Title

Dengue Fever: Lessons from Epidemiology

Creator

Sushil Kabra

Yogesh Jain

Date

1998

extracted text

1

Background Paper for the MFC Annual Meet, 1 -3, January 1998

Dengue Fever: Lessons from Epidemiology
Sushil Kabra & Yogesh Jain
Dept, of Pediatrics, AllMS, New Delhi.
History of Dengue: Dengue fever (DF) is a disease caused by 4 serotypes of dengue virus. It
is a disease of antiquity. The first few scientific descriptions in literature include epidemic of knee
fever in 1779 in Cairo and its suburbs described by Aljabarti; an epidemic in Batavia (Djakart)
described by David Bylon in 1779, and an epidemic in Philadelphia in 1780 described by
Benjamin Rush. Till the 1950's the disease was reported infrequently from various tropical and
subtropical countries. DF has been reported with increasing frequency subsequently. At present
more than 2.5 billion people live in dengue endemic areas of the world. Each year an estimated
50-100 million cases of dengue fever occur annually throughout the world.

DF acquired public health importance due to the increasing incidence of dengue
hemorrhagic fever reported since 1950’s. The epidemics of DHF first occured in South Asian
region in 1950's, spread to South Pacific islands in the 1970's and reached the Caribbean basin in
the 1980‘s.
In 1953-54 the first major outbreak of DHF was reported from Manila. Subsequently it
became endemic in Philippines with increasing number of cases each year. From Philippines the
disease spread to other South East Asian countries.
In 1958 an outbreak of'DHF occured in Bangkok affecting 2706 patients with a case
fatality rate of 10.94%, majority of affected patients being children below 10 years. In the next few
years the disease spread to the suburbs of Bangkok. It involved adjacent provinces in the central
region of Thailand in 1961 and by 1964 major outbreak occurred in big cities in northern and north
eastern Thailand. Over next ten years DHF became endemic throughout Thailand. Now every
. year DHF cases are reported from all provinces of Thailand with periodic epidemics. With good
case management the case fatality has decreased from 11% in 50's to less than 1% in 1990's. The
experience from Thailand suggest that the disease first spread to adjacent areas and finally
involved all major cities to make the whole country/province endemic for DHF. Outbreaks of DHF
like illness were reported from Hanoi and Hochiminh city of Vietnam in 1958 and 1960
respectively. Dengue type 2 virus was isolated from these epidemics and there were serological
evidence to suggest presence of dengue 1 virus also. The disease has become endemic in
Vietnam with frequent epidemics from mid 70s.

In 1960 DHF was reported first time in Singapore. Subsequently it became endemic with
frequent outbreaks. In 1969 a nationwide control programme was launched to decrease dengue
by environmental control which gave encouraging results with sharp decline in DHF. However,
from mid 80's there is resurgence of DHF in Singapore. Outbreaks of DHF/DSS were reported
from Kampuchia in 1961 with isolation of dengue 1 and dengue 4. In Malaysia, DHF first occurred
in 1962. In 1971 the disease was made notifiable. Subsequent to this the number of cases
increased with epidemics every 4 years. Cases of DHF from Myanmar was reported in 1963 and
subsequently more cases were reported, every alternate year with epidemics every 4-5 years.
In 1963-64 DHF outbreak was reported from Calcutta. Dengue 2 virus was isolated from
affected patients. Subsequently DHF was reported from Vishakapatnam in 1964, Vellore
between 1960-68 Kanpur in 1968, Ajmer in 1969, Jalore 1985, Delhi 1988 and Prabhani 1988.
From 1990 onwards epidemics of DHF were reported from Jammu, Surat, Shajahanpur, Lucknow,
Delhi, Ludhiana, Panipat, Hissar and Jaipur. Between 1969 to 1985 epidemics of DF were
recorded from Delhi, Gwalior, Hardoi, Jaipur, Bangalore, Pune, Trichur and Amalner (Table I). All 4
serotypes were isolated from different parts of India. DHF epidemics/outbreaks have been
reported from all the states except Bihar, Orissa & Kerala. In the first two decades after the
Philippines outbreak, DHF was localized to few countries in South East Asia. Then it spread to
other regions. In the period 1974-1980 three epidemics occurred in Southern coastal area of
Peoples Republic of China.

)
<.!. i> L' '

2
In 1977 a dengue pandemic began in the Caribbean. Following outbreaks on many
islands including Puerto Rico, classical dengue was introduced into Southern Eastern Mexico in
1978. In the Americas the first major epidemic of DHF/DSS occured in 1981 in Cuba. In recent
years clinically compatible cases with DHF with or without laboratory confirmation has been
reported from many countries including Mexico, El-Salvador, Nicaragua, Jamaica, Dominican
Republic, Puerto Rico, St Luca, Aruba, Brazil, Surnam, Colombia, Haiti and the US.
The knowledge of dengue in Africa is incomplete. Dengue has been reported from
coastal areas and islands of East and South Africa and from most of West Africa. There is no record
of illness from central Africa. All four dengue serotypes have been involved, but to date, epidemic
DHF has not been reported in Africa or the middle East. However, sporadic cases of disease
clinically compatible with DHF have been reported from Mozambique, Djiboute and Saudi Arabia.
From review of DHF epidemics in various countries it seems that there is increasing
number of cases from various parts of the world in last 2 decades. The disease got more attention
due to the high case fatality rate when it occurs the first time. The disease first occurs as a small
outbreak, becomes endemic and then periodic epidemics. The period between two epidemics
decrease with time. In between the two epidemic in a geographic area there may be increased
number of cases every alternate year.

Importance of Epidemiological information
The review of dengue infection in India suggests the disease has became endemic.
Attempts to control the DF/DHF problem should ideally be based on a good understanding of its
determinants, the factors which affect disease spread. In a very simplistic model, if the vector
(aedes aegypti mosquito) and the virus (multiple strains of dengue virus) are present and can
proliferate and spread in a suitable environment with susceptible hosts (human beings), the
disease is likely to occur and occur repeatedly. This disease may occur, if conditions permit, in an
epidemic form. We have all the ingredients, susceptible hosts, the vector, the virus and the
environmental conditions. Disease control theoretically should then be possible given the
presence of political and technical commitment.

However, dengue fever epidemiology has many unanswered questions. While a review
of publications reveal a large number of redundant epidemiological investigations which should
be discarded, many important factors in transmission have been overlooked. These need to be
studied in a planned manner. We shall briefly dwell upon the present knowledge about dengue
epidemiology.

It is our belief, based on a review of available literature, that like malaria, factors important in
dengue transmission have to be studied at a local/regional level. There can be no universal
epidemiological pattern which explain transmission of this disease in various parts of the same
country or continents. And therefore, if we wish to study and then control our own epidemics, we
need to study our problem ourselves.

Epidemiological factors could be discussed under three headings: agent, host and
environmental factors.
(a) Agent: The disease is caused by a virus. It is not firmly established what determines the
occurrence of complications like DHF in areas where multiple strains of dengue virus are present.
Is it that some strains are more pathogenic (and therefore lead to complications) or is it that the
sequential infections with multiple strains that leads to this dreaded disease. However, one thing
is clear that DHF occurs in an area where dengue is already established for some years.

The virus is transmitted by the female aedes aegypti during her blood feeding activity.
Thus all factors which govern its survival, proliferation and feeding habits would assume

importance. Aedes is a domesticated mosquito, prefers clean water collections, rests indoors and

does not like to move too much (25-50 metres). When it gets infected after feeding on a patient
having virus in the blood, it remains infectious for its entire life span of around 10 days (range 8-42
days). Then it can bite vigorously and painlessly a large number of people in a small area. Since it is
a day biting mosquito, it will bite children when they are in school, day care centres, at home and
adults at their workplace or other areas of congregation. It can travel in rails, buses, ships and
aeroplanes and therefore spread over short and long distances.
The factors outlined above have been shown to play their role in some very interesting
epidemiological investigations of outbreaks or in experimental designs. The mosquito needs a
warm temperature to grow (and the virus too), but it is not clear which temperature, outdoor or •
indoor temperature. In the Ajmer epedemic of 1969, low grade transmission persisted even in
December inspite of prevailing low temperature. This total dependence of Aedes on man allows it
to persist in these ‘hot islands' (indoors) in cold weather. Wherever man collects fresh water (for
drinking, washing, cooking or for holding flowers/plants), Aedes can proliferate. Most human
dwellings in developing countries have high population density-especially in urban areas. This
suits aedes-which has a short flight range and can bite multiple hosts to spread the disease.
The spread of the Aedes mosquito has been well documented by all means of
communication. In fact, a unique way of transmission is by used automobile tyres - Aedes
albopictus finds it very useful for transmission. Movement of viremic patients is a mechanism of
spread. The Ajmer epidemic of 1969 referred to above, started at the time of annual 'Urs Mela' in
the central zone of the city and spread radially to peripheral zones along the busy routes of human
movement. Probably pilgrims got the virus and the presence of high vector density caused an
explosive outbreak. Pushkar, a nearby place which attracted an equally large number of pilgrims
during the same period remained unaffected due to absence of Aedes^

The mosquito and the vector has spread widely now- Dengue/Aedes have been seen in
Shahjahanpur, Vellore, Mangalore, Western Coast upto Pune, parts of Maharashtra, Surat, Jaipur
and many other un-reported areas. So we have the disease/vector/virus in all the 4 zones of the
country.
(b) Host factors: The most important host factor is clustering of human beings. Enormous rise
in population of major cities (Delhi is more than 1 crore) and urbanization of rural areas in many
developing countries including us, over last 20-40 years has undoubtedly contributed to frequent
recrudescence of greater magnitude. Urbanization has clearly been shown in many analyses to be
responsible as in Malaysia, Thailand. It is possible to get periods of intense transmission with
epidemics of DF/DHF even in small populations e.g. islands, but endemicity does not typically
occur. But if the population of an area is larger, the disease may linger between epidemics in an
endemic form. It is very likely that the critical community size lies somewhere between 1.5 to 10
lakhs. All cities with population above this have a high chance to remain endemic once dengue
reaches them.
Even within households, multiple infections is the rule. Secondary attack rate was 44% in
households in a philipino outbreak. In Ajmer again, it was observed that in central wards practically
all the family members of a house were affected.

How the epidemic will behave depends on the immunity of the people. It has been
estimated that the basic reproductive rate of the disease is close to 2 in early part of an outbreak.
This rate does not drop to below 1 (i.e, the epidemic starts dying down) until 50% of the
population becomes immune (due to infection). How frequently epidemics occur in areas of
dengue endemicity will depend on number and proportion of susceptible people and the level of
herd immunity. If the growth rate of a population is rapid and there is a large significant population
movement, the proportion of susceptible people will change with consequent effects on disease
incidence.

Environmental factors: The larvae need water to grow and therefore epidemics are more
likely to occur following or during rainy season. However, epidemics have been reported in hot

4

summers or during absent rain period if mosquitoes find water collections for drinking and other
domestic purposes e.g, in Northeast Thailand, Ajmer.
The temperature issue has already been alluded to. However, in areas where seasonal
changes in temperature are clear, dengue transmission usually declines with the approach of cold
temperature.

Surveillence: In view of dramatic emergence of DHF in last 2 decades an effective surveillance
system to monitor the disease in community is desirable. Dengue fever has some clinical features
similar to other viral infections, a laboratory based surveillence system is more useful. The lab
based surveillance system allows public health authorities to accurately monitor the activity of a
number of infectious disease agents that present clinically as viral syndrome including
dengue/dengue hemorrhagic fever.
In spite of the information mentioned above, we need to answer many questions:
(I)
(ii)
(iii)
(iv)
(v)

How frequently can epidemics of DF/DHF occur ?
Is there a correlation between vector density and dengue incidence ?
How does temperature affect the growth of mosquitoes ?
What determines DHF epidemics when multiple strains of dengue virus are present ?
What is the best short term way of managing increased vector density ?

The lab based surveillance programme has three components Sentinel clinics/physicians
1.
Fever alert
2.
3.
Sentinel hospitals
Sentinel Clinics/Physician: Some dispensaries, clinics, physicians can be identified from
the existing system that provides primary health care to community. After a short training they may
start keeping records of patients presenting with nonspecific viral syndrome and collect blood
samples for further studies. The blood is sent for tests to central/regional laboratory for tests.
Fever alert: It relies on community health and sanitation workers. On observing an increase in
fever cases they notify it to the designated authority responsible for monitoring. The outbreak
may be investigated by public health department.
Sentinel Hospitals: These are hospitals which admit sick patients of infectious diseases.
Patients admitted with various clinical symptoms suggestive of viral infections including DHF are
investigated. The category of patients which need investigation for DHF and other infection
include , patients with
any hemorrhagic manifestations
1)
an admission diagnosis of viral encephalitis, aseptic meningitis and meningococcal
2)
shock.
a fatal outcome following a viral prodrome.
3)

To the above list, patients presenting with other clinical manifestates thought to be
associated with DHF may be added. These can be identified by clinical studies of confirmed
dengue cases.

More Musings on unknown Infectious Diseases
I am in complete agreement with Sridhar, not only because I am been on understanding
Infections, or because I am a pediatrician presently working in an academic institution. Today I am
convinced that finding answers to clinicial questions is as important an trying to understand the
socio-economic determinants of disease.
Even though the information and knowledge available on infections is vast, there are so many
unanswered questions> In the last four decades, the growth of knowledge about infectious

5

diseases has not been at the same rate as that in other spheres of human ill health. This could be
the consequence ofthe West guiding the agenda for research for themselves and not necessarily
for the benefit of developing countries. The time has come for us to take the 'bull by the horns’ as
it were and set our own agenda and attempt tofind answers to our questions.
For instance, what proportion of fever cases are due to malaria ? due to viral fever ? to urinary tract
infections ? What is the correct definition of fever: axillary temperature of +1 degree Fahrenheit
equal to core temperature ? what is the natural history of a child with positive tuberculin test ? What
proportion of acute dysentry in children is due to amoebiasis ? What are the causes and
determinants of encephalitis - endemic or epidemic ?
The background papers for this meet, I notice, are many small attempts at understanding different
aspects of various infectious diseases. A still more heartening fact is that most of these are based
on individual experiences. It appears that we have been trying to grapple with many important
clinical questions at an individual level. If answering these questions is considered essential at a
collective level, then a forum for addressing these should be established. True, MFC has
traditionally and essentially been a group fro sharing thoughts and personal experiences. But just
the way the PHC cell and Women & Health cell have been established, we could think of doing
something proactive about infectious diseases as well. We cannot leave it all to the existing
academic institutions where the agenda is likely to be determined by other concern, or
professional bodies like the API, IAP ( which aften function more like clubs) to answer these
questions. Individual attempts are important and will continue but sharing skills to ask the right
questions, to design studies and to plan, conduct, and analyze the results in a collective manner
has become the need of the day.
Another potential activity I envisage is to do regional surveillance. My epidemiologist friends could
probably throw more light on this, but I am sure we could get member organizations to maintain
regular surveillance of disease patterns, vital events, state of environment ( water, sewage) vector
density, etc. Is it necessary to wait for te NICD of some other institution to provide us the much
needed information ? We have our own resources. For instance, Dr. BR Chatterjee's amazing
laboratory at the Leprosy Field Research Unit, Jhalda, is capable of carrying out microbiology
investigations of the highest order given a bit or financial inputs. I am also sure that certain
governmental laboratory setups would be keen to collaborate in such work.
Microbiology as a discipline has been ignored for too long. Even in the 'conscious' and well
meaning voluntary organizations it has taken a back seat. Cost considerations can only partially
explain this attitude. The more likely reasons are (i) Most infections are still self limiting (ii) Most
antimicrobials are by and large very safe (one can give penicillin in doses of 1000 units to 24
million units without causing any problems save an occasional anaphylaxis) and (iii) till recently
most microbials had been affordable.
We could discuss the scope of establishing such a group within MFC on Day 3 of the annual
meet.

YogeshJain.