Concordance: Clinical Events and Specialties – PCORnet — cnc_sp_process

This is a check that will assess quality of specialty data in a study sample The user will provide a clinical codeset of interest with an associated domain and will be able to stratify results by: visit type (user will provide groupings) time

Usage

cnc_sp_process_pcornet(
  cohort,
  multi_or_single_site = "multi",
  age_groups = NULL,
  codeset_tbl = NULL,
  care_site,
  provider,
  visit_type_tbl = NULL,
  time = FALSE,
  time_span = c("2012-01-01", "2020-01-01"),
  time_period = "year",
  vocab_tbl = NULL
)

Arguments

cohort

A dataframe with the cohort of patients for your study. Should include the columns: site | person_id | start_date | end_date

multi_or_single_site

Option to run the function on a single vs multiple sites - 'single': run on a single site, or treat all of the sites as one - 'multi': run on a group of sites, treating each site separately

age_groups

If you would like to stratify the results by age group, create a table or CSV file with the following columns and include it as the age_groups function parameter:

min_age: the minimum age for the group (i.e. 10)
max_age: the maximum age for the group (i.e. 20)
group: a string label for the group (i.e. 10-20, Young Adult, etc.)

If you would not like to stratify by age group, leave the argument as NULL

codeset_tbl

table in the specs directory with the columns:

domain: name of the domain
domain_tbl: name of the cdm_tbl
concept_field: column name in the domain_tbl for which to search the codeset concept_ids
date_field: column name in the domain_tbl to be used for time-based filtering
codeset_name: name of a codeset that exists as a csv file in the specs directory. The codeset can optionally contain a cluster column specifying subgroups of the codeset, and if so, the results will be stratified by cluster

care_site

TRUE if want to look at care_site specialty FALSE if do not want to look at care_site specialty

provider

TRUE if want to look at provider specialty FALSE if do not want to look at provider specialty IF both provider and care_site are both TRUE, provider specialty will be prioritized if provider and care_site are discordant for the visit

visit_type_tbl

a csv file that defines available visit types that are called in visit_types. defaults to the provided conc_visit_types.csv file, which contains the following fields: - visit_concept_id: the visit_concept_id that represents the visit type of interest (i.e. 9201) - visit_type: the string label to describe the visit type; this label can be used multiple times within the file if multiple visit_concept_ids represent the visit type
```
                    This CSV can be altered to fit the users needs, or another csv with the same columns and formatting can be supplied.
```

time

TRUE if results should be over time. Defaults to FALSE

time_span

if time=TRUE, vector containing minimum and maximum dates over which to measure

time_period

if time=TRUE, indicates time period (e.g. 'year', 'month') over which to measure

vocab_tbl

location of vocabulary table containing concept_id to concept_name mapping. If a vocabulary table is not available, will default to NULL

Value

2 tables: 1 table stored as a csv file in the results directory containing all of the specialties in the results of the DQ check, with the columns: - specialty_concept_id: an identifier for the specialty based on the data model - specialty_concept_name: if a vocab_tbl is provided, the name of the specialty that corresponds to each specialty_concept_id. If no vocab_tbl is provided, defaults to 'No vocabulary table input' 1 table containing counts of visits, optionally stratified by visit and/or time period, with each specialty for the visits meeting criteria (i.e. those with the clinical fact provided)