Introduction

SIR model is an acronym of three concepts involved in the model: S for Susceptible, I

for infected and R for the removed. This model is fascinating for me as I was amazed by the

way how a simple model could be used to predict a deathly epidemic. In order to approach

this topic, we firstly need to understand the real life situation which is the cholera outbreak in

South Sudan and we need to understand what SIR modelling actually is. After we have

understood these two crucial areas for this investigation we can bring them into a relation and

investigate this real life case through this mathematical area which is most definitely

appropriate.

Cholera
First of all, by World Health organization cholera is defined as an acute diarrheal

disease that can kill within hours if left untreated. Researchers have estimated that each year

there are 1.3 million to 4.0 million cases of cholera, and 21 000 to 143 000 deaths worldwide

due to cholera. Most of those infected will have no or mild symptoms, and can be

successfully treated with oral rehydration solution. Severe cases will need rapid treatment

with intravenous fluids and antibiotics. Provision of safe water and sanitation is critical to

control the transmission of cholera and other waterborne diseases. Safe oral cholera vaccines

should be used in conjunction with improvements in water and sanitation to control cholera

outbreaks and for prevention in areas known to be high risk for cholera. A global strategy on

cholera control with a target to reduce cholera deaths by 90% was launched in 20171.

Cholera outbreak in South Sudan

1
World Health Organization page visited on 28.10.2018
http://www.who.int/news-room/fact-sheets/detail/cholera
 The fight against cholera in South Sudan has involved a range of agencies such as

World Health Organization, UN etc. working together to enhance surveillance, deploy rapid

response teams to investigate and respond to cases, provide clean water, promote good

hygiene practices and treat cholera patients. To enhance outbreak response efforts, the

government worked with the European Union Humanitarian Aid (ECHO), GAVI, the

Vaccine Alliance, the United States Agency for International Development (USAID) and the

World Health Organization (WHO), securing 2.2 million doses of the Oral Cholera Vaccine

(OCV) from the Gavi-funded global stockpile. In 2017 more than 885,000 people at higher

risk of cholera were immunized in the first round and nearly 500,000 people also received a

second round of the vaccine. Due to security challenges, not everyone was able to receive the

recommended two doses, which would significantly decrease their risk of being affected by

cholera. “Cholera is a virulent disease which spreads when hygiene and sanitation are

inadequate,” said Evans Liyosi, WHO Acting Representative to South Sudan. “The outbreak

was declared on 18 June 2016 and spread to many parts of the country, including the capital

Juba. By the time the last confirmed cholera case was discharged on 18 December 2017, over

20,000 suspected cholera cases and 436 deaths were reported.

The country is dealing with several complex health emergencies with 5.1 million

people in need of health assistance. Armed conflict has forced almost 4 million people to flee

their homes. Nearly 5 million people, more than 40% of the population, are severely food

insecure. These challenges place a huge burden on the country’s health system, while the

sanitation infrastructure needs further strengthening2.

SIR modelling

2
World Health Organization Regional Office for Africa page visited on 28.10.2018
https://afro.who.int/news/south-sudan-declares-end-its-longest-cholera-outbreak
 SIR model is a quantative model that explains the dynamics of epidemics. This model

takes into account the possibility of a disease-contracted patient to recover and become

immune to infection in the future.

After we have introduced our real life situation and understood what cholera is we

may proceed to investigation about SIR modelling which is the mathematical part of this

investigation. An SIR model is an epidemiological model that computes the theoretical

number of people infected with a contagious illness in a closed population over time. The

name of this class of models derives from the fact that they involve coupled equations

relating the number of susceptible people S(t), number of people infected I(t), and number of

people who have recovered R(t). One of the simplest SIR models is the Kermack-

McKendrick model. In order to put this real life example into the SIR modelling we firstly

need to get few numbers. So we need to have number of susceptible people to the cholera in

South Sudan, after a short investigation we have come to a conclusion that around 973,980

people were susceptible to cholera in South Sudan during this outbreak. Furthermore, we

have investigated as well that around 620,122 people were infected and about 413,873 people

recovered3. After knowing these stats we need to know the time which is required for

equations in order to plot the graph as precisely as possible. Cholera outbreak was officially

clamed on 18th June 2016 and it finished on 18th December 2017 making it 548 days. After

we know how many people are infected, how many recovered and how many susceptible and

we know the time in days we may proceed further.

This is an ordinary differential equation model, described by the following equation:

3
World Health Organization Regional Office for Africa page visited on 28.10.2018
https://afro.who.int/news/south-sudan-declares-end-its-longest-cholera-outbreak
After we know all of this we may proceed to plotting a graph.

S0= 973,980

I0=620,122

R0=413,873

T=548

Methods of SIR
The SIR model is the following system of quadratic ODEs:
  dS = −β S I (1) dt

 dI =βSI−νI (2) dt

 dR = ν I, (3) dt

where the disease transmission rate β > 0 and the recovery rate ν > 0 (or in other words, the

duration of infection D = 1/ν).

The bi-linear incidence term β S I for the number of new infected indi- viduals per unit time

corresponds to homogeneous mixing of the infected and susceptible classes. The total

population size should remain constant, and this easily follows from the SIR system: that the

sum of the left hand sides of the three equations is the derivative of the total population size

and the sum of the right hand sides is zero. We denote the total population size by N. Since

R(t) = N − S(t) − I(t), the system can be reduced to a system of two ODEs: (1) and (2).

Suppose that each infected individual has κ contacts (each sufficient for transmission) per

unit time and κ is independent of the population size. Then κ S/N of these contacts are with

susceptible individuals. If the fraction τ of adequate contacts result in transmission, then each

infected individual infects κ τ S/N susceptible individuals per unit time. Thus β = b/N where

b = κ τ . The parameter τ is called the transmissibility of the infectious disease.

Analysis of the SIR Model

Since the right hand side of (1) is negative and the right hand side of (3) is positive,

this implies that dS/dt ≤ 0 and dR/dt ≥ 0. Since 0 ≤ S(t) ≤ S(0) ≤ N and 0 ≤ R(0) ≤ R(t) ≤ N,

this implies that the limits S(∞) = limt→∞ S(t), R(∞) = limt→∞ R(t), and thus I(∞) =

limt→∞ I(t) = N − S(∞) − R(∞) exist.

The disease always dies out

It is also easy to prove that the disease always dies out, I(∞) = 0 for all initial conditions,

without having a formula for I(t). If not, (3) implies that for t sufficiently large, dR/dt >

νI(∞)/2 > 0, and this implies that R(∞) = ∞, a contradiction.

Epidemic threshold theorem

We define the effective reproductive number Re = (S(0)/N)b/ν and the basic reproductive

number R0 = b/ν. If the entire population is initially suscep- tible, i.e., S(0) = N − 1, I(0) = 1,

R(0) = 0, and large (recall this is a model assumption), then Re = ((N − 1)/N ) b/ν is

approximately equal to R0. Henceforth, to beautify formulas involving R0, we will assume

that the quantity (N − 1)/N is equal to 1.

We now show that Re is the threshold value or tipping point that deter- mines whether an

infectious disease will quickly die out or whether it will invade the population and cause an

epidemic.

Theorem 2.1. 1. If Re ≤ 1, then I(t) decreases monotonically to zero as t → ∞.

2. If Re > 1, then I(t) starts increasing, reaches its maximum, and then decreases to zero as t

→ ∞. We call this scenario of increasing numbers of infected individuals an epidemic.

It follows that an infection can invade and cause an epidemic in an entirely susceptible

population if R0 > 1 or b > ν.

Proof. Equation (2) and the discussion in Section 2.2.1 imply that dI/dt = (β S − ν) I ≤ (β S(0)

− ν) I = ν (Re − 1) I ≤ 0 for Re < 1. This observation together with I(∞) = 0 (see Section

2.2.2) proves the first statement.

Equation (2) implies (dI/dt)(0) = ν (Re − 1) I(0) > 0 for Re > 1. Thus I(t) is increasing at t = 0.

Equation (2) also implies that I(t) has only one non-zero critical point. These observations,

together with I(∞) = 0 imply the second statement.

Figure 1 contains solutions of the SIR system simulating a highly virulent (Re = 3.5) flu

epidemic in a town of 50, 000 people.

Figure 1: Solutions of SIR system of ODEs with β = 0.7/50000,ν = 1/5,S(0) = 49955, I(0)
= 5, R(0) = 0

We stress that the existence of a threshold for infection is far from obvious and was missed

by many public health and infectious disease experts. The reason is that such a threshold can

not be discerned from data; it requires a mathematical model to illuminate.

Above we observed that (dI/dt)(0) = ν (Re − 1) I(0), which implies that the number of

infected individuals initially starts growing/decreasing expo- nentially at rate ν (Re − 1). The

next section will provide strong intuition for the exponential growth.
Up to this point practically every differential equation that we’ve been presented with could

be solved. The problem with this is that these are the exceptions rather than the rule. The vast

majority of first order differential equations can’t be solved.

In order to teach you something about solving first order differential equations we’ve had to

restrict ourselves down to the fairly restrictive cases of linear, separable, or exact differential

equations or differential equations that could be solved with a set of very specific

substitutions. Most first order differential equations however fall into none of these

categories. In fact, even those that are separable or exact cannot always be solved for an

explicit solution. Without explicit solutions to these it would be hard to get any information

about the solution.

So, what do we do when faced with a differential equation that we can’t solve? The answer

depends on what you are looking for. If you are only looking for long term behavior of a

solution you can always sketch a direction field. This can be done without too much difficulty

for some fairly complex differential equations that we can’t solve to get exact solutions.

The problem with this approach is that it’s only really good for getting general trends in

solutions and for long term behavior of solutions. There are times when we will need

something more. For instance, maybe we need to determine how a specific solution behaves,

including some values that the solution will take. There are also a fairly large set of

differential equations that are not easy to sketch good direction fields for.

In these cases, we resort to numerical methods that will allow us to approximate solutions to

differential equations. There are many different methods that can be used to approximate

solutions to a differential equation and in fact whole classes can be taught just dealing with
the various methods. We are going to look at one of the oldest and easiest to use here. This

method was originally devised by Euler and is called, oddly enough, Euler’s Method.

Let’s start with a general first order IVP

where f(t,y)f(t,y) is a known function and the values in the initial condition are also known

numbers. From the second theorem in the Intervals of Validity section we know that

if ff and fyfy are continuous functions then there is a unique solution to the IVP in some

interval surrounding t=t0t=t0. So, let’s assume that everything is nice and continuous so that

we know that a solution will in fact exist.

We want to approximate the solution to (1) near t=t0t=t0. We’ll start with the two pieces of

information that we do know about the solution. First, we know the value of the solution

at t=t0t=t0 from the initial condition. Second, we also know the value of the derivative

at t=t0t=t0. We can get this by plugging the initial condition into f(t,y)f(t,y) into the

differential equation itself. So, the derivative at this point is.

Now, recall from your Calculus I class that these two pieces of information are enough for us

to write down the equation of the tangent line to the solution at t=t0t=t0. The tangent line is

Take a look at the figure below

If t1t1 is close enough to t0t0 then the point y1y1 on the tangent line should be fairly close to

the actual value of the solution at t1t1, or y(t1)y(t1). Finding y1y1 is easy enough. All we

need to do is plug t1t1 in the equation for the tangent line.

Now, we would like to proceed in a similar manner, but we don’t have the value of the

solution at t1t1 and so we won’t know the slope of the tangent line to the solution at this

point. This is a problem. We can partially solve it however, by recalling that y1y1 is an

approximation to the solution at t1t1. If y1y1 is a very good approximation to the actual value

of the solution then we can use that to estimate the slope of the tangent line at t1t1.

So, let’s hope that y1y1 is a good approximation to the solution and construct a line through

the point (t1,y1t1,y1) that has slope f(t1,y1f(t1,y1). This gives

Now, to get an approximation to the solution at  t=t2t=t2 we will hope that this new line will

be fairly close to the actual solution at t2t2 and use the value of the line at t2t2 as an

approximation to the actual solution. This gives.

We can continue in this fashion. Use the previously computed approximation to get the next

approximation. So,

In general, if we have tntn and the approximation to the solution at this point, ynyn, and we

want to find the approximation at tn+1tn+1 all we need to do is use the following.

yn+1=yn+f(tn,yn)⋅(tn+1−tn)yn+1=yn+f(tn,yn)⋅(tn+1−tn)

If we define fn=f(tn,yn)fn=f(tn,yn) we can simplify the formula to

yn+1=yn+fn⋅(tn+1−tn)(2)(2)yn+1=yn+fn⋅(tn+1−tn)

Often, we will assume that the step sizes between the points t0t0 , t1t1 , t2t2 , … are of a

uniform size of hh. In other words, we will often assume that

tn+1−tn=htn+1−tn=h

This doesn’t have to be done and there are times when it’s best that we not do this. However,

if we do the formula for the next approximation becomes.

yn+1=yn+hfn(3)(3)yn+1=yn+hfn

So, how do we use Euler’s Method? It’s fairly simple. We start with (1)(1) and decide if we

want to use a uniform step size or not. Then starting with (t0,y0)(t0,y0) we repeatedly

evaluate  (2) or  (3) depending on whether we chose to use a uniform step size or not. We

continue until we’ve gone the desired number of steps or reached the desired time. This will

give us a sequence of numbers y1y1 , y2y2 , y3y3 , … ynyn that will approximate the value

of the actual solution at t1t1 , t2t2 , t3t3 , … tntn.

What do we do if we want a value of the solution at some other point than those used here?

One possibility is to go back and redefine our set of points to a new set that will include the

points we are after and redo Euler’s Method using this new set of points. However, this is

cumbersome and could take a lot of time especially if we had to make changes to the set of

points more than once.

Another possibility is to remember how we arrived at the approximations in the first place.

Recall that we used the tangent line

y=y0+f(t0,y0)(t−t0)y=y0+f(t0,y0)(t−t0)

to get the value of y1y1. We could use this tangent line as an approximation for the solution

on the interval [t0,t1][t0,t1]. Likewise, we used the tangent line

y=y1+f(t1,y1)(t−t1)y=y1+f(t1,y1)(t−t1)

to get the value of y2y2. We could use this tangent line as an approximation for the solution

on the interval [t1,t2][t1,t2]. Continuing in this manner we would get a set of lines that, when

strung together, should be an approximation to the solution as a whole.

Conclusion
To conclude we may see that mathematics may help us in different aspects of life. For

example, thanks to the SIR model we are able to understand more efficiently cholera

outbreak in South Sudan as many other epidemic situations. It helps us predict situation and it

is more than useful. However there are few limitations with this model such as the fact that
this model takes the population as homogeneous and fully mixed etc. Nonetheless this model

and this work helped me connect mathematics with a real life situation.