Should DDT continue to be recommended for malaria vector control?

Extracts from "Should DDT continue to be recommended for malaria
vector control?" by C. F. Curtis published in Medical and Veterinary Entomology (1994) 8, 107--112

The banning of DDT in developed countries

In the early 1970s the use of DDT was banned in the U.S.A. and some
European countries except for certain emergency pest control
situations. The ban was based on the toxicity of DDT to fish, its
persistence in the environment and evidence that it could be passed up
food chains. Very high doses of DDT applied against the beetle
vectors of Dutch elm disease led to concentration of residues in
earthworms and hence killing of American robins which ate the worms
(data reviewed by Cooper, 1991). There is much literature on the
passage of DDT through food chains causing thinning of the egg shells
of certain raptorial birds and hence interference with their
reproduction (Cooke, 1973; Cooper, 1991). As pointed out by Jukes
(1983), some of this evidence is now realized not to have been
conclusive because of the earlier failure to distinguish analytically
between DDT (or its derivatives such as DDE) and polychlorinated
bi-phenyls (PCBs) which are also persistent organochlorines and which
were used as electrical insulators between the 1930s and
1970s. Another complication, pointed out by Cooper (1991), is that in
some cases the timing of the declines of bird populations, the onset
of egg shell thinning and of large-scale DDT usage do not coincide,
and some of the more sensational bird kills in the 1950s and 1960s are
attributable to cyclodiene insecticides such as dieldrin. Though the
evidence against DDT was not altogether scientific at the time of its
banning, the recovery of populations of the sparrow hawk and peregrine
falcon since the banning of DDT and cyclodienes in the U.S.A. and
Europe are circumstantial evidence that the use of organochlorines on
field crops was harmful to some wild life.

At the time of the hearings in the U.S.A. leading to the banning of
DDT, much emphasis was given to the detection of DDT in human breast
milk. Its detection has been reported in well over 100 papers (see
tabulation by Smith, 1991). DDT is concentrated in breast milk because
it is lipophilic and the secretion of stored DDT into milk is the main
route for excretion of DDT in lactating women.

Support for the continued use of DOT for malaria control

The World Health Organization and many malariologists argued strongly
that the ban should not be extended to its use against DDT-susceptible
malaria vectors. W.H.O. (1984) recommended DDT as the insecticide of
choice for such vectors. Data tabulated by Pant (1988) indicate that,
in 1979, 77% of the tonnage of insecticide used for malaria vector
control world wide was DDT. It has frequently been emphasized that it
is used against these vectors for residual spraying inside houses and
scarcely enters natural food chains (Galley, 1971), provided that
spraymen follow instructions and do not pour surplus DDT into water
courses. Another proviso is that significant proportions of the DDT
intended for house spraying are not illicitly diverted to agricultural
use - unfortunately such diversions cannot be disregarded: they may
lead to residues in crops and livestock making them unacceptable for
export.

In supporting the continued anti-malarial use of DDT where anophelines
remain susceptible to it, W.H.O. and others have emphasized that it is
remarkably free of reported acute adverse side-effects for
spraymen. This contrasts with many alternative insecticides which may:
(i) lower cholinesterase, which requires that the blood of spraymen is
carefully monitored and those with signs of trouble are rested from
spraying duties until they recover (W.H.O., 1973,1984); (ii) be
contaminated with dangerous by-products (e.g. iso-malathion; Baker et
al., 1978); (iii) cause lachrymation and skin paraesthesia
(e.g. Moretto, 1991; Chester et al., 1992; Bayer, 1993;
Kalyanasundaram etal., 1993).

A major consideration in the case for continuing DDT for anti-malaria
spraying has been its cheapness per unit weight and its durability,
thus limiting the number of spray rounds required per year to one or
two. Estimates by W.H.O. (1990) of the increased cost of insecticide
(exclud- ing operational costs) for protection of a given population
for a year if organophosphates or carbamates were substituted for DDT
are: malathion 2x, fenitrothion 7x and propoxur 23 X. More detailed
economic comparisons in Nepal by Phillips & Mills (1991) found the
overall cost factors relative to DDT for programmes using malathion
and bendiocarb wettable powders to be 2.9x and 3.7x. It is frequently
contended that unless DDT is used, extensive malaria vector control
will not be affordable in low-income countries in the absence of large
and sustained donations of a suitable alternative insecticide. Such
donations have been made in a few cases, for example malathion has
been donated to Sri Lanka to enable extensive use of this compound
since DDT resistance was shown to be seriously interfering with
effective use of DDT (Clarke et al., 1974).

Data of Abdalla & El-Zorgani (1988) showed that, in very hot tropical
climates, DDT was less persistent in soil (half-life of 5 weeks) than
in temperate climates.

Regarding the possible dangers of DDT to humans, studies of spraymen
in India and Brazil (W.H.O., 1973) and workers in DDT factories in the
U.S.A. (Laws et al., 1967; W.H.O., 1973) showed elevated levels of DDT
in serum or body fat, but medical follow-up showed that this did not
appear to have done them any harm.

DDT in breast milk was considered to have arisen from the background
contamination of the environment due to widespread outdoor use of DDT
in the past.

It has long been known that DDT is carcinogenic in mice, but until
recently it has been emphasized (e.g. by Hayes, 1971, and Jukes, 1983)
that there was no valid evidence for its carcinogenicity in humans.

...

Effectiveness and cost of new methods of malaria vector control

The above evidence of human health hazard from DDT for anti-malaria
spraying would perhaps not yet be sufficiently compelling to demand a
change in spraying policy, were it the case that DDT spraying is still
the only effective method of malaria control which is affordable by
most developing countries. However, recent field studies with
pyrethroids and carbamates and/or the use of impregnated bednets and
low volume spraying with motorized mist blowers are increasingly
indicating that there are now more effective methods than conventional
DDT spraying for malaria control, and furthermore that these methods
are affordable. The fact that the newer alternative insecticides are
naturally degradable in soil and mammalian tissue and do not leave
long-term residues is re-assuring but not conclusive proof about their
long-term safety. Information about their toxicity has been reviewed
by Ray (1991), Baron (1991) and Eriksson (1991); more data are
available in submissions to regulatory authorities by the
manufacturers of the newer insecticides.

An entomological study by Sharp et al. (1994) of house spraying with
conventional suspensions of wettable powder of DDT versus
lambda-cyhalothrin ('Icon') against Anopheles arabiensis showed that
about the same number of mosquitoes exited alive (fed or unfed) from
DDT-sprayed houses as from unsprayed houses, but far fewer did so from
lambda-cyhalothrin-sprayed houses. The poor performance of DDT was not
because of physiological resistance, but presumably reflects the
well-known excito-repellency of DDT (first reported by Kennedy, 1947),
especially to a rather exophilic species like An. arabiensis. This is
not to deny that DDT can still have a major insecticidal impact on
some anopheline populations (see e.g. Asinas et al., 1991).

Deltamethrin and cyfluthrin were found to be much superior to DDT, HCH
or malathion in vector control in trials in India (Ansari et al.,
1990; Schofield, 1993). However, these data are not entirely relevant
to the question under discussion because they were in areas where the
vector (An.culicifacies) was resistant to DDT, and it should be
recalled that the W.H.O. recommendation of the use of DDT only applies
to susceptible populations (W.H.O., 1984).

In a combined entomological and economic study in Mexico,
Arrendondo-Jimenez et al. (1993b) reported persistence of the
insecticidal effect over 9 weeks or more after low volume spraying of
bendiocarb, cyfluthrin or deltamethrin. The operational costs of these
treatments were comparable with conventional spraying of DDT. The
overall costs of low volume spraying benefit from the fact that a
sprayman can cover 3 times as many houses per day with the low volume,
as opposed to the conventional, method. This means that if the low
volume method were adopted spraymen could be deployed on other work
for much of the year and that there could be more rapid response to
malaria epidemics. However, in very low income countries the relative
contribution of spraymen's wages to total costs might be less, and the
more sophisticated maintenance required for motorized mist blowers
might create problems.

Several recent studies have shown greater mean reductions in malaria
with newer chemicals and/or application methods than with conventional
DDT spraying, though the differences were not necessarily
statistically significant in all cases. These studies include
comparisons with DDT by: (i) Arrendondo-Jimenez et dl. (1993a) of low
volume spraying of bendiocarb in Mexico, (ii) Asinas et al.
(unpubl. b) with lambda-cyhalothrin' wettable powder (WP) in the
Philippines, (iii) Asinas et al. (unpubl. a) with etofenprox WP in the
Philippines, (iv) Mnzava et al. (1993) with lambda-cyhalothrin WP in
Tanzania, (v) Yuyi Anti-Epidemic Station, Jiangsu, China (cited in
Curtis, 1991) with pyrethroid impregnated bednets, and (vi)
deltamethrin-impregnated nets in Hainan, China, against An.dirus
transmitted malaria which DDT failed to control (Li Zuzi in Curtis,
1991), and (vii) Kere (1992) with permethrin-impregnated bednets in
the Solomon Islands. In the latter study there were seven sectors
with DDT and seven with bednets, and the worst malaria control result
among the bednet sectors was better than the best among the DDT
sectors.

Numerous studies with pyrethroids on bednets or used for house
spraying have shown high mortality of nuisance insects such as
bedbugs, whereas with DDT these are poorly killed (or even stimulated
to cause extra biting nuisance). This suppression of nuisance,
together with the lack of a visible deposit on walls, have led to
better acceptance of pyrethroid applications by householders (and
hence higher coverage rates) than with DDT. This is probably a major
reason for the. better malaria control achieved with pyrethroids.

Regarding the cost of the alpha-cyano pyrethroids which can be used
for house spraying at the very low dosage of 25 mg a.i./m2, Schofield
(1992) and White (1994) state that lambda-cyhalothrin in the form of
'Icon' 10% WP can be cost competitive with DDT 75% WP for the
protection of a given population for a given time. This is based on a
required dosage of the DDT WP 10.6x greater than 'Icon' WP and a
selling price of the latter, ex-factory, about 19x more per unit
weight ($67,000 versus $3500 per tonne of WP if ordered though
W.H.O.). These authors and Asinas et al. (1991) put much emphasis on
the saving in transport costs because of the smaller amount of a
pyrethroid which is needed. In fact, intercontinental sea freight and
land haulage may add about $1480 to the cost per tonne (G. B. White,
pers. comm.), i.e. 42% to the cost of DDT, but only 2% to the cost of
a pyrethroid. This difference would somewhat redress the balance of
cost in favour of pyrethroids. However, a more important uncertainty
is the true long-term market price of the insecticides. W.H.O. (1990)
and Schofield (1992, 1993) quote prices and dosages which lead to cost
factors for a pyrethroid versus DDT ranging from less than Ix to more
than 3x. In some countries the availability of home-produced DDT as
against a foreign exchange requirement for alternatives is an
important consideration.

With regard to the cost of impregnated bednets, information from China
(data of Li Zuzi in Curtis, 1991; Curtis, 1992) show that one annual
treatment with deltamethrin of the bednets already owned by a given
community is 50--75% cheaper than one annual round of DDT spraying of
the same houses. Where nets have to be provided by the health
authorities and are not customarily bought by householders, cost
competitiveness of permethrin-impregnated bednets with DDT house
spraying was reported from the Solomon Islands by Kere & Kere
(1992). These authors comment that the relatively easy task of net
re-impregnation and the increasing willingness of village volunteers
to take part in re-impregnation, but not in house spraying, brings the
operational costs of the bednet method down.

Pyrethroid resistance

Pyrethroid resistance is becoming a serious problem in several
agricultural pests, such as the horn fly (Sparks et al., 1985) where
there is cross resistance from DDT to pyrethroids. However, it was not
found in the two malaria vector species in Sichuan, China, where up to
2 million nets have been impregnated with deltamethrin annually over a
5-year period (Curtis, 1992; Cheng et al., 1994). However, there has
been a report of a rise in level of pyrethroid tolerance during a
village-scale trial of impregnated material in Kenya (Vulule et al.,
1994). Pyrethroid resistance has been reported, associated with
organophosphate resistance, in An.albimanus in Central America (Beach
et al., 1989), and it has been selected in the laboratory in
An.Stephens! from India, where it had a degree of cross resistance to
DDT (Chakravorthy & Kalyanasundaram, 1992). It has also been selected,
independently of DDT resistance, in An.stephensi from Dubai (Ladonni,
1988). This example of resistance was found to have only limited
impact on the effectiveness of impregnated nets under realistic
conditions (Curtis et al., 1993). Studies are in progress on: (i)
whether bedbugs, whose control has been a welcome side-effect of use
of impregnated nets, are building up pyrethroid resistance; (ii)
bendiocarb sprayed curtains as a possible alternative impregnated
material in the event of the failure of pyrethroids; and (iii)
mixtures of unrelated compounds as a means of delaying the appearance
of pyrethroid resistance (Curtis et al., 1993), on the same principle
as multi-drug therapy against tuberculosis and leprosy.

Cautions about a possible switch-over from DDT to alternative methods of malaria vector control

(i) The evidence that DDT may be harmful to human health quoted above
is not conclusive and demands corroboration.

(ii) Better training of spraymen will be required to avoid the
short-term side-effects (paraesthesia, cholinesterase depression,
etc.) of almost all substitutes for DDT.

(iii) The price competitiveness of substitute insecticides may vary
from country to country and depends on pricing policies of
multi-national chemical companies which may take advantage of the
disappearance of competition from DDT to increase the selling price
of alternatives; conversely the increased demand for alternatives
might bring down their price because of economies of larger scale
manufacture.

(iv) The relative effectiveness of DDT and alternatives may vary
between different Anopheles species and levels of malaria endemicity;
more widespread trials with adequate sample sizes are urgently
required.

(v) Fewer studies have been published on possible long term adverse
effects on human health of alternatives than are in the extensive
literature on DDT; such possibilities need to be kept under urgent
review.

(vi) A widespread switch to pyrethroid usage might lead to evolution
of resistance in vectors, as well as nuisance insects; it is possible
that the resistance selected might confer cross resistance to DDT
which might preclude an effective switch back to DDT; development of
practical pyrethroid resistance management is highly desirable.

(vii) It is important that governments do not ban DDT precipitately
without having ensured that alternatives are available to ensure
continuity of vector control.

Conclusion

With due regard to the above seven cautions, but in view of the facts
that: (i) it can no longer be confidently stated that DDT
anti-malarial spraying is harmless to human health, and (ii)
affordable alternatives are becoming available, the author considers
that DDT should no longer be recommended as the insecticide of choice
for malaria vector control.