ChessProcessingPipeline: Process chess data

In terminological/uk-covid-datatools: Functions for sourcing, cleaning and calculating Rt for uk covid timeseries data

Description

Process chess data

Process chess data

Super classes

ukcovidtools::PassthroughFilesystemCache -> ukcovidtools::DataProvider -> ChessProcessingPipeline

Methods

Public methods

• ChessProcessingPipeline$new()

• ChessProcessingPipeline$chessQuality()

• ChessProcessingPipeline$chessItuSubset()

• ChessProcessingPipeline$chessAdmissionSubset()

• ChessProcessingPipeline$chessDefaultFilter()

• ChessProcessingPipeline$cleanseCHESSData()

• ChessProcessingPipeline$clone()

Inherited methods

• ukcovidtools::PassthroughFilesystemCache$getDaily()

• ukcovidtools::PassthroughFilesystemCache$getHashCached()

• ukcovidtools::PassthroughFilesystemCache$getSaved()

• ukcovidtools::PassthroughFilesystemCache$unloadCache()

• ukcovidtools::DataProvider$ageCatToFactor()

• ukcovidtools::DataProvider$ageToAgeCat()

• ukcovidtools::DataProvider$breakFromCats()

• ukcovidtools::DataProvider$cutByAge()

• ukcovidtools::DataProvider$download()

• ukcovidtools::DataProvider$downloadAndUntar()

• ukcovidtools::DataProvider$downloadAndUnzip()

• ukcovidtools::DataProvider$downloadDaily()

• ukcovidtools::DataProvider$labelsFromBreaks()

• ukcovidtools::DataProvider$normaliseAgeCat()

• ukcovidtools::DataProvider$normaliseGender()

---

Method new()

New chess processing pipeline

Usage

ChessProcessingPipeline$new(providerController, ...)

Arguments

providerController

the provider controller

...

for compatibility

Returns

the processor

---

Method chessQuality()

Chess data metrics on a hospital by hospital basis

Usage

ChessProcessingPipeline$chessQuality(
  CHESSdf,
  updatedWithin = 14,
  date = max(CHESSdf$dateupdated, na.rm = TRUE),
  unknownDatesPercent = 0.2
)

Arguments

CHESSdf

- the raw chess data set

Details

the chess data has some issues as different hospitals have different reporting strategies

Returns

a data frame of stats by trust representing reporting quality.

---

Method chessItuSubset()

The subset of the CHESS data of patients which are admitted to ITU

the chess data has some issues as different hospitals have different reporting strategies

Usage

ChessProcessingPipeline$chessItuSubset(
  CHESSdf = self$getCHESS(),
  updatedWithin = 14,
  date = max(CHESSdf$dateupdated, na.rm = TRUE),
  unknownDatesPercent = 0.2
)

Arguments

CHESSdf

- the raw chess data set

date

- the date of the data set

Returns

a data frame of stats by trust representing reporting quality.

---

Method chessAdmissionSubset()

The subset of the CHESS data of patients from hospitals that appear to be reporting all admissions

the chess data has some issues as different hospitals have different reporting strategies

Usage

ChessProcessingPipeline$chessAdmissionSubset(
  CHESSdf = self$getCHESS(),
  updatedWithin = 14,
  date = max(CHESSdf$dateupdated, na.rm = TRUE),
  unknownDatesPercent = 0.2
)

Arguments

CHESSdf

- the raw chess data set

date

- the date of the data set

Returns

a data frame of stats by trust representing reporting quality.

---

Method chessDefaultFilter()

Usage

ChessProcessingPipeline$chessDefaultFilter(CHESSdf = self$getCHESS(), ...)

---

Method cleanseCHESSData()

Cleanses CHESS data set and adds a few fields

* Detects trusts that have not been updating records * establishes a censoring date on a trust by trust basis * checks all outcomes have an outcomedate * checks outcome dates are after admission date * excludes patients whose tests are >10 days after admission e.g. hospital acquired * excludes paediatric patients, and those with no known gender * adds censoring status (as status) * adds time from admission to outcome (as timeToOutcome)

Usage

ChessProcessingPipeline$cleanseCHESSData(CHESSdf, date, removeOutliers = TRUE)

Arguments

path

- path to the ff100 file

Returns

cleansed CHESS data set

---

Method clone()

The objects of this class are cloneable with this method.

Usage

ChessProcessingPipeline$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

terminological/uk-covid-datatools documentation built on June 24, 2021, 8:16 p.m.