Predictors of smoking cessation in 50–66-year-old male Taiwanese smokers:
A 7-year national cohort study
Alan Chung-hong Tsai
a,b,*, Yun-Ann Lin
a, Hsin-Jen Tsai
caDepartmentofHealthcareAdministration,AsiaUniversity,500LiufengRd.,Wufeng,Taichung41354,Taiwan
bDepartmentofHealthServices,SchoolofPublicHealth,ChinaMedicalUniversity,91Hsueh-ShihRoad,Taichung40402,Taiwan
cDepartmentofHealthManagement,I-ShouUniversity,No.8,YidaRd.,Yanchao,Kaohsiung82445,Taiwan
1. Introduction
Cigarettesmokinghaslongbeenidentifiedasamajorcauseof premature death and is a major public health issue in most developedanddevelopingcountries(Petoetal.,1992;Molariuset al., 2001; WHO, 2009a). Itis associated withincreased risk of vasculardiseasesandactssynergisticallywithotherriskfactors.
For example, the relative risk of stroke among hypertensive smokersisfivetimesthatofnormotensivesmokersbut20times thatofnormotensivenon-smokers(OckeneandMiller,1997;Kato et al.,2001).Smokingstatuswasanindependentdeterminantof multiplesilentcerebrovasculardiseases(Eguchiet al.,2004).A significant association was also observed between cigarette smokingand the incidenceof cancer ofthe lungand transient ischemicattacks,intermittentclaudicationandtotalcardiovascu- lardisease,especiallyinmen(Freundetal.,1993).Thus,theWorld HealthOrganization(WHO)hasrepeatedlylaunchedinitiativesto
intensify efforts to control theglobal tobacco epidemic(WHO, 2009b,2010).
Themostcommonandperhapsthemostimportantreasonfor quitting smoking appears to be the concern over health (Hymowitzetal.,1997;Hylandetal.,2004;Suwalaetal.,2005, 2006).Mostsmokershaveincreaseddesiretoquitsmokingwith advancing ageperhapsdue toincreasing concernover health.
Thus,olderindividualsarelesslikelytosmoke(Pomerleauet al., 2004)andmostsmokers(87%)quitsmokingbefore65yearsofage (Suwalaet al.,2006).
Age isamajorfactorofsmokingcessation(Hymowitzetal., 1997;OslerandPrescott,1998;Hendersonetal.,2004;Pomerleau et al.,2004;Grotvedtand Stavem,2005;vanLoonet al.,2005;
Sachs-Ericssonet al.,2009;Ayo-YusufandSzymanski,2010)and wasfoundtobeapredictorofallsevenmostcommonreasonsto quitsmoking(GrotvedtandStavem,2005).Mostsmokersattempt toquitsmokingatallagesbutstrongermotivationappearstobe around 50 years of age perhaps related to the increasing development of aging-associated health conditions (Suwala et al., 2006). Many older-smokers have tried to quit smoking repeatedly without success (Breitling et al., 2009). Although anyageisagoodtimetoquitsmoking,theperiodbetween50and 65 years of age appears to be a critical time for intervention ARTICLE INFO
Articlehistory:
Received5June2011
Receivedinrevisedform10August2011 Accepted11August2011
Availableonline1September2011
Keywords:
Smoking Smokingcessation Longitudinalcohortstudy Dependenceofsmoking Functionalimpairment Predictorsofsmokingcessation
ABSTRACT
The study was aimed to determine the predictorsof smoking cessation in 50–66-year-old male Taiwanesesmokers.Thestudyanalyzeddatasetsofthe‘‘SurveyofHealthandLivingStatusoftheElderly inTaiwan’’(SHLSET),apopulation-basedlongitudinalcohortstudyconductedbytheBureauofHealth PromotionofTaiwan.Binarylogisticregressionanalysiswasperformedtodeterminetheassociationof demographic,socioeconomic,lifestyleandhealth-relatedvariableswithchangesinsmokingstatusat baseline,orwithsubsequentchangesinsmokingstatusin50–66-year-oldmaleTaiwanese.Functional impairmentwasthestrongestpredictorofquittingsmokingfor50–66-year-oldTaiwanesemen.Other factorsincludingafirsthospitalization,emergencyvisit,ordiagnosisofheartdisease,quittingdrinking, livingwithaspouseandolderagewereassociatedwithincreasedlikelihoodofquittingsmoking.Men withlongsmokinghistory,heavydailycigaretteconsumptionandmoreformaleducationwereless likelytoquit.Resultssuggestthatfunctionaldeclineisthemajorcauseforquittingsmokingforolder Taiwanesemen.Physicalimpairmentandtraumaticdiseasesthatcausephysicalimpairmenthavethe mostimpactwhereas ‘‘silentdiseases’’ suchashypertension ordiabetes havelittle impact.These findingsshouldbeusefulfordesigningtarget-specificinterventionstrategiesforoldermaleTaiwanese smokers.
ß2011ElsevierIrelandLtd.Allrightsreserved.
*Corresponding author at: Department of Healthcare Administration, Asia University,500LiufengRd.,Wufeng,Taichung41354,Taiwan.
Tel.:+886423323456x1943;fax:+886423321206.
E-mailaddress:[email protected](A.C.Tsai).
ContentslistsavailableatScienceDirect
Archives of Gerontology and Geriatrics
j ou rna l h om e pa ge : w w w. e l s e v i e r. co m/ l oc a t e / a rch ge r
0167-4943/$–seefrontmatterß2011ElsevierIrelandLtd.Allrightsreserved.
doi:10.1016/j.archger.2011.08.004
becausepeoplehaveincreasingconcernovertheirhealthduring thisperiod(Szantoetal.,2005;Suwalaet al.,2006).Smokerswho quitsmokingduringthisage-rangealsohavethepotentialtoreap abighealthbenefitaftermerelyfiveyears(Menzinet al.,2009).
The determinants of smoking cessation vary according to sociodemographicandhealth-relatedfactors(Vangeliand West, 2008)buttheever-changingsocialandculturalenvironments,the tobaccocontrol policies(Changand Chiang, 2009)and tobacco industry activities also play a major role in shaping smoking behaviorand beliefs(Siahpushet al.,2008). Factorsassociated withsmokingcessation havebeen extensivelystudied on both WesternandAsianpopulationsbuttherealreasonsarefarfrom wellunderstood (Freundet al., 1992; Pomerleau et al., 2004).
Identification of population-specific cessation determinants is importantfordevelopingeffectiveinterventionprograms(Chung et al., 2009). The present study was aimed to identify the predictors of discontinuing smoking in 50–66 years old male Taiwanesesmokers.
2. Subjectsandmethods 2.1. Sourceofdata
Thisstudyanalyzedthedatasetofthecohortsampledin1996 for the SHLSET. SHLSET is a prospective cohort study being conducted by the Bureau of Health Promotion of Taiwan for gaininganunderstandingoftheimpactofdemographic,socioeco- nomic,environmentalandlifestylechangesonhealth,healthcare useandqualityoflifeofolderTaiwanese.Subjectsweredrawn withathree-stageprobabilitysamplingprocess(BureauofHealth Promotion,2007a).Thefirststageinvolvedastratificationofthe Taiwanesepopulationinto361primarysamplingunits(PSUs)and then a proportion-to-size random selection of 56PSUs from 331PSUs (after excluding 30 lightly populated mountainous areas). The second stage was a proportion-to-size random selection of residential blocks from each selected PSU and the finalstage was a random selection of two persons from each selectedblock.Theprocessselected3041menandwomen,50–66 yearsof age,and 2462of thesepersons (81.2%)completed the initialsurveyin1996.Thesesameparticipantswereinterviewed again in 1999, 2003 and 2007. Data up to 2003 survey were availableforacademicstudies.Theresponseratewas90.6%forthe 1999surveyand92.1% forthe2003survey.Theprotocolofthe survey was reviewed and approved by government-appointed representatives and the study wasconducted according to the ethicalstandardssetforthintheHelsinkiDeclaration.
2.2. Datacollection
In each survey, trained interviewers conducted in-home in- person interviews using a structured questionnaire to elicit demographic, socioeconomic, lifestyle, health and healthcare- relatedinformationoneachsubject(BureauofHealthPromotion, 2007b).
2.2.1. Predictorvariables
The covariates were sociodemographic, health condition, medical history and lifestyle-related variables (including self- reportsofconfirmedhypertension,diabetes,heartdisease,stroke, cancer,respiratorydiseases,gastrointestinaldisorders,hospitali- zation, emergency visit, health examination, routine physical activity;smoking,betelnut-chewingandalcohol-drinkinghabits;
andphysicalfunctionalstatus).
Functionalstatuswasassessedwiththeactivitiesofdailyliving (ADL) score, adapted from the1984 National Health Interview SurveySupplementonAging(FittiandKoval,1987). Twoscores
could be derived from the scale: one measured difficulty on carryingthefunctionalitemsandtheothermeasuredthedegreeof difficulty on performing the items. To assess ADL, participants were asked whether they had difficulty bathing, dressing, transferring,eating,walkingortoiletingwithouthelp.Participants alsoratedthelevelofdifficultyforperformingeachofthesixitems onascalefrom0to3points(0=nodifficulty,1=somedifficulty, 2=much difficulty, and 3=cannot do). Individual scores were summedacrosseachofthesixitemstoyieldatotalscorefrom0to 18,withahigherscorereflectingahigherlevelofADLdifficulty (Johnsonetal.,2007).Wealsodefinedthatatotalscoreof0–2isno impairmentwhile3ormoreisimpaired.
2.3. Outcomemeasures
Changesin smokingstatus and otherinformation related to cigarette-smokingwerederivedfromtheitemsthatspecifically askedwhethereachsubjectwasacurrentsmokerandtheaverage number of cigarettes consumed per day. Those who werenot current smokers were also asked whether he/she was a past- smokerandthenumberofyearshe/shesmoked.
2.4. Statisticalanalysis
Logisticregressionwasperformedtodeterminetheabilityof thesociodemographic,lifestyleandhealth-relatedvariablesand thechangesofthesevariables(suchasquittingalcohol-drinkingor havinganewdiagnosisofdiabetesoranewfunctionalimpairment (occurredaftertheinitialsurveyin1996)topredictachangein smokingstatus in 1999or 2003.Thestabilityof behavior-and health-relatedvariableswasevaluatedwithSpearman’scorrela- tionanalysis.
Onlymalesmokerswereanalyzedbecausetherewereonly37 femalesmokersinthecohortatbaseline,notsufficienttohavea meaningfulanalysis.Statisticalanalyseswereperformedwiththe
Table1
Characteristicsof673malesmokersoftheSHLSETstudyin1996inTaiwan.
Variables n(%)
Age,years
50–59 417(62.0)
60–66 256(38.0)
Formaleducation,years
6 480(71.3)
7 193(28.7)
Maritalstatus
Married 582(86.5)
Notmarried 91(13.5)
Numberofcigarettesmoked/day
1–15 240(35.7)
16–30 342(50.8)
>30 91(13.5)
Smokinghistory,years
<15 21(3.3)
15.1–30 140(22.0)
>30 474(74.6)
Drink1time/week 348(51.7)
Betelnutchewing 173(25.7)
Exercise3days/week 263(39.1)
Duringpast12months
Hadaphysicalexam 235(34.9)
Hospitalization 75(11.2)
Emergencyvisit 46(6.9)
Hadfunctionalimpairment 11(1.6)
Self-reportedmorbidconditions
Hypertension 124(18.4)
Diabetes 66(9.8)
Heartdisease 56(8.3)
Respiratorydisease 66(9.8)
GIproblems 111(16.5)
Statistical Package for the Social Sciences (SPSS version 15.0., Chicago,IL).Statisticalsignificanceforallanalyseswasevaluated at
a
=0.05.3. Results
Table 1 shows the characteristics of subjects. Among 2462 subjectsatbaseline,1267weremaleandamongthem,673were current smokers. Two-thirds smoked>15 cigarettes/dayand a greatmajority(97%)had15yearsormoreofsmokinghistory.
Table 2 shows the results of the binary logistic regression analysis.Duringthefirstthreeyears(1996to1999)ofthestudy period,becomingfunctionallyimpairedincreasedthelikelihoodof discontinuing smoking by 12-fold (odds ratio (OR)=12.9;
95%CI=3.07–54.4,p<0.001);olderage(60–66vs.50–59years) nearly doubled (p<0.05) the likelihood; but heavy cigarette consumption(>30cigarettes/day)decreasedthelikelihoodby4- fold(p<0.05)andalongsmokinghistoryof15–30yearsreduced the likelihood by11 times and a history of 30 years or longer reducedthelikelihoodby30times.Duringthelaterfour(4th–7th) years,becomingfunctionallyimpairedincreasedthelikelihoodof discontinuingsmokingbynearly14-fold(OR=13.7,95%CI=3.26–
57.7,p<0.001);livingwithaspouseincreasedthelikelihoodby about10-fold(OR=10.21,95%CI=2.47–42.16,p<0.001);quitting alcohol-drinking,havingafirsthospitalization,afirstemergency visitoranewdiagnosisofheartdiseaseallsignificantlyincreased thelikelihood ofdiscontinuingsmoking(allp<0.05). Havinga firsthealth examinationwasonlynearlysignificant(p<0.057).
Longeryearsofformaleducation(7years)andlongersmoking history (15–30 years) significantly reduced the likelihood of discontinuingsmoking.Othervariables includingalcohol-drink- ing, betel-nut chewing and physical activity, hypertension, diabetes, heart disease, respiratory diseaseand gastrointestinal disorders showed no significant association with changes in smokingstatus.Thereweresignificant(p<0.05)interactionswith timeinsomevariables.Olderage,moreformaleducation,living withaspouse,quittingdrinking,havingthefirsthealthcheck-up, firsthospitalizationorfirstemergencyvisit,becomingfunctionally dependentandhavinganewdiagnosisofCHDallproducedgreater impactsonchangingthesmokingstatusduringthelastfouryears thanduringthefirstthreeyearsofthestudy.Thereversewastrue forquantityandyearsofsmoking.Allbehavior-andhealth-related variables examined showedrelatively goodstability duringthe studyperiods(allp<0.001)(Table3).
Table2
Binarylogisticregressionmodelsevaluatingtheprobabilityofquittingsmokingin1999or2003predictedby1996sociodemographicandlifestylevariablesandchangesin healthstatusesin50–66yearsoldmaleTaiwanesesmokersoftheSHLSETstudy.
Variables Groups
1996–1999(n=569)a 1999–2003(n=426)a
%quit OR(95%CI) p< %quit OR(95%CI) p<
Agein1996
50–60 13.4 1 17.3 1
60–66 17.7 1.91(1.01–3.60) 0.047 26.6 1.60(0.86–2.96) 0.135*,b
Yearsofformaleducation
0 12.9 1 31.9 1
1–6 16.1 1.34(0.57–3.15) 0.502 19.0 0.51(0.24–1.09) 0.081
7 14.5 0.93(0.34–2.52) 0.885 17.5 0.39(0.16–0.95) 0.038***
Livingwithaspouse
No 15.0 1 7.3 1
Yes 15.1 0.88(0.38–2.04) 0.758 22.6 10.21(2.47–42.16) 0.001**
Smokingquantity
0–15 21.2 1 20.1 1
16–30 13.0 0.64(0.34–1.19) 0.157 21.9 1.16(0.61–2.22) 0.651
>30 6.8 0.25(0.07–0.86) 0.028 17.2 0.74(0.27–2.03) 0.564***
Smokinghistory,years
0–15 67.3 1 33.3 1
16–30 15.4 0.09(0.04–0.20) 0.001 16.8 0.21(0.05–0.92) 0.039***
>30 7.8 0.03(0.01–0.07) 0.001 21.9 0.29(0.07–1.18) 0.084
Quitdrinking
No 14.6 1 17.7 1
Yes 18.9 1.30(0.46–3.69) 0.617 37.3 3.92(1.74–8.85) 0.001***
Startedhealthcheck-up
No 15.0 1 19.4 1
Yes 14.9 0.76(0.35–1.65) 0.491 24.0 2.38(0.97–5.83) 0.057**
Firsttimehospitalized
No 13.4 1 16.4 1
Yes 26.7 1.60(0.60–4.27) 0.344 46.0 2.56(1.09–5.99) 0.030***
Firsttimeemergencyvisit
No 14.0 1 16.5 1
Yes 23.8 1.36(0.46–4.03) 0.582 48.1 3.67(1.53–8.83) 0.004***
Becamefunctionallydependent
No 13.8 1 18.7 1
Yes 58.8 12.92(3.07–54.36) 0.001 70.6 13.72(3.26–57.66) 0.001*
NewCHDdiagnosis
No 14.7 1 18.9 1
Yes 20.7 2.24(0.66–7.57) 0.194 39.5 2.79(1.11–7.00) 0.029**
aBothmodelsalsoincludedthefollowingvariables:alcohol-drinking,betel-nutchewing,physicalactivity,healthcheck-up,hypertension,diabetes,heartdisease, respiratorydisease,gastrointestinaldisordersandnewdiagnosisofhypertension,diabetes,respiratorydisease,andgastrointestinaldisorders.Allthesevariableswerenot significantlyassociatedwithachangeinsmokingstatus.
b Indicatingthesignificantdifferencesinthestrengthofassociationbetweenthetwosurveys.
*p<0.05.
**p<0.01.
*** p<0.001.
4. Discussion
4.1. Majorpredictorsofchangingsmokingstatus
Resultssuggestthatconcernoverhealthisthechiefreasonfor quittingsmokingin50–66-year-oldmaleTaiwanesesmokers,but notallhealthconcernshavethesamelevelofimpact.Havinganew functionalimpairment(animpairmentoccurredaftertheinitiation ofthestudy)wasthestrongestpredictorofchangingsmokingstatus.
Havingafirstemergencyvisitorhospitalizationalsosignificantly increasedthelikelihoodofchangingsmokingstatus,especiallyin oldersmokers.However,havinganewdiagnosisofhypertension, diabetesorgastrointestinaldisordershowednosignificantimpact on changing smoking status. These results suggest that merely havinga‘‘silent’’healthconditionwithoutfunctionalimpairmentor a traumatic experience would not induce enough ‘‘fear or motivation’’tocausea changeinsmokingstatus.Thesefindings aregenerallyinlinewithobservationsmadebyHonjoet al.(2010) whohaveobservedthatnewhealthconcerns(theinitiationofa prescribed drug and the development of a new disease) were significantpredictorsofchangingsmokingstatusin40–59yearsold Japanese.Ourfindingsarealsoinlinewiththeobservationsmadein theFraminghamstudywhichshowedthatrecenthospitalization andthedevelopmentofcoronaryheartdiseasewerepredictorsof changingsmoking status,but diagnosis ofcancer or changesin pulmonaryfunctionwerenot(Freundet al.,1992).Twardellaetal.
(2006)alsoobservedthatthediagnosisofasmoking-relateddisease was the strongest predictor of smoking cessation in Germans.
However, the relative cessation rates in the year of disease occurrence were different among various conditions: 11.2 for myocardialinfarction,7.2forstroke,2.5fordiabetesmellitusand4.8 for cancer relative to years before diagnosis of the respective diseases.Dubeet al.(2009)andSachs-Ericssonet al.(2009)also observedthatsmokingcessationwasassociatedwithhigherlevels ofdistressandhealthproblems.Olderadultsmokerswithhigher levels of psychological distress and health problems are more motivatedtoquitsmoking thanthosewhohave fewerof these problems.But,theconnectionbetweenhealthstatusandquitting intentionsappeartoweakenafterage65(Szantoet al.,2005).
Interestingly,arecentstudybyYanget al.(2009)hasshown thathealthconcernforfamilymemberswasbuthealthconcernfor selfwasnotasignificantreasonforquittingsmokinginadultmale Chinese.Theseresultssuggestthatalthoughconcernoverhealthis themajorreason forquittingsmokinginolderadults, different ethnicities/populationsmayreactdifferentlytothesamemedical conditions(Hymowitzetal.,1997;Hylandet al.,2004).
4.2. Otherpredictors
Manysociodemographicfactorsalsohavebeenobservedtobe associatedwithsmokingcessation.Thepresentstudyfoundthat
oldermen(60–66yearsold)generallyquitsmokingatahigherrate comparedtoyoungercounterparts(50–60yearsold),afindingin linewithmanyearlierobservations(Hymowitzetal.,1997;Osler andPrescott,1998;Hendersonetal.,2004;Pomerleauetal.,2004;
GrotvedtandStavem,2005;vanLoonetal.,2005;Sachs-Ericsson etal.,2009;Ayo-YusufandSzymanski,2010).Menwhohadmore than6yearsofformaleducationwerelesslikelytoquitsmokingat leastduringthelaterfouryearsofthestudy.Educationorsocial status has generally been found to be a positive factor with smokingcessationinsome(OslerandPrescott,1998;Twardellaet al., 2006) but not all studies (Giordano and Lindstrom, 2010).
Livingwithaspouseincreasesthelikelihoodofquittingsmoking but only during the later four years. Support mechanisms (via marriageoremployment)andelementsofsocialcapital(measured by ‘trust’ and ‘social participation’) have been shown to be independentlyandpositivelyassociatedwithsmokingcessation (Osleretal., 1999;Hyland etal., 2004;Pomerleau etal.,2004;
Honjoet al.,2006;Ayo-YusufandSzymanski,2010;Giordanoand Lindstrom,2010). However,beingunmarriedwasfoundtobea factor associatedwithquitting smokingin adultChinesemales (Yang et al., 2009). The differences of these effects among populationsagainsuggesttheinteractivenatureofthevariables andtheroleofcultureinshapingsmokingcessation.
Lifestylefactorsespeciallyalcohol-drinkinganddailyexercise havealsobeenshowntobesignificantlyassociatedwithsmoking cessation.Alcohol-drinkingisnegativelyassociatedwithsmoking cessationanddrinkersdelaytheirsmokingcessationtolateryears compared to non-drinkers. Quitting drinking is positively associated with smoking cessation perhaps because these behaviors are similar lifestyle vices and have similar health effects(Osler etal., 1999;Hylandet al.,2004).Thus,quitting alcohol-drinking may be considered a surrogate of smoking cessation rather than a causal factor of smoking cessation.
Smokingcessation hasbeen showntobeinfluenced by socio- environmentalandlifeeventsaccordingtostagesoflife(Kvizet al.,1994;GrotvedtandStavem,2005).
Thepredictivevalueofabehavior-orhealth-relatedvariableis highlydependentonthestabilityofthevariableovertime.Inthe present study most of these variables examined showed a relatively high stability during the study duration. However, physicalactivityandsomehealthconditions,asexpected,areless stable.Physicalactivitywilldeclineasfunctionalabilitydecreases withagingwhereasahealthconditioncanchangeasacondition progresses.
Contrary to the expectation, quitting chewing betel-nut (another lifestyle-vice for Taiwanese and Southeastern Asians, similartochewingtobaccoinWesterncountries)isnotassociated withquittingsmoking.Thelownumberofbetel-nutchewerswho quitsmokingcouldbeareasonbuttherealanswerisnotknown.
Havingafirstphysicalcheckupwasnearlypositivelyassociated withquittingsmoking(p=0.057).Anincreaseintheallowanceof thefrequencyoffreehealthexaminationsinthisagerangebythe Universal Health Insurance Policy in Taiwan during the study periodmighthaveobscuredthepossibleassociation.
Similartofindingsofothers(Hymowitzetal.,1997;Osleretal., 1999; Yanget al., 2009)thecurrentstudyshowsthat quitting smoking is negatively associated with the length of smoking historyanddailycigaretteconsumptionespeciallyduringthelater period.Longer-termandheavier-volumesmokersaremorelikely to postpone smoking cessation to later years. Daily cigarette consumptionisafactorbutnotascriticalasthelengthofsmoking history.Thesefindings supportthefindingthatnicotinedepen- dence is a major predictor of smoking cessation in long-term smokers(Hylandetal.,2004).Lowervolume-andshortersmoking history-smokersaremorelikelytoquitsmokingthanheavierand longer-timesmokers(Osleretal.,1999;Hendersonet al.,2004).
Table3
Stabilityofbehavioralandhealth-relatedvariablesbetween1996–1999and1996–
2003surveysaccordingtoSpearman’scorrelation.
Variables 1996–1999 1999–2003
r n r n
Alcohol-drinking 0.557 582 0.523 502
Betel-nutchewing 0.554 582 0.581 502
Physicalactivity 0.280 583 0.329 503
Hypertension 0.590 583 0.654 503
Diabetes 0.670 583 0.607 503
Heartdisease 0.440 581 0.195 502
Respiratorydisease 0.287 583 0.318 503
Gastrointestinaldisorders 0.334 580 0.300 501
Allcorrelationsaresignificantatp<0.001.
Thepresent studyalsoshows that several variables interact with time in effecting a decision to quit smoking. These interactions could be the result of many factors including but notlimitedtochangesingovernmentaltobaccocontrolpolicies, health-promotionstrategies,cigarettetaxburdenandtheUniver- salHealthInsurancepolicies,theanti-smokingsentimentofthe public and the tobacco industry activities. Unfortunately, the datasetprecludestheanalysisofthesevariables.
4.3. Implicationoffindings
Thehealthbenefitofquittingsmokinghasbeenshowntobebig even in older smokers. Smoking cessation reduces overall mortality,cardiovascular mortalityand cancer-relatedmortality within5yearsofquitting,andinsomecases,therisksarereduced tothelevelsofneversmokers(BjartveitandTverdal,2009;Menzin et al., 2009). A recent study has shown that older smokers, particularlythose hospitalized withcardiovascular disease, can quitathighrateswhenprovidedanintervention.Furthermore,use ofnicotinereplacementtherapyhasbeenshowntobesafeand efficacious among older smokers with cardiovascular disease (Doolan and Froelicher, 2008). However, it is important that smoking cessation programs be tailored to the relevant target groups because there are significant age-related differences in reasons of cessation among smokers over the course of their smoking careers. Developing effective methods to provide behavioraland/orpharmacologicalsupportforthosewhoattempt toquitsmokingisthekeytogreatsuccess.Inolderadultssincethe concern over health is the chief reason for quitting smoking, physician counseling especially at the time of a new medical diagnosisshouldbeapriority(Murrayet al.,2009).
4.4. Strengthsandlimitations
Themajorstrengthofthisstudyisthatthedatasetisfroma longitudinalcohortstudyinvolvinga nationallyrepresentative sample.Butthestudyalsohassomelimitations:(a)dataareself- reportswhicharegenerallyacceptable butunavoidably would have some shortcomings; (b) lack of more detailed data on smokinghistoryandpastattemptsofquitting;(c)thereasonfor quittingsmokingwasnotspecifiedandwhetheritwasintentional wasnotknown;and(d)thelengthoftimeapersonstayedasapast smokerwasnotknown.Thelack ofsuchinformation,tosome degree,limitstheanalysisandinterpretationofthesedata.
5. Conclusion
Resultsofthepresentstudysuggestthatfunctionalimpairment isthestrongestpredictorofsmokingcessationin50–66-yearmale Taiwanesesmokers.Havingafirstemergencyvisit,hospitalization or a first diagnosis of heart disease was also associated with increasedsmokingcessation.Resultsofthepresentstudyshould provide useful information for designing effective smoking cessation programsfor thelate middle ageor young-old male Taiwanesesmokers.
Conflictofintereststatement None.
Acknowledgements
TheauthorswishtothanktheBureauofHealthPromotionof TaiwanforprovidingthedataoftheSHLSETforthisanalysis.The presentstudyreceivednospecificgrantfromanyfundingagency inthepublic,commercialornot-for-profitsectors.
References
Ayo-Yusuf,O.A.,Szymanski,B.,2010.Factorsassociatedwithsmokingcessationin S.Africa.S.Afr.Med.J.100,175–179.
Bjartveit,K.,Tverdal,A.,2009.Healthconsequencesofsustainedsmokingcessation.
Tob.Control18,197–205.
Breitling,L.P.,Rothenbacher,D.,Stegmaier,C.,Raum,E.,Brenner,H.,2009.Older smokers’motivationandattemptstoquitsmoking:epidemiologicalinsight intothequestionoflifestyleversusaddiction.Dtsch.Arztebl.Int.106,451–
455.
BureauofHealthPromotion,2007a.Introductiontothe1996SurveyofHealthand LivingStatusoftheElderlyinTaiwan. http://www.bhp.doh.gov.tw/BHPnet/
Portal/file/ThemeDocFile/200712270515350682.pdf(accessed10.08.10).
BureauofHealthPromotion,2007b.Questionnaireofthe1996SurveyofHealthand LivingStatusoftheElderlyinTaiwan. http://www.bhp.doh.gov.tw/BHPnet/
Portal/Them_Show.aspx?Subject=200712270002&Class=2&No=200712270015 (accessed10.08.10).
Chang,H.-H.,Chiang,T.-L.,2009.Depressivesymptoms,smoking,andcigarette priceelasticity:resultsfromapopulation-basedsurveyinTaiwan.Int.J.Public Health54,421–426.
Chung,W.,Kim,H.,Lim,S.,Lee,S.,Cho,K.,2009.Factorsinfluencingcigarette smokingandquantifiedimplicationsforanti-smokingpolicy:evidencefrom SouthKorea.Int.J.PublicHealth54,409–419.
Doolan,D.M., Froelicher,E.S.,2008. Smokingcessation interventionandolder adults.Progr.Cardiovasc.Nurs.23,119–127.
Dube,S.R.,Caraballo,R.S.,Dhingra,S.S.,Pearson,W.S.,McClave,A.K.,Strine,T.W., Berry,J.T.,Mokdad,A.H.,2009.Therelationshipbetweensmokingstatusand seriouspsychologicaldistress:findingsfromthe2007BehavioralRiskFactor SurveillanceSystem.Int.J.PublicHealth54,S68–S74.
Eguchi,K.,Kario,K.,Hoshide,S.,Hoshide,Y.,Ishikawa,J.,Morinari,M.,Hashimoto, T.,Shimada,K., 2004.Smoking isassociatedwithsilentcerebrovascular diseaseinahigh-riskJapanesecommunity-dwellingpopulation.Hypertens.
Res.27,747–754.
Fitti,J.E.,Koval,M.G.,1987.Thesupplementonagingtothe1984NationalHealth InterviewSurvey.VitalHealthStat.121,1–115.
Freund,K.M.,D’Agostino,R.B.,Belanger,A.J.,Kannel,W.B.,Stokes,J.,1992.Predictors ofsmokingcessation:theFraminghamStudy.Am.J.Epidemiol.135,957–964.
Freund,K.M.,Belanger,A.J.,D’Agostino,R.B.,Kannel,W.B.,1993.Thehealthrisksof smoking.TheFraminghamStudy:34yearsoffollow-up.Ann.Epidemiol.3, 417–424.
Giordano,G.N.,Lindstrom,M.,2010.Theimpactofsocialcapitalonchangesin smoking behavior: a longitudinal cohort study. Eur. J. Public Health, doi:10.1093/eurpub/ckq048.
Grotvedt,L.,Stavem,K.,2005.Associationbetweenage,genderandreasonsfor smokingcessation.Scand.J.PublicHealth33,72–76.
Henderson,P.N.,Rhoades,D.,Henderson,J.A.,Welty,T.K.,Buchwald,D.,2004.
SmokingcessationanditsdeterminantsamongolderAmericanIndians:The StrongHeartStudy.EthnicDis.14,274–279.
Honjo,K.,Tsutsumi,A.,Kawachi,I.,Kawakami,N.,2006.Whataccountsforthe relationshipbetweensocialclassandsmokingcessation?Resultsofapath analysis. Soc.Sci.Med.62,317–328.
Honjo,K.,Iso,H.,Inoue,M.,Tsugane,S.,2010.Smokingcessation:predictivefactors amongmiddle-agedJapanese.NicotineTob.Res.12,1050–1054.
Hyland, A.,Li, Q.,Bauer,J.E.,Giovino,G.A.,Steger,C.,Cummings,K.M.,2004.
Predictorsofcessationinacohortofcurrentandformersmokersfollowed over13years.NicotineTob.Res.6(Suppl.3),S363–S369.
Hymowitz,N.,Cummings,K.M.,Hylan,A.,Lynn,W.R.,Pechacek,T.F.,Hartwell,T.D., 1997.Predictorsofsmokingcessationinacohortofadultsmokersfollowedfor fiveyears.Tob.Control6(Suppl.2),S57–S62.
Johnson,J.K.,Lui,L.Y.,Yaffe,K.,2007.Executivefunction,morethanglobalcogni- tion,predictsfunctionaldeclineandmortalityinelderlywomen.J.Gerontol.A:
Biol.Sci.Med.Sci.62,M1134–M1141.
Kato,J.,Aihara,A.,Kikuya,M.,Matsubara,M.,Ohta,M.,Ohkubo,T.,Tsuji,I.,Sekina, H.,Mequro,T.,Imai,Y.,2001.Riskfactorsandpredictorsofcoronaryarterial lesionsinJapanesehypertensivepatients.Hypertens.Res.24,3–11.
Kviz,F.J.,Clark,M.A.,Crittenden,K.S.,Freels,S.,Warnecke,R.B.,1994.Ageand readinesstoquitsmoking.Prev.Med.23,211–222.
Menzin,J.,Lines,L.M.,Marton,J.,2009.Estimatingtheshort-termclinicaland economicbenefitsofsmokingcessation:dowehaveitright? ExpertRev.
Parmacoeco.OutcomesRes.9,257–264Retrievedfrom:http://dx.doi.org.prox- y.lib.umich.edu/10.1586/erp.09.28.
Molarius,A.,Parsons,R.W.,Dobson,A.J.,Evans,A.,Fortmann,S.P.,Jamrozik,K., Kuulasmaa,K.,Moltchanov,V.,Sans,S.,Tuomilehto,J.,Puska,P.,WHOMONICVA Project,2001.Trendsincigarettesmokingin36populationsfromtheearly 1980stothemid-1990s:FindingsfromtheWHOMONICSProject.Am.J.Public Health91,206–212.
Murray, R.L.,Lewis,S.A.,Coleman,T., Britton,J.,McNeill,A.,2009. Unplanned attemptstoquitsmoking:missedopportunitiesforhealthpromotion? Addic- tion104,1901–1909.
Ockene,I.S., Miller,N.H.,1997.Cigarette smoking,cardiovascular disease,and stroke:a statementfor healthcareprofessionals fromtheAmericanHeart Association:AmericanHeartAssociationTaskForceonRiskReduction.Circu- lation96,3243–3247.
Osler,M.,Prescott,E.,1998.Psychosocial,behavioural,andhealthdeterminantsof successful smoking cessation:a longitudinalstudyof Danishadults.Tob.
Control7,262–267.
Osler,M.,Prescott,E.,Godtfredsen,N.,1999.Genderanddeterminantsofsmoking cessation:alongitudinalstudy.Prev.Med.29,57–62.
Peto,R.,Lopez,A.D.,Boreham,J.,Thun,M.,Heath,C.,1992.Mortalityfromtobaccoin developedcountries:indirectestimationfromnationalvitalstatistics.Lancet 339,1268–1278.
Pomerleau,J.,Gilmore,A.,McKee,M.,Rose,R.,Haerpfer,C.W.,2004.Determinantsof smokingineightcountriesoftheformerSovietUnion:resultsfromtheLiving Conditions,LifestylesandHealthStudy.Addiction99,1577–1585.
Sachs-Ericsson,N.,Schmidt,N.B.,Zvolensky,M.J.,Mitchell,M.,Collins,N.,Blazer,D.G., 2009.Smokingcessationbehaviorinolderadultsbyraceandgender:theroleof healthproblemsandpsychologicaldistress.NicotineTob.Res.11,433–443.
Siahpush,M., Borland, R., Yong,H.H., Kin,F., Sirirassamee, B., 2008.Socio- economicvariationsintobacco consumption,intention toquitand self- efficacytoquitamongmalesmokersinThailandandMalaysia:resultsfrom theInternationalTobaccoControl-South-EastAsia(ITC-SEA)survey.Addic- tion103,502–508.
Suwala,M.,Gerstenkorn,A.,Kaczmarczyk-Chalas,K.,Drygas,W.,2005.Tobacco smokingbyelderlypeopleaccordingtoCINDIWHOresearch.Przeglad.Lekarski 62(Suppl.3),55–59.
Suwala,M.,Drygas,W.,Gerstenkorn,A.,2006.Subjectiveandobjectivehealthstatus ex-smokersamongelderlypersons.CINDIWHOsurvey.Przeglad.Lekarski63, 1095–1098.
Szanto,Z., Susanszky,E., Kopp,M.,2005.Relationshipsbetweenunfavourable healthstatusandsmokingcessationattemptsinHungary.Soz.Preventivmed.
50,324–333.
Twardella,D.,Loew,M.,Rothenbacher,D.,Stegmaier,C.,Ziegler,H.,Brenner,H., 2006.Thediagnosisofasmoking-relateddiseaseisaprominenttriggerfor smokingcessationinaretrospectivecohortstudy.J.Clin.Epidemiol.59,82–89.
vanLoon,A.J.M.,Tijhuis,M.,Surtees,P.G.,Ormel,J.,2005.Determinantsofsmoking status:cross-sectiondataonsmokinginitiationandcessation.Eur.J.Public Health15,256–261.
Vangeli,E.,West,R.,2008.Sociodemographicdifferencesintriggerstoquitsmok- ing:findingsfromanationalsurvey.Tob.Control17,410–415.
WHO(WorldHealthOrganization),2009a.WHOReportontheGlobalTobacco Epidemic.ImplementingSmoke-FreeEnvironments. Retrievedfrom:http://
whqlibdoc.who.int/hq/2009/WHO_NMH_TFI_09.1_eng.pdf.
WHO(WorldHealthOrganization),2009b.TobaccoFreeInitiative:WHOFrame- workConventiononTobaccoControl. Retrievedfrom:http://www.who.int/
tobacco/mpower/2009/b_gtcr_whofctc.pdfindex.html.
WHO(WorldHealthOrganization),2010.TobaccoFreeInitiative.Retrievedfrom:
http://www.who.int/tobacco/en/index.html.
Yang,T.,Abdullah,A.S.,Mustafa, J.,Chen, B.,Yang,X.,Feng,X.,2009. Factors associatedwithsmokingcessationamongChineseadultsinruralChina.Am.
J.HealthBehav.33,125–134.
