CONTROLLING AND PREVENTING DISEASE

Chapter 3

Disease in the population

This chapter introduces the dynamics of communicable disease in a population.

We look at immunity, endemic and epidemic occurrence of disease, some epide-
miological concepts, and we considermortality and morbidity rates in a popula-
tion in both stable and emergency situations.

Immunity in the population

Immunity plays a crucial role in the dynamics of disease transmission. The more
people are immune, the less likely it is that a pathogen will find a susceptible
person. If enough people are immune, the chance of the pathogen causing an
infection becomes so small that transmission stops, even though there are still
susceptible people. This is called herd immunity (41). With poliomyelitis, for
example, if 80 to 85 per cent of the entire population is immune, the virus will
disappear (50). A population can lose its herd immunity through births, migration
of susceptible people into the population, or waning immunity in the population
over time. Figure 3.1 presents a model of immunity in the population.

Two important points can be deduced from Figure 3.1:

„ The immunity in a population is the result of people either overcoming the

infection or being immunised (vaccinated).

„ The susceptibility of the population in an area increases through the influx of

non-immune people, birth, and from people losing their immunity (through
time or another reason, such as HIV infection).

28
 DISEASE IN THE POPULATION

Susceptible persons
Births
from outside region

Susceptible persons

Infection

Effective immunisation Persons with infection Disappearing immunity

and reduced resistance

Recovery

Death

Immune persons

Death

Figure 3.1. Model of immunity in a population (adapted from 37)

3.1 Endemic and epidemic occurrence of disease

The occurrence of an infection in a population is determined by many factors,
including the pathogen, its persistence and/or latency, whether it has a high or low
infectious dose, whether it can multiply outside the host, and whether it infects
both humans and animals or only humans. It depends on hosts, how many
pathogens they shed, and whether their behaviour favours transmission. It de-
pends on the environment in which transmission occurs, its climate, and its
human physical environment, which may favour direct transmission, vectors, or
intermediate hosts. It depends on potential new hosts and their behaviour, resist-
ance, and immunity against the infection. (These factors were covered in Chapter 2.)

29
 CONTROLLING AND PREVENTING DISEASE

These factors can either favour, or oppose, the transmission of the pathogen from
a host to a potential new host. Favouring and opposing factors balance each other.
Three situations are possible:

„ The opposing factors are stronger than the favouring factors: the infection
disappears or does not occur. This situation is what we try to achieve.
„ The opposing and favouring factors are in balance: there is a continuous
presence and transmission of the infection in the population. The disease is
endemic.
„ The opposing factors are weaker than those that favour transmission: the
occurrence of the infection increases in the population. If the occurrence is
clearly more than normally expected, then the infection is epidemic.

This balancing between the opposing and favouring factors is a dynamic process
that can easily alter with changes in the pathogen, hosts, environment, or potential
new hosts.

Communicable diseases are usually either absent, endemic, or epidemic in a

population (although sporadic or imported cases can occur). Most infections can
be both endemic and epidemic, but only some can cause explosive, severe
epidemics. Even though epidemics can be dramatic, endemic disease is often
worse for the population (51).

In health programmes it is the eradication of frequent, severe, and preventable or

controllable infections that should receive priority (71).

3.1.1 Endemic occurrence of disease

An infection is endemic when it is always present in a population (3). How often
the infection occurs depends on the factors mentioned in Section 3.1, but seasonal
fluctuations of infections are also common.

When an infection is common and results in long-lasting immunity, disease will

usually occur in childhood, as adults will have built up immunity. If the infection
is highly endemic, it is unlikely that an epidemic will occur, unless several
subtypes of pathogens can cause the same disease and the population is immune
against only one of these, which can happen with dengue fever, for example (3).

Disease is often unevenly distributed in a population. Depending on people’s

occupation, environment, and behaviour, some may be more exposed to patho-
gens than others. Children may be more at risk because of their behaviour (e.g.
schistosomiasis caused by playing in water) (16) ; the poor may be more at risk

30
 DISEASE IN THE POPULATION

because of the conditions in which they live (e.g. poor housing resulting in Chagas
disease (73)); people with certain occupations or living in specific locations may be
more exposed (e.g. farmers or sewage workers would come in contact more easily
with leptospirosis). It is important to identify the people who are most at risk, and
why to know who to target and what preventive measures to take.

3.1.2 Epidemic occurrence of disease

An epidemic, or outbreak, occurs if there are clearly more cases of an infection
than would be expected in a given area over a given period of time or season (71).

Outbreaks can occur if the following features are combined (10):

„ a pathogen must be introduced or be present in the area;

„ the environment must be favourable to transmission; and
„ there must be enough susceptible people in the population.

There is a large risk of an outbreak when:

„ infected people enter a non-immune population, in an environment favourable

to transmission (e.g. infected refugees or migrants enter a non-endemic area);
„ susceptible people move into an endemic area (e.g. non-immune refugees or
migrants enter an endemic area; people enter an area where a zoonosis occurs
in an animal population (79));
„ the population has lost its resistance or immunity, and the pathogen is reintro-
duced (e.g. people’s immunity has diminished over time; babies have been
born; or people are suffering from disease or malnutrition (73)); and
„ the environment has changed, and has become more favourable to transmis-
sion (e.g. construction of a dam has produced an environment favourable to
mosquitoes (a malaria vector) or snails (the intermediate host of schistosomia-
sis)).

An outbreak can become an emergency if the infection is severe, if the society is

disrupted because of the number of cases occurring, or if medical infrastructure is
unable to cope because of lack of personnel, material, or organisational skills (10).

The most severe epidemics are those caused by infections which are easily
transmitted, have short incubation periods (71), and have a potentially severe
outcome. The main iInfections that cause severe outbreaks are diarrhoeal diseases
(e.g. cholera, bacillary dysentery), yellow fever, malaria, epidemic louse-borne
typhus fever, and louse-borne relapsing fever, but other infections can cause
emergencies too.

31
 CONTROLLING AND PREVENTING DISEASE

Most of the infections covered in this manual can cause epidemics which impact
hard on society or individuals. They will not normally cause emergencies though,
as they develop slowly, are less serious, or people have high levels of immunity.
Where an infection is endemic it is impossible to give a threshold level that marks
the beginning of an epidemic, as this depends on what is ‘normal’ in a given
population, in that area, in that season. Where cholera is not endemic, one case of
locally acquired cholera will be declared an epidemic (10). Where cholera is
endemic, two new cases in a week would not necessarily cause concern. An
epidemic would be confirmed if more cases occur than occurred in the same
season in the recent past (55). Table 3.1 presents the epidemic threshold level for
several diseases.

As with endemic occurrence of disease, outbreaks may be limited to specific

groups of the population. Analysis of outbreaks are covered in the next section.

Table 3.1. Threshold levels of epidemics (10)

Infection In a non-endemic In an endemic population

population

Salmonellosis A group of cases with one common source of infection.

Cholera One locally infected confirmed A ‘significant’ increase over what

case (a) is normal in that season

Yellow fever One confirmed case in a non- A ‘significant’ increase in the

immune population with a number of cases over a ‘limited
presence of vectors time’

Mosquito-borne A group of cases in a non- A ‘significant’ increase in the

arboviral immune population (the first number of cases caused by that
encephalitis case is a warning) specific pathogen over a ‘limited
time’

Malaria A group of cases occurring in a Rare

specific area

Plague One confirmed case A cluster of cases caused by

domestic rodents or respiratory
transmission or an epidemic in
rodents

Epidemic louse- One confirmed case in a louse- A ‘significant’ increase in the

borne typhus fever infested, non-immune number of cases over a ‘limited
population time’

(a)
A ‘confirmed case’ is an infection confirmed by laboratory tests.

32
 DISEASE IN THE POPULATION

3.2 Introduction to epidemiology

Epidemiology is the study of the distribution, occurrence, and causes of a disease
in a population to improve the existing health situation (54). Epidemiology covers
endemic as well as epidemic occurrence of disease, and the approach taken to both
is similar.

A full epidemiological study will consist of four phases (71):

1. An assessment of the distribution and frequency of a disease in a population.

2. Determining and analysing the causes of the disease.
3. Conducting an intervention to try to reduce the occurrence of the infection.
4. Evaluating the effectiveness of the intervention.

Conducting a full and methodical epidemiological inquiry is complex, and should

remain the domain of an epidemiologist, but an intuitive form of epidemiology is
already used by WES specialists. Identifying poor personal hygiene, caused by
lack of water, as a cause of trachoma or diarrhoea, and installing a water supply to
improve water availability, is intuitive epidemiology. Although intuitive epidemi-
ology does not have the scientific rigour of classic epidemiology, it is more
practical for fieldworkers.

The following sections will help water and sanitation specialists to take this
intuitive approach to disease prevention.

3.2.1 Data collection

Epidemiology is about information. The mortality and morbidity rates of those
diseases which are an important health problem will have to be collected from
local medical staff or authorities, or from medical agencies working in the area.
The rates may not be very accurate, but for a WES specialist accuracy is not
essential. Collect the monthly or weekly incidence rates of diseases related to
WES, going back for some years if possible (i.e. the number of new cases
occurring in a population per unit of time (3)).

The incidence rates, combined with a questionnaire or survey on the background

and the environment of the cases, should answer the following questions:
„ What infection is being investigated?
„ Who is affected? What are their characteristics: socio-economic background
(e.g. income level), occupation (e.g. agricultural workers), age, sex, ethnic
group, specific behaviour (e.g. use of one specific source of water) or other
characteristic (e.g. HIV infected)?

33
 CONTROLLING AND PREVENTING DISEASE

„ Where does the disease occur (place of exposure)? What is the geographical
distribution (e.g. altitude) and the environment (e.g. slums, swamps, forests,
poor sanitation).
„ When does the disease occur? Is there a season (e.g. wet season, when many
vectors are present), a specific occasion (e.g. one week after a feast, or visit to
a town, or a strong increase of the disease in years after the construction of a
dam) (71)?

Being aware of the risk factors which can cause transmission will help to identify
relevant information. More detailed information about risk factors concerning
WES can be found in Chapter 5.

The local risk factors that cause transmission will have to be identified by
surveying the environment and human behaviour. It is also important to look at
local attitudes and beliefs regarding the disease and its prevention, as these could
affect potential interventions. The survey will also have to assess the risk the
infection poses, and the capacity of the local authorities to deal with the existing
situation or with a potential outbreak. Then the relative importance of the different
risk factors will have to be determined.

While endemic occurrence will usually exist in a relatively static situation, an

epidemic is always the result of some kind of change which favours transmission.
Either a pathogen is introduced into a susceptible population, the population has
become more susceptible, or the environment has changed in a way which favours
transmission. This change should be identified.

In an outbreak, the primary transmission, or the way the initial cases are infected,
may be different from the secondary transmission, or the way the pathogens are
transferred from the initial cases to new cases (8). An outbreak of typhoid fever
may originate with infected drinking-water, while secondary transmission may
occur through infected food handlers. Similarly, with endemic diseases not all
cases need to be infected in the same way.

Once the local risk factors are identified and their importance assessed, the
potential effects of eliminating or controlling these factors has to be estimated. By
combining this information with what is known about local limitations and
resources, it is possible to come up with an indication of what type of intervention
would be appropriate in a particular place. When an outbreak results in an
emergency, all feasible measures that could potentially reduce transmission should
be taken.

34
 DISEASE IN THE POPULATION

This analysis will usually be enough to choose an intervention for endemic and
epidemic diseases. Trying to analyse an outbreak can be more complex, as the
process is more dynamic. The following aids can help analyse an outbreak.

3.2.2 Aids in analysing an outbreak

The minimum requirement to follow and analyse an outbreak is up-to-date
information. A sufficient number of competent and motivated medical staff must
be present to identify cases, and reliable and regular reports on cases must be
collected at a central point.

It is usually qualified medical personnel who will analyse an outbreak, but the
WES specialist has to understand some of the basic aids that can be used, with a
questionnaire or survey, to assess the risks and extent of the outbreak, and the
possible sources of the epidemic.

3.2.2.1 The spot map: mapping the outbreak

A ‘spot map’ is a map on which the location of the cases is marked. The spot map
shows both the distribution and trend of the outbreak, and potential sources of
infection (55). It also indicates which villages or neighbourhoods are most at risk of
further transmission. If possible, the map should show where people became
infected to help locate the source of infection. If the infection is easily spread from
person to person, it may be useful to map where the cases live or work to predict
where there is the greatest risk of secondary transmission.

3.2.2.2 The epidemic incidence curve: following the outbreak in time

During an outbreak an ‘epidemic incidence curve’ should be drawn. It is a graph
that plots the number of new cases day by day, or week by week. The curve can
then be extrapolated to show when the initial infection occurred. Looking at the
whereabouts and activities of the initial cases will help to pinpoint the cause of the
infection.

The curve can highlight a trend and the nature of the outbreak (71).

The point-source or common-source outbreak

A point-source outbreak is caused by a particular incident that infects a group of
people almost simultaneously. It is typical of water-borne and food-borne out-
breaks, or outbreaks caused by handling infected material (55). This type of out-
break could be caused by contaminated food served at a feast, for example, or
travellers drinking from a contaminated stream. By plotting the incidence curve of
an outbreak of a known disease, the approximate time of primary infection can be
determined. The first cases that appear are the ones with the shortest incubation

35
 CONTROLLING AND PREVENTING DISEASE

period; the last ones are those with the maximum incubation period. Going back
in time for the length of the incubation period indicates when infection occurred.
By looking at where the people were and what they were doing at that time, the
source of infection can be identified (73).

Figure 3.2 shows a point-source outbreak of diarrhoea in a village. The first cases
of diarrhoea appear on the morning of 16 July. The diarrhoea is identified as
salmonellosis. As the incubation period of salmonellosis is between six hours and
three days (3), people were probably infected on the evening of 15 July. A survey
shows that on the evening of the 15th all the known cases attended a funeral. At
this funeral food was served, and the majority of those who ate meat have fallen
ill, while those who did not have no problems. In this case it is probable that the
meat served at the funeral was the source of infection.

Cases with a 'short' incubation

period of 6 hours to 1 day

Infection of all cases Cases with a 'long' incubation

period of 3 days

Number of
cases

12 13 14 15 16 17 18 19 20 21 22 23 24
(July)
Time

Infection on the evening of the 15th results

in persons developing the disease in the
morning of the 16th to the 18th of July.
(incubation period: 6 hours to 3 days)

Figure 3.2. The epidemic incidence curve of a

point-source outbreak of salmonellosis

36
 DISEASE IN THE POPULATION

The extended point-source outbreak

These outbreaks are caused by specific sources that have infected people over a
period of time. The onset is comparable to a point-source outbreak, but cases
continue to appear over a longer period (71). This type of outbreak could be caused
by sewage leaking into a water supply system, for example.

The process of finding the source of infection is similar to that with the point-
source outbreak. The probable time of initial infection is determined by going
back to the time the first cases appear and back further for the shortest incubation
period of that infection. A survey of where the first cases occurred, and what those
people were doing will normally indicate the probable cause of infection (73).

Figure 3.3 shows the epidemic incidence curve of an extended point-source

outbreak of cholera. As the shortest possible incubation period of cholera is one

Persons with a 'short'

incubation period of 1 day Persons with a 'long'
infected on the 9th May Transmission ends
incubation period of 5 days
infected on the 14th May
Transmission begins

Number of
cases

8 9 10 11 12 13 14 15 16 17 18 19 (May)

Time

Infection on the 9th results in average onset of disease around the 12th
(average incubation period is 2-3 days)

Figure 3.3. The epidemic incidence curve of an extended

point-source outbreak of cholera

37
 CONTROLLING AND PREVENTING DISEASE

day, the initial infection probably occurred on 9 May. A survey shows that all the
cases ate at a particular food stall on the local market. The stall was closed the
evening of the 14th. Cases continued to appear until the 19 May because some of
the people infected on the 14th will have had an incubation period of five days.

The propagated-source outbreak

This type of outbreak is the result of progressive transmission. One case, the
primary case, will infect a cluster of cases, the secondary cases, who will infect the
next cluster, the tertiary cases, and so on. Usually the onset and decline of the
outbreak will be more gradual than a point-source outbreak. This type of outbreak
is possible with most infections covered in this manual.

Every cluster of cases will show a peak in the incidence curve. The period
between the peaks is called the ‘serial interval’ (73). The serial interval will depend
on the latent period, the period of communicability of the host, and the time it
takes for the pathogen to develop in a vector or intermediate host. This will often
be about the average incubation period, plus, if applicable, the period of develop-
ment in the vector or intermediate host. The longer the latent period, the longer
the period of communicability, and the longer the time the pathogen needs to
develop in the vector or intermediate host, the more spread out over time the
curves will be.

The number of cases that will occur will depend on how effective transmission is.
The presence of risk factors such as overcrowding, behaviour which favours
transmission, a large susceptible population, or an environment favourable to
vectors or intermediate hosts, will increase the number of cases (55,71).

Figure 3.4 is an example of a propagated-source outbreak. This is a theoretical

example of an infection that has an incubation period of seven to 10 days and
period of communicability of two days.

3.2.2.3 Limitations of the spot map and the epidemic incidence curves
The spot map and epidemic incidence curves have several limitations:

„ The reported rates always lag at least one incubation period behind the actual
situation of the infection. The cases identified now were infected one incuba-
tion period earlier. People infected since then are developing the infection, but
do not show any symptoms yet (even if transmission were to stop abruptly, new
cases would continue to appear for the length of the incubation period). Delay
is also likely because of communication problems between the field and the
central registration point.

38
 DISEASE IN THE POPULATION

Tertiary cases
Serial interval
Secondary cases (period between peaks)

Primary case

Number of
cases

Transmission resulting
in secondary cases
Time
(in days)
Transmission resulting
in tertiary cases

Figure 3.4. The epidemic incidence curve of a propagated-source outbreak

„ Reliable and up-to-date information identifying the infection and recording

cases is vital. Problems in identifying or reporting cases make analysing an
outbreak difficult.
„ Normally only symptomatic cases will be registered, and asymptomatic infec-
tions will not be identified. This means that you may only be seeing the tip of
the iceberg.
„ The epidemic incidence curves only indicate when the initial infection prob-
ably occurred. The actual cause of the outbreak must be identified by people in
the field assessing the cases and their environment.
„ When cases infected by primary transmission spread the pathogen to others,
they may do so through a different route than the one that infected them. With
secondary transmission every new case becomes a potential new source of
infection for others, behaving as little (extended) point-source outbreaks them-

39
 CONTROLLING AND PREVENTING DISEASE

selves. The incidence curve may be the result of an accumulation of these many
little outbreaks. Cases can often transmit the infection over long periods of
time, which will ‘smear out’ the distinct peaks in a propagated-source out-
break, so the epidemic incidence curves found in practise will not rarely look
like the neat models shown here.

3.3 Mortality and morbidity rates in a population

The mortality and morbidity rates of infections are an indication of health prob-
lems in a population. Combined with an environmental assessment, the rates of
important infections will help to identify health risks relating to WES. The
importance of a disease will depend on:

„ its frequency in the population (i.e how common it is, or how big is the risk of
an epidemic); and
„ its severity ( i.e. whether the infection causes disability or death) (71);

Seasonal rates are important in identifying seasonal health risks and potential
epidemics.

3.3.1 Mortality rates in a stable population

Table 3.2 presents the Crude Mortality Rate (CMR) and the Infant Mortality Rate
(IMR) common to a stable population. The CMR is the total number of deaths in
the population due to disease, injury, and malnutrition. The IMR is the total
number of deaths in children under the age of one year per total births.

The figures for poor communities are not threshold levels, but give an idea of
what to expect. The rates for these communities are not acceptable at these levels,
and should be brought down, preferably to the CMR of developed communities.

Table 3.2. Mortality rates in a stable population

Situation Mortality rates

Crude Mortality Rate Developed communities 10 deaths/1,000/year (71)

(CMR) (0.3 deaths/10,000/day)

Poor communities 18-25 deaths/1,000/year (11,47)

(± 0.6 deaths/10,000/day)

Infant Mortality Rate Poor communities 60-150 inf.deaths/1,000births/year (71)

(IMR) (± 3 deaths/10,000/day)

40
 DISEASE IN THE POPULATION

It is not possible to give ‘acceptable’ incidence rates for specific diseases in a

population, as this will depend heavily on the local situation, but the figures
should be lower than the ones presented in Table 3.4. In practise the rates will have
to be compared with the feasibility of reducing morbidity by improving the
situation through an intervention.

3.3.2 Mortality and morbidity rates in emergency situations

In an emergency situation the CMR is the most practical indicator of the health
status of a population. As long as the CMR in a population is more than 1 death/
10,000/day the situation remains an emergency (47). Mortality rates in the initial
phases of an emergency can be much higher than this (11,47). Table 3.3 presents
figures of what would be acceptable upper threshold levels in the post-emergency
phase in camps for displaced people or refugees.

Table 3.3. Threshold levels of Crude Mortality Rate and Infant Mortality Rate in
camps

Mortality rates

Crude Mortality Rate 1 death/10,000/day

Infant Mortality Rate ± 2.5 deaths/10,000/day

Even though it is difficult to give concrete figures on incidence rates of diseases,

Table 3.4 gives an indication of acceptable rates in camps for displaced people or
refugees.

Table 3.4. Indicative acceptable incidence rates and specific mortality rates in camps
for displaced persons or refugees (72)

Incidence rate Mortality rate

(in cases/10,000/week) (in deaths/10,000/week)

Diarrhoea total 60
Acute watery diarrhoea 50 1
Bloody diarrhoea 20
Cholera Every suspected case must be reacted upon
Fever of unknown origin 100 0.5
Malaria 20
Skin infections 40 -
Eye infections 35 -

Children under five are more likely to develop disease, and incidence rates of
roughly 1½ times those presented here would be acceptable in this group (72).

41