Knowledge of mosquitos in relation to public and domestic control

Knowledge of mosquitos in relation to publicand domestic control activities in the cities ofDar es Salaam and TangaC. Stephens,1 E.T. Masamu,2 M.G. Kiama,2 A.J. Keto,2 M. Kinenekejo,2K. Ichimori,23 & J. Lines1

A study of community awareness of mosquitos and related subjects in the residential areas of two Tan-zanian cities (Dar es Salaam and Tanga) showed that residents were well aware of mosquitos. Almostall claimed to use some form of domestic mosquito control product for their personal protection, andmany spend a significant portion of the household income on this.

The problems of nuisance-biting and malaria transmission are usually not separated and are con-sidered to be the result of poor environmental hygiene, for which both residents and local authoritiesare responsible. Although Culex mosquitos are not a primary target of the Urban Malaria Control Project(UMCP), the persistence of nuisance-biting has made residents sceptical and dissatisfied with insecti-cide spraying. The residents' priorities are evidently not the same as those of the health authorities, yetmutual cooperation is essential. In order to maintain community support, campaigns aimed at malariavectors should consider the need for additional measures to control Culex mosquitos, such as thosenow being tried by the UMCP.

Mosquito breeding sites are non-specifically associated with rubbish and standing water of all kinds,and so the actions that the community considers necessary for mosquito source reduction tend to bepoorly targeted. Residents do not recognize that some sources produce malaria mosquitos while othersproduce nuisance mosquitos. The environmental anti-mosquito measures currently promoted by healtheducation and other forms of propaganda are also poorly targeted. While some of them are directed atimportant Culex breeding sites, others are aimed at sites of little importance for mosquitos of any kind.Almost no attention is paid to the most productive breeding sites for Anopheles malaria vectors.

IntroductionMost organized mosquito control strategies requirepublic support of one kind or another, and the extentof people's cooperation can determine the success orfailure of the entire campaign (1). To be successfuland sustainable, therefore, an intervention has tomeet the expectations of both the public health plan-ners and the local people.

Community reactions to vector control are great-ly influenced by the perceived impact of interven-tions on insect pests and nuisance-biting (2, 3). Inrural areas of Africa, Anopheles mosquitos are

I Departments of Public Health & Policy and of Medical Parasi-tology, London School of Hygiene and Tropical Medicine, KeppelStreet, London WC1 E 7HT, England. Requests for reprintsshould be sent to C. Stephens or J. Lines at this address.2 Dar es Salaam and Tanga Urban Malaria Control Project,United Republic of Tanzania.3 Present address: c/o WHO, PO Box 77, Apia, WesternSamoa.Reprint No. 5573

responsible not only for the transmission of malariaand filariasis, but also for most of the nuisance-bitingat night. There is therefore no practical need to dis-tinguish between the problems of nuisance-bitingand disease transmission. In urban areas, on the otherhand, anophelines are relatively rare. This is becausethey breed in unpolluted breeding sites-sunlit pud-dles, pools and ricefields in the case of Anophelesgambiae s.l., and swamps in the case of A. funestus.In towns, of course, most such places are either builtover or polluted, and organic pollution favoursanother mosquito, Culex quinquefasciatus, which cantransmit filariasis but not malaria parasites. This spe-cies breeds abundantly in pit latrines, soakage pits,and in drains blocked by rubbish (4). Several studiesindicate that it greatly outnumbers malaria vectors,often by more than 100 to 1 in African towns andcities (5, 9).

Malaria is none the less an important publichealth problem, even in towns, firstly because A.gambiae is such an efficient vector, and secondlybecause urban residents, being less exposed, tend tobe less immune than their rural counterparts (10-13).

Bulletin of the World Health Organization, 1995, 73 (1): 97-104 © World Health Organization 1995 97

C. Stephens et al.

In Dar es Salaam, for example, malaria reportedlyaccounts for 13% of outpatient attendances, nearly10% of admissions, and over 5% of hospital deaths.a

The contrast between the mosquitos of urban andrural areas turns out to be a mixed blessing for urbanhealth planners. On the one hand, Anopheles breed-ing sites in urban areas tend to be fewer, easier toidentify and more accessible than those in the coun-tryside, and well-targeted attacks can thereforeachieve a worthwhile degree of control; this wouldbe unthinkable in most rural areas, where breedingsites are numerous, shifting, and scattered all over thecountryside. The good news for town dwellers, there-fore, is that malaria is potentially more manageableby environmental control measures in towns than inthe country. The bad news is that actions directedagainst urban malaria vectors will usually have littleeffect on nuisance-biting, which in towns is mainlydue to other mosquitos breeding in other places.

Nevertheless, malaria takes precedence over allother mosquito-related problems, from the point ofview of public health authorities (14). Do residentsshare this sense of priority? This question wasrecently addressed by an Urban Malaria Control Pro-ject (UMCP), which has been carrying out chemicallarviciding and residual house-spraying againstmalaria vector mosquitos in the cities of Dar esSalaam and Tanga in the United Republic of Tanza-nia. By 1990, there were already signs in the pressand elsewhere that public support for mosquito con-trol efforts was waning. Project staff noted that"people are becoming reluctant and uncooperativein the question of self-help, leading to minimal com-munity participation in vector control, e.g., clearingdrains, taking out their goods during residual housespraying, etc." (unpublished report of the UMCP,1990). One suggested explanation for this lack ofsupport was that residents had seen no reduction innuisance-biting.

We therefore undertook a qualitative study ofthe knowledge, attitudes and practice of residentstowards mosquitos and mosquito control measures.Our methods and findings are described in thisarticle.

MethodsStudy areas

Dar es Salaam and Tanga are both located on theTanzanian coast. Dar es Salaam is the largest urban

a World Health Organization. Malaria control programme inDar es Salaam and Tanga. Feasibility study to justify proposal.Dar es Salaam, Ministry of Health, 1986 (unpublished report).

centre in the country, with a population of about 1.9million; Tanga has a population of about 250 000(15). The study was carried out in four areas of Dares Salaam (Ilala Shauri Moyo, Temeke Wailes,Magomeni, and Mikocheni 'A') and three areas ofTanga (Kiomoni, Amboni, and Mafuriko) in Augustand September 1991. The study areas in Dar esSalaam were neither in the very centre nor at theedge of the city, and are typical of urban residentialareas of many Tanzanian towns and cities. The studyareas in Tanga were more peripheral and less urban-ized than those in Dar es Salaam.

In both cities, malaria is the most importantmosquito-bome disease and a public health problemof primary importance. Entomological sampling(UMCP, unpublished data) indicates that, as expect-ed, malaria transmission is more intense in the semi-rural study areas in Tanga than in the more centralareas studied in Dar es Salaam. Filariasis, which inurban areas is transmitted mainly by C. quinquefas-ciatus, is a problem of secondary importance in bothcities, the more serious manifestations of elephantia-sis being confined mainly to older people. Yellowfever has not been reported for many years in eitherDar es Salaam or Tanga.

Since 1988, with financial support from the Jap-anese govemment, the Dar es Salaam and TangaUMCP has been conducting a malaria vector controlcampaign, using the conventional approach of chem-ical larviciding in central areas, and house-sprayingin semi-rural peripheral areas. The insecticide feni-trothion has been used for both purposes. The larvi-ciding has been directed primarily against the breed-ing sites of Anopheles but also, in some places,against C. quinquefasciatus. However, C. quinque-fasciatus populations in both cities are known to behighly resistant to a wide range of organophosphatesincluding fenitrothion (16, 17). Entomological sam-pling has confirmed that C. quinquefasciatus den-sities remain high.

Interview methods

Methods were selected because they were non-intrusive to householders, and simple and quick forexisting staff to perform and analyse. Direct inter-views with residents were conducted in the Swahililanguage using focus group discussion techniques(e.g. 18). Two kinds of discussion took place. First,discussions were conducted in pre-arranged andsometimes rather formal meetings with communityleaders, party officials and balozis (i.e., the politicalrepresentatives of groups of about ten house-holds). There were five such discussions in Dar esSalaam and three in Tanga, with a combined total of94 community leaders. Second, in-depth focused dis-

98 WHO Bulletin OMS. Vol 73 1995

Knowledge of mosquitos in relation to public and domestic control

cussions were conducted informally with individualsand small groups of residents encountered on thestreet, in shops, and at home. These concentrated ongroups that were poorly represented in the meetingswith,community leaders, i.e., on women, the youngand the old; 28 such conversations took place in Dares Salaam, and 44 in Tanga, with totals of 80 and 95residents, respectively.

The interviewing team carried field guides giv-ing key prompts to direct interviews, but notes weretaken after the meetings rather than at the time. Eachinterviewer independently wrote case notes on a pre-structured form, including key words and phrasesused by interviewees on each topic, and details onthe mood and surroundings of the interview. Up tothree independent reports were produced for eachinterview or meeting.

ResultsThe interview notes were analysed for commentsrelevant to the following topics: knowledge of mos-quitos and mosquito-bome disease, understanding ofbreeding sites of mosquitos, attitudes to the controlmeasures carried out by the UMCP, and domesticmethods of mosquito control.

Knowledge of mosquitosAll residents interviewed were aware of mosquitos,which were often present and attempting to bitewhen interviews were conducted in the evening.Mosquitos were distinguished by their biting habitand the noise made by flying adults. The existence ofmore than one kind of mosquito was recognized byonly a minority of respondents; in each case, bitingwas stressed as a feature of all mosquitos. The name"Anopheles" was mentioned more frequently than"Culex", but no one commented on which kind wasmore common. Variations in the size and colour ofmosquitos were sometimes noted, but it was clearthat for most people "a mosquito is a mosquito".Typical comments included:- "Different types? I haven't really looked. But

now you come to mention it, there is a smallblack one now. Small and weak-looking andblack: I haven't seen that one before" (old man,Magomeni);

- "Some are big and make noise, others are smalland cause itching when they bite" (young man,Kiomoni);

- "All are alike" (young man, Mafuriko).

Knowledge of mosquito-borne diseaseMost respondents made an explicit connection be-tween mosquitos and malaria. Malaria is seen as an

important health problem for the community. Fever,and less frequently anaemia, were mentioned assymptoms. A number of respondents, particularlymothers and community elders, noted that malaria isa particular danger for young children, and protec-ting them was seen as an important reason for usingnets and insecticides.- "They cause malaria... (so) I bought the net for

the new baby when he was only three monthsold" (young woman, Magomeni);

- "I've got a net for the baby... otherwise he'll getfever" (young woman, Ilala);

- "Children - they're the ones who get fever ifthey're bitten" (old man, Magomeni).In Tanga, a greater proportion of respondents

reported recent episodes of malaria in themselves ormembers of their family:- "They cause fever and weakness. Even today I

have taken chloroquine tablets because I havefever" (middle-aged man, Kiomoni);

- "My child got a blood transfusion because ofmalaria which led to a lack of blood" (younglady, Kiomoni);

- "Many of our children have died because ofmalaria" (old lady, Kiomoni).Filariasis and hydrocoele were mentioned rather

more frequently in Tanga than in Dar es Salaam, butin both places much less than malaria. Yellow feverwas mentioned by only two respondents.

The link with disease appeared to be withhuman-biting mosquitos generally. Only threerespondents out of 80 in Dar es Salaam namedAnopheles as the vector for malaria; while in Tanga,one resident stated that Anopheles mosquitos carrymalaria, Culex transmits filariasis, and Aedes carriesyellow fever. In general, very few people associateddifferent mosquitos with the different diseases theymentioned. A group of balozis in Mafuriko reportedthat "mosquitos cause malaria, filariasis and hydro-coele", "they also cause yellow fever", and "they area nuisance too".

The nuisance role of the mosquito was repeatedlyemphasized, and all residents cited this as a key reasonfor spending money on protection against mosquitobites. From the range of graphic comments made byresidents, three main portraits of the mosquito emerged,the first two being mentioned more frequently:

(1) Mosquitos as disease agents: "they bringdisease, malaria..."; "when they bite, you get feverand have to go to bed..."; "we have no choice but toprotect ourselves from mosquitos"; "malaria is kill-ing us".

WHO Bulletin OMS. Vol 73 1995 99

C. Stephens et al.

(2) Mosquitos as irritants: "at night you can'tsleep"; "their bites cause rashes and itching..."; "theykeep you awake all night..."; "some people usebooze so as not to feel the bites..."; "sometimes youcan't sleep for the noise of people in houses aroundtrying to kill mosquitos..."; "they itch so you can'tsleep".

(3) Mosquitos as debilitators: "they suck yourblood..."; "they inject you with toxins..."; "they eatpeople's blood..."; "they reduce your blood supply".

Breeding sitesMost people said that mosquitos breed in one ormore types of water, especially "standing water" and"dirty water". Latrines, cess pits, drains, ditches,ponds and swamps were often specifically identified,and tins and coconut shells were also mentioned.Some respondents cited sources not specifically rela-ted to water, although it was sometimes unclearwhether these were thought of as breeding sitesor as hiding places of mosquitos. In Tanga, thesenon-water-related sites were mostly vegetation suchas banana trees, bushes, grass, and overgrown areas,while in Dar es Salaam, they were mainly associatedwith dirt and rubbish inside and outside houses.None of the interviewees, even those who named dif-ferent mosquito species, distinguished the breedingsites of different kinds of mosquito.

Mosquitos are apparently regarded as one symp-tom of wider environmental deterioration, and whenthese concems emerged, the subject of mosquitostended to be subsumed in broader discussions of dirtand rubbish, drainage and sanitation, and councilservices in general.

Attitudes to organized mosquito controlactivitiesRespondents judged the success of UMCP controlactivities in terms of the impact on mosquitos as awhole; no one distinguished between efforts tocontrol malaria vectors and those aimed at Culexmosquitos. Although some people recognized thatmosquito control is not an easy task, favourablecomments were greatly outnumbered by complaints.The most fundamental complaints, particularly inDar es Salaam, related to the scope and focus of cur-rent control strategies. Many older respondents remi-nisced about broader municipal services and mosqui-to control measures in the past, which they said hadbeen more effective:

"In the old days drains, puddles, pits and rubbishwere cleared up and inspectors came; now all thishas died and they just come to spray every two orthree months... spraying those chemicals inbreeding sites is good, but it's not a permanent

solution... health officers never come here any-more, and health regulations are not enforced...people neglect their surroundings and let breed-ing sites come up... it's just not like the old days"(group of balozis, Ilala);

- "There used to be tin collectors, mosquito-findersgoing from house to house, people who cut andcleaned drains, people who sprayed chemicals inthe water, and even spraying from the air" (balozi,Magomeni);

- "Environmental sanitation is good, like weedinground our houses, cutting down sugar cane, bana-na trees and long grass" (young woman, Amboni).

In Dar es Salaam, some people reported havingjoined in community-based source-reduction activ-ities, clearing drains of silt and rubbish by hand.These activities were instigated by the UMCP andlocal authorities, and participation in them wasreported with pride. However, there were severalcomplaints that the Council had repeatedly broken itspromise to send a lorry to take away the debris theyhad dug out so that, with the next heavy rain, it allwashed back into the drain.

Many residents in Dar es Salaam complainedabout the spray teams responsible for weekly larvi-ciding, particularly the long and irregular intervalsbetween their visits:- "What activities? Oh, yes-those people with

pumps on their backs like babies. They did comeand spray the place, but later I only saw themwalking along the street... I don't know wherethey were going" (female secretary, Ilala);

- "Nowadays the mosquito people don't go intothe small streets, they come once and don't comeback... not like the old days" (balozi, Ilala).

In Dar es Salaam, the consensus was that mos-quito densities declined for only a few days follow-ing larviciding activities, and a few people specifi-cally cited resistance as a reason for this.

"Some workers came and put chemicals in ourlatrines and along the drains. Yes, the mosquitosdid decrease... for two or three days" (youngman, Mikocheni);"They used to use a better spray. DDT?-yes, Iread that it wasn't used anymore, but it workedbetter, it used to kill mosquitos properly" (oldman, Magomeni);"The blocked drain is our worst problem. Thechemicals used to work but they're not good any-more-they don't work" (Balozi, Temeke).

Similarly in Tanga, where the study areas receivedhouse-spraying at intervals of several months, several

WHO Bulletin OMS. Vol 73 1995100


respondents stated that this treatment had worked wellon the first occasion, but that subsequent spray-roundshad been much less effective. There were also com-ments on the smell of the insecticide:- "The first spraying was very effective but what

followed was like limewash, it painted our walls"(old man, Mafuriko);

- "Spraying in houses is useless as it does notbring any relief, the first phase was OK, perhapsnowadays they put too much water" (old man,Amboni);

- "The chemical you are spraying is useless, it justmakes mosquitos more aggressive, it does notkill them. We are fed up with this job of remov-ing our household items, washing the floor andputting back the furniture. That is why some peo-ple keep their doors closed. The chemical stinks"(young women, Amboni).

It may be noted that this last comment echoesthe objections to house spraying often voiced inmalaria control programmes (19).

The only control activity to gain as manyfavourable as unfavourable comments was insecti-cide fogging carried out in some parts of Dar esSalaam:-"When the machine works by fogging, there are no

mosquitos for up to a week" (young man, Ilala).

Personal protection against mosquitosAlmost everyone claimed to be using domestic mos-quito control products of some sort. The degree towhich different individuals used such products appa-rently depended mainly on their perceptions ofnuisance-biting. However, an explicit link with pro-tection from malaria was made frequently, particu-larly with reference to mosquito nets for protectingchildren. Cost was by far the most frequently re-ported constraint. In Dar es Salaam, the four respon-dents who reported not using mosquito control intheir homes were elderly, and three cited shortageof money as the reason. The fourth said that protec-tion from mosquitos was important for children butnot for old people. As factors influencing choice, re-spondents cited a variety of factors including cost,effectiveness, practicality, and the side-effects asso-ciated with some methods.

The most frequently cited method was the burn-ing of repellents-almost entirely mosquito coils inDar es Salaam, and both coils and local herbs of thegenus Ocimum in Tanga. Coils were seen as beingonly partially effective, but better than nothing. Itwas noted that a coil can protect the whole room, butalso that it does not last the whole night. There wasstrong criticism of the perceived respiratory side-

effects of burned repellents, especially coils, and forsome people these side-effects outweighed the bene-fits.- "We normally burn kivumbasi [Ocimum spp.]

and sometimes coils, but coils are expensive"(old man, Kiomoni);

- "I use coils but they are not effective and causecough and flu" (old man, Mafuriko);

- "I don't like mosquito coils: they're not very good,and the smoke intoxicates me and makes me heavyand sleep too much, giving mosquitos more time tosuck my blood freely" (young man, Temeke).

Most coil users in Dar es Salaam claimed to use acoil every night, and although this was probably anexaggeration in some cases, it implies a monthlyexpenditure of about 360 shillings (approximatelyU$ 1.00). This can be compared with the monthly rentfor one room, as reported by residents of Mikocheniand Magomeni, which varied around 750 shillings.

Nets were generally recognized as being themost effective method of personal protection, andespecially important for the protection of babies. InDar es Salaam, three out of five women who hadvery young babies had bought nets with the specificintention of protecting their infants from mosquitobites. Some respondents noted that nets need to betucked in properly, that they do not protect youbefore going to bed, and that mosquitos can still biteif the occupants sleep with an arm or leg touchingthe net. A few commented that it was very hot tosleep under a net. However, the main constraint onthe use of nets was clearly the cost (2500-4000 shil-lings, US$ 7-11). Several people remarked that netsget holes quickly and so do not last long, in relationto the cost. However, even assuming that a net lastsonly for one year, this amounts to a monthly expen-diture of about 265 shillings (US$ 0.75), less thanthe cost of using one coil a night. The usual com-ment on this, often made spontaneously, was thatcoils are more affordable in practice because theycan be bought one by one, whereas a net requires alarge initial outlay.- "Nets are best-you don't need education to

know that. You just lie there and it doesn't mat-ter if the house is full of mosquitos. But if youearn only three or four thousand shillings amonth, and a net is two thousand, what do youeat?" (balozi, Ilala);

- "Nets are best-but if you don't have the where-withal, you have to try and buy bit by bit"(young woman, Magomeni);

- "Nets are good because you sleep nicely but theyare very expensive and I cannot afford one" (oldwoman, Amboni);


C. Stephens et al.

-"I do not use coils because they have very seri-ous side-effects on me. Instead I am using a netwhich I bought in 1985" (old man, Amboni).

Some respondents in Dar es Salaam reportedusing a knockdown spray, consisting either of a liq-uid insecticide sprayed with a hand pump or an aero-sol can, which was used in the evening before goingto bed. Sprays were appreciated for their immediateeffectiveness, but were criticized for their cost andlack of sustained benefit. If a can lasts one weekwith nightly use, the monthly cost would be aboutUS$ 3.30 - much more than a net or coils.

A few people said they used nothing but a fan.Fans were said to cause colds, and to be good only inthe immediate vicinity of the fan; they were regardedas being too expensive for most people (10 000 to15 000 shillings, US$ 28-42). Very few respondentsfavoured window gauze which, being expensive andeffective only in combination with other efforts toseal the house, was said to be something for richpeople only. These methods were not reported in themore rural areas in Tanga, although several peoplementioned that they would use them if they couldafford it.

DiscussionThe residents of Dar es Salaam and Tanga regardmosquitos as a nuisance, as debilitators, and asdisease vectors. They place a high priority on domes-tic control, almost everyone claiming to make use ofone or more of the available products. Cost was themost important constraint on choice and degree ofuse. Even allowing for some exaggeration, it appearsthat a considerable portion of household income isspent on this. However, price was not the only consi-deration; the relative efficacy of different productswas also taken into account, as were questions ofdurability and both positive and negative side-effects.

While coils were described by many as beingless satisfactory but more affordable, nets were con-sistently the most desirable option, a view thatshould be taken into account by those involved inlarge-scale operations using pyrethroid-treated nets.Indeed, on several occasions, people wondered whenthe UMCP would introduce treated nets into its pro-gramme-as it has done now..

The residents' understanding of the role of mos-quitos as disease vectors was high. Nevertheless,with the exception of some mothers who used nets toprotect their babies specifically against malaria, mostpeople appeared to use domestic control products asprotection first against nuisance-biting and seconda-rily against malaria. Both this and the high level of

spending on protection are consistent with the resultsof studies in other African cities (20-24). While theanswers to questions about mosquitos sometimesmentioned malaria, and vice versa, filariasis wasmentioned by fewer people, and yellow fever onlyrarely. This reflects the relative importance of thesediseases in terms of public health.

A significant minority knew the name "Anophe-les", and some the name "Culex"; but for most people"mosquitos are mosquitos"; and while they may varyin appearance, different kinds of mosquitos were notlinked to different diseases or to different controlmeasures. This kind of distinction is unnecessary forthe use of domestic control products, most of whichare more or less equally effective against the twomain mosquito problems of Dar es Salaam andTanga: the nuisance-biting C. quinquefasciatus andthe less numerous but more dangerous A. gambiae.

Discrimination between mosquito types isimportant, however, when it comes to control meas-ures that affect one of these problems but not theother. This includes the efforts of the UMCP, whichhave been directed almost entirely against malariavectors, but employing an insecticide to which themain nuisance-biting mosquito is resistant. As origi-nally suspected, residents' critical attitudes towardsUMCP activities stem largely from the lack of anynoticeable reduction in nuisance-biting. This con-firms that public support for urban mosquito con-trol schemes is likely to depend on a reduction innuisance-biting, and sometimes additional measuresmay be necessary specifically to ensure this.

Many residents contrasted current control effortswith those recalled from "the good old days". Theysaid that local health officers in the past used to visitregularly to enforce measures of environmental sani-tation, which seem to have been aimed more at gen-eral hygiene than at mosquitos. This accords withthe community's perception of mosquito breedingsites. Most people have a sketchy knowledge of mos-quito breeding sites, being able to identify some categ-ories but not others, and often naming sites that areunimportant. Moreover, there is no recognition thatmosquitos causing different problems come fromdifferent places. This generalized view appears tobe shared by almost everyone, including teachers,politicians, health educators and expatriates.

Tanzanians are frequently exhorted by healtheducation messages of all kinds to clear up theirenvironment for the sake of malaria control. Unfortu-nately, the specific actions recommended by suchcampaigns, while they may have other benefits, areunlikely to have this effect. For example, the cuttingdown of vegetation around houses-particularlymaize and bananas is widely believed to be a sim-ple and moderately effective form of malaria mos-

0WHO Bulletin OMS. Vol 73 1995102


quito control (25), although there seems to be no evi-dence to support this suggestion (26), and at leastsome evidence against it (27). The most commonsuggestion, and the one that most often leads to posi-tive action, is the clearance of silt and rubbish frompolluted drains, which are major breeding sites forCulex but not Anopheles mosquitos. Swamps areoften cited as another target to be tackled with com-munity participation, but in fact the swamps them-selves are relatively harmless, as long as they remaincovered in tall reeds. Often no mention is made ofricefields and cultivation ridges, which are almostcertainly the most productive Anopheles breedingsites by far. The fact that such information is notavailable to the residents of Dar es Salaam potential-ly hampers effective action for mosquito control.

People are concerned about mosquitos andwould control them if they could. Public sources ofinformation tend to reinforce prevailing views onmosquitos and where they come from. Consequently,community-based efforts for mosquito control aremostly aimed at Culex breeding sites, which are themost conspicuous, and rarely at the less obviousAnopheles sites, which are the usually stated target.It might be argued that a more specific messagewould be too complicated, and that people are just asconcerned about nuisance-biting as they are aboutmalaria. However, while C. quinquefasciatus and A.gambiae s.l. larvae are sometimes found together inAfrican cities, their most productive breeding sitesare actually quite separate. In order to target sourcereduction measures effectively, some difficultchoices have to be made-for example, whetherto concentrate on the nuisance-biting Culex or themalaria vector Anopheles; and if the latter, whether toforego crops of rice, yam and sweet potato for thesake of malaria control.

These decisions, if they are to gain thecommunity's support and participation, must bemade through some kind of social consensus. By-laws governing urban cultivation, for example, willremain unenforceable unless they are backed upsocially in this way. This cannot happen withoutmore information. More specific awareness, amongresidents and community leaders, of which mosquitoproblems come from which sites must be promoted.This may not be complicated: a three-way distinctionbetween small containers, foul water, and unpollutedpuddles and streams would be enough. This wouldhelp both local communities and public healthauthorities to act according to their own environmen-tal priorities.

AcknowledgementsThis study was supported by the Overseas DevelopmentAdministration of the British government through the

Urban Health and Tropical Disease Control Programmesat the London School of Hygiene and Tropical Medicine,as well as by the Japanese government and the JapaneseInternational Co-operation Agency through the Dar/TangaUrban Malaria Control Project. We thank the TanzanianMinistry of Health, the city authorities of Dar es Salaamand Tanga, and the residents of our study areas for theirassistance and hospitality, and most of all our colleaguesin the UMCP for their encouragement and help.

Resume

La connaissance des moustiques et sesrapports avec les activit6s generales etdomestiques de lutte a Dar es-Salaam et aTangaUne 6tude portant sur la connaissance des mous-tiques et autres questions en relation avec lesmoustiques a 6t6 r6alis6e dans divers quartiersd'habitation de deux villes de Tanzanie, Dar es-Salaam et Tanga. Des discussions en groupe ontete menees avec les notables locaux et avec deshabitants et ont port6 sur les moustiques, lesmaladies qu'ils transmettent et les methodes delutte.

Les r6sidents ont une bonne connaissancedes moustiques; le paludisme et, dans une moin-dre mesure, la filariose sont reconnus comme ma-ladies transmises par les moustiques, et sont davan-tage mentionn6s dans les zones p6riurbaines deTanga que dans les quartiers urbains de Dar es-Salaam. La quasi-totalite des personnes inter-rogees utilisent un produit de lutte contre lesmoustiques et nombre d'entre elles declarent yconsacrer une part notable de leur revenu. Laplupart des habitants, a 1'exception des meres detres jeunes enfants, mentionnent comme raison del'utilisation de tels produits la protection contrela gene occasionnee par les piqOres plut6t quecontre le paludisme.

Toutefois, le lien perqu entre les moustiqueset le paludisme est tres net, comme le montrentles r6ponses aux deux types de questions. Lesmoustiques et le paludisme sont tous deux perquscomme la cons6quence d'une mauvaise hygienede l'environnement, dont sont responsables a lafois les habitants et les autorit6s locales. Bien queles moustiques du genre Culex ne fassent paspartie des cibles principales du Projet de luttecontre le paludisme urbain, la persistance demoustiques nuisants provoque le scepticisme et lem6contentement des habitants vis-a-vis des pulv6-risations d'insecticides dans le cadre de ce projet.


C. Stephens et al.

Les sites de reproduction des moustiquessont associes de fagon non sp6cifique aux detri-tus et aux eaux stagnantes de tous types, et lesmesures jugees necessaires pour reduire les gitesde moustiques tendent a etre mal ciblees. 11 n'estpas reconnu que certains gites produisent desmoustiques vecteurs du paludisme et d'autres desmoustiques nuisants. Les mesures de lutte contreles moustiques fondees sur la gestion de 1'envi-ronnement, actuellement pr6n6es par l'educationsanitaire et par d'autres campagnes d'information,sont 6galement mal ciblees. Certaines sont diri-g6es contre d'importants sites de reproduction deCulex tandis que d'autres visent des sites de peud'int6ret. Les sites de reproduction les plus impor-tants pour les anopheles vecteurs du paludismene sont presque jamais mentionnes.

References1. Service MW. Community participation in vector-

borne disease control. Annals of tropical medicineand paratisology, 1993, 87: 223-234.

2. Rafatjah H. The problem of resurgent bedbug infes-tation in malaria eradication programmes. Journal oftropical medicine and hygiene, 1971, 74: 53-56.

3. Njunwa KJ et al. Trial of pyrethroid impregnatedbednets in an area of Tanzania holoendemic formalaria. Part l: Operational methods and acceptabil-ity. Acta tropica, 1991, 49: 87-96.

4. Subra R. Etudes ecologiques sur Culex pipiens fati-gans dans une zone urbaine de savane souda-nienne Ouest-Africaine. Cahiers ORSTOM s6rieentomologie m6dicale et parasitologie, 1971, 9:73-102.

5. Bang YH et al. Changes in mosquito populationswith urbanisation in Tanzania. East African medicaljournal, 1977, 54: 403-410.

6. Robert V et al. Le paludisme urbain a Bobo-Dioulasso. I. Etude entomologique de la transmis-sion. Cahiers ORSTOM s6rie entomologie m6dicaleet parasitologie, 1986, 24: 121-128.

7. Trape J-F et al. Vector density gradients and theepidemiology of urban malaria in Dakar, Senegal.American journal of tropical medicine and hygiene,1992, 47:181-189.

8. Fondjo E et al. Le paludisme urbain a Yaound6(Cameroun). II. Etude entomologique dans deuxquartiers peu urbanises. Bulletin de la Soci6t6 dePathologie exotique, 1992, 85: 57-63.

9. Trape J-F, Zoulani A. Malaria and urbanisation inCentral Africa: the example of Brazzaville. Part Il:Results of entomological surveys and epidemiologi-cal analysis. Transactions of the Royal Society ofTropical Medicine and Hygiene, 1987, 81: 34-42.

10. Gardiner C et al. Malaria in urban and rural areasof Southern Ghana: a survey of parasitaemia, anti-bodies, and antimalarial practices. Bulletin of theWorld Health Organization, 1984, 62: 607-613.

11. Benasseni R et al. Le paludisme urbain a Bobo-Dioulasso. I. Etude de la morbidite palustre. CahiersORSTOM serie entomologie m6dicale et parasitolo-gie, 1987, 25: 165-170.

12. Watts TE et al. Malaria in an urban and a rural areaof Zambia. Transactions of the Royal Society ofTropical Medicine and Hygiene, 1990, 84: 196-200.

13. Trape J-F et al. Malaria morbidity among childrenexposed to low seasonal transmission in Dakar,Senegal, and its implications for malaria control intropical Africa. American journal of tropical medicineand hygiene, 1993, 48: 748-756.

14. Kilama WL, Kihamia CM. Malaria. In: MwakuloGMP et al., eds. Health and disease in Tanzania.London, Harper Collins, 1991: 117-132.

15. Mwakulo GMP et al. Urbanisation and health. In:Mwakulo GMP et al., eds., Health and disease inTanzania. London, Harper Collins, 1991: 289-295.

16. Curtis CF, Pasteur N. Organophosphate resistancein vector populations of the complex of Culex pipi-ens L. (Diptera: Culicidae) Bulletin of entomologicalresearch, 1981, 71: 153-161.

17. Curtis CF et al. Assessment of the impact of chlor-pyrifos resistance in Culex quinquefasciatus on acontrol scheme. Insect science and its application,1984, 5: 263-267.

18. Krueger RA. Focus groups: a practical guide forapplied research. London, Sage publications, 1988.

19. Hongvivatana T et al. An observational study of DDThouse spraying in a rural area of Thailand. Journal oftropical medicine and hygiene, 1982, 85: 245-250.

20. Desfontaine M et al. Evaluation des pratiques etdes coOts de lutte antivectorielle a I'echelon familialen Afrique Centrale. II. Ville de Douala (Cameroun),juillet 1988. Annales de la Soci6t6 belge de M6de-cine tropicale, 1990, 70: 137-144.

21. Desfontaine M et al. Evaluation des pratiques etdes coOts de lutte antivectorielle a l'chelon familialen Afrique Centrale. I. Ville de Yaounde (Came-roun). Bulletin de la Societ6 de Pathologie exotique,1989, 82: 558-565.

22. Zandu A et al. Pratiques et depenses pour la pro-tection contre les moustiques dans les menages aKinshasa, Zaire. Annales de la Soci6t6 belge deM6decine tropicale, 1991, 71: 259-266.

23. Louis JP et al. Le paludisme-maladie dans la villede Yaounde (Cameroun). Prise en charge et lutteantivectorielle au niveau familial. Bulletin de laSoci6t6 de Pathologie exotique, 1992, 85: 26-30.

24. Songsore J, McGranahan G. Environment, wealthand health: towards an analysis of intra-urban dif-ferentials within the Greater Accra MetropolitanArea, Ghana. Environment and urbanisation, 1993,5: 10-34.

25. UNICEF. Women and children in Tanzania. A situa-tion analysis. Dar es Salaam, Government of theUnited Republic of Tanzania and UNICEF, 1990.

26. Hackett LW et al. The present use of naturalisticmeasures in the control of malaria. Bulletin of theHealth Organisation of the League of Nations, 1938,7:1016-1064.

27. Ribbands CR. The effects of bush clearance onflighting of West African anophelines. Bulletin ofentomological research, 1946, 37: 33-41.

104 WHO Buletin OMS. Vol 73 1995

Documents

Knowledge of mosquitos in relation to public and domestic control