In Moohan/slfhelper: Useful functions for working with the Source Linkage Files

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

Selecting only specified variables

It is recommended to only choose the variables you need when reading in a Source Linkage File. This can be achieved by specifying a column argument to the relevant read_slf_ function.

This will result in the data being read in much faster as well as being easy to work with. The full episode and individual files have 200+ and 100+ variables respectively!

library(slfhelper)

library(slfhelper)

ep_data <- read_slf_episode(year = 1920, columns = c("year", "anon_chi", "recid"))

indiv_data <- read_slf_individual(year = 1920, columns = c("year", "anon_chi", "nsu"))

Looking up variable names

To help with the task of picking which variables you might need for your analysis, as well as getting the spelling correct, we provide lists of the variable names in the package.

# Show the first few variables from the episode file
head(ep_file_vars)

# Do the same for the individual file
head(indiv_file_vars)

Variable packs

This is great but it can still be a lot of effort and copy/pasting every time, especially if you need quite a few variables for your analysis.

To assist with this, there are a number of 'variable packs', these are groups of variables which would commonly be needed together which can be accessed with a simple name. Currently there are four packs; demog_vars, ltc_vars, ep_file_bedday_vars and ep_file_cost_vars. Let's see what they contain.

Demographic variables

These are demographic variables which are specific to CHI and can be used with episode or individual file.

demog_vars

Long Term Condition (LTC) variables

These are the Long Term Condition flag variables which are specific to CHI and can be used with episode or individual file.

ltc_vars

Bedday variables

These are variables detailing beddays, they are specific to an episode and can only be used with the episode file.

ep_file_bedday_vars

Cost variables

These are variables detailing costs, they are specific to an episode and can only be used with the episode file.

ep_file_cost_vars

Using variable packs

These variable packs can be used in the column selection to simplify your code substantially.

For example to take some demographic data and LTC flags from the individual file.

library(slfhelper)

indiv_ltc_data <- read_slf_individual(year = 1920, columns = c("year", demog_vars, ltc_vars))

Or to get bedday information about Acute records from the episode file.

library(slfhelper)

acute_beddays <- read_slf_episode(
  year = 1920,
  columns = c("year", "anon_chi", "hbtreatcode", "recid", ep_file_bedday_vars, "cij_pattype"),
  recid = c("01B", "GLS")
)

Conclusion

You should be using the column argument when reading in data to increase the read speed, and reduce the amount of data you are loading into R. slfhelper provides a number of helpers to make picking and using the variables you need easier.

If you would like any changes made to any existing packs, please open an issue on GitHub.

If you would like to suggest any additional variable packs, either open an issue, or even submit a pull request!

Moohan/slfhelper documentation built on April 25, 2024, 5:24 a.m.