Applied modelling and its utilization in analysis and planning in the OCP: report of a meeting held in Ouagadougou,19 - 21 January 1987

-a WORLD HEALTH ORGAI{IZATION ORGAT'IISATION MONDIALE DE I..A SANTE ONCHOCERCIASIS CONTROL PROGRAII,IIIIIE IN wEsT AFRICA PROGRAMI'IE DE LIITTE CONTRE LIONCHOCERCOSE EN AFRIQUE DE LTOUEST EXPERT ADVISORY COMMITTEE Eighth session Ouagadougou, 16-19 June 1987 INFORMATION DOCTJMENT APPLIED MODELLINC AND ITS UTILIZATION IN ANALYSIS AND PLANNING IN THE OCP Repor of a meet held in 19-2t J 1987 Page I OUTLINB OF TEE REPOR:I The applled modelllng meetlng, whlch ls che subJecE of the present rePort 'focussed on quesElons of long t,erm onchocerclasls conErol and t,he uElllzatlon of a transmlsslon model ln the plannlng of long cerm conErol acclvlcles. In addltlon co addresslng che technlcal aspecc6 of a transmlsslon model for onchocerclasls, the oeeElng provtded an opport,unlEy for the Eechnlcal staff of the OCP Co llst and dlscuss alt,ernaElve future sEracegies and survelllance mechanlsms, Eo revlew recent, resulEs of relaEed research, and Eo draw up plans and prtorltles for further fleld research, lncludlng experlmental lnterruptlons of vector concrol ln the cenEral OCP area. Though all chese polnEs were dlscussed ln an lntegraEed way, lE was declded to glve decalled reporE6 on t.he dlfferent lEems ln separace appendlces to che overall report, ln order co allow their lndependent utlllzat,lon wherever approprlate. @I{TEIITS: - l. Introductlon 2. Sunmary of recent, work on applled modelllng ln the OCP 3. Posslble sErat,eglee for long t,erm control. 4. Recent, research and analysls relaElng t,o transmisston and cont,rol 5. Full transnlsslon nodel 6. Plans for applled research 7. Concluslons APPENDICES: I. Alt,ernaclve st,racegles for long Eerm cont,rol and lndlces for survell1ance. II. Plans and prlorlcles for applled research. III. Vector-ParaslEe model. IV. SpeclflcaElons for a t,ransmtsslon model. V. Llst of parc, lclpants. VI. Background documencs Page 2 I.IITTRODUCTION During che prevtous modelllng meetlng ln Ouagadougou in March 19g5, it was agreed that applled modelllng, whlch had recently been lncroduced ln the oCp, would be an lmportanB and pracElcal cool for data analysls, evaluat,lon andlong term p1annlng. Thls prevlous meet,lng propoeed a pf.n'for furcher nodeldevelopment, conslsElng of t,hree phases: the development of a host-paraslEe nodel, Ehe developmenc of a vect,or-paraslte oodel and, once chese Ewo submodels had been sufftclent,ly cesEed, Ehelr comblnatlon in60 a fulltransmlsslon oodel. Thls proposal for a phased proJect was subnlcted co the EAC whlch strongly supporced 1c and underllned Ehe pracE,lcal import,ance of theoodelllng project for che oCP. Slnce then nodel development has proceeded according E,o Plan. A host-paraslEe model has been deveioped successfully andhas become a basic tool for the tnEerprecaElon of the epliemlologlcal daEa andthe predicElon of epldemlologlcal Erends ln t,he cencral OCp area. The secondphase is underway and a ftrsE vecEor-paraslte model has been developed whlchls presenely being cested ln Ouagadougou ln the analysls of che encomologtcaldata. Durlng 1986 lt was reallzed thaE a full transmlsslon model would berequlred much sooner than lnltlally foreseen and that lt, was necessary Eo tryto speed up the thlrd and flnal phase of the proJeet. The three mosElnportant, reaaons for thls are t,he followlng: - All the lateet, lnformatlon 1nd1cate6 that che reservolr of lnfectlon lsdytng out, very fast ln che cencral oCP area and ir ls predlcted thatthere w111 be vast areas where Ehe lnfecElon will be vlrcually ellmlnaced aft,er 2-3 years from now. Declslons on reducing ordlsconElnulng vecEor cont,rol operatlons ln Ehls area wlll therefore haveto be taken ln the near fuEure. - The programme ls under severe flnanclal consEralnEs whlch wlII resulc lnlncreased pressure Eo reduce vector control operatlons in the central area. - IvermecEin will be avallable for uEllizaclon on a mass scale from early1988 onward when tc will provlde for the flrst cime an alternatlve method for disease conErol and posslbty also for transmlsslon conErol. It ls generally accept,ed t.haE che presenE vecEor control operaElons ln the cenEral OCP area w111 have Eo cease once Ehe reservolr of infecElon will havevtrcually died out, and that it wllt have to be followed up by a sysgem of survelllance and long t,erm conErol (probably by uslng tvermecElo) LnderdevoluE,lon. The detalled plannlng of thls change of conErol and the deslgn ofthe follo$r-uP acEivlcles wlll have Eo be based on realisclc est,imates of Ehelong term lmpact, coscs and rlsks of uhe possible alE,ernative approaches. Such est,lmat.es are presenEly not yec available and tn order co provide t,hem it w111 be necessary to lncenslfy some operaclonally relevanE fleld research andto develop a valld, and sufflclenEly speclfic, t.ransmission moder. Because of the above, Ehe meeclng declded t,o concenEraEe on quanEltac,ivequesElons relat.lng t.o onchocerelasls survelllance and eonErol on ghe long t,ermln the cencral oCP area, and on the development of a realistlc Eransmlsslon nodel whlch would be approprlaEe for che study of Ehe speclflc questlonsfaclng the OCP ln rhls respecc. Page 3 2. SI'I,II.IARY OP SBCENT WORK ON APPLIED I{ODELLING IN TEE OCP The cransmlsslon cyc le of O.volvulus ls glven ln the dtagram below. It showe whlch part6 of the t,ransnLsslon cycle have been descrlbed ln che models developed sofar, and whlch element.s sctll need co be addressed. VECTOR HOST-VECTON HAST CONTAC LSHeod L1 L3\ o L4 \ mmoture Hoct-Porocito ModclI BodyL3 r t L2 wormMotu inMF re It MF /ln sk Fin eeyI + Vcctor-Pororitc Madcl 0culor Pothotogy The maln elements in the transmission cycle whlch stlll need to be modelled, are(e) tne relaElon beE,ween the uptake of mlcrofllarlae ln the fly and the dlscrlbuclon of skln microfllarlae ln che human populaElon.(S) fhe relatlon beEween Ehe tnocculatlon race of L3 larvae and enEomologlcal lndices (sueh as the ATP) on lnfectlve larvae ln che biElng fly populaElon.(C) ftre probablllcy of lnoculated L3 larvae Eo develop lnt,o mature hrorms. FurEhermore, Ehe lncldence of ocular pachology has noE yet been lneluded ln the host,-paraslc,e model. Before conslderlng t,hese necessary exEenslons and chelr lncluslon inEo aful1 t,ransmlsslon model, Ehe experlence wlEh che Host-Paraslt,e model was shorEly summar Lzed. Furt,hermore a subgroup meeElng (whtch was attended by allparclclpants) dlscussed Ehe flrsc verslon of the vector-parastEe model. 2.1. Eogt-Paraalte rcdel. Thls model has now been used successfully for more Ehan one and a half years as a rouclne cool ln che analysls and lnEerpretatlon of che epldemlologlcal dat,a. The predleted epldemlologlcal Erends st{11 flt very well co chose observed ln che cenEral OCP area. Lc seems cherefore t,hac thls rnodel adequately descrlbes the parasltologlcal facEors ln the human populatlon and thaE Ehls submodel can be lncorporaced tn Ehe cransntsslon nodel for a vaIld descrtpclon of chls part of t,he Eransmlsslon cycle. However, Ehere uere two paramecers for whlch che lntclal estlmates had co be Page 4 changed constderably Eo make che predlcced results flc the observed epldemlologlcal daEa. These were: - The varlabllltv ln woru survival ls less Ehen was lnlctally esElmaEed.Even afEer correcElon for thts, the prevalenee Eendsblt faster than predlcted. to fall sclll a - The varlabllltv ln lnt,ensltv ln lnfecclon bet,ween lndlvlduals ls fargreat,er Ehan can be explalned by random factors alone. The explanaElon should probably be sought ln lndlvldual tmnunologlcal dlfferences because lt ls choughc unllkely that behavloural dlfferences can explaln the ext,ent of chls varlablIlcy. These lndlvldual dlfferences may be of major lmport,ance for quesElons of long Eern control. AE prevlous modelllng meeclngs some ext,enslons Eo Ehe hosE-paraslE,e model have been dlscussed, such as the simulatlon of male worms, mat,ing of worms, lncldence of eye lestons and blindness. However, Lt was concluded Ehat not all of t,hese extenslons qrere urgenE and that the mosE inportant excensions(such as bllndness) should be directly incorporaEed ln the full transmLsslon model. 2.2. Vector-Paraslte oodel Durlng a meeElng ln RoEterdam ln December 1985 che baslc contents of a vecEor-ParaslEe subrmodel were escabllshed. A flrst test verslon of a coltrpuEer slmulatlon program of Ehis model has been developed durlng 1986, andhas recencly been EesEed by che OCP. The result,s of chese Eescs were dlscusseddurlng a subgroup meet, lng and chelr cornmenEa and recoornendat, lons are given 1n appendlx II. However, t,wo major commenEa may be ment, loned here: - The presenc model si.mulat,es only a local fly populaElon and does no6 yeE, allow for lmmigraElon of (lnfecrlve) slmulltds whlch ltmlrs irs utlllzatlon ln Ehe analysis of Ehe entomologlcal dat,a ln Ehe area under vecEor conErol. A major lmprovemenc would be the slmulaElon of both alocal and a dlscanc populaclon as well as a mlgraElon process from Ehedlscant co che local breeding slEe, and Ehls exEenslon ts planned for the next verslon of Ehe model. It ls recognlzed EhaE chis ls sclll a slmpllfled approach co the quesclon of Eransmlsslon ln relatlon Eo Ehedispersal and mlgratlon of onchocerclasls vectors, buE so lltcle is known about thls phenomenon Ehac lc would be unreallstlc Eo aEtemp6 a nore detalled quantltaElve descrlpclon. It was reeognized that this quesElon of vecEor mlgraclon may remaln one of the weaker polnEs ln any attempt of modelllng onchocercLasls t,ransmLsslon. - The vecEor-paraslEe model was lntcially EesEed against enEomological evaluatlon daEa from Ehe non-conErolled area ln Ehe souEhern extenslon. There was e remarkable dlfference beEween the number of Ll and L2 sEage larvae predlcted by the model and che number accually foGaErrnST[!_dlssectlons. The predlcted number was 2 ro 3 tlmes hlgher and'li ts assumed EhaE the dlfference ts due Eo serlous undercounE,lng of Ll and L2larvae ln rourlne dissectlons. Thls hypochests ls supporced by comparlsons of rout,lne dlssecclon daca from che OCP wlch research data from che same caEchlng poincs ln Togo, wtth daca from Ehe Sudan, and IiliEh dat,a from recenE OCP scudles ln whlch the flles were preserved and 6talned before dlssecElon. Thls systemaElc undercountlng of Ll. and L2 larvae greacly llmlcs the use of che hlstorlcal entomologlcal data for a deCalled analysls of the age sEructure and migraElon stacus of the vecEor populatlon and the conErlbutlon of relnvaslon Eo Cransmtsslon. 3 POSSIBLE STRATBGTES FOR IONG TER}I CONTROL An lnvent,ory was made of possible sEraceSles whlch should be constdered as alEernatlves 1n plannlng and declslon maklng for the long Eerm control of onchocerclasls. The meetlng addressed also t,he quest,ton of survelllance andl1sted lndlees for monltorlng the encomo-epldemlologtcal slEuaElon. The Page 5 dlscusslon was sErlcEly llmlted Eo possible scrategles whlch are to follow a prolonged perlod of successful vecEor conErol, wlth the vlrtual ellmtnatlon of the paraslce reservotr, as wlll be the case for Ehe cencral OCP area aft,er a fen more years. A detalled descrlpclon of c,he varlous optlons ls glven ln appendlx III. The mosE llkely long t,erm conE,rol strategy to be followed ln the cent,ral OCP area ls a sEracegy wlth heavlly reducedr or complecely lnterrupted, veccor conErol, survelllance under devoluclon, and creacment wtth lvermecgln ln case recrudescence problems are detected. However, beyond thls broad ouEllne, no speclflc plans exlet as yet, and many essent,lal questlons st,lll need Eo be answered, such as: InLq!!upclon of larvlcldlng. When , where and how f ast. [.IhaE are E,he rlsks lnvolved. The baslc quesElon here concerns t,he pot,entlal of the remalnlng reservolr of lnfeeElon Eo contrlbute Eo Eransmlsslon, and thls needs t,o be studled uslng xenodlagnosls and experloenEal lnt,errupglons of larvlcldlng. Furchermore, t,he cosc-effecttveness of contlnulngllmlted larvlctdlng operaElons along a few breedlng slces of maJor lmporEance needs Eo be lnvesElgated. - Survelllance. EnEomologlcal survelllance rnay provlde the earliesE warnlng of recrudescence of Eransmisslon, but only if lt, alms at che aasessment of transmlsslon potenttals. Thls w111 requlre hlgh qualltydlssectlons of ext,remely large nunbers of flles, whlch makes lE less approprlate for devoluElon. Epldemlologlcal evaluatlone uelng ektn snlps would be simpler and more speclfle, b,ut would detect posslble recrudescence of EransmlssLon conslderably lat,er. It ls not yec clear how serlous such a delay would be. A better understandlng of recrudescence dynamlcs ls required t,o answer quesElons abouE what survelllance ls required, where and how ofcen, and to quantlfy Ehe rlskslnvolved ln lncompleEe surveillance. - Ivermect,ln. lt Is doubted lf lvermectln has serious poEentlal for transmlsslon control and thls quesElon needs Eo be studled urgencly ln conmunlEy Erlals. If posslble, some trlals should also be underEaken in a recrudescence sltuaclon nhere the dynamlcs of transmlsson wl11 bequlEe dlfferenc. A major quesclon for recrudescence control concerns Ehe method of dlagnosls and drug admlnlsEratlon, and Ehe quanEltaglve lmporcance of delayed or lneomplete EreatmenE. Thls polnt relages closely t,o Ehe quest,lons ralsed above under survelllance. 4.RECETfr RESEARCE AXD ANA.LYSIS REIATTNG TO TBANSI{TSSTOII AIID COI{TROL The meetlng dlscussed Ehe resulEs of recenE research and analysls concernlng Ehe misslng llnks (A),(B) or (C) ln rhe Eransmlsslon cycle shown lnthe dlagram, as well as the ocular pathology conponent. 4.[. t'llcrof llarlal uptake !a the yg..1qg. prelimlnary xenodlagnosElcexperlments' uslng wtld caughE flles on volunEeers wlch low lntenslcles or 1nrec"iimi:'iiT::"1.;r%:::.;T:i:l?r"lil,x'o*ffr.i,,an MF,oad at the tllac cresE and average upEake of MF. - In che range of l-20 mfs, les MF.(see also Appendtx tI, polnc l) s then L07 of. che fltes engorged Varlaclons ln hosc-vector Sg!jg,1!. An exaurple of slgnlf lcanu varlaElonln conEact rat,ea between man and vecEor, and Ehe resulElng varlaElon lnlnfectlon rates, ls found along che dam at Yabo ln north Burkina Faso. 4.2. UnlnEerrupced breedlng durlng t986 resulted ln an ABR of 20rO0O but only Page 6 ln an ATP of 100. Transmlsslon ls probably noc, lnEerrupted ln thls area but ls concent,raced ln a few flsherman famllles. The average prevalence of mfs ls only 5Z Ln Ehe surroundlng vlllages but 652 ln the flshermam aet,ElemenEs. Thls varlablllcy ln conEacE, rates because of behavloural dlfferences nlghr be an imporEant faccor durlng the devoluElon pertod. 4.3. Ivermeccln and larval developmenE ln the vector. There exlsEs d lsagreemenc about, the pocenElal of mlcrofllarlae, engorged from pat,lenEs treat.ed wlth lvermectln, to develop lnto lnfect,lve larvae ln che fly. FurEher studles wlEh WesE-Afrlcan vector specles are requlred E,o clarlfy thls polnt,. 4.4. Ivermeccln and larval development, 1!g g. At Ehe momenE lE 16 noE clear wheEher lvermeccln affecEs che developtrent, of lnoculaced L3 larvae and L4 larvae tnEo tmmaEure worms. Ic 18 presenEly assumed that,, even lf chls mlghc be the case, uhe effect w111 only last, for a negllglble perlod. 4.5. Immunologlcal defense agalnsE infecEton. Longlcudlnal epldemlologlcal lcal surveys have been done ln a hyper endemlc vlllage ln anand lmmunolog area nlEh ongolng transmlsslon ln Mall. In thls vlllage che prevalence of nfs ls sttll low ln the underflves, but rlses steeply to close co L00Z ln che age group of 5-9 years. The lntenslcy of lnfectlon reaches a peak around 15 years of age and remalns constanE from t,hen onward. However, when for t,he same populatlon t,he level of onchocercal spectftc anc,lbodles ls ploEEed agatnst, age, lt, shows a sharp lncrease co the age of about l0 years, but,lE decreases again after thls age and falls to a much lower level 1n the older age groups. From t,hese resulEs, Ehe hypoEhesls can be derlved thac the lmmunodefenses agalnst, onchocerclaels lnfectlon bulld up in che younger age Sroups as a resulc of che challenge provlded by lnoculared L3 larvae and some flrsc, actlve, lnfecElons. However, when chls challenge becomes too hlgh, Ehere w111 occur a breakdown ln the lnmune defenses, and subsequent lnoculaced L3 larvae wtll have a hlgher probablllcy of developmenE lnEo maEure lrorms. Thls hypoEhesls was of greaE lnEerest Eo the blomathemaclclans present, because of 1ts quant,lcaEive lmporcance for t,ransmission dynamtcs. If Ehe hypoEhesis ls correcE, lc is expecEed EhaE ln areas with a lower endemlclty, t,he moment, of breakdown would occur ag a much laEer age Ehan the l0 years of age observed ln rhe hyper endemlc vlllage. Howeverr lf Ehe shape of the immunoresponse curve ls only a result of behavloral facE,ors, then lE ls expec.Eed chac che age range at whtch che breakdown occurs ls lndependenE of Ehe endemlclcy level. IE was Eherefore suggesEed Eo repeaE Ehe study ln vlllages wlch lower endemlclty lavels. 4.6 Relaclon bet.ween entomo ical and epidemlologlcal data ln the Southern exEenslon. Some resulEs were shown of an analysls of Ehe relaEionship between epldemtologlcal lndlces, llke the prevalence of mfs and the CMFL, and varlous ent.omological lndlces for closely relat,ed palrs of EPI vlllages and nearby catchlng points ln che Sout,hern exr.enslon. Though there exlsCs a staElstlcally slgnlflcant, relaElonshlp bet,ween che CttFL and ATP, the varlablllcy ls exEremely large. Thls 1s noE, surprlslng glven Ehe varlous llmlEaElons of chese data, whlch cooe from an area wlEh an unknown nixt,ure of vect,ors and paraslEe sc,ratns (see also appendlx II, paragraph 5.2). 4.7. Incldence of g.E leslons after vector eonErol. Reeent analysls of longltudlnal ophEhaloologlcal daEa for a perlod of t0-12 year6 concrol, as well as reanalysls of pre-conErol daEa, have resulted ln the followlng concluslons of dlrect relevance t,o quest, lons of long terrrr concrol: - The mlcrofllarlal load ln the cornea and ln the antertor chamber of t,he eye ls llnearly relaced, durtng Ehe pre-concrol perlod, Eo the mfs load. If a cercaln mf load ln the eye ls reached, leslons of the Page 7 anEerlor segment, of Ehe eye st,arc Eo develop, and Ehese may ulElmaEely develop lnto bllndness. - Also the post,erlolr segmenE of the eye can become lnfected but, as a rule thle developnent, ls less acuce. The raEe of progreselon aft,er lnlclal Bcart,up le, cont,rary t,o ant,erlor segmenE leslons, not, very dependenc on che mf load and much slower. - Followlng lncerrupt,ion of transmlsslon, Ehe mf load ln Ehe eye drops ouch fasEer than Ehe mf load tn che skln. FurEhermore, che mf load ln the cornea falls agaln mueh fasEer Ehan the mf load ln Ehe antertor chamber of the eye. - Anterlor segment lestons: - Do not occur at low endemlclE,y levels - Can show regresslon lf causaElve agenE, i.e. mf, ls removed - Have a rat,e of progresslon proportlonal t,o the MF load in Ehe eye - Post,erlor segmenE leslons: - Can occur even ln low endemlclEy areas - Cannot, show any regresslon. - Rate of progreselon le less clearly dependent on mf load, and may contlnue to show slgn{flcant progresslon even after removal of the caueatlve agent,. - A emall proportlon of heavlly lnfecced people apparently do not develop severe eye leslons or bllndness. 4.7. Effect, of tvermee.Eln. Mlcrofllaria ln the skln are reduced to very low i'e-Grs ilrcfrGf-ffi followlng treatmenE,. However, ivermect,ln doe6 not enEer the ant,ertor charaber of the eye, and ocular mlcrofllartae are only ellmlnaced afcer a perlod of 6 monEhs. rt, appears t,hat, lvernecglnlnterrupts Ehe release of mlcrofllarlae durlng the flrsc 6 monchs afEer creat,menc by causlng a paralysls of t,he uterus of che adult, worq. A fewpat,lents do not respond t,o treaEmenc at all. 5.FIILL TRATTSUISSION I,IODEL In response to Ehe need for speedlng up t.he developmenE of a full t,ransmlsston nodel, Ehe Eeam froo Rocc,erdam universlEy had already developed an exploracory transmlsslon model for a feaslbtllcy study of dlfferent approaches and goprovtde an example of Ehe possible use of such a model durlng Ehe meeElng. IE was st,ressed chaE chls exploraEory model cont.alned many overslmpliflcactons and chat lt should Eherefore not yec be t,aken as a reallsclc model for onchocerclasls cransmlsslon and cont,rol. 5.l.Exploratory transmlsslon model. The Ewo mosE lmporEanE new quest,lons scudied In che exploratory model, concerned the modelllng of che populatlondynanlcs of che veccor and che unic, of E,lme t,o be used for Ehe lnt,erface bet.ween the t$ro eubmodels developed sofar. Though lt, was agreed in a previous meeElng thaE no aLEempt would be made at extenslve modelllng of vector- breedlng and migratlon, a certaln amounE, of quantlflcaclon of Ehe reproductlon dynamlcs of the vecEor ls essent,lal lf t,he model ls Eo be used for che stmulaclon of the lmpact of varlous cont,rol straEegles wlt,h a componeng of vector conErol. The compromlse proposed ln che exploratory model was Ehat,for a cerEaln area to be slmulaEed, boch the ecologlcally decermtned maxlmumposslble biclng rat,e would need Eo be speclfled for each mont.h of the year Eo allow for seasonal varlatlons. The model would Ehen, after comblnatlon wtththe vector-ParaslEe submodel, slmulate btttng and E,ransmlsslon rat,es, andthelr alEeratlona as a resulE of vecEor conErol, on a weekly basls. The aggregaced resulrs per monch wlll be used to slmulace lnfect,lon ln the humanpopulatlon. The encomologlscs conflrmed EhaE reasonable est,imaEes of Ehe requlred daEa could be provlded for several.breedlng sltes in the OCp and che meeElng supported chls compromlse as a suff{r:lenEly reallstlc approach. Noc acceptable were the aEt,empt,ed s{mpllflcaclons of modelllng at the human populaclon level. The maln alm ln Ehese aLtempts was co clreumvenE thevery laborlous calculat,lons lnvolved ln Ehe s(mulatlon of tndlvldual llfe Page 8 hlstorles (a technlque referred to as mlcroslmulaclon) whlch consumes large anount,s of conpucer t,lme and whtch would be a maJor technlcal constralnc of a full transmlsslon model whlch uses chls technlque. However, Ehe slnpllfled nethods could clearly not, correctly deecrlbe Ehe epldemlology of onchocerclasle, ae has been done so euccessfully ln che host-paraslEe model. Furthermore, Ehe technique of mlcroslmulatlon provides an ext,remely lmporcant flexlblllcy ln the practlcal use of the model by enabllng Ehe speclflcaclon and almulatlon of reallst,lc controt strategles and the lncluslon of faccors operat,lng on the lndlvldual leve1, Buch as sysEemaElc non-conpllance or excluslon from treactrent, ln chemotherapy based control. 5 . 2 . Re q u llqqqq! s for a full transmlsslon ggdel. It was therefore agreed chat the full transmlsslon model should be a comblnatlon of a mlcroslmulaclon model for the human part of the t,ransmlsslon cycle whlch should lnclude Ehe relevant aspect,s of che hosE-parastte model, and che modelllng approach t,o Ehe veeEorlal part of the cycle as proposed above under Ehe exploracory model. It was noEed EhaE the acEual development, and programrntng of such a model would noE BlBply be a matt,er of combinlng a few exlstlng computer prograomes but, would requlre a conpletely dlfferent approach to Ehe compuEer Eechnlcal quesElons. The reason for rhls ts the need to slmulace slmulEaneously a hunan populatlon, a vector populatlon as well es thelr lnEeracElon for perlods of several decades. Second1y, the whole aln of the modelllng exercize ls Eo develop a pract,lcal cool for use ln analysls and operaElonal plannlng ln che OCP. The compuEer program should therefore be able to run on a mlcrocomputer ln che OCP area and chls lnEroducea some speclflc t,echnlcal consEraint,s. A eubgroup of maEhematlclans constdered the6e requtremencs ln decall and concluded Ehat lE would noE be cechnlcally feaslble Eo run such a programme on the tBl'l XT compucers, whlch are preeent,ly avallable ln Ouagadougou, because of Ehelr llmltaclons ln adressable RAl.( memory and processlng speed. The only alEernat,lve would be to design the conputer prograo for a mlcrocomput,er wlEh a 32 blc INTEL 80386 nlcroprocessoro It was therefore proposed chat OCP purchases such a microcompuEer during 1987. The requtred conflguraclon, wlth at leasE 2 Mb RAM and a hard disk, would cost, around USD t0r000, and STAT sEaEed Ehac lE had made sufflclenc budgetary provlslons for such a purchase. 5.3.lpgglj;lggEg of the conEencs of a full t,ransmlsslon model. Based on Ehe precedlng dlscusslons on posalble cont,rol strategtes and vartous aspecEs of t,he cransmlsslon cycle, a ltst was made of che facEors whlch should be lncluded ln the final cransmlsslon oodel. Thls llsc ls glven ln appendlx IV. 16. PIJINS FOB APPLfED RESEARCE. The concent,raClon on quesElons of long Eerm cont,rol, wlE,h speclal reference t,o the central OCP area, helped to highllghu the oosE lmport,ant, quest,ions whtch need to be addressed by applled fleld research, before maJor changes can be made ln the presenc control scrategy and decalled plans can be made for long tero control actlvlEles. It ls of lmmedlate concern to know Eo what. extent the resldual reservolrs const,ltut,e a rlsk for transmlsslon, and lt was unanimously agreed thac applled research lnto thts questlon should gec copprlorlt,y. Xenodlagnostlc experlmenEs rrere cruclal t,o enabte Ehe lnterpretac,lon of the exEenslve skln snlp data from the central OCP area ln terms of Eransmlsslon rlsk. Top prlorlty should be gtven co experlmental lnEerruptlons of vector conErol ln epldemlologtcally well-deftned areas, where t,he reservotr of lnfectton has already fallen Eo low levels, such as ln the Farako. A llsE of plans and prlortt,les for applled research, as dlscussed durlng Ehe meetlng, ls glven ln appendlx LI. Pagt, 9 7 I{AIN CONCLUSIONS AND RECOHHENDATIONS. IE was recogrllzed tlraE, glven the rapldly decltnlng reservolr of lnfecElon tn t.he cenE.ral OCP area, declslons on major changes in control operaE. lons wtll need Eo be Eaken ln t,he near fuEure. Furchermore, flnanclal const,raincs will necesslt.at,e t,he reduct,ion or tnterruptton of vecEor cont,rol operaElons as soon as t,hls would be scienElflcally justlfied. IE ls also likely thac excernal Pressur€'s c,o cuE vect,or control operatlons wlll lncrease conslderably during nexE few years and posslbly rather lrrespectlve of che publlc health rlskslnvolved. The meeclng agreed cherefore EhaE. lE. rlas essenElal chac che OCp accelerat,es iEs preparatlons ln th(. central OCP area ln order to be able co change conc,rol st,rac,egies as soon as posslble, or t.o be able co demonstrat,e why cert.ain changes are not yet warranced ln certain slcuaElons. The Long Term Stracegy foresees Ehe lnEerrupElon of vecEor eonEroL, once Ehlsls epidcmlologtcally jusEtftabre, and iEs successton by a system of aurvelllance under devoluElon and tntervent,ton by chemocherapy urhenever requlred. It was agreed thac chts was Ehe mosc llkely scenarto co be followedln the future, buE Ehat, Ehere remalned a mulclcude of quanElcaElve quesElonsEo be resolved before chls staEement, of lncenc can be Eurned inEo a suecessfullong Eerm scracegy of onchocerelasts concrol. The mosE importang questlons inthls respec-c are: when t,o lncerrupE veeE,or control, what. are che relat.ed r isksof reerudescence of cransmlsslon, whac ls the speed of reerudescence, whaElevel of surveillance ls required r..,nslderlng Che dynamlcs of recrudescence, whaE ls the lopacc of lnE,ervenE,lon and whaE is che cosE-ef f ecc.l veness of chevarlous combinaE. ions. Mosc of Ehese vtcal quesEtons cannot, yet be answered, buE very relevant research projects, such as t.he xenodlagnosEic experlmenEs, thL r-arako experlment., che lvermect.in eommunlcy crlars ecc., are underway and are expeeEed Eo produce signlficanc results wlthin Ehe nexc one E; Evro years.However' even when chese dlfferenc results becorne available, they will needEo be used ln an lnEegraEed way fu. order co arrive at valid predlctlons of t.helmpacE of various sEracegies and the opttmal planrrlng of the long Eerm contr.rrof onchoeerclasls ln Ehe ocP area. rn oEher words, *huc ls rrrgencly rrceded isa pracCleal and reaIIst. lc tr;rnsrnlsslon model whleh ts suf f tciclC ly spt-,ctt 1r.f <.lr E,lte quesclotrs f acing t.tre oCP to be <-lf dlrect trse l-rr a cost- eL Iecc.iveplanning of the conElnued, long term c.ontrol of onchocerclasis. The development of sueh a transmlss tor, rnodel has natural Ly bcc'n one of the mosE lmport.anc objeecives of the applied modelling projecc sirrce iEsbeglnning. However, a major guldellne, whlch has been adopcecl throughout Ehemodel development, has been co synchronlze rnodelling wtctr itre progress in dacaanalysls ln order Eo errable continrrous EesEing of t[e model" "!"in"c ctuobserved daEa ln Ehe oCP. The raclonale behlnd Ehls approa.h las obvlously toensure chaE modelling would remaLn a relevanE and praeElcal activiEy for theoCP and would not, dtverE to become an academlc exerclse. The lncreaslng urgency Eo acceleraE,e Ehe developmenE of a fu11 Eransmlssion model *." i".y undersEandable and the proJect had been modtfted_accordtngly. Neverthelcss,thc meeclng lnsisced thar a properly tesced, reaLlscle nnJ i.11.ll.altransmlsslon model remained the prlrrrary alm of che projeet for the oCp, andEhaE chls prlnelple should noc be sacrlflced co demanas ror prernacure resu I Es . The condiclons for a reallstle and pr,r, clcal transrntsslon model for the OCpcan only be mec by uslng a mlcrosimulaclon approach which will provide Ehenecessary flexlblllEy ln stmulatlng all posslble scenarlos of relevance t,o EheOCP. It r.r111 not lre teehntcally posslhle co program and run sueh a model ont,he presenc IBM- ltT comput,ers ln Ou.rgadotrgou, and ic is recomrnended thac OCt,purehases a mlcroeomputer wlth an INTF:L 803t16 proeessor and 2 MB RAI-1 to enablethe utlllzaelon of the Eransmtsslon model wlchln the programme area.Concernlng che moCel development, 1t was agreecl t.haE the team from Rot,terdam Page l0 should now give full prtorlty Eo che cransmlsslon model, and Ehe followtng aceelerated c,lmecable was agreed upon: - l'tarch/Aprtl 87: completlon technleaL deslgn; to be dlscussed wlch STAT. - June 87: dlscusslon of deslgn L,ransmlsslon model wlch EAC - AugusE 87: dlscusslon progress computer programmtng by R'dam wich STAT. - December 87: Compleclon of a prellmlnary model and smali neec,lng on t.hls nodel 1n R-dam followlng JPC. - Jan.87-April 87: Quantiflcatlon and EesElng of prellmlnary model ln OCP, and lmprovement of model and compuEerprogrammes by R-dam. - June 88: Complete model and flrsc results Eo be presenE,ed Eo EAC; compuEer Program oPeraclonal ln Ouagadougou and used ln analysts and planning. Appendlx i, pagc I ALTERNATIVE STRATECTBS FOR U)NG TERX CONTROL AND SURVEILIJINCE. The meet,lng on applled modelllng dtscussed whlch posslble strateglesshould be consldered as reallstlc alEernat,lves ln Ehe long t,errn eonE,rol ofonchocerclasls- The quesclon of survelllance was also addressed and a llscwas drawn up of lndices for monlEortng of the impacc of Ehe dlfferencaccivlcles. The dlscusslons referred excluslvely co a sl6uatlon whereprolonged vector conErol has resulced ln che vlrtual ellmlnaEion of thereservolr of infecElon. AII avallable evidence suggesEs EhaE large parts ofthe cenErar ocP area wlll reach such a slEuatlon wlthln Lhe nexE 2-3 years,and lc is particularly to thls area thac Ehe varlous opElons, whlch weredlscussed, apply. The objecclve of the long Eerm conErol acttvtEles menEtonedln chls report, ls cherefore Eo ensure chat, c,here wlrl be no recrudescence ofonchocerclaels as a dlsease of publlc healch loportanee 1n areas where thedlsease has been successfully controlled and where the reservolr of lnfecElonls nearly ellmlnaEed. I. LONG ISRH iOrfiROL STRATEGIES The dlfferent st,racegles dlscussed can be grouped as tarvlcrdlngoperaElons, chemoE,herapy or combtnatlons of chfse cwo. l.l. Larvlcldtng The posslble larvlclding opclons are: ( a) t'to reduct ion ln rarvicldlng. Thls would be the saf es E alcernar lve whlchn-rsTEZ*n-G"Trffidur1ngchelasEl2yearsbuc1c1sf1nane1alIy and pollElcalry unaccepEabre. Nevercheress, this opclon shourd belncluded ln the model for comparlson pu.po""". (b) gomplete lgcerruptlon of larvlqldlng. Thls ls rhe oprlon foreseen ln cher'otrgjerm sEE.ffiaJsEseali-[he assumpEron EhaE rhere wlrr bevlrcually no risk ln recrudescence of Eransmlsslon glven the nearrydepleEed reservolr of lnfecclon. However, thls .o.oln" todat.e still orrlyan assumpE. lon. (c), t'lg{rYaJ larylclding. Though manual larvlcldlng has been successful inserecEeo pracesr t$renEy years of experlence lndlcaces thac thts 1s noc. avlable solutlon for che larger part of Ehe cenEral Ocp area. (d) 9ontinrrred larytei4ing ar a f ew Ereedlng srlg_s of major irnporcance. Atvpicar e*i*pffifr"luf Sreeairu srteffinJsEoigrrffi-F che majorsource of annual relnvaslon lnEo the tocal Red and whi[e Volta bastn, andparElcularly so durtng perlods $rlthout. vector cont,rol. Evldence for thlsmlght be che surprlslngly low vector denslcles ln che Red and whlce voltabasln ln t975 when larvrtrarng had not yet starEed here, buc when uhe BUrhad come under contror for che flrsc t,loe. Larvrcldlng'rn ",.r.t-speclarbreedtng slEes eould posslbly be qulte cost_effeetlve. concllugl uuc lelliea l3fvlcl,dlng ar q lower cost. Thls rntgh6, tnvorve rheuse of longcr lncer-rpp-t1".aEi-p"ifoT"T-reffiTng Erearmefc in che dryscason' However' :1n:u chese opclons impry trrat rirvrcrarng-rru, aocontlnue for an unlimlted perlod over a wlde area and aE eonslderableexpense' Ehey hrere not consldered co be very rearlsElc artern.cires. "Bi€ Bang"-larvlcldiE. Thls lmprres larvlcldlng for a brlef perlod lnorder co wlpe ou. a locar vec.or popuratlon and to subsequen.ly sroplpr:vicldlng and lec che breedlng slte be repopulated. Even lf thls BlgBang ls glven only once per yeai, but aE Eh; itgtc ror"nt, ra-rry resulErn a slgnlflcanc reductlon ln che ABR arong *.ny b."uarng'siie", ,na ,:speelal ly along c,hose locaced more co che norcir. (e) (f) Appendlx [, page 2 General comments on larvlcldlng. The vlrtual ellmlnaclon of che reservolr of lnfectton ln the eentral OCP area w111 creaEe a complet,ely new slt,uaclon.It ls aB yet, unknown whac che rlsk of recrudescence of Eransmlsslon would belf che vector te allowed t,o repopulaEe Ehe numerous, and poEenclally hlghly product,lve, breedlng slEes ln chls area. The meerlng scressed chac lncerruPtlon of vecEor conErol needs Eo be carefully prepared and monlEored. This lncludes a quanttcat,lve assessmenc of the risks lnvolved through a combinaElon of applled research (see appendlx VI) and applled modetllng(appendlx V), as well as che developmenE of sound conEtngency plans. A htghprlortty for Ehe eentral OCP area are the planned experlmental lnterrupElons of veccor cont,rol, and che effect of t,hese lnterrupclons should be closely nonlEored by lncenslfled entomologlcal evaluaElons. Plannlng of fut,ure strategles should also lnclude a cost,-beneflc analysls of the posslble cont,lnuatlon of llmlced veccor conErol operaElons, and 1n parEleular of Ehe operaclons mencloned under (d) and (f). 1.2. Chemtherapy. (a) l.{acrof 11ar tc1de. The presenE poslElon wlEh regard t,o Ehe search for a macrofllarlclde does not. permlc a predlcE, lon of the poinc ln t,lme at, whtch a drug may become avallable. But thls opclon should be lncluded ln Ehe Eransmlsslon model co allow comparisons wlth a sErategy based on such a radlcally dlfferent lnt,ervenEion. In parElcular cosc comparlsons mlghc be useful for questlons on conEinued fundlng of OCT. ( b) Ivermec-e ln The only reallsclc alEernat,lve cont,rol mechod tn the near future ls EreaEmenE wtch lvermecEln whlch has been shown t.o be an effecElve mlcrofllarlclde wlth almosE no adverse reactlons. The only sympcoms t,haE have been observed t,odaEe are m(ld e.ardtovascular changes and llghr fever, but the number of cases treaEed ls llmlted. Treatment, ls conEralndtcaced ln pregnant $romen, ln lacEaEing women durlng che flrst monE,h of lactatlon and ln chlldren below 5 years. More experlence wich the drug ls needed before lt, can be applled Ln mass campalgns, buE, lf all goes well, Ehe drug w111 be avallable for,mass dlscrtbutlon from mld 1988 onward. Since t,he use of lvermect,tn is a complecely new lnEervenElon mechod, the use of mode1l1ng may be espeelally helpful ln opcimal and reallsElc planntng of long Eerm control sEraEegles using lvermecEin under devoluElon. ObjecElves of lvermect,ln based eont,rol straEegles ean be dtsease eoncrol, t.ransmlsslon control or a comblnatlon of t.hese two durlng the post-OCP perlod, whlch we wlll refer co aa recrudescence conErol: - The alm ln dlsease concrol ls secondary prevenElon by uslng lvermeec.lnln treaElng lnfected patlent,s at, rlsk of developlng (severe) pathology. Thls requlres a htgh level of survelllance co enable Ehe clmely ldenElficatlon of t,hese pat,lents, and Ehe feastblllty and cost of Ehls survelllance needs to be crlElcally evaluaE,ed. - Transmlsslon conErol lmp1les Ehe use of {vermectln for conErolllng t,ransmlssion by regular t,reaEment, of che populaClon whlch 1s che source of t,he lnfecclon ln Ehe vect,or. Populatlons wlch a relaElvely hlgh prevalence of lnfeet.lon wlll sc1ll be found durlng devoluclon and some of chese have already been ldenclfled, such as Pendle, Goreba-Somun and vlllages ln che BUI area. The potenclal for tvermeccln for Eransmlsslon cont,rol ls noc yeE known, but varlous st,udles are planned for Ehe near fut,ure and oosc ansqrera co thls questlon should be avallable by Ehe second half of 1988. - The obJectlve ln recrudescence concrol ls tnlElally one of transmlsslon cont,rol but, lf chls falls, lc wlll be necessary Eo fall back more and more on Ehe use of lvermect,ln for secondary preventlon only. Appendlx I, page 3 A maJor characEerlsclc of an lvermect,ln based control strat,egy ts Ehe method of creacmcnr, used, whlch may be any of the followlng: - TreaEment of htgh rlsk cases followlng dlagnosls. TreaEmenc of hlgh rtsk groups such as adulE males or flshermen. - TreaEmenc of paElent,s, dlagnosed as havlng onchocerclasls lnfecElon. - Blanket EreaEnenc ln htgh rlsk vlllages followlng communlcy dlagnosls. - Blanket treaEment, t lchouE communlt,y dlagnoele ln all villages where onchocerclasls mlghc c,heoretlcally occur. IE was argued chat blankec treaEment wtthout any dlagnostlc JusEtflcatlonmlghc be rather unechlcal and costly ln the future central OCp area, where t,he vasc majorlty of che populatlon will have no onchocere.iasts infecclon what,soever. l'lass treaE,ment, wlc,h tvermectln ln chls area can probably only be underEaken lf lndlcaced by unaoblguous eptdemlologlcal evldence. rndividual EreaE,menE may be glven followlng an lnEervlew and theldenclflcaElon of certaln sympt,oms, such as ltchlng. However, experlence suggests chac Ehls method ls very subjecEtve and Eoo unreliabie co be of muchpracElcal use. More objecclve, but at the same Elme more lnvaslve, are EheresulEs of skln snlp examlnatlons as a basls for treatment,. The lmporganEquesElon here ls which health factlltles wlll do the examlnaclons and provldetreatmenE. The PHC worker may be ln Ehe besc poslclon co do so, but, ,p".cfrom already belng overburdened wlth other t."Le, he ls not, yeE cralned forthls work and laeks Ehe necessary equlpmenE, such as a micro".op". TheslcuaElon would lmprove constderably 1f a pracclcal lmmunodlagnoscic EesEbecame avallable. Anocher approach would be lndlvtdual or coilmunlcy dlagnoslsby moblle epldeurlologlcal t,eams. A slmpl(fled meEhod of communlty dlagnoslsmlghc lnvolve skin snlp examlnaElon ln hlgh rlsk groups, sueh as adult males.Based on Ehe prevalence of mfs ln these gioups lt, can t,hen be declded Eo EreaEonly the lnfecced persons or the Eotal ,iltage. The effectlveness of lvermeccln ln recrudescence control wlll depend to alarge excenc on che effle.lency of the survelllance syscem ln dec,ecclng andtreaclng new cases of infectlon, and one lmportant oL1"ctlve of modelltng ls acomparatlve st,udy of che rlsks lnvolved ln inadequate executlon ofsurvelllance. The potenclar of lvermectln for transmlsslon control depends onthe. coverage of Ehe lnfected populat,lon wlt,h repeaced annual EreaEmenEs overan exEended perlod, on the conErlbuElon co transmlsslon of Ehe mlcrofllarlaesurvlvlng lvermectln treaEment and of Ehe lmportance for transmlsslon of thenon-EreaEed populaElon. In thls conEexE lE ls lmporcant to lncrease ourknowledge abouE the degree of mf-lntake and larval derelopmenr ln fltesfeedlng on Persons who have only a low tncenslty of lnfecllon, wheEher naE,urally or as a resulE of lvermectln treaEment. IvermecEln crlals are presently underway co lnvesclgace 1f repeaced admlnlsEraclon of lvermectln (ac 6-monthly intervals) has a cumulaElve andscerlllzlng effecE on the feeundlcy of thl adult feurale worm. rf conflrmed,thls would SreaEly lncrease Ehe pocenclal of lvermect.lrr for onc.hocerclaslsconErol and thls poss1blllty should Eherefore be lncluded 1n che model. 1.3. Comblnatlons of larvlctdlng and chemotherapy. Comblned sErateSles uslng both larvlcLdlng and chemotherapy would appearto be very cosEly beeause of che need Eo malncaln Ewo dlffe."nt op".aclonalstrucEures. Modelllog mlght be helpful ln lnvestlgaclng tf Ehe lncelllgenr use of a speclal comblnacton of Ehese two mechods olghc noE be cosE-effecclve afEer all. Such comblned sEraLegles mlghE be necessary for cercaln speclalslt,uatlons where t,he separate conErol methods cannoE achteve on chelr own aeufflclent degree of control. one poselble exanple of such an area ls the BUI where a Ertal of the cooblned use of lvernectln and larvlcldlng w111 be underEaken Ehls year. 0f speclal lnterest, also for the "*t"nJron areas, are the comblnaElon of lvermecctn Ereat,menE and less cost,ly larvlcldlng meEhods, such as che "Big Bang" method dlscussed above. Appendlx [, page 4 ln skln snlps. The sktn snlp cal meEhod whle-h has proven lts the deserlpElon of boch che Cs dlsadvantages are thac lE 1s s not detecE pre-patenC ry senslElve ln paE(ents wlEh 2. INDTCES POR. SURVEILLIINCE. As parE of the discusstons on the cechntcal aspects of survetllance underdevoluElon, a llst was made of the varlous lndlces whlch mlghc be used EononlEor the entomo-epldemlologlcal sltuatlon and the lmpact of long cermcontrol acElviE,les. These indlces nay be ranked according Eo Ehetrchronologlcal value ln the det,ectton of recrudescence of onchoeerelasls as: encomologlcal lndlces' meaeures of tnfecclon and meagures of paEhology. 2.1. EnEorcloglcal lndlcee: (a) gtttng Ya":. These are useless for most devoluclon slEuarlons, when t,hevecEor density wll1 be very hlgh followlng the repopulation of thebreedlng slt,es by S.damnosum, buE when che hlgh bicing rages provlde nolnformatlon on t,ransmlsslon levels. only as ieedback for speelallarvtcldlng acclvltles may blElng rates be of some value. (b) Transnlsston Potenclals. Because of the very small reservolr of infect,lonln man durlng che devoluElon perlod, lt ls expected that only a very smallproporElon of Ehe vectors can become lnfecttve, and transrnlsslonpoEenclals (or relaEed indlces) wlll therefore be, more t,han ever, Ehepreferred enEomologlcal tndex for che devoluElon perlod. Unfort.unately, t,he use of chls lndex requlres rather advanced dlssecclon cechnlqu"" ..d,because of the very low lnfectlvtcy raEes, Ehe dlssecElon of *.r"h lr.ger- numbers of flles Ehan is presently belng done by Ehe ocp. (c) Esclmiced rnfecclve BlElng Rate. A compromlse may be co use an esElmat,efor che IBR by comblnlng oUservea blting rat,es and epldemlologlcallnformaclon on lnfecclon ln the local human populatlln. IE i; presenEly noE clear how such lnformation can be comblned, whfle such an es6lmated rBR would definlcely noE be rellable in a relnvaslon sltuaElon.Nevert.heless, lc may be very worchwhlle to pursue t,hts ldea in thetransmlsslon mode1. (d)'Eellec TraPs. Lc is sometlmes suggesced thac craps are a perfecc soluElonfor enE,omological survelllance under devolut,lon because they are cheap and requlre litcle lnpuE of manpower. However, unless a DNA probe lsdeveloped for Ehe dececclon of L3 larvae ln bacches of blackflles, Ehe crap ls even less useful Ehan blEing rat,es ln a eltuatlon wlEhouc vecEorcontrol, because 1t, captures both ant,hropophlllc and zoophlllc flies.Furthermore, lE appears thaE a loE of expertence ls needed for the exacEplaclng of che E,rap and EhaE for nany "atchlng slEes the E,rap cannog beused durlng che peak of t,he ralny season when vect,or breedlng ls hlghesg. 2.2. lleasures of onchocerclaels tnfectloo. (a) Im$!-rngdlagnosclc lndlces. The advancages of a practteal lmmunodtagnost1c meEhod for onchocerclasts lnfect,1on, lf lt ever becomes avallable, are exPeec,ed co be ease of use ln the fleld and early dlagnosls of lnfecclons when chey are Bt,lrl ln che pre-pat,enE st,age. The import,ance of earrydlagnosls depends on che dynamtcs of recrudescence of t,ransmlsslon andEhls quesElon needs to be studted furEher. (b) Prevatence and lnEensley of mlcrof lLar tae examlnaElon ls at present Ehe only pract,l value as a valld epldemlologlcal cool for prevalence and lntenslEy of lnfecEton. I lnvas1ve, requlres spectal equlpraenc, doe lnfecttons and thaE lc ls probably not ve Appendlx [, page 5 very llghE lnfect.lons. However, slnce lt ls assumed thac very llghclnfectlons are not of lmmedlaEe publlc healch lmportance, the usefulnessof a sensltlve dlagnosElc meEhod for llghcly lnfecced cases wlll depend onEhe exEenE Eo whlch such cases may conErtbute Eo t,ransmtsslon. Thlsquestlon ls preeenEly belng studled ln xenodlagnosE,lc experlmenEs and Eheresult,s ehould be lncruded ln the model ln order to predicE Eheslgnlflcance of low senslElvlty and eepecially of faise-negagives ln anlvermeccln based control sErategy. 2.3. Heasures of pathology. (t) P&"...€.sEln leslons. Flne papular rashes of Ehe skln mlghE, occurdurlng early lnfecclon wlth onchocerclasls and t,he appearance qf a slzableproportlon of such cases in the communlt,y may be an indlcatlon ofrecrudescence of onchocerclasls. Howevei, such rashes may be due Eovarlous other causes and are noE speclflc for onchocerctasrs. rn general,t,he meeclBg sras crtElcal abouE the use of dermatologlcal slgns and sympcortrs whlch have shown to be unrellable ln the epldemtoligrcal evaluat,lone of t,he OCp. (b) (c) ,rncldencg an4 prevalgnce of ocular rggrgns. Appearance of eye pachorogymay be vlewed as an rmp.rrant srgn ifiTc-frTrgn.i" rt," srarc ofrecrudescence of onchocerclasls as a publlc healch problem, and chls faccalone jusclfles an ophthalmologlcal cooponenE ln onchocerclaslssurvelllance. Furt,hermore, lt should be recalred chat leslons of t,heposEerl0r segmenE of-che eye may devel0p even when the rnt,enslty oflnfecElon ls low. cllnlcal trlals have demonscraced chat Ehe use oflvermecEln lowers Ehe mlcrofllarlal load slgntflcantly buc lc ls noE yetknown to erhaE ext,ent, Ereatment wlth lverr""Iln prevents Ehe lncldence ofposcerlor segmenc lesrons ln lnfected lndlvlduais, and Ehls maynecesslE'ate some conLlnued evaluaElon of posEerlor segmenE leslons durlnga perlod of lvermecEln based conBrol. fE ls recogntsid chac che dlagnoslsof oeular leslons requlres ophthalmologlscs wlch tralnlng a1d experleneeln onchocerclasls, as well as speelal ophttralmologlcal equlpment, and chlsEype of survelllance can t.herefore only be undertiken ,c'a very llmicedscale durlng che devoluElon perlod. Ilil9l"= !?S"".- These are essenLral for measurlng rhe frnar pubrrc.nealEh lmpacc of control act,lv1t,les. Twelve years of experience ln cheocP have shown thac the sJogren hand tesc is a very practrcar, rerrabreand repeatable cest. rcs disadvanEage ls that iE neasures only one aspec.of the visual funcElonl 1.€. che vlsrral aculty, and glves no lnformaclonabouE detorlatlons ln Ehe vlsual fleld resutting from leslons of EheposEerlor segmenE of the eye. unfortunately, a proper vlsual flerd cescls exEremely tlme-consumtng and and ie therliore noE prae.Elcar forutlllzaclon on a rarge s.aie tn che fleld. Nevercheless, . p.ort"ron forfunctlonal bllndness should be made in che model co arrow for theslgnlfleanc,ly dlfferenc epldemlology of posEerror segmenc lesrons. Appendlx II, page t PUINS A"ITII PRIORTTIES POR A"PPLIED RESEARCE l,llth Ehe rapldly decreaslng reservolr of lnfect,lon ln the central OCP area' quesElons of lnEerruPElng vector control and of long t.erm cont.rol underdevolutlon, are becoming lncreaslngly relevant. It ls generally recognlzed Ehat urajor changes ln che concrol strat,egy need Eo be carefully prepared using the results of applled fleld research and experlmental lnEerrupt.lons of vecEor conErol ln spectflc zones ln che central OCP area. The meec,lng on appLled nodelltng was a good opporEunlEy to dlscuss these research acElviEies, and E.o revlew and propose speciflc plans for the comlng year. The followtng ls a Buomary of proposals for applled research whlch were glven the hlghestprlorlty by che meeclng: I . X,eoodtagnoe tLc exper Lments. A cruclal questlon for the cent.ral OCP area concerns che exEent to whlchthe remalnlng'reservolrs of lnfectlon constltut,e a rlsk for t,ransmlsslon lf the veccor would be allowed Eo reCurn. The results of che epldemlological evaluaEton have shown thac che prevalence and lntenslty of lnfect.ion arefalllng rapldly and chac presenE levels of lnfecclon are no longer of publlchealth lmportance. However, before chese epldenlologlcal data can be lnterpreced ln Eerms of Eransmlsslon rlsks lE erllt be necessary co quanclfy che PoEentlal of llghcly lnfected cases and people, who have jusE become skln snlp negaEtve, t,o conE,rlbut,e Co Eransmisslon. Xenodiagnosls ls an approprlaEe t,ool for scudles of thls kind. C1assical xenodlagnosEle expertment,s have always lnvolved heavlly infeeEed volunEeers but, OCP has recenEly started with xenodlagnosls using llghtly lnfected volunt,eers from t,he Farako ln Ehe cent,ral ocp area and from a non-controlled area ln t,he wescern ext,enslon. Prellminary resulcs lndteaEe chac che Proport,ton of fl1es, whlch take up olcrofllarlae durlng a bloodmeal,ls very much reduced when Ehese flles are fed on llghtly tnfecced pat,lencs. On che oEher hand, limiced resulEs of one small study seem t,o suggesc c,haE c,heprobabillty of developmenE tnEo lnfectlve larvae ls slgnlflcantty hlgher for nlcrofilarlae, Eaken up from llghtly lnfected people, fld chac chlsprobablllcy ls not reduced for mlcrofllarlae orlglnaclng from old adulc woruls. The meetlng emphaslzed thac thls Eype of study ls exEremely relevanEfor'che OCP and should be glven a htgh research prlorlEy. Ic was noEed EhaE xenodlagnostic experlnents are relatlvely lnexpenslve, requirellctle manpoerer and can produce lmportant, results wlthln a shorc period. Further xenodiagnosElc experlmenEs were urgenEly needed to quant,ify the relaclonship bet,ween microfllarlal denslEy ln the skln and che uprake of mlcrofllariae and development lnco L[ and L3 larvae. These experiments should be undertaken jolncly by VCU and EPI, and lnvolve a considerably larger number of volunteers over a wlder range of skln snlp denslcles, and larger numbers offlles Per Person. A small, buE well-deslgned, study should be uodertaken Eo conflrm that larval developmenc ls lndependenc of the age dlstrlbucton of the adulc $rorms. 2.Faraho experLnent. Of parcleular relevance for t,he phaslng ouc of vect,or conc.rol operatlons, are exPeriment.al lnt,errupclons of larvtctdlng ln areas Hhere Ehe reservolr oflnfecclon has become low, ln order t,o study che rlsk of recrudescence of transmlsslon 1n relacton Eo che epldemlologlcal slEuaElon. Such an experlmenEls planned for the area around and upstream of Bamabougou ln the Farako. The Farako has been well-protected slnce 1975 and has also beneflc,ced fromparElally suceessfull veccor concrol under the FED campalgn durlny, several years before Ehe scart of OCP. In preparaElon of the expertment, ErI has undertaken an lntenslve serles of slmple evaluatlons ln the area and has found prevalences of mfs ranglng from 0.82 co t3Z ln early 1986. Appendtx II, page 2 The deslgn of the Farako experlment, foresees the lnEerrupclon oflarvlcldlng early tn the year, chough noc before the water level aE Bamabougouls hlgh enough to allow large seale breedlng and che rapld butld-up of the ,vector populatlon to an equllibrtuo level. once thaE level has been reached,larvlcldlng w111 contlnue to be suspended for at, leasE anocher t,hree weeks, inorder co allow the sEudy of a vect,or populatlon wlth a relaElvely stable agedlscrlbuclon. A very tntenslve enEomologlcal evaluaElon will be undertakenduring thls perlod to allow a detalled quantlflcatton of entomologlcar indteesof transmlsslon. Aft,er thls three week perlod, the need for lntervent,ion willbe determlned on the basls of ent,omologtcal evaluatlon daca and larvlcldingwtll be resumed as soon as Ehe cumulat,lve transmlss(on poEenELal has reachedthe classtcal threshold level of 100. rE ls of partlcular tnteresg t,o scudyhow long lt w111 cake for Ehe t,ransmlsslon potentlal to bulld up aga1n.Though lnlc{ally lntended for 1986, t,he Farako was postponei when abar,ereelstance ln savanna epeclee was for the flrsc Elme encountered ln Mall.Furthermore, noc all oCP staff dld agree chac Ehe prevalence was sufflelenclylour Eo JusElfy such an experlment. However, two ,ill"guo upstream ofBamabougou, t,he one whlch had che prevalence of l3z.r,J "noiher wlch aprevalence of. 77r were reexamlned ln novenber 1986. The prevalences had byEhen fallen Eo 82 and 32 respectlvely, and lE 18 now unanlmously agreed thaEt'he experlment ls justtfled and should be undercaken thls year Lefore thereservolr of lnfectton has fallen so low that chls experlmenc loses its re levance. - The meeElng sEressed t,hat. the Farako experlment ls essent,lal for fuEureplannlng and chac lt should recelve top prtorlty. 3.Pendle trlal rt was also proposed Eo undertake a sEudy wlth experlmencal lnterrupt.ionof larvicldlng along che Dlenkoa near Pendle, and to comblne chls wlrh acommunlty Erlar of che use of lvermeccln ln Eransmlsslon cont,ror. Thlscomblned scudy would lnvolve the lnEerrupElon of larvrcldlng durlng the ralnyaeason ln 1987 in order co allow repopulitlon of che breedlig slces and theescabllshment of a sEable local vecEor populaElon. rt should be accompagnledby lntenslve enEomologlcal evaluatlon to allow a decalled scudy of lnfecclonand lnfecElvlEy rates. These dat,a may slgnlflcantly contrlbute Eo ourunderstandlng of t,he dynamlcs of .e..Ld"""ence of transmlsslon ln a sit,uat,ionwhere recrudescence has actually occurred and the prevalence of mfs ls stillrather hlgh' Durlng the subsequent year rarvlcldlng wrll agaln be lnEerrupEeddurlng the same perlod buE lE w111 clren be proc-eeded by blanket lvermecElncreacment ln all Ehe affecEed onchocerclasls vlllag." in che area. AnlnEenslve enEomologlcal evaluat,lon wl11 also be malnEalned throughou6 Ehlssecond year and the comparlson of the lnfeet,lon and tnfeeElvtty rates ln thevecEor bet'ween Ehe tgro years wlll provlde unlque informaclon on the pocentlalof lvermecEln for Eransmlsslon conErol ln a real slEuaElon wlch recrudescene.eof lnfectlon. The quesElon was raised lf lncerruptlon of vector concrol durtng l9g7 wasnot unec,hlcal, because of che near certalncy that chls will resurc lnonchocerelasls transmlsslon. ft was recalled thac the avatlable evldeneeshows chat slgnlfleanc Eransntsslon had Eaken place ln thls area beEween l9g0and 1985 whlch, though noE yec resultlng ln a severe publlc healch problem,nay have caused relnfecclon of abouE half che populacion rn tne vtllage ofPendle' Transmlsslon ls presenEly agaln tnteriupced chrough ve"tor cont.rolbuc no oEher lnEervent,ion ls planned for the ne*l f"* y.".i. it" poprraclonln chls area wrll be becrer off wlth the proposed t.t"i and a specrarryorganlzed lvermecEin dlstrlbuElon progrartrme ln chls area. Howeier, ocp shouldtake the responslblllty of ensurlng a proper syscem of tvermee-cln dlstrlbuclonin the Pendle area durlng the years followtng the crlal. The mee3lng very mu(:h supporEed Ehe proposal for che Pendte cital and was of the oplnlon chaElt shourd recelve an equally hlgh prlor(ty as che Farako experlment. The bestmoment Eo starE lnterruPElng vecEor conE.rol was thoughc to L" "".1y June,and lvermeccln should be glven durlng April t9gg. Appendlx II, page 3 4.BuL trlal The proposed BUr trtal was seen as a good Eest of a strategy, whlch combines vect,or control and lvermeccln creatmenE. Lt ls especlally relevant .for areas where control has always been dtfftcult and whlch are of speclal concern for the long term control of onchocerclasls ln Wesc-Afrlca. Adetalled protocol of thls trtal ls given eleewhere. It was also noted Ehat varloue oBher trlale of use of lvermeccln for transmlsslon control are belngplanned for che near future. 5. Iotegrated epldenlologlcal/entorologlcal studles (a) t'Iestern extenslon. Reference rdas made t,o varlous old proposals fordetaIlFiitGEI-EPr/vcu scudles ln non-concrolled, meso- or hypo-endemlc, areas ln the western ext,enslon. The obJectlve of t,hese studles would be tolnprove our underscandlng of the eplden(ologlcal slgnlftcance of encomologlcallndlces ln areas wlch a low lntenslty of Eransmlsston. It 16 unlikely thac t,he funds, ltme and manporrer can stlll be found to do the elaborat,e studles referred co by t,he research meetlng of 1985. However, lt might be opporcune Eo revlew the lnformaElon collected recenEly ln the west,ern extenslon and t,o seeto what, exEent slmple, addlElonal, studles could provlde Ehe necessary lnformaElon. (b) Souchern Extenslon. In t,he concrolled oCP area, the epldemlologlcal data reflect nearly exclustvely lnfecttons acqulred durlng the pre-control perlod, whlle Ehe ent.omologlcal evaluatlon rras only establlshed around the clme veeEor control sEart,ed. WtEh Ehe except,lon of sone well-known relnvaslon or problem sPots, the epldemlologlcal and ent,omologlcal daca do therefore not, relaEe. The sltuat,ton ls compleEely dlfferent ln the southern ext,enslon where entomologlcal data are avallable for a perlod of 8 years without, vecEor cont,rol and a large number of epldemlologlcal evaluatlons have been done in 1978 ln vlllages whlch are locaEed close to the catchlng potnEs. However, allulcatlon of chese dat.a for the lncegraced analysls ls the facc chat all rhe sub-specles of S.damnosum, as well as t,he dlfferenc epldemlologlcal pacterns, are found lntermlxed. The results of an analysls of the relatlon beEween the epldemtologlcaldata of 1.978 and t,he subsequent entomologlcal daca for 1979-t986 was presenEeddurlng t,he meet,lng. Early Ehis year, EPi w111 underEake follow-up surveys ln several of the vlllages ln the Sout,hern excenston and thls w111 enable a far more lnterest,lng analysls of longitudlnal epldemiologlcal and entomologlcal dat,a whlch refer Eo the same perlod ln tlme. Because of Ehe lmport,ance of these longlcudlnal data, t,he rneetlng agreed that addlclonal surveys would bejustlfled 1f thls would lncrease the slze of thls longltudlnal data base. General renarks In all t.he experlmenEs menrloned above, the effect w111 be assessed uslng encomologlcal measuremencs of lnfecElon and lnfectlvlEy raEes ln t,he vecEors.It wlll therefore be essenElal chat the ent,omologlcal daea are of a very hlghquallty, sd that, every efforc ls taken co avold common problems such as undercountlng of Lt and L2 larvae. It was agreed thac the best technlctans would be made avallable for che sEudles, t,hat etalntng wlch liayer's Actd Haemalum w111 be used throughouE, and t,hat quallcy control mechanlsms ehould be tnc.luded ln all experlmencs. Because of che present, flnanclal problems of the programme, iE can be expecced thaC obJect,lons wtll be ralsed agalnst excra expendlcures for the enEomologieal evaluatlons requlred for these scudles. However, the meet.lng warned EhaE savlngs on t,he enLomologlcal evaluaE,lon are not, necessartly savlngs for the programme, and thaE such savlngs would deflnlcely be counterproduct,lve lf applled co che above studles. lnadequate evaluaElon durlng Ehe experlmental lnterruptlons over Ehe next few years would certatnly cause a delay ln the large-ecale reducElons ln larvlctdlng operatlons and chus result ln slgnlftcanE losses for the programme. Appendlx III, page I YECTOR-PARASTE MODBL Durlng a meeclng ln RotEerdam ln December 1.985 che baelc contencs of a veetor- ,paraslte eubmodel Lrere establlehed. A flrec cegt verston of thls model(called vEcPAR) has been developed durlng 1986, and has recenrly been resEedby the ocP. A summary descrtpcion of the nodei, t,aken from the manual, hadbeen dlstrtbuted auong the paiticlpants before ihe meetlng. -iotro*r.,g ,presentatlon of the model acructure and the flrsc test results, Eheoeet'lng dlscussed che model and lt,s potentlal use ln some det.il, and thlsresulced ln a number of recommendaclons. Reegsrueldatlons on modifleatlone, of the vector-paraslEe nodel and on iEsuttrrzaiionE-tEE =:'usr o" - - - In the currenE model lt ls assumed thac all survtvtng larvae ln the fly wltlulcimarely become an infect,lve 1arvae ln the head (L3H),.ia-at"a Ehls ls cheonly stage whlch can be cransmltted. However, from entomologrcalobeervaElons, lt can be concluded-thar 13 scaie rarvae rn thI body (L3B) mayalso be transmlEted as lnfectlve larvae. Furfhermore, there ls no reason whyL3 rarvae should not be able to move co the head and vlce versa. Therefore,the present translcory L3B stage and che flnal L3H stage stouid be dlfferenrlysubdlvlded lnto 3 cat,egorles, ,r1r., l. L3B in the nuscles of the body, 2.L3B free ln the body (lnfective) and 3. LiH-i."" in the head (1nfecr,1ve). - one baslc llmrtaclon in the model ls the assumptron that afEer rwobloodmeals followlng che engorgement of MF atl survlvlng larvae have reachedthe L3-sEage' Thls assuoptlon-nlght be quesclonable because xenodlagnostlcexperlmenEs have shown thac larvat aevetopnent ln the flles can beconsiderably slowed down by low t,emperatures. However, lc ls nog expecgedthat thls facEor wlll slgnlflcantly affect che obserr"i .ntorologlcar daEa,and, anyhow, lt ls noE feaslble co adapt the present nodel ln thls reapecg. - A remarkable dlfference beEween some tesc-runs of the oodel and fleldobservaElons concerns Ehe number of Ll and L2 larvae ln the flles Ehac havebeen dlssecEed after a caEchlng sesslon. rn the model many more Ll/2 larvaeare carculaced th-an,were really found ln the flies. ft ls assumed chac chlslack of observed Ll/2 musc be iue co the dlssectlon method in whlch much moreacEenclon ls paid Eo larvae ln the head than to larvae ln che body and flleswlchout lnfectlve larvae are not furEher lnspecced for Ll/2 larvae. Thlsassumptlon ls a'o. support,ed by research data from Togo, the wesEern extensionand by data garhered in Sudan, where abouE 3x as aany-t i/Z were found as lnthe oCP-data. It ts expected that ln the l.IesEern- extenslon, due co moreadvanced dlssectlon meEhods (stalnlng), the presence of the [t /2 stage wlll bedeCected much bet,Eer. - rn order to allow a beEEer comparlson between Ehe slmulat,ed resurcs and Eheobserved daEa on doubre lnfecttons, the moder output wlrl separate theresult's for L2 larvae from Ehls sectlon and only provlde scaEl6glcs on Lland L3 larvae for double infectlons. - rnstead of modelllng an t'Ll-1ntake", {t would be better co include a modulefor the slmulaclon of real MF-lntake and the proporclon of Ehose mlcrofllarlaewhlch pass the gutwall and become classlfled as Lt larvae. Thls extra modulecould be quanElfled uslng Ehe resulcs of Ehe ongolng xenodlrgio"cr. experlments and enable che lntegraElon of chls tnfoimarlon ii the analysts ofentomologlcal data. AnoEher advantage of chls modtflcatlon """ia be moreflexlblllcy ln studylng Ehe effect, of varlatlons ln endemlcliy levels on theencomologlcal results. Ehe current model would beIt was proposed to modlfy the lncluslon ofthe nodel ln - One of che maJor tmprovemenEs on t,he process of mlgratlon of flles. Appendlx III, page 2 auch a way chat Ewo fly-populat,lone could be slmulated ln one run. In chatway lt would be posslble co mlx a local fly-populatton wlrh an immtgrarlngfly-populaclon of varlable elze and age arstitlucton. It was recogntzed thatthls was stlll an enormous overslmpltflcaclon of the act,ual phenomenon ofblack-fly ntgraclon buE lE was agreed that oore detalled nodelllng would notbe reallsctc and ehould not be attenpt,ed. , Appenclix IV, page I SPECIFTCATIOrIS POB. A TNAiISHISSIOf, I{ODEL This appendix provldes a ltst of che factors whlch should be caken lnco accounE ln a full transmleslon nodel of onchocerclasls. IE should be used nconjunctlon wlEh che tnlclal llst provlded durlng rhe 4ch meeE,lng on appllednodelllng for onchocerclasls, whlch wae held ln decenber l9g5 ln RoEterdam.This ltst ls only lntended as a checkllst for nodel developmenE, and does not)et touch the quesElon of quanElflcatlon of the varlous p"i"r"c".s llsEed. FACTORS TO BE INCLUDED IN A TRANSMISSION MODEL: - Human populaclon dytramlce: agelng, b1rth, death, lnmtgrarlon of persons wlthepeclflc worn loade.(refer H/p nodel) - Potentlal of local breedlng and Ehe aasoclated reproductlon capaclcy of thefl5populailon. seasonal l.ct"rr," in reproductlon capaclty. (refer explorat,or y trans .modeL ) - Immlgratlon of flles (refer update V/p nodel) - Relaclon beEween personal bltlng rat,e and Ehe slze of che human popuratlonln dependence of Ehe capaclty oi che fl5breedlng stte and seasonar Erends. - Age-, sex- (and behaviour-?) dependent exposure. - Human hosr cholce probablrlry (dependent of fry specles) - MF-lntake by che flles ae a functlon of the mf denelcy ln the human hosE. - Parous raee of flles; flysurvlval. (refer V/p model) - Varlablllry in MF-intalce - Relatlon L3H released/MF engorged (non-llnear, includtng excess-morrallty,refer revlsed V/p model). - Incldence of new lnfectlons and blclng fl-y. ln relatlon t,o parasiElc sEat,us of human hosC - Immunologlcal defenee lnfectlons. ln relatton to past-hlotory of lncldence of - Matlng probability (lmportanc in slEuacion wlth low worm-loads). - D'evelopnent of worns, morcalrcy of worms (refer H/p moder) - ltF-producElon of worms (ln dependence of worm-age?, refer H/p moder). - Development of anterlor and of posterlor leeions ln dependence of the(hlscory of) MF-toad. - Development of eye-leslons, possdlblllty of regreselon (ancertor), MF-dependent progresslon (ant,erior), .,rcor,irous piog.esslon (posEerlor). - FracElon of the human populatlon wlth lrnmunlcy agatnst, bllndness. - Reduced llfe-expecEancy of bltnd people (rn Eerms or excess mo,.t?.rl.y race,refer 11tt). . * varlable coverage * varlable lnter-appllcatlon perlod - Larvlclde appllcatlon: Appendlx IV, page 2 * Beason dependence * costs (per appltcatton + overhead). - Trlggers for larviclde-appllcaclon: Blclng rat,e Transmlss lon poEent lal Estlnated IBR. - Ivermect,ln: * coverage (blanket treat,ment) * trend ln MF (skln and eye separately) followlng * cumulative effect of successlve doses ' * (posslble) effect on MF-tnt,ake and larval ,the fly. - IvermecEln admlniet,raElon: * lndlvldual monltorlng and t,reaEnenc * blanket treatment, of a village * blanket treatnenc of a rleh-group (endemrclty subpopulaclon) * excluston of persons wlCh contra-indlcat,ions * co8ts. - Trlggers for ivermectln treatment: * lnrenslty of lnfect,ion (CMFL) * prevalence ln populatlon based on elcln snlps (or lnmunologlcal EesE).* dlsease status of the populatlon based on ophthalnologicar surveys orlnEerviewfng (?) * prevalence tn a rtsk group wlthtn the populat,lon. - Macrofllarlclde: slmpLe provlelon (low prlorlty) treaEment developmenE in Appendlx Vt B. Phlllppon D.G. Zerbo J. Grunewald F. ['lalsh D. Kur t.a k R. Baker G. De Sole M. Karam Y. Dadzle J.B. Zongo J.H. Renme A.E. Souobey J.D.F. Habbema K. Dletz G.J. Van Oortnarsaen A.P. Plalsler LIST OF PARTTCIPANTS (vcu) (vcu )(vcu) (vcu )(vcu) (vcu) (EPT) (EPI ) ( EPT)(Eco)(srAr)(srAr)(External, Unlverslcy of(Excernal, Unlverslty of(External, Unlverslty of(External, Unlverslty of Rotterdan) TUblngen) Rotterdan) Rotcerdam) Appendlx VI a BACKGROUND I'IATERIAL. -Applled nodelllng Report of the Report of the Report of the Report, of the as an ald ln analyels and plannlng for OCP': Eeet,tng 1n Tllblngen, 29-30 June 1984 rueeting ln Rott,erdam, 27-28 Sepcember 1984 oeetlng ln Ouagadougou, 25-28 March 1985 neetlng ln Rotterdan, 16-18 Decerober 1985. 'Applled modelllng; progress report, and propoeale for furcher research-(OCP/EAC6.l0, Ouagadougou, June 1985)o - Meoorandun on 'l.leetlng on applled nodelllng , lg-22 Januar y l9B7'(ocP, 5-l-87). G. van Oortmarsaen, J. Dlk, F. Habbena and M. OsEendorf : '-ONCHOSII.|', A compuEer slnulatlon program for slnulat,ing the llfe-hlstory of onchocerca volvurue- (Technlcal reporE, Rot,Eerdao, June r.985). A. Plalsler, G. van oorEnarssen and J.D.F. Habbema: '-vECpARl.o-, A computer Program for sloulatlng the llfe-c1rcles of Slnullum damnosum8.1. and Onchocerca volvuLus, and thelr lnteractlon- (Technlcal report, RoEcerdan, Seprember t986). K. DleEz: 'The populatlon dynamlcs of onchocerclasls- ln: R.M. Anderson: -Populatlon dynamlcs of lnfectlous dleease', London, 1982. K. Dtetz and H. Renner: 'SlmulaElon of selecttve chemotherapy for t,he control of helmtnth dlseases'. J.Remne, O.BA, K.Y.Dadzle and M.Karam: -A Force-of-Infectlon ruodel for onchocerclasls and lcs appllcatlone in the epldemlologlcal evaluatlon of the Onchocerctasls Control Programe ln the Volta Rtver Basln area.-Bulletln of rhe World Healch Organlzacion, 64: 667-681 (19g6).