm$pharyngitis |>
dplyr::filter(cohort_definition_id == 99)
expect_true(nrow(summariseCohortCodeUse(list(cs = 4134304),
cdm = cdm,
cohortTable = "pharyngitis",
timing = "any")) == 0)
# expected errors
expect_error(summariseCohortCodeUse(4134304,
cdm = cdm,
cohortTable = "not_a_cohort",
timing = "any"))
expect_error(summariseCohortCodeUse(list(4134304),
cdm = cdm,
cohortTable = "not_a_cohort",
timing = "any"))
expect_error(summariseCohortCodeUse(list(cs = 4134304),
cdm = cdm,
cohortTable = "not_a_cohort",
timing = "any"))
expect_error(summariseCohortCodeUse(list(cs = 4134304),
cdm = cdm,
cohortTable = "pharyngitis",
timing = "not_a_option"))
expect_error(summariseCohortCodeUse(list(cs = 4134304),
cdm = cdm,
cohortTable = "pharyngitis",
timing = c("any", "entry")))
CDMConnector::cdmDisconnect(cdm)
})
test_that("summarise code use - redshift", {
testthat::skip_if(Sys.getenv("CDM5_REDSHIFT_DBNAME") == "")
db <- DBI::dbConnect(RPostgres::Redshift(),
dbname = Sys.getenv("CDM5_REDSHIFT_DBNAME"),
host = Sys.getenv("CDM5_REDSHIFT_HOST"),
port = Sys.getenv("CDM5_REDSHIFT_PORT"),
user = Sys.getenv("CDM5_REDSHIFT_USER"),
password = Sys.getenv("CDM5_REDSHIFT_PASSWORD"))
cdm <- CDMConnector::cdmFromCon(cdmName = "cdm",
con = db,
cdmSchema = Sys.getenv("CDM5_REDSHIFT_CDM_SCHEMA"),
writeSchema = Sys.getenv("CDM5_REDSHIFT_SCRATCH_SCHEMA"),
cdmVersion = "5.3")
asthma <- list(asthma = c(317009, 257581))
results <- summariseCodeUse(asthma,
cdm = cdm,
byYear = TRUE,
bySex = TRUE,
ageGroup = list(c(0,17),
c(18,65),
c(66, 100)))
# column names
expect_true(inherits(results, "summarised_result"))
# overall record count
expect_true(results |>
dplyr::filter(variable_name == "overall" &
strata_name == "overall" &
strata_level == "overall",
estimate_name == "record_count") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$condition_occurrence |>
dplyr::filter(condition_concept_id %in% !!asthma[[1]]) |>
dplyr::tally() |>
dplyr::pull("n"))
# overall person count
expect_true(results |>
dplyr::filter(variable_name == "overall" &
strata_name == "overall" &
strata_level == "overall" &
estimate_name == "person_count") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$condition_occurrence |>
dplyr::filter(condition_concept_id %in% !!asthma[[1]]) |>
dplyr::select("person_id") |>
dplyr::distinct() |>
dplyr::tally() |>
dplyr::pull("n"))
# by year
# overall record count
expect_true(results |>
dplyr::filter(variable_name == "overall" &
strata_name == "year" &
strata_level == "2008",
estimate_name == "record_count") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$condition_occurrence |>
dplyr::filter(condition_concept_id %in% !!asthma[[1]]) |>
dplyr::filter(year(condition_start_date) == 2008) |>
dplyr::tally() |>
dplyr::pull("n"))
# overall person count
expect_true(results |>
dplyr::filter(variable_name == "overall" &
strata_name == "year" &
strata_level == "2008",
estimate_name == "person_count") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$condition_occurrence |>
dplyr::filter(condition_concept_id %in% !!asthma[[1]]) |>
dplyr::filter(year(condition_start_date) == 2008) |>
dplyr::select("person_id") |>
dplyr::distinct() |>
dplyr::tally() |>
dplyr::pull("n"))
# by age group and sex
# overall record count
expect_true(results |>
dplyr::filter(variable_name == "overall" &
strata_name == "sex" &
strata_level == "Male",
estimate_name == "record_count") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$condition_occurrence |>
dplyr::filter(condition_concept_id %in% !!asthma[[1]]) |>
PatientProfiles::addSex() |>
dplyr::filter(sex == "Male") |>
dplyr::tally() |>
dplyr::pull("n"))
expect_true(results |>
dplyr::filter(variable_name == "overall" &
strata_name == "age_group &&& sex" &
strata_level == "18 to 65 &&& Male",
estimate_name == "record_count") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$condition_occurrence |>
dplyr::filter(condition_concept_id %in% !!asthma[[1]]) |>
PatientProfiles::addAge(indexDate = "condition_start_date") |>
PatientProfiles::addSex() |>
dplyr::filter(sex == "Male" &
age >= "18" &
age <= "65") |>
dplyr::tally() |>
dplyr::pull("n"))
# overall person count
expect_true(results |>
dplyr::filter(variable_name == "overall" &
strata_name == "age_group &&& sex" &
strata_level == "18 to 65 &&& Male",
estimate_name == "person_count") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$condition_occurrence |>
dplyr::filter(condition_concept_id %in% !!asthma[[1]]) |>
PatientProfiles::addAge(indexDate = "condition_start_date") |>
PatientProfiles::addSex() |>
dplyr::filter(sex == "Male" &
age >= "18" &
age <= "65") |>
dplyr::select("person_id") |>
dplyr::distinct() |>
dplyr::tally() |>
dplyr::pull("n"))
results <- summariseCodeUse(asthma,
cdm = cdm, countBy = "person",
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL)
expect_true(nrow(results |>
dplyr::filter(estimate_name == "person_count")) > 0)
expect_true(nrow(results |>
dplyr::filter(estimate_name == "record_count")) == 0)
results <- summariseCodeUse(asthma,
cdm = cdm, countBy = "record",
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL)
expect_true(nrow(results |>
dplyr::filter(estimate_name == "person_count")) == 0)
expect_true(nrow(results |>
dplyr::filter(estimate_name == "record_count")) > 0)
# domains covered
# condition
expect_true(nrow(summariseCodeUse(list(cs = c(317009)),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))>1)
# visit
expect_true(nrow(summariseCodeUse(list(cs = 9201),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))>1)
# drug
expect_true(nrow(summariseCodeUse(list(cs = 19071493),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))>1)
# measurement
expect_true(nrow(summariseCodeUse(list(cs = 2212542),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))>1)
# procedure and condition
expect_true(nrow(summariseCodeUse(list(cs = c(4261206,317009)),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))>1)
# no records
expect_message(results <- summariseCodeUse(list(cs = c(999999)),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))
expect_true(nrow(results) == 0)
# expected errors
expect_error(summariseCodeUse(list(cs = "not a concept"),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))
expect_error(summariseCodeUse(list(cs = 123),
cdm = "not a cdm",
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))
expect_error(summariseCodeUse(list(cs = 123),
cdm = cdm,
byYear = "Maybe",
bySex = FALSE,
ageGroup = NULL))
expect_error(summariseCodeUse(list(cs = 123),
cdm = cdm,
byYear = FALSE,
bySex = "Maybe",
ageGroup = NULL))
expect_error(summariseCodeUse(list(cs = 123),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = 25))
expect_error(summariseCodeUse(list(cs = 123),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = list(c(18,17))))
expect_error(summariseCodeUse(list(cs = 123),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = list(c(0,17),
c(15,20))))
CDMConnector::cdmDisconnect(cdm)
})
test_that("summarise code use - eunomia source concept id NA", {
skip_on_cran()
con <- DBI::dbConnect(duckdb::duckdb(), dbdir = CDMConnector::eunomiaDir())
cdm <- CDMConnector::cdmFromCon(con, cdmSchema = "main", writeSchema = "main")
acetiminophen <- c(1125315, 1127433, 40229134,
40231925, 40162522, 19133768, 1127078)
cdm$drug_exposure <- cdm$drug_exposure |>
dplyr::mutate(drug_source_concept_id = NA_character_)
cs <- list(acetiminophen = acetiminophen)
results <- summariseCodeUse(cs,
cdm = cdm)
expect_true(all(omopgenerics::splitAdditional(results) |>
dplyr::filter(variable_name != "overall") |>
dplyr::pull("source_concept_name") == "NA"))
expect_true(all(omopgenerics::splitAdditional(results) |>
dplyr::filter(variable_name != "overall") |>
dplyr::pull("source_concept_id") == "NA"))
CDMConnector::cdmDisconnect(cdm)
})
test_that("summarise cohort code use - eunomia source concept id NA", {
skip_on_cran()
con <- DBI::dbConnect(duckdb::duckdb(), dbdir = CDMConnector::eunomiaDir())
cdm <- CDMConnector::cdmFromCon(con, cdmSchema = "main", writeSchema = "main")
pharyngitis <- c(4112343)
cdm$condition_occurrence <- cdm$condition_occurrence |>
dplyr::mutate(condition_source_concept_id = NA_character_)
cdm <- CDMConnector::generateConceptCohortSet(cdm = cdm,
conceptSet = list(pharyngitis = pharyngitis),
name = "pharyngitis",
end = "observation_period_end_date",
overwrite = TRUE)
results_cohort <- summariseCohortCodeUse(list(cs = 4134304),
cdm = cdm,
cohortTable = "pharyngitis",
timing = "any")
expect_true(all(omopgenerics::splitAdditional(results_cohort) |>
dplyr::filter(variable_name != "overall") |>
dplyr::pull("source_concept_name") == "NA"))
expect_true(all(omopgenerics::splitAdditional(results_cohort) |>
dplyr::filter(variable_name != "overall") |>
dplyr::pull("source_concept_id") == "NA"))
CDMConnector::cdmDisconnect(cdm)
})
test_that("empty cohort", {
skip_on_cran()
con <- DBI::dbConnect(duckdb::duckdb(),
dbdir = CDMConnector::eunomiaDir())
cdm <- CDMConnector::cdmFromCon(con,
cdmSchema = "main",
writeSchema = "main",
cdmName = "test")
cdm <- CDMConnector::generateConceptCohortSet(cdm = cdm,
conceptSet = list(a = 260139,
b = 1127433),
name = "cohorts",
end = "observation_period_end_date",
overwrite = TRUE)
results_cohort_mult <- summariseCohortCodeUse(list(cs = as.numeric()),
cdm = cdm,
cohortTable = "cohorts",
timing = "entry")
expect_true(inherits(results_cohort_mult, "summarised_result"))
expect_true(nrow(results_cohort_mult) == 0)
CDMConnector::cdmDisconnect(cdm)
})
oxford-pharmacoepi/CodelistGenerator documentation built on June 9, 2025, 9:38 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
test_that("summarise code use - eunomia", {
skip_on_cran()
con <- DBI::dbConnect(duckdb::duckdb(), dbdir = CDMConnector::eunomiaDir())
cdm <- CDMConnector::cdmFromCon(con, cdmSchema = "main", writeSchema = "main")
acetiminophen <- c(1125315, 1127433, 40229134,
40231925, 40162522, 19133768, 1127078)
poliovirus_vaccine <- c(40213160)
cs <- list(acetiminophen = acetiminophen,
poliovirus_vaccine = poliovirus_vaccine)
startNames <- CDMConnector::listSourceTables(cdm)
results <- summariseCodeUse(cs,
cdm = cdm,
byYear = TRUE,
bySex = TRUE,
ageGroup = list(c(0,17),
c(18,65),
c(66, 100)))
endNames <- CDMConnector::listSourceTables(cdm)
expect_true(length(setdiff(endNames, startNames)) == 0)
expect_no_error(results_no_by_concept <- summariseCodeUse(cs,
cdm = cdm,
byYear = TRUE,
bySex = TRUE,
byConcept = FALSE))
expect_true(all(results_no_by_concept |> dplyr::pull("group_level") |> unique() == c("acetiminophen","poliovirus_vaccine")))
expect_true(all(results_no_by_concept |>
dplyr::filter(group_level == "acetiminophen", strata_level == "overall") |>
dplyr::pull("estimate_value") == c("14205", "2679")))
# min cell counts:
expect_true(
all(
omopgenerics::suppress(results) |>
dplyr::filter(
variable_name == "overall",
strata_level == "1909",
group_level == "acetiminophen"
) |>
dplyr::pull("estimate_value") == "-"
))
# check is a summarised result
expect_true("summarised_result" %in% class(results))
expect_equal(omopgenerics::resultColumns(),
colnames(results))
# overall record count
expect_true(results |>
dplyr::filter(group_name == "codelist_name" &
strata_name == "overall" &
strata_level == "overall" &
group_level == "acetiminophen" &
estimate_name == "record_count",
variable_name == "overall") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$drug_exposure |>
dplyr::filter(drug_concept_id %in% acetiminophen) |>
dplyr::tally() |>
dplyr::pull("n"))
# overall person count
expect_true(results |>
dplyr::filter(group_name == "codelist_name" &
strata_name == "overall" &
strata_level == "overall" &
group_level == "acetiminophen" &
estimate_name == "person_count",
variable_name == "overall") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$drug_exposure |>
dplyr::filter(drug_concept_id %in% acetiminophen) |>
dplyr::select("person_id") |>
dplyr::distinct() |>
dplyr::tally() |>
dplyr::pull("n"))
# by year
# overall record count
expect_true(results |>
dplyr::filter(group_name == "codelist_name" &
strata_name == "year" &
strata_level == "2008" &
group_level == "acetiminophen" &
estimate_name == "record_count",
variable_name == "overall") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$drug_exposure |>
dplyr::filter(drug_concept_id %in% acetiminophen) |>
dplyr::filter(year(drug_exposure_start_date) == 2008) |>
dplyr::tally() |>
dplyr::pull("n"))
# overall person count
expect_true(results |>
dplyr::filter(group_name == "codelist_name" &
strata_name == "year" &
strata_level == "2008" &
group_level == "acetiminophen" &
estimate_name == "person_count",
variable_name == "overall") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$drug_exposure |>
dplyr::filter(drug_concept_id %in% acetiminophen) |>
dplyr::filter(year(drug_exposure_start_date) == 2008) |>
dplyr::select("person_id") |>
dplyr::distinct() |>
dplyr::tally() |>
dplyr::pull("n"))
# by age group and sex
# overall record count
expect_true(results |>
dplyr::filter(group_name == "codelist_name" &
strata_name == "sex" &
strata_level == "Male" &
group_level == "acetiminophen" &
estimate_name == "record_count",
variable_name == "overall") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$drug_exposure |>
dplyr::filter(drug_concept_id %in% acetiminophen) |>
PatientProfiles::addSex() |>
dplyr::filter(sex == "Male") |>
dplyr::tally() |>
dplyr::pull("n"))
expect_true(results |>
dplyr::filter(group_name == "codelist_name" &
strata_name == "age_group &&& sex" &
strata_level == "18 to 65 &&& Male" &
group_level == "acetiminophen" &
estimate_name == "record_count",
variable_name == "overall") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$drug_exposure |>
dplyr::filter(drug_concept_id %in% acetiminophen) |>
PatientProfiles::addAge(indexDate = "drug_exposure_start_date") |>
PatientProfiles::addSex() |>
dplyr::filter(sex == "Male" &
age >= "18" &
age <= "65") |>
dplyr::tally() |>
dplyr::pull("n"))
# overall person count
expect_true(results |>
dplyr::filter(group_name == "codelist_name" &
strata_name == "age_group &&& sex" &
strata_level == "18 to 65 &&& Male" &
group_level == "acetiminophen" &
estimate_name == "person_count",
variable_name == "overall") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$drug_exposure |>
dplyr::filter(drug_concept_id %in% acetiminophen) |>
PatientProfiles::addAge(indexDate = "drug_exposure_start_date") |>
PatientProfiles::addSex() |>
dplyr::filter(sex == "Male" &
age >= "18" &
age <= "65") |>
dplyr::select("person_id") |>
dplyr::distinct() |>
dplyr::tally() |>
dplyr::pull("n"))
# Check date range
results <- summariseCodeUse(cs,
cdm = cdm,
byYear = TRUE,
bySex = TRUE,
ageGroup = list(c(0,17),
c(18,65),
c(66, 100)),
dateRange = as.Date(c("2010-01-01","2015-01-01")))
expect_equal(results |>
omopgenerics::settings() |>
dplyr::select("date_range_start", "date_range_end") |>
as.character(),
c("2010-01-01","2015-01-01"))
# overall record count
expect_true(results |>
dplyr::filter(group_name == "codelist_name" &
strata_name == "overall" &
strata_level == "overall" &
group_level == "acetiminophen" &
estimate_name == "record_count",
variable_name == "overall") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$drug_exposure |>
dplyr::filter(drug_exposure_start_date >= as.Date("2010-01-01"),
drug_exposure_start_date <= as.Date("2015-01-01")) |>
dplyr::filter(drug_concept_id %in% acetiminophen) |>
dplyr::tally() |>
dplyr::pull("n"))
# overall person count
expect_true(results |>
dplyr::filter(group_name == "codelist_name" &
strata_name == "overall" &
strata_level == "overall" &
group_level == "acetiminophen" &
estimate_name == "person_count",
variable_name == "overall") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$drug_exposure |>
dplyr::filter(drug_concept_id %in% acetiminophen) |>
dplyr::filter(drug_exposure_start_date >= as.Date("2010-01-01"),
drug_exposure_start_date <= as.Date("2014-12-31")) |>
dplyr::select("person_id") |>
dplyr::distinct() |>
dplyr::tally() |>
dplyr::pull("n"))
# by year
expect_true(
results |>
dplyr::filter(strata_name == "year") |>
dplyr::pull("strata_level") |>
unique() |>
as.numeric() |> max() == 2014)
expect_true(
results |>
dplyr::filter(strata_name == "year") |>
dplyr::pull("strata_level") |>
unique() |>
as.numeric() |> min() == 2010)
results <- summariseCodeUse(list("acetiminophen" = acetiminophen),
cdm = cdm, countBy = "person",
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL)
expect_true(nrow(results |>
dplyr::filter(estimate_name == "person_count")) > 0)
expect_true(nrow(results |>
dplyr::filter(estimate_name == "record_count")) == 0)
results <- summariseCodeUse(list("acetiminophen" = acetiminophen),
cdm = cdm, countBy = "record",
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL)
expect_true(nrow(results |>
dplyr::filter(estimate_name == "person_count")) == 0)
expect_true(nrow(results |>
dplyr::filter(estimate_name == "record_count")) > 0)
# domains covered
# condition
expect_true(nrow(summariseCodeUse(list(cs= c(4112343)),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))>1)
# visit
expect_true(nrow(summariseCodeUse(list(cs= c(9201)),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))>1)
# drug
expect_true(nrow(summariseCodeUse(list(cs= c(40213160)),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))>1)
# measurement
expect_true(nrow(summariseCodeUse(list(cs= c(3006322)),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))>1)
# procedure and condition
expect_true(nrow(summariseCodeUse(list(cs= c(4107731,4112343)),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))>1)
# no records
expect_message(results <- summariseCodeUse(list(cs= c(999999)),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))
expect_true(nrow(results) == 0)
# expected errors
expect_error(summariseCodeUse("not a concept",
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))
expect_error(summariseCodeUse("123",
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))
expect_error(summariseCodeUse(list("123"), # not named
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))
expect_error(summariseCodeUse(list(a = 123),
cdm = "not a cdm",
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))
expect_error(summariseCodeUse(list(a = 123),
cdm = cdm,
byYear = "Maybe",
bySex = FALSE,
ageGroup = NULL))
expect_error(summariseCodeUse(list(a = 123),
cdm = cdm,
byYear = FALSE,
bySex = "Maybe",
ageGroup = NULL))
expect_error(summariseCodeUse(list(a = 123),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = 25))
expect_error(summariseCodeUse(list(a = 123),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = list(c(18,17))))
expect_error(summariseCodeUse(list(a = 123),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = list(c(0,17),
c(15,20))))
expect_error(summariseCodeUse(list(a = 123),
cdm = cdm,
dateRange = c("a","b")))
CDMConnector::cdmDisconnect(cdm)
})
test_that("summarise cohort code use - eunomia", {
skip_on_cran()
con <- DBI::dbConnect(duckdb::duckdb(), dbdir = CDMConnector::eunomiaDir())
cdm <- CDMConnector::cdmFromCon(cdmName = "cdm", con, cdmSchema = "main", writeSchema = "main")
pharyngitis <- c(4112343)
cdm <- CDMConnector::generateConceptCohortSet(cdm = cdm,
conceptSet = list(pharyngitis = pharyngitis),
name = "pharyngitis",
end = "observation_period_end_date",
overwrite = TRUE)
# any
results_all <- summariseCodeUse(list(cs = 4134304),
cdm = cdm)
results_cohort <- summariseCohortCodeUse(list(cs = 4134304),
cdm = cdm,
cohortTable = "pharyngitis",
timing = "any")
expect_no_error(summariseCohortCodeUse(list(cs = 4134304),
cdm = cdm,
cohortTable = "pharyngitis",
timing = "any",
byConcept = FALSE))
expect_true(inherits(results_cohort, "summarised_result"))
expect_true(all(c("result_id", "result_type", "package_name", "package_version", "timing") %in%
colnames(omopgenerics::settings(results_cohort))))
expect_true(results_cohort |>
dplyr::filter(variable_name == "overall" &
strata_name == "overall" &
strata_level == "overall" &
estimate_name == "person_count") |>
dplyr::pull("estimate_value") |>
as.numeric() <
results_all |>
dplyr::filter(variable_name == "overall" &
strata_name == "overall" &
strata_level == "overall" &
estimate_name == "person_count") |>
dplyr::pull("estimate_value") |>
as.numeric())
# at entry - everyone in the cohort should have the code
results_cohort <- summariseCohortCodeUse(list(pharyngitis = pharyngitis),
cdm = cdm,
cohortTable = "pharyngitis",
timing = "entry")
results_cohort |>
dplyr::filter(variable_name == "overall" &
strata_name == "overall" &
strata_level == "overall" &
estimate_name == "person_count") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
CDMConnector::cohortCount(cdm$pharyngitis) |>
dplyr::pull("number_subjects")
# 260139
# on index
index_260139 <- cdm$pharyngitis |>
dplyr::left_join(cdm$condition_occurrence,
by=c("subject_id"="person_id")) |>
dplyr::filter(condition_start_date == cohort_start_date) |>
dplyr::filter(condition_concept_id == 260139) |>
dplyr::select("subject_id") |>
dplyr::distinct() |>
dplyr::count() |>
dplyr::pull()
results_cohort_260139 <- summariseCohortCodeUse(list(cs = 260139),
cdm = cdm,
cohortTable = "pharyngitis",
timing = "entry")
expect_equal(results_cohort_260139 |>
dplyr::filter(variable_name == "overall" &
strata_name == "overall" &
strata_level == "overall" &
estimate_name == "person_count") |>
dplyr::pull("estimate_value") |>
as.numeric(), index_260139)
# 260139 or 19133873 or 1127433
# on index
index_260139_19133873_1127433 <- dplyr::union_all(
cdm$pharyngitis |>
dplyr::left_join(cdm$condition_occurrence,
by=c("subject_id"="person_id")) |>
dplyr::filter(condition_start_date == cohort_start_date) |>
dplyr::filter(condition_concept_id == 260139) |>
dplyr::select("subject_id"),
cdm$pharyngitis |>
dplyr::left_join(cdm$drug_exposure,
by=c("subject_id"="person_id")) |>
dplyr::filter(drug_exposure_start_date == cohort_start_date) |>
dplyr::filter(drug_concept_id %in% c(19133873,1127433)) |>
dplyr::select("subject_id")) |>
dplyr::count() |>
dplyr::pull()
results_cohort_260139_19133873_1127433<- summariseCohortCodeUse(list(cs = c(260139,19133873,1127433)),
cdm = cdm,
cohortTable = "pharyngitis",
timing = "entry")
expect_equal(results_cohort_260139_19133873_1127433 |>
dplyr::filter(variable_name == "overall" &
strata_name == "overall" &
strata_level == "overall" &
estimate_name == "record_count") |>
dplyr::pull("estimate_value") |>
as.numeric(),
index_260139_19133873_1127433)
expect_equal(results_cohort_260139_19133873_1127433 |>
dplyr::filter(stringr::str_detect(variable_name, "Acute bronchitis")) |>
dplyr::filter(strata_name == "overall" &
strata_level == "overall" &
estimate_name == "person_count") |>
dplyr::pull("estimate_value") |>
as.numeric(),
index_260139)
# multiple cohorts
cdm <- CDMConnector::generateConceptCohortSet(cdm = cdm,
conceptSet = list(a = 260139,
b = 1127433 ),
name = "cohorts",
end = "observation_period_end_date",
overwrite = TRUE)
results_cohort_mult <- summariseCohortCodeUse(list(cs = c(260139,19133873,1127433)),
cdm = cdm,
cohortTable = "cohorts",
timing = "entry")
expect_true(nrow(results_cohort_mult |>
dplyr::filter(stringr::str_detect(variable_name, "Acute bronchitis")) |>
dplyr::filter(strata_name == "overall" &
strata_level == "overall" &
estimate_name == "person_count")) == 2)
expect_equal(c("a", "b"), results_cohort_mult |>
dplyr::filter(stringr::str_detect(variable_name, "Acute bronchitis")) |>
dplyr::filter(strata_name == "overall" &
strata_level == "overall" &
estimate_name == "person_count") |>
visOmopResults::splitGroup() |>
dplyr::pull("cohort_name"))
# empty cohort - no results
cdm$pharyngitis <- cdm$pharyngitis |>
dplyr::filter(cohort_definition_id == 99)
expect_true(nrow(summariseCohortCodeUse(list(cs = 4134304),
cdm = cdm,
cohortTable = "pharyngitis",
timing = "any")) == 0)
# expected errors
expect_error(summariseCohortCodeUse(4134304,
cdm = cdm,
cohortTable = "not_a_cohort",
timing = "any"))
expect_error(summariseCohortCodeUse(list(4134304),
cdm = cdm,
cohortTable = "not_a_cohort",
timing = "any"))
expect_error(summariseCohortCodeUse(list(cs = 4134304),
cdm = cdm,
cohortTable = "not_a_cohort",
timing = "any"))
expect_error(summariseCohortCodeUse(list(cs = 4134304),
cdm = cdm,
cohortTable = "pharyngitis",
timing = "not_a_option"))
expect_error(summariseCohortCodeUse(list(cs = 4134304),
cdm = cdm,
cohortTable = "pharyngitis",
timing = c("any", "entry")))
CDMConnector::cdmDisconnect(cdm)
})
test_that("summarise code use - redshift", {
testthat::skip_if(Sys.getenv("CDM5_REDSHIFT_DBNAME") == "")
db <- DBI::dbConnect(RPostgres::Redshift(),
dbname = Sys.getenv("CDM5_REDSHIFT_DBNAME"),
host = Sys.getenv("CDM5_REDSHIFT_HOST"),
port = Sys.getenv("CDM5_REDSHIFT_PORT"),
user = Sys.getenv("CDM5_REDSHIFT_USER"),
password = Sys.getenv("CDM5_REDSHIFT_PASSWORD"))
cdm <- CDMConnector::cdmFromCon(cdmName = "cdm",
con = db,
cdmSchema = Sys.getenv("CDM5_REDSHIFT_CDM_SCHEMA"),
writeSchema = Sys.getenv("CDM5_REDSHIFT_SCRATCH_SCHEMA"),
cdmVersion = "5.3")
asthma <- list(asthma = c(317009, 257581))
results <- summariseCodeUse(asthma,
cdm = cdm,
byYear = TRUE,
bySex = TRUE,
ageGroup = list(c(0,17),
c(18,65),
c(66, 100)))
# column names
expect_true(inherits(results, "summarised_result"))
# overall record count
expect_true(results |>
dplyr::filter(variable_name == "overall" &
strata_name == "overall" &
strata_level == "overall",
estimate_name == "record_count") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$condition_occurrence |>
dplyr::filter(condition_concept_id %in% !!asthma[[1]]) |>
dplyr::tally() |>
dplyr::pull("n"))
# overall person count
expect_true(results |>
dplyr::filter(variable_name == "overall" &
strata_name == "overall" &
strata_level == "overall" &
estimate_name == "person_count") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$condition_occurrence |>
dplyr::filter(condition_concept_id %in% !!asthma[[1]]) |>
dplyr::select("person_id") |>
dplyr::distinct() |>
dplyr::tally() |>
dplyr::pull("n"))
# by year
# overall record count
expect_true(results |>
dplyr::filter(variable_name == "overall" &
strata_name == "year" &
strata_level == "2008",
estimate_name == "record_count") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$condition_occurrence |>
dplyr::filter(condition_concept_id %in% !!asthma[[1]]) |>
dplyr::filter(year(condition_start_date) == 2008) |>
dplyr::tally() |>
dplyr::pull("n"))
# overall person count
expect_true(results |>
dplyr::filter(variable_name == "overall" &
strata_name == "year" &
strata_level == "2008",
estimate_name == "person_count") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$condition_occurrence |>
dplyr::filter(condition_concept_id %in% !!asthma[[1]]) |>
dplyr::filter(year(condition_start_date) == 2008) |>
dplyr::select("person_id") |>
dplyr::distinct() |>
dplyr::tally() |>
dplyr::pull("n"))
# by age group and sex
# overall record count
expect_true(results |>
dplyr::filter(variable_name == "overall" &
strata_name == "sex" &
strata_level == "Male",
estimate_name == "record_count") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$condition_occurrence |>
dplyr::filter(condition_concept_id %in% !!asthma[[1]]) |>
PatientProfiles::addSex() |>
dplyr::filter(sex == "Male") |>
dplyr::tally() |>
dplyr::pull("n"))
expect_true(results |>
dplyr::filter(variable_name == "overall" &
strata_name == "age_group &&& sex" &
strata_level == "18 to 65 &&& Male",
estimate_name == "record_count") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$condition_occurrence |>
dplyr::filter(condition_concept_id %in% !!asthma[[1]]) |>
PatientProfiles::addAge(indexDate = "condition_start_date") |>
PatientProfiles::addSex() |>
dplyr::filter(sex == "Male" &
age >= "18" &
age <= "65") |>
dplyr::tally() |>
dplyr::pull("n"))
# overall person count
expect_true(results |>
dplyr::filter(variable_name == "overall" &
strata_name == "age_group &&& sex" &
strata_level == "18 to 65 &&& Male",
estimate_name == "person_count") |>
dplyr::pull("estimate_value") |>
as.numeric() ==
cdm$condition_occurrence |>
dplyr::filter(condition_concept_id %in% !!asthma[[1]]) |>
PatientProfiles::addAge(indexDate = "condition_start_date") |>
PatientProfiles::addSex() |>
dplyr::filter(sex == "Male" &
age >= "18" &
age <= "65") |>
dplyr::select("person_id") |>
dplyr::distinct() |>
dplyr::tally() |>
dplyr::pull("n"))
results <- summariseCodeUse(asthma,
cdm = cdm, countBy = "person",
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL)
expect_true(nrow(results |>
dplyr::filter(estimate_name == "person_count")) > 0)
expect_true(nrow(results |>
dplyr::filter(estimate_name == "record_count")) == 0)
results <- summariseCodeUse(asthma,
cdm = cdm, countBy = "record",
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL)
expect_true(nrow(results |>
dplyr::filter(estimate_name == "person_count")) == 0)
expect_true(nrow(results |>
dplyr::filter(estimate_name == "record_count")) > 0)
# domains covered
# condition
expect_true(nrow(summariseCodeUse(list(cs = c(317009)),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))>1)
# visit
expect_true(nrow(summariseCodeUse(list(cs = 9201),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))>1)
# drug
expect_true(nrow(summariseCodeUse(list(cs = 19071493),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))>1)
# measurement
expect_true(nrow(summariseCodeUse(list(cs = 2212542),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))>1)
# procedure and condition
expect_true(nrow(summariseCodeUse(list(cs = c(4261206,317009)),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))>1)
# no records
expect_message(results <- summariseCodeUse(list(cs = c(999999)),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))
expect_true(nrow(results) == 0)
# expected errors
expect_error(summariseCodeUse(list(cs = "not a concept"),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))
expect_error(summariseCodeUse(list(cs = 123),
cdm = "not a cdm",
byYear = FALSE,
bySex = FALSE,
ageGroup = NULL))
expect_error(summariseCodeUse(list(cs = 123),
cdm = cdm,
byYear = "Maybe",
bySex = FALSE,
ageGroup = NULL))
expect_error(summariseCodeUse(list(cs = 123),
cdm = cdm,
byYear = FALSE,
bySex = "Maybe",
ageGroup = NULL))
expect_error(summariseCodeUse(list(cs = 123),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = 25))
expect_error(summariseCodeUse(list(cs = 123),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = list(c(18,17))))
expect_error(summariseCodeUse(list(cs = 123),
cdm = cdm,
byYear = FALSE,
bySex = FALSE,
ageGroup = list(c(0,17),
c(15,20))))
CDMConnector::cdmDisconnect(cdm)
})
test_that("summarise code use - eunomia source concept id NA", {
skip_on_cran()
con <- DBI::dbConnect(duckdb::duckdb(), dbdir = CDMConnector::eunomiaDir())
cdm <- CDMConnector::cdmFromCon(con, cdmSchema = "main", writeSchema = "main")
acetiminophen <- c(1125315, 1127433, 40229134,
40231925, 40162522, 19133768, 1127078)
cdm$drug_exposure <- cdm$drug_exposure |>
dplyr::mutate(drug_source_concept_id = NA_character_)
cs <- list(acetiminophen = acetiminophen)
results <- summariseCodeUse(cs,
cdm = cdm)
expect_true(all(omopgenerics::splitAdditional(results) |>
dplyr::filter(variable_name != "overall") |>
dplyr::pull("source_concept_name") == "NA"))
expect_true(all(omopgenerics::splitAdditional(results) |>
dplyr::filter(variable_name != "overall") |>
dplyr::pull("source_concept_id") == "NA"))
CDMConnector::cdmDisconnect(cdm)
})
test_that("summarise cohort code use - eunomia source concept id NA", {
skip_on_cran()
con <- DBI::dbConnect(duckdb::duckdb(), dbdir = CDMConnector::eunomiaDir())
cdm <- CDMConnector::cdmFromCon(con, cdmSchema = "main", writeSchema = "main")
pharyngitis <- c(4112343)
cdm$condition_occurrence <- cdm$condition_occurrence |>
dplyr::mutate(condition_source_concept_id = NA_character_)
cdm <- CDMConnector::generateConceptCohortSet(cdm = cdm,
conceptSet = list(pharyngitis = pharyngitis),
name = "pharyngitis",
end = "observation_period_end_date",
overwrite = TRUE)
results_cohort <- summariseCohortCodeUse(list(cs = 4134304),
cdm = cdm,
cohortTable = "pharyngitis",
timing = "any")
expect_true(all(omopgenerics::splitAdditional(results_cohort) |>
dplyr::filter(variable_name != "overall") |>
dplyr::pull("source_concept_name") == "NA"))
expect_true(all(omopgenerics::splitAdditional(results_cohort) |>
dplyr::filter(variable_name != "overall") |>
dplyr::pull("source_concept_id") == "NA"))
CDMConnector::cdmDisconnect(cdm)
})
test_that("empty cohort", {
skip_on_cran()
con <- DBI::dbConnect(duckdb::duckdb(),
dbdir = CDMConnector::eunomiaDir())
cdm <- CDMConnector::cdmFromCon(con,
cdmSchema = "main",
writeSchema = "main",
cdmName = "test")
cdm <- CDMConnector::generateConceptCohortSet(cdm = cdm,
conceptSet = list(a = 260139,
b = 1127433),
name = "cohorts",
end = "observation_period_end_date",
overwrite = TRUE)
results_cohort_mult <- summariseCohortCodeUse(list(cs = as.numeric()),
cdm = cdm,
cohortTable = "cohorts",
timing = "entry")
expect_true(inherits(results_cohort_mult, "summarised_result"))
expect_true(nrow(results_cohort_mult) == 0)
CDMConnector::cdmDisconnect(cdm)
})
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.