Domestic Smoke Pollution and Acute Respiratory Infection in a Rural Community of the Hill Region of Nepal

Abstract

Acute respiratory infections (ARI) are the cause of death for at least five million children per year under five years of age. Most of these deaths occur in developing countries. Domestic smoke pollution is very common in many parts of the developing world, and appropriate technology, such as smokeless stoves, is available to reduce this type of pollution. The present study has been undertaken in a rural community of the hill region of Nepal to find out if there is any association between domestic smoke pollution and ARI in infants and children younger than two years of age. This preliminary study showed that episodes of moderate and severe ARI increased with increases in the level of exposure to domestic smoke pollution, thus suggesting domestic smoke pollution to be an important, preventable risk factor of ARI.

Full text

Environment International,
Vol. 15, pp. 337-340, 1989 0160-4170/8953.00 +.00
Printed in the U.S.A. Allrishts reserved. Copyright @1989 Pergamon Press plc
DOMESTIC SMOKE POLLUTION AND ACUTE
RESPIRATORY INFECTIONS IN A RURAL
COMMUNITY OF THE HILL REGION OF NEPAL
M.R. Pandey, R.P. Neupane, A.
Gautam, and I.B. Shrestha
Mrigendra Medical Trust, ARI Project, P.O. Box 2587, Kathmandu, Nepal
E1 87-385 (Received
11
November 1987; Accepted 5 May 1989)
Acute respiratory infections (ARI) arc the cause of death for at least five million children per
year under five years of age. Most of these deaths occur in developing countries. Domestic smoke
pollution is very common in many parts of the developing world, and appropriate technology,
such as smokeless stoves, is available to reduce this type of pollution. The present study has been
undertaken in a rural community of the hill region of Nepal to find out if there is any association
between domestic smoke pollution and ARI in infants and children younger than two years of
age. This preliminary study showed that episodes of moderate and severe ARI increased with
increases in the level of exposure to domestic smoke pollution, thus suggesting domestic smoke
pollution to be an important, preventable risk factor of ARI.
INTRODUCTION
Out of about 15 million children under five years
of age who die in the world each year, approximately
one-third die from acute respiratory infections (ARI).
Most of the deaths from acute respiratory infections
arc caused by pneumonia (Lancet 1985; WHO 1983).
The overwhelming majority of these deaths occur in
developing countries. Nepal, one of the least devel-
oped countries in the world, has a high infant and
childhood mortality. The infant mortality rate based
on a 1986 Nepal Fertility and Family Planning Sur-
vey is estimated at 108 per 1000 live births (FP/MCH
1987). ARI, in conjunction with acute diarrhoeal dis-
eases and/or malnutrition, is a major cause of mor-
bidity and mortality among children under five years
of ago. This is the case especially in the hill and
mountain regions of the country, where domestic
smoke pollution and other risk factors may exacer-
bate ARI (Pandey 1987).
It is generally assumed that air pollution is a
problem of industrially developed countries. But the
problem of indoor air pollution in rural locations in
developing countries, where combustion of bio-
mass fuels is the principal source of energy, is
considerable. This problem is common in the north-
¢rn Indian belt, Nepal, large parts of China, north-
orn Burma, northern Thailand, Papua Now Guinea,
the Irianjaya region of Indonesia, Guatemala, and
Central and South America. It is estimated that about
half of the world's households cook daily with bio-
mass fuels. Most of this cooking is done indoors
using unvented stoves. The highest exposures are
probably experienced by women, infants, and young
children. Measured levels of air pollution in these
houses in developing countries greatly exceed indoor
and outdoor concentrations found in developed coun-
tries (Smith 1987). An epidemiological study on chronic
bronchitis among adults has shown a significant as-
sociation between the prevalence of chronic bronchi-
tis and exposure to domestic smoke pollution (Pandey
1984).
There have been very few studies that quantify the
effect of environmental factors such as air pollution
as a risk factor for ARI in developing countries. ARI
is a particularly important problem in Nepal, perhaps
because of the climate, terrain, and living conditions
of the people. The present community-based prelim-
inary study has been undertaken to find out if there
is any relationship between domestic smoke pollu-
tion and ARI.
METHODS
The study area comprised two contiguous village
panchayats, Talku Dundechaur and Chhaimale, and
the adjacent villages of Dakshinkali and Phakhel,
which are situated on the southwest edge of
337

338 M.R. Pandey et al.
Kathmandu Valley (about 18 to 24 km from the city
centre at an altitude of 1200 to 2000 m above sea
level). The area is totally free from industrial and
general atmospheric pollution. The terrain and living
conditions of the area are typical of rural communi-
ties in the hill region of Nepal. The villagers are
predominantly farmers with a subsistence economy.
In the study area, as in most other parts of hilly
Nepal, traditional stoves that burn biomass fuels such
as firewood and agricultural residues are used for
cooking and heating in ill-ventilated houses. Domes-
tic smoke pollution is considerable because the houses
have no chimneys.
The eligible population for this study comprised
everyone under two years of age who was perma-
nently residing in the area. In the scattered rural
population, eligible infants and children were visited
every two weeks by trained community health work-
ers under tue supervision of health assistants.
ARI information was recorded on the basis of its
severity classified as Grade I, II, Ill, or IV. This
classification was based on studies originally devel-
oped in Papua New Guinea (Riley 198 I). According
to
the Goroka
classification,
ARI grades
were
de-
fined as follows:
Grade I: Cough with sore throat or nasal discharge
or fever
Grade II: Cough with chest pain or purulent spu-
tum or a history of shortness of breath
Grade III: Grade II plus intercostal indrawing (or
use of other accessory muscles) or increased respira-
tory rate or strider
Grade IV: Grade Ill plus heart failure or cyanosis
or shock.
The World Health Organization (WHO) system of
classification (mild, moderate, or severe) was devel-
oped subsequent to this study and, thus, was not used,
although it is comparable to the Goroka classifica-
tion.
In order to record different episodes of ARI the
following working definition was adopted: If a child
suffering from ARI of any grade was reported to be
completely free from signs and symptoms of ARI for
at least 48 hours and contracted ARI again, then it
was considered a new episode of ARI (Miller; Brid-
ges-Webb 1984).
Data from the observed infants and children under
two years of age were analyzed to determine whether
there was any relationship between domestic smoke
pollution and ARI. Exposure to domestic smoke was
assessed by asking the mothers about the average
amount of time per day spent near the fireplace by
infants and children under two years of age. In older
children, aged 2 to 4 years, estimates of exposure to
smoke could not be made because the children were
too mobile. For this reason, no attempt was made to
collect data on smoke exposure in this age group. In
contrast to the first substudy, which covered an ear-
lier six-month period (February to July 1984), the
new substudy covered three months (November 1984
to January 1985) spent among the same population.
It, however, employed separate investigators to de-
termine ARI and smoke exposure times, thereby en-
suring a greater degree of independence in observation
of the two variables. Random checks by senior med-
ical staff were also undertaken in order to raise the
level of quality control.
RESULTS
A summary of the general pattern of the relation-
ship between ARI episodes among infants and chil-
dren under two years of age and exposure to domestic
smoke pollution is shown in Tables 1 and 2. The
average time spent near the fireplace per day by a
child is used as a measure of the extent of exposure
to domestic smoke pollution.
Table 1 shows the results of the study covering a
six-month period (first substudy, February through
July 1984). Among infants, the episodes of ARI per
child increased with the increased level of smoke
exposure. Although this is true for all grades of ARI,
in the case of life-threatening ARI (Grades III and
IV), the increase is more pronounced. To confirm this
finding, observations were recorded for a second
time in the same population covering a three-month
period (second substudy, November 1984 through
January 1985). Among infants, the results indicated
that the relationship is strong between life-threaten-
ing ARI episodes and exposure to domestic smoke
pollution; for mild to moderate grades of ARI, the
relationship is not that marked (Table 2).
Among children between one and two years of age,
a similar association was observed in ARI grades II,
Ill, and IV, but not in the mild ARI Grade I, in the
first substudy. In this age group, an association was
found in all the ARI grades observed during the
second substudy.
Thus, both substudies show a consistent, direct
relationship between life-threatening ARI Grades III
and IV and domestic smoke exposure.
DISCUSSION
Most ARI research so far has focused on microbi-
ological causes, immunization, and case management.
There have been relatively few studies to assess the

Domestic smoke pollution in Nepal 339
Table 1. ARI episodes according to time spent per day near the fireplace (February to July 1984).
Avsrazs O-1 year 1-2 years
time
Number Number
par
day of
ARI episodes by ffrade of ARI aplsodea by @fade
($n hra,) infants
r
lI Ill/IV ©hil- I II III/IV
dren
o-o.9 33
~O
9 1 17 ~2 ~ I
(1.21) (0.27) (0.03) (2.~7) (o.2~) (0.06)
I-1.9 90 12o 33 6 6~ 118 15 8
(1.33) (o.37) (0.07) (1.8~) (o.23) (o.z3)
2-3.9 94 170 48 16 95 179 32
11
(1.81) (0.51) (0.17) (1.88) (0.3~) (0.12)
4+ 16 30 13 9 ~0 81 31 11
(1.88) (0.81) (0.56)
(2.03)
(0.78) (0.28)
TotaX 233 360 103 32 216 ~20 82 31
(1.55) (0.~ (O.14) (1.9~) (0.38) (o.14)
Note: Figures in parentheses are average ARI episodes per ohild during the six-
month period.
Table 2.
ARI
episodes according to time spent per day near the fireplace (November 1984 to January 1985).
O-1 year 1-2 years
Avarage
time
Number
Number
wet day of
ARI
eplsodes by ~rade of
ARI
episodes by ffrade_
(in hrs.) infants
I II III/IV
chil-
I lI Ill/IV
dren
0-0.9 I~2 118 2h o 70 62
3
0
(0.83) (o.17) - (o.89) (0.04)
-
1-1.9
61 63 10 1 75 90 8 1
(1.03) (0.16) (0.02) (1.20) (0.11) (0.01)
2-3.9 24 42 5 4 36 69 9 3
(1.75) (o.21) (o.!v) (1.92) (0.25) (0.08)
4+ 20 36 ~ lO 27 45 9 8
(1.8o) (o.2o) (o.5o) (1.67) (o.33~ (o,3o)
Total 247 259 43 15 208 266) 29 12
(1.o5) (o.17) (0.06) (1.27) (o.14) (0.o6)
P<0.Ol when the group exposed to more than two hours (exposure present) is compared
with the (nonexposed) group that is exposed to 0-1.9 hours in both the age strata.
Note: Figures in parentheses are average ARI episodes per child during three
month-period.
effect of environmental factors such as air pollution
as a risk factor. The present study has focused partic-
ular attention on domestic smoke pollution and its
relationship to ARI.
The possible mechanisms by which passive inha-
lation of smoke may contribute to respiratory infec-
tions in infants and children are: cilial paralysis; the
facilitation of attachment of respiratory bacteria to
mucosa; and the depression of immune responsive-
heSS.
The possible role of domestic cooking fuel in con-
tributing to ARI has also been investigated in Britain
and the USA, where it seems that exposure to nitro-
gen dioxide from gas cooking stoves may increase
risk (Melia et al. 1982; Speizer et al. 1980). Biomass
fuels produce high NO 2 levels as well as other toxic
pollutants, and it is probable that these pollutants
have a range of effects on a child's respiratory de-
fense mechanisms. The problem is believed to be
especially critical in countries such as Nepal, where
biomass fuels are burned in unrented fireplaces and
stoves resulting in very high levels of indoor air
pollution. Nitrogen dioxide levels in these homes are
much higher than in gas-cooking houses of the USA
and UK and are associated with a complicated mix-
ture of other toxic pollutants (Davidson et al. 1986).
The present study's findings have shown a consis-
tently positive association between domestic smoke
pollution and ARI, especially life-threatening mod-

340 M.R. Pandey et al.
crate and severe grades, suggesting that domestic
smoke pollution is an important risk factor in ARI.
There were no obvious hidden correlations that
might have produced these findings. More than 98%
of the population have a subsistence economy. So
socioeconomic condition is not an important factor.
Almost all the houses are homogenous, similar in
construction, and have thatched roofs. Traditional
stoves that burn biomass fuels such as firewood and
agricultural residues are used for cooking and heat-
ing in all the houses and there are no chimneys.
Children are not provided with warm clothing even
in the winter. Cooking is usually done by mothers,
and infants and children younger than two years of
age are often with the mother. The duration of expo-
sure to domestic smoke of infants and children de-
pends upon the cooking and other behavioural patterns
of individual mothers.
Prior to and during the study period there were no
epidemics of flu or infectious diseases such as mea-
sles and pertussis. In order to confirm and validate
the findings of the first substudy, the observations
were repeated among the same population in a second
substudy. This second substudy, however, employed
separate observers to determine ARI episodes and
smoke exposure times, thereby ensuring a greater
degree of independence in observation of the two
variables. If all these facts are taken into consider-
ation, it is unlikely that the findings are compromised
by any other factors.
A study on acute respiratory infections among chil-
dren under five years of age in Kenya has shown no
correlation between ARI incidence and pollution lev-
els or housing characteristics. But no definite conclu-
sion could be made from this study, since the pollution
levels were very homogeneously distributed among
the houses (Boleij et al. 1987).
Ideally, smoke exposure assessment for important
pollutants in smoke should be done by conducting
extensive personal air pollution monitoring involv-
ing several infants and children in each of the homes
over several seasons. Only in this way could overall
exposure be quantitatively and completely defined.
Such an active monitoring programme, however, was
beyond the resources of the present study.
The present preliminary study has shown that ex-
posure to domestic smoke pollution is an important
risk factor in ARI, which is a leading cause of infant
and child mortality in developing countries. The ideal
way of dealing with the problem of life-threatening
moderate and severe ARI is the control of risk fac-
tors. Appropriate technology is available for this pur-
pose. Well-designed smokeless stoves not only reduce
pollution, they are also fuel-efficient. This helps in
the conservation of the forest, which is very impor-
tant in preventing environmental degradation in most
developing countries.
This preliminary study indicates that a longitudi-
nal intervention study lasting at least two years that
includes all seasons before and after intervention
would most likely lead to results that are scientifi-
cally sound. While arranging to conduct such re-
search, the present information seems to justify the
implementation of ameliorative measures to reduce
exposures by utilizing available appropriate technol-
ogies such as smokeless stoves as part of rural en-
ergy, housing, and health programs.
Acknowledgment ~ We would like to express our thanks to
UNICEF for its financial support and to the Mrigendra Medical
Trust for providing other necessary support in carrying out this
study. We would also like to express our appreciation and thanks
to all the hard-working members of the field team and the village
community for their cooperation in conducting this study. We also
want to thank Mr. Hem L. Shrestha for advice and secretarial help
in preparing this manuscript. Our grateful thanks are due also to
Dr. Kirk R. Smith for his invaluable suggestions.
REFERENCES
Boleij, J.S.M. et al. Indoor air quality in the Maregua area, Kenya.
In: Indoor Air '87, Proceedings of the 4th international confer-
ence on indoor air quality and climate, Vol 2, Institute
for
Water, Soil and Air Hygiene, Berlin, 50-54; 1987.
Davidson, C.I.; Lin, S.F.; Osborn, J.F.; Pandey, M.R.; Rasmussen,
R.A.; Khalil M.A.K. Indoor and outdoor air pollution in the
Himalaya. Environ. Sci. Technol. 20:561-567; 1986.
Miller, D.; Bridges-Webb, C. Nomenclature and statistical issues
in regional and international comparisons. In: Douglas, R.M.;
Kerby-Eaton, E., eds. Proceedings of an international work-
shop on acute respiratory infections in childhood, p. 164; 1984.
Editorial note, Acute respiratory infections in under fives, 15
million deaths a year. Lancet 699-700; 1985.
Melia, I.B. et al. Childhood respiratory illness and the home
environment II. Association between respiratory illness and
nitrogen dioxide, temperature and relative humidity. Int. J.
Epidemiol. 1:164-9; 1982.
Nepal FP/MCH Project Nepal fertility and family planning survey
report 1986, HMG, Ministry of Health, Katmandu; 79-80; 1987.
Pandey, M.R. A social response to respiratory infections: Nepal's
initiative, future development perspective on children. UNI-
CEF, New Delhi, 20:18-21; 1987.
Pandey, M.R. Domestic smoke pollution and chronic bronchitis in
a rural community of the hill region of Nepal. Thorax 39:337-
339; 1984.
Riley, R.D. Technical bases for the standardization of clinical
management of ARI in Papua New Guinea, Geneva, WHO
(WHO/RSD/81.7); 1981.
Smith, K.R. Biofuels air pollution and health: a global review.
New York: Plenum; 1987.
Spiezer, F.E.; Ferris, B.G.; Bishop, Y.M.M.; Spongier, J.D. Respi-
ratory disease rates and pulmonary function in children associ-
ated with No2 exposure. Am. Roy. Respir. Dis. 121:3-10; 1980.
World Health Organization (WHO) Global medium-term pro-
gramme Acute reapimtoryinfections document TRI/ARI/MTP/83.1,
Geneva; 1983.