T5b - Interventions (detecting typhoid carriers)

This tutorial shows how to use a routine diagnostic intervention, that administers a test (the product) to a fraction of the eligible population. By default the eligible population is acute individuals. The eligibility criteria can can optionally passed to the intervention as a function that returns the indices (uids) of the eligibile individuals, or directly as an array of uids.

The sensitivity of the diagnostic product (the test), understood as the probability of receiving a positive diagnostic given that the individual is infected, can be set when instantiating the product.

import starsim as ss
import typhoidsim as ty

# Define high-level simulation parameters
pars = dict(
    start    = 2000,  # Starting year
    dur      = 2.0,   # Duration of the simulation in years
    dt       = 7.0/365.0,     # Timestep of 1 week, expressed in years
    verbose  = 0,             # Do not print details of the run
)

# The population
ppl = ss.People(50_000)

# Person to person transmission
random_p2p = ss.RandomNet({'n_contacts': 10})

# The disease
typhoid = ty.Typhoid(pars={'tppi':0.1, 'init_seroprev':0.8, 'init_prev':0.001})

# Create an intervention that routine screens symptoimatic (acute) inidividuals. 
# The diagnostic product is a test that is equivalent to a blood test with sensitivity of 0.65
blood_test   = ty.typhoid_test(pars=dict(sensitivity=ss.bernoulli(p=0.65)))
screen_acute = ty.routine_acute_screening(product=blood_test, prob=0.3, start_year=2001.0) # Screen 30% of acute individuals every year starting in 2001


# Now create two sims: a baseline sim and one with the intervention
sim = ss.Sim(pars=pars, 
             people=ppl, 
             networks=random_p2p,
             diseases=typhoid, 
             interventions=screen_acute)
sim.run()

Sim(n=50000; 2000—2002.0; networks=randomnet; diseases=typhoid; interventions=routine_acute_screening)

sim.plot()

Figure(1152x864)

import matplotlib.pyplot as plt

plt.figure()
plt.plot(sim.timevec, sim.interventions.routine_acute_screening.results.new_screened, label='New screened acute carriers')
plt.plot(sim.timevec,  sim.interventions.routine_acute_screening.results.new_positive, label='New screened who tested positive')

plt.axvline(x=2001, color='k', ls='--')
plt.legend()
plt.show()

plt.figure()
plt.plot(sim.timevec, sim.interventions.routine_acute_screening.results.positivity, label='Instantaneous Positivity')

plt.axvline(x=2001, color='k', ls='--')
plt.legend()
plt.show()