Vulnerability to HIV among older men who have sex with men users of dating apps in Brazil

(1)

w w w . e l s e v ie r . c o m / l o c a t e / b j i d

The

Brazilian

Journal

of

INFECTIOUS

DISEASES

Original

article

Vulnerability

to

HIV

among

older

men

who

have

sex

with

men

users

of

dating

apps

in

Brazil

Artur

Acelino

Francisco

L.N.

Queiroz

a

_,

_Alvaro

_Francisco

_Lopes

_de

_Sousa

a,b,∗

_,

Sandra

Brignol

c

_,

_Telma

_Maria

_Evangelista

_Araújo

d

_,

_Renata

_K.

_Reis

a

a_Universidade_de_São_Paulo,_Escola_de_Enfermagem_de_Ribeirão_Preto,_Ribeirão_Preto,_SP,_Brazil

b_Global_Health_and_Tropical_Medicine_(GHTM),_Instituto_de_Higiene_e_Medicina_Tropical,_Universidade_Nova_de_Lisboa,_Lisboa,_Portugal c_Universidade_Federal_Fluminense,_Departamento_de_Saúde_Pública,_Rio_de_Janeiro,_RJ,_Brazil

d_Universidade_Federal_do_Piauí,_Departamento_de_Enfermagem,_Teresina,_PI,_Brazil

a

r

t

i

c

l

e

i

n

f

o

Articlehistory:

Received13May2019

Accepted28July2019

Availableonline28August2019

Keywords:

Menwhohavesexwithmen

HIV/Aids

Sexuallytransmittedinfections

Mobilephoneapplications

Middle-agedmen

a

b

s

t

r

a

c

t

Introduction:The elderlypopulationis increasinglybeneﬁting fromrecenttechnological

advances.Inthisscenario,geolocation-baseddatingapplicationsprovideaviablealternative

forﬁndingpartnersinapracticalandtimelymanner,butmaybeaccompaniedbycertain

riskbehaviorsforHIVinfection.Althoughthereareconsiderablenumberofusersover50on

theseapplications,nostudieshaveaddressedthisproblem.Theaimofthepresentstudy

wastoanalyzefactorsofvulnerabilitytoHIV/Aidsamongthepopulationofmenwhohave

sexwithmen(MSM)age50yearsorolderwhousedatingapps.

Methods:Thiswasacross-sectional,population-survey-based,analyticalstudy,conducted

exclusivelyonlinewithasampleof412MSM.Thedatawascollectedfromthefollowing

apps:Grindr®,Hornet®,Scruff® andDaddyHunter®.

Results:Factors associated with a higher chance of having HIV were: sexual relations

with an HIV-infected partner (ORa=5.53; 95%CI=2.23–13.73); chemsex (ORa=3.97;

95%CI=1.72–8.92); and, above all, having an HIV-infected partner (ORa=8.02;

95%CI=2.01–32.01). The belief that apps increase protection against sexually

trans-mittedinfections(ORa=0.43;95%CI=0.19–0.95)andnotbeingfamiliarwithpost-exposure

prophylaxis (ORa=0.43; 95%CI=0.19–0.95) were associated with decreased chances of

havingHIV.

Conclusions:WehighlightsomeimportantfactorsthatstructurethevulnerabilityoftheMSM

surveyedinrelationtoHIVinfection.Theﬁndingsshouldbeusedtocustomizecareforthis

population,whichcouldbringtheminmoreforhealthcareservices.

anopenaccessarticleundertheCCBY-NC-NDlicense(http://creativecommons.org/

licenses/by-nc-nd/4.0/).

∗ _{Corresponding}_author_at:_Escola_de_Enfermagem_de_Ribeirão_Preto,_Universidade_de_São_Paulo,_Avenida_dos_{Bandeirantes,}₃₉₀₀_Campus

Universitário,BairroMonteAlegre,RibeirãoPreto,CEP:14040-902SP,Brazil.

E-mail addresses: arturqueiroz@usp.br (A.A. Queiroz), alvarosousa@usp.br (A.F. Sousa), sandrabrignol@gmail.com (S. Brignol),

telmaevangelista@gmail.com(T.M.Araújo),rkreis@eerp.usp.br(R.K.Reis).

https://doi.org/10.1016/j.bjid.2019.07.005

(2)

Introduction

Inrecentdecades,thepopulationofmiddle-agedandolder

men(50yearsorolder)hasenjoyedthebeneﬁtsofimproved

qualityoflife,whichhasincreasedtheirlifeexpectancy.

Fur-thermore, technological advances in the ﬁeld of medicine,

especiallythoserelativetohormonereplacementtreatment

anderectiledysfunctionmedications,haveallowedthesemen

tocontinuetobesexuallyactive.1

Agingandtheexpressionofsexualitycommonlyreferto

theconﬂuencebetweenbodyandcultureandareconsidered

mutuallyexclusive inWesternsocieties,especiallybecause

ofthe lossofphysicalactivity and thesense ofasexuality

experiencedinoldage.However,nowadays,theseideasare

constantlychallenged. This conﬂictreaches its apexwhen

consideringmalegay culture,whichis markedby

compla-centhedonismassociatedwithbeingobsessedwithphysical

attributescapableofelicitingattractionanddesire.2,3

Aspartofthisrediscoveryoftheirownbodiesand

sexual-ity,middle-agedandoldermenwhohavesexwithmen(MSM)

havegraduallyfreedthemselvesofthesocialrestrictionsand

normshistoricallyimposedonpeopleofthisage regarding

sexuality,recreatingandremodelingnormstobettersuittheir

experiences,sexualneeds,anddesires.4

Theuseofonlinedigitalsocialmedia,especiallygeosocial

datingappsforsexualpurposes,helpsmeetthedemandof

thismalepopulation,providinganeffectiveandconvenient

mechanismtoquicklylocatesexualpartners,incomparison

withother,moretraditionalorofﬂinemethods.5,6

Thesemobileappsallowuserstoﬁlterpossiblepartners

accordingtotheirpreferences(age,sexpositions,race,and

physicalattributes,amongothers),streamliningthecreation

ofpartnerships,especiallysexual.5,6_This_agility_in

establish-ingpartnerships,however,cancontributetocondomlessanal

sexandotherpracticesthatcanincreaseexposuretosexually

transmittedinfections(STIs),especiallythehuman

immunod-eﬁciencyvirus(HIV).7,8_Furthermore,_lack_of_discussions_about

thesexualhealthofMSMaged50yearsoroldercontributesto

thisscenario,becauseitreinforcestheinvisibilityoftheLGBT

communityandthesexualityofolderpersons.9

TheliteratureaboutfactorsrelatedtoHIVinfectionamong

middle-agedandolderMSMisscarce,especiallywhen

con-sidering the use of social media, even though data from

internationalofﬁces,suchastheUSCentersforDisease

Con-trol, show that the incidence of the infection has grown

considerablyamongtheMSMpopulation.9–12_The_contextual

factorsthatcouldexplainthispopulation’sgreater

vulnerabil-itytoHIVarestillunknown.Theobjectiveofthepresentstudy

wastoanalyzefactorsofvulnerabilitytoHIV/Aidsamongthe

populationofMSMaged50yearsorolderwhousedatingapps.

Theconceptofvulnerabilityadoptedhereconsistsofthree

fundamentalaspectsordimensions—individual,social,and

programmatic—thatareusedtodescribethesusceptibilityof

populationgroupstoimportanthealthproblems.Inthe

con-textoftheHIV/Aidsepidemic,vulnerabilityisrelatedtothe

ideathatpeople’schancesofbeingexposedtothevirus(and

ofdevelopingAids)aretheresultofasetofindividual,

col-lective,andcontextualaspects,whichtogetherleadtogreater

susceptibilitytoinfectionanddisease.13

Individual vulnerability is of a cognitive nature and

includes biological, emotional, cognitive and attitudinal

aspects relative to social relationships, understood as the

quantityandqualityoftheinformationavailableto

individ-uals,inassociationwiththeirabilitytoprocessit.Thesocial

aspectofvulnerabilityisdeﬁnedbycultural,social,and

eco-nomicaspectsthatdeterminepeople’sopportunitiestoaccess

goods andservices.Programmatic(healthcare)vulnerability

referstothenecessarysocialresourcestoprotectindividuals

fromriskstotheirphysical,psychological,andsocialintegrity

and well-being. Any study ofvulnerability depends on the

analysisofthecombinationofthethreedomains.14

Methods

Thisarticleispartofamulticenterstudycalled“Behaviors,

practices,andvulnerabilitiesamongmenwhohavesexwith

menanduse geolocation(geosocial)datingappsinBrazil”.

Itwasacross-sectional,population-survey-based,analytical

studycarriedoutwithMSMaged50yearsorolderwhouse

datingapps,conductedexclusivelyonlineinalltheregionsof

Brazil.

Sample

For sampledeﬁnitionweusedamodiﬁcation oftime

loca-tion sampling (TLS). This technique seeks to approach

probabilistic sampling by mapping the universe of places

where the target population can be found in large

num-bers and then randomly selecting aspeciﬁc day, time and

place for recruitment and systematically selecting

partici-pantsfromthatlocation.15_Volunteers_who_were_{speciﬁcally}

approached through the applications and who agreed to

respond to the online questionnaire were considered.The

inclusion criteria were: identiﬁed as a cisgender male;

being 50 years or older; and being online at the time of

data collection. Users who did not reside in Brazil were

excluded.

Speciﬁcprocedures

Toﬁndparticipants,fourappscommonlyusedbyMSMwere

selected: Grindr®,Hornet®, Scruff® and DaddyHunter®.8

The computer-assistedself-interview (CASI) technique was

usedfordatagathering.15_The_participants_were_approached

throughtheselectedappsforfourconsecutivemonths

(Febru-arytoMayof2017),intheafternoon,eveningandatnight.

Theparticipantswereinformedabouttheresearchobjectives

andtheimportanceoftheirparticipation.Aftergaining

con-sent,theresearcherssharedthehyperlinktothesurveyform,

wheretheparticipantsansweredspeciﬁcquestionsofinterest

tothisstudy.

To cover theparticipants’ socialand demographic

char-acteristics, sexual behavior, and health information, the

questionnairewasdividedintofoursections,whichincluded

bothmandatory andoptionalquestions: personal

informa-tion; sociocultural information; health issues; and sexual

practices.Thesystemonlyallowedparticipantstoproceedto

(3)

Furthermore,thequestionnaireswereonlycomputedif,atthe

end,alltheconditionsweremet.Incompletesurveyswerenot

savedbythesystem.

Toanswer thequestionnaire,participantsprovidedtheir

email address, preventing duplicated entries. If a

partici-pantwasunwillingtoparticipate,thenextonlineuserwas

approached.Recruitmentwasconductedbytwoadult

cisgen-dermaleresearcherswhoareexpertsinthestudiedﬁeld.They

createdaccountswithpublicproﬁles(open,withaphoto)in

theappstogainaccesstotheusers.

Theanalysisofthedatawasbasedontheadopted

vulner-abilityframework.13,14 Theresearchvariables weregrouped

accordingtothethreedomainsofvulnerability:

1) Individual: age, marital status, living situation, number

ofpartners,typeofsexualpartnership established,

sex-ual practices(sex positions, consistentuse ofcondoms,

barebacksex,groupsex),formsofprevention(condoms,

withdrawal,nopenetration),chemsex(usingdrugsduring

sexualrelations),useoferectiledysfunctiondrugs,andthe

presenceofSTIs;

2) Social:educationallevel,sexualorientation,religionand

goingtosaunas;and

3) Health care (programmatic): being familiar with

post-exposureprophylaxis(PEP)andpre-exposureprophylaxis

(PrEP);goingtohealthcareservices; andpriortestingfor

HIV.

Variablesthat allowedmorethan oneresponse,suchas

preventionandsexualposition,werepresentedtothe

partic-ipantsinmultiple-choiceformat.

Dataanalysis

Descriptiveanalysiswasemployedtodescribethe

vulnerabil-ityfactorsandotherfactorsofinterest,forboththenumerical

andcategoricalvariables.ThedatawereanalyzedusingIBM®

SPSSversion23.0.Associationsbetweenpositiveserological

statusforHIVandthecategorizedvulnerabilityfactorswere

assessedusing the Pearson’s chi-squared test and Fisher’s

exact test, with a 0.05 signiﬁcance level. To construct the

conﬁdenceinterval,thereliabilitywassetat95%.The

out-comeofinterestwaspositiveHIVstatus,whichwasassessed

usingthequestion:DoyouknowyourHIVstatus?Theanswer

optionswere: ÏamHIV_{positive¨, Ï}amHIV_negative¨,and Ïdo

notknow/I’mnot_sure¨.

Multivariate logistic model wasadopted to producethe

adjustedoddsratios(ORa),consideringthefactorsthat

pre-sentedstatisticalsigniﬁcanceinunivariateanalysis.

Ethicalaspects

The study was approved by the CEP/CONEP system (no.

1.523.003)andstrictlycompliedwithalltheethicalprecepts

thatguide research withhumansubjects. Theparticipants

readthe free and informed consentformand then signed

it,indicating theiragreementwith theproposed objectives

andwillingnesstoparticipateinthestudy.Thisconsentwas

appliedandobtainedonline.

Table1–Sociodemographiccharacteristicsofmenaged 50yearsorolderwhohavesexwithmen(MSM)anduse datingapps.Brazil,2017.

Variables n % Age 50–59 177 43.0 60–69 154 37.4 70–79 61 14.8 80ormore 20 4.8 Educationlevel Elementaryeducation 28 6.8 Secondaryeducation 130 31.6 Highereducation 168 40.8 Graduatestudies 86 20.8 Practicesareligion Yes 195 47.3 No 217 52.7 Livingsituation Alone 167 40.5 Family 162 39.3 Peer/friend 33 8.0 Withpartner 50 12.2 Maritalstatus Single 253 61.4 Married 42 10.2 Stablerelationship 57 13.8 Separated/Divorced/Widowed 60 14.6 Sexualorientation Heterosexual 10 2.4 Bisexual 80 19.4 Homosexual 319 77.5 Others 03 0.7

Results

In all,412 participantswere recruited;Table 1shows their

sociodemographiccharacteristics.Therewasapredominance

ofindividualsbetween50and59yearsold(42%),meanage

of61.6years(rangingbetween50and97,SD=9.2).Mosthad

ahighereducation(61.7%)anddidnotpracticeanyreligion

(52.7%). Most were single (61.4%), had not been in a

rela-tionshipinthelast30days(71.1%),identiﬁedthemselvesas

homosexuals(77.4%),andlivedalone(40.5%).

Themostcommontypeofsexualpartnershipestablished

was casual(63.1%),which resultedinameanof2.7sexual

partnerships via the apps inthe 30 days priorto the

sur-vey.ThemostpopularappswereGrindr(59.7%)andHornet

(45.1%),whichwereusedprimarilytofacilitatesexual

encoun-ters(88%).

Regardingfrequencyofuse,emphasisgoestothedailyuse

oftheapps(60%),especiallyintheevening(63%)andon

week-days(64.8%).IntermsofthepresenceofSTIs,22.1%reported

atleastoneintheyearpriortothesurvey,withemphasison

infectionbyHIV(11.7%)andsyphilis(10.2%).

The most common form of prevention was condoms

(86.2%).However,itisworthnotingthat38%optedfor

with-drawal. Furthermore, 31.3% of the participants chose the

insertivesexualpositionexclusively,27.7%,thereceptive

(4)

thosewhohadreceptiveintercourse,25.7%haddoneso

con-domlessatleastonceinthe30dayspriortothesurvey.HIV

prevalencewas11.7%(48),and30.8%(127)didnotknowtheir

serologicalstatus.Alsonoteworthyisthefactthat47.3%(195)

ofthesubjectshadnothadHIVtestingrecently(inthe

previ-ous12months).

Univariateanalysisofpositiveserologicalstatus forHIV

showedassociationswiththefollowingfactors:educational

level(p=0.039);sexposition(p<0.001);HIVtesting(p=0.020);

barebackpractice (p=0.010); chemsex (p=0.003); group sex

(p=0.004);useoferectiledysfunctiondrugs(p=0.002);

pres-ence of STIs (p<0.001); knowledge of PEP (p<0.001) and

PrEP(p<0.001);goingtosaunas(p=0.040);usingwithdrawal

(p=0.010)asaformofprevention;useofPrEP(p=0.050)and

testing(p=0.020);andknowingthepartners’status(p<0.001)

(Table2).

Multivariateanalysisrevealedthatthefollowingvariables

wereindependentlyassociatedwithincreasedchancesof

hav-ingHIV:havingrelationswithsomeonewithHIV(ORa=5.53;

95%CI=2.23–13.73); chemsex (ORa=3.97; 95%CI=1.72–8.92);

and, especially, having a partner with HIV (ORa=8.02;

95%CI=2.01–32.01). The belief that apps increase

protec-tionagainstSTIs(ORa=0.43;95%CI=0.19–0.95)andnotbeing

familiarwithPEP(ORa=0.43;95%CI=0.19–0.95)were

associ-atedwithreducedchancesofhavingHIV(Table3).

Discussion

TheresultsshowedahighprevalenceofHIVamongMSMaged

50yearsorolder(11.7%)whoareusersofdatingappsinBrazil,

especiallywhencomparedwiththegeneralBrazilian

popula-tion(0.39%).16_However,_this_prevalence_is_still_lower_than_that

foundinarecentstudywithMSMin12Braziliancities,which

reportedaprevalenceof18.4%.17_It_is_worth_noting_that_these

studiesuseddifferentmethodologicalapproachestoestimate

prevalencerates.

Theprevalenceinourstudywasalsolowerthaninother

studies conducted with MSM in Latin America. A study

conducted in large urban centers in Colombia revealed a

prevalence of 15%,18 _while _another _carried _out _in _Mexico

showedaprevalenceof16.9%.19_However,_these_two_studies

addressedyoungMSM,withmeanagesof29.7and25years,

respectively.Theseratesarequiteelevatedwhencomparedto

HIVprevalenceamongMSMinEuropeancountries(6.0%).20

InBrazil,thereislittleinformationabouttheHIVepidemic

andriskandvulnerabilityfactorsforSTIsandHIVinfection

amongMSMover50yearsold.Themostrecentstudieshave

foundprevalenceratesandotherinformationaboutthe

gen-eralpopulationofMSM,withasmallrepresentationofmen

inolderagegroups.8,15,17

Intheinternationalliteraturethereare alsofewstudies

thatfocus onthis MSM subgroup, whichlimits knowledge

about the speciﬁccharacteristics ofthesemen and makes

theirvulnerabilityandhealthconditionsinvisible.21–23_To_our

knowledge,thisistheﬁrststudytoprovideoriginaldata

focus-ingonthevulnerabilityfactorsforHIVinfectionamongMSM

aged50and olderwho are usersofgeosocialdating

appli-cations.Studiespreviouslycarriedout onthissubject have

concentratedonthegeneralMSMpopulation(young/adult).5

The sociodemographic proﬁle of the study participants

revealedthatmostwerehomosexualandpresentedhigh

edu-cational levels (universitydegree),a ﬁnding similarto that

inarecentliteraturereview.5_This_proﬁle_can_help_MSM

bet-termanagetheirvulnerabilities,becausewhenequippedwith

knowledgeandotherresources,theyareabletoselectthe

pre-ventionstrategiesthatbetterﬁttheirlifestyle.15 _There_was

alsoapredominanceofparticipantswholivedaloneorwho

werenotinastablerelationship,afactorthatcaninﬂuence

theirsocialsupport.20

Appuseproﬁle–socialvulnerability

Althoughmostofthesocialvulnerabilityfactorsrelatedtoapp

use didnotpresentstatisticalsigniﬁcanceinthe

multivari-ateanalysis,it isimportanttodiscusstheseﬁndings,since

theydifferedfromotherstudies24_and_have_direct_implications

forthesexualpracticesofMSM.Analysisofappuseshowed

characteristics that clearlydeﬁneddifferences betweenthe

consumptionofdifferentgenerationsofMSM,whichcan

con-tributetotheunderstandingofsomeaspectsofthesemen’s

vulnerability. For example,in relationto reasons for using

these apps, younger MSM used them for sexual purposes

(64.7%),8 _a_lower_rate_than _that_found_among_MSM_aged₅₀

yearsorolder(88.0%).Thismayreﬂecttheimportanceofthis

toolinthedailylivesoftheparticipantsandcorroboratesother

studiesindicatingthatolderadultstendtobetteradopt

tech-nologiesthatservespeciﬁcobjectives.25,26

These factors can affect the sexual behavior of users

because themoreanapp isused,the greater the

familiar-ity with its tools, and the greater the ease and speed of

ﬁndingpartners,asevidencedbypreviousstudies.5,27,28_This

familiaritycaninﬂuencethespeedand abilityof

establish-ing partnerships.Astudyrevealedthatthe averagetimeof

use oftheseappsamongyouthswas50% lessthanamong

participantsofthepresentstudy(12monthsand27months,

respectively).5Accordingly,thenumberofpartnerships

estab-lishedthroughtheappswasalsohigherinthepresentstudy

(2.7partnersinthelast30days)whencomparedwithother

studieswithyoungerMSMwhousedappsintheUnitedStates

(2.03and1.9).29,30

It is important to highlight that the safety that comes

from theanonymityand discretionprovidedbytheappsis

useful forsomeMSM,especially those who identify

them-selvesasheterosexualorbisexualandwhodonotchooseto

disclosetheiraffectiveandsexualbehaviortoprotect

them-selvesagainstviolence,stigma,discrimination,homophobia

andfamilyrejection.31,32_The_use_of_new_digital_technologies

makes it easier to contactfriends and establish new

rela-tionships(affectiveandsexual),andthisisconstantlybeing

updated.Consequently,newissuesemerge,suchas

relation-shipsbetweenolderandyoungerMSM.33

It is worth remembering that “believing that apps help

preventHIVinfection,”thatis,believingthatusingappsto

establishpartnershipscansomehowhelppreventHIV

infec-tion,wasstatisticallyassociatedwithbeingaprotectingfactor

againstHIVinfection. Inother words,theparticipants

per-ceivedthatthisformofdigitalinteractionmediatedbysocial

(5)

Table2–HIVserologicalstatusaccordingtovulnerabilityfactors.Brazil,2017.

HIVstatus p-value*

Positive Negative Unknown Total

n % n % n % n % Individualvulnerability Agegroup 50–59yearsold 17 35.4 107 45.1 53 41.7 177 43 0.870 60–69yearsold 21 43.7 86 36.3 47 37 154 37.4 70–79yearsold 8 16.7 34 14.4 19 15 61 14.8 80yearsoldor+ 2 4.2 10 4.2 8 6.3 20 4.8 Maritalstatus 0.249 Single 28 58.3 142 59.9 83 65.4 253 61.4 Stablerelationship 14 29.2 53 22.4 32 25.2 99 24.0 Divorced/Widowed 6 12.5 42 17.7 12 9.4 60 14.6 Typeofpartnerships Casual 27.0 56.2 151 63.7 82 64.6 260 63.1 0.104 Steady 2 4.2 29 12.2 17 13.4 48 11.7

Steadyandcasual 19 39.6 57 24.1 28 22.0 104 25.2

Numberofpartners

Nopartner 4 8.3 31 13.1 19 15.0 54 13.1 0.342

1partner 10 20.8 75 31.6 37 29.1 122 29.6

>1partner 34 70.8 131 55.3 71 55.9 236 57.3

Numberofpartnersviaapp

Nopartner 11 22.9 78 32.9 38 29.9 48 30.8 0.381

>1partner 10 20.8 63 26.6 34 26.8 237 26.0

1partner 27 56.2 96 40.5 55 43.3 127 43.2

Frequencyofuseperweek 0.414

Daily 23 47.9 142 59.9 84 66.1 249 60.4 2daysaweek 9 18.8 40 16.9 18 14.2 67 16.3 3daysaweek 5 10.4 23 9.7 10 7.9 38 9.2 5daysaweek 11 22.9 32 13.5 31 11.8 91 14.1 Sexposition <0.001 Receptiveanal 41 85.4 130 54.9 81 63.8 252 61.2 Insertiveanal 35 72.9 160 67.5 72 56.7 267 64.8

Consistentuseofcondomb _0.972

Yes 41 9.95 205 49.75 109 26.45 355 86.16 No 7 1.6 32 7.76 18 4.36 57 13.83 Barebacksexb _0.010 Yes 13 27.1 40 16.9 34 26.8 87 21.1 No 35 72.9 197 83.1 93 73.2 325 78.9 Chemsexb _0.003 Yes 12 26.7 20 8.7 15 12.0 47 11.7 No 33 73.3 211 91.3 110 88.0 354 88.3 Groupsexb _0.004 Yes 31 64.6 93 40.1 48 38.1 172 42.4 No 17 35.4 139 59.9 78 61.9 234 57.6

Useoferectiledysfunctiondrugsb _0.002a

Never 20 41.7 141 59.5 72 56.7 233 56.6 Rarely 5 10.4 41 17.3 11 8.7 57 13.8 Sometimes 14 29.2 23 9.7 20 15.7 57 13.8 Often 2 4.2 4 1.7 10 7.9 16 3.9 Always 7 14.6 28 11.8 14 11.0 49 11.9 PresenceofSTIsc _<0.001a Yes 27 56.2 45 19.0 19 15.0 91 22.1 No 17 35.4 158 66.7 77 60.6 252 61.2 Don’tknow 4 8.3 34 14.3 31 24.4 69 16.7

(6)

–Table2(Continued)

HIVstatus p-value*

Positive Negative Unknown Total

n % n % n % n % Methodsofprevention Condom 0.972 Yes 41 85.4 205 86.5 109 85.8 355 86.2 No 7 14.6 32 13.5 18 14.2 57 13.8 Withdrawal 0.010 Yes 20 41.7 76 32.1 61 48.0 157 38.1 No 28 58.3 161 67.9 66 52.0 255 61.9 Nopenetration 0.314 Yes 14 29.2 46 19.4 26 20.5 86 20.9 No 34 70.8 191 80.6 101 79.5 326 79.1 UseofPrEP 0.050 Yes 4 8.3 6 2.5 2 1.6 12 2.9 No 44 91.7 231 97.5 125 98.4 400 97.1 Testing 0.020 Yes 9 18.8 15 6.3 12 9.4 36 8.7 No 39 81.2 222 93.7 115 90.6 376 91.3

Knowspartner’sstatus? <0.001a

Nopartner 10 20.8 59 24.9 31 24.4 100 24.3 Yes.Positive 10 20.8 8 3.4 1 0.8 19 4.6 Yes.Negative 8 16.7 68 28.7 15 11.8 91 22.1 Don’tknow 20 41.7 102 43.0 80 63.0 202 49.0 Socialvulnerability Educationlevel 0.014a ElementaryorSecondary 16 33.3 80 33.8 62 48.8 158 38.3 University 32 66.7 157 66.2 65 51.2 254 61.7 Goestosaunas? 0.040 Yes 22 46.8 74 31.6 33 26.6 129 31.9 No 25 53.2 160 68.4 91 73.4 276 68.1 Sexualorientation 0.401a Heterosexual 1 2.1 6 2.5 3 2.4 10 2.4 Homosexual 42 87.5 186 78.5 94 74.0 322 19.4 Bisexual 5 10.4 45 19.0 30 23.6 80 78.2 Other HealthcareVulnerability

FamiliarwithPrEP? 19 39.6 61 25.7 16 12.6 96 23.3 <0.001

Yes 29 60.4 176 74.3 111 87.4 316 76.7

No

FamiliarwithPEP? 35 72.9 105 44.3 35 27.6 175 42.5 <0.001

Yes 12 27.1 132 55.7 92 72.4 237 57.5

No

Testingc ₉ _18.8 ₁₅ _6.3 ₁₂ _9.4 ₃₆ _8.7 _0.020

Yes 39 81.2 222 93.7 115 90.6 376 91.3

No 16 33.3 80 33.8 62 48.8 158 38.3

∗ _Statistical_{signiﬁcance}_was_set_at_p_≤_0.05. a _Fisher’s_exact_test.

b _In_the_last₃₀_days. c _Last₁₂_months.

(7)

Table3–MultivariateanalysisoffactorsassociatedwithvulnerabilitytoHIVinfectionbyMSMaged50yearsorolder andwhousedatingapps.Brazil,2017.

Variables ORa 95%CI p-value*

NotbeingfamiliarwithPEP 0.41 0.18–0.93 0.034

HavinganHIV-positivepartner 8.02 2.01–32.01 <0.001

WouldhaverelationswithanHIV-positiveperson 5.53 2.23–13.73 <0.001

Beliefthatappsincreaseprevention 0.43 0.19–0.95 0.037

Chemsex 3.97 1.77–8.92 <0.001

Visitinghealthservicesyearly 0.05 0.09–0.32 <0.001

∗ _Statistical_{signiﬁcance}_was_set_at_0.05.

Individualvulnerability

The sex position preferred by men, whether insertive or

receptive, is an important factor in the sexual experience

of MSM. Preference for the insertive position

(insertive-only or versatile-insertive position) is associated with an

imageofgreatermasculinity,whichishighlyvaluedingay

sociability,34–36_and_also_presents_a_lower_risk_of_infection_by

HIVandSTIs.20,37

Eventhoughtheparticipantsinthepresentstudyreported

preferringthe insertive position, other risk factors for HIV

(Table2)presentedhigh frequency,which canincreasethe

chancesoftransmissionofHIV.38_These_results_were_similar

tothosefoundintheliterature,wheremanyoftheﬁndings

ofthepresentstudy(erectiondifﬁculties,usingcondomsonly

withcasualpartnersandnotwithsteadypartnersorwhen

engaginginchemsex)arefactorsassociatedwithHIVinfection

amongMSM.39,40

However,there isalso evidencethat people with

unde-tectableHIVviral load,or whouse PrEP,can maintainsafe

sexual relations even with serodiscordant partners.41 _The

repertoireofsexualpracticesinvolvesfactorsfromthe

individ-ualdimensionofvulnerability,issuesthatmustbeaddressed

topreventHIVandSTIs,42 because“notbeingfamiliarwith

PEP,”“havingapartnerwithHIV,”andopennessto“having

relationswithsomeonewithHIV”and“chemsex”emergedas

signiﬁcantlyassociatedwithHIVinfectioninthisstudy.

Chemsexisadifferentiated,non-unitaryandintentional

sexualbehaviorthatcombinessexualactivityand

consump-tionofpsychoactivesubstances.Thisisanimportantfactor

associatedwithpositive HIVstatus, and hasrecentlybeen

addressedbytheliterature,whichhasidentiﬁedanincrease

inthis behavioramongMSM.43 _Among_the_reasons _for_this

increasearegrowingcapacityforengaginginthedesiredform

ofsex(increased arousaland libido,sexualconﬁdenceand

increasedsexualstamina),coupledwiththeimprovementof

qualitiesvaluedinsex,suchasintensiﬁedsensation,reduced

inhibitions,andgreatereaseofhavingsexualadventures.44

Theseeffectsareduetothesubstancesusedinchemsex,

whichprovidemenwiththeabilitytoimprovetheir

perfor-manceandsexualexperiences,increasingarousal,endurance,

andpleasure.Foroldermen,theinvigoratingeffectofthese

substancesseems tobethe mainmotivationfortheir use,

enablingimprovedsexual performancethatcorresponds to

theidealsofmasculinitysharedbythegaycommunity,

espe-ciallywhenhavingrelationswithyoungerpartners.45

Programmaticvulnerability

InBrazil,MSMfacedifﬁcultieswhenaccessinghealthservices

because ofissuesrelative tostigmaand socialviolence,in

additiontodiscriminationsufferedinhealthservices.46_Even

though a set ofnationalhealth policiesis directedtoward

theLGBTcommunity,47_effective_{implementation}_of_these

ser-vicesandreducingdiscriminationarestillachallengeforthe

BrazilianUniﬁedHealthSystem.48_The_results_of_the_present

study showed that“visiting healthservicesatleast oncea

year”wasaprotectivefactoragainstHIVinfection,which

rein-forcestheimportanceofimprovingaccesstohealthservices

forthesemen.

One of the importantissues in the context of the HIV

epidemicthatstillneedtobeaddressedislatediagnosisof

HIVinfection amongMSM,46 _especially_those_over₅₀ _years

old,49,50_since_this_is_an_important_marker_for_access_to_HIV_and

STIpreventionstrategiesinhealthservices.46,51_Programmatic

vulnerabilityisassociatedwithbarrierstoaccesstohealth

ser-vices,whicharestilldiscriminatoryenvironmentsforMSM,in

additiontoﬂawsinpreventionpolicies,sincedissemination

ofandaccesstotheseservicesandsuppliesforMSMisstill

notablylimited.

It isimperative that Braziliansociety and public health

managers face thereality ofthe vulnerability ofMSM

pre-sented by various studies in terms of the HIV and Aids

epidemics.Itisurgentthatexistinghealthpoliciesbe

imple-mented,and effortsbemadetostrengthenpreventionand

interventionactionsthateffectivelyreducethespreadofHIV

andAidsinBrazil.

The present study has somelimitations: a convenience

sampleconsistingofMSMaged50yearsorolderwasused;a

portionofthekeyvariablesforthestudy(testingandHIV

sta-tus)wereself-reported;incompleteresponseswerenotsaved

inGoogleForms;andtheﬁndingthatthemajorityof

partici-pantsinthisstudyhadahighlevelofeducation.

Furthermore,thestudywaslimitedtoreportingﬁndings

foraspeciﬁcpopulationwhousedmobilesocialnetworksfor

givenpurposes,whichcaninﬂuenceresultsandhinder

gener-alizationsabouttheMSMpopulationover50yearswhodonot

usetheseapps.Muchofthediscussionwasbasedonresearch

involvingyoungerMSM,duetolackofresearchwithMSM50

yearsoldorolderwhoareusersofdatingapps.

Despite these limitations, the results effectively reﬂect

importantaspectsofindividual,socialandhealthcare

(8)

Conclusion

AhighprevalenceofHIVwasobservedamongMSM50years

old and older who use dating apps. Furthermore,some of

the factors that affect vulnerability ofMSM regardingHIV

infection were: types of partnership established; partner’s

serologicalstatus;knowledgeofpreventionmeasures;

chem-sex;educationallevels; sexposition;groupsex;presenceof

STIsandtheuse ofPrEP;useoferectiledysfunctiondrugs;

going to saunas; prevention through withdrawal; and

fre-quencyofvisitstohealthservicesandHIVtesting.

Authors’

contributions

AAFLNQandAFLScontributedtodesignandexecutionofthe

study.TMEA,SBandRKRcontributedtoexecutionofthestudy.

Allauthorscontributedtotheinterpretationofﬁndings,read

andapprovedtheﬁnalversionofthearticleforsubmission.

Conﬂicts

of

interest

Theauthorsdeclarenoconﬂictsofinterest.

Acknowledgement

Thisresearchdidnotreceiveanyspeciﬁcgrantfromfunding

agenciesinthepublic,commercial,ornot-for-proﬁtsectors.

r

e

f

e

r

e

n

c

e

s

1. SnyderPJ,BhasinS,CunninghamGR,MatsumotoAM, Stephens-ShieldsAJ,CauleyJA,etal.Effectsoftestosterone treatmentinoldermen.NEJM.2016;374:611–24.

2. MojolaSA,WilliamsJ,AngottiN,Gómez-OlivéFX.HIVafter 40inruralSouthAfrica:alifecourseapproachtoHIV vulnerabilityamongmiddleagedandolderadults.SocSci Med.2015;143:204–12.

3. SimõesJA.Corpoesexualidadenasexperiênciasde envelhecimentodehomensgaysemSãoPaulo.Aterceira idade.2011;22:7–19.

4. LyonsA,CroyS,BarretC,WhyteC.Growingoldasagayman: howlifehaschangedforthegayliberationgeneration. AgeingSoc.2015;35:2229–50.

5. QueirozAAFLN,SousaAFL,AraújoTME.Areviewofrisk behaviorsforHIVinfectionbymenwhohavesexwithmen throughgeosocialnetworkingphoneapps.JAssocNurses AIDSCare.2017;28:807–18.

6. SousaAFL,QueirozAAFLN,FronteiraI,LapãoLV,MendesIA, BrignolS.TestingforHIVamongmiddle-agedandoldermen whohavesexwithmen(MSM):ablindspot?AmJMens Health.2019;13:1557988319863542.

7. WhitﬁeldDL,KattariSK,WallsNE,GrindrAl-TayyibA.Scruff,

andonthehunt:predictorsofcondomlessanalsex,internet

use,andmobileapplicationuseamongmenwhohavesex

withmen.AmJMensHealth.2017;11:775–84,

http://dx.doi.org/10.1177/1557988316687843.

8. QueirozAA,MatosMC,AraújoTM,ReisRK,SousaAF.Sexually transmittedinfectionsandfactorsassociatedwithcondom useindatingappusersinBrazil.ActaPaulEnferm.2019;32(5).

9.ValdesB,GattamortaK,JonesSG,FenklEA,DeSantisJP. Socialsupport,loneliness,depressivesymptoms,and high-risksexualbehaviorsofmiddle-agedhispanicmenwho havesexwithmen.JAssocNursesAIDSCare.2019;30:98–110.

10.EmletCA,ShiuC,KimHJ,Fredriksen-GoldsenK.Bouncing back:resilienceandmasteryamongHIV-positiveoldergay andbisexualmen.Gerontologist.2017;57suppl1:

S40–9.

11.TsangEY,QiaoS,WilkinsonJS,FungAL,LipelekeF,LiX. MultilayeredstigmaandvulnerabilitiesforHIVinfectionand transmission:aqualitativestudyonmalesexworkersin Zimbabwe.AmJMensHealth.2019;13,1557988318823883.

12.UNAIDS,Geneva.Availablefrom:TheGlobalGendergapGap

report.1sted;2018

http://reports.weforum.org/global-gender-gap-report-2018/the-global-gender-gap-index-2018/

13.MannJM,TarantolaDJM,NetterTW.AIDSintheworld.1sted. Boston:HarvardUniversity;1992.

14.JRCMAyres,Franc¸aJI,CalazansGJ,SallettiH.Vulnerabilidade eprevenc¸ãoemtemposdeAIDS.In:BarbosaRM,ParkerR, editors.Sexualidadepeloavesso:direitos,identidadeepoder, 34.RiodeJaneiro:Editora;1999.

15.QueirozAAF,SousaAF,MatosMC,AraújoTM,ReisRK,Moura MEB.KnowledgeaboutHIV/AIDSandimplicationsof establishingpartnershipsamongHornet® users.RevBras Enferm.2018;71:1949–55.

16.BenzakenAS,OliveiraMCP,PereiraGFM,GiozzaSP,Souza FMA,SouzaARC.PresentingnationalHIV/AIDSandsexually transmitteddiseaseresearchinBrazil.Medicine.2018;97 Suppl1:S1–2.

17.KerrL,KendallC,GuimarãesMDC.HIVprevalenceamong menwhohavesexwithmeninBrazil:resultsofthe2nd nationalsurveyusingrespondent-drivensampling.Medicine. 2018;97Suppl1:S9–15.

18.RubioMendozaML,JacobsonJO,Morales-MirandaS,Sierra AlarcónCÁ,LuqueNú ˜nezR.HighHIVburdeninmenwho havesexwithmenacrossColombia’slargestcities:ﬁndings fromanintegratedbiologicalandbehavioralsurveillance study.PLoSOne.2015;10:e0131040.

19.SempleSJ,PitpitanEV,Goodman-MezaD,StrathdeeSA, ChavarinCV,RangelG,etal.Correlatesofcondomlessanal sexamongmenwhohavesexwithmen(MSM)inTijuana, Mexico:theroleofpublicsexvenues.PLoSOne.

2017;12:e0186814.

20.BeyrerC,SullivanPS,SanchezJ,DowdyD,AltmanD, TrapenceG.AcalltoactionforcomprehensiveHIVservices formenwhohavesexwithmen.Lancet.2012;380:424–38.

21.SchildcroutJ.Queerjustice:theretrialsofLeopoldandLoeb.J PopCult.2011;34:175–88.

22.BudhwaniH,HearldKR,BarrowG,PetersonSN,

Walton-LevermoreK.Acomparisonofyoungerandolder menwhohavesexwithmenusingdatafromJamaicaAIDS SupportforLife:characteristicsassociatedwithHIVstatus. IntJSTDAIDS.2016;27:769–75.

23.KuppratSA,KrauseKD,OmpadDC,HalkitisPN.Substance useandcognitivefunctionasdriversofcondomlessanalsex amongHIV-positivegay,bisexual,andothermenwhohave sexwithmenaged50andolder:thegoldstudies.LGBT Health.2017;4(6):434–41.

24.HollowayIW,PulsipherCA,GibbsJ,Barman-AdhikariA,Rice E.Networkinﬂuencesonthesexualriskbehaviorsofgay, bisexualandothermenwhohavesexwithmenusing geosocialnetworkingapplications.AIDSBehavior.2015;19 Suppl2:112–22.

25.SchulzR,WahlHW,MatthewsJT,DabbsAV,BeachSR,Czaja SJ.Advancingtheagingandtechnologyagendain

gerontology.Gerontologist.2015;55:724–34.

26.HubersC,LyonsG.Newtechnologiesfortheold:potential implicationsoflivinginlaterlifefortraveldemand.Transport Policy.2013;30:220–8.

(9)

27.WinetrobeH,RiceE,BauermeisterJ,PeteringR,HollowayIW. AssociationsofunprotectedanalintercoursewithGrindr-met partnersamongGrindrusingyoungmenwhohavesexwith meninLosAngeles.AIDSCare.2014;26:1303.

28.HollowayIW,PulsipherCA,GibbsJ,Barman-AdhikariA,Rice E.Networkinﬂuencesonthesexualriskbehaviorsofgay, bisexualandothermenwhohavesexwithmenusing geosocialnetworkingapplications.AIDSBehavior. 2015;19:112–22.

29.LehmillerJJ,IoergerM.Socialnetworkingsmartphone applicationsandsexualhealthoutcomesamongmenwho havesexwithmen.PLoSOne.2014;9:e86603.

30.LandovitzRJ,TsengCH,WeissmanM,HaymerM,Mendenhall B,RogersK.Epidemiology,sexualriskbehavior,andHIV preventionpracticesofmenwhohavesexwithmenusing GRINDRinLosAngeles,California.JUrbanHealth. 2013;90:729–39.

31.AltmanD,AggletonP,WilliamsM,KongTSK,ReddyV,Harrad

D.Homophobia:Whatisit?Whatareitseffects?Howcanit

bereduced?In:InternationalAIDSConference.International

AIDSSociety,2012.TheConference’swebsiteislocatedat

http://pag.aids2012.org/session.aspx?s=650.

32.OliveiraFBM,QueirozAAFLN,SousaAFL,MouraMEB,ReisRK. Sexualorientationandqualityoflifeofpeoplelivingwith HIV/Aids.RevBrasEnferm.2017;70:1004–10.

33.BirdJD,LaSalaMC,HidalgoMA,KuhnsLM,GarofaloR.“Ihad togotothestreetstogetlove”:pathwaysfromparental rejectiontoHIVriskamongyounggayandbisexualmen.J Homosex.2017;64:321–42.

34.IribarrenSJ,GhazzawiA,SheinﬁlAZ,FrascaT,BrownW, Lopez-RiosJ.Mixed-methodevaluationofsocialmedia-based toolsandtraditionalstrategiestorecruithigh-riskand hard-to-reachpopulationsintoanHIVPrevention InterventionStudy.AIDSBehav.2018;22:347–57.

35.DangerﬁeldDT,SmithLR,WilliamsJ,UngerJ,BluthenthalR. Sexualpositioningamongmenwhohavesexwithmen:a narrativereview.ArchSexBehav.2017;46:869–84.

36.OliveiraJMD,CostaCG,NogueiraC.Theworkingsof homonormativity:Discoursesoflesbian,gay,bisexualand queerpeopleondiscriminationandpublicdisplayof affectionsinPortugal.JHomosex.2013;60:1475–93.

37.MengX,ZouH,FanS,ZhengB,ZhangL,DaiX.Relativerisk forHIVinfectionamongmenwhohavesexwithmen engagingindifferentrolesinanalsex:asystematicreview andmeta-analysisonglobaldata.AIDSBehav.2015;19:882–9.

38.DavisO.AspecialissueofSexualities:barebacksexand queertheoryacrossthreenationalcontexts(France,UK, USA).Sexualities.2015;18:120–6.

39.SurkanPJ,LiY,JacobsonLP,CoxC,SilvestreA,GorbachP. Unsafesexualbehavioramonggay/bisexualmenintheeraof combinationantiretroviraltherapy(cART).AIDSBehav. 2017;21:2874–85.

40.JacobsRJ,KaneMN,OwnbyRL.Condomuse,disclosure,and riskforunprotectedsexinHIV-negativemidlifeandolder menwhohavesexwithmen.AmJMensHealth. 2013;7:186–97.

41.RodgerAJ,CambianoV,BruunT,VernazzaP,CollinsS,van LunzenJ.SexualactivitywithoutcondomsandriskofHIV transmissioninserodifferentcoupleswhentheHIV-positive partnerisusingsuppressiveantiretroviraltherapy.JAMA. 2016;316:171–81.

42.BelusJM,KlineT,CarneyT,MyersB,WechsberG.Measuring relationshipfunctioninginSouthAfricancouples:astrategy forimprovingHIVpreventionefforts.SexRelationTher. 2018;1:1–13.

43.GlynnRW,ByrneN,O’DeaS,ShanleyA,CoddM,KeenanE. Chemsex,riskbehavioursandsexuallytransmitted infectionsamongmenwhohavesexwithmeninDublin, Ireland.IntJDrugPolicy.2018;52:9–15.

44.SewellJ,MiltzA,LampeFC,CambianoV,SpeakmanA, PhillipsAN.Polydruguse,chemsexdruguse,and

associationswithsexualriskbehaviourinHIV-negativemen whohavesexwithmenattendingsexualhealthclinics.IntJ DrugPolicy.2017;43:33–43.

45.GurevichM,MercerZ,CormierN,LeedhamU.Responsibleor recklessmen?Sexuopharmaceuticalmessagesdifferentiated bysexualidentityofusers.PsycholMenMascul.

2017;18:341–51.

46.DouradoI,SilvaLAVD,MagnoL,LopesM,CerqueiraC,Prates A.Buildingbridges:interdisciplinarityinpractice.PopTrans Study:astudywithtransvestitesandtranssexualwomenin Salvador,BahiaState,Brazil.CadSaudePublica.

2016;32:e00180415.

47.MinistériodaSaúde,Brasília.Availablefrom:PolíticaNacional

deSaúdeIntegraldeLésbicas,Gays,Bissexuais,Travestise

Transexuais.1sted;2013http://bvsms.saude.gov.br/bvs/

publicacoes/politicanacionalsaudelesbicasgays.pdf

48.GomesSM,SousaLMP,VasconcelosTM,NagashimaAMS.SUS outofthecloset:conceptionsofmunicipalhealthmanagers ontheLGBTpopulation.SaudeSoc.2018;27:1120–33.

49.OrakaE,MasonS,XiaM.Toooldtotest?Prevalenceand correlatesofHIVtestingamongsexuallyactiveolderadults.J GerontolSocWork.2018;61:460–70.

50.CentersforDiseaseControlandPrevention,Availableat:

https://http://www.cdc.gov/nchhstp/newsroom/docs/ factsheets/cdc-msm-508.pdf.AccessedApril13,2017CDC

FACTSHEET:HIVAmongGayandBisexualMen.2016;2016.

51.TanK,BlackBP.Asystematicreviewofhealthcareprovider perceivedbarriersandfacilitatorstoroutineHIVtestingin primarycaresettingsintheSoutheasternUnitedStates.J AssocNursesAIDSCare.2017;29:357–70.