Scope of the possible with R

R
overview
Published

November 21, 2024

Session materials

Slides for this session / .pdf slides for this session

No feedback found for this session

Welcome

  • this session is a non-technical overview designed for service leads

Session outline

  • Why R, and why this session?
  • R demo - take some data, load, tidy, analyse
  • Strengths and weaknesses
    • obvious
    • less obvious
  • Alternatives
  • Skill development

R

  • free and open-source
  • multi-platform
  • large user base
  • prominent in health, industry, biosciences

Why this session?

  • R can be confusing
    • it’s code-based, and most of us don’t have much code experience
    • it’s used for some inherently complicated tasks
    • it’s a big product with lots of add-ons and oddities
  • But R is probably the best general-purpose toolbox we have for data work at present
    • big user base in health and social care
    • focus on health and care-like applications
    • not that hard to learn
    • extensible and flexible
    • capable of enterprise-y, fancy uses

R demo

  • this is about showing what’s possible, and give you a flavour of how R works
  • we won’t explain code in detail during this session
  • using live open data
    https://www.opendata.nhs.scot/dataset/weekly-accident-and-emergency-activity-and-waiting-times

Load that data

library(readr)
ae_activity <- read_csv("data/weekly_ae_activity_20240609.csv")

One small bit of cheating: renaming

names(ae_activity) <- c("date", "country", "hb", "loc", "type", "attend", "n_within", "n_4", "perc_4", "n_8", "perc_8", "n_12", "perc_12")

Preview

Preview of data
date country hb loc type attend n_within n_4 perc_4 n_8 perc_8 n_12 perc_12
20220109 S92000003 S08000024 S319H Emergency Department 733 695 38 94.8 1 0.1 0 0.0
20240310 S92000003 S08000024 S314H Emergency Department 2273 1078 1195 47.4 384 16.9 177 7.8
20170611 S92000003 S08000024 S314H Emergency Department 2265 2181 84 96.3 7 0.3 1 0.0
20190310 S92000003 S08000028 W107H Emergency Department 120 118 2 98.3 0 0.0 0 0.0
20180617 S92000003 S08000022 H212H Emergency Department 246 235 11 95.5 0 0.0 0 0.0

Removing data

ae_activity <- ae_activity |>
    select(!c(country, contains("perc_")))
Preview of data
date hb loc type attend n_within n_4 n_8 n_12
20210502 S08000020 N101H Emergency Department 904 752 152 14 1
20230115 S08000022 C121H Emergency Department 138 129 9 0 0
20220313 S08000032 L302H Emergency Department 1198 744 454 87 41
20240414 S08000016 B120H Emergency Department 584 364 220 63 39
20240114 S08000031 G513H Emergency Department 1295 1235 60 0 0

Tidying data

ae_activity <- ae_activity |>
    mutate(date = lubridate::ymd(date))
Preview of data
date hb loc type attend n_within n_4 n_8 n_12
2024-01-28 S08000025 R103H Emergency Department 126 117 9 0 0
2021-02-21 S08000024 S308H Emergency Department 831 667 164 41 7
2020-12-13 S08000028 W107H Emergency Department 70 68 2 0 0
2021-05-16 S08000032 L106H Emergency Department 1412 1216 196 25 5
2020-01-19 S08000031 G513H Emergency Department 1197 1177 20 0 0

Subset data

  • we’ll take a random selection of 5 health boards to keep things tidy
ae_activity <- ae_activity |>
    filter(hb %in% boards)
Preview of data
date hb loc type attend n_within n_4 n_8 n_12
2022-10-23 S08000019 V217H Emergency Department 1099 409 690 379 211
2022-12-11 S08000029 F704H Emergency Department 1356 772 584 168 45
2024-03-17 S08000020 N101H Emergency Department 1062 474 588 279 69
2017-09-17 S08000020 N121H Emergency Department 304 300 4 0 0
2021-03-07 S08000024 S314H Emergency Department 1825 1468 357 39 3

Basic plots

library(ggplot2)
ae_activity |>
    ggplot() +
    geom_line(aes(x = date, y = attend, colour = hb, group = loc))

Joining data

ae_activity |>
    left_join(read_csv("data/boards_data.csv"), by = c("hb" = "HB")) |>
    select(!any_of(c("_id", "HB", "HBDateEnacted", "HBDateArchived", "Country"))) |>
    ggplot() +
    geom_line(aes(x = date, y = attend, colour = HBName, group = loc))

and again…

Add to a map

ae_activity_loc |>
    leaflet::leaflet() |>
    leaflet::addTiles() |>
    leaflet::addMarkers(~longitude, ~latitude, label = ~HospitalName)

Then make that map more useful

ae_activity_loc |>
    group_by(HospitalName) |>
    summarise(attend = sum(attend), n_within = sum(n_within), longitude = min(longitude), latitude = min(latitude)) |>
    mutate(rate = paste(HospitalName, "averages", scales::percent(round(n_within / attend, 1)))) |>
    leaflet::leaflet() |>
    leaflet::addTiles() |>
    leaflet::addMarkers(~longitude, ~latitude, label = ~rate)

Then add to reports, dashboards…

Strengths

  • enormous scope and flexibility
  • a force-multiplier for fancier data work
    • helps collaboration within teams, between teams, between orgs
    • reproducible analytics
    • modular approaches to large projects
  • decreasing pain curve: the fancier the project, the better

Weaknesses

  • harder to learn than competitors
  • very patchy expertise across H+SC Scotland
  • complex IG landscape
  • messy skills development journey