SUMMARY
summary Theaimofthisthesiswastodescribethedevelopmentandevaluationofanindividual basedlifestyleinterventionamongworkersintheconstructionindustrywithanelevated risk of cardiovascular disease (CVD). In chapter 1, an introduction to this topic was provided. CVD is a chronic and degenerative disease, of which the risk is strongly influencedbylifestylebehaviorssuchasdiet,physicalactivity,andsmoking.thereis evidencefortheeffectivenessofindividualbasedinterventionsaimedatlifestylebehavior changeamongworkers.intheconstructionindustry,theproportionofworkerswhoare overweightoratriskofcvdisrelativelylarge,andnoguidelineforcvdriskreductionyet exists.inthisstudy,thefeasibilityand(cost)effectivenessofalifestyleinterventioninthe construction industry was investigated. In this thesis, the following objectives were addressed: 1) Toprovideanoverviewoftheevidencefortheeffectivenessofworkplace lifestyleinterventionsonprecursorsofcvd; 2) To describe the design of the Health under Construction study, the characteristicsoftheparticipants,andtheevaluationoftheintervention process; 3) TopresenttheshortandlongtermeffectsoftheHealthunderConstruction studyonlifestyleandprecursorsofcvd,aswellasaneconomicevaluation. Chapter2concernedasystematicliteraturereviewontheeffectivenessofworkplace lifestyle interventions. To the 31 studies that fulfilled the inclusion criteria, a best evidence system was applied, taking into account the quality of the study and the consistencyofeffects.strongevidencewasfoundforapositiveeffectofworkplace lifestyleinterventionsonbodyfat,oneofthestrongestpredictorsofcvdrisk.among populationswithanelevatedriskofcvd,therewasstrongevidenceforapositiveeffect onbodyweightaswell.duetoinconsistenciesinresultsbetweenstudies,therewasno evidencefortheeffectivenessofworkplacelifestyleinterventionsonbloodpressure, cholesterol,triglycerides,andhba1c.populationswithanelevatedriskofcvdseemedto benefitmorefromlifestyleinterventionsthanpopulationsnotatrisk,andsupervised exerciseinterventionsappearedlesseffectivethangrouporindividualcounseling.for futureinterventionstudieswerecommendtoreportparticipants compliancewiththe interventionandthelifestylechangesachieved,inordertogainbetterinsightintothe mechanismsthatledtotheinterventioneffects. Inchapter3,thedesignofHealthunderConstructionstudywasthoroughlydescribed. Health under Construction is a randomized controlled trial for male workers in the constructionindustry,boththoseinvolvedinconstructionactivitiesandinadministration, supervision, and management, with an elevated risk of CVD, in which usual care is comparedtoanindividualbasedlifestyleintervention.theinterventionwasbasedon opinionsofemployers,employees,andoccupationalphysicians,andonaliteraturesearch onthecurrentevidencefortheeffectivenessoflifestyleinterventions.theintervention 172
summary consistedofthreefacetofaceandfourtelephonecontactswithanoccupationalphysician ornurse,oneitherimprovingdietorphysicalactivitybehavior,orsmokingcessation.the counselors followed a stepwise protocol and used motivational interviewing as a counselingstyle.measurementstookplaceatbaseline,6,and12months.bymeansof questionnaires,datawerecollectedone.g.lifestylebehaviorandabsenteeism.doctors assistantsmeasuredbodyweight,hdlandtotalcholesterol,systolicanddiastolicblood pressure,andhba1c. Inchapter4,weprovidedinsightintothefactorsassociatedwithnonparticipationand dropoutinthehealthunderconstructionstudy.toexaminetheassociationsbetween (non)participationandcvdriskfactors,andtheassociationsbetween(non)dropoutand CVDriskfactors,weusedcrudeandmultiplelogisticregressionmodels.Bymeansof questionnaires, the reasons for nonparticipation and dropout were assessed. The participants,i.e.20%ofallinvited,hadaworsecvdriskprofilethannonparticipantswith respecttobloodpressure,cholesterol,tirednessand/orstress,andchestpainand/or shortnessofbreath.theworsecvdriskprofilewasmainlyexplainedbythedifferencein age;participantswere3.8yearsolderthannonparticipants.dropoutswere4.6years youngerthanthosewhocompletedthestudy,andmorelikelytosmoke.thus,thestudy completerswereonaverageolderandlesslikelytobeasmokeratbaselinethanthose invited.themainreasonsfornonparticipationwere nointerest, current(para)medical treatment, and feeling healthy, and for dropout the main reason was a lack of motivation. Inchapter5,weevaluatedthepracticalexecutionoftheintervention,aswellasthe opinionsofthecounselorsandparticipants.theadherenceofthe27counselorstothe interventionprotocolwasdeterminedbyregistrationofthenumberofsessionsanditems discussed,andbymeasuringthequalityofmotivationalinterviewingusingexpertscoring ofrandomsegmentsof19counselingsessions.counselors'competencewasratedby participantsandcounselorsseparately.associationsbetweenthreeprocessindicatorsand bodyweightlossbetweenbaselineand6monthsweredeterminedusinglinearregression analyses. Twothirds of all participants attended five or more sessions, and 38.5% attendedallsevensessions.in90.2%ofallcases,thecounselordiscussedallobligatory itemsinthefirstsession.adherencetomotivationalinterviewingwasreachedinonlyone audiotapedfragment.86.3percentofallparticipantsagreedwiththecounselorbeing competent.neitherperceivedcounselors competence,nornumberofsessionsoritems discussed,wassignificantlyassociatedwithbodyweightloss. In chapter 6, the effects of the intervention on lifestyle behaviors were described. Completedatawereavailablefor595participants.Participantswhohadchosentoaimat dietandphysicalactivity(energybalance)wereanalyzedseparatelyfromtheoneswho aimedatsmokingcessation.effectsizesweredeterminedbylinearandlogisticregression analysesinwhichthebaselinevaluewasaddedasacovariate.intheenergybalance 173
summary subgroup,theinterventionhadasignificantbeneficialeffectonsnackintake(piecesper week:1.9,95%ci3.7;0.02)andfruitintake(piecesperweek:1.7,95%ci0.6;2.9)at 6months.Theeffectonsnackintakeremainedsignificantat12months(1.9,95%CI 3.6;0.2).At6months,31.3%ofparticipantshadquitsmoking,ascomparedto13.4%in thecontrolgroup(orforsmoking0.3,95%ci0.1;0.7),butthiseffectwasnotsustained until12months(or0.8,95%ci0.4;1.6).bothcontrolandinterventiongroupparticipants substantiallyincreasedtheirleisuretimephysicalactivitybyalmost1.5and2.5hoursper weekrespectively. Inchapter7,wedescribedtheinterventioneffectsonprecursorsofCVD,i.e.thebiological riskfactors.completedatawereavailablefor517participants.theinterventionhad significanteffectsat6monthsonbodyweight(1.9,95%ci2.6;1.2)anddiastolicblood pressure(1.7,95%ci3.3;0.1).theeffectonbodyweighteffectwassustaineduntil12 months(1.8,95%ci2.6;1.1).amongparticipantswhohadaimedatenergybalance, theinterventionhadasignificanteffectonbodyweightat6(2.1,95%ci2.9;1.3)and 12months(2.2,95%CI3.1;1.3),andonHDLcholesterol(0.05,95%CI0.01;0.10)and HbA1c(,0.06,95%CI0.12;0.001)at12months.TheeffectsonHDLcholesteroland HbA1covertime,asdeterminedbylongitudinalanalysis,werenotsignificant.Onaverage, theobeseparticipantsachievedthelargestimprovementsinbodyweight,bloodpressure, and HDL cholesterol. Among participants who had aimed at smoking cessation, the intervention had no significant effects; beneficial changes in blood pressure and cholesterol were found in both groups, and even body weight decreased in the interventiongroup. Inchapter8,thecosteffectivenessfromthesocietalperspective,andthecostbenefit fromtheemployer sperspectiveweredescribed.weincludedonlythoseparticipantswho hadchosentoaimatenergybalance,i.e.573.allmissingdatawereimputedbymultiple imputations.forthecosteffectivenessanalyses,allcostsfortheintervention,healthcare useparticipantspurchasesrelatedtolifestyle,andabsenteeismweretotaledanddivided bytheincrementaleffectonbodyweight.forthecostbenefitanalysis,thecostsforthe interventionweresubtractedfromtheincrementalbenefitsassociatedwithabsenteeism. Anincrementalcosteffectivenessratiowascalculated,ofwhichtheuncertaintywas estimatedbybootstrappingcost/effectpairs.theinterventionwasmoreeffectivebut alsomoreexpensivethanusualcare.foroneadditionalkgofbodyweightloss,thecosts forthesocietywouldbe 145.Incaseofawillingnesstopayof 2,000,theprobabilityof costeffectivenesswouldbe0.95.thenetemployercostsresultingfromtheintervention were 254(95%CI;1,070;1,536);thustheinterventioncannotberegardedascost saving. Inchapter9,wesummarizedanddiscussedourfindings,outlinedthestrengthsand limitations of our study, described the public and occupational health context, and providedrecommendationsforresearchandpractice.inconclusion,thissystematically 174
summary developed, wellappreciated, and partly feasible intervention led to sustained improvementsindietandbodyweight.theresultscanbegeneralizedtotheoldermale workersintheconstructionindustrywithanelevatedriskofcvd,boththeonesinvolved inconstructionactivitiesandthoseinsupervisionandmanagement.at12months,the intervention was more effective but also more expensive than usual care. Still, a workplaceinterventionthatiseffectiveinmakingsustainedchangesinlifestyleandhealth is of high relevance for occupational health, as it may contribute to sustained employabilityinthisagingpopulation.therefore,afterincorporatingoursuggestionsfor improvement,werecommendimplementationofthisintervention. 175