ryngeal515hh

OriginalarticleDemographicandbehaviouralproleofadultsinfectedwithchlamydia:acase-controlstudyKWRadcliVe,SAhmad,GGilleran,JDCRossObjectives:Todeterminewhichdemographicandbehaviouralparametersareindependentlyassociatedwithchlamydialinfectioninadults.
Methods:Subjectswererecruitedprospectivelyfrommaleandfemaleattendeesatalargeclinicforsexuallytransmittedinfections(STI).
Allsubjectsweretestedforchlamydiaandgonorrhoeaandaskedtocompleteaquestionnaireaddressingdemography,sexualandnon-sexual(includingdrugtaking)behaviour,andhistoryofSTI.
Caseswerethoseattendingwithanewclinicalepi-sodeandfoundtobeinfectedwithchlamydia,butwhodidnothavegonorrhoea.
Acontrolgroupwasselectedrandomlyfromthosefoundtobenegativeonscreeningforbothinfections.
Results:986casesand1212controlswererecruitedoveronecalendaryear.
Thefollowingwerefoundtobeindependentriskfactorsforchlamydialinfectiononmultivariateanalysis(oddsratioswith95%condenceintervalsinparentheses):beingunmarried(1.
8;1.
1–3.
1);blackCaribbeanethnicity(2;1.
5–2.
7).
Increasingage,fewerpartners,andhigherreporteduseofcondomswereassociatedwithalowerriskofinfection.
Conclusion:BlackCaribbeansareatincreasedriskfromchlamydiaaftercontrollingforsexualbehaviourandsocioeconomicstatus.
Futureresearchshouldseekanexplanationelsewhere—forexample,intermsofdiVerencesinsexualmixingoreVectivenessofhealthcareinterventions.
(SexTransmInf2001;77:265–270)Keywords:chlamydia;epidemiology;sexualbehaviourIntroductionGenitalinfectionwithChlamydiatrachomatisisanimportantsexuallytransmittedinfection(STI)whichcanresultinpelvicinfectionwithresultantchronicpelvicpain,tubalinfertilityandincreasedriskofectopicpregnancy;sexuallyacquiredreactivearthritis;epididymo-orchitis;conjunctivitisandbronchopneumoniainbabiesborntoinfectedmothers.
1Theidenticationofriskfactorssignicantlyassociatedwithgenitalchlamydialinfectionmayaidintheunderstandingofthetransmis-siondynamicsofthisSTIandalsohelptodevelopinterventionstargetedatthosemostatrisk.
VariousstudiesconductedinSTIclinicsaswellasingeneralpracticeintheUnitedKingdomhaveshownthatanumberofdemo-graphicfactorsareassociatedwithincreasedriskofinfectionwithchlamydia,includingfemalesex,23youngadultage,4particularlyteenagegirls,256andblackethnicorigin.
457Behaviouralriskfactorsincludethosedirectlydescribingaspectsofsexualbehaviour,andotherswhich,althoughnotdirectlyrelatedtosexualbehaviour,mayrepresentsurrogatemarkersofhighrisksexualbehaviour,suchasselectionofhighriskpartners.
Directsexualbehaviourriskfactorsincludeanincreasednumberofsexualpartners4–6andnon-useofbarriermethodsofcontraception,suchascon-doms.
8OtherriskfactorsforchlamydiahavebeendemonstratedonlyinWesterncountriesotherthantheUnitedKingdom.
Theseincludenewsexualpartners(UnitedStates9);hetero-sexualorientation(Australia,10Netherlands,11andCanada12);smokingcrackcocaine(UnitedStates13);andcigarettesmoking(Sweden14).
AdditionalriskfactorshavebeenshowntobeassociatedwithotherSTIbutfewdataexistwithrespecttochlamydia—increasedalcoholconsumptionandgonorrhoeainYugoslavia,15andwithSTIingeneralintheUnitedStates,16previousgonorrhoeaorotherSTIwithgonor-rhoeaintheUnitedKingdom,17Netherlands,11andtheUnitedStates.
18ToourknowledgeonlyonestudyhasreportedonriskfactorsassociatedwithchlamydialinfectionfromaUKcentreotherthaninLondon,andthiswasaretrospectivestudywhichincludednobehaviouraldata.
19Wethereforeundertookastudytoinvesti-gatewhichdemographicandbehaviouralriskfactorsareassociatedwithchlamydialinfectionamongattendeesatalargeSTIclinicinthecentreofBirmingham,UnitedKingdom(populationapproximatelyonemillion).
MethodsEthicalapprovaltoconductthestudywasobtainedfromthelocalresearchethicscom-mittee.
AllmaleandfemalepatientsattendingtheWhittallStreetClinicinBirminghamduringthecalendaryear14June1997to13June1998inclusivewereeligibletotakepartinthestudyiftheypresentedasanewclinicalepisodeandacceptedscreeningforgonococcalandchlamy-dialinfection.
TestingforChlamydiatrachoma-tiswasdoneusingligasechainreaction(LCX,Abbott)onaurethralswaborurinesampleinmales,andonacervicalswabinfemales.
Test-ingforNeisseriagonorrhoeaewasdonebyplatingswabsdirectlyontoselectiveculturemedia:swabswereroutinelytakenfromtheSexTransmInf2001;77:265–270265WhittallStreetClinic,Birmingham,UKKWRadcliVeSAhmadGGilleranJDCRossCorrespondenceto:DrKWRadcliVe,WhittallStreetClinic,WhittallStreet,BirminghamB46DH,UKKeith.
RadcliVe@bscht.
wmids.
nhs.
ukAcceptedforpublication22May2001www.
sextransinf.
comonDecember25,2020byguest.
Protectedbycopyright.
http://sti.
bmj.
com/SexTransmInfect:firstpublishedas10.
1136/sti.
77.
4.
265on1August2001.
Downloadedfromurethrainmalepatientsandfromboththeurethraandthecervixinfemalepatients.
Pha-ryngealandrectalcultureswerealsoobtainedifindicatedbythesexualhistory.
CasesweredenedassubjectsfoundtobeinfectedwithchlamydiabutnotwithNgonor-rhoeae.
Asimilarnumberofcontrolswereselected,usingrandomnumbertables,fromthosepatientsfoundtobefreeofbothinfections.
Alleligiblepatientsweregivenastructuredquestionnaireonarrivalattheclinicreception,whichtheywereaskedtocompleteontheirownandreturnbeforeleaving.
Thequestion-naireincludedquestionson:+demographicattributes—age,sex,maritalstatus,selfidentiedethnicgroup,andoccupationalclassication(asasurrogatemarkerofsocioeconomicstatus)+sexualbehaviour—numberofselfdenedregularandcasualpartnersintheprecedingyear,orientation,useofcondoms,numberoflifetimeforeignpartners+recentuseoftobacco,alcohol,andillicitdrugs+selfreportedpreviousdiagnosesofgonor-rhoeaandchlamydia.
ThedatafromthequestionnaireswereenteredintoanAccess97database(Microsoft,UK)beforetransfertoSPSS(v.
10,SPSSInc)foranalysis.
Aunivariatecomparisonofpatientswithchlamydiainfectionandcontrolpatientswasperformedusing2test.
Variableswithasignicancelevellessthan0.
1onunivariateanalysiswereenteredintoamultivariatelogis-ticregressionmodelusingaforwardcondi-tional(likelihoodratio)method.
Apoweranalysis(EPI-INFOv,5.
01a,CDC,Atlanta,GA,USA)suggestedthattodetectanoddsratioof1.
5foravariablewithaprevalenceof20%(power80%,condencelevel95%)wouldrequireasamplesizeof1124(562ineachgroup).
Toallowformissingdatawithinthestudygroupsarecruitmenttargetof2000wasset.
ResultsAtotalof986patientswithchlamydiaand1212controlpatientswereenteredintothestudy.
Thisrepresentsarecruitmentrateof83%ofthe1181chlamydiacasesseenoverthestudyperiod.
Thecharacteristicsofthestudypopulationandcompletenessofthedatasetareshownintable1.
Onunivariateanalysis(table2)allthevariableswerefoundtobesigni-cantlyassociatedwithchlamydialinfectionexcepthavingahistoryofgonorrhoeaorchlamydia.
Onlyonevariableonunivariateanalysishadasignicancelevelgreaterthan0.
1(historyofchlamydialinfection)andthiswasexcludedfromthemultivariateanalysis.
Alargeproportionofpatientswithnocasualpartnershadnotrepliedtothequestionaboutcondomuseinthissituationandthisvariablewasthereforealsoexcludedfromthemultivari-ateanalysis.
Datafrom1371subjects(526casesand845controls)whohadacompletedatasetwereenteredintoamultivariatelogisticprogressionmodel.
Sevenvariableswerefoundtobesignicantlyassociatedwithchlamydialinfec-tion(table3):+Malesexwasassociatedwithanincreasedriskofinfection+Ageover20wasassociatedwithalowerriskofinfection+Havingfewerpartnerswasassociatedwithalowerriskofinfection+AlwaysusingcondomswasassociatedwithalowerriskofinfectionTable1CharacteristicsofthestudypopulationVariableMaleFemaleAgeUnder2088(8%)253(23%)20–24261(24%)332(30%)25–30316(29%)258(23%)Over30419(39%)271(24%)MaritalstatusSingle910(84%)944(85%)Married103(10%)85(8%)Other*57(5%)68(6%)Noreply14(1%)17(2%)EthnicgroupWhite589(54%)659(59%)BlackCaribbean367(34%)295(26%)Other113(10%)131(12%)Noreply15(1%)29(3%)Totalnoofpartners(pastyear)None67(6%)21(2%)1194(18%)423(38%)2185(17%)236(22%)3ormore380(35%)242(22%)Noreply258(24%)192(18%)Noofcasualpartners(pastyear)None349(32%)646(58%)1215(20%)162(14%)2ormore237(22%)88(8%)Noreply283(26%)218(20%)CondomusewithallpartnersNever187(17%)239(22%)Sometimes454(42%)563(50%)Always103(10%)90(8%)Noreply340(31%)222(20%)CondomusewithcasualpartnersNever151(14%)95(8%)Sometimes160(15%)69(6%)Always170(16%)131(12%)Noreply603(56%)819(74%)HistoryofgonorrhoeaNo805(74%)802(72%)Yes117(11%)84(8%)Noreply162(15%)228(20%)HistoryofchlamydiaNo802(74%)740(66%)Yes118(11%)154(14%)Noreply164(15%)220(20%)HistoryofforeignpartnerNo831(77%)781(70%)Yes214(20%)226(20%)Noreply39(4%)107(10%NoofsamesexpartnersNone758(70%)872(78%)1ormore68(6%)15(1%)Noreply258(24%)227(20%)AlcoholNone347(32%)344(31%)1–5units/week216(20%)425(38%)>5units/week505(47%)326(29%)Noreply16(2%)19(2%)SmokingNone250(24%)200(18%)1–5/day464(44%)537(49%)Over5/day349(33%)364(33%)Noreply21(2%)13(1%)IllicitdruguseinpastyearNo489(45%)648(58%)Yes345(32%)268(24%)Noreply250(23%)198(18%)OccupationalgroupHomemaker0(0%)108(10%)Student118(11%)232(21%)Unemployed188(17%)121(11%)Working512(47%)439(39%)Noreply266(24%)214(19%)*Divorced,separated,livingtogether,widowed.
266RadcliVe,Ahmad,Gilleran,etalwww.
sextransinf.
comonDecember25,2020byguest.
Protectedbycopyright.
http://sti.
bmj.
com/SexTransmInfect:firstpublishedas10.
1136/sti.
77.
4.
265on1August2001.
Downloadedfrom+Beingsinglewasassociatedwithanin-creasedriskofinfection+BlackCaribbeanethnicitywasassociatedwithanincreasedriskofinfection+Smokinglessthanvecigarettesperdaywasassociatedwithareducedriskofinfectioncomparedwithhighsmokingrates,butnon-smokersalsohadanincreasedriskofinfec-tion.
Toinvestigatepotentialinteractionswithsex,age,andethnicityseparatelogisticregressionanalyseswereperformedonthefollowingsubgroups—men,women,ageunder25,age25andover,blackCaribbeanethnicity,whiteethnicity(tables4–6).
Inwomentherewasanassociationwithinfectioninthoseclassiedashomemakerscomparedwiththoseunem-ployedorworking.
Condomuseandmaritalstatuswerenotassociatedwithinfectioninthisgroup.
Formalepatientsillicitdrugusewasassociatedwithanincreasedriskofinfection,butagewasnolongerasignicantvariable.
Thoseundertheageof25(table5)nolongerhadanassociationbetweenmaritalstatusorsexwithinfection,whilenumberofpartners,condomuse,smoking,andethnicityremainedsignicant.
Inthoseaged25oroverthetotalnumberofpartners,ethnicity,andmartialsta-tuswereassociatedwithinfection.
Inwhitepatientsahistoryofoneormoresamesexpartnerswasassociatedwithareducedriskofinfectionandsignicantassociationswerealsoseenfortotalnumberofpartners,condomuse,smoking,sex,andage.
InblackCaribbeanpatientsdrinkingonetoveunitsofalcoholperweekwasassociatedwithareducedriskofinfectioncomparedwithnon-drinkers,andtheonlyothersignicantvariablewasage,demon-stratingareducedriskinthoseover20.
DiscussionThislargeprospectivestudyhasdemonstratedblackCaribbeanethnicitytobeanindepend-entpredictorofchlamydialinfectioninbothmenandwomen,andinthoseagedunder25and25orover.
Thisaddstoearlier,retrospec-tivestudiesconductedinLeeds20andLondon21showinganincreasedrateofgonorrhoeainblackethnicgroupsindependentofsocioeco-nomicstatusandasimilarndinghasalsobeenreportedforchlamydiafromCoventry.
19OnepreviousUKstudy,performedinaLondonSTIclinic,hasreportedanincreasedriskofgonorrhoeaandchlamydiainblackmeninde-pendentofbothsocioeconomicstatusandsexualbehaviour.
7Thepresentstudyalsosuggeststhatyoungadults(thatis,teenagers)arethegroupmostatriskfromchlamydiaandTable2UnivariateanalysisofdemographicandbehaviouralfactorswithchlamydialinfectionVariableChlamydiaControls2pvalueSexFemale471(48%)643(53%)0.
014Male515(52%)569(47%)AgeUnder20214(22%)127(10%)5units/week302(31%)529(44%)SmokingNone358(37%)92(8%)comonDecember25,2020byguest.
Protectedbycopyright.
http://sti.
bmj.
com/SexTransmInfect:firstpublishedas10.
1136/sti.
77.
4.
265on1August2001.
Downloadedfromconrmsthattheseindividualsareanimpor-tanttargetforhealtheducationinterventionstoencouragebehaviouralchange.
Theassociationbetweeninfectionwithincreasingnumberofpartners,decreasedcondomuse,andbeingsingleareconsistentwithknownriskbehav-ioursfortheacquisitionofinfection.
Infectionwasfoundtobeassociatedwithbeingmale,despitethelikelihoodthatinfectioninwomenresultsinalargeasymptomaticreservoir.
Sincethepopulationwasselectedfromclinicattendersitwouldbeunwisetoextrapolatethisndingtothegeneralpopula-tionhowever.
Individualswhosmokemaybethoughtofasbeing"risktakers"andanassoci-ationwithinfectionwasindeedfoundinthosewhosmokemoreheavily.
However,anin-creasedriskwasalsoseeninnon-smokerswhichwouldnottwiththishypothesis.
Thecondenceintervalsforthisndingwere,how-ever,verywideandneedtobeinterpretedwithcaution.
Althoughperformedprospectivelythecase-controldesignofthisstudyhasintrinsiclimita-tionswhichmayaVecttheinterpretationoftheresultsandtheirapplicabilitytothegeneralpopulation.
Boththecasesandcontrolsweretakenfromasexuallytransmitteddiseaseclinicpopulation,representingahighriskgroupforinfection.
Thecontrolgroupmaythereforehavehadahigherriskofunsafesexualbehaviourthanthegeneralpopulation.
ItishoweverlikelythatthiswouldreducethediVerencebetweencasesandcontrols,ratherthanfalselyelevatingit.
Thenon-responseratesareshownintable2andvarywithindividualquestions.
Thenon-responderswereexcludedfromthemultivariateanalysiswhichraisesthepossibilityofbias,particularlyforthosecharacteristicswithhighnon-responserates—totalnumberofpartners,numberofcasualpartners,condomusewithallpartners,condomusewithcasualpartners,illicitdruguseinthepastyear,andoccupationalgroup.
ThepossibilityofrecallbiaswasreducedbyensuringthatallparticipantscompletedthequestionnairebeforetheirconsultationwiththemedicalstaV.
Itshouldalsoberecognisedthattheuseofoccupationalclassprovidesonlyacrudeestimateofeconomicstatusandraisesthepossibilityofidentifyingadditionalassocia-tionsifamorediscriminatingclassicationwasused.
Table4Multivariatelogisticregressionanalysisofvariablesassociatedwithchlamydiainfection—subanalysisbysex*VariableOR(95%CI)Women(n=747)OccupationalgroupHomemaker1Student0.
5(0.
3–1)Unemployed0.
4(0.
2–0.
9)Working0.
4(0.
3–0.
7)AgegroupUnder20120–240.
3(0.
2–0.
5)25–300.
3(0.
2–0.
4)Over300.
1(0.
07–0.
2)Totalnumberofpartners00.
4(0.
03–4.
9)10.
4(0.
3–0.
6)20.
6(0.
4–0.
9)Over21SmokingNone12(1.
4–98)1–5/day0.
7(0.
5–1.
04)Over5/day1EthnicgroupWhite1BlackCaribbean1.
9(1.
2–2.
8)Other1.
4(0.
9–2.
4)Men(n=624)IllicitdruguseinpastyearYes1.
5(1.
1–2.
2)Totalnumberofpartners00.
001(NS)10.
5(0.
3–0.
8)20.
8(0.
5–1.
2)Over21CondomusewithallpartnersNever1Always0.
4(0.
2–0.
8)Sometimes/often0.
9(0.
6–1.
3)SmokingNone1180(NS)1–5/day0.
8(0.
6–1.
1)Over5/day1EthnicgroupWhite1BlackCaribbean2(1.
4–3.
1)Other1(0.
5–1.
9)MaritalstatusSingle3.
4(1.
6–7.
3)Married1Other0.
8(0.
2–3.
1)*Variablesinthemodel:age,maritalstatus,ethnicgroup,totalnumberofpartners,condomusewithpartners,historyofgon-orrhoea,historyofforeignpartner,numberofsamesexpartners,alcohol,historyofillicitdruguse,occupationalgroup,smoking,numberofcasualpartners.
Divorced,separated,livingtogether,widowed.
Table5Multivariatelogisticregressionanalysisofvariablesassociatedwithchlamydiainfection—subanalysisbyage*VariableOR(95%CI)Ageunder25(n=628)Totalnumberofpartners00.
6(0.
05–7.
4)10.
4(0.
3–0.
6)20.
5(0.
3–0.
8)Over21CondomusewithallpartnersNever1Always0.
3(0.
2–0.
7)Sometimes/often0.
7(0.
4–1.
1)SmokingNone8(1–65)1–5/day0.
6(0.
4–0.
9)Over5/day1EthnicgroupWhite1BlackCaribbean2.
3(1.
4–3.
5)Other1.
2(0.
7–2.
1)Age25orover(n=743)Totalnumberofpartners00.
001(NS)10.
5(0.
3–0.
8)20.
9(0.
6–1.
4)Over21SmokingNone1407(NS)1–5/day0.
8(0.
6–1.
1)Over5/day1EthnicgroupWhite1BlackCaribbean1.
9(1.
3–2.
8)Other1.
5(0.
9–2.
7)MaritalstatusSingle2.
2(1.
2–4)Married1Other0.
8(0.
3–1.
9)*Variablesinthemodel:sex,maritalstatus,ethnicgroup,totalnumberofpartners,condomusewithpartners,historyofgon-orrhoea,historyofforeignpartner,numberofsamesexpartners,alcohol,historyofillicitdruguse,occupationalgroup,smoking,numberofcasualpartners.
Divorced,separated,livingtogether,widowed.
268RadcliVe,Ahmad,Gilleran,etalwww.
sextransinf.
comonDecember25,2020byguest.
Protectedbycopyright.
http://sti.
bmj.
com/SexTransmInfect:firstpublishedas10.
1136/sti.
77.
4.
265on1August2001.
DownloadedfromAlternativefactors,suchasdiVerencesinsexualmixingpatternsorintheeVectivenessofhealthcareinterventions(forexample,partnernoticationandfollowup),needtobeexploredtoseekanexplanationforthesendings.
Unmarriedstatuswasfoundtobeanindependentriskfactoronmultivariateanaly-sis.
Toourknowledgethisisanoriginalnding—smallercontrolledstudiesfromtheUnitedKingdom,6UnitedStates,9Sweden,14andAustralia22failedtondthisassociation.
Theonlyassociationfoundwithsocio-economicstatus,asmeasuredbythenatureofemployment,wasinwomenwhowereatincreasedriskofinfectionwhentheywere"homemakers.
"Thereasonforthisisnotclearalthoughotherstudieshavealsofoundlittleassociationwithsocioeconomicstatus.
1423–26However,itshouldbenotedthatinthispresentstudyarelativelyimprecisemeasureofsocio-economicstatusintheformofemploymentwasutilised.
Illicitdrugmisusewasfoundtobeassociatedwithanincreasedriskofchlamydialinfectioninmen.
Thismayreectanincreasedpredispositiontorisktakingbehaviourandalsoraisesthepossibilityofincreasedhighriskbehaviourwhileundertheinuenceofillicitdrugs.
Theriskfactorsforchlamydialinfectionforthoseagedunderandover25weresimilaralthoughcondomusewasonlysignicantfortheyoungerpatientsandmaritalstatusforthoseagedover25.
InblackCaribbeanpatientstheonlyfactorssignicantlyassociatedwithchlamydiawereageunder20andbeingteeto-tal.
Thiswashoweverthesmallestsubgroupanalysis(n=316),raisingthepossibilitythatalargestudypopulationwithgreaterpowermayidentifymoresignicantvariables.
Inrelationtosexualbehaviourandinagree-mentwithpreviousresearch,thisstudyshowsthatanincreasingnumberofsexualpartnersisariskfactorforinfectionbutthathomosexual-ityandconsistentcondomuseareprotective.
Perhapscounterintuitively,anincreasednumberofcasualsexualpartners(denedaspeoplewithwhomtherewasasinglesexualencounter)didnotincreasetheriskofbeinginfected.
ThisndinghasbeenpreviouslyreportedinastudyconductedinaLondonSTIclinic.
8Althoughoverallconsumptionofalco-holdidnotcorrelatewithchlamydialinfection,useofillicitdrugsdid.
Toourknowledgethisassociationhasnotpreviouslybeendemon-strated.
Theresultsofthisresearchhaveimplicationsforsexualhealthpromotionactivities.
Infor-mationconcerningchlamydiaanditspreven-tionshouldbetargetedtothosemostatrisk—namely,young,unmarriedheterosexualsandpeopleofblackCaribbeanethnicity.
Healtheducationmessagesshouldstressthatriskofinfectionriseswithincreasingnumberofsexualpartnersandshouldseektodispelthewidelyheldviewthatcasualpartnersarethesourceofgreatestrisk.
Thesoberingndingthat44%ofcasesand30%ofcontrolsinthisstudyadmittedtousingillicitdrugsinthepre-vious12monthsshowsclearlythathealthedu-cationaimedatSTIclinicattendeesshouldaddressthisissueaswellasthepreventionofinfection.
1StammWE.
Chlamydiatrachomatisinfectionsoftheadult.
In:HolmesKK,MardhP-A,SparlingPF,etal,eds.
Sexu-allytransmitteddiseases.
3rded.
NewYork:McGraw-Hill,1999:407–22.
2SimmsI,CatchpoleM,BrughaR,etal.
EpidemiologyofgenitalChlamydiatrachomatisinEnglandandWales.
GenitourinMed1997;73:122–6.
3MatondoP,WallR,MorganK,etal.
Epidemiologyofgono-coccalandchlamydialinfectionsinHarrowandBrent.
GenitourinMed1996;72:352–7.
4OakeshottP,KerryS,HayS,etal.
Opportunisticscreeningforchlamydialinfectionattimeofcervicalsmeartestingingeneralpractice:prevalencestudy.
BMJ1998;316:351–2.
5EvansBA,TaskerT,MacRaeKD.
Riskprolesforgenitalinfectioninwomen.
GenitourinMed1993;69:257–61.
6GrunL,Tassano-SmithJ,CarderC,etal.
Comparisonoftwomethodsofscreeningforgenitalchlamydialinfectioninwomenattendingingeneralpractice:crosssectionalsur-vey.
BMJ1997;315:226–30.
7EvansBA,BondRA,MacRaeKD.
Racialorigin,sexualbehaviour,andgenitalinfectionamongheterosexualmenattendingagenitourinarymedicineclinicinLondon(1993–4).
SexTransmInf1998;74:40–4.
8EvansBA,KellPD,BondRA,etal.
Heterosexualrelationshipsandcondom-useinthespreadofsexuallytransmitteddiseasestowomen.
GenitourinMed1995;71:291–4.
9HandseldHH,JasmanLL,RobertsPL,etal.
CriteriaforselectivescreeningforChlamydiatrachomatisinfectioninwomenattendingfamilyplanningclinics.
JAMA1986;255:1730–4.
10HartG.
Riskprolesandepidemiologicinterrelationshipsofsexuallytransmitteddiseases.
SexTransmDis1993;20:126–36.
11VanDuynhovenYTHP,vandeLaarMJW,SchopWA,etal.
DiVerentdemographicalandsexualcorrelatesforchlamy-dialinfectionandgonorrhoea.
IntJEpidemiol1997;26:1373–85.
12VinceletteJ,BarilJG,AllardA.
Predictorsofchlamydialinfectionandgonorrhoeaamongpatientsseenbyprivatepractitioners.
CanMedAssocJ1991;144:713–21.
Table6Multivariatelogisticregressionanalysisofvariablesassociatedwithchlamydiainfection—subanalysisbyethnicity*VariableOR(95%CI)White(n=913)Totalnumberofpartners00.
3(0.
03–3.
3)10.
4(0.
3–0.
6)20.
6(0.
4–0.
9)Over21Noofsamesexpartners1ormore0.
2(0.
04–0.
8)CondomusewithallpartnersNever1Always0.
5(0.
3–0.
8)Sometimes/often0.
7(0.
5–0.
9)SmokingNone11(1.
3–91)1–5/day0.
6(0.
5–0.
8)Over5/day1SexMale1.
4(1.
03–1.
9)AgegroupUnder20120–240.
4(0.
2–0.
6)25–300.
3(0.
2–0.
5)Over300.
2(0.
1–0.
3)BlackCaribbean(n=316)Noofsamesexpartners1ormore2047(NS)SmokingNone1477(NS)1–5/day1.
1(0.
6–1.
9)Over5/day1AlcoholNone11–5units/week0.
4(0.
2–0.
8)>5units/week0.
5(0.
2–1.
1)AgegroupUnder20120–240.
4(0.
2–0.
8)25–300.
4(0.
2–0.
8)Over300.
2(0.
1–0.
4)*Variablesinthemodel:sex,age,maritalstatus,totalnumberofpartners,condomusewithpartners,historyofgonorrhoea,his-toryofforeignpartner,numberofsamesexpartners,alcohol,historyofillicitdruguse,occupationalgroup,smoking,numberofcasualpartners.
Divorced,separated,livingtogether,widowed.
Demographicandbehaviouralproleofadultsinfectedwithchlamydia269www.
sextransinf.
comonDecember25,2020byguest.
Protectedbycopyright.
http://sti.
bmj.
com/SexTransmInfect:firstpublishedas10.
1136/sti.
77.
4.
265on1August2001.
Downloadedfrom13EdlinBR,IrwinKL,FarqueS,etal.
Intersectingepidemics-crackcocaineuseandHIVinfectionamonginner-cityyoungadults.
NEnglJMed1994;331:1422–7.
14JonssonM,KarlssonR,PerssonK,etal.
TheinuenceofsexualandsocialfactorsontheriskofChlamydiatrachomatisinfections:apopulation-basedserologicstudy.
SexTransmDis1995;22:355–63.
15BjekicM,VlajinacH,SipeticS,etal.
Riskfactorsforgonorrhoea:case-controlstudy.
GenitourinMed1997;73:518–21.
16ShaferM-A,HiltonJF,EkstrandM,etal.
Relationshipbetweendruguseandsexualbehaviorsandtheoccurrenceofsexuallytransmitteddiseasesamonghigh-riskmaleyouth.
SexTransmDis1993;20:307–13.
17SherrardJ,BarlowD.
Menwithrepeatedepisodesofgonor-rhoea1990–1992.
IntJSTDAIDS1996;7:281–3.
18SchwarzSK,BolanGA,FulliloveM,etal.
Crackcocaineandtheexchangeofsexformoneyordrugs:riskfactorsforgonorrhoeaamongblackadolescentsinSanFrancisco.
SexTransmDis1992;19:7–13.
19WinterAJ,SriskandabalanP,WadeAAH,etal.
Sociodemog-raphyofgenitalChlamydiatrachomatisinCoventry,UK,1992–96.
SexTransmInf2000;76:103–9.
20LaceyCJN,MerrickDW,BensleyDC,etal.
AnalysisofthesociodemographyofgonorrhoeainLeeds,1989–93[seecomments].
BMJ1997;314:1715–18.
21LowN,Daker-WhiteG,BarlowD,etal.
GonorrhoeaininnerLondon:resultsofacross-sectionalstudy.
BMJ1997;314:1719–23.
22HartG.
Factorsassociatedwithgenitalchlamydialandgonococcalinfectionsinfemales.
GenitourinMed1992;68:217–20.
23EllenJM,KohnRP,BolanGA,etal.
SocioeconomicdiVer-encesinsexuallytransmitteddiseaseratesamongblackandwhiteadolescents,SanFrancisco,1990to1992.
AmJPub-licHealth1995;85:1546–8.
24McCormackWM,RosnerB,McCombE,etal.
InfectionwithChlamydiatrachomatisinfemalecollegestudents.
AmJEpidemiol1985;121:107–15.
25JollyAM,OrrPH,HammondG,etal.
Riskfactorsforinfec-tioninwomenundergoingtestingforChlamydiatracho-matisandNeisseriagonorrhoeaeinManitoba,Canada.
SexTransmDis1995;22:289–95.
26RamstedtK,ForssmanL,GieseckeJ,etal.
RiskfactorsforChlamydiatrachomatisinfectionin6810youngwomenattendingfamilyplanningclinics.
IntJSTDAIDS1992;3:117–22.
270RadcliVe,Ahmad,Gilleran,etalwww.
sextransinf.
comonDecember25,2020byguest.
Protectedbycopyright.
http://sti.
bmj.
com/SexTransmInfect:firstpublishedas10.
1136/sti.
77.
4.
265on1August2001.
Downloadedfrom