Package 'afrihealthsites'

Title: Geographic locations of African health facilities from different sources
Description: African healthsite locations from healthsites.io and WHO/KEMRI-WELLCOME.
Authors: Andy South [aut, cre]
Maintainer: Andy South <[email protected]>
License: GPL (>= 3) | file LICENSE
Version: 0.1.0.9008
Built: 2024-11-11 05:46:50 UTC
Source: https://github.com/afrimapr/afrihealthsites

Help Index

africa healthsite points from healthsites.io

Description

africa healthsite points from healthsites.io

Usage

sf_healthsites_af

Format

sf

An object of class sf (inherits from tbl_df, tbl, data.frame) with 56854 rows and 35 columns.

Slots

data

healthsite points

Source

https://www.healthsites.io

africa healthsite points from WHO

Description

africa healthsite points from WHO

Usage

df_who_sites

Format

dataframe

An object of class tbl_df (inherits from tbl, data.frame) with 98745 rows and 12 columns.

Slots

data

healthsite points

Source

https://www.who.int/malaria/areas/surveillance/who-cds-gmp-2019-01-eng.xlsx

african country names and iso 3 letter country codes

Description

african country names and iso 3 letter country codes

Usage

afcountries

Format

dataframe

An object of class data.frame with 53 rows and 2 columns.

Slots

data

dataframe

afrihealthsites : healthsite locations for Africa

Description

locations from WHO and healthsites.io. Part of afrimapr project.

returns healthsite locations for specified countries and optionally plots map

Usage

afrihealthsites(
  country,
  datasource = "healthsites",
  plot = "mapview",
  hs_amenity = "all",
  who_type = "all",
  type_filter = "all",
  returnclass = "sf",
  type_column = "Facility Type",
  label_column = "Facility Name",
  lonlat_columns = c("Longitude", "Latitude"),
  admin_level = NULL,
  admin_names = NULL
)

Arguments

country

a character vector of country names or iso3c character codes.
datasource

data source, 'healthsites' predownloaded, 'who', 'healthsites_live' needs API, 'hdx' not working yet
plot

option to display map 'mapview' for interactive, 'sf' for static
hs_amenity

filter healthsites data by amenity. 'all', 'clinic', 'dentist', 'doctors', 'pharmacy', 'hospital' to exclude dentist hs_amenity=c('clinic', 'doctors', 'pharmacy', 'hospital')
who_type

filter by Facility type
type_filter

filter by facility type (from the type_column, or internally defined for healthsites & who)
returnclass

'sf' or 'dataframe', currently 'dataframe' only offered for WHO so that can have points with no coords
type_column

for user provided files which column has information on type of site, default : 'Facility Type'
label_column

for user provided files which column has information on name of site, default : 'Facility Name'
lonlat_columns

for user provided files which columns contain longitude, latitude. option of NULL if no coords
admin_level

what admin level to filter regions from FALSE or NULL if no filtering
admin_names

names of admin regions to filter NULL if no filter

Value

sf

See Also

afrihealthsites compare_hs_sources

Examples

sfnga <- afrihealthsites("nigeria", datasource='who', plot='sf')

afrihealthsites('chad', datasource='who', plot='sf')
afrihealthsites('chad', datasource='healthsites', plot='sf')

sfnga <- afrihealthsites("nigeria", plot='mapview')

#to return raw dataframe for WHO data including any rows with no coordinates
dfzaf <- afrihealthsites("south africa", datasource='who', plot=FALSE, returnclass='dataframe')
#note that ISO 3 letter codes and country names with upper case letters can also be used
#afrihealthsites("ZAF")
#afrihealthsites("South Africa")


#filter healthsites data by amenity type
afrihealthsites('chad',datasource = 'healthsites', hs_amenity=c('clinic','hospital'))
#filter who data by Facility type
afrihealthsites('chad',datasource = 'who',who_type=c('Regional hospital','Health Centre'))

#filter by admin regions
afrihealthsites('togo', admin_level=1, admin_names=c('Maritime Region', 'Centrale Region'))

Check whether to install rhealthsites and install if necessary

Description

If the rhealthsites package is not installed, install it from GitLab using remotes.

Usage

check_rhealthsites()

compare healthsite points from different sources for a country

Description

main aim to plot map comparing different sources

Usage

compare_hs_sources(
  country,
  datasources = c("healthsites", "who"),
  plot = "mapview",
  plotshow = TRUE,
  plotcex = c(6, 4),
  col.regions = list(RColorBrewer::brewer.pal(9, "YlGn"), RColorBrewer::brewer.pal(9,
    "BuPu")),
  alpha = c(0.1, 0.1),
  alpha.regions = c(0.7, 0.7),
  layer.names = NULL,
  plotlegend = TRUE,
  hs_amenity = "all",
  who_type = "all",
  canvas = FALSE,
  plotlabels1 = FALSE,
  plotlabels2 = FALSE,
  map.types = c("CartoDB.Positron", "OpenStreetMap.HOT"),
  type_column = "Facility Type",
  label_column = "Facility Name",
  lonlat_columns = c("Longitude", "Latitude"),
  admin_level = NULL,
  admin_names = NULL
)

Arguments

country

a character vector of country names or iso3c character codes.
datasources

vector of 2 datasources from 'healthsites' predownloaded, 'who', 'healthsites_live' needs API, 'hdx' not working yet
plot

option to display map 'mapview' for interactive, 'sf' for static
plotshow

whether to show the plot, otherwiser just return plot object
plotcex

sizes of symbols for each source default=c(6,3), helps view symbol overlap
col.regions

list of two colour palettes to pass to mapview
alpha

list of two alphas to pass to mapview - low keeps borders light
alpha.regions

list of two alpha.regions to pass to mapview
layer.names

allow mapview layer.names to be set c('a','b')
plotlegend

whether to add legend to mapview plot
hs_amenity

filter healthsites data by amenity. 'all', 'clinic', 'dentist', 'doctors', 'pharmacy', 'hospital'
who_type

filter by Facility type
canvas

mapview plotting option, TRUE by default for better performance with larger data
plotlabels1

whether to add static labels for source1
plotlabels2

whether to add static labels for source2
map.types

optional specification of background map tiles for mapview, default c('CartoDB.Positron','OpenStreetMap.HOT')
type_column

just for user provided files which column has information on type of site, default : 'Facility Type'
label_column

just for user provided files which column has information on name of site, default : 'Facility Name'
lonlat_columns

for user provided files which columns contain longitude, latitude. option of NULL if no coords
admin_level

what admin level to filter regions from FALSE or NULL if no filtering
admin_names

names of admin regions to filter NULL if no filter

Value

sf

Examples

#compare_hs_sources("nigeria", datasources=c('who', 'healthsites'), plot='mapview')
#compare_hs_sources(c('malawi','zambia'))

#filter by admin regions
compare_hs_sources('togo', admin_level=1, admin_names=c('Maritime Region', 'Centrale Region'))

conversion from country names to iso3c code

Description

#todo vectorise

Usage

country2iso(country)

Arguments

country

a character vector of country names

Value

character vector of iso3c codes

Examples

iso3c <- country2iso("nigeria")

get & plot freq of facility types for a country

Description

get & plot freq of facility types for a country

Usage

facility_types(
  country,
  datasource = "healthsites",
  datasource_title = NULL,
  type_filter = "all",
  type_column = "Facility Type",
  label_column = "Facility Name",
  brewer_palette = "BuPu",
  lonlat_columns = c("Longitude", "Latitude"),
  admin_level = NULL,
  admin_names = NULL,
  plot_title = "default",
  plot = TRUE
)

Arguments

country

a character vector of country names or iso3c character codes.
datasource

data source, 'healthsites' predownloaded, 'who', 'healthsites_live' needs API, 'hdx' not working yet
datasource_title

optional title for datasource to be used in plots - particularly if a filname has been passed for datasource
type_filter

filter by facility type - will depend on the data source
type_column

just for user provided files which column has information on type of site, default : 'Facility Type'
label_column

just for user provided files which column has information on name of site, default : 'Facility Name'
brewer_palette

ColorBrewer palette default 'BuPu',
lonlat_columns

just for user provided files which columns contain longitude, latitude
admin_level

what admin level to filter regions from FALSE or NULL if no filtering
admin_names

names of admin regions to filter NULL if no filter
plot_title

title for plot, 'default', string or NULL for no title
plot

whether to display plot

Value

ggplot2 object

Examples

ggnga <- facility_types("nigeria", datasource='who')

facility_types('chad', datasource='who')
facility_types('chad', datasource='healthsites')


#filter healthsites data by amenity type
facility_types('chad',datasource = 'healthsites', type_filter=c('clinic','hospital'))
#filter who data by Facility type
facility_types('chad',datasource = 'who', type_filter=c('Regional hospital','Health Centre'))

# from an sf object
data(sfssd)
ggssd <- facility_types("south sudan",
                        datasource=sfssd,
                        type_column = "type")

# using consistent 9 class facility types for WHO data, specify type_column='facility_type_9'
facility_types('all', datasource='who', type_column='facility_type_9')

Install the rhealthsites package after checking with the user

Description

Install the rhealthsites package after checking with the user

Usage

install_rhealthsites()

conversion from iso3c code to country names

Description

#todo vectorise

Usage

iso2country(iso3c)

Arguments

iso3c

a character vector of country codes

Value

character vector of country names

Examples

name <- iso2country("nga")

merge healthsite points from different sources for a country

Description

IN DEVELOPMENT aim is to identify and remove duplicates first version just tests distance, between points in layer1 and 2 BUT problem maybe that there are points in each layer that are duplicates too may want to move this to separate package later

Usage

merge_points(
  country,
  datasources = c("healthsites", "who"),
  hs_amenity = c("clinic", "doctors", "pharmacy", "hospital"),
  dist_same_m = 50,
  toreturn = "summary"
)

Arguments

country

a character vector of country names or iso3c character codes.
datasources

vector of 2 datasources from 'healthsites' predownloaded, 'who', 'healthsites_live' needs API, 'hdx' not working yet
hs_amenity

filter healthsites data by amenity. 'all', 'clinic', 'dentist', 'doctors', 'pharmacy', 'hospital' to exclude dentist hs_amenity=c('clinic', 'doctors', 'pharmacy', 'hospital')
dist_same_m

distance below which a site from source 1 and 2 is considered same
toreturn

whether to return 'summary' or the merged 'sf' object

Value

sf

Examples

#merge_points("nigeria", datasources=c('who', 'healthsites'), plot='mapview')
#merge_points(c('malawi','zambia'))

name of label column according to datasource

Description

name of label column according to datasource

Usage

nameof_labcol(datasource, label_column)

Arguments

datasource

vector of 2 datasources from 'healthsites' predownloaded, 'who', 'healthsites_live' needs API, 'hdx' not working yet
label_column

just for user provided files which column has information on name of site, default : 'Facility Name'

Value

character column name

Examples

nameof_labcol('who')

name of z column according to datasource

Description

name of z column according to datasource

Usage

nameof_zcol(datasource, type_column = NULL)

Arguments

datasource

'healthsites' predownloaded, 'who', 'healthsites_live' needs API, 'hdx' not working yet
type_column

just for user provided files which column has information on type of site, default : 'Facility Type'

Value

character column name

Examples

nameof_zcol('who')

availability of a national master facility list by country

Description

returns data availability dataframe for 'all' countries or 1 row for a single country TODO add calculation of stats on numbers of countries satisfying criteria

Usage

national_list_avail(country = "all")

Arguments

country

'all' (default) or a single country name or iso3c character code.

Value

dataframe or single value

Examples

national_list_avail('Togo')

data url for national master facility lists by country

Description

returns urls for 'all' available countries or url for a single country

Usage

national_list_url(country = "all")

Arguments

country

'all' (default) or a single country name or iso3c character code.

Value

dataframe or single value

Examples

national_list_url('South Sudan')

#example of reading in data direct from a url and mapping
#will only work for countries where "machine_readable" is TRUE
#dfgha <- read.csv(national_list_url("Ghana"))
#sfgha <- sf::st_as_sf(dfgha, coords=c("Longitude","Latitude"), crs=4326, na.fail=FALSE)
#mapview::mapview(sfgha, zcol='Type')
#afrihealthsites('gha',datasource=sfgha, type_column='Type')
#compare_hs_sources("ghana", datasources=list(sfgha,"who"), type_column='Type')
#3756 facilities from mfl
#facility_types('gha',datasource=sfgha, type_column='Type')
#1878 facilities in who-kemri data
#facility_types('gha',datasource='who', type_column='Type')


#to return a dataframe with all countries that have urls
national_list_url()

find near points for a country initially in one datasource

Description

IN DEVELOPMENT aim is to help identify and remove duplicates

Usage

near_points(
  country,
  datasources = c("healthsites", "who"),
  hs_amenity = c("clinic", "doctors", "pharmacy", "hospital"),
  dist_same_m = 50,
  toreturn = "summary"
)

Arguments

country

a character vector of country names or iso3c character codes.
datasources

vector of 2 datasources from 'healthsites' predownloaded, 'who', 'healthsites_live' needs API, 'hdx' not working yet
hs_amenity

filter healthsites data by amenity. 'all', 'clinic', 'dentist', 'doctors', 'pharmacy', 'hospital' to exclude dentist hs_amenity=c('clinic', 'doctors', 'pharmacy', 'hospital')
dist_same_m

distance below which a site from source 1 and 2 is considered same
toreturn

whether to return 'summary' or the merged 'sf' object

Value

sf

Examples

#near_points("burundi", datasources=c('who', 'healthsites'), plot='mapview')

to run the shiny web application.

Description

to run the shiny web application.

Usage

runviewer()

south sudan health facility points from moh

Description

south sudan health facility points from moh

Usage

sfssd

Format

sf

An object of class sf (inherits from data.frame) with 2889 rows and 18 columns.

Slots

data

health facility points

Source

https://www.southsudanhealth.info/facility/fac.php?list&s=0&p=0&ps=2889

takes a vector of facility types and classes them into 4 tiers following Falchetta type definition

Description

to reclass facility types into a 4 tier classification, any types not passed will be classed as 0

Usage

tiers4(
  types,
  tier1 = NULL,
  tier2 = NULL,
  tier3 = NULL,
  tier4 = NULL,
  nametier0 = "Tier0 unknown",
  nametier1 = "Tier1 health post",
  nametier2 = "Tier2 health centre",
  nametier3 = "Tier3 provincial hospital",
  nametier4 = "Tier4 central hospital",
  to_return = "Tier"
)

Arguments

types

facility types, one per health facility as a vector
tier1

types to be classed as tier1
tier2

types to be classed as tier2
tier3

types to be classed as tier3
tier4

types to be classed as tier4
nametier0

name for tier0 i.e. when it isn't classed into other tiers
nametier1

name for tier1
nametier2

name for tier2
nametier3

name for tier3
nametier4

name for tier4
to_return

"Tier" or "Tier_name" potentially others later

Value

vector of numeric values or strings representing tiers

Examples

tiers4(c("a","b","c"), tier1=c("a","b"), tier2="c", to_return="Tier_name")

lookup table to convert 173 WHO facility types to 9 broad categories

Description

lookup table to convert 173 WHO facility types to 9 broad categories

Usage

who_type_lookup

Format

dataframe

An object of class tbl_df (inherits from tbl, data.frame) with 318 rows and 3 columns.

Slots

data

who type lookup

Source

https://link.springer.com/article/10.1186/s12916-019-1459-6