DrugUtilisation
Conduct Drug Utilisation Studies using the OMOP Common Data Model
Introduction
Drug utilisation studies (DUS) were defined by the World Health Organization as studying the marketing, distribution, prescription, and use of medicinal products in a society, with special emphasis on the resulting medical and socioeconomic consequences (WHO, 2003).
This package aims to standardise and provide the tools to conduct Drug Utilisation studies as of the Darwin-EU Catalogue of Standard Analytics.
Functionalities
Create cohort of drug user
Compute daily dose for a given exposure.
Identify potential indications
Summarise drug use
Identify alternative treatments
Analyse treatment discontinuation (under development)
Create a reference to a cdm object
For this example we are going to use
library(DBI)
library(duckdb)
library(CDMConnector)
library(dplyr)
library(here)
# downloadEunomiaData(pathToData = here(), overwrite = TRUE)
# Sys.setenv("EUNOMIA_DATA_FOLDER" = here())
db <- dbConnect(duckdb(), dbdir = eunomia_dir())
cdm <- cdmFromCon(con = db, cdmSchema = "main", writeSchema = "main")Fix eunomia problems with drugs
Add a drug_strength table:
x <- read.csv("https://raw.githubusercontent.com/darwin-eu-dev/DrugUtilisation/main/extras/mock_drug_strength.csv")
cdm <- insertTable(cdm = cdm, name = "drug_strength", table = x)Generate a drug cohort
To generate a drug cohort we need a conceptSet (named list of concepts). Each element will be a different cohort, so we can instantiate multiple cohorts following the same logit at the same time.
Methods to create a conceptSet:
read it from JSON CodelistGenerator::codesFromConceptSet
create from an ingredient CodelistGenerator::getDrugIngredientCodes
create from a text search CodelistGenerator::getCandidateCodes
custom creation
codelist |> omopgenerics::newCodelist()
- concept1 (4 codes)
- concept2 (3 codes)Generate a drug cohort
For this example we are going to use all records of acteaminophen:
library(CodelistGenerator)
codelist <- getDrugIngredientCodes(cdm, "acetaminophen")
codelist$acetaminophen
[1] 1125315 1127078 1127433 40229134 40231925 40162522 19133768Generate a drug cohort
Once we have our conceptSet we can create a simple cohort:
library(DrugUtilisation)
cdm <- generateDrugUtilisationCohortSet(
cdm = cdm,
name = "dus_cohort",
conceptSet = codelist
)The results is the cdm object with the new cohort instantiated.
cdm── # OMOP CDM reference (duckdb) of Synthea synthetic health database ──────────────────────────────────────────────────• omop tables: person, observation_period, visit_occurrence, visit_detail, condition_occurrence, drug_exposure,
procedure_occurrence, device_exposure, measurement, observation, death, note, note_nlp, specimen, fact_relationship,
location, care_site, provider, payer_plan_period, cost, drug_era, dose_era, condition_era, metadata, cdm_source,
concept, vocabulary, domain, concept_class, concept_relationship, relationship, concept_synonym, concept_ancestor,
source_to_concept_map, drug_strength• cohort tables: dus_cohort• achilles tables: -• other tables: -cdm$dus_cohort# Source: SQL [?? x 4]
# Database: DuckDB v0.10.0 [martics@Windows 10 x64:R 4.2.3/C:\Users\martics\AppData\Local\Temp\Rtmp6LHYNc\filec9d465b41608.duckdb]
cohort_definition_id subject_id cohort_start_date cohort_end_date
<int> <int> <date> <date>
1 1 5 1971-03-12 1971-03-26
2 1 9 2003-02-23 2003-03-09
3 1 17 1971-10-27 1971-12-01
4 1 65 2001-09-15 2001-09-22
5 1 79 1990-08-05 1990-08-12
6 1 115 1960-12-25 1961-01-08
7 1 135 2001-01-06 2001-02-03
8 1 160 2016-08-05 2016-09-04
9 1 162 1979-07-29 1979-08-05
10 1 177 1973-10-15 1973-11-04
# ℹ more rowsGenerate a drug cohort
We can check some metadata of the cohort:
settings(cdm$dus_cohort)# A tibble: 1 × 11
cohort_definition_id cohort_name duration_range_min duration_range_max impute_duration gap_era prior_use_washout
<int> <chr> <chr> <chr> <chr> <chr> <chr>
1 1 acetaminophen 1 Inf none 0 0
# ℹ 4 more variables: prior_observation <chr>, cohort_date_range_start <chr>, cohort_date_range_end <chr>, limit <chr>cohortCount(cdm$dus_cohort)# A tibble: 1 × 3
cohort_definition_id number_records number_subjects
<int> <int> <int>
1 1 13908 2679Generate a drug cohort
attrition(cdm$dus_cohort)# A tibble: 1 × 7
cohort_definition_id number_records number_subjects reason_id reason excluded_records excluded_subjects
<int> <int> <int> <int> <chr> <int> <int>
1 1 13908 2679 1 Initial qualifying e… 0 0Generate a drug cohort
summary(cdm$dus_cohort)# A tibble: 6 × 13
result_id cdm_name group_name group_level strata_name strata_level variable_name variable_level estimate_name
<int> <chr> <chr> <chr> <chr> <chr> <chr> <chr> <chr>
1 1 Synthea syntheti… cohort_na… acetaminop… overall overall number_recor… <NA> count
2 1 Synthea syntheti… cohort_na… acetaminop… overall overall number_subje… <NA> count
3 2 Synthea syntheti… cohort_na… acetaminop… reason Initial qua… number_recor… <NA> count
4 2 Synthea syntheti… cohort_na… acetaminop… reason Initial qua… number_subje… <NA> count
5 2 Synthea syntheti… cohort_na… acetaminop… reason Initial qua… excluded_rec… <NA> count
6 2 Synthea syntheti… cohort_na… acetaminop… reason Initial qua… excluded_sub… <NA> count
# ℹ 4 more variables: estimate_type <chr>, estimate_value <chr>, additional_name <chr>, additional_level <chr>Generate a drug cohort
Let’s explore the parameters:
?generateDrugUtilisationCohortSet
Generate a drug cohort
Generate a drug cohort
Generate a drug cohort
Generate a drug cohort
Generate a drug cohort
Generate a drug cohort
cdm <- generateDrugUtilisationCohortSet(
cdm = cdm,
name = "dus_new_user",
conceptSet = codelist,
gapEra = 30, # 30 day separated prescriptions will be joined
priorUseWashout = 365, # no use in the prior 365 days
priorObservation = 365, # in the database for more than 365 dyas before start date
cohortDateRange = as.Date(c("2000-01-01", "2019-12-31")), # range of allowed dates
limit = "first" # limit to only the first record per person
)Generate a drug cohort
Rows: 1
Columns: 11
$ cohort_definition_id <int> 1
$ cohort_name <chr> "acetaminophen"
$ duration_range_min <chr> "1"
$ duration_range_max <chr> "Inf"
$ impute_duration <chr> "none"
$ gap_era <chr> "30"
$ prior_use_washout <chr> "365"
$ prior_observation <chr> "365"
$ cohort_date_range_start <chr> "2000-01-01"
$ cohort_date_range_end <chr> "2019-12-31"
$ limit <chr> "first"Generate a drug cohort
cohortCount(cdm$dus_new_user) |> glimpse()Rows: 1
Columns: 3
$ cohort_definition_id <int> 1
$ number_records <int> 1979
$ number_subjects <int> 1979Generate a drug cohort
Rows: 7
Columns: 7
$ cohort_definition_id <int> 1, 1, 1, 1, 1, 1, 1
$ number_records <int> 13908, 13860, 12907, 12533, 3332, 3332, 1979
$ number_subjects <int> 2679, 2679, 2679, 2674, 1979, 1979, 1979
$ reason_id <int> 1, 2, 3, 4, 5, 6, 7
$ reason <chr> "Initial qualifying events", "join exposures separated by 30 or less days", "require prio…
$ excluded_records <int> 0, 48, 953, 374, 9201, 0, 1353
$ excluded_subjects <int> 0, 0, 0, 5, 695, 0, 0Generate a drug cohort
library(CohortCharacteristics)
cdm$dus_new_user |>
summariseCohortAttrition() |>
plotCohortAttrition()Some helpers
?generateIngredientCohortSet?generateAtcCohortSetYour turn
Can you use the generateIngredientCohortSet to create an amoxicillin new users cohort with a gapEra of 30 days, with at least a prior observation of 365 days and no prior use of the drug in the last 365 days.
Your turn
Dealing with daily dose
This function works if:
- The relation between the
drug_concept_idandingredient_concept_id, can be established thoughdrug_strengthtable by one of the 126 identified valid patterns. And if they have a formula assigned. Currently 41 patterns have a formula assigned you can check them here:
patternsWithFormula |> print(n = Inf)# A tibble: 41 × 9
pattern_id amount amount_unit numerator numerator_unit denominator denominator_unit formula_name formula
<dbl> <chr> <chr> <chr> <chr> <chr> <chr> <chr> <chr>
1 1 <NA> <NA> number microgram number hour time based … if (de…
2 2 <NA> <NA> number milligram number hour time based … if (de…
3 3 <NA> <NA> number unit number hour time based … if (de…
4 4 <NA> <NA> number microgram <NA> hour time based … 24 * n…
5 5 <NA> <NA> number milligram <NA> hour time based … 24 * n…
6 6 number international unit <NA> <NA> <NA> <NA> fixed amoun… quanti…
7 7 number microgram <NA> <NA> <NA> <NA> fixed amoun… quanti…
8 8 number milliequivalent <NA> <NA> <NA> <NA> fixed amoun… quanti…
9 9 number milligram <NA> <NA> <NA> <NA> fixed amoun… quanti…
10 10 number milliliter <NA> <NA> <NA> <NA> fixed amoun… quanti…
11 11 number unit <NA> <NA> <NA> <NA> fixed amoun… quanti…
12 12 <NA> <NA> number international unit number milligram concentrati… quanti…
13 13 <NA> <NA> number international unit number milliliter concentrati… quanti…
14 14 <NA> <NA> number milliequivalent number milliliter concentrati… quanti…
15 15 <NA> <NA> number milligram number Actuation concentrati… quanti…
16 16 <NA> <NA> number milligram number liter concentrati… quanti…
17 17 <NA> <NA> number milligram number milligram concentrati… quanti…
18 18 <NA> <NA> number milligram number milliliter concentrati… quanti…
19 19 <NA> <NA> number milligram number square centimet… concentrati… quanti…
20 20 <NA> <NA> number milliliter number milligram concentrati… quanti…
21 21 <NA> <NA> number milliliter number milliliter concentrati… quanti…
22 22 <NA> <NA> number unit number Actuation concentrati… quanti…
23 23 <NA> <NA> number unit number milligram concentrati… quanti…
24 24 <NA> <NA> number unit number milliliter concentrati… quanti…
25 25 <NA> <NA> number unit number square centimet… concentrati… quanti…
26 26 <NA> <NA> number international unit <NA> milligram concentrati… quanti…
27 27 <NA> <NA> number international unit <NA> milliliter concentrati… quanti…
28 28 <NA> <NA> number mega-international … <NA> milliliter concentrati… quanti…
29 29 <NA> <NA> number milliequivalent <NA> milligram concentrati… quanti…
30 30 <NA> <NA> number milliequivalent <NA> milliliter concentrati… quanti…
31 31 <NA> <NA> number milligram <NA> Actuation concentrati… quanti…
32 32 <NA> <NA> number milligram <NA> liter concentrati… quanti…
33 33 <NA> <NA> number milligram <NA> milligram concentrati… quanti…
34 34 <NA> <NA> number milligram <NA> milliliter concentrati… quanti…
35 35 <NA> <NA> number milligram <NA> square centimet… concentrati… quanti…
36 36 <NA> <NA> number milliliter <NA> milligram concentrati… quanti…
37 37 <NA> <NA> number milliliter <NA> milliliter concentrati… quanti…
38 38 <NA> <NA> number unit <NA> Actuation concentrati… quanti…
39 39 <NA> <NA> number unit <NA> milligram concentrati… quanti…
40 40 <NA> <NA> number unit <NA> milliliter concentrati… quanti…
41 41 <NA> <NA> number unit <NA> square centimet… concentrati… quanti…daily dose
We can add daily dose to drug_exposure like this:
cdm$drug_exposure |>
addDailyDose(ingredientConceptId = 1125315) |>
glimpse()Rows: ??
Columns: 25
Database: DuckDB v0.10.0 [martics@Windows 10 x64:R 4.2.3/C:\Users\martics\AppData\Local\Temp\Rtmp6LHYNc\filec9d465b41608.duckdb]
$ drug_exposure_id <int> 47027, 10459, 10524, 9646, 9653, 31813, 13969, 60871, 14061, 60753, 6923, 47028, …
$ person_id <int> 3895, 871, 877, 806, 806, 2636, 1169, 5063, 1175, 5051, 582, 3896, 877, 2815, 174…
$ drug_concept_id <int> 40162522, 1127078, 1127433, 1127078, 40229134, 1127433, 1127433, 1127433, 4023192…
$ drug_exposure_start_date <date> 1982-03-30, 1967-11-09, 1992-05-28, 1962-07-02, 2009-01-18, 2017-08-22, 1937-12-…
$ drug_exposure_start_datetime <dttm> 1982-03-30, 1967-11-09, 1992-05-28, 1962-07-02, 2009-01-18, 2017-08-22, 1937-12-…
$ drug_exposure_end_date <date> 1982-03-30, 1968-01-08, 1992-06-11, 1962-07-30, 2009-01-25, 2017-09-05, 1937-12-…
$ drug_exposure_end_datetime <dttm> 1982-03-30, 1968-01-08, 1992-06-11, 1962-07-30, 2009-01-25, 2017-09-05, 1937-12-…
$ verbatim_end_date <date> NA, 1968-01-08, 1992-06-11, 1962-07-30, 2009-01-25, 2017-09-05, 1937-12-17, 2010…
$ drug_type_concept_id <int> 38000177, 38000177, 38000177, 38000177, 38000177, 38000177, 38000177, 38000177, 3…
$ stop_reason <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, N…
$ refills <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
$ quantity <int> 0, 60, 14, 28, 7, 14, 14, 14, 0, 35, 7, 14, 14, 7, 14, 30, 7, 14, 7, 21, 14, 7, 1…
$ days_supply <int> 0, 60, 14, 28, 7, 14, 14, 14, 0, 35, 7, 14, 14, 7, 14, 30, 7, 14, 7, 21, 14, 7, 1…
$ sig <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, N…
$ route_concept_id <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
$ lot_number <chr> "0", "0", "0", "0", "0", "0", "0", "0", "0", "0", "0", "0", "0", "0", "0", "0", "…
$ provider_id <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
$ visit_occurrence_id <int> 259023, 57955, 58440, 53468, 53505, 174469, 77028, 336572, 77481, 335907, 38579, …
$ visit_detail_id <int> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
$ drug_source_value <chr> "857005", "282464", "313782", "282464", "1043400", "313782", "313782", "313782", …
$ drug_source_concept_id <int> 40162522, 1127078, 1127433, 1127078, 40229134, 1127433, 1127433, 1127433, 4023192…
$ route_source_value <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, N…
$ dose_unit_source_value <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, N…
$ daily_dose <dbl> NA, 157.37705, 303.33333, 154.48276, 18.98750, 303.33333, 303.33333, 303.33333, N…
$ unit <chr> "milligram", "milligram", "milligram", "milligram", "milligram", "milligram", "mi…!! As drug exposure is a big table it is not recommended to add daily dose to the whole drug_exposure table, first it is important to subset to the subjects and concepts of interest.
daily dose
To see the coverage of an specific ingredient you can compute it as:
dailyDoseCoverage(cdm = cdm, ingredientConceptId = 1125315) |>
select(-c("result_id", "cdm_name", "group_name", "group_level")) |>
print(n = Inf)# A tibble: 60 × 9
strata_name strata_level variable_name variable_level estimate_name estimate_type estimate_value additional_name
<chr> <chr> <chr> <chr> <chr> <chr> <chr> <chr>
1 overall overall number recor… <NA> count integer 14205 overall
2 overall overall daily_dose <NA> count_missing integer 697 overall
3 overall overall daily_dose <NA> percentage_m… percentage 4.90672298486… overall
4 overall overall daily_dose <NA> mean numeric 234.787349787… overall
5 overall overall daily_dose <NA> sd numeric 104.686522784… overall
6 overall overall daily_dose <NA> min numeric 18.9875 overall
7 overall overall daily_dose <NA> q05 numeric 18.9875 overall
8 overall overall daily_dose <NA> q25 numeric 152.727272727… overall
9 overall overall daily_dose <NA> median numeric 284.375 overall
10 overall overall daily_dose <NA> q75 numeric 303.333333333… overall
11 overall overall daily_dose <NA> q95 numeric 315.972222222… overall
12 overall overall daily_dose <NA> max numeric 375 overall
13 unit milligram number recor… <NA> count integer 14205 overall
14 unit milligram daily_dose <NA> count_missing integer 697 overall
15 unit milligram daily_dose <NA> percentage_m… percentage 4.90672298486… overall
16 unit milligram daily_dose <NA> mean numeric 234.787349787… overall
17 unit milligram daily_dose <NA> sd numeric 104.686522784… overall
18 unit milligram daily_dose <NA> min numeric 18.9875 overall
19 unit milligram daily_dose <NA> q05 numeric 18.9875 overall
20 unit milligram daily_dose <NA> q25 numeric 152.727272727… overall
21 unit milligram daily_dose <NA> median numeric 284.375 overall
22 unit milligram daily_dose <NA> q75 numeric 303.333333333… overall
23 unit milligram daily_dose <NA> q95 numeric 315.972222222… overall
24 unit milligram daily_dose <NA> max numeric 375 overall
25 route &&& unit NA &&& mill… number recor… <NA> count integer 14205 overall
26 route &&& unit NA &&& mill… daily_dose <NA> count_missing integer 697 overall
27 route &&& unit NA &&& mill… daily_dose <NA> percentage_m… percentage 4.90672298486… overall
28 route &&& unit NA &&& mill… daily_dose <NA> mean numeric 234.787349787… overall
29 route &&& unit NA &&& mill… daily_dose <NA> sd numeric 104.686522784… overall
30 route &&& unit NA &&& mill… daily_dose <NA> min numeric 18.9875 overall
31 route &&& unit NA &&& mill… daily_dose <NA> q05 numeric 18.9875 overall
32 route &&& unit NA &&& mill… daily_dose <NA> q25 numeric 152.727272727… overall
33 route &&& unit NA &&& mill… daily_dose <NA> median numeric 284.375 overall
34 route &&& unit NA &&& mill… daily_dose <NA> q75 numeric 303.333333333… overall
35 route &&& unit NA &&& mill… daily_dose <NA> q95 numeric 315.972222222… overall
36 route &&& unit NA &&& mill… daily_dose <NA> max numeric 375 overall
37 unit &&& route … milligram &… number recor… <NA> count integer 1993 overall
38 unit &&& route … milligram &… number recor… <NA> count integer 12212 overall
39 unit &&& route … milligram &… daily_dose <NA> count_missing integer 2 overall
40 unit &&& route … milligram &… daily_dose <NA> count_missing integer 695 overall
41 unit &&& route … milligram &… daily_dose <NA> percentage_m… percentage 0.10035122930… overall
42 unit &&& route … milligram &… daily_dose <NA> percentage_m… percentage 5.69112348509… overall
43 unit &&& route … milligram &… daily_dose <NA> mean numeric 19.6295316423… overall
44 unit &&& route … milligram &… daily_dose <NA> mean numeric 271.982731911… overall
45 unit &&& route … milligram &… daily_dose <NA> sd numeric 0.63291464828… overall
46 unit &&& route … milligram &… daily_dose <NA> sd numeric 58.8791704932… overall
47 unit &&& route … milligram &… daily_dose <NA> min numeric 18.9875 overall
48 unit &&& route … milligram &… daily_dose <NA> min numeric 148.571428571… overall
49 unit &&& route … milligram &… daily_dose <NA> q05 numeric 18.9875 overall
50 unit &&& route … milligram &… daily_dose <NA> q05 numeric 149.333333333… overall
51 unit &&& route … milligram &… daily_dose <NA> q25 numeric 18.9875 overall
52 unit &&& route … milligram &… daily_dose <NA> q25 numeric 284.375 overall
53 unit &&& route … milligram &… daily_dose <NA> median numeric 20.2533333333… overall
54 unit &&& route … milligram &… daily_dose <NA> median numeric 303.333333333… overall
55 unit &&& route … milligram &… daily_dose <NA> q75 numeric 20.2533333333… overall
56 unit &&& route … milligram &… daily_dose <NA> q75 numeric 303.333333333… overall
57 unit &&& route … milligram &… daily_dose <NA> q95 numeric 20.2533333333… overall
58 unit &&& route … milligram &… daily_dose <NA> q95 numeric 316.447368421… overall
59 unit &&& route … milligram &… daily_dose <NA> max numeric 20.2533333333… overall
60 unit &&& route … milligram &… daily_dose <NA> max numeric 375 overall
# ℹ 1 more variable: additional_level <chr>Characterisation
Characterise drug use
We have a function to add columns related to drugUse:
cdm$dus_new_user |>
addDrugUse(
ingredientConceptId = 1125315,
duration = FALSE,
quantity = FALSE,
dose = FALSE
)# Source: table<og_012_1718884000> [?? x 6]
# Database: DuckDB v0.10.0 [martics@Windows 10 x64:R 4.2.3/C:\Users\martics\AppData\Local\Temp\Rtmp6LHYNc\filec9d465b41608.duckdb]
cohort_definition_id subject_id cohort_start_date cohort_end_date number_exposures number_eras
<int> <int> <date> <date> <dbl> <dbl>
1 1 5 2003-09-10 2003-09-17 1 1
2 1 9 2003-02-23 2003-03-09 1 1
3 1 30 2004-11-28 2004-12-05 1 1
4 1 65 2001-09-15 2001-09-22 1 1
5 1 143 2000-08-16 2000-08-23 1 1
6 1 154 2017-07-15 2017-07-20 1 1
7 1 160 2003-06-06 2003-06-13 1 1
8 1 162 2001-06-25 2001-07-02 1 1
9 1 177 2002-05-14 2002-05-14 1 1
10 1 184 2019-03-06 2019-03-13 1 1
# ℹ more rowsnumber_exposure and number_eras are always computed.
Characterise drug use
duartion = TRUE adds duration and impute_duration_percentage fields
cdm$dus_new_user |>
addDrugUse(
ingredientConceptId = 1125315,
duration = TRUE,
quantity = FALSE,
dose = FALSE
) |>
select(-c("number_exposures", "number_eras"))# Source: SQL [?? x 6]
# Database: DuckDB v0.10.0 [martics@Windows 10 x64:R 4.2.3/C:\Users\martics\AppData\Local\Temp\Rtmp6LHYNc\filec9d465b41608.duckdb]
cohort_definition_id subject_id cohort_start_date cohort_end_date duration impute_duration_percentage
<int> <int> <date> <date> <dbl> <dbl>
1 1 5 2003-09-10 2003-09-17 8 0
2 1 9 2003-02-23 2003-03-09 15 0
3 1 30 2004-11-28 2004-12-05 8 0
4 1 65 2001-09-15 2001-09-22 8 0
5 1 143 2000-08-16 2000-08-23 8 0
6 1 154 2017-07-15 2017-07-20 6 0
7 1 160 2003-06-06 2003-06-13 8 0
8 1 162 2001-06-25 2001-07-02 8 0
9 1 177 2002-05-14 2002-05-14 1 0
10 1 184 2019-03-06 2019-03-13 8 0
# ℹ more rowsCharacterise drug use
quantity = TRUE adds initial_quantity and cumulative_quantity fields
cdm$dus_new_user |>
addDrugUse(
ingredientConceptId = 1125315,
duration = FALSE,
quantity = TRUE,
dose = FALSE
) |>
select(-c("number_exposures", "number_eras"))# Source: SQL [?? x 6]
# Database: DuckDB v0.10.0 [martics@Windows 10 x64:R 4.2.3/C:\Users\martics\AppData\Local\Temp\Rtmp6LHYNc\filec9d465b41608.duckdb]
cohort_definition_id subject_id cohort_start_date cohort_end_date cumulative_quantity initial_quantity
<int> <int> <date> <date> <dbl> <dbl>
1 1 5 2003-09-10 2003-09-17 7 7
2 1 9 2003-02-23 2003-03-09 14 14
3 1 30 2004-11-28 2004-12-05 7 7
4 1 65 2001-09-15 2001-09-22 7 7
5 1 143 2000-08-16 2000-08-23 7 7
6 1 154 2017-07-15 2017-07-20 7 7
7 1 160 2003-06-06 2003-06-13 7 7
8 1 162 2001-06-25 2001-07-02 7 7
9 1 177 2002-05-14 2002-05-14 0 0
10 1 184 2019-03-06 2019-03-13 7 7
# ℹ more rowsCharacterise drug use
dose = TRUE adds impute_daily_dose_percentage, initial_daily_dose and cumulative_dose fields
cdm$dus_new_user |>
addDrugUse(
ingredientConceptId = 1125315,
duration = FALSE,
quantity = FALSE,
dose = TRUE
) |>
select(-c("number_exposures", "number_eras"))# Source: SQL [?? x 7]
# Database: DuckDB v0.10.0 [martics@Windows 10 x64:R 4.2.3/C:\Users\martics\AppData\Local\Temp\Rtmp6LHYNc\filec9d465b41608.duckdb]
cohort_definition_id subject_id cohort_start_date cohort_end_date impute_daily_dose_percentage initial_daily_dose_m…¹
<int> <int> <date> <date> <dbl> <dbl>
1 1 5 2003-09-10 2003-09-17 0 19.0
2 1 9 2003-02-23 2003-03-09 0 303.
3 1 30 2004-11-28 2004-12-05 0 284.
4 1 65 2001-09-15 2001-09-22 0 284.
5 1 143 2000-08-16 2000-08-23 0 284.
6 1 154 2017-07-15 2017-07-20 0 284.
7 1 160 2003-06-06 2003-06-13 0 19.0
8 1 162 2001-06-25 2001-07-02 0 284.
9 1 177 2002-05-14 2002-05-14 100 NA
10 1 184 2019-03-06 2019-03-13 0 19.0
# ℹ more rows
# ℹ abbreviated name: ¹initial_daily_dose_milligram
# ℹ 1 more variable: cumulative_dose_milligram <dbl>Characterise drug use
We can make some decisions to solve how we threat gaps (not exposed periods):
Characterise drug use
Characterise drug use
cdm$dus_new_user |>
addDrugUse(
ingredientConceptId = 1125315,
eraJoinMode = "zero",
sameIndexMode = "sum"
)# Source: table<og_026_1718884018> [?? x 13]
# Database: DuckDB v0.10.0 [martics@Windows 10 x64:R 4.2.3/C:\Users\martics\AppData\Local\Temp\Rtmp6LHYNc\filec9d465b41608.duckdb]
cohort_definition_id subject_id cohort_start_date cohort_end_date duration number_exposures cumulative_quantity
<int> <int> <date> <date> <dbl> <dbl> <dbl>
1 1 5 2003-09-10 2003-09-17 8 1 7
2 1 9 2003-02-23 2003-03-09 15 1 14
3 1 30 2004-11-28 2004-12-05 8 1 7
4 1 65 2001-09-15 2001-09-22 8 1 7
5 1 143 2000-08-16 2000-08-23 8 1 7
6 1 154 2017-07-15 2017-07-20 6 1 7
7 1 160 2003-06-06 2003-06-13 8 1 7
8 1 162 2001-06-25 2001-07-02 8 1 7
9 1 177 2002-05-14 2002-05-14 1 1 0
10 1 184 2019-03-06 2019-03-13 8 1 7
# ℹ more rows
# ℹ 6 more variables: initial_quantity <dbl>, impute_duration_percentage <dbl>, number_eras <dbl>,
# impute_daily_dose_percentage <dbl>, initial_daily_dose_milligram <dbl>, cumulative_dose_milligram <dbl>Characterise drug use
We can summarise all this information grouped by cohorts in a summarisedResult object:
cdm$dus_new_user |>
addDrugUse(ingredientConceptId = 1125315) |>
summariseDrugUse() |>
select(-"strata_name", -"strata_level") |>
print(n = Inf)# A tibble: 101 × 11
result_id cdm_name group_name group_level variable_name variable_level estimate_name estimate_type estimate_value
<int> <chr> <chr> <chr> <chr> <chr> <chr> <chr> <chr>
1 1 Synthea sy… cohort_na… acetaminop… number recor… <NA> count integer 1979
2 1 Synthea sy… cohort_na… acetaminop… number subje… <NA> count integer 1979
3 1 Synthea sy… cohort_na… acetaminop… duration <NA> min numeric 1
4 1 Synthea sy… cohort_na… acetaminop… duration <NA> q05 numeric 1.90000000000…
5 1 Synthea sy… cohort_na… acetaminop… duration <NA> q25 numeric 8
6 1 Synthea sy… cohort_na… acetaminop… duration <NA> median numeric 15
7 1 Synthea sy… cohort_na… acetaminop… duration <NA> q75 numeric 15
8 1 Synthea sy… cohort_na… acetaminop… duration <NA> q95 numeric 38
9 1 Synthea sy… cohort_na… acetaminop… duration <NA> max numeric 128
10 1 Synthea sy… cohort_na… acetaminop… duration <NA> mean numeric 16.4537645275…
11 1 Synthea sy… cohort_na… acetaminop… duration <NA> sd numeric 15.5892764641…
12 1 Synthea sy… cohort_na… acetaminop… duration <NA> count_missing integer 0
13 1 Synthea sy… cohort_na… acetaminop… duration <NA> percentage_m… percentage 0
14 1 Synthea sy… cohort_na… acetaminop… number_expos… <NA> min numeric 1
15 1 Synthea sy… cohort_na… acetaminop… number_expos… <NA> q05 numeric 1
16 1 Synthea sy… cohort_na… acetaminop… number_expos… <NA> q25 numeric 1
17 1 Synthea sy… cohort_na… acetaminop… number_expos… <NA> median numeric 1
18 1 Synthea sy… cohort_na… acetaminop… number_expos… <NA> q75 numeric 1
19 1 Synthea sy… cohort_na… acetaminop… number_expos… <NA> q95 numeric 1
20 1 Synthea sy… cohort_na… acetaminop… number_expos… <NA> max numeric 3
21 1 Synthea sy… cohort_na… acetaminop… number_expos… <NA> mean numeric 1.02475997978…
22 1 Synthea sy… cohort_na… acetaminop… number_expos… <NA> sd numeric 0.16180659945…
23 1 Synthea sy… cohort_na… acetaminop… number_expos… <NA> count_missing integer 0
24 1 Synthea sy… cohort_na… acetaminop… number_expos… <NA> percentage_m… percentage 0
25 1 Synthea sy… cohort_na… acetaminop… cumulative_q… <NA> min numeric 0
26 1 Synthea sy… cohort_na… acetaminop… cumulative_q… <NA> q05 numeric 7
27 1 Synthea sy… cohort_na… acetaminop… cumulative_q… <NA> q25 numeric 7
28 1 Synthea sy… cohort_na… acetaminop… cumulative_q… <NA> median numeric 14
29 1 Synthea sy… cohort_na… acetaminop… cumulative_q… <NA> q75 numeric 14
30 1 Synthea sy… cohort_na… acetaminop… cumulative_q… <NA> q95 numeric 35.1999999999…
31 1 Synthea sy… cohort_na… acetaminop… cumulative_q… <NA> max numeric 100
32 1 Synthea sy… cohort_na… acetaminop… cumulative_q… <NA> mean numeric 15.5219807983…
33 1 Synthea sy… cohort_na… acetaminop… cumulative_q… <NA> sd numeric 15.4241972689…
34 1 Synthea sy… cohort_na… acetaminop… cumulative_q… <NA> count_missing integer 0
35 1 Synthea sy… cohort_na… acetaminop… cumulative_q… <NA> percentage_m… percentage 0
36 1 Synthea sy… cohort_na… acetaminop… initial_quan… <NA> min numeric 0
37 1 Synthea sy… cohort_na… acetaminop… initial_quan… <NA> q05 numeric 7
38 1 Synthea sy… cohort_na… acetaminop… initial_quan… <NA> q25 numeric 7
39 1 Synthea sy… cohort_na… acetaminop… initial_quan… <NA> median numeric 14
40 1 Synthea sy… cohort_na… acetaminop… initial_quan… <NA> q75 numeric 14
41 1 Synthea sy… cohort_na… acetaminop… initial_quan… <NA> q95 numeric 35
42 1 Synthea sy… cohort_na… acetaminop… initial_quan… <NA> max numeric 100
43 1 Synthea sy… cohort_na… acetaminop… initial_quan… <NA> mean numeric 15.3956543708…
44 1 Synthea sy… cohort_na… acetaminop… initial_quan… <NA> sd numeric 15.2265534027…
45 1 Synthea sy… cohort_na… acetaminop… initial_quan… <NA> count_missing integer 0
46 1 Synthea sy… cohort_na… acetaminop… initial_quan… <NA> percentage_m… percentage 0
47 1 Synthea sy… cohort_na… acetaminop… impute_durat… <NA> min numeric 0
48 1 Synthea sy… cohort_na… acetaminop… impute_durat… <NA> q05 numeric 0
49 1 Synthea sy… cohort_na… acetaminop… impute_durat… <NA> q25 numeric 0
50 1 Synthea sy… cohort_na… acetaminop… impute_durat… <NA> median numeric 0
51 1 Synthea sy… cohort_na… acetaminop… impute_durat… <NA> q75 numeric 0
52 1 Synthea sy… cohort_na… acetaminop… impute_durat… <NA> q95 numeric 0
53 1 Synthea sy… cohort_na… acetaminop… impute_durat… <NA> max numeric 0
54 1 Synthea sy… cohort_na… acetaminop… impute_durat… <NA> mean numeric 0
55 1 Synthea sy… cohort_na… acetaminop… impute_durat… <NA> sd numeric 0
56 1 Synthea sy… cohort_na… acetaminop… impute_durat… <NA> count_missing integer 0
57 1 Synthea sy… cohort_na… acetaminop… impute_durat… <NA> percentage_m… percentage 0
58 1 Synthea sy… cohort_na… acetaminop… number_eras <NA> min numeric 1
59 1 Synthea sy… cohort_na… acetaminop… number_eras <NA> q05 numeric 1
60 1 Synthea sy… cohort_na… acetaminop… number_eras <NA> q25 numeric 1
61 1 Synthea sy… cohort_na… acetaminop… number_eras <NA> median numeric 1
62 1 Synthea sy… cohort_na… acetaminop… number_eras <NA> q75 numeric 1
63 1 Synthea sy… cohort_na… acetaminop… number_eras <NA> q95 numeric 1
64 1 Synthea sy… cohort_na… acetaminop… number_eras <NA> max numeric 2
65 1 Synthea sy… cohort_na… acetaminop… number_eras <NA> mean numeric 1.00404244567…
66 1 Synthea sy… cohort_na… acetaminop… number_eras <NA> sd numeric 0.06346762761…
67 1 Synthea sy… cohort_na… acetaminop… number_eras <NA> count_missing integer 0
68 1 Synthea sy… cohort_na… acetaminop… number_eras <NA> percentage_m… percentage 0
69 1 Synthea sy… cohort_na… acetaminop… impute_daily… <NA> min numeric 0
70 1 Synthea sy… cohort_na… acetaminop… impute_daily… <NA> q05 numeric 0
71 1 Synthea sy… cohort_na… acetaminop… impute_daily… <NA> q25 numeric 0
72 1 Synthea sy… cohort_na… acetaminop… impute_daily… <NA> median numeric 0
73 1 Synthea sy… cohort_na… acetaminop… impute_daily… <NA> q75 numeric 0
74 1 Synthea sy… cohort_na… acetaminop… impute_daily… <NA> q95 numeric 50
75 1 Synthea sy… cohort_na… acetaminop… impute_daily… <NA> max numeric 100
76 1 Synthea sy… cohort_na… acetaminop… impute_daily… <NA> mean numeric 5.59204985683…
77 1 Synthea sy… cohort_na… acetaminop… impute_daily… <NA> sd numeric 21.8615099438…
78 1 Synthea sy… cohort_na… acetaminop… impute_daily… <NA> count_missing integer 0
79 1 Synthea sy… cohort_na… acetaminop… impute_daily… <NA> percentage_m… percentage 0
80 1 Synthea sy… cohort_na… acetaminop… initial_dail… <NA> min numeric 18.9875
81 1 Synthea sy… cohort_na… acetaminop… initial_dail… <NA> q05 numeric 18.9875
82 1 Synthea sy… cohort_na… acetaminop… initial_dail… <NA> q25 numeric 284.375
83 1 Synthea sy… cohort_na… acetaminop… initial_dail… <NA> median numeric 303.333333333…
84 1 Synthea sy… cohort_na… acetaminop… initial_dail… <NA> q75 numeric 303.333333333…
85 1 Synthea sy… cohort_na… acetaminop… initial_dail… <NA> q95 numeric 316.281067251…
86 1 Synthea sy… cohort_na… acetaminop… initial_dail… <NA> max numeric 321.782178217…
87 1 Synthea sy… cohort_na… acetaminop… initial_dail… <NA> mean numeric 248.791088712…
88 1 Synthea sy… cohort_na… acetaminop… initial_dail… <NA> sd numeric 108.199579381…
89 1 Synthea sy… cohort_na… acetaminop… initial_dail… <NA> count_missing integer 91
90 1 Synthea sy… cohort_na… acetaminop… initial_dail… <NA> percentage_m… percentage 4.59828196058…
91 1 Synthea sy… cohort_na… acetaminop… cumulative_d… <NA> min numeric Inf
92 1 Synthea sy… cohort_na… acetaminop… cumulative_d… <NA> q05 numeric <NA>
93 1 Synthea sy… cohort_na… acetaminop… cumulative_d… <NA> q25 numeric <NA>
94 1 Synthea sy… cohort_na… acetaminop… cumulative_d… <NA> median numeric <NA>
95 1 Synthea sy… cohort_na… acetaminop… cumulative_d… <NA> q75 numeric <NA>
96 1 Synthea sy… cohort_na… acetaminop… cumulative_d… <NA> q95 numeric <NA>
97 1 Synthea sy… cohort_na… acetaminop… cumulative_d… <NA> max numeric -Inf
98 1 Synthea sy… cohort_na… acetaminop… cumulative_d… <NA> mean numeric NaN
99 1 Synthea sy… cohort_na… acetaminop… cumulative_d… <NA> sd numeric <NA>
100 1 Synthea sy… cohort_na… acetaminop… cumulative_d… <NA> count_missing integer 1979
101 1 Synthea sy… cohort_na… acetaminop… cumulative_d… <NA> percentage_m… percentage 100
# ℹ 2 more variables: additional_name <chr>, additional_level <chr>Your turn
Get the mean and median duration and initial daily dose for our amoxicillin cohort:
addDrugUse to the amoxicillin cohort (we have to find the ingredientConceptId for amoxicillin!!)
summarise drug use
filter to find the information that we need
Your turn
ingredientConceptId <- cdm$concept |>
filter(concept_class_id == "Ingredient") |>
filter(concept_name == "Amoxicillin") |>
pull(concept_id)
result <- cdm$my_cohort |>
addDrugUse(ingredientConceptId = ingredientConceptId) |>
summariseDrugUse()
result |>
filter(
variable_name %in% c("duration", "initial_daily_dose_milligram"),
estimate_name %in% c("mean", "median")
)# A tibble: 4 × 13
result_id cdm_name group_name group_level strata_name strata_level variable_name variable_level estimate_name
<int> <chr> <chr> <chr> <chr> <chr> <chr> <chr> <chr>
1 1 Synthea syntheti… cohort_na… amoxicillin overall overall duration <NA> median
2 1 Synthea syntheti… cohort_na… amoxicillin overall overall duration <NA> mean
3 1 Synthea syntheti… cohort_na… amoxicillin overall overall initial_dail… <NA> median
4 1 Synthea syntheti… cohort_na… amoxicillin overall overall initial_dail… <NA> mean
# ℹ 4 more variables: estimate_type <chr>, estimate_value <chr>, additional_name <chr>, additional_level <chr>Indication
First we need to instantiate some indication cohorts, in this case we are going to explore sinusitis and bronchitis:
library(CohortConstructor)
cdm$indications <- conceptCohort(
cdm = cdm,
conceptSet = list(sinusitis = c(4294548, 40481087, 257012), bronchitis = c(258780, 260139)),
name = "indications"
)
settings(cdm$indications) |>
select("cohort_definition_id", "cohort_name") |>
inner_join(cohortCount(cdm$indications), by = "cohort_definition_id")# A tibble: 2 × 4
cohort_definition_id cohort_name number_records number_subjects
<int> <chr> <int> <int>
1 1 sinusitis 18629 2688
2 2 bronchitis 8232 2546Indication
We have to define a gap (period before index date) that we are interested to assess indication:
Indication
cdm$dus_new_user <- cdm$dus_new_user |>
addIndication(
indicationCohortName = "indications",
unknownIndicationTable = c("condition_occurrence"),
indicationGap = c(0, 30, Inf)
)
glimpse(cdm$dus_new_user)Rows: ??
Columns: 16
Database: DuckDB v0.10.0 [martics@Windows 10 x64:R 4.2.3/C:\Users\martics\AppData\Local\Temp\Rtmp6LHYNc\filec9d465b41608.duckdb]
$ cohort_definition_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,…
$ subject_id <int> 5, 9, 17, 30, 114, 115, 143, 154, 162, 177, 184, 213, 225, 264, 273, 287, 332, 3…
$ cohort_start_date <date> 2003-09-10, 2003-02-23, 2012-11-25, 2004-11-28, 2009-11-30, 2001-06-30, 2000-08…
$ cohort_end_date <date> 2003-09-17, 2003-03-09, 2012-12-09, 2004-12-05, 2009-12-07, 2001-07-14, 2000-08…
$ indication_gap_0_bronchitis <dbl> 1, 0, 1, 0, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 1, 0, 1,…
$ indication_gap_0_sinusitis <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,…
$ indication_gap_0_none <dbl> 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0,…
$ indication_gap_0_unknown <dbl> 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 1, 1, 0, 0, 1, 0,…
$ indication_gap_30_sinusitis <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,…
$ indication_gap_30_bronchitis <dbl> 1, 0, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0, 0, 0, 1, 1, 0, 1,…
$ indication_gap_30_none <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,…
$ indication_gap_30_unknown <dbl> 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 1, 1, 0, 0, 1, 0,…
$ indication_gap_inf_bronchitis <dbl> 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,…
$ indication_gap_inf_sinusitis <dbl> 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,…
$ indication_gap_inf_none <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,…
$ indication_gap_inf_unknown <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,…Indication
We can join the indications if we need later to stratify by them:
cdm$dus_new_user |>
indicationToStrata()# Source: SQL [?? x 7]
# Database: DuckDB v0.10.0 [martics@Windows 10 x64:R 4.2.3/C:\Users\martics\AppData\Local\Temp\Rtmp6LHYNc\filec9d465b41608.duckdb]
cohort_definition_id subject_id cohort_start_date cohort_end_date indication_gap_0 indication_gap_30
<int> <int> <date> <date> <chr> <chr>
1 1 5 2003-09-10 2003-09-17 Bronchitis Bronchitis
2 1 9 2003-02-23 2003-03-09 Unknown Unknown
3 1 17 2012-11-25 2012-12-09 Bronchitis Bronchitis
4 1 30 2004-11-28 2004-12-05 None Bronchitis
5 1 114 2009-11-30 2009-12-07 Bronchitis Bronchitis
6 1 115 2001-06-30 2001-07-14 Bronchitis Bronchitis
7 1 143 2000-08-16 2000-08-23 Bronchitis Bronchitis
8 1 154 2017-07-15 2017-07-20 Bronchitis Bronchitis
9 1 162 2001-06-25 2001-07-02 Bronchitis Bronchitis
10 1 177 2002-05-14 2002-05-14 None Unknown
# ℹ more rows
# ℹ 1 more variable: indication_gap_inf <chr>Indication
Summarise indication into a summarisedResult object:
cdm$dus_new_user |>
summariseIndication() |>
select("variable_name", "variable_level", "estimate_name", "estimate_value") |>
print(n = Inf)# A tibble: 26 × 4
variable_name variable_level estimate_name estimate_value
<chr> <chr> <chr> <chr>
1 number records <NA> count 1979
2 number subjects <NA> count 1979
3 Indication on index date Bronchitis count 1373
4 Indication on index date Bronchitis percentage 69.3784739767559
5 Indication on index date Sinusitis count <NA>
6 Indication on index date Sinusitis percentage <NA>
7 Indication on index date None count 48
8 Indication on index date None percentage 2.42546740778171
9 Indication on index date Unknown count 558
10 Indication on index date Unknown percentage 28.1960586154624
11 Indication during prior 30 days Sinusitis count 24
12 Indication during prior 30 days Sinusitis percentage 1.21273370389085
13 Indication during prior 30 days Bronchitis count 1395
14 Indication during prior 30 days Bronchitis percentage 70.4901465386559
15 Indication during prior 30 days None count <NA>
16 Indication during prior 30 days None percentage <NA>
17 Indication during prior 30 days Unknown count 580
18 Indication during prior 30 days Unknown percentage 29.3077311773623
19 Indication any time prior Bronchitis count 1902
20 Indication any time prior Bronchitis percentage 96.1091460333502
21 Indication any time prior Sinusitis count 1958
22 Indication any time prior Sinusitis percentage 98.9388580090955
23 Indication any time prior None count 0
24 Indication any time prior None percentage 0
25 Indication any time prior Unknown count <NA>
26 Indication any time prior Unknown percentage <NA> Alternative treatments
You can summarise other treatments after a certain cohort, for example we have our cohort of acetaminophen, which are the treatments that are used.
getDrugIngredientCodes(cdm = cdm)$ondansetron
[1] 35605482 1000560
$meperidine
[1] 1102527 40165015
$loratadine
[1] 1107830 1107882 19125062
$morphine
[1] 35605858 1110410
$aspirin
[1] 19059056 1112807
$naproxen
[1] 1115008 19019273 1115171
$celecoxib
[1] 1118084
$dextromethorphan
[1] 1119510 40229134
$diclofenac
[1] 1124300
$oxycodone
[1] 1124957 40231925
$acetaminophen
[1] 1125315 1127433 40229134 40231925 40162522 19133768 1127078
$dornase_alfa
[1] 19029476 1125443
$diphenhydramine
[1] 1129625 40232448
$salmeterol
[1] 1137529 40169281
$cetirizine
[1] 1149196 40228214 40228230
$fluticasone
[1] 1149380 40169216 40169281
$astemizole
[1] 1150770 1150771
$terfenadine
[1] 1150836 1150837
$fexofenadine
[1] 1153428 40223821 40223834
$fentanyl
[1] 1154029
$albuterol
[1] 1154615 1154343
$hydrocodone
[1] 40162522 1174888 19133768
$ibuprofen
[1] 19019979 19078461 1177480
$chlorpheniramine
[1] 43012036 1192710 19112599
$methotrexate
[1] 19007333 1305058
$verapamil
[1] 40171902 1307863
$vitamin_b_12
[1] 964261 1308738
$amiodarone
[1] 1310034 1309944
$warfarin
[1] 1310149 40163554
$clopidogrel
[1] 1322184 19075601
$digoxin
[1] 1326303 19018935
$amlodipine
[1] 1332419 1332418
$epinephrine
[1] 46275916 1343916 1594382
$alteplase
[1] 40220960 1347450
$nitroglycerin
[1] 1361711 19126352
$heparin
[1] 1367571
$ferrous_sulfate
[1] 19077344 1396131
$medroxyprogesterone
[1] 40224132 40224805 1500211
$levothyroxine
[1] 40174880 1501700
$mestranol
[1] 1503184 19128065
$drospirenone
[1] 1512674 19127904
$norgestimate
[1] 19128263 19128022 1515774
$norelgestromin
[1] 44816277 1518198
$etonogestrel
[1] 43012433 1519937 1519936
$norethindrone
[1] 42902245 19133679 19127922 19128065 1521369 19129144
$simvastatin
[1] 1539411 1539403
$atorvastatin
[1] 1545959 1545958
$estradiol
[1] 40221415 1548195
$ethinyl_estradiol
[1] 42902245 43012433 44816277 19128263 19127904 19127918 19128009 19128022 1549786
$prednisone
[1] 1551192 19089061 1551099
$alendronate
[1] 1557272 40173590
$levonorgestrel
[1] 1718412 46221626 1589505 19127918 19128009 19132304 43012754
$ferrous_fumarate
[1] 42902245 1595799
$amoxicillin
[1] 19073183 19073188 1713671 1713332
$ampicillin
[1] 19129655 1717327
$penicillin_g
[1] 1728416 19006318
$penicillin_v
[1] 1729720 19133873 19133905
$doxycycline
[1] 1738521 46233710 46233988
$tazobactam
[1] 46275444 1741122
$piperacillin
[1] 46275444 1746114
$clavulanate
[1] 1713671 1759842
$cefaclor
[1] 19074841 19074843 1768849
$cefuroxime
[1] 19075001 1778162
$desflurane
[1] 19023423 19002770
$rocuronium
[1] 19003953 42707627
$cyclosporine
[1] 19010482 19134077
$remifentanil
[1] 19016749
$sevoflurane
[1] 19039298 19023398
$dienogest
[1] 40221415 19054876
$alfentanil
[1] 19059528
$sufentanil
[1] 19078219
$inert_ingredients
[1] 40221415 19128263 19127904 19127918 19128022 19127890 19128065
$memantine
[1] 701322 40222846 45892894
$methylphenidate
[1] 705944 40236446
$midazolam
[1] 708298 19078924
$donepezil
[1] 40223774 40223768 45892894 715997
$diazepam
[1] 723013 19076374 19115351
$doxylamine
[1] 40229134 738818
$carbamazepine
[1] 19074679 740275
$atomoxetine
[1] 742185 19123231
$propofol
[1] 753626 40220386
$galantamine
[1] 757627 19077572
$isoflurane
[1] 782043 782047
$lorazepam
[1] 791967 19057271
$tacrine
[1] 836654 836659
$atropine
[1] 40243436 914335
$nitrofurantoin
[1] 920293 920300 920334
$pancreatin
[1] 926487 19045231
$phenazopyridine
[1] 933724 40236824
$sodium_chloride
[1] 19079524 967823
$hydrocortisone
[1] 19008572 975125Alternative treatments
x <- cdm$dus_new_user |>
summariseTreatmentFromConceptSet(
treatmentConceptSet = getDrugIngredientCodes(cdm = cdm),
window = list(c(0, 90), c(91, 180), c(181, 270), c(271, 360))
)
x |>
select("variable_name", "estimate_name", "estimate_value", "additional_name", "additional_level")# A tibble: 736 × 5
variable_name estimate_name estimate_value additional_name additional_level
<chr> <chr> <chr> <chr> <chr>
1 acetaminophen count 1979 window 0 to 90
2 acetaminophen percentage 100 window 0 to 90
3 albuterol count 2 window 0 to 90
4 albuterol percentage 0.101061141990904 window 0 to 90
5 alendronate count 1 window 0 to 90
6 alendronate percentage 0.0505305709954522 window 0 to 90
7 alfentanil count 1 window 0 to 90
8 alfentanil percentage 0.0505305709954522 window 0 to 90
9 alteplase count 0 window 0 to 90
10 alteplase percentage 0 window 0 to 90
# ℹ 726 more rowsAlternative treatments
library(ggplot2)
xx <- x |>
filter(estimate_name == "percentage") |>
mutate(estimate_value = as.numeric(estimate_value)) |>
select(variable_name, window = additional_level, estimate_value) |>
mutate(window = factor(window, levels = c("0 to 90", "91 to 180", "181 to 270", "271 to 360")))
xx |>
filter(variable_name %in% unique(xx$variable_name[xx$estimate_value > 1])) |>
ggplot(aes(x = variable_name, y = estimate_value, color = variable_name, fill = variable_name)) +
geom_bar(stat="identity") +
coord_flip() +
facet_grid( ~ window) +
theme(legend.position = "none")Alternative treatments
Treatment discontinuation
We are working on a treatment discontinuation where we use the CohortSurvival package that we will see later.
Oxford Summer School 2024