https://github.com/asenetcky/distiller
Distill your wrangled data down to the CDC's EPHT XML format
https://github.com/asenetcky/distiller
cdc epht r r-package rstats rstats-package xml
Last synced: 5 months ago
JSON representation
Distill your wrangled data down to the CDC's EPHT XML format
- Host: GitHub
- URL: https://github.com/asenetcky/distiller
- Owner: asenetcky
- License: other
- Created: 2024-10-29T13:21:50.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2025-03-18T20:25:20.000Z (over 1 year ago)
- Last Synced: 2025-03-18T21:26:36.794Z (over 1 year ago)
- Topics: cdc, epht, r, r-package, rstats, rstats-package, xml
- Language: R
- Homepage: https://asenetcky.github.io/distiller/
- Size: 1.78 MB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.Rmd
- Changelog: NEWS.md
- License: LICENSE
- Code of conduct: CODE_OF_CONDUCT.md
Awesome Lists containing this project
README
---
output: github_document
---
```{r, include = FALSE}
knitr::opts_chunk$set(
collapse = TRUE,
comment = "#>",
fig.path = "man/figures/README-",
out.width = "100%"
)
```
# distiller
[](https://github.com/asenetcky/distiller/actions/workflows/R-CMD-check.yaml)
[](https://app.codecov.io/gh/asenetcky/distiller)
## Motivation
As a newbie who has to submit data to the CDC's EPHT program, I was
dismayed to find out that the documentation is buried under many layers
inside their SharePoint. It is also highly fragmented, convoluted and in many
cases, conflicts with itself.
My goal is to make this process easier and reproducible for myself, and others.
So who is this highly specific package for?
* Do you submit data to the CDC's EPHT program?
* Do you use R? Or are interested in incorporating R into your workflow?
* Do you struggle with the CDC's EPHT documenation and/or tooling?
* Do you want to make your submission process more reproducible?
If you answered yes to the first question and any of the others,
then this package might be for you.
## What does this package do?
I think it's important to state up front what this package _doesn't_ do - and
that is, it will not wrangle your data for you. There are a few helpers, and
and a whole slew of checks `distiller` will run on your data and metadata to
ensure that everything is reasonably close to the correct format
for submission to the CDC's EPHT program.
`distiller` still expects your data to have specific variable names, and
to have the required variables for each type of data. However, if you've
ever wondered why the epht requires different _variable names_
in a _different order_ for the same types of data, even for the _same disease_
you'll be pleased to know that distiller takes care of the
facility-type-specific naming conventions and the ordering for you. Users just
need to bring the data and now they can spend less time worrying about
XML semantics and more time polishing their data products.
`disitller` is __no__ replacement for the CDC EPHPT Test Submission portal,
however, creating the XML, and shuffling files around and then
dropping them into the portal and waiting an indeterminate amount of time for
feedback eats up time and is a pain.
`distiller` aims to provide feedback on your data and metadata
before you send it off to the CDC. This way, you can fix any obvious issues
before you sink 20+ minutes waiting to find out you forgot to replace your `NA`'s
with "U".
## What's in the box?
`distiller` contains the following core functions:
* `check_submission()` - a function that checks your data and metadata and
provides quick feedback
* `make_xml_document()` - a function that creates an xml document for
submission based on your data and the metadata your provide it
`distiller` also contains functions for:
* collapsing race and ethnicity values into the CDC's required format
* converting month integers to 0-padded character strings
* return the proper health outcome identifier for a given content group identifier
* Starting from scratch? Most of the mini-functions that make up the two core
ones are exposed to the user, so you can check your work in pieces as you make
progress with your data wrangling
## `distiller` expectations and scope
`distiller` works for the following content group identifiers:
- AS-HOSP
- AS-ED
- CO-HOSP
- CO-ED
- MI-HOSP
- HEAT-HOSP
- HEAT-ED
- COPD-HOSP
- COPD-ED
`distiller` expects the following variables in your data:
For every content group identifier:
- agegroup
- county
- sex
- ethnicity
- race
- health_outcome_id,
- monthly_count
- month
- year
For content group identifiers CO-HOSP and CO-ED, the above plus the following:
- fire_count
- nonfire_count
- unknown_count
## Installation
You can install the development version of distiller from [GitHub](https://github.com/) with:
``` r
# install.packages("pak")
pak::pak("asenetcky/distiller")
```
## Example
Here is a basic example of how to use it:
```{r example}
library(distiller)
# Take you already-wrangled data
# note the specific variable names
data <-
mtcars |>
dplyr::rename(
month = mpg,
agegroup = cyl,
county = disp,
ethnicity = hp,
health_outcome_id = drat,
monthly_count = wt,
race = qsec,
sex = vs,
year = am
) |>
dplyr::select(-c(gear, carb))
# And your metadata
content_group_id <- "AS-HOSP"
mcn <- "1234-1234-1234-1234-1234"
jurisdiction_code <- "two_letter_code"
state_fips_code <- "1234"
submitter_email <- "submitter@email.com"
submitter_name <- "Submitter Name"
submitter_title <- "Submitter Title"
# Optionally check your submission data structure and metadata
check_submission(
data,
content_group_id,
mcn,
jurisdiction_code,
state_fips_code,
submitter_email,
submitter_name,
submitter_title
)
# This can also be checked with `check_first = TRUE` in `make_xml_document()`
# And then make your xml document
make_xml_document(
data,
content_group_id,
mcn,
jurisdiction_code,
state_fips_code,
submitter_email,
submitter_name,
submitter_title
)
```