wranglEHR

Overview

wranglEHR is a data wrangling tool for EMAP. It is designed to run against OMOP CDM 5.3.1. Please see the R vignettes for further details on how to use the package to perform the most common tasks:

• extract() produces a rectangular table in a “long” format that is suitable for most statistial packages.
• clean() cleans the above table in accordance with pre-defined standards (pending)

Installation

# install directly from github with
remotes::install_github("inform-health-informatics/wranglEHR")

Usage

More documentation to follow

library(wranglEHR)

ctn <- ctn <- DBI::dbConnect(
  RPostgres::Postgres(),
  host = "****", # Host target for the UDS
  port = 5432,
  user = "****",
  password = rstudioapi::askForPassword(),
  dbname = "uds")

# Extract variables.
# Rename on the fly.
# Dynamically set time cadance.
ltb <- extract(
  connection = ctn,
  target_schema = "ops_dev",
  visit_occurrence_ids = 600000:600005,
  concept_names = c(3013502, 44809212),
  relabel = c("spo2", "spo2_target"),
  coalesce_rows = dplyr::first,
  chunk_size = 5000,
  cadance = 1
)

head(ltb)

# Add in bed movement data
ltb <- attach_locations(ctn, ltb)
head(ltb)

# Regularise the underlying time cadance of the table
ltb <- regularise(ltb, cadance = 1)

# Don't forget to switch of the lights after you leave
DBI::dbDisconnect(ctn)

Getting help

If you find a bug, please file a minimal reproducible example on github.