perfOrman ~ e of health workers after training in

integrated management of childhood illness inGondar, E hiopia

E.A.F. Simoes,S. Gove4

t. Desta,2 T. Tessema,2 T. Gerbresellassie,2 M. Dagnew,3 &

The performance of s x primary health workers was evaluated after following a 9-day training course onintegrated manageme t of childhood illness (IMCI). The participants were selected from three primary healthcentres in the Gonda District, Ethiopia, and the course was focused on assessment, classification, andtreatment of sick chit ren (aged 2 months to 5 years) and on counselling of their mothers. Immediatelyfollowing this training, a 3-week study was conducted in the primary health centres to determine howwell these workers p rformed in assessing, classifying and treating the children and in counselling themothers.

A total of 449 sic children who Rresented at the three primary health centres during the study periodwere evaluated. Most f the complaints (87%) volunteered by the mothers (fever, cough, diarrhoea, and earproblems) were cover d by the IMCI charts. The assessment of commonly seen signs (tachypnoea or earpain) or easily identifi ble signs (slow return after skin pinch, wasting, or pedal oedema) was good, withsensitivities of 67-91° , whereas the assessment of uncommonly seen signs (dry mouth, corneal clouding)or less easily quantifia le signs (eyelid pallor, absence of tears) had a fair or poor sensitivity of 20-45%. Theclassification of pneu onia, diarrhoea with signs of dehydration, and malnutrition showed sensitivities of88%, 76%, and 85% nd specificities of 87%, 98%, and 96%, respectively. However, the classification offebrile illnesses had a sensitivity of only 39% due to problems in using the draft algorithm in areas with amixture of high, low, nd no malaria risk, and due to confusion between axillary and rectal temperaturethresholds. Of 39 chit ren classified as having severe disease, 9 were misclassified, mostly by one nurse.Treatment of patients .mproved over the three weeks of observation, their completeness increasing from69% to 88%.

Health workers u ally communicated appropriate advice to the mother. They learned to use checkingquestions but failed to adequately solve problems in the majority of cases. The mother's counselling card,which summarized rec mmendations on feeding and home fluids, and advice on when to return, was widelyused to aid communic tion. The time taken to perform the complete management of children did not changesignificantly (20 to 19 inutes) during the study. Lessons from our findings have been incorporated into animproved version of th IMCI charts.

IntroductioniThis article describes t e first test of the draft train-

ing materials for inte ated management of child-hood illness (IMCI) y first-level primary health

workers. Previous studies on the IMCI algorithmhave assessed the performance of trained workers touse only the first case management chart, which wasto assess and classify the sick child (1, 2). In GondarDistrict, Ethiopia, the health workers were trained inthe entire process of case management (assessment,classification and treatment of the sick child andcounselling of the mother). Qualitative informationcollected during the course provided feedback forimproving the training materials.

The primary objective of the study was to deter-mine how well the health workers could assess, clas-sify and treat ill children (aged 2-59 months) andcounsel their mothers after receiving training usingthe draft version of the WHO/UNICEF course onIMCI. Our observations included assessing whetherthe mother was adequately taught to deliver key

1 Department of Pediatrics, I fectious Diseases Section, Univer-

sity of Colorado Health Scie ces Center, Denver, CO, USA. Re-quests for reprints should be ent to Dr Eric A.F. Simoes, Sectionof Infectious Diseases, Box B 70, The Children's Hospital, 1056 E19th Avenue, Denver, CO 8 18, USA.2 Department of Pediatrics, ondar College of Medical Sciences,

Gondar, Ethiopia.3 Division of Community H alth, Gondar College of Medical

Sciences, Gondar, Ethiopia.4 Division of Child Health and Development, World Health Organi-

zation, Geneva, Switzerland.

Reprint No.5819

Bulletin of the World Health Orga1ization, 1997, 75 (Supplement 43-53 @ World Health Organization 1997 43

Training in integrated management of childhood illness in Gondar, Ethiopia

mended foods for others unable to read Amharic(the local language .Common, potentially modifi-able feeding proble s were described in the adaptedtraining course; the e included too much reliance onbreast milk (it was ommon not to give complemen-tary foods in the fi st year and to give them infre-quently in the seco d year); complementary foodsthat were too dilute ( diluted milk or muk) or of poorvariety; lack of vita in-A-rich foods for the children;use of bottle-feedin sometimes; sudden weaning ofthe child when the mother becomes pregnant; andfailure to give th infant or young child meat,chicken or fish whe these are eaten by the family.

weeks of follow-up. During this period, the twohealth workers at each centre were exclusively as-signed to seeing all the sick children who came tothat centre. The study was conducted during normalworking hours from 08:30 to 16:30. Patients present-ing in the evening and at night for emergency treat-ment were not evaluated in this study. Threepaediatricians (TD, 1T, TG) who spoke Amharicwere assigned, one to each centre while the paedia-trician (EAFS) who did not speak Amharic assistedin the physical examination and diagnosis of thechildren.

Any sick child who presented at the primaryhealth centre during the study hours was evaluatedby one of the two trained health workers, using thethree case management charts (taped to the walls ofthe health centre) or the chart booklet. They useda draft recording form which included only the as-sessment. For the purposes of the study, the healthworkers were provided with the appropriate medica-tions utilized by this course ( oral sulfamethoxazole+ trimethoprim ( co-trimoxazole ), chloroquine,paracetamol, iron, vitamin A, tetracycline eye oint-ment, oral rehydration solution, and chloramphe-nicol and quinine for intramuscular use).

The paediatricians observed the health workers'performance in assessing, classifying, and treatingthe children and in counselling the mother for everyalternate patient, and filled out detailed forms foreach of these four aspects of management. A four-channel timer was used to measure how long theassessment, classification, treatment and counsellingtook. The assessment of counselling included anestimation of whether the mother was advised ap-propriately (using simple language and appropriateexplanations), whether the nurse asked checkingquestions to make sure the mother understood theadvice, whether the nurse avoided leading questions,and whether the nurse asked about problems andhelped to solve them. After the interaction with thenurse, the paediatrician (assisted by another paedia-trician in some cases) conducted an independent as-sessment of the child using the IMCI classificationscheme to make a clinical diagnosis and an expertassessment of the need for admission. No laboratoryresults were available to the paediatrician. In ap-proximately half the number of children, observationof the nurse's case management was not performedand only the paediatrician's independent assessmentof the child was carried out. Any changes in manage-ment caused by a disagreement in classification andtreatment were administered by the physician. For asample of mothers, an exit interview was carried outby Amharic-speaking nurses trained in interviewingtechniques to determine their understanding and re-call of the health worker's advice.

Training

Two paediatricians a community health clinician,and an experienced acilitator from.WHO were firstgiven 5 days of pr paration in the content of thecourse and in fac litator techniques. They thenserved as course di ctor and facilitators for the sixhealth workers fro the three primary health cen-tres in Gondar Distr ct (Gondar polyclinic, Teda andKoladuba). Althou h the written materials were inEnglish, the course was conducted in Amharic and

English.Training was c mpleted in 9 days using the pre-

test version of the I CI course. This draft version ofthe training course i cluded modules on assessmentand classification o a sick child, how to identifytreatment, treat the child, counsel the mother, andfollow-up; no mater als were available on the younginfant. After the firs day of the course, each morningwas spent in an outp tient clinic and on the inpatientwards. In addition t the module, training materialsincluded three case anagement wall charts (Assessand classify the sic child aged 2 months up to 5years; Treat the c 'Id; and Counsel the mother),booklets of the wall charts, recording forms for theassessment of the si k child, and a draft video andphoto booklet. Th facilitator techniques for themodules (including instruction in individual feed-back on written exe cises, role plays, group discus-sions, and drills) an clinical practice were describedin draft facilitator g ides.

Clinical trainin was carried out in the out-patient department of Gondar hospital and theGondar town polyc inic. Severely ill children wereassessed and classifi d in the inpatient services of thePaediatric Departm nt of Gondar College of Medi-cal Sciences.

Observational stu t y Following the 9 day of training, the health workers

returned to their rimary health centres for 3

WHO Bulletin OMS Val 75, S~ppl. 1997 45

Training in integrated management of childhood illness in Gondar, Ethiopia

Table 2: RecognitiO~~Cr~ signs by the health worker, compared with that by the paediatrician

Sensitivity (%) Specificity (%)Main symptom No. identified

Cough or difficult breathing(n = 254)"

Diarrhoea (n = 217)

111

18

21

10

9

8

4

17

41

41

6

55

57

4

4

16

TachypnoeaChest indrawing

General condition

Dry eyesAbsence of tearsDry mouth

Drinking poorlySkin pinch slow

Ear painEar discharge

Visible severe wasting

Conjunctival pallorPalmar pallorCorneal clouding

Foamy patches (Bitot's spots)Bipedal oederzta

Ear problem (n = 52: 8366

674537255069

Malnutrition and anaemia(n = 449)

, Figures in parentheses are t~~-n-umbers assessed for each main symptom

ers, but in 3 of them the clouding had occurred manyyears previously. Bitot's spots (foamy patches on theeyes) were identified 50% of the time. Bipedaloedema was reliably detected.

Classification of illness

Table 3 shows a summary of the classification ofchildren who presented at the three health centres.While 254 children were assessed with cough or dif -

ficult breathing, 112 were classified with pneumoniaor severe pneumonia. Compared with the paediatri-cians, the health worker correctly identified pneumo-nia or severe pneumonia with a sensitivity of 88%and a specificity of 87%. In the 217 children withdiarrhoea, dehydration and dysentery were correctlyclassified in the majority of children, but a history ofpersistent diarrhoea was often missed. The classifica-tion of the febrile child was problematic, with thecorrect classification occurring in only 39%. Acuteear infections characterized by ear pain were cor-rectly diagnosed more frequently than chronic earinfection characterized by pus in the ear. Overall,the recognition of some or severe malnutrition wasexcellent.

assessed in the study. ealth workers obtained ahistory of ear pain almos as frequently as physicians,but ear Iftischarge was issed in almost a third of

patients.The recognition of f ver and the classification of

malaria had a poor s nsitivity (39% ), comparedto the paediatrician's c assification of fever. Mosterrors occurred for seve al reasons. Health workersused the rectal temper ture cut-offs on the chart(38 °C) while taking an a illary temperature (withoutconverting to the 37.5 C cut-off), resulting in asignificant loss in sensiti ity compared to the paedia-trician's classification of ever. The malaria classifica-tion table for high and low malaria risk areas, towhich were added inst uctions for areas with nomalaria risk, were not c rrectly used. The presenceof a significant number of children with causes offever that were not incl ded on the charts, includinghepatitis, abscesses, and gum infections, also caused

problems.All children were a sessed for malnutrition and

anaemia. The nutrition I status of many children inGondar is poor, kwashio kor is common, and 38% ofchildren have a weight- or-age Z-score of less than-2. The weight-for-age chart was used for everychild to classify "some alnutrition". Two out of sixchildren with visible s vere wasting were missedby the health workers. n only two instances, whenthe weight fell on the orderline area close to theline, were patients mis assified. Palmar pallor wasmissed in two-thirds of he patients, mainly becausethe health workers com ared the palms of the childand the mother. Often he mother also had pallor,invalidating the compa son. All four children withcorneal clouding were i entified by the health work-

Classification of severe disease

Comparison of health worker's use of the IMCI algo.rithm with paediatrician's use of the IMCI algorithm(Table 4). There were 38 children with 41 classifica.tions of severe disease who should have been re.ferred, based on the IMCI algorithm. The healthworkers misclassified 14 as non.severe, of which 11

47WHO Bulletin OMS Vo175, Suppll1, 1997

Training in integrated management of childhood illness in Gondar, Ethiopia

Table 5: Quality of treatm~nt given by the healthworker, as observed by thelpaediatrician

Table 6: Performance of trained health workers in

giving antimicrobial treatment

Week Week 2 Week 3

88%12%0%

Watchedmother give

the first doseDose shown to

the motherCompleteIncompleteIncorrect

69%28%

3%

77%20%

3% 89%92%50%

48%64%17%

Pneumonia

DysenteryMalaria

Correcttreatmentselected

positive predictive value 7 % ). Of the five patientsnot referred using the al rithm, three had bron-chiolitis (two with audible wheezing) and two hadpneumonia. All had fast reathing but none hadchest indrawing. Of the 11 atients who would havebeen referred by the I CI algorithm 'but werejudged by tfue paediatricia not to need admission, 7had pneumonia, one had croup, two had visiblesevere wasting and one ha febrile seizures.

Comparison of health work r's use of IMCI algorithmwith paediatrician's expe estimation of need forreferral. Using the IMCI al orithm the health work-ers would have referred 1 patients correctly, notreferred 11 patients who n eded admission, and re-ferred 10 patients who di not require admission(sensitivity 66%, specificit 98%, positive predictivevalue 68% ). Of the 11 patie ts not referred using thealgorithm, 5 were the sam ones missed by the doc-tor's use of the algorithm, had chest indrawing and3 had pedal oedema. Of t e 10 patients who wouldhave been over-referred, were also over-referredby doctors using the algorit m (five with pneumoniaand two with visible seve e wasting); additionally,chest indrawing (in two pe sons) and pedal oedema(in one) were identified b the health worker usingthe IMCI algorithm, but these were not recom-mended for referral by the paediatrician.

the first two weeks, 3% of children were treatedincorrectly; in the third week no one was treatedincorrectly.

The antimicrobial treatment of pneumonia,dysentery and malaria was studied closely (Table6): whether the correct dose was calculated, andwhether the health worker demonstrated how to ad-minister it and actually watched the mother adminis-tering the first dose. Treatment and dosage followedthe IMCI guidelines in 95-98% of children. The cor-rect dose was demonstrated to the mother, but thehealth worker's witnessing the administration ofthe first dose was more common for pneumonia anddysentery than for malaria. Of the 449 children, 178required antimicrobial therapy, 98 for pneumonia,56 for dysentery, 20 for malaria, and 4 for acute earinfection (Table 7). Oral antimicrobials were notadministered in 12 of these children on account ofsevere disease and referral to hospital (6 children),or they went home with pneumonia (2), dysentery(2), malaria (1) and ear infection (1) without appro-priate antimicrobial treatment (Table 7).

The recommended treatment of diarrhoea in-cluded the advice to increase fluids during the illness,to reduce milk or give yogurt, and to recommendhome ORS (oral rehydration salts) therapy for chil-dren with some dehydration. Most of the children(94% ) with some or no dehydration were advised toincrease fluids during their illness and 82% of chil-dren with diarrhoea were recommended home useof ORS, while 73% with persistent diarrhoea wereadvised to reduce milk or give yogurt. Every childwith some or severe dehydration received oralrehydration therapy.

Treatment of children

~During the study period, t e completeness of treat-ment of a patient improved steadily from 69% in thefirst week to 88% in the hird week (Table 5). In

Table 7: Distribution of pati nts requiring but not given antimicrobial treatment by the healt~~rs

No. sent home withoutantimicrobial treatment

No. with severe diseasereferred to hospitalNo. requiring jreatment No. not given treatment

22

Pneumonia

DysenteryMalariaEar infection

WHO Bulletin OMS. Vol 75, Suppl. 1, 1 ~97

Training in integrated management of childhood illness in Gondar, Ethiopia

Discussion

Overall, the health wor ers performed well in theclinic after IMCI trainin Their use of the case man-agement guidelines resul ed in the correct classifica-tion of children and ap ropriate antimicrobial andoral rehydration therapy when compared to the as-sessment by the paediat cian. More problems wereencountered in correctly identifying children for re-ferral. In'this study, 87 '10 of the presenting com-plaints volunteered by th mothers were covered bythe charts, which were d signed to include the ma-jority of childhood con itions seen in developingcountries; to keep a bala ce between simplicity andcompleteness entailed le ving out some conditions.Our test of the draft tr ining materials identifiedspecific problems. Many hanges were tberefore in-corporated into the co rse materials which weresubsequently field-tested in the United Republic ofTanzania (3).

The classification o illness performed betterthan the detection of individual clinical signs(Table 4). Thus, the clas ification of pneumonia orsevere pneumonia, dehy ration, dysentery, and mal-nutrition was good (sensi ivities in the range 76-94%and specificities 87-98% .Dependence on a historyto make a classification e.g. duration of diarrhoeafor persistent diarrhoea, duration of ear drainagefor ear infection) or use f a complex chart (fever)led to poorer classificatio of the illness (sensitivitiesof 39-65% ). Modificati ns in the algorithm weremade in order to redu e these errors in diseaseclassification.

Signs of dehydration often changed in the timebetween exmination by t e health worker and by thephysician, which explain differences in the assess-ment of the signs and p rtly accounts for the lowsensitivity (20-25% ). Th a child who was quiet butrousable initially may ha e become irritable at theend of the physician's xamination. A child withabsence of tears and a dr mouth may, after drinkingfluids, have developed ars and a moist mouth.Signs that changed less q ickly and were more easilyquantifiable such as dry eyes and skin pinch weremore reliably identified sensitivity, 71-90% ). Thegeneral condition of the child is a very importantsign, and many misclassi ations were due to misin-terpretation of an irritab e child or a drowsy child.Subsequent versions of t e algorithm excluded drymouth and absent tears, nd the training materialshaye been strengthened t improve the clinical rec-ognition of the child's ge eral condition.

Some of the reasons or misclassification of thefebrile child have been emedied by changing theaxillary temperature thre holds on the IMCI charts,by improving the recogni .on of malaria risk, and by

simplifying the malaria classification tables. The cur-rent algorithm still relies on reminders to assess andtreat other causes of fever which are not includedin the algorithm ( e.g. periodontal disease, abscesses,hepatitis, etc;). These will continue to result in over-treatment for malaria. More research is urgentlyrequired for the accurate detection of malaria byclinical means and by simple laboratory tests such as

dipsticks.Health workers readily identified severe mal-

nutrition using visible severe wasting and bipedaloedema, however there was a greater lack of appre-ciation for signs of xerophthalmia. Thus, foamypatches in the eyes (Bitot's spots) were only detected50% of the time. Of the four children with cornealclouding, one child with a new onset was correctlyclassified as severe malnutrition. The clouding wasidentified in the other three, but no importancewas ascribed to this finding since they had hadscarring for a number of years. This was, in fact, theright decision. Subsequent versions of the courseteach that only recent corneal clouding is a sign forreferral.

Conjunctival pallor was poorly detected, partlybecause the children in Gondar often had ablepharo-conjunctivitis that obscured the pallor.Palmar pallor was defined by comparing the child'spalms with the mother's in order to account for pig-mentation. This may have accounted for the poorsensitivity and specificity of this sign in our study,since both mothers and children were often pale. Itmay be more appropriate to compare the child'spalms with those of a normal health worker. Specifictraining materials to support recognition of someand severe palmar pallor (such as photos and videoexamples) were not available for our test and havesubsequently been added to the course.

The complaint of ear pain was accurately as-sessed, but discharge from the ear was not detectedin a third of the patients. Health workers in Gondardid not pay much attention to this common problemwhich even the parents often ignored.

Children with severe disease who are mis-classified could potentially suffer a serious outcome.A significant number of severe classifications weremissed, but some of these children were referredbecause of another condition. In our study most ofthe missed referrals were due to inaccurate observa-tion of chest indrawing by one health worker. Em-phasis needs to be placed on the accurate recognitionof this sign based on sufficient clinical practice duringtraining. Since this is one of the first signs taught inthe course, the longer duration of the course canprovide more practice on this sign.

Local paediatricians, who are familiar with thelimited hospital inpatient capacity, would not have

WHO Bulletin OMS Vol 75 Jppl 1997

E.A.F. Simoes et al.

'I referred some of the children with chest indrawing

and thus disagreed with the conclusion based on thealgorithm. More emphasis needs to be placed onchest indrawing during training and on further re-search to explore ways to safely limit the referral ofchildren with chest indrawing who have no othersigns of severity.

Overall, correct treatment was delivered to ahigh proportion of children, rising from 69% in thefirst week to 88% in the third week (Table 5) as thehealth workers grew more accustomed to the charts.Antimicrobial treatment was especially good, with95-98% of children receiving correct treatment(Table 6). No child in this study received unneces-sary antibiotics, one major advantage of the IMCIclassification being the clear identification ofchildren who do not require antibiotics. Priorsensitization from the widespread use of oralrehydration therapy in Gondar District through allthree health centres probably facilitated correcttreatment of diarrhoeal disease and dysentery.

After training using the draft course materials,health workers were more likely to give appropriateadvice on treatment and feeding than on when toreturn to the clinic. They usually did not try to solveproblems, but in almost two-thirds of cases they usedchecking questions to determine whether the motherunderstood the advice given. Training in communi-ation, including more clinical practice, has been

.mproved in subsequent versions of the course.The time taken with each child improved slowly

rom about 20min at the start of the study to about8min at the end of the third week. A follow-upnterview with the six participants, a year later (July995), revealed that it took about 3 months of con-tant working with the charts to become thoroughlyamiliar with them. At this time they reported thathey could complete the management of one child in-10min.

Twenty minutes may seem to be a long time topend on one child in some circumstances, but in ourtudy there were an average of 10 sick children for

.nitial visits who were seen each day (450 patients

.the three centres over 15 days); other childrenho came to the health centre for immunizationr trauma did not need this case management pro-ess applied to them. Thus, effectively each healthorker saw 5-6 new cases of sick children, using the

MCI approach, from 08:30 till 13:30 without signifi-ant delays or longer clinic hours. Our youngestealth worker showed steady improvement in assess-ent over the three weeks (20min to 15min), but

s e took longer to counsel the mother. Good man-gement of sick children requires effective counsel-

1 ng of the mother, for which sufficient time must beYen.

It is clear that health workers learned and im-proved during the 3 weeks of the study, and reducedtheir dependence on the recording form. From thisexperience we estimate that, after such training, itwould take an average health worker several monthsto incorporate thoroughly and efficiently the IMCIprocess in their daily practice. Our results demon-strate that this training course can be effective inpreparing health workers from first-Ievel healthfacilities to take good care of sick children under 5years of age in developing countries in an integratedfashion.

AcknowledgementsThe authors are grateful to the following participants in thestudy: Gebrehwot, Zerajion, Tsehay, Bosena, Girmai andAsmara. We thank Mary Lung'Aho for assistance withnutritional counselling; Karen Mason for the statisticalanalysis; Fitaw and Hebretu for conducting the exit inter-views; and Georgina Armendariz and Chandrika John forassistance in preparation of the manuscript. The studywas supported by the World Health Organization (Divisionof Child Health and Development and the Special Pro-gramme for Research and Training in Tropical Diseases),Geneva, Switzerland.

Resume

Evaluation des prestations des agents

de sante, apres formation a la prise en

charge integree des maladies de I'enfanta Gondar (Ethiopie)

La qualification de six agents de soins de santeprimaires a ete evaluee apres un cours de formationde 9 jours a la prise en charge integree des mala-dies de I'enfant (IMCI). Les participants ont eteselectionnes dans les trois centres de soins desante primaires du district de Gondar (Ethiopie); lecours a porte essentiellement sur I'evaluation, laclassification et le traitement d'enfants malades (de2 mois a 5 ans), ainsi que sur le conseil maternel.Des la fin du cours, une etude de 3 semainesa ete realisee dans les centres de soins desante primaires, pour determiner la qualite desprestations fournies par ces agents, en matiered'evaluation, de classification et de traitement desenfants, et de conseil maternel.

Un total de 449 enfants malades qui se sontpresentes aux centres pendant la periode d'etudeont ete evalues. La plupart des symptomes (87%)rapportes spontanement par les meres, a savoirfievre, toux, diarrhee et affections de I'oreille,etaient inclus dans les tableaux de I'IMCI.

WHO Bulletin OM: 75, Suppl

Training in integrated management of childhood illness in Gondar, Ethiopia

L'evaluation de sig es couramment observes(tachypnee ou otalgi ), ou facilement identifiables(signe du pli, emaci tion, cedeme pretibial) etaitbonne, avec une sen ibilite de 67-91%, alors queI'evaluation des sign s peu courants (secheressede bouche, opacites corneennes) ou de signesmoins facilement quantifiables (paleur despaupieres, absence d larmes) avait une sensibilitefaible a moyenne, de 20-45%. La sensibilite de laclassification des pn umopathies, de la diarrheeavec signes de desh ratation et de la malnutritionetait respectivement d 88%, 76% et 85%, pour desspecificites de 87%, 8% et 96% respectivement.La sensibilite de la lassification des pathologiesfebriles n'etait toutefo s que de 39%, en raison desdifficultes d'utilisation du projet d'algorithme dansles secteurs ou le risq e d'impaludation est variable(eleve, faible ou abs nt), et de la confusion entreles seuils de temper ure axillaire et rectale. On adenombre 9 erreurs d classification sur 39 enfantsranges dans la categ rie «maladie grave", presquetoutes dues a la mem infirmiere. Le traitement despatients a ete ameli re au cours de la perioded'observation, la co pletude passant de 69% a88%.

La carte de conseil maternel, qui resumait lesrecommandations concernant I'alimentation etles liquides administres au domicile, ainsi que lesconseils sur la necessite de ramener I'enfant, alargement ete utilisee pour faciliter la communi-cation. Le temps necessaire a la prise en chargecomplete de I'enfant n'a pas ete significativementmodifie pendant I'etude (20/19 minutes). Lesenseignements tires de nos resultats ont servia produire une version amelioree des tableaux deI'IMCI.

References

1 .Weber MW et al. Evaluation of an algorithm for theintegrated management of childhood illness in an areawith seasonal malaria in the Gambia. Bulletin of theWorld Health Organization, 1997, 75 (Suppl. 1): 113-118.

2. Zucker JR et al. Clinical signs for the recognition ofchildren with moderate or severe anaemia in westernKenya. Bulletin of the World Health Organization, 1997.75 (Suppl. 1 ): 97-102.

3. WHO Division for Child Health and Development &WHO Regional Office for Africa. Integrated manage-ment of childhood illness: field test of the WHO/UNICEF training course in Arusha, United Republic ofTanzania. Bulletin of the World Health Organization.1997. 75 (Suppl. 1 ): 55-64.

Les agents de sf te ont en general donne de

bons conseils a la m' re. Ils ont appris a utiliser les

questions de control, mais n'ont pas reussi aresoudre les problem s dans la majorite des cas.

!ulletin OMS. Val 75, Suppl 1997 53