,
Marti Catala [aut] ,
Edward Burn [aut] ,
Nuria Mercade-Besora [aut] ,
Xihang Chen [aut] | Title: | Creation of Mock Observational Medical Outcomes Partnership Common Data Model |
|---|---|
| Description: | Creates mock data for testing and package development for the Observational Medical Outcomes Partnership common data model. The package offers functions crafted with pipeline-friendly implementation, enabling users to effortlessly include only the necessary tables for their testing needs. |
| Authors: | Mike Du [aut, cre] ,
Marti Catala [aut] ,
Edward Burn [aut] ,
Nuria Mercade-Besora [aut] ,
Xihang Chen [aut] |
| Maintainer: | Mike Du <[email protected]> |
| License: | Apache License (>= 2) |
| Version: | 0.3.1.9000 |
| Built: | 2024-10-25 16:20:24 UTC |
| Source: | https://github.com/ohdsi/omock |
This function takes an existing CDM reference (which can be empty) and a list of additional named tables to create a more complete mock CDM object. It ensures that all provided observations fit within their respective observation periods and that all individual records are consistent with the entries in the person table. This is useful for creating reliable and realistic healthcare data simulations for development and testing within the OMOP CDM framework.
mockCdmFromTables(cdm = mockCdmReference(), tables = list(), seed = NULL)mockCdmFromTables(cdm = mockCdmReference(), tables = list(), seed = NULL)
cdm |
A 'cdm_reference' object, which serves as the base structure where all additional tables will be integrated. This parameter should already be initialized and can contain pre-existing standard or cohort-specific OMOP tables. |
tables |
A named list of data frames representing additional tables to be integrated into the CDM. These tables can include both standard OMOP tables such as 'drug_exposure' or 'condition_occurrence', as well as cohort-specific tables that are not part of the standard OMOP model but are necessary for specific analyses. Each table should be named according to its intended table name in the CDM structure. |
seed |
An optional integer that sets the seed for random number generation used in creating mock data entries. Setting a seed ensures that the generated mock data are reproducible across different runs of the function. If 'NULL', the seed is not set, leading to non-deterministic behavior in data generation. |
Returns the updated 'cdm' object with all the new tables added and integrated, ensuring consistency across the observational periods and the person entries.
library(omock) library(dplyr) # Create a mock cohort table cohort <- tibble( cohort_definition_id = c(1, 1, 2, 2, 1, 3, 3, 3, 1, 3), subject_id = c(1, 4, 2, 3, 5, 5, 4, 3, 3, 1), cohort_start_date = as.Date(c( "2020-04-01", "2021-06-01", "2022-05-22", "2010-01-01", "2019-08-01", "2019-04-07", "2021-01-01", "2008-02-02", "2009-09-09", "2021-01-01" )), cohort_end_date = cohort_start_date ) # Generate a mock CDM from preexisting CDM structure and cohort table cdm <- mockCdmFromTables(cdm = mockCdmReference(), tables = list(cohort = cohort)) # Access the newly integrated cohort table and the standard person table in the CDM print(cdm$cohort) print(cdm$person)library(omock) library(dplyr) # Create a mock cohort table cohort <- tibble( cohort_definition_id = c(1, 1, 2, 2, 1, 3, 3, 3, 1, 3), subject_id = c(1, 4, 2, 3, 5, 5, 4, 3, 3, 1), cohort_start_date = as.Date(c( "2020-04-01", "2021-06-01", "2022-05-22", "2010-01-01", "2019-08-01", "2019-04-07", "2021-01-01", "2008-02-02", "2009-09-09", "2021-01-01" )), cohort_end_date = cohort_start_date ) # Generate a mock CDM from preexisting CDM structure and cohort table cdm <- mockCdmFromTables(cdm = mockCdmReference(), tables = list(cohort = cohort)) # Access the newly integrated cohort table and the standard person table in the CDM print(cdm$cohort) print(cdm$person)
This function initializes an empty CDM reference with a specified name and populates it with mock vocabulary tables based on the provided vocabulary set. It is particularly useful for setting up a simulated environment for testing and development purposes within the OMOP CDM framework.
mockCdmReference(cdmName = "mock database", vocabularySet = "mock")mockCdmReference(cdmName = "mock database", vocabularySet = "mock")
cdmName |
A character string specifying the name of the CDM object to be created.This name can be used to identify the CDM object within a larger simulation or testing framework. Default is "mock database". |
vocabularySet |
A character string that specifies the name of the vocabulary set to be used in creating the vocabulary tables for the CDM. This allows for the customization of the vocabulary to match specific testing scenarios. Default is "mock". |
Returns a CDM object that is initially empty but includes mock vocabulary tables.The object structure is compliant with OMOP CDM standards, making it suitable for further population with mock data like person, visit, and observation records.
library(omock) # Create a new empty mock CDM reference cdm <- mockCdmReference() # Display the structure of the newly created CDM print(cdm)library(omock) # Create a new empty mock CDM reference cdm <- mockCdmReference() # Display the structure of the newly created CDM print(cdm)
This function generates synthetic cohort data and adds it to a given CDM (Common Data Model) reference. It allows for creating multiple cohorts with specified properties and simulates the frequency of observations for individuals.
mockCohort( cdm, name = "cohort", numberCohorts = 1, cohortName = paste0("cohort_", seq_len(numberCohorts)), recordPerson = 1, seed = NULL )mockCohort( cdm, name = "cohort", numberCohorts = 1, cohortName = paste0("cohort_", seq_len(numberCohorts)), recordPerson = 1, seed = NULL )
cdm |
A CDM reference object where the synthetic cohort data will be stored. This object should already include necessary tables such as 'person' and 'observation_period'. |
name |
A string specifying the name of the table within the CDM where the cohort data will be stored. Defaults to "cohort". This name will be used to reference the new table in the CDM. |
numberCohorts |
An integer specifying the number of different cohorts to create within the table. Defaults to 1. This parameter allows for the creation of multiple cohorts, each with a unique identifier. |
cohortName |
A character vector specifying the names of the cohorts to be created. If not provided, default names based on a sequence (e.g., "cohort_1", "cohort_2", ...) will be generated. The length of this vector must match the value of 'numberCohorts'. This parameter provides meaningful names for each cohort. |
recordPerson |
An integer or a vector of integers specifying the expected number of records per person within each cohort. If a single integer is provided, it applies to all cohorts. If a vector is provided, its length must match the value of 'numberCohorts'. This parameter helps simulate the frequency of observations for individuals in each cohort, allowing for realistic variability in data. |
seed |
An integer specifying the random seed for reproducibility of the generated data. Setting a seed ensures that the same synthetic data can be generated again, facilitating consistent results across different runs. |
A CDM reference object with the mock cohort tables added. The new table will contain synthetic data representing the specified cohorts, each with its own set of observation records.
library(omock) cdm <- mockCdmReference() |> mockPerson(nPerson = 100) |> mockObservationPeriod() |> mockCohort( name = "omock_example", numberCohorts = 2, cohortName = c("omock_cohort_1", "omock_cohort_2") ) cdmlibrary(omock) cdm <- mockCdmReference() |> mockPerson(nPerson = 100) |> mockObservationPeriod() |> mockCohort( name = "omock_example", numberCohorts = 2, cohortName = c("omock_cohort_1", "omock_cohort_2") ) cdm
This function inserts new concept entries into a specified domain within the concept table of a CDM object.It supports four domains: Condition, Drug, Measurement, and Observation. Existing entries with the same concept IDs will be overwritten, so caution should be used when adding data to prevent unintended data loss.
mockConcepts(cdm, conceptSet, domain = "Condition", seed = NULL)mockConcepts(cdm, conceptSet, domain = "Condition", seed = NULL)
cdm |
A CDM object that represents a common data model containing at least a concept table.This object will be modified in-place to include the new or updated concept entries. |
conceptSet |
A numeric vector of concept IDs to be added or updated in the concept table.These IDs should be unique within the context of the provided domain to avoid unintended overwriting unless that is the intended effect. |
domain |
A character string specifying the domain of the concepts being added.Only accepts "Condition", "Drug", "Measurement", or "Observation". This defines under which category the concepts fall and affects which vocabulary is used for them. |
seed |
An optional integer value used to set the random seed for generating reproducible concept attributes like 'vocabulary_id' and 'concept_class_id'. Useful for testing or when consistent output is required. |
Returns the modified CDM object with the updated concept table reflecting the newly added concepts.The function directly modifies the provided CDM object.
library(omock) library(dplyr) # Create a mock CDM reference and add concepts in the 'Condition' domain cdm <- mockCdmReference() |> mockConcepts( conceptSet = c(100, 200), domain = "Condition") # View the updated concept entries for the 'Condition' domain cdm$concept |> filter(domain_id == "Condition")library(omock) library(dplyr) # Create a mock CDM reference and add concepts in the 'Condition' domain cdm <- mockCdmReference() |> mockConcepts( conceptSet = c(100, 200), domain = "Condition") # View the updated concept entries for the 'Condition' domain cdm$concept |> filter(domain_id == "Condition")
This function simulates condition occurrences for individuals within a specified cohort. It helps create a realistic dataset by generating condition records for each person, based on the number of records specified per person.The generated data are aligned with the existing observation periods to ensure that all conditions are recorded within valid observation windows.
mockConditionOccurrence(cdm, recordPerson = 1, seed = NULL)mockConditionOccurrence(cdm, recordPerson = 1, seed = NULL)
cdm |
A ‘cdm_reference' object that should already include ’person', 'observation_period', and 'concept' tables.This object is the base CDM structure where the condition occurrence data will be added. It is essential that these tables are not empty as they provide the necessary context for generating condition data. |
recordPerson |
An integer specifying the expected number of condition records to generate per person.This parameter allows the simulation of varying frequencies of condition occurrences among individuals in the cohort, reflecting the variability seen in real-world medical data. |
seed |
An optional integer used to set the seed for random number generation, ensuring reproducibility of the generated data.If provided, it allows the function to produce the same results each time it is run with the same inputs.If 'NULL', the seed is not set, resulting in different outputs on each run. |
Returns the modified 'cdm' object with the new 'condition_occurrence' table added. This table includes the simulated condition data for each person, ensuring that each record is within the valid observation periods and linked to the correct individuals in the 'person' table.
library(omock) # Create a mock CDM reference and add condition occurrences cdm <- mockCdmReference() |> mockPerson() |> mockObservationPeriod() |> mockConditionOccurrence(recordPerson = 2) # View the generated condition occurrence data print(cdm$condition_occurrence)library(omock) # Create a mock CDM reference and add condition occurrences cdm <- mockCdmReference() |> mockPerson() |> mockObservationPeriod() |> mockConditionOccurrence(recordPerson = 2) # View the generated condition occurrence data print(cdm$condition_occurrence)
This function simulates death records for individuals within a specified cohort. It creates a realistic dataset by generating death records according to the specified number of records per person. The function ensures that each death record is associated with a valid person within the observation period to maintain the integrity of the data.
mockDeath(cdm, recordPerson = 1, seed = NULL)mockDeath(cdm, recordPerson = 1, seed = NULL)
cdm |
A ‘cdm_reference' object that must already include ’person' and 'observation_period' tables.This object is the base CDM structure where the death data will be added. It is essential that the 'person' and 'observation_period' tables are populated as they provide necessary context for generating death records. |
recordPerson |
An integer specifying the expected number of death records to generate per person. This parameter helps simulate varying frequencies of death occurrences among individuals in the cohort, reflecting the variability seen in real-world medical data. Typically, this would be set to 1 or 0, assuming most datasets would only record a single death date per individual if at all. |
seed |
An optional integer used to set the seed for random number generation, ensuring reproducibility of the generated data. If provided, it allows the function to produce the same results each time it is run with the same inputs. If 'NULL', the seed is not set, which can result in different outputs on each run. |
Returns the modified ‘cdm' object with the new ’death' table added. This table includes the simulated death data for each person, ensuring that each record is linked correctly to individuals in the ' person' table and falls within valid observation periods.
library(omock) # Create a mock CDM reference and add death records cdm <- mockCdmReference() |> mockPerson() |> mockObservationPeriod() |> mockDeath(recordPerson = 1) # View the generated death data print(cdm$death)library(omock) # Create a mock CDM reference and add death records cdm <- mockCdmReference() |> mockPerson() |> mockObservationPeriod() |> mockDeath(recordPerson = 1) # View the generated death data print(cdm$death)
This function simulates drug exposure records for individuals within a specified cohort. It creates a realistic dataset by generating drug exposure records based on the specified number of records per person. Each drug exposure record is correctly associated with an individual within valid observation periods, ensuring the integrity of the data.
mockDrugExposure(cdm, recordPerson = 1, seed = NULL)mockDrugExposure(cdm, recordPerson = 1, seed = NULL)
cdm |
A ‘cdm_reference' object that must already include ’person' and 'observation_period' tables. This object serves as the base CDM structure where the drug exposure data will be added. The 'person' and 'observation_period' tables must be populated as they are necessary for generating accurate drug exposure records. |
recordPerson |
An integer specifying the expected number of drug exposure records to generate per person. This parameter allows for the simulation of varying drug usage frequencies among individuals in the cohort, reflecting real-world variability in medication administration. |
seed |
An optional integer used to set the seed for random number generation, ensuring reproducibility of the generated data. If provided, this seed enables the function to produce consistent results each time it is run with the same inputs. If 'NULL', the seed is not set, which can lead to different outputs on each run. |
Returns the modified ‘cdm' object with the new ’drug_exposure' table added. This table includes the simulated drug exposure data for each person, ensuring that each record is correctly linked to individuals in the 'person' table and falls within valid observation periods.
library(omock) # Create a mock CDM reference and add drug exposure records cdm <- mockCdmReference() |> mockPerson() |> mockObservationPeriod() |> mockDrugExposure(recordPerson = 3) # View the generated drug exposure data print(cdm$drug_exposure)library(omock) # Create a mock CDM reference and add drug exposure records cdm <- mockCdmReference() |> mockPerson() |> mockObservationPeriod() |> mockDrugExposure(recordPerson = 3) # View the generated drug exposure data print(cdm$drug_exposure)
This function simulates measurement records for individuals within a specified cohort. It creates a realistic dataset by generating measurement records based on the specified number of records per person. Each measurement record is correctly associated with an individual within valid observation periods, ensuring the integrity of the data.
mockMeasurement(cdm, recordPerson = 1, seed = NULL)mockMeasurement(cdm, recordPerson = 1, seed = NULL)
cdm |
A ‘cdm_reference' object that must already include ’person' and 'observation_period' tables. This object serves as the base CDM structure where the measurement data will be added. The 'person' and 'observation_period' tables must be populated as they are necessary for generating accurate measurement records. |
recordPerson |
An integer specifying the expected number of measurement records to generate per person. This parameter allows for the simulation of varying frequencies of health measurements among individuals in the cohort, reflecting real-world variability in patient monitoring and diagnostic testing. |
seed |
An optional integer used to set the seed for random number generation, ensuring reproducibility of the generated data. If provided, this seed enables the function to produce consistent results each time it is run with the same inputs. If 'NULL', the seed is not set, which can lead to different outputs on each run. |
Returns the modified ‘cdm' object with the new ’measurement' table added. This table includes the simulated measurement data for each person, ensuring that each record is correctly linked to individuals in the 'person' table and falls within valid observation periods.
library(omock) # Create a mock CDM reference and add measurement records cdm <- mockCdmReference() |> mockPerson() |> mockObservationPeriod() |> mockMeasurement(recordPerson = 5) # View the generated measurement data print(cdm$measurement)library(omock) # Create a mock CDM reference and add measurement records cdm <- mockCdmReference() |> mockPerson() |> mockObservationPeriod() |> mockMeasurement(recordPerson = 5) # View the generated measurement data print(cdm$measurement)
This function simulates observation records for individuals within a specified cohort. It creates a realistic dataset by generating observation records based on the specified number of records per person. Each observation record is correctly associated with an individual within valid observation periods, ensuring the integrity of the data.
mockObservation(cdm, recordPerson = 1, seed = NULL)mockObservation(cdm, recordPerson = 1, seed = NULL)
cdm |
A ‘cdm_reference' object that must already include ’person', 'observation_period', and 'concept' tables. This object serves as the base CDM structure where the observation data will be added. The 'person' and 'observation_period' tables must be populated as they are necessary for generating accurate observation records. |
recordPerson |
An integer specifying the expected number of observation records to generate per person. This parameter allows for the simulation of varying frequencies of healthcare observations among individuals in the cohort, reflecting real-world variability in patient monitoring and health assessments. |
seed |
An optional integer used to set the seed for random number generation, ensuring reproducibility of the generated data. If provided, this seed enables the function to produce consistent results each time it is run with the same inputs. If 'NULL', the seed is not set, which can lead to different outputs on each run. |
Returns the modified ‘cdm' object with the new ’observation' table added. This table includes the simulated observation data for each person, ensuring that each record is correctly linked to individuals in the 'person' table and falls within valid observation periods.
library(omock) # Create a mock CDM reference and add observation records cdm <- mockCdmReference() |> mockPerson() |> mockObservationPeriod() |> mockObservation(recordPerson = 3) # View the generated observation data print(cdm$observation)library(omock) # Create a mock CDM reference and add observation records cdm <- mockCdmReference() |> mockPerson() |> mockObservationPeriod() |> mockObservation(recordPerson = 3) # View the generated observation data print(cdm$observation)
This function simulates observation periods for individuals based on their date of birth recorded in the 'person' table of the CDM object. It assigns random start and end dates for each observation period within a realistic timeframe up to a specified or default maximum date.
mockObservationPeriod(cdm, seed = NULL)mockObservationPeriod(cdm, seed = NULL)
cdm |
A ‘cdm_reference' object that must include a ’person' table with valid dates of birth. This object serves as the base CDM structure where the observation period data will be added. The function checks to ensure that the 'person' table is populated and uses the date of birth to generate observation periods. |
seed |
An optional integer used to set the seed for random number generation, ensuring reproducibility of the generated data. If provided, this seed allows the function to produce consistent results each time it is run with the same inputs. If 'NULL', the seed is not set, which can lead to different outputs on each run. |
Returns the modified ‘cdm' object with the new ’observation_period' table added. This table includes the simulated observation periods for each person, ensuring that each record spans a realistic timeframe based on the person's date of birth.
library(omock) # Create a mock CDM reference and add observation periods cdm <- mockCdmReference() |> mockPerson(nPerson = 100) |> mockObservationPeriod() # View the generated observation period data print(cdm$observation_period)library(omock) # Create a mock CDM reference and add observation periods cdm <- mockCdmReference() |> mockPerson(nPerson = 100) |> mockObservationPeriod() # View the generated observation period data print(cdm$observation_period)
This function creates a mock person table with specified characteristics for each individual, including a randomly assigned date of birth within a given range and gender based on specified proportions. It populates the CDM object's person table with these entries, ensuring each record is uniquely identified.
mockPerson( cdm = mockCdmReference(), nPerson = 10, birthRange = as.Date(c("1950-01-01", "2000-12-31")), proportionFemale = 0.5, seed = NULL )mockPerson( cdm = mockCdmReference(), nPerson = 10, birthRange = as.Date(c("1950-01-01", "2000-12-31")), proportionFemale = 0.5, seed = NULL )
cdm |
A 'cdm_reference' object that serves as the base structure for adding the person table. This parameter should be an existing or newly created CDM object that does not yet contain a 'person' table. |
nPerson |
An integer specifying the number of mock persons to create in the person table. This defines the scale of the simulation and allows for the creation of datasets with varying sizes. |
birthRange |
A date range within which the birthdays of the mock persons will be randomly generated. This should be provided as a vector of two dates ('as.Date' format), specifying the start and end of the range. |
proportionFemale |
A numeric value between 0 and 1 indicating the proportion of the persons who are female. For example, a value of 0.5 means approximately 50 the generated persons will be female. This helps simulate realistic demographic distributions. |
seed |
An optional integer used to set the seed for random number generation, ensuring reproducibility of the generated data. If provided, this seed allows the function to produce consistent results each time it is run with the same inputs. If 'NULL', the seed is not set, which can lead to different outputs on each run. |
A modified ‘cdm' object with the new ’person' table added. This table includes simulated person data for each generated individual, with unique identifiers and demographic attributes.
library(omock) cdm <- mockPerson(cdm = mockCdmReference(), nPerson = 10) # View the generated person data print(cdm$person)library(omock) cdm <- mockPerson(cdm = mockCdmReference(), nPerson = 10) # View the generated person data print(cdm$person)
This function simulates condition occurrences for individuals within a specified cohort. It helps create a realistic dataset by generating condition records for each person, based on the number of records specified per person.The generated data are aligned with the existing observation periods to ensure that all conditions are recorded within valid observation windows.
mockProcedureOccurrence(cdm, recordPerson = 1, seed = NULL)mockProcedureOccurrence(cdm, recordPerson = 1, seed = NULL)
cdm |
A ‘cdm_reference' object that should already include ’person', 'observation_period', and 'concept' tables.This object is the base CDM structure where the procedure occurrence data will be added. It is essential that these tables are not empty as they provide the necessary context for generating condition data. |
recordPerson |
An integer specifying the expected number of condition records to generate per person.This parameter allows the simulation of varying frequencies of condition occurrences among individuals in the cohort, reflecting the variability seen in real-world medical data. |
seed |
An optional integer used to set the seed for random number generation, ensuring reproducibility of the generated data.If provided, it allows the function to produce the same results each time it is run with the same inputs.If 'NULL', the seed is not set, resulting in different outputs on each run. |
Returns the modified 'cdm' object with the new 'condition_occurrence' table added. This table includes the simulated condition data for each person, ensuring that each record is within the valid observation periods and linked to the correct individuals in the 'person' table.
library(omock) # Create a mock CDM reference and add condition occurrences cdm <- mockCdmReference() |> mockPerson() |> mockObservationPeriod() |> mockProcedureOccurrence(recordPerson = 2) # View the generated condition occurrence data print(cdm$procedure_occurrence)library(omock) # Create a mock CDM reference and add condition occurrences cdm <- mockCdmReference() |> mockPerson() |> mockObservationPeriod() |> mockProcedureOccurrence(recordPerson = 2) # View the generated condition occurrence data print(cdm$procedure_occurrence)
Function to generate visit occurrence table
mockVisitOccurrence(cdm, seed = NULL)mockVisitOccurrence(cdm, seed = NULL)
cdm |
the CDM reference into which the mock visit occurrence table will be added |
seed |
A random seed to ensure reproducibility of the generated data. |
A cdm reference with the visit_occurrence tables added
library(omock)library(omock)
This function adds specified vocabulary tables to a CDM object. It can either populate the tables with provided data frames or initialize empty tables if no data is provided. This is useful for setting up a testing environment with controlled vocabulary data.
mockVocabularyTables( cdm = mockCdmReference(), vocabularySet = "mock", cdmSource = NULL, concept = NULL, vocabulary = NULL, conceptRelationship = NULL, conceptSynonym = NULL, conceptAncestor = NULL, drugStrength = NULL )mockVocabularyTables( cdm = mockCdmReference(), vocabularySet = "mock", cdmSource = NULL, concept = NULL, vocabulary = NULL, conceptRelationship = NULL, conceptSynonym = NULL, conceptAncestor = NULL, drugStrength = NULL )
cdm |
A 'cdm_reference' object that serves as the base structure for adding vocabulary tables. This should be an existing or a newly created CDM object, typically initialized without any vocabulary tables. |
vocabularySet |
A character string that specifies a prefix or a set name used to initialize mock data tables. This allows for customization of the source data or structure names when generating vocabulary tables. |
cdmSource |
An optional data frame representing the CDM source table. If provided, it will be used directly; otherwise, a mock table will be generated based on the 'vocabularySet' prefix. |
concept |
An optional data frame representing the concept table. If provided, it will be used directly; if NULL, a mock table will be generated. |
vocabulary |
An optional data frame representing the vocabulary table. If provided, it will be used directly; if NULL, a mock table will be generated. |
conceptRelationship |
An optional data frame representing the concept relationship table. If provided, it will be used directly; if NULL, a mock table will be generated. |
conceptSynonym |
An optional data frame representing the concept synonym table. If provided, it will be used directly; if NULL, a mock table will be generated. |
conceptAncestor |
An optional data frame representing the concept ancestor table. If provided, it will be used directly; if NULL, a mock table will be generated. |
drugStrength |
An optional data frame representing the drug strength table. If provided, it will be used directly; if NULL, a mock table will be generated. |
Returns the modified 'cdm' object with the new or provided vocabulary tables added.
library(omock) # Create a mock CDM reference and populate it with mock vocabulary tables cdm <- mockCdmReference() |> mockVocabularyTables(vocabularySet = "mock") # View the names of the tables added to the CDM names(cdm)library(omock) # Create a mock CDM reference and populate it with mock vocabulary tables cdm <- mockCdmReference() |> mockVocabularyTables(vocabularySet = "mock") # View the names of the tables added to the CDM names(cdm)