Skip to contents

Mass Treatment

Source: vignettes/MassTreatment.Rmd
MassTreatment.Rmd

ramp.control has utilities to set up and simulate mass treatment, including:

  • mass drug administration (MDA)
  • mass screen and treat (MSAT)

In the following, we demonstrate how, and then we discuss the implementation.

Basic Setup

The first step is basic setup. Before setting up MDA, it’s a good idea to check that module is capable of handling mass treatment. To see if a module supports it, look at the skill set. For example, we can check whether the SIS supports MDA or MSAT.

skill_set_XH("SIS")$mda
## [1] TRUE

The SIS module supports MDA.

The next step is basic setup:

base_model <- xds_setup_eir(Xname = "SIS", 
                       eir=1/365, 
                       season_par = makepar_F_sin(phase=120))

Next we solve it.

base_model <- burnin(base_model) 
base_model <- xds_solve(base_model, 1095,5)
xds_plot_PR(base_model)

Mass Treatment Events

Adding mass treatment is an advanced setup option. The function call is setup_mass_treat_events. To use it, we need to pass:

  • start – the julian date(s)

  • span – the length of the event

  • frac_tot – the fraction that will be treated (for mda) or tested (for msat)

  • test – whether testing is done

    • test=TRUE configures MSAT

    • test=FALSE configures MDA

This sets up two rounds of MDA starting on days 180 and 480. Each round lasts 10 days, and treats 90% of the population.

start = c(180, 480)
span = c(10,10)
frac_tot = c(0.9, 0.9)
test = c(FALSE, FALSE) 
mda_model <- setup_mass_treat_events(base_model, start, span, frac_tot, test)

Now we solve the base_model with MDA and compare it to the baseline:

mda_model <- xds_solve(mda_model, 1095,5)
xds_plot_PR(base_model)
xds_plot_PR(mda_model, add=T, clrs = "darkred")

Add Events

To add another round of mass treatment, use add_mass_treat_events. This adds an extra round on day 600 that is like the others 2 rounds.

tt <- seq(0, 1095, by = 1)
mda_model_1 <- add_mass_treat_events(mda_model, 600, 10, .9, FALSE)
mda_model_1 <- xds_solve(mda_model_1, 1095,5)
show_mda(tt, mda_model_1)

If no mass treatment events have been set up, the function calls setup_mass_treat_events. This adds a mass treatment event to the base_model

mda_model_2 <- add_mass_treat_events(base_model, 600, 10, .9, FALSE)
mda_model_2 <- xds_solve(mda_model_2, 1095,5)
show_mda(tt, mda_model_2)

This plot compares all four models:

xds_plot_PR(mda_model_2, clrs="darkblue")
xds_plot_PR(mda_model_1, add=T, clrs = "purple3")
xds_plot_PR(base_model, add=T)
xds_plot_PR(mda_model, add=T, clrs = "darkred")

Notes

Mass Treatment Ports

To understand how the mass treatment ports are set up, we can look inside the SIS module. The relevant lines in dXHdt.SIS are these:

r_t <- r + mda(t) + d_rdt*msat(t)
dI <- foi*(H-I) - r_t*I + D_matrix %*% I

In this case, both mda and msat work like the parameter \(r,\) so inside the code, a term is computed: r_t = r+mda(t)+d_rdt*msat(t), where d_rdt is a parameter that describes the probability of testing positive by RDT. Perhaps most importantly, mass treatment functions return a per-capita rate.

This model explained in the documentation for the function:

help(dXHdt.SIS)

During setup, The functions mda and msat are assigned to a function that returns \(0.\) Advanced setup replaces mda or msat with a new function.

Mass Treatment Functions

Treatment during the span of an event occurs at a fixed rate. The shape of the treatment curve is a variant of a sharkfin function:

tt <- seq(598, 612, by = .1)
show_mda(tt, mda_model_1)

The total area under the curve is calibrated to reach the right fraction treated (for mda) or tested (for msat): in this case, 90%. We can check this:

mda <- mda_model_1$mass_treat_obj$mda_obj$F_treat
t_frac = 1-exp(-integrate(mda, 595, 615)$val)
approx_equal(.9, t_frac, tol = 1e-6) 
## [1] TRUE