v0.3.2

• Important bugfix of age and los bin edges
  • Prior versions had a subtle bug that resulted in age and los being assigned incorrect categories when the values were on the boundaries.
• Merge #42, #43 from @georgm8
• Fix #48

v0.3.1

• Merge #35 for ICD10 validation
• Merge #40 for edchiefcomplaint grouping which closes #36
• Improve regex matching of diag_NN, opertn and opdate
• Change edwaittime,timeined dtypes - fixes #32
• Update documentation

Contributors to code and testing:

• @ccarenzoIC
• @georgm8
• @MattStammers (maybe there will be a v0.4.0 after all 😄 )
• @quindavies

And as always, thanks to Sheffield SCHARR Institute, HDRUK and the entire team for a great collaborative project. 🔥 💣 🚀

v0.3.0

Major updates to validation and feature engineering

validate_dataframe function has been updated to allow customising validation by allowing the user to set column validation to be skipped entirely or to be modified. This should help temporarily bypass significant DQ issues with SNOMED codes but should be used sparingly.

Feature maps have been reconfigured to make maintenance easier. As part of this multiple strategies for dealing with missing and unmapped codes have been implemented. Read the documentation on error codes for more information and mitigation steps.

Documentation has been updated to reflect these changes.

v0.2.1-alpha

Fix #28 - Change visit_id dtype to str to allow hashed ids.
Merge #29 - Fix typo in docs

Contributors:

@georgm8

v0.2.0-alpha

Pre-release of avoidable_admissions v0.2.0-alpha as a pip installable package.

pip install "avoidable_admissions @ git+https://github.com/LTHTR-DST/[email protected]"

Please see README.md and project documentation at https://lthtr-dst.github.io/hdruk_avoidable_admissions/ for more information on installation and usage.

Please use GitHub issues for reporting bugs or requesting features. Fork, clone and contribute code to this collaborative effort.

Contributors to code and testing:

• @quindavies
• @MattStammers
• @ccarenzoIC
• @dfleming9

And special thanks to SCHARR Institute at Sheffield University for putting together excellent data and analysis specification documents that makes this approach even possible.

v0.1.0-alpha

Pre-release of avoidable_admissions as a pip installable package. Please see README.md and project documentation at https://lthtr-dst.github.io/hdruk_avoidable_admissions/ for more information for installation and usage.

Please use GitHub issues for reporting bugs or requesting features. Fork, clone and contribute code to this collaborative effort.