Malaria control in Vietnam: the Binh Thuan experience

Malaria control in Vietnam: the Binh Thuan experience

Nguyen Van Nam1, Peter J. de Vries2, Le Van Toi1 and Nico Nagelkerke3

1 Binh Thuan Provincial Malaria Station, Phan Thiet, Binh Thuan Province, Vietnam2 Division of Infectious Diseases, Tropical Medicine and AIDS, Academic Medical Center, Amsterdam, The Netherlands3 Department of Medical Statistics, Leiden University Medical Center, Leiden, The Netherlands

Summary objective The National Malaria Control Program (NMCP) in Vietnam is based on application of

insecticide-treated bed nets (ITNs), spraying of insecticides and early microscopic diagnosis of malaria

and treatment (EDTM) with artemisinin drugs. This study explores the implementation of the NMCP at

provincial level and its impact on malaria incidence (mi) and prevalence in Binh Thuan in southern

Vietnam.

methods Data on implementation of EDTM, distribution of ITNs, annual mi and Plasmodium index

(pi) were derived from intervention logbooks and surveillance records kept by the provincial Malaria

Station since 1988. The relation between interventions and the change of pi over time was analysed with

Generalized Estimating Equations.

results Control activities focused on the highly endemic zones where ITNs were distributed free of

charge to ethnic minority groups, including twice yearly re-impregnation, from 1992 onwards. This

almost completely replaced insecticide spraying. Complete ITN coverage of these groups was achieved in

1995, constituting 40% of the entire population. In all malaria endemic communes, primary health care

posts were consecutively upgraded or installed, mainly between 1992 and 1995, offering EDTM with

artemisinin drugs free of charge. Before 1994, mi peaked to over 50/1000, pi to over 16% in the highly

endemic zones. In 1998, these had decreased to below 9/1000 and 4% respectively. The effects of the

interventions could not be discerned with statistical significance.

conclusion Malaria incidence and prevalence declined significantly in Vietnam, possibly due to the

malaria control efforts, but coinciding with rapid socioeconomic changes.

keywords malaria, Vietnam, early diagnosis, treatment, artemisinins, insecticide-treated bed nets,

health policy

Introduction

Malaria control currently focuses on two interventions, the

use of insecticide-treated bed nets (ITNs), and early

diagnosis and treatment of malaria (EDTM). These are the

pillars of the global malaria control campaign (WHO

2000). ITN programmes effectively prevent infection in

low and high malaria transmission regions, can even

reduce transmission if coverage is high, but can be

expensive, potentially compromising their sustainabili-

ty.(Verle et al. 1999). EDTM denotes a wide range of

passive and active case-finding activities and therapeutic

strategies that all aim at halting progress of disease to

severe stages (WHO 2000).

The recent Vietnamese malaria control experience, much

quoted but hardly studied in a quantitative fashion,

suggests that an integrated approach based on community

participation can be successful (Ettling 2002). Vietnam is

endemic for malaria, especially in its mountainous-forested

areas. Between reunification in 1975 and 1985 the number

of malaria cases declined, followed by a resurgence with

thousands of fatalities annually (Thi Phan et al. 1999). The

health care system was weak and supply of antimalarial

drugs irregular. Chloroquine and sulphadoxine/pyrimeth-

amine (SP) were the mainstay of treatment yet only

available in limited amounts. They were increasingly

replaced by quinine because of widespread multi drug

resistance. Severe malaria was common but parenteral

treatment with quinine was not available at the primary

health care level.

The new National Malaria Control Program (NMCP)

was adopted in 1991, based on ITNs, spraying of residual

insecticides and EDTM. National malaria morbidity and

mortality figures showed a steady decline since 1994

(Nguyen 1999; Warrell 2001). A similar pattern can be

observed at provincial level (Ettling 2002).

The implementation of the NMCP at provincial level is

delegated to provincial Malaria Stations (MS) which are

Tropical Medicine and International Health

volume 10 no 4 pp 357–365 april 2005

ª 2005 Blackwell Publishing Ltd 357

also responsible for surveillance. In Binh Thuan, a province

in the mountainous parts of southern Vietnam, MS started

recording malaria incidence (mi) and prevalence before the

launching of the NMCP, as an adjunct to the standard

health information system. In this study, we explore this

valuable data source with the aim to document and analyse

the relation between malaria control efforts and the

changes in mi and prevalence in Binh Thuan.

Data sources and methods

Study site

Binh Thuan (Figure 1) is traversed by a forested mountain

ridge in north–south direction, parallel to the coast. The

average annual rainfall ranges from 1000 to 1400 mm

mostly between May and November. Binh Thuan com-

prises eight administrative districts, including the capital

Phan Thiet, and 106 communes. The total population rose

from 767 000 in 1989 to 1 041 000 in 2000 (population

density 120/km2), of which three-quarters live in rural

areas (data from Provincial Statistics Department). Half of

the provincial domestic product is generated by agricul-

ture, forestry and fishery, 20% by industry and construc-

tion (increasing) and 30% by services (increasing). The

average income rose from US$217 in 1995 to US$278 in

1998 in urban areas and from US$137 to US$230 in rural

areas.

Malaria epidemiology

The Vietnamese epidemiological classification of malaria is

based on a compilation of criteria (Vu Thi Phan et al.

1999) The epidemiological classification of communes in

Binh Thuan is shown in Table 1. Improvement, sustained

for 5 consecutive years, leads to reclassification. Figure 1

shows the geographical position of communes in Binh

Thuan and their epidemiological classification until 1994.

Malaria transmission is mainly confined to the moun-

tainous forested regions inhabited by poor tribes, of the

highly endemic zones 3 and 4, but occasionally extends to

other villages in these zones. Especially remote ethnic

1 2

3 4

Low endemic communes:

Highly endemic communes:

•Hanoi

•

Ho Chi MinhCity

Phan Thiet

Figure 1 Map of Vietnam and Binh Thuan Province and their epidemiological classification until 1994.

Tropical Medicine and International Health volume 10 no 4 pp 357–365 april 2005

N. Van Nam et al. Malaria control in Vietnam

358 ª 2005 Blackwell Publishing Ltd

minority communes often lacked any health care facility

until the early 1990s. Inhabitants had to seek curative

care at other communes’ health facilities. Outside zones 3

and 4 malaria is mainly imported by people exposed to

mosquitoes in the forest. In 1999, a program was

launched to further the development of forested regions,

leading to increased migration and a resurgence of

malaria.

Malaria control interventions

The implementation of the NMCP at provincial level is

delegated to the MS. The MS is part of the provincial

health services but also accountable to the National

Institute of Malariology, Parasitology and Entomology

(NIMPE). Malaria control activities focused on zones 3

and 4 and were organized through the district hospitals and

health posts, which are run by the respective People’s

Committees. In principle, all communes have their own

health post with responsibilities for diagnosis and treat-

ment of malaria, prevention and surveillance. Until

recently health posts were staffed by professionals of

different educational backgrounds such as second degree

doctors, nurses, midwives and technicians.

The new 1991 national malaria policy aimed at revital-

izing existing or installing new health posts in all com-

munes, and providing subsidized essential drugs (‘social

drugs’) such as artemisinin drugs for malaria. To finance

these measures, cost sharing was introduced. However,

poor people were exempted from this and continue to

receive free health care from the public sector. This

effectively covers all ethnic minorities and most other

inhabitants of the highly endemic malaria regions (zones 3

and 4).

Early diagnosis and treatment of malaria

The implementation of the NMCP in Binh Thuan was a

stepwise process of consecutively upgrading and staffing

health posts. Supplementary to the NMCP, Binh Thuan

MS provided all health posts in zones 3 and 4 with

equipment, staff and training for microscopic malaria

diagnosis, including refresher courses twice yearly. This

coincided with the introduction of artemisinin drugs. Thus,

in the following our interpretation of EDTM includes early

EDTM with artemisinin drugs, all free of charge, 24 h/day,

within walking distance for the entire population.

In many instances, the first participation of a health post

in randomized-treatment studies was grasped as an

opportunity to launch EDTM, thereby combining efforts,

sharing resources and ensuring quality control and

adequate documentation of treatment (Giao et al. 2001,

2002, 2003, 2004; Le et al. 1997; Hung et al. 2004).

Prior to the introduction of EDTM, health posts did not

prescribe artemisinin drugs; quinine was the mainstay of

treatment for falciparum malaria and chloroquine for

vivax malaria. Between 1992 and 1994 artemisinin deriv-

atives were introduced in all districts. Artemisinin and later

artesunate were mainly used in combination with meflo-

quine as a single dose treatment (500 mg plus 500 mg

mefloquine and later 200 mg artesunate plus 750 mg

mefloquine), but also 5-day courses of monotherapy

artesunate or artemisinin were used. In 1999–2000, a fixed

combination of dihydroartemisinin, piperaquine, trimeth-

oprim and primaquine (CV8�) became the first-line

treatment (Giao et al. 2004).

A national policy to upgrade all health posts and to have

at least one university-trained medical doctor among the

staff of all health posts started later, when EDTM had

Table 1 The classification of malaria epidemiology in Binh Thuan Province

Epidemiogicalzone

Plasmodiumindex (%) Main vector/BR

Descriptives

Public healthcare system Economic situation

I 0 Number of anophelines Relatively well High (urban area)II <0.5, mainly

P. vivaxAnopheles minimus,BR < 0.4

Good Stable (rural area)

III >0.5, P. falciparum >P. vivax

Mainly A. minimus,BR > 0.4 some A. dirus

Average or weak Average or low(forests and mountains,

some ethnic minorities)

IV >1, mainly

P. falciparumA. minimus and A. dirus,BR > 0.4, DDT resistance

Weak/absent Low (like III with slash

and burn agriculture,mainly ethnic minority

population)

Plasmodium index, the proportion of positive blood smears in malaria surveys; biting rate, bites of anophelines per hour per person;

DDT, dichlorodiphenyltrichloroethane; BR, biting rate.




already been installed at all health posts. Between 1995

and 2000 the number of fully qualified doctors doubled.

Malaria treatment in the private sector

In Binh Thuan, treatment for malaria was an exclusive

activity of the public sector until 1999. Most formulations

of artemisinin drugs and mefloquine are listed as ‘social

drugs’ and thus not allowed in the private sector. In

addition, since 1990, following national principles, the

Binh Thuan MS has been persuading the private sector to

refer suspected malaria cases to the public health posts,

which were made more attractive by offering their services

free. Thus, the contribution of the private sector to malaria

treatment was limited.

In zones 1 and 2, where people tended to be more

affluent, the (more developed) private sector retained a role

in malaria treatment, albeit small. Since 1999, the national

policy again allowed the private sector to sell antimalarial

agents except social drugs. In that same year CV8� was

introduced but it was only available as a social drug to be

used in the public health sector.

Vector control

Residual spraying and ITNs were introduced into zone 3

and 4 communes, starting in 1992. Before that, bed nets

were not used in ethnic minority communes and rarely in

the ethnic Vietnamese communes. Spraying and distribu-

tion of ITNs were discontinued as soon as a commune was

reclassified to epidemiological zone 2 but surveillance

continued for 5 years. Spraying and re-impregnation of

ITNs was done twice a year, just before and after the rainy

seasons, concurrent with health education and surveillance

campaigns. In 1992 and 1993, dichlorodiphenyltrichloro-

ethane (DDT) was used for spraying. Because of the rapid

decline of the mi after the introduction of ITNs, spraying

was largely abandoned after 1995.

To combat the resurgence of malaria in 1999, spraying

of k-cyhalothrin (ICON�) and occasionally deltamethrine

was resumed. The main reason for the resurgence was that

settlers migrated into the forested zones and often had to

stay in semipermanent huts or sheds. ITNs were not always

suitable in these conditions or available in sufficient

quantities for the expanded households. This was covered

by spraying residual insecticides. From 1997 insecticide

sensitivity tests were performed irregularly, mainly to guide

the purchase of insecticides (WHO 1992) The main

vectors Anopheles dirus and A. minimus were sufficiently

sensitive throughout. However, the residual effects were

short-lived especially on brick walls, prompting the

replacement of k-cyalothrin by a-cypermethrin in 2002.

The ITNs were rectangular (height 150 cm, length 180–

190 cm, width 100–140 cm for two persons or 70–80 cm

for single persons), of knitted Denier 70–100 nylon multi-

filament and mesh 156 (12 · 13 holes/in.2 or hole size 1.2–

1.5 mm), mostly blue or green (white was avoided because

of its association with mourning). ITNs were produced in

Vietnam and purchased by the NMCP at approximately

US$3. They were distributed by the MS free of charge to all

inhabitants of the ethnic minority villages in zones 3 and 4.

The ITNs were re-impregnated by soaking with per-

methrin-emulsified concentrate. In 1999, a permethrin

formulation appeared to be substandard and as then

deltamethrin suspension concentrate became the main

insecticide. The distribution of free ITNs to the ethnic

minority villages in zones 3 and 4 and the number of ITNs

per household were gradually increased, with available

resources, from one ITN for every four to one for every

two household members. The proportion of the population

protected by ITNs was calculated as the number of people

living in a house with at least one ITN per four household

members. ITN distribution campaigns always aimed at

providing ITNs to all households of an ethnic minority

hamlet/village.

More affluent villages inhabited by ethnic Vietnamese

did not receive ITNs, even if they were located in zones 3

or 4. Inhabitants of these villages were encouraged to

purchase ITNs themselves, following the health promotion

campaigns, but we have no data on the coverage rate of

self-purchased ITNs and do not know how often people re-

impregnated their bed nets. During the study period,

dengue vector control was in its infancy and unlikely

affected malaria epidemiology.

Health education and community participation

Education sessions on recognition and prevention of mal-

aria, guided by periodic knowledge, attitude and practice

(KAP)-surveys, were organized for the population in zones 3

and 4, starting with the Cham minority in 1992, with

support fromCARE International.Other groups followed in

1993 and 1994. These sessions were integrated with the

twice yearly ITN re-impregnation campaigns and surveys.

Several health programmes in Vietnam, including the

malaria and dengue control programmes, the family plan-

ning programme, the malnutrition programme and the

HIV/AIDS prevention programme, make use of co-workers.

They are lay members of a commune, who receive short

extra training in certain health problems. Some co-workers

participate in more than one programme. To date there are

380 health co-workers working in 77 villages.

In some highly endemic communes, co-workers partici-

pated in the malaria control programme. Initially, they




received a 1-week training course in order to assist in

health communication and supervise the use and coverage

of ITNs. Later their training and responsibilities were

expanded to include early detection of febrile patients,

taking blood smears and referring patients to the health

posts, and participation in the surveillance and ITN

distribution and re-impregnation campaigns.

Malaria data sources

Between 1988 and 1994 the recording of mi, separately

from the general health management information system,

was introduced at the public health posts. The Vietnamese

public health system has been used to recording data and

even at a time when facilities for diagnosis and treatment of

malaria were not optimal, malaria cases were recorded.

The diagnosis was often based on clinical grounds or by

sending a blood slide for confirmation to the MS or district

hospital. Where microscopy was available, the causative

parasite was also recorded. In communes without health

posts, notification was non-existent and in the low endemic

zones it was less complete.

These data, collected monthly by the MS, were the basis

for this study. Annual mi was calculated as the total of

annually reported cases divided by the total population of

the communes which participated in notification. At the

higher levels of the health care system, recording of malaria

cases has long been in place. The data on hospital

admissions for malaria, severe malaria, and malaria-

attributable mortality were extracted from the district and

provincial hospital records.

Malaria prevalence

Surveillance of malaria prevalence is carried out by the MS

and by the commune health posts. Classic malariometric

surveys are held by the MS in the highly endemic zones (3

and 4), at the end of the dry and of the wet season (Hung

et al. 2002). To measure local transmission in a village,

they commonly sample those present there, including the

majority of children and women. Because of the rather

uniform methodology of these surveys, their results were

used for further study. Health posts in zones 3 and 4 also

took fixed annual quotas of blood smears. Because of the

variable nature of the sampled population these data were

not used for analysis.

Analysis and statistics

The association between the malaria control measures and

the malaria data was studied, based on the previously

explained assumption that the interventions, EDTM and

ITNs/spraying, can be interpreted as instantaneous changes

introduced in all communes in a stepwise fashion. We

hypothesized that the malaria control interventions had a

significant effect on the decline of prevalence per commune.

The interventions were entered as nominal value dividing

before and after the start of the intervention. The first

introduction of the ITNs in a commune,with a density of one

ITN per four members of every household, was taken as the

start of this intervention. The later increase of the number of

ITNs per household was not incorporated in the analysis.

The effect of the interventions on malaria prevalence

data was analysed on a time scale, taking dependencies

between observations from the same health posts into

account using Generalized Estimating Equations (GEE;

sas version 8.2). Year and epidemiological classification

were incorporated in the model as potential confounders.

Other changes over time, such as socioeconomic improve-

ments, usually do not follow a stepwise pattern and are

thus less likely to confound the outcome. The surveys at

the end of the dry season and the surveys at the end of the

wet season were analysed separately, taking the blood

smear results as dependent variable and the number of

slides per survey as a weight variable.

In some ethnic minority communes, the erection of a

new health post before 1995 coincided with the introduc-

tion of EDTM. This confounded the reported incidence. In

addition, the introduction of EDTM probably caused some

redirection of help seeking from the private sector to public

health services. Therefore, incidence was not further

analysed.

Results

Interventions

By the end of 1995, all health posts of the malaria endemic

zones 3 and 4 offered EDTM free of charge (Figure 2). In

the other zones diagnosis and treatment were less regulated

by the health authorities.

The population protected by ITNs/spraying programme,

calculated as the number of subjects living in a household

protected by the ITNs programme divided by the total

population of the communes, is shown in Figure 2. After

the introduction of ITNs additional spraying was limited,

focusing on new comers in the high transmission. This

maximally concerned 12 villages with a total population of

approximately 14 000.

Malaria incidence

The number of notified cases of malaria is shown in

Figure 3. Because of incomplete data from communes




without health posts before 1994, and the redirection of

health-seeking behaviour after the introduction of EDTM,

there is a spurious increase in incidence. The reverse is seen

in the district hospital data, which also served as primary

health facilities in the absence of commune health posts

before 1994. After 1994, the hospital data became more

stable and show that severe disease and mortality declined

rapidly after 1994, faster than the total of admitted cases.

Malaria prevalence

Malaria prevalence, shown in Figure 4, started to decline

significantly after 1994. Notably, epidemiological zone 2

data only reflect communes or hamlets, being surveyed

after a recent epidemiological reclassification. The provin-

cial average of the non-endemic zones 1 and 2 is much

lower.

Association between interventions and malaria data

Analysis by GEE of the prevalence data supports what can

be gleaned from Figure 4, viz a significant decline in

malaria prevalence after 1994 in the surveys in the wet

as well as in the dry season. A significant independent effect

of either intervention (ITNs or EDTM) could not be

confirmed.

Discussion

We explored the success of Vietnam’s malaria control

interventions at provincial level during the last decade.

Although the interventions jointly may have contributed to

the improvements, they could not be attributed to any

individual element of the control activities. Health service

data are difficult to interpret. There are many factors

influencing the reported incidence, which cannot be

quantified and which may confound statistical analysis. In

this study, the incidence data collected before 1994 are also

subject to these drawbacks. However, from 1994 onwards,

the data are reliable. The Vietnamese public health services

have a tradition of recording data and the number of

malaria patients seeking help outside the public health

posts after the introduction of EDTM was probably very

small. The people in the endemic regions were very eager to

start using the well-tolerated and highly effective artemis-

inin drugs and there was prohibitive pressure on the private

sector to be involved in malaria diagnosis and treatment.

Prevalence data are also subject to errors, notably a bias

in selecting subjects for surveys. The population which is

routinely surveyed by the MS may contain a dispropor-

tionate number of women and children: men perform most

of the outdoor work and tend to be absent at surveys, but

are more often exposed to forest-related malaria. However,

the trends observed in this study paralleled the decline of the

malaria prevalence observed in a series of surveys in one

ethnic minority hamlet, which aimed at surveying the entire

population, including males, and a 2-year prospective study

carried out in the Mekong Delta region (Hung et al. 2002;

Erhart et al. 2004a).

With respect to EDTM, the approach was to improve

awareness and responsiveness of both the population and

the health care sector to the extent that EDTM was

19901992

19941996

19982000

2002Years

19931995

19971999

20011991

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1989

residual spraying, highly endemic communes,

Only ITNs, highly endemic communes

ITNs, low endemic communesHighly endemic communes Low endemic communes

% population protected by:

% commune health posts offering EDT:

ITNs + Spraying of residual insecticides, highly endemic communes

Figure 2 Progress of malaria control

interventions in Binh Thuan, Vietnam.

Population protected by insecticide-treatedbed nets free of charge or residual spraying

( , residual spraying in highly endemic

communes; , insecticide-treated bed nets

(ITNs) in highly endemic communes; ,ITNs and residual spraying in highly

endemic minority communes; , ITNs in

low endemic communes) and proportion of

health posts providing early diagnosis andtreatment of malaria ( , highly endemic

communes; , low endemic communes).




permanently available within walking distance to most of

the population. It should be noted that the centralized

introduction of EDTM by the health authorities in Binh

Thuan was possible because the private sector was almost

non-existent in the regions inhabited by the ethnic minority

groups and because these groups, being recognized as

vulnerable, are used to participate in government-supported

programmes. In other regions, the private sector may be

much stronger, requiring different approaches (Erhart et al.

2004a).

Hospital admission data show a decrease after 1989

which is probably an artefact. In the absence of a health

post, district hospitals tend to take over much of the

notification. The increase until 1994 is a real reflection of

0

2

4

6

8

10

12

14

16

18

1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002d w d w d w d w d w d w d w d w d w d w d w d w

Highly endemic communesLow endemic communes

Years, dry season (d ) and wet season (w )

Mal

aria

pre

vale

nce

in s

urve

ys (

%)

Figure 4 Malaria prevalence in Binh

Thuan, Vietnam. Prevalence of malariaparasite carriers in twice yearly malario-

metric surveys. h bars, highly endemic

communes; bars, low endemiccommunes.

0

10

20

30

40

50

60

19891990

19911992

19931994

19951996

19971998

19992000

20012002

Com

mun

e he

alth

pos

t-in

cide

nce

(/10

00)

Hos

pita

l dat

a

Highly endemic communes Low endemic communes

0

50

100

150

200

250

Severe malariaHospital admissions for malaria (x100)

Malaria attributablemortality

Hospital data:

Commune health post incidence:

69%

33% 31

%19

%86

%45

%90

%

93%

97%

100%

100%

100%

100%

100%

100%

100%

100%

72%

76%

81%

92%

94%

91%

87% 98

%

98%

100%

100%

Figure 3 Malaria incidence, severe mor-

bidity, mortality and prevalence in Binh

Thuan, Vietnam. Malaria incidence at

primary health care level and in-hospitalmalaria attributable morbidity and

mortality. On the left ordinate the incidence

at health posts: h bars, highly endemic

communes; bars, low endemic com-munes. The proportion of commune health

posts reporting incidence (%) is indicated

by the numbers on top of the bars. On theright ordinate the hospital data: triangles,

hospital admissions for malaria (·100);asterisk (*), number of cases with severe

malaria; crosses, malaria attributablemortality.




the increased burden of malaria attributable disease in

Vietnam in those years. The marked decrease of severe

morbidity, preceding the decline of total malaria hospital

admission rate after 1994, suggests that timely diagnosis,

treatment and referral improved. There is prior evidence

that an antimalarial treatment policy can affect mortality.

When chloroquine resistance became widespread in West

Africa, mortality increased (Trape et al. 1998). In contrast,

other methods of delivering some form of EDTM by

mothers, school teachers or community health workers

significantly reduced mortality in children under 5 (Kidane

& Morrow 2000). Other methods of delivering early

diagnosis and treatment by community health workers or

teachers have also proved successful.

In Vietnam, affordable and sustainable distribution of

twice yearly re-impregnated ITNs is the mainstay of vector

control. ITNs offer individual protection thereby reducing

morbidity and mortality, especially in infants, but carry-

over to the community by reducing transmission has also

been demonstrated (D’Alessandro et al. 1995; Lengeler

2002; Gimnig et al. 2003; ter Kuile et al. 2003a,b; Phillips-

Howard et al. 2003). Re-impregnation enhances the effic-

acy (D’Alessandro et al. 1995; Gimnig et al. 2003). It is

not easily adopted as a routine practice and therefore

requires some form of central delivery free of charge (Snow

et al. 1999). Nevertheless, untreated nets are better than

none and social marketing of bed nets reduced under 5

mortality in a holoendemic area (Clarke et al. 2001;

Schellenberg et al. 2001). Concerns about rebound

mortality after introduction of ITNs have not been

substantiated (Binka et al. 2002; ter Kuile et al. 2003a;

Phillips-Howard et al. 2003; Diallo et al. 2004).

The epidemiological and socioeconomic conditions in

Binh Thuan were such that a strategy of covering only the

population in highly endemic foci, i.e. the ethnic minority

hamlets and villages, and offering EDTM to all com-

munes, was effective in reducing malaria. However, the

separate effects of EDTM and ITNs could not be

discriminated, probably because they were put into place

in a rather narrow time span. They coincided with the

rapid transformation of a society committed to develop-

ment, education and community participation. Further-

more, other factors came into play. For example, 1998

was a dry year and 1999 very wet. Excessive rains

promote mosquito growth (especially A. dirus). Para-

doxically, droughts also tend to increase mi as farmers try

to compensate their reduced rice crop by forest work. In

addition, the epidemiology of malaria in Southeast Asia is

patchy, correlated with the distribution of forest. This

reduced transmission by targeting the risk population, the

ethnic minority groups, which constitutes only 40% of

the total population in zones 3 and 4. Moreover, the

main vectors, A. minimus and A. dirus, in these

communes were relatively easily deterred by ITNs

because most inhabitants went to bed very early before

the introduction of electricity. Although hard to quantify,

ongoing deforestation may also have affected malaria

transmission. Similarly hard to quantify is the effect of

socioeconomic improvements, such as road and electrical

grid constructions.

Malaria transmission in Binh Thuan is nowadays mainly

confined to the forested regions, requiring permanent

vigilance and special approaches (Erhart et al. 2004b).

This is illustrated by the resurgence of incidence in 1999,

which can largely be explained by increased movement of

people into the forested regions, especially workers in

infrastructural projects such as new roads and hydroelec-

tric plants. Many of these workers come from other

provinces, live in groups in temporary camps, do not apply

appropriate preventive measures and were difficult to reach

for the health services. To date these groups receive special

attention by the MS.

Which lessons can be learned from this example from

Vietnam? The rapid reorganization of the health sector in

Vietnam and, with respect to malaria, the rapidly achieved

success, may have frustrated the statistical inference to

evidence-based health policy. However, the key features of

the Vietnamese approach can easily be distinguished. These

are a flexible and responsive organization, community

participation and surveillance, and clear objectives shared

by health service and population. Human migration to and

from the endemic foci in the forested regions remains a

challenge for malaria control and a permanent pressure on

resources put for adequate malaria control and will. These

general features, rather than details, indicate the way

forward.

Acknowledgement

Thanks are due to Dr Allan Shapira for his valuable

suggestions on the first draft of this manuscript.

References

Binka FN, Hodgson A, Adjuik M & Smith T (2002) Mortality in a

seven-and-a-half-year follow-up of a trial of insecticide-treated

mosquito nets in Ghana. Transactions of the Royal Society of

Tropical Medicine and Hygiene 96, 597–599.

Clarke SE, Bogh C, Brown RC et al. (2001) Do untreated bednets

protect against malaria? Transactions of the Royal Society of


D’Alessandro U, Olaleye BO, McGuire W et al. (1995) A com-

parison of the efficacy of insecticide-treated and untreated bed

nets in preventing malaria in Gambian children. Transactions of




the Royal Society of Tropical Medicine and Hygiene 89, 596–

598.

Diallo DA, Cousens SN, Cuzin-Ouattara N et al. (2004) Child

mortality in a West African population protected with insecti-

cide-treated curtains for a period of up to 6 years. Bulletin of the

World Health Organization 82, 85–91.

Erhart A, Thang ND, Bien TH et al. (2004a) Malaria epidemiol-

ogy in a rural area of the Mekong Delta: a prospective com-

munity-based study. Tropical Medicine and International

Health 9, 1081–1090.

Erhart A, Thang ND, Hung NQ et al. (2004b) Forest malaria in

Vietnam: a challenge for control. American Journal of Tropical

Medicine and Hygiene 70, 110–118.

Ettling MB (2002) The Control of Malaria in Vietnam from 1980

to 2000: What Went Right? Report of Consultancy for World

Health Organization Regional Office for the Western Pacific,

Manila, Philippines.

Giao PT, Binh TQ, Kager PA et al. (2001) Artemisinin for treat-

ment of uncomplicated falciparum malaria: is there a place for

monotherapy? American Journal of Tropical Medicine and

Hygiene 65, 690–695.

Giao PT, de Vries PJ, Tran BQ et al. (2002) Artemisinin or chlo-

roquine for blood stage Plasmodium vivax malaria in Vietnam.

Tropical Medicine and International Health 7, 858–864.

Giao PT, de Vries PJ, Hung LQ et al. (2003) Atovaquone-

proguanil for recrudescent Plasmodium falciparum in

Vietnam. Annals of Tropical Medicine and Parasitology 97,

575–580.

Giao PT, de Vries PJ, Hung LQ et al. (2004) CV8, a new combi-

nation of dihydroartemisinin, piperaquine, trimethoprim and

primaquine, compared with atovaquone-proguanil against fal-

ciparum malaria in Vietnam. Tropical Medicine and Interna-

tional Health 9, 209–216.

Gimnig JE, Vulule JM, Lo TQ et al. (2003) Impact of permethrin-

treated bed nets on entomologic indices in an area of intense

year-round malaria transmission. American Journal of Tropical


Hung LQ, de Vries PJ, Giao PT et al. (2002) Control of malaria: a

successful experience from Vietnam. Bulletin of the World

Health Organization 80, 660–666.

Hung LQ, de Vries PJ, Binh TQ et al. (2004) Artesunate with

mefloquine at various intervals for non-severe Plasmodium

falciparum malaria. American Journal of Tropical Medicine and

Hygiene 71, 160–166.

Kidane G & Morrow RH (2000) Teaching mothers to provide

home treatment of malaria in Tigray, Ethiopia: a randomised

trial. Lancet 356, 550–555.

ter Kuile FO, Terlouw DJ, Kariuki SK et al. (2003a) Impact of

permethrin-treated bed nets on malaria, anemia, and growth in

infants in an area of intense perennial malaria transmission in

western Kenya. American Journal of Tropical Medicine and

Hygiene 68, 68–77.

ter Kuile FO, Terlouw DJ, Phillips-Howard PA et al. (2003b)

Impact of permethrin-treated bed nets on malaria and all-cause

morbidity in young children in an area of intense perennial

malaria transmission in western Kenya: cross-sectional survey.

American Journal of Tropical Medicine and Hygiene 68,

100–107.

Le NN, De VP, Le TD et al. (1997) Single dose artemisinin-

mefloquine versus mefloquine alone for uncomplicated

falciparum malaria. Transactions of the Royal Society of


Lengeler C (2002) Insecticide-treated Bednets and Curtains for

Preventing Malaria. The Cochrane Library (Issue 2), Update

Software, Oxford, UK.

Nguyen VK (1999) Malaria in Vietnam 1957–1997: strategies

and tactics. Malaria Course on Intensified Control Measures,

Cho Ray Hospital, Ho Chi Minh City, 1–10.

Thi Phan V, Quoc Tuy T & Xuan Hung L (1999) Malaria epi-

demics in Vietnam (1955–1997). Mekong Forum 1, 12–16.

Phillips-Howard PA, Nahlen BL, Kolczak MS et al. (2003) Effic-

acy of permethrin-treated bed nets in the prevention of mortality

in young children in an area of high perennial malaria trans-

mission in western Kenya. American Journal of Tropical


Schellenberg JR, Abdulla S, Nathan R et al. (2001) Effect of large-

scale social marketing of insecticide-treated nets on child sur-

vival in rural Tanzania. Lancet 357, 1241–1247.

Snow RW, McCabe E, Mbogo CN et al. (1999) The effect of

delivery mechanisms on the uptake of bed net re-impregnation

in Kilifi District, Kenya. Health Policy and Planning 14, 18–25.

Trape JF, Pison G, Preziosi MP et al. (1998) Impact of chloroquine

resistance on malaria mortality. Comptes Rendus de l’Academie

des Sciences (Series III) – Sciences de la Vie 321, 689–697.

Verle P, Lieu TT, Kongs A, Van der SP & Coosemans M (1999)

Control ofmalaria vectors: cost analysis in a province of northern

Vietnam.TropicalMedicine and InternationalHealth4, 139–145.

Warrell DA (2001) ‘To search and study out the secret of tropical

diseases by way of experiment’. The Lancet 358, 1983–1988.

WHO (1992) Vector Resistance to Pesticides. 15th Report of the

WHO Expert Committee on Vector Biology and Control.

Technical Report Series 818. WHO, Geneva.

WHO (2000) 20th Report of the WHO Expert Committee on

Malaria. Technical Report Series 892. WHO, Geneva.

Authors

Nguyen Van Nam (Vietnamese corresponding author) and Le Van Toi, Malaria Control Center, 133A Hai Thuong Lan Ong, Phan

Thiet, Binh Thuan Province, Vietnam. E-mail: [email protected]

Peter J. de Vries (English corresponding author), Division of Infectious Diseases, Tropical Medicine and AIDS, Academic Medical

Center F4-217, PO Box 22700, 1100 DE Amsterdam, The Netherlands. E-mail: [email protected]

Nico Nagelkerke, Department of Medical Statistics, Leiden University Medical Center, Postbox 9604, 2300 RC Leiden, The

Netherlands




Malaria control in Vietnam: the Binh Thuan experience

Documents

Transcript of Malaria control in Vietnam: the Binh Thuan experience