In ianbatesncc/aafractions.ncc: Alcohol Attributable Fractions Lookup Tables

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

options(knitr.kable.NA = "")

library("aafractions.ncc")
library("dplyr")

Executive Summary

Aims and Objectives

• Describe data items in this package

• Describe how this package can be used in Alcohol Attributable Fractions calculations

• Describe various Alcohol Attributable Fraction analyses

Methods

Results

Package Outputs

Discussion

Introduction

Underlying the Alcohol Attributable Fractions calculations are the lookup tables relating alcohol conditions to ICD10 diagnosis codes and those conditions to their respective (age, sex) specific attributable fractions.

There have been a number of revisions to initial lookup tables. This package provides a number of recent lookup tables in a machine readable and processable format for use in local alcohol attributable fraction calculations.

Methods

The original lookup tables are available in a number of formats including .pdf documents and .xlsx spreadsheets. Their extraction is (to be) described in another document.

This document shows how theses tables are to be accessed and queried.

Results

Load package with:

library("aafractions.ncc")

Three datasets are now available.

aa_conditions, aa_versions and aa_fractions.

Conditions

aa_conditions is a data.frame object that lists all the alcohol attributable conditions that are present in the lookup tables.

glimpse(aa_conditions)

For each condition there is a unique identifier, condition_uid, by which that condition is referenced. Each condition has textual description (the desc field) and a definition in terms of ICD10 diagnosis codes (the codes field). This allows a condition to be derived from an ICD10 code.

There are in addition two levels of further categorisation to indicate the nature of the condition which is of use in later analysis. cat1 indicates wholly attributable, or partially attributable - chronic or partially attributable - acute. cat2 indicates wholly attributable or one of eleven other conditions.

require("dplyr", warn.conflicts = FALSE)
set.seed(20180219) ; invisible(sample(32))
knitr::kable(
    caption = "aa_conditions (pseudo-random rows)"
    , sample_n(aa_conditions, 4)
)

Fractions

aa_fractions is a data.frame object that lists all the alcohol attributable fractions for each of the alcohol attributable conditions.

glimpse(aa_fractions)

list("aa_ageband", "sex", "analysis_type", "version") %>% 
    lapply(function(x, y){unique(select(y, x))}, y = aa_fractions) %>% 
    knitr::kable(row.names = FALSE)

versions

aa_versions is a data.frame object that summarises which conditions are used in each version of the lookup tables.

#knitr::kable(data.table::dcast(aa_versions, ... ~ version, value.var = "condition_fuid", fun = function(x){any(as.logical(x))}))

knitr::kable(
    caption = "aa_versions (pseudo-random rows)"
    , sample_n(aa_versions, 6)
)

Discussion

ianbatesncc/aafractions.ncc documentation built on May 3, 2019, 3:34 p.m.