In nhs-pt/hospitals: Portuguese 'NHS' Hospitals

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%"
)
library(hospitals)

hospitals

This R data package contains one single data set --- the eponymous hospitals tibble --- comprising the current r nrow(hospitals) hospitals in the Portuguese National Health Service.

For a description of each variable and data sources used to assemble this data set browse the documentation with ?hospitals.

Installation

Install {hospitals} from CRAN:

install.packages("hospitals")

You can instead install the development version of {hospitals} by setting nhs-pt's universe repository:

options(repos = c(
    nhspt = 'https://nhs-pt.r-universe.dev',
    CRAN = 'https://cloud.r-project.org'))

install.packages('hospitals')

Citing this work

To generate a citation for this publication from within R:

citation('hospitals')

The hospitals data set

The Portuguese NHS hospitals

Here are the r nrow(hospitals) hospitals currently included in the Portuguese National Health Service:

library(hospitals)

# Hospital identifier and name
hospitals[c(1, 7)] %>% print(n = Inf)

Variables

The hospitals data set contains the following variables:

| Variable | Description | |:------------------|:-------------| | hospital_id | A surrogate identifier created by us that ought to help researchers refer to hospitals in an unambiguous way. The idea is for these identifiers to never change. So if an hospital disappears, or is merged with another one, we will deprecate old identifiers and create new, accordingly. | | hospital_acronym | The acronym or initialism of the hospital name. | | hospital_legal_status | Either an administrative public sector ('SPA') entity, a corporate public entity ('EPE') or a public-private partnership ('PPP') entity. | | hospital_integration | The integration level of the hospital institution: a single hospital unit ('H'); Centro Hospitalar ('CH'), comprising two or more individual Hospital units (a case of horizontal integration); and Unidade Local de Saúde ('ULS'), comprising both Hospital entities and primary care health centres (a case of vertical integration). | | hospital_group | One of the groups defined by Ordinance 82/2014: 'I', 'II', 'III', 'IV-a', 'IV-b' or 'IV-c'. This classification is based on the hospital catchment area and the scope of medical specialities provided. Broadly speaking, group I includes local hospitals, group II corresponds to regional hospitals, and group III to central hospitals. Group IV corresponds to specialised hospitals: IV-a, oncology institutes; IV-b, physical medicine and rehabilitation hospitals; and IV-c, psychiatry and mental health hospitals. | | hospital_short_name | An abbreviated version of the hospital name. | | hospital_full_name | The name of the hospital. | | hospital_region_full_name | The name of the Portuguese health region. | | hospital_region_short_name | A shortened version of hospital_region_full_name. | | hospital_vatin | VAT identification number (VATIN). In Portuguese, Número de Identificação de Pessoa Colectiva (NIPC). | | hospital_latitude | Latitude of the location of the hospital entity expressed in decimal degrees. | | hospital_longitude | Longitude of the location of the hospital entity expressed in decimal degrees. | | hospital_region_colour | Colour associated with the region. These colours are assigned by this package and are provided as a convenience colour palette for plotting in R. | | hospital_bh_group | One of the groups defined by the project Benchmarking of Hospitals. Source: BH \| Grupos e Instituições. | | hospital_bh_group_colour | Colour associated with the hospital_bh_group. These colours are assigned by this package and are provided as a convenience colour palette for plotting in R. | | hospital_regex | A regular expression that can be used to identify hospitals from their names, even when the names are not exactly as in hospital_full_name. This is used by the function normalise(). |

Helper functions

normalise()

The names of hospitals found in the wild can have subtle variations. The normalise() function matches hospital names found elsewhere to the hospitals included in the data set hospitals.

For example, with normalise you can convert hospital names found elsewhere to the names found in hospitals$hospital_short_name:

normalise(c('Hospital do Alto Minho'))

Also normalise is aware of old hospital names, and can map to the new ones:

normalise(c('Hospital do Alto Ave'))

The method behind normalise for matching hospital names to their names in hospital_short_name is based on an heuristic that uses a minimal set of keywords to identify the hospital. So it is pretty tolerant to variations in the name as long as one of the critical keywords is found in the name, e.g., to identify the Hospital of Algarve as such only the keyword Algarve needs to be present in the name:

normalise(c('Algarve', 'H Algarve', 'Hospital do Algarve'))

Also normalise should be lenient with typos associated with accented characters:

normalise('Hospital de São João')

# Without the tildes the mapping still works fine
normalise('Hospital de Sao Joao')

Instead of the hospital_short_name you may ask for the hospital_full_name or the hospital_id:

# 'hospital_short_name' is the default 
normalise('Matosinhos', return = 'hospital_short_name')

normalise('Matosinhos', return = 'hospital_full_name')

normalise('Matosinhos', return = 'hospital_id')

get_hospital_attribute()

get_hospital_attribute() is a thin wrapper function allowing you to easily retrieve hospital attributes from the data set hospitals.

You specify the hospitals by indicating what type of key you are providing: "hospital_id", "hospital_acronym", "hospital_short_name" or "hospital_full_name". Contrarily to the matching performed by normalise(), get_hospital_attribute() matches exactly the key provided.

And specify the column variable in hospitals with the value argument (default is "hospital_short_name").

# By default you get the hospital short name
get_hospital_attribute('h0001')

# Same as above
get_hospital_attribute('h0001', value = 'hospital_short_name')

# Or get instead the full name
get_hospital_attribute('h0001', value = 'hospital_full_name')

# Map the hospital short name to its full name
get_hospital_attribute('IPO de Lisboa', key = 'hospital_short_name', value = 'hospital_full_name')

Code of Conduct

Please note that the hospitals project is released with a Contributor Code of Conduct. By contributing to this project, you agree to abide by its terms.

nhs-pt/hospitals documentation built on Dec. 22, 2021, 2:10 a.m.