Financial fraud and health: the case of Spain

(1)

GacSanit.2017;31(4):313–319

Original article

Financial fraud and health: the case of Spain

Maria Victoria Zunzunegui

^a,^∗

, Emmanuelle Belanger

^a

, Tarik Benmarhnia

^b

, Milena Gobbo

^f

, Angel Otero

^c

, Franc¸ ois Béland

^a

, Fernando Zunzunegui

^d

, Jose Manuel Ribera-Casado

^e

aInstitutdeRechercheenSantéPubliquedel’UniversitédeMontréal(IRSPUM),UniversitédeMontréal,Montreal,Quebec,Canada

bDepartmentofFamilyMedicineandPublicHealth&ScrippsInstitutionofOceanography,UniversityofCalifornia,SanDiego,California,UnitedStates

cDepartamentodeMedicinaPreventiva,FacultaddeMedicina,UniversidadAutónomadeMadrid,Madrid,Spain

dDepartmentofPrivateLaw,FacultaddeDerecho,UniversidadCarlosIII,Madrid,Spain

eServiciodeGeriatría,HospitalClínicodeSanCarlos,Madrid,Spain

fFinsalud,Madrid,Spain

a r t i c l e i n f o

Articlehistory:

Received2November2016 Accepted23December2016 Availableonline2March2017 Keywords:

Financialfraud Physicalhealth Mentalhealth Sleepandqualityoflife

a b s t r a c t

Objective:Toexaminewhetherfinancialfraudisassociatedwithpoorhealthsleepingproblemsandpoor qualityoflife.

Methods:Pilotstudy(n=188)conductedin2015–2016inMadridandLeón(Spain)byrecruitingsubjects affectedbytwotypesoffraud(preferredsharesandforeigncurrencymortgages)usingvenue-based sampling.Informationonthemonetaryvalueofeachcaseoffraud;thedateswhensubjectsbecame awareofbeingswindled,lodgedlegalclaimandreceivedfinancialcompensationwerecollected.Inter- groupcomparisonsoftheprevalenceofpoorphysicalandmentalhealth,sleepandqualityoflifewere carriedaccordingtotypeoffraudandthe2011–2012NationalHealthSurvey.

Results: Inthisconventionalsample,victimsoffinancialfraudhadpoorerhealth,morementalhealth andsleepingproblems,andpoorerqualityoflifethancomparablepopulationsofasimilarage.Those whohadreceivedfinancialcompensationforpreferredsharelosseshadbetterhealthandqualityoflife thanthosewhohadnotbeencompensatedandthosewhohadtakenoutforeigncurrencymortgages.

Conclusion: Theresultssuggestthatfinancialfraudisdetrimentaltohealth.Furtherresearchshould examinethemechanismsthroughwhichfinancialfraudimpactshealth.Ifourresultsareconfirmed psychologicalandmedicalcareshouldbeprovided,inadditiontofinancialcompensation.

Fraudesfinancierosysalud:elcasodeEspa ˜na

Palabrasclave:

Fraudesfinancieros Saludfísica Saludmental Sue˜noycalidaddevida

re s um e n

Objetivo:Explorarsilosfraudesfinancierosseasocianalamalasalud,problemasdesue˜noymalacalidad devida.

Métodos:Estudiopiloto(n=188)realizadoen2015-2016enMadridyLeónreclutandopersonasafectadas pordostiposdefraudes(preferentesehipotecasmultidivisas),porelmétodovenue-sampling.Serecogió informaciónsobreelvalormonetariodelfraude,lasfechasenquelapersonaconocíaquehabíasido estafada,habíainiciadounademandayhabíarecibidounacompensacióneconómica.Secompararonlas prevalenciasdemalasaludfísicaymental,sue˜noycalidaddevidaentregrupossegúntipodefraudey conlaEncuestaNacionaldeSaludde2011-2012.

Resultados:Enestamuestraconvencional,lasvíctimasdefraudefinancieropresentaronpeorsalud,más problemasdesaludmentalydesue˜no,ypeorcalidaddevidaquelaspoblacionescomparablesdela mismaedad.Aquellosquehabíanrecibidounacompensacióneconómicaporlaspérdidasenpreferentes tuvieronmejorsaludycalidaddevidaquelosquenohabíanrecibidocompensaciónyqueaquellosque habíancontratadohipotecasmultidivisas.

Conclusión:Losresultadossugierenquelosfraudesfinancieroscausanda˜nosalasalud.Deberíaninves- tigarselosmecanismosporlosquelosfraudesfinancieroscausanda˜nosdesalud.Silosresultadosse confirman,debeproveerseasistenciapsicológicaymédica,ademásdelascompensacioneseconómicas.

∗ Correspondingauthor.

E-mailaddress:[email protected](M.V.Zunzunegui).

Introduction

Itappearsthatprevailingbusinesscultureinthebankingpro- fession has promoteddishonesty. A randomized trial involving employeesatalargeSwissbankdemonstratedthat,onaverage, http://dx.doi.org/10.1016/j.gaceta.2016.12.012

(2)

314 M.V.Zunzuneguietal./GacSanit.2017;31(4):313–319 bankingemployeesbehavedhonestlywhentheydidnotidentified

themselvesassuch;however,whenputinacontextwherethey identifiedthemselvesasmembersofthebankingindustry,asig- nificantproportionofthembehaveddishonestly.¹Suchdishonest behaviourscan inducebankcustomers tobuy highrisks finan- cialproducts,leadingthemtounforeseenindebtedness,andeven personal bankruptcy. The United States Financial Crisis Inquiry Commissionconcluded“Widespreadfailuresinfinancialregula- tionand supervision proved devastating tothe stability of the nation’sfinancialmarkets...Therewasasystemicbreakdownin accountabilityandethics”.² InEurope,bankingderegulationhas beenidentifiedasamaincauseoftherecenteconomicrecession andofmassiveindebtedness.³

QuotingtheInternationalOrganizationofSecuritiesCommis- sion “Complex products were often sold to elderly and senior investorswithlittleinvestingexperienceandmarketknowledge”.⁴ Financialfraudshavecausedlossoflifetimesavingstomillionsof citizens,bothsmallsaversandthoseacceptingabusivemortgages.² Such financial frauds may have harmful consequences for the healthofthoseaffected,butlittleresearchhasbeenconductedon theseconsequences.

Weadvancethehypothesisthatfraudulentbehavioursbyfinan- cialinstitutionsareassociatedwithsubstantialphysicalandmental health problems in the affectedpopulations. Data from a pilot studyin Spainwereusedtoassesshealthstatus ofpopulations affectedbyfinancialfraudscomparingwiththehealthofthegen- eral populations to which they belong. We alsoexamined the potentialrelationshipbetweenfinancialcompensationobtained throughlegalprocessesandadversephysicalandmental health outcomesandpoorqualityoflife.

Contextofthestudy

Spain has been strongly affected by the recent economic crisis caused by the behaviour of financial institutions. Many instancesoffinancialfraudoccurredduringtheperiodbetween 2007and2014.^5–7 Spainwasvulnerabletowidespreadfraudu- lentbehavioursinthebankingsectordueto:1)collusionbetween politicalpowerandboardofdirectorsofsavingbanks;2)anexac- erbationof therecessionlinked tothefalling valueof thereal estatemarketcausedbyirresponsiblelending.⁸Manyfraudulent behaviourswerecarriedoutbytrustedfinancialadvisorsatlocal bankbrancheswithwhomsmallsavershadlong-standingtrusting relations.

Wewillconsiderheretwomainformsoffraud.First,thosethat originatedfromthesaleofcomplexproductstoconsumerswith thepurposeofmobilizingthesavingsofpotentialsmallinvestors torescuebanks.ThisfraudiscalledpreferentesinSpanishforpre- ferredshares,hybridproductswhicharebestdefinedascomplex debtinstrumentsaccordingtotheEuropeanSecuritiesandMar- ketAuthority.⁹TheSpanishParliamentaryCommissionappointed toinvestigatethecommercializationofpreferentesestimatedthat about3millionsmallsaverswereaffectedintheperiodbetween from1998to2012.¹⁰Thisconstitutesasizeableproportion(8%)of the37millionadultsover20yearsofageinSpain(Census,2010).A secondtypeoffraudconsistedininducingclientstocontractmort- gagesinaforeigncurrency(JapaneseYensorSwissFrancs),with theargumentthattheinterestrateapplicabletothecurrencyin whichthemortgagewasdenominatedwaslowerthantheinterest rateapplicableineuros,forexample,0.5%forloansinSwissFranc loansinsteadof5%fortheEuribor.Banksdidnotinformclientson theforecastedinstabilityoftheeurowithregardtotheJapanese yenandSwissfranc.Withdecreasingexchangerates,mortgages labelledinthesecurrenciescouldhitskylinesvaluesineuros,lead- ingtoindebtednessandcapitallossinrealterms.Asanillustration, bySeptember2011,thosewhohadcontractedanaverageloanof

200,000eurosin2008,hadseentheirmortgageincreaseto280,000, eventhoughmonthlypaymentswerehittingnewhighswithout foreseeableend.Anequivalenthomeloanineuroswouldimply equalmonthlypaymentsandamortgagevaluereducedto180,000 euros.Thisfraudisreferredtoasaforeigncurrencyhomeloanor multidivisasinSpanish.

Methods Design

Thisisapilotstudy.DatawascollectedbetweenJuly2015and July2016inmeetingplaceslocatedmostlyintheprovinceofLeon (cityofLeonand countyLaBaneza)and theMadridmetropoli- tanregion.Sincethelistofsmallinvestorsinvolvedinbankfrauds isnotavailable,werecruitedvolunteersthroughthemembersof associationsofpeoplewhose savingswereaffectedbyfinancial frauds,followingtheprinciplesofvenuesamplingforhardtoreach populations.¹¹Weaskedvolunteerswhobelongtoassociationsof citizensaffectedbythesefraudstorecruitotherpeopletheyknow whosharethatcondition,irrespectiveoftheabsoluteamountof thefraudoroftheirmembershipinanyassociation.Duetobud- getaryandtimerestrictionsweincludedcollaboratorsinMadrid andinLeon.Forthepreferentesgroup,participantswererecruited atMadridandLeon,but thestudysample includesparticipants whoresidedinotherprovincesfromtheregionofCastilla-Leon.

Forthemultidivisasgroup,datawascollectedattheASUFINoffice (www.asufin.com)locatedindowntownMadridandwhilemost participantsresidedinMadrid,afewresidedincitiesofAndalucia andtheCanaryIslands.Mostparticipantsarefromurbanareasof morethan50,000inhabitants.

Externaldatabaseforcomparisonandage-adjustment

The 2011-2012 National Health Survey is accessible at the SpanishMinisteryofHealthwebsite.

Datacollection

Data were collectedby self-administeredquestionnaire. The questionnaireisavailableuponrequest.Mostquestionsonsociode- mographicvariables,andhealthoutcomesweredrawnfromthe 2011-2012NationalHealth Survey toincrease comparability.A sectiononfinancialfraudscharacteristicswasadded.

Outcomes

Psychological distress wasmeasured by the General Health Questionnaire(GHQ-28),apsychiatricscreeninginstrumentcom- posedoffoursubscales(somaticsymptoms,anxietyandinsomnia, socialdysfunctionanddepression),eachofsevenitems.¹²Individ- ualswithaGHQ-28aboveorequalto5wereconsideredashaving clinicallyrelevantsymptomsindicativeofneedofmentalhealth care.¹³Self-reporteddiagnosesofpsychiatricdiseaseafter2008 andfrequencyofanxietycrisesafter2008werealsorecorded.

Resilience,thecourageandadaptabilityinfaceoflifemisfor- tune,wasassessedwiththeWagnildandYoung¹⁴resiliencescale, whichiscomposedof14itemsinascalefrom1to7.¹⁴

Self-ratedhealthisavalidindicatorofhealthaccordingtothe generalliterature.¹⁵ Participants wereasked: “Wouldrate your health asvery good,good, fair, poor,orvery poor?”.Self-rated healthwasconsideredgoodifrespondentsanswered“verygood”or

“good”,anditwasconsideredpoorifrespondentsanswered“fair”,

“poor”or“verypoor”.

Chronic conditions diagnosed after theyear of thereported fraud(highbloodpressure,diabetes,cancer,chroniclungdisease,

(3)

M.V.Zunzuneguietal./GacSanit.2017;31(4):313–319 315 heartdisease,strokeandrheumaticdiseases)wereself-reported

basedonthequestion“Hasadoctorornurseevertoldyouthatyou have...”.Whatwastheyearofthatdiagnoses?

Painwasassessedwithonegeneralquestionaboutbeingboth- eredbypaininthelastmonth.¹⁶Thosewhoansweredpositively wereaskedtoreportwheretheyhadpainfromalistofanatomic sites.

Sleepduration,efficiencyandqualitywereobtainedwiththe PittsburghSleepQualityScale,¹⁷validatedforSpain.¹⁸

Qualityoflifewasassessedwithavisualanaloguescaleran- gingfrom1to10.Givenourpreviousexperiencewiththisscale, agoodqualityoflifewasestimatedat8outof10orabove,which correspondsroughlyto75%oftheolderadultspopulation.^19,20 Measuringtheexposuretofrauds

Forpreferentes,foursectionswereincludedinthequestionnaire:

A)natureofthefinancialproduct:amount,amountasproportion oftotalsavings,nameoffinancialentityanddateoftheoperation;

B)dateatwhichthepersonwasmadeawareofthelossofsavings;

C)dateatwhichthepersonstartedalegalclaim,ifany;D)dateat whichthepersonwasfinanciallycompensated,iftherewasacom- pensationandbywhatamount.Thechronicityoftheexposurewas calculatedasthedifferencebetweenthedatetheaffectedperson wasmadeawareofthelossandthedateofcompensationorthe studyinterviewdate.Formultidivisas,wecomputedthedifference betweentheamountoftheinitialloanandtheamounttheperson owestothebankatthetimeoftheinterview.Atthetimeofwrit- ing,onlytwoofthesubjectsaffectedbytheseloanshadreachedan arrangementwiththeirbanktochangethecurrencyoftheirloan fromaforeigncurrencytoeuros;thus,wecouldnotexaminethe effectofthereconversionofthehomeloantoeuros.

Covariates

Age, sex, education, occupation, income and sufficiency of incometocoverbasicneedswereconsideredpotentialconfounders astheycouldbeassociatedwithhealthandlossofsavings.

Statisticalanalyses

First,descriptiveandbivariatestatisticswereusedtocompare thethreeexposuregroups(preferredshareswithfinancialcom- pensation,preferredshareswithoutcompensation,andhomeloans inaforeigncurrency)usingtheChi-squaredtest forcategorical variablesandtheF-testforcontinuousvariablesfollowingapprox- imatelynormaldistributions.Age-adjustedprevalencerateswere computedusingthetotalpopulationofn=188participantsasthe standardpopulation.Second,healthoutcomeswerecomparedto similarindicatorsincomparableagepopulationsparticipatingin the2011-2012NationalHealthSurvey(NHS)whichcontainsinfor- mationon18734subjectsagedbetween29and92years,theage rangeofparticipantsinthispilotstudy.ThissubsampleoftheNHS wasusedtocomputeweightedage-adjustedprevalenceforthe pilotstudyhealthoutcomes.

Ethicalconsiderations

Thisprojectwasapprovedby theinstitutionalreview board oftheHospitalLaPaz,UniversidadAutonomadeMadrid.Written informedconsentwasobtainedfromallrespondents.

Results

The final sample is composed of 188 respondents. Approx- imately 60% come from the Madrid recruiting sites and the

remainingsubjectscomefromtheLeonrecruitingsites.Thereare importantdemographicdifferencesbetweenthethreecomparison groups(Table1).Thoseaffectedbypreferentesareapproximately 20 yearsolder,less educatedand havelowerincomesthan the multidivisagroup,whicharemoreactiveintheworkforce.

Inthepreferentesgroup,theaverageamountlostwas60660 euros(rangefrom3000to300000).Theproportionoftotallifesav- ingslostreachedmorethan80%for25%ofparticipantsandwas lessthan20%for20%ofparticipants.Thetimebetweentheaware- nessoffraudandreceivingcompensationwas2.4(SD=1.4)years forthe60peoplewhohadreceivedcompensationfortheirlossof savings;amongthe47peoplewhowerestillwaitingforfinancial compensation,thistimewasonaverage4.4years(SD=1.8).Eleven subjectshadnotstartedlegalprocedures.Amongthe70subjectsin themultidivisagroup,onlytwohadachievedreconversionoftheir loantoeuros.

Two thirds(67%) of those in themultidivisa groupdeclared having poorhealth, compared tothree fourths (75%) for those who acquired preferentes and were compensated, and to 85%

among those withpreferentes without compensation (Table 2).

Theage-standardizedprevalenceforpoorhealthwas67%(95%CI:

48-87) for multidivisas, 56% (95%CI: 40-73) for those with preferenteswhowerecompensatedand84%(95%CI:61-108)for thosewithpreferenteswithoutfinancialcompensation.

Pain was reported by almost all of those in the preferentes groups,andfouroutoffiveofthemultidivisasamplereportedsome chronicpain.Thenumberofanatomicsiteswithpainwassimilarin themultidivisaandinthepreferentesgroups.Thetwositesthatwere reportedmostofteninthepreferentesgrouparelegsandhands, possiblyrelatedtothehighfrequencyofarthritisinolderpopu- lations.Headachesweresignificantly morefrequent among the multidivisasample,whichmayreflectstressheadaches.Abouthalf ofthepreferentesgrouphadbeendiagnosedwithatleastonenew chronicconditionafterawarenessofthefraudwithnosignificant differenceaccordingtocompensation; thecorrespondingfigure forthemultidivisagroupwas19%(Table2).Age-standardization prevalencesofchronicpainandchronicdiseasesweresimilarto theseobservedproportions.

Onefourthoftheparticipantshadreceivedadiagnosisofpsy- chiatricdiseasesincebeingawareofthefraudandmorethan75%of eachgroupobtainedscorescompatiblewithapsychiatricdisease intheGHQ(Table2).Morethanhalfofparticipantshadeverexpe- riencedanxietycrises;abouthalfofthemhadahistoryofanxiety crisesbeforethefraudandhavecontinuedtohavethemthereafter, whiletheotherhalfexperiencedtheirfirstcrisisafterbecoming awareofthefraud.Allgroupsshowedsimilarlyhighlevelsofanx- ietyandneedforpsychiatriccare.Standardizationchangesvery littletheobservedproportions.Forexample,theagestandardized ratesofpsychiatricdiseasebyGHQare82%(95%CI: 61-100)for multidivisas,77%(95%CI:55-100)forpreferenteswithoutcompen- sationand69%(95%CI:48-90)forpreferenteswithcompensation comparedwiththeoriginalcorrespondingestimatesof86%,79%

and70%.

Resilience,asassessedat thetime oftheinterview,wassig- nificantlyloweramongthosewhohadnocompensationandthe multidivisasgroupcomparedwiththepreferentesgroupwithcom- pensation(Table2).Theageadjustedpercentageoflowresilience was31%(95%CI:16-46)amongthepreferenteswithoutcompensa- tionand26%(95%CI:15-37)amongthemultidivisascomparedwith 10%(95%CI:3-17)amongthepreferentesgroupwithcompensation.

Theaveragenumberofhoursofsleepwassixwithoutsignifi- cantdifferencesacrossgroups.Halfofrespondentsreportedvery goodsleepefficiencywithnodifferencesacrossgroups(Table3).

Wefoundadifferenceinthequalityofsleepaccordingtocompen- sationstatus.Morethanhalfofthosewithcompensationreported havinggoodorverygoodqualityofsleep,whilethispercentage

(4)

316 M.V.Zunzuneguietal./GacSanit.2017;31(4):313–319 Table1

Sociodemographiccharacteristicsofthethreestudysamples.

Preferredshares-Preferentes Homeloansinforeigncurrency-Multidivisas p-value^a With

compensation (n=60)

Without compensation (n=58)

(n=70)

Age,mean(SD),(range) n 66.3(11.3),(29-89) 62.3(12.7),(37-92) 44.0(7.9),(29-60) <0,001

Sex

Men 100 50.8% 56.9% 52.1% 0.785

Women 88 49.2% 43.1% 47.9%

Maritalstatus

Married 125 69.0% 60.3% 70.4% 0.007

Widowed 18 13.8% 17.2% 0.0%

Other 44 17.2% 22.4% 29.6%

Meetneeds

Alotofdifficulty 48 12.1% 22.4% 40.0% <0.001

Quitedifficulty 37 20.7% 13.8% 24.3%

Littledifficulty 66 32.8% 51.7% 24.3%

Nodifficulty 35 34.5% 12.1% 11.4%

Monthlyincome(euros)

<1000 22 14.0% 21.1% 2.9% <0.001

1000-2000 83 61.4% 42.1% 34.8%

2000-3000 40 19.3% 24.6% 21.7%

Morethan3000 38 5.3% 12.3% 40.6%

Cohabitants

Alone 30 14.0% 21.1% 14.5% 0.002

With1 54 49.1% 28.1% 14.5%

With2 55 22.8% 28.1% 37.7%

With3andmore 44 14.0% 22.8% 33.3%

Education

Primaryorless 66 63.2% 42.1% 8.6% <0,001

Secondary 64 19.3% 31.6% 50.0%

University 54 17.5% 26.3% 41.4%

Workstatus

Activeinlaborforce 99 19.0% 41.4% 91.4% <0.001

Notinlaborforce 87 81.0% 58.6% 8.6%

ap-valueforF-testofmeansorforChi-squaretestforequalityofproportions.

issignificantlyloweramongthosewithoutcompensationandin themultidivisasgroup(Table3).Theage-adjustedproportionwith poorqualityofsleepwas45%(95%CI:28-63)forthoseintheprefer- entesgroupwithcompensationcomparedwith69%(95%CI:47-91) amongthosewithoutcompensationand73%(95%CI:52-94)inthe multidivisasgroup.

Theaverage proportion reporting a poorquality of life was significantly lower in the preferentes group that had received compensation(64%)comparedwiththosewithoutcompensation (84.2%)andthoseinthemultidivisasgroup(84.9%)(p=0.027for equalityofproportions).Changesintheseproportionsafterage- adjustmentwerenegligible.

Comparisonwithpopulationhealthsurveys(Table4)

Self-perceivedhealthwaspooramongpopulationsaffectedby financial frauds and stands in sharp contrast with theSpanish NationalHealthSurveyresults.Only32%ofthe2011-2102NHS populationwithages between29and 92 reportedpoorhealth, thecorrespondingage-adjustedfigureswere47%forthoseinthe preferentesgroupwithfinancialcompensation,63%forthosewith compensationand66%forthoseinthemultidivisasgroup.

SuspectedmentalillnessaccordingtotheGHQwasmuchhigher inallcomparisonsgroups thaninNationalHealthsurveysesti- mates:theprevalencewas21.8%intheNHSpopulationcompared with73%(preferenteswithcompensation),77%(preferenteswithout compensation)and84%(multidivisas).

Theprevalenceofhavingeverreceivedapsychiatricdiagnosisof depressionoranxietywas12.1%intheNHSpopulationcompared withtheageadjustedprevalencesofthesediagnosesafterthefraud of25%(preferenteswithcompensation),31%(preferenteswithout compensation)and29%(multidivisas).

Theproportionsleepinglessthan7hourswas24.8%inthe2011 NHScomparedwiththeage-adjustedproportionsof55%(preferenteswithcompensation),78%(preferenteswithoutcompensation) and65%(multidivisas).

Thethreegroupsratedtheirqualityoflifeaspoorcompared withgeneralpopulationreports.^19–21

Discussion

Thispilotstudysuggeststhatthoseaffectedbyfinancialfrauds havepoorerhealththanthereferencegeneralpopulationpartici- patinginthe2011-2012nationalhealthsurveyofSpain.Resultsof thispilotstudysupportthehypothesisthatvictimsoffraudshave suffereddamagetotheirphysicalandmentalhealth,totheirsleep andqualityoftheirlife.Inspiteofthedifferentagedistribution,the peopleaffectedbymultidivisaloanswerenotverydifferentfrom thoseaffectedbythepreferentesfraudinmostofthehealthand qualityoflifeindicatorsconsidered.Thepreferentesrespondents whohadreceivedfinancialcompensationhadoverallbetterhealth andbetterqualityoflifethanthosewithoutcompensation.Thegap betweendiagnosedpsychiatricillnessandsuspectedpsychiatric illnessbasedonGHQscoressuggestsconsiderableunderdiagnoses ofmentalproblems.Ourresultsprovidepreliminaryevidenceofthe harmfuleffectsoffinancialfraudforthehealthoffraudvictims.

Mechanisms

Consideringtheliteratureonrelationshipsbetweeneconomic strainandsocialstressandhealth,threemajormechanismscould explain theeffects of financial frauds onhealth.^22–25 First, loss of money and property may leadto chronicfinancial stress or indebtedness. Populationsaffected by preferentes have suffered

(5)

M.V.Zunzuneguietal./GacSanit.2017;31(4):313–319 317 Table2

Physicalandmentalhealthstatusindicatorsinthethreestudysamples.

Preferredshares-Preferentes Homeloansinforeign currency-Multidivisas

n With

compensation (n=60)

(n=70) p-value^a

Physicalhealth

Self-ratedhealth 0.036

Verygood 4 3.3% 0.0% 2.9%

Good 43 21.3% 15.5% 30.0%

Fair 93 41.0% 65.5% 42.9%

Poor 35 29.5% 12.1% 14.3%

Verypoor 14 4.9% 6.9% 10.0%

0.078

Verygood,good 46 24.6% 15.5% 32.9%

Fairtoverypoor 142 75.4% 84.5% 67.1%

Pain 0.010

No 19 6.8% 3.6% 19.1%

Yes 164 93.2% 96.4% 80.9%

Numberofpainlocations

(mean,sd) 3.2(2.5) 4.0(2.7) 3.6(2.6) 0.28

Chronicconditionssinceawareoffraud <0.001

No 118 45.9% 56.9% 81.4%

Yes 71 54.1% 43.1% 18.6%

Mentalhealth

Psychiatricdiagnosissinceawareoffraud 0.930

No 138 73.8% 74.1% 71.4%

Yes 51 26.2% 25.9% 28.6%

ProbablepsychiatriccaseGHQ 0.105

No 40 29.5% 21.1% 14.3%

Yes 148 70.5% 78.9% 85.7%

GHQ(mean,SD) 11.4(8.2) 11.5(6.8) 13.4(7.8) 0.245

Anxietycrises 0.331

No 91 53.6% 40.4% 44.3%

Yes 99 46.4% 59.6% 56.7%

Anxietycriseshistory(byfraudawareness) 0.095

Firstcrisesafterfraud 46 27.9% 20.7% 24.3%

Crisesbeforeandafterfraud 45 11.5 31.0% 28.6%

Afterbyunknownifbefore 7 3.3% 6.9% 1.4%

Nocrises 91 57.4% 41.4% 45.7%

Resilience 0.006

Lowerquartile(<58) 45 13.1% 29.3% 28.6%

Middlelower(58-70) 48 16.4% 25.9% 32.9%

Middleupper(71-83) 48 29.5% 22.4% 24.3%

Upperquartile(84+) 48 41.0% 22.4% 14.3%

ap-valueforChi-squaretestforequalityofproportionsorp-valueforFtestforequalityofmeans.

Table3

Sleepcharacteristicsinthestudysamples.

Preferredshares-Preferentes Homeloansinforeigncurrency-Multidivisas With

compensation (n=60)

(n=70) p-value^a

Hours(mean,SD) n 6.3(1.3) 5.9(1.1) 6.1(1.2) 0.210

Hours 0.116

7andover 69 42.6% 25.9% 40.6%

Lessthan7 119 57,4% 74,1% 59,4%

Sleepefficiency 0.778

85%andmore 85 48.1% 46.4% 50.0%

75-85% 40 30.5% 39.3% 30.3%

65-75% 22 18.6% 30.4% 24.2%

<65% 29 8.5% 7.1% 6.1%

Sleepquality 0.022

Verygood 10 8.3% 3.6% 4.5%

Good 62 48.3% 28.6% 25.4%

Fair 89 31.7% 60.7% 53.7%

Poor 22 11.7% 7.1% 16.4%

ap-valueforChi-squaretestforequalityofproportions.

thepartialortotallossoflifesavings,whilepopulationsaffected bymultidivisa are actually indebtedfor unpredictableamounts.

Financial strain and material deprivation are well known risk factorsforpoorerhealththroughoutthelifecourse.^26,27Second,

experiencingtheabuseoftrustmayleadtofeelingsofshame,guilt andfamilyconflict,allofthempotentriskfactorsformentalhealth.

Poorhealth behaviourswould constitutethethird mechanisms.

Materialdeprivationasaconsequenceoffinancialfraudcanhave

(6)

318 M.V.Zunzuneguietal./GacSanit.2017;31(4):313–319 Table4

Weightedageadjustedprevalence(95%confidenceinterval)ofhealthoutcomes.

Preferenteswithcompensation Preferenteswithoutcompensation Multidivisas 2011-2012NHS

Ageinyears(range) (29-89) (37-92) (29-60) (29-92)

Fair,poor,verypoorself-rated health

54%(37%-72%) 85%(74%-97%) 66%(54%-79%) 32%

Psychologicaldistress,GHQ12 75%(62%-89%) 80%(66%-94%) 77%(64%-89%) 22%

Diagnosesofpsychiatric diseases(everinNHSand afterawarenessofthefraud inthispilot)

26%(7%-45%) 30%(14%-46%) 27%(14%-40%) 12%

Lessthan7hours(%) 57%(36%-79%) 78%(65%-91%) 68%(56%-81%) 25%

bothadirecteffectonhealth,andanindirecteffectviapsychosocial factorsandtheadoptionofunhealthycopingbehaviours.Lastly, reversecausalitymaybeapartialexplanationsincepeoplewith poorphysicalhealth or mentaland cognitivedisorders maybe morelikelytobetakenadvantageof.Inaddition,thefraud-health associationsinTable4maybeunderestimatedsincetheNHSsurvey populationthatweuseascomparisonmaycontainfraudvictims andthiswillresultinhigherprevalenceofpoorhealthoutcomes.

Testingthesehypotheseswouldrequirealongitudinalpopu- lationstudywithalargeenoughsampleofindividualswillingto providebiologicalsamplesandsensitivepersonalinformationand tobefollowed-upduringseveralyears.Thatresearchwouldneed substantialeconomicsupportduringanextendedperiodoftime.

Strengthsandlimitations

Themain limitationof this studyis thevoluntarynature of participation.Atthispoint,weareunabletoassessifthosewho volunteertoparticipatearemoreorlessaffectedthanthosewho decidednottoparticipate.Twointerpretationsarepossible.First, ifthemostaffectedpopulationwaslesslikelytoparticipate,our findingswouldunderrepresenttheburdenoffinancialfraudson populationhealth.Second,ifthepopulationnotinvolvedinasso- ciationsofcitizensaffectedbyfraudsislesslikelytosuffersevere fraudsorrelatedhealth consequences,ourrecruitmentstrategy wouldleadtoanoverestimationoftheaverageeffectsoffraudson health.Asecondlimitationisthesmallsamplesizesincedueto budgetaryconstraintswerestrictedfieldresearchtooneyear.The smallsamplesizehindersourabilitytoexaminedose-response relationships. We attemptedtoassess categoriesof duration of exposurebythetime of exposuretostress;we usedthe years betweenbeenawareofthefraudandinterviewtimeforthosewho havenotreceivedcompensation,orthenumberofyearswaiting afteraclaimwasfiledanduntilthecasewasresolvedforthosewho hadacompensation,buttherewaslackofstatisticalpowerusing suchsmallgroups.Thisdesignandrestrictedsampledonotallow toexaminethemoderatingeffectsofsocialsupportandresilience ontheassociationsbetweenfraudandhealthandthemediating mechanismsoutlinedabove.Lastly,non-differentialmisclassifica- tionbiascouldaffectthecomparisonofthestudysampleswiththe NHSparticipantsiftheexperienceoffraudhadaffectedself-reports ofhealthoutcomes.Amongthestrengths,wewouldliketomen- tiontheoriginalityofthestudyandthevalidityandextendeduse ofthehealthmeasurementtools,whichenhancethecomparability ofresultswiththegeneralpopulation.

Futureresearch

Researchontherelationshipbetweenfinancialfraudsandpop- ulationhealthwillpromotetherecognitionofthisseriouspublic healthissueandprovideevidencetodevelopappropriatecompen- sationforaffectedindividualsandtheirfamiliesthroughdedicated social programmes. Questions aboutfinancial frauds shouldbe

includedingeneralhealth surveys.Researchcouldprovideevi- dence to support legal restitution and financial compensation.

Qualitativeresearchcouldalsofurtherinvestigatethemechanisms throughwhichtheexperienceoffinancialfraudmanifestsacross familiesandgroups.

Conclusion

Ourresearchsuggestsfinancialfraudharmshealthandquality oflife.Adedicatedresearchprogramontherelationshipbetween healthandfinancialfraudsisthusneededinordertoprovideevi- dencethatmayhelprecognizingthisissueandultimatelybenefit theaffectedpopulations.Iftheresultsofthisresearchareconfirmed in largerpopulationstudies,interventions providing healthcare andlegalassistanceshouldbeimplemented,inadditiontofinancial compensation.

Whatisknownaboutthetopic?

Financial frauds have caused the Great Recession but theirconsequencesforthehealthofaffectedpopulationsare unknown.

Whatdoesthisstudyaddtotheliterature?

Thoseaffectedbyfinancialfraudshavepoorerhealththan thereferencegeneralpopulationascomparedwithnational surveys.Thispilotstudysuggeststhatfinancialfraudsharm thephysicalandmentalhealth,thesleepandthequalityoflife ofaffectedpeople.Epidemiologicresearchwithlongitudinal designsareneededtostrengthencausalinference.

Editorincharge CarlosÁlvarez-Dardet.

Transparencydeclaration

Thecorrespondingauthoronbehalfoftheotherauthorsguar- antee the accuracy, transparency and honestyof the data and informationcontainedinthestudy,thatnorelevantinformation hasbeenomittedandthatalldiscrepanciesbetweenauthorshave beenadequatelyresolvedanddescribed.

Authorshipcontributions

M.V.Zunzunegui,A.Oteroand F.Zunzuneguiconceptualized thisstudy.M.Gobbo,F.ZunzuneguiandJ.M.RiberaCasadocollab- oratedinitsconductandthewritingofthispaper.M.V.Zunzunegui, E.Belanger,F.BélandandT.Benmarhniadidthedataanlyses.All

(7)

M.V.Zunzuneguietal./GacSanit.2017;31(4):313–319 319 authorshaveparticipatedwiththeircommentstodifferentver-

sionsofthemanuscriptandhaveapproveditsfinalversion.

Acknowledgements

Weacknowledgethepeopleaffectedbyfinancialfraudswho haveparticipated in this research,andtheFinsalud Foundation whose financial contribution has allowed the conduct of this study.

Funding

ThefoundationFinsalud(www.finsalud.com)hasfinanciaded thisstudiy.

Conflictsofinterest

Authorsdeclarenoconflictofinterest.F.ZunzuneguiandJ.M.

RiberaCasadoaremembersoftheBoardofDirectorsoftheFoun- dationFinancesandHealth(Finsalud).

References

1.CohnA,FehrE,MarechalM.Businesscultureanddishonestyinthebanking industry.Nature.2014;516:86–9.

2.FinancialCrisisInquiryCommission.FinancialCrisisInquiryReport.Washing- ton:USGovernment;2011.

3.MullerM,DevnaniS,HyasR,etal.Consumerprotectionaspectsoffinancialser- vices.Brussels:EuropeanParliament.Directorate-GeneralforInternalPolicies;

2014.

4.InternationalOrganizationofSecuritiesCommission(IOSCO).In:Suitability requirementswithrespecttothedistributionofcomplexfinancialproducts.

InternationalOrganizationofSecuritiesCommission;2012.

5.ZunzuneguiF.Comercializacióndeparticipacionespreferentesentreclientela minorista.RevistadeDerechoBancarioyBursátil.2013.

6.ZunzuneguiF.Latoxicidaddelashipotecasmultidivisa.RevistadeDerechodel MercadoFinanciero.2015.

7.ZunzuneguiF.Distributionofpreferredsharesamongretailclients.2013.Avail- ableat:http://ssrn.com/abstract=2242520

8.MisseA.Lagranestafadelaspreferentes.Madrid:AlternativasEconomicas;

2016.

9.EuropeanSecurityandMarketAuthorityFinalreport:Guidelinesoncomplex debtinstrumentsandstructureddeposits.https://www.esma.europa.eu/sites/

default/files/library/2015-1787-guidelinesoncomplexdebtinstruments and structured deposits.pdf;2015.

10.Comisióndeseguimientosobrecomercializacióndelosinstrumentoshíbri- dos de capital y deuda subordinada. http://graphics.thomsonreuters.com/

13/05/cnmvhibridos.pdf;2013.

11.MuhibFB,LinLS,StueveA,etal.Avenue-basedmethodforsamplinghard-to- reachpopulations.PublicHealthReports.2001;116(Suppl1):216–22.

12.LoboA,Perez-EcheverriaMJ,ArtalJ.ValidityofthescaledversionoftheGen- eralHealthQuestionnaire(GHQ-28)inaSpanishpopulation. Psychological Medicine.1986;16:135–40.

13.DelAmoJ,JarrinI,Garcia-FulgueirasA,etal.MentalhealthinEcuadorian migrantsfroma population-basedsurvey:theimportanceofsocialdeter- minantsandgenderroles.SocialPsychiatryandPsychiatricEpidemiology.

2011;46:1143–52.

14.WagnildGM,YoungHM.Developmentandpsychometricevaluationofthe ResilienceScale.JournalofNursingMeasurement.1993;1:165–78.

15.MosseyJM,ShapiroE.Self-ratedhealth:apredictorofmortalityamongthe elderly.AmJPublicHealth.1982;72:800–8.

16.PatelKV,GuralnikJM,DansieEJ,etal.Prevalenceandimpactofpainamong olderadultsintheUnitedStates:findingsfromthe2011NationalHealthand AgingTrendsStudy.Pain.2013;154:2649–57.

17.GrandnerMA,KripkeDF,YoonIY,etal.CriterionvalidityofthePittsburgh SleepQualityIndex:investigationinanon-clinicalsample.SleepandBiological Rhythms.2006;4:129–39.

18.Notario-PachecoB,Solera-MartinezM,Serrano-ParraMD,etal.Reliabilityand validityoftheSpanishversionofthe10-itemConnor-DavidsonResilience Scale(10-itemCD-RISC)inyoungadults.HealthandQualityofLifeOutcomes.

2011;9:63.

19.BelangerE,AhmedT,VafaeiA,etal.Sourcesofsocialsupportassociatedwith healthandqualityoflife:across-sectionalstudyamongCanadianandLatin Americanolderadults.BMJOpen.2016;6:e011503.

20.BowlingA,GabrielZ,DykesJ,etal.Let’saskthem:anationalsurveyofdefinitions ofqualityoflifeanditsenhancementamongpeopleaged65andover.IntJAging HumDevelop.2003;56:269–306.

21.BowlingA.Justonequestion:ifonequestionworks,whyaskseveral.JEpidemiol CommunHealth.2005;59:342–5.

22.AldabeB,AndersonR,Lyly-YrjanainenM,etal.Contributionofmaterial,occupa- tional,andpsychosocialfactorsintheexplanationofsocialinequalitiesinhealth in28countriesinEurope.JEpidemiolCommunHealth.2011;65:1123–31.

23.NiedhammerI,KerradS,SchutteS,etal.Material,psychosocialandbehavioural factorsassociatedwithself-reportedhealthintheRepublicofIreland:cross- sectionalresultsfromtheSLANsurvey.BMJOpen.2013;3.

24.SchutteS,ChastangJF, Parent-ThirionA,etal. Associationbetweensocio- demographic,psychosocial,materialandoccupationalfactorsandself-reported health among workers in Europe. J Public Health (Oxford, England).

2014;36:194–204.

25.van Oort FV, van Lenthe FJ, Mackenbach JP. Material, psychosocial, and behaviouralfactorsintheexplanationofeducationalinequalitiesinmortality inTheNetherlands.JEpidemiolCommunHealth.2005;59:214–20.

26.AdlerNE,NewmanK.Socioeconomicdisparitiesinhealth:pathwaysandpoli- cies.HealthAffairs(ProjectHope).2002;21:60–76.

27.Ben-ShlomoY,KuhD.Alifecourseapproachtochronicdiseaseepidemiology:

conceptualmodels,empiricalchallengesandinterdisciplinaryperspectives.Int JEpidemiol.2002;31:285–93.