From f47e30d09ab6da86cd10c55d38c6c6b8eb1fa23c Mon Sep 17 00:00:00 2001
From: Peter Winskill
Date: Wed, 20 Nov 2024 16:45:53 +0000
Subject: [PATCH] Port historical foresite vignettes
---
DESCRIPTION | 5 +-
vignettes/Interventions.Rmd | 276 +++++++++++++++++++++++++++++++++++
vignettes/Metadata.Rmd | 50 +++++++
vignettes/Seasonality.Rmd | 30 ++++
vignettes/historical_epi.Rmd | 46 ++++++
vignettes/nature.csl | 154 +++++++++++++++++++
vignettes/pop_demog.Rmd | 63 ++++++++
vignettes/references.bib | 173 ++++++++++++++++++++++
vignettes/vectors.Rmd | 65 +++++++++
9 files changed, 861 insertions(+), 1 deletion(-)
create mode 100644 vignettes/Interventions.Rmd
create mode 100644 vignettes/Metadata.Rmd
create mode 100644 vignettes/Seasonality.Rmd
create mode 100644 vignettes/historical_epi.Rmd
create mode 100644 vignettes/nature.csl
create mode 100644 vignettes/pop_demog.Rmd
create mode 100644 vignettes/references.bib
create mode 100644 vignettes/vectors.Rmd
diff --git a/DESCRIPTION b/DESCRIPTION
index 2d67968..bfa0168 100644
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -27,4 +27,7 @@ Imports:
withr
Suggests:
testthat (>= 3.0.0),
- fs
+ fs,
+ rmarkdown,
+ knitr
+VignetteBuilder: knitr
diff --git a/vignettes/Interventions.Rmd b/vignettes/Interventions.Rmd
new file mode 100644
index 0000000..49c58de
--- /dev/null
+++ b/vignettes/Interventions.Rmd
@@ -0,0 +1,276 @@
+---
+title: "Interventions"
+output: rmarkdown::html_vignette
+bibliography: references.bib
+csl: nature.csl
+link-citations: true
+vignette: >
+ %\VignetteIndexEntry{Interventions}
+ %\VignetteEngine{knitr::rmarkdown}
+ %\VignetteEncoding{UTF-8}
+---
+
+```{r, include = FALSE}
+knitr::opts_chunk$set(
+ collapse = TRUE,
+ comment = "#>"
+)
+```
+
+
+`site_file$interventions`
+
+
+Interventions contains the historical intervention information for a site. It is
+also the section of the site file that you would modify with intervention
+information for future scenarios. Details, references and methods for individual
+interventions are shown below:
+
+## ITNs
+
+### ITN use
+
+`site_file$interventions$itn_use`
+
+Due to differences in the availability of data sources the approach for countries
+within sub-Saharan Africa differs to countries outside of sub-Saharan Africa:
+
+#### Within sub-Saharan Africa
+
+The population at risk weighted mean ITN use estimates for each site are taken
+from the malaria atlas project raster entitled:
+"Insecticide treated bednet (ITN) use version 2020". This, and other elements
+of the [netz R package](https://mrc-ide.github.io/netz/) are based on work by
+Bertozzi-Villa *et al* @bertozzi
+
+#### Outside of sub-Saharan Africa
+
+ITN use is much more heterogeneous outside of SSA and data are less systematically
+collected. As a result, there are strong assumptions associated with the historical
+scale and magnitude of ITN distributions. We make the assumption that any reported
+ITN distributions (as detailed by the world malaria report @WMR) are tarted to areas
+that have > 1% PfPr or > 1% PvPr at baseline. Annual distributions in these regions
+are then scaled so that that number of ITNs distributed is aligned with the
+world malaria report.
+
+### Missing years
+
+Available data on ITN use (via MAP @MAP or the world malaria report @ WMR) will not extend
+to the present year. Missing ITN use estimates to present are filled assuming a
+constant, continuing level of coverage. To respect the multi-year cyclical
+nature of ITN distribution cycles any missing estimates are filled in assuming
+that coverage is constant with respect to 3 years prior. For example if years
+2019, 2020 and 2021 are missing then 2019 == 2016, 2020 == 2017 and 2021 == 2018.
+
+### Net type
+
+`site_file$interventions$net_type`
+
+It is assumed that all nets distributed prior to 2025 are standard pyrethroid
+(`pyrethroid_only`) bed nets. Current available net types are: pyrethroid_only,
+pyrethroid_pbo and pyrethroid_pyrrole.
+
+### Net input distribution
+
+`site_file$interventions$itn_input_dist`
+
+We need to estimate the annual ITN distributions that would cumulatively result
+in our observed timeseries of ITN usage. For the we use the `fit_usage()`
+function from the [netz R package](https://mrc-ide.github.io/netz/).
+We an impose 3 year cyclical limits to the number of nets that are
+distributed to avoid unrealistic over-distribution.
+
+### Pyrethroid resistance
+
+`site_file$interventions$pyrethroid_resistance`
+
+For each site we include an estimated level of pyrethroid insecticide resistance.
+This has been estimated by Tom Churcher and colleagues using spatio-temporally
+distributed bioassay mortality data. This work is not yet published. Therefore
+for attribution/citation and further information on the methods used please contact
+Pete Winskill or Tom Churcher.
+
+### ITN efficacy parameters
+
+`site_file$interventions$dn0`
+`site_file$interventions$rn0`
+`site_file$interventions$gamman`
+`site_file$interventions$rnm`
+
+Given an ITN type and level of pyrethroid insecticide resistance, we can link to the
+corresponding estimates of the key ITN efficacy parameters. These have been
+estimated by Ellie Sherrard-Smith *et al* @sherrard2022.
+Please note that gamman is provided in units of years here, these will need to be converted to days for use in
+malariasimulaiton.
+
+## IRS
+
+### IRS coverage
+
+`site_file$interventions$irs_cov`
+
+As with ITNs, due to differences in the availability of data sources the approach
+for countries within sub-Saharan Africa differs to countries outside of
+sub-Saharan Africa:
+
+#### Within sub-Saharan Africa
+
+The population at risk weighted mean IRS coverage estimates for each site are
+taken from the malaria atlas project raster entitled:
+"Indoor Residual Spraying (IRS) coverage version 2020" @MAP. Coverage estimates
+are rescaled such that the country-level estimate of the number of persons protected
+by IRS matches the number reported in the world malaria report @WMR.
+
+#### Outside of sub-Saharan Africa
+
+IRS coverage is much more heterogeneous outside of SSA and data are less systematically
+collected. As a result, there are strong assumptions associated with the historical
+scale and magnitude of IRS campaigns. We make the assumption that any reported
+IRS campaigns (as detailed by the world malaria report @WMR) are tarted to areas
+that have > 1% PfPr or > 1% PvPr at baseline. Coverage in these regions
+are then scaled so that that number of persons protected by IRS is aligned with the
+world malaria report.
+
+### IRS insecticide
+
+`site_file$interventions$irs_insecticide`
+
+It is assumed that a DDT-type insecticide is used prior to 2017, after which
+there is a switch to an actellic-like insecticide.
+
+Current available IRS insecticide options are: ddt, actellic, bendiocarb,
+sumishield.
+
+### Number of rounds of IRS per year
+
+`site_file$interventions$irs_spray_rounds`
+
+We assume a single IRS spray round per year.
+
+### IRS efficacy parameters
+
+`site_file$interventions$ls_theta`
+`site_file$interventions$ls_gamma`
+`site_file$interventions$ks_theta`
+`site_file$interventions$ks_gamma`
+`site_file$interventions$ms_theta`
+`site_file$interventions$ms_gamma`
+
+Given an IRS insecticide type we can link to the
+corresponding estimates of the key IRS efficacy parameters. These have been
+estimated by Ellie Sherrard-Smith *et al* @sherrard2018.
+
+### IRS households sprayed
+
+It is often helpful to convert the number of persons protected by IRS into an
+estimate of the number of households covered. To aid this conversion we have
+included country-levels estimates of the average household size obtained from
+the UN @UNHH.
+
+`site_file$interventions$hh_size`
+
+### Missing years
+
+Available data on IRS coverage (via MAP @MAP or the world malaria report @WMR) will not extend
+to the present year. Missing IRS coverage estimates to present are filled assuming a
+constant, continuing level of coverage.
+
+## Treatment
+
+### Coverage
+
+`site_file$interventions$tx_cov`
+
+The population at risk weighted mean treatment coverage of an effective
+antimalarial for each site are taken from the malaria atlas project raster entitled:
+"Effective treatment with an Antimalarial drug version 2020" @MAP.
+
+### Drug type
+
+`site_file$interventions$prop_act`
+
+We estimate the proportion of treatments that are with an ACT from
+DHS StatCompiler data @DHS, using the indicator:
+"Children who took any ACT" (ID: ML_AMLD_C_ACT).
+For SSA estimates by year are expanded by linear interpolation between data
+points and an assumption of constant coverage after the most recent data point.
+We assume that ACT coverage is zero before 2006, when the WHO recommendation was
+first issued. For outside of SSA the DHS indicator is confounded by treatment
+for Plasmodium vivax, and we therefore assume the mean values by year from data
+within SSA.
+
+### Drug provider
+
+`site_file$interventions$prop_public`
+
+We include an estimate of the proportion of treatments that are from the public
+sector `prop_public`. This is useful for costing.
+We use the DHS StatCompiler @DHS
+indicator "Children with fever for whom advice or treatment was sought,
+the source was a public sector facility" (ID: ML_FEVA_C_PUB). We assume a
+constant proportion over time by country, estimated as the mean from all
+country survey estimates since 2010. For countries without survey data, we
+assume the median across all estimates.
+
+## Seasonal malaria chemoprevention (SMC)
+
+### SMC coverage
+
+`site_file$interventions$smc_cov`
+
+Historical SMC implementation and coverage estimates are fragmented. We
+identify historical SMC implementation areas from maps presented by both
+Access SMC @access_SMC and more recently
+SMC alliance @SMC_alliance. We assume a linear increase in
+coverage post implementation initiation up to a maximum of 80% to capture an
+increasing number of smaller sub-national units being targeted over time.
+
+### SMC drug
+
+`site_file$interventions$smc_drug`
+
+We assume that SP-AQ is used for SMC. This is currently the only available drug
+option.
+
+### Number of SMC rounds delivered annualy
+
+`site_file$interventions$smc_n_rounds`
+
+We assumed that historical SMC is delivered over 4 rounds `smc_n_rounds`.
+
+### SMC age range
+
+`site_file$interventions$smc_min_age`
+`site_file$interventions$smc_max_age`
+
+We assume SMC is delivered to children aged between 3 months and 5 years.
+
+## RTS,S vaccine
+
+`site_file$interventions$rtss_cov`
+
+We include historical RTS,S coverage that has occurred as part of the MVIP
+implementation trial, sub-nationally in Malawi, Ghana and Kenya. The spatial
+distribution is informed from an
+MVIP briefing presentation @MVIP
+
+## Perennial malaria chemoprevention (PMC).
+
+This intervention has been known in the past as intermittent preventative treatment
+of infants (IPTi).
+
+### PMC coverage
+
+`site_file$interventions$pmc_cov`
+
+Due to the very limited (non-trial setting) implementation of PMC historically,
+we assume 0 coverage.
+
+### PMC drug
+
+`site_file$interventions$pmc_drug`
+
+We assume PMC would be implemented with SP. This is currently the only available
+drug option.
+
+## Citations
diff --git a/vignettes/Metadata.Rmd b/vignettes/Metadata.Rmd
new file mode 100644
index 0000000..6aa727f
--- /dev/null
+++ b/vignettes/Metadata.Rmd
@@ -0,0 +1,50 @@
+---
+title: "Metadata"
+output: rmarkdown::html_vignette
+bibliography: references.bib
+csl: nature.csl
+link-citations: true
+vignette: >
+ %\VignetteIndexEntry{Metadata}
+ %\VignetteEngine{knitr::rmarkdown}
+ %\VignetteEncoding{UTF-8}
+---
+
+```{r, include = FALSE}
+knitr::opts_chunk$set(
+ collapse = TRUE,
+ comment = "#>"
+)
+```
+
+## Country
+
+`site_file$country`
+
+The ISO3c code identifying the country. The R package `countrycode` @countrycode
+is useful for converting between ISO3c and country names.
+
+## Version
+
+`site_file$version`
+
+The year and version of the site file.
+
+## Admininstrative level
+
+`site_file$admin_level`
+
+The administrative level (e.g. state, region, province) of the sites within the site file.
+
+## Sites
+
+`site_file$sites`
+
+The sites within the site file. These are the named sites disaggregated to the
+specified administrative level. Further disaggregation may include, for example,
+an [urban rural split](https://mrc-ide.github.io/site/articles/pop_demog.html).
+We use the GADM @GADM
+version 4.04 administrative boundary simple feature spatial files at the first
+or second administrative unit level.
+
+## Citations
diff --git a/vignettes/Seasonality.Rmd b/vignettes/Seasonality.Rmd
new file mode 100644
index 0000000..0e8fa21
--- /dev/null
+++ b/vignettes/Seasonality.Rmd
@@ -0,0 +1,30 @@
+---
+title: "Seasonality"
+output: rmarkdown::html_vignette
+bibliography: references.bib
+csl: nature.csl
+link-citations: true
+vignette: >
+ %\VignetteIndexEntry{Seasonality}
+ %\VignetteEngine{knitr::rmarkdown}
+ %\VignetteEncoding{UTF-8}
+---
+
+```{r, include = FALSE}
+knitr::opts_chunk$set(
+ collapse = TRUE,
+ comment = "#>"
+)
+```
+
+## Seasonality
+
+`site_file$seasonality`
+
+Daily rainfall global rasters for the period 2019-2021 were obtained from
+CHIRPS @CHIRPS using the
+[umbrella R package](https://mrc-ide.github.io/umbrella/). For each site we
+estimate the fourier series parameters representing general seasonal profiles.
+Please see the umbrella website for more information.
+
+## Citations
diff --git a/vignettes/historical_epi.Rmd b/vignettes/historical_epi.Rmd
new file mode 100644
index 0000000..8a5626c
--- /dev/null
+++ b/vignettes/historical_epi.Rmd
@@ -0,0 +1,46 @@
+---
+title: "Historical Epidemiological Data"
+output: rmarkdown::html_vignette
+bibliography: references.bib
+csl: nature.csl
+link-citations: true
+vignette: >
+ %\VignetteIndexEntry{Historical Epidemiological Data}
+ %\VignetteEngine{knitr::rmarkdown}
+ %\VignetteEncoding{UTF-8}
+---
+
+```{r, include = FALSE}
+knitr::opts_chunk$set(
+ collapse = TRUE,
+ comment = "#>"
+)
+```
+
+## Cases and deaths
+
+`site_file$cases_deaths`
+
+Where available this includes the time series of cases and deaths for the country
+as reported in previous editions of the world malaria report. Other useful data
+include the world malaria report @WMR population at risk and uncertainty intervals for
+estimates.
+
+## Prevalence
+
+`site_file$prevalence`
+
+Where available this includes the population at risk weighted average parasite
+prevalence time series from the Malaria Atlas Project @MAP
+
+_Plasmodium falciparuim_ parasite prevalence (`PfPr`) is reported for children aged
+between 2-10 years, summarized from the malaria atlas project raster entitled:
+"Plasmodium falciparum PR2 - 10 version 2020".
+
+_Plasmodium vivax_ parasite prevalence (`PvPr`) is reported for individuals aged
+between 1-99 years, summarized from the malaria atlas project raster entitled:
+"Plasmodium vivax PR1-99 version 2020".
+
+Prevalence estimates are used for [calibrating the baseline EIR](https://mrc-ide.github.io/site/articles/Calibration.html).
+
+## Citations
diff --git a/vignettes/nature.csl b/vignettes/nature.csl
new file mode 100644
index 0000000..7b058ba
--- /dev/null
+++ b/vignettes/nature.csl
@@ -0,0 +1,154 @@
+
+
diff --git a/vignettes/pop_demog.Rmd b/vignettes/pop_demog.Rmd
new file mode 100644
index 0000000..19c2fc7
--- /dev/null
+++ b/vignettes/pop_demog.Rmd
@@ -0,0 +1,63 @@
+---
+title: "Population and Demography"
+output: rmarkdown::html_vignette
+bibliography: references.bib
+csl: nature.csl
+link-citations: true
+vignette: >
+ %\VignetteIndexEntry{Population and Demography}
+ %\VignetteEngine{knitr::rmarkdown}
+ %\VignetteEncoding{UTF-8}
+---
+
+```{r, include = FALSE}
+knitr::opts_chunk$set(
+ collapse = TRUE,
+ comment = "#>"
+)
+```
+
+## Population projections
+
+`site_file$population`
+
+Site-level population estimates are extracted at the pixel level from
+WorldPop @worldpop. We use the "Unconstrained individual
+countries 2000-2020 UN adjusted ( 1km resolution )" rasters.
+
+### Urban rural
+
+Urban rural splits within administrative unit are defined using a threshold
+density of 1500 people per square km.
+
+### Population at risk
+
+`site_file$population$par_pf`
+`site_file$population$par_pv`
+`site_file$population$par`
+
+Population at risk from _Plasmodium falciparum_, _Plasmodium vivax_ or both are
+estimated by masking the total population raster by areas with active transmission
+(prevalence >0%) in year 2000 @MAP.
+
+### Population growth and urbanisation
+
+We provide annual population projections to 2050 using UN data on total population
+growth @unwpp combined with UN projections of
+the levels of urbanisation @UNWUP.
+We use these projections to produces estimates of the rate of urban and rural
+growth (relative to a baseline year) which are then applied to our spatial-unit
+population estimates. Final estimates are rescaled to ensure that
+population total match UN totals.
+
+## Demography
+
+Population demographics (age structure) are defined over time for each site.
+Demography is obtained using the [peeps R package](https://github.com/mrc-ide/peeps),
+and are based on data from the UN WPP @unwpp.
+
+Mortality rates are specified for neonates (0-30 days), young infants
+(31 days - 1 year), older infants (1 year - 5 years) and then in five year age
+bands.
+
+# Citations
diff --git a/vignettes/references.bib b/vignettes/references.bib
new file mode 100644
index 0000000..66f46b3
--- /dev/null
+++ b/vignettes/references.bib
@@ -0,0 +1,173 @@
+@Article{countrycode,
+ title = {countrycode: An R package to convert country names and country codes},
+ author = {Vincent Arel-Bundock and Nils Enevoldsen and CJ Yetman},
+ journal = {Journal of Open Source Software},
+ year = {2018},
+ volume = {3},
+ number = {28},
+ pages = {848},
+ url = {https://doi.org/10.21105/joss.00848},
+}
+
+@misc{GADM,
+ title={GADM maps and Data},
+ url={https://gadm.org/}
+}
+
+@misc{MAP,
+ title={Malaria Atlas Project},
+ url={https://malariaatlas.org/}
+}
+
+@Article{MAP_package,
+ author = {Daniel Pfeffer and Tim Lucas and Daniel May and Joseph Harris and Jennifer Rozier and Katherine Twohig and Ursula Dalrymple and Carlos Guerra and Catherine Moyes and Mike Thorn and Michele Nguyen and Samir Bhatt and Ewan Cameron and Daniel Weiss and Rosalind Howes and Katherine Battle and Harry Gibson and Peter Gething},
+ journal = {Malaria Journal},
+ volume = {17},
+ number = {1},
+ title = {malariaAtlas: an R interface to global malariometric data hosted by the Malaria Atlas Project.},
+ pages = {352},
+ year = {2018},
+ doi = {10.1186/s12936-018-2500-5},
+ publisher = {BioMed},
+}
+
+@report{WMR,
+ Author = {{World Health Organization}},
+ isbn = {9789240040496},
+ title = {World Malaria Report},
+ year = {2021},
+}
+
+@article{bertozzi,
+ abstract = {Insecticide-treated nets (ITNs) are one of the most widespread and impactful malaria interventions in Africa, yet a spatially-resolved time series of ITN coverage has never been published. Using data from multiple sources, we generate high-resolution maps of ITN access, use, and nets-per-capita annually from 2000 to 2020 across the 40 highest-burden African countries. Our findings support several existing hypotheses: that use is high among those with access, that nets are discarded more quickly than official policy presumes, and that effectively distributing nets grows more difficult as coverage increases. The primary driving factors behind these findings are most likely strong cultural and social messaging around the importance of net use, low physical net durability, and a mixture of inherent commodity distribution challenges and less-than-optimal net allocation policies, respectively. These results can inform both policy decisions and downstream malaria analyses.},
+ author = {Amelia Bertozzi-Villa and Caitlin A. Bever and Hannah Koenker and Daniel J. Weiss and Camilo Vargas-Ruiz and Anita K. Nandi and Harry S. Gibson and Joseph Harris and Katherine E. Battle and Susan F. Rumisha and Suzanne Keddie and Punam Amratia and Rohan Arambepola and Ewan Cameron and Elisabeth G. Chestnutt and Emma L. Collins and Justin Millar and Swapnil Mishra and Jennifer Rozier and Tasmin Symons and Katherine A. Twohig and T. Deirdre Hollingsworth and Peter W. Gething and Samir Bhatt},
+ doi = {10.1038/s41467-021-23707-7},
+ issn = {20411723},
+ issue = {1},
+ journal = {Nature Communications},
+ pages = {1-12},
+ pmid = {34117240},
+ publisher = {Springer US},
+ title = {Maps and metrics of insecticide-treated net access, use, and nets-per-capita in Africa from 2000-2020},
+ volume = {12},
+ url = {http://dx.doi.org/10.1038/s41467-021-23707-7},
+ year = {2021},
+}
+
+@article{sherrard2022,
+ abstract = {Background: Concern that insecticide resistant mosquitoes are threatening malaria control has driven the development of new types of insecticide treated nets (ITNs) and indoor residual spraying (IRS) of insecticide. Malaria control programmes have a choice of vector control interventions although it is unclear which controls should be used to combat the disease. The study aimed at producing a framework to easily compare the public health impact and cost-effectiveness of different malaria prevention measures currently in widespread use. Methods: We used published data from experimental hut trials conducted across Africa to characterise the entomological effect of pyrethroid-only ITNs versus ITNs combining a pyrethroid insecticide with the synergist piperonyl butoxide (PBO). We use these estimates to parameterise a dynamic mathematical model of Plasmodium falciparum malaria which is validated for two sites by comparing simulated results to empirical data from randomised control trials (RCTs) in Tanzania and Uganda. We extrapolated model simulations for a series of potential scenarios likely across the sub-Saharan African region and include results in an online tool (Malaria INtervention Tool [MINT]) that aims to identify optimum vector control intervention packages for scenarios with varying budget, price, entomological and epidemiological factors. Findings: Our model indicates that switching from pyrethroid-only to pyrethroid-PBO ITNs could averted up to twice as many cases, although the additional benefit is highly variable and depends on the setting conditions. We project that annual delivery of long-lasting, non-pyrethroid IRS would prevent substantially more cases over 3-years, while pyrethroid-PBO ITNs tend to be the most cost-effective intervention per case averted. The model was able to predict prevalence and efficacy against prevalence in both RCTs for the intervention types tested. MINT is applicable to regions of sub-Saharan Africa with endemic malaria and provides users with a method of designing intervention packages given their setting and budget. Interpretation: The most cost-effective vector control package will vary locally. Models able to recreate results of RCTs can be used to extrapolate outcomes elsewhere to support evidence-based decision making for investment in vector control. Funding: Medical Research Council, IVCC, Wellcome Trust. Translation: For the French translation of the abstract see Supplementary Materials section.},
+ author = {Ellie Sherrard-Smith and Peter Winskill and Arran Hamlet and Corine Ngufor and Raphael N'Guessan and Moussa W. Guelbeogo and Antoine Sanou and Rebecca K. Nash and Alexander Hill and Emma L. Russell and Mark Woodbridge and Patrick Tungu and Mara D. Kont and Tom Mclean and Christen Fornadel and Jason H. Richardson and Martin J. Donnelly and Sarah G. Staedke and Samuel Gonahasa and Natascha Protopopoff and Mark Rowland and Thomas S. Churcher},
+ doi = {10.1016/S2542-5196(21)00296-5},
+ issn = {25425196},
+ issue = {2},
+ journal = {The Lancet Planetary Health},
+ pages = {e100-e109},
+ pmid = {35065707},
+ title = {Optimising the deployment of vector control tools against malaria: a data-informed modelling study},
+ volume = {6},
+ year = {2022},
+}
+
+@article{sherrard2018,
+ abstract = {Indoor residual spraying (IRS) is an important part of malaria control. There is a growing list of insecticide classes; pyrethroids remain the principal insecticide used in bednets but recently, novel non-pyrethroid IRS products, with contrasting impacts, have been introduced. There is an urgent need to better assess product efficacy to help decision makers choose effective and relevant tools for mosquito control. Here we use experimental hut trial data to characterise the entomological efficacy of widely-used, novel IRS insecticides. We quantify their impact against pyrethroid-resistant mosquitoes and use a Plasmodium falciparum transmission model to predict the public health impact of different IRS insecticides. We report that long-lasting IRS formulations substantially reduce malaria, though their benefit over cheaper, shorter-lived formulations depends on local factors including bednet use, seasonality, endemicity and pyrethroid resistance status of local mosquito populations. We provide a framework to help decision makers evaluate IRS product effectiveness.},
+ author = {Ellie Sherrard-Smith and Jamie T. Griffin and Peter Winskill and Vincent Corbel and Cédric Pennetier and Armel Djénontin and Sarah Moore and Jason H. Richardson and Pie Müller and Constant Edi and Natacha Protopopoff and Richard Oxborough and Fiacre Agossa and Raphael N’Guessan and Mark Rowland and Thomas S. Churcher},
+ doi = {10.1038/s41467-018-07357-w},
+ issn = {2041-1723},
+ issue = {1},
+ journal = {Nature Communications},
+ keywords = {Epidemiology,Malaria,Population dynamics,Public health},
+ pages = {4982},
+ pmid = {30478327},
+ title = {Systematic review of indoor residual spray efficacy and effectiveness against Plasmodium falciparum in Africa},
+ volume = {9},
+ url = {http://www.nature.com/articles/s41467-018-07357-w},
+ year = {2018},
+}
+
+@misc{UNHH,
+ Author = {{United Nations}},
+ title={Household Size and Composition},
+ url={https://www.un.org/development/desa/pd/data/household-size-and-composition}
+}
+
+@misc{DHS,
+ Author = {{USAID}},
+ title={The DHS Program},
+ url={https://dhsprogram.com/}
+}
+
+@misc{DHS,
+ Author = {{USAID}},
+ title={The DHS Program},
+ url={https://dhsprogram.com/}
+}
+
+@misc{access_SMC,
+ title={Access SMC},
+ url={https://www.access-smc.org/}
+}
+
+@misc{SMC_alliance,
+ title={SMC Alliance},
+ url={https://www.smc-alliance.org/}
+}
+
+@misc{MVIP,
+ Author = {{World Health Organization}},
+ title={MVIP briefing},
+ url={https://www.givewell.org/files/DWDA%202009/PATH/WHO_WHO_Malaria_Vaccine_Implementation_Program_Briefing_2021.pdf}
+}
+
+@misc{worldpop,
+ title={WorldPop},
+ url={https://www.worldpop.org/}
+}
+
+@misc{UNWPP,
+ Author = {{United Nations}},
+ title={World Population Prospects},
+ url={https://population.un.org/wpp/}
+}
+
+@misc{UNWUP,
+ Author = {{United Nations}},
+ title={World Urbanization Prospects},
+ url={https://population.un.org/wup/}
+}
+
+@misc{CHIRPS,
+ title={CHIRPS},
+ url={https://www.chc.ucsb.edu/data/chirps}
+}
+
+@article{Sinka2016,
+ abstract = {Background: Malaria remains a heavy burden across sub-Saharan Africa where transmission is maintained by some of the world's most efficient vectors. Indoor insecticide-based control measures have significantly reduced transmission, yet elimination remains a distant target. Knowing the relative abundance of the primary vector species can provide transmission models with much needed information to guide targeted control measures. Moreover, understanding how existing interventions are impacting on these relative abundances highlights where alternative control (e.g., larval source management) is needed. Methods: Using the habitat suitability probabilities generated by predictive species distribution models combined with data collated from the literature, a multinomial generalized additive model was applied to produce relative abundance estimates for Anopheles arabiensis, Anopheles funestus and Anopheles gambiae/Anopheles coluzzii. Using pre- and post-intervention abundance data, estimates of the effect of indoor insecticide-based interventions on these relative abundances were made and are illustrated in post-intervention maps. Results: Conditional effect plots and relative abundance maps illustrate the individual species' predicted habitat suitability and how they interact when in sympatry. Anopheles arabiensis and An. funestus show an affinity in habitat preference at the expense of An. gambiae/An. coluzzii, whereas increasing habitat suitability for An. gambiae/An. coluzzii is conversely less suitable for An. arabiensis but has little effect on An. funestus. Indoor insecticide-based interventions had a negative impact on the relative abundance of An. funestus, and a lesser effect on An. arabiensis. Indoor residual spraying had the greatest impact on the relative abundance of An. funestus, and a lesser effect on An. gambiae/An. coluzzii. Insecticide-treated bed nets reduced the relative abundance of both species equally. These results do not indicate changes in the absolute abundance of these species, which may be reduced for all species overall. Conclusions: The maps presented here highlight the interactions between the primary vector species in sub-Saharan Africa and demonstrate that An. funestus is more susceptible to certain indoor-based insecticide interventions than An. gambiae/An. coluzzii, which in turn, is more susceptible than An. arabiensis. This may provide An. arabiensis with a competitive advantage where it is found in sympatry with other more endophilic vectors, and potentially increase the need for outdoor-based vector interventions to deal with any residual transmission barring the way to malaria elimination.},
+ author = {Marianne E. Sinka and Nick Golding and N. Claire Massey and Antoinette Wiebe and Zhi Huang and Simon I. Hay and Catherine L. Moyes},
+ doi = {10.1186/s12936-016-1187-8},
+ issn = {14752875},
+ issue = {1},
+ journal = {Malaria Journal},
+ keywords = {Africa,Anopheles,Insecticide control,Malaria,Mosquito,Relative abundance,Vector},
+ month = {3},
+ pmid = {26945997},
+ publisher = {BioMed Central},
+ title = {Modelling the relative abundance of the primary African vectors of malaria before and after the implementation of indoor, insecticide-based vector control},
+ volume = {15},
+ year = {2016},
+}
+
+@article{Sinka2012,
+ abstract = {BACKGROUND: Global maps, in particular those based on vector distributions, have long been used to help visualise the global extent of malaria. Few, however, have been created with the support of a comprehensive and extensive evidence-based approach.\n\nMETHODS: Here we describe the generation of a global map of the dominant vector species (DVS) of malaria that makes use of predicted distribution maps for individual species or species complexes.\n\nRESULTS: Our global map highlights the spatial variability in the complexity of the vector situation. In Africa, An. gambiae, An. arabiensis and An. funestus are co-dominant across much of the continent, whereas in the Asian-Pacific region there is a highly complex situation with multi-species coexistence and variable species dominance.\n\nCONCLUSIONS: The competence of the mapping methodology to accurately portray DVS distributions is discussed. The comprehensive and contemporary database of species-specific spatial occurrence (currently available on request) will be made directly available via the Malaria Atlas Project (MAP) website from early 2012.},
+ author = {Marianne E Sinka and Michael J Bangs and Sylvie Manguin and Yasmin Rubio-Palis and Theeraphap Chareonviriyaphap and Maureen Coetzee and Charles M Mbogo and Janet Hemingway and Anand P Patil and William H Temperley and Peter W Gething and Caroline W Kabaria and Thomas R Burkot and Ralph E Harbach and Simon I Hay},
+ doi = {10.1186/1756-3305-5-69},
+ isbn = {1756-3305 (Electronic)\n1756-3305 (Linking)},
+ issn = {1756-3305},
+ issue = {January 2008},
+ journal = {Parasites & vectors},
+ pages = {69},
+ pmid = {22475528},
+ title = {A global map of dominant malaria vectors.},
+ volume = {5},
+ year = {2012},
+}
+
diff --git a/vignettes/vectors.Rmd b/vignettes/vectors.Rmd
new file mode 100644
index 0000000..ebbda0f
--- /dev/null
+++ b/vignettes/vectors.Rmd
@@ -0,0 +1,65 @@
+---
+title: "Mosquito Vectors"
+output: rmarkdown::html_vignette
+bibliography: references.bib
+csl: nature.csl
+link-citations: true
+vignette: >
+ %\VignetteIndexEntry{Mosquito Vectors}
+ %\VignetteEngine{knitr::rmarkdown}
+ %\VignetteEncoding{UTF-8}
+---
+
+```{r, include = FALSE}
+knitr::opts_chunk$set(
+ collapse = TRUE,
+ comment = "#>"
+)
+```
+
+
+`site_file$vectors`
+
+## Within sub-Saharan Africa
+
+For countries in sub-Saharan Africa a statistical model of relative abundance of
+major vectors has been produced
+Sinka *et al* @Sinka2016
+For countries we extract the relative abundance of
+arabiensis, funestus and gambiae using the malaria atlas project
+"Anopheles arabiensis Patton, 1905", "Anopheles funestus" and
+"Anopheles gambiae Giles, 1902" rasters.
+Proportions are normalised to sum to one (as other species may be present in the data).
+
+## Outside of sub-Saharan Africa
+
+For countries outside of sub-Saharan Africa we do not have estimates of the relative
+abundance of vectors species. There are however probability of occurrence maps for
+a large number of species
+Sinka et al @Sinka2012.
+For these locations we select up to a maximum of three species ranked by their
+average probability of occurrence across a site. Each species is then given an
+equally weighted relative abundance.
+
+## Vector bionomics
+
+`site_file$vectors$blood_meal_rates`
+`site_file$vectors$foraging_time`
+`site_file$vectors$Q0`
+`site_file$vectors$phi_bednets`
+`site_file$vectors$phi_indoors`
+`site_file$vectors$mum`
+
+The bionomics parameters for vectors in sub-Saharan Africa are parameterised
+using the fitted parameters in
+[malarisimulation](https://mrc-ide.github.io/malariasimulation/).
+
+All other vectors are parameterised from an (unpublished) literature review
+by Arran Hamlet. Any unknown species are assigned the median vector bionomics
+parameters across other species at the site. Please contact Pete Winskill
+directly if using these estimates.
+
+The blood_meal_rates, foraging_time and mum parameters are assumed fixed at
+previously fitted values.
+
+## Citations