Alcohol-Related Road Casualties in Official Crash Statistics. International Traffic Safety Data and Analysis Group.

Transcription

1 Alchl-Related Rad Casualties in Official Crash Statistics Internatinal Traffic Safety Data and Analysis Grup Research Reprt

2 Alchl-Related Rad Casualties in Official Crash Statistics Internatinal Traffic Safety Data and Analysis Grup Research Reprt

3

4 The Internatinal Transprt Frum The Internatinal Transprt Frum is an intergvernmental rganisatin with 57 member cuntries. It acts as a think tank fr transprt plicy and rganises the Annual Summit f transprt ministers. ITF is the nly glbal bdy that cvers all transprt mdes. The ITF is plitically autnmus and administratively integrated with the OECD. The ITF wrks fr transprt plicies that imprve peples lives. Our missin is t fster a deeper understanding f the rle f transprt in ecnmic grwth, envirnmental sustainability and scial inclusin and t raise the public prfile f transprt plicy. The ITF rganises glbal dialgue fr better transprt. We act as a platfrm fr discussin and prenegtiatin f plicy issues acrss all transprt mdes. We analyse trends, share knwledge and prmte exchange amng transprt decisin-makers and civil sciety. The ITF s Annual Summit is the wrld s largest gathering f transprt ministers and the leading glbal platfrm fr dialgue n transprt plicy. The Members f the ITF are: Albania, Armenia, Argentina, Australia, Austria, Azerbaijan, Belarus, Belgium, Bsnia and Herzegvina, Bulgaria, Canada, Chile, China (Peple s Republic f), Cratia, Czech Republic, Denmark, Estnia, Finland, France, Frmer Yugslav Republic f Macednia, Gergia, Germany, Greece, Hungary, Iceland, India, Ireland, Israel, Italy, Japan, Krea, Latvia, Liechtenstein, Lithuania, Luxemburg, Malta, Mexic, Republic f Mldva, Mntenegr, Mrcc, the Netherlands, New Zealand, Nrway, Pland, Prtugal, Rmania, Russian Federatin, Serbia, Slvak Republic, Slvenia, Spain, Sweden, Switzerland, Turkey, Ukraine, the United Kingdm and the United States. Internatinal Transprt Frum 2, rue André Pascal F Paris Cedex 16 cntact@itf-ecd.rg Any findings, interpretatins and cnclusins expressed herein are thse f the authrs and d nt necessarily reflect the views f the Internatinal Transprt Frum r the OECD. Neither the OECD, ITF nr the authrs guarantee the accuracy f any data r ther infrmatin cntained in this publicatin and accept n respnsibility whatsever fr any cnsequence f their use. This dcument and any map included herein are withut prejudice t the status f r svereignty ver any territry, t the delimitatin f internatinal frntiers and bundaries and t the name f any territry, city r area.

5 4 ACKNOWLEDGEMENTS Authrs f the reprt This reprt was prepared by Mr Luuk Vissers (SWOV, the Netherlands), Mr Sjerd Huwing (SWOV, the Netherlands) and Mr Fred Wegman (Delft University f Technlgy). Acknwledgements This study is carried ut with the indispensable help f thers. We gratefully acknwledge the help frm bth the IRTAD and the OISEVI Secretariats. We especially thank Vernique Feypell, fr her advice in and the distributin f the questinnaire t all the IRTAD-members. Als the help f Crina Pupp and Virginia Alvarez in translating the survey int Spanish and distributing and cllecting the questinnaire and respnses fr the OISEVI-members is highly appreciated. Klaus Machata, David Silcck, Kathy Stewart and Pilar Zri are als gratefully acknwledged fr their recmmendatins n the initial draft f the survey. We als thank the reviewers f the final draft f this paper, Hrst Schulze and nce mre Kathy Stewart fr their critical and helpful review. Finally ur acknwledgements g t Hansje Weijer f SWOV fr her help in making this reprt well readable. IRTAD Since 1988, the Internatinal Traffic Safety and Analysis Grup (IRTAD) has been a permanent wrking grup f the Jint Transprt Research Centre f the OECD and the Internatinal Transprt Frum (ITF). It is cmpsed f rad safety experts and statisticians frm renwned safety research institutes, natinal rad and transprt administratins, internatinal rganizatins, universities, autmbile assciatins, the autmbile industry, and thers frm OECD and nn-oecd cuntries. IRTAD is bth a wrking grup and a database. Its main bjectives are t: Be a frum f exchange n rad safety data cllectin and reprting systems, and n trends in rad safety develpment and rad safety plicies. Cllect accident data and cnduct data analysis t cntribute t the wrk f the ITF/OECD, as well as t prvide advice n specific rad safety issues. Cntribute t internatinal c-peratin n rad accident data and its analysis. IRTAD currently includes mre than 70 rganizatins frm 37 cuntries which all have a direct interest in rad safety. It is the ambitin f IRTAD t be a wrld leader in rad safety by building and maintaining a high quality database n rad safety infrmatin, by being a netwrk fr rad safety specialists, by mnitring and analysing rad safety develpments in its member states, and by carrying ut rad safety data-related research. IRTAD has the aspiratin t further expand its cverage by making mre cuntries an active member f IRTAD. IRTAD ffers a mechanism fr the integratin f prspective member cuntries while assisting with imprvement f rad safety data cllectin systems, where needed. The IRTAD Grup c-perates with ther internatinal rganizatins, such as the Wrld Bank t invlve lw- and middle-incme cuntries in the wrk f the IRTAD Grup.

6 TABLE OF CONTENTS 5 Table f cntents Executive summary... 7 Chapter 1. Intrductin... 9 The underreprting issue Defining an alchl-related crash Cntent f the reprt References Chapter 2. Research methd Descriptin f the wrk Data cllectin and participants Questinnaire Chapter 3. Results f the survey Descriptin f respndents Legislatins, definitins and fficial statistics Data cllectin methds and surces Quality f the data Expert best estimate n the real number f alchl-related rad casualties Main results References Chapter 4. Methds fr adjusting the fficial number f alchl-related rad casualties fr underreprting Methds fr adjusting the fficial number f alchl-related rad fatalities Methds fr adjusting the fficial number f alchl-related serius rad injuries References Chapter 5. Cnclusins and recmmendatins Cnclusins Recmmendatins References Annex A. Changes in legislatin n bld alchl cncentratin Annex B. Questinnaire Figures Figure 3.1. Share f alchl-related rad fatalities in 2000, 2005 and Figure 3.2. Share f alchl-related serius rad injuries in 2000, 2005 and

7 6 TABLE OF CONTENTS Tables Table 3.1. Respnse f cuntries by rganisatin membership in percentage Table 3.2. Organisatin type f the respndents Table 3.3. Unit f measurement Table 3.4. Legal limits n bld alchl cncentratin levels definitin f an alchl-related rad fatality Table 3.5. Definitin f an alchl-related rad fatality Table 3.6. Definitin used fr alchl-related serius rad injuries Table 3.7. Share f alchl-related rad fatalities in cuntries with a present legal limit f 0.0 g/l frm fficial statistics Table 3.8. Share f alchl-related rad fatalities in cuntries with a present legal limit f 0.2 g/l frm fficial statistics Table 3.9. Share f alchl-related rad fatalities in cuntries with a present legal limit between 0.3 g/l and 0.4 g/l frm fficial statistics Table Share f alchl-related rad fatalities in cuntries with a present legal limit f 0.5 g/l frm fficial statistics Table Share f alchl-related rad fatalities in cuntries with a present legal limit f 0.8 g/l Frm fficial statistics Table Cuntries publishing fficial statistics n alchl-related serius rad injuries Table Cnditins t carry ut an alchl test Table Reasns fr the plice nt t carry ut alchl tests Table Cmparisn f expert best estimate with fficial statistics n alchl-related data n alchl-related rad casualties... 35

8 EXECUTIVE SUMMARY 7 Executive summary What we did This study examines hw imprving insights regarding the real number f alchl-related rad casualties wrldwide can help t save lives. Every year 1.25 millin peple die in rad crashes accrding t the Wrld Health Organizatin. It is widely recgnised that drink driving is an imprtant riskincreasing factr and cntributes t many rad deaths. With great certainty, the real number f alchlrelated rad casualties is higher than reprted in the fficial statistics. Better insights int reprting prcedures is f the utmst relevance t arrive at cmparable and reliable data. Fr this study, a ttal f 45 cuntries were surveyed with the help f an nline questinnaire. The survey was facilitated thrugh the members f the Internatinal Transprt Frum s permanent wrking grup n rad safety, knwn as the Internatinal Traffic Safety Data and Analysis Grup (IRTAD), the Iber-American Rad Safety Observatry (OISEVI) and the Internatinal Center fr Alchl Plicies (Internatinal Alliance fr Respnsible Drinking (IARD), frmerly ICAP), Washingtn.. The questinnaire lked at drinking and driving legislatin and at definitins f alchl-related rad fatalities and serius injuries. Based n the infrmatin prvided, the methds f recrding alchlrelated rad casualties and the quality f the data were reviewed, with specific attentin t the issue f underreprting. What we fund Previus research fund large differences in the share f alchl-related rad fatalities fr different cuntries, ranging between 2% and 38% f the ttal. This is cnfirmed by this study, which fund a share f alchl-related fatalities ranging frm apprximately 5% t 35%. The fficial data f the cuntries surveyed fr this study shw that a weighted average f 21.8% amng rad deaths are alchl-related. This prprtin remains cnstant ver the years ( ). Accepting this figure as a reasnable estimate fr all cuntries in the wrld and based n 1.25 millin annual rad fatalities wrldwide, the alchl-related deaths amng fatally injured rad users can be put at arund peple every year. There are indicatins, hwever, that this number underestimates the prblem, because fficial statistics are affected by underreprting f alchl-related crashes and casualties. As a result, the actual number f alchl-related fatalities is prbably higher still. (Serius injuries are even mre prne t underreprting). The vast majrity f cuntries surveyed (89%) still base their fficial data upn nly a single data surce. In mst cases these are the plice recrds (87.5%) fr which this study fund evident shrtcmings. These deficiencies in data cllectin negatively influence the accuracy and reliability f the fficial data regarding alchl-related rad casualties, which cmplicates meaningful internatinal cmparisns. Official statistics d nt give a cmplete picture f alchl-related fatalities and serius injuries if nly based plice recrds f fatal crashes. Furthermre, different cuntries use different definitins f what cnstitutes a rad crash casualty. Definitins regarding alchl-related serius rad injuries in particular differ substantially between cuntries. This reinfrces the distrtins created thrugh inaccurate recrding f crash data when

9 8 EXECUTIVE SUMMARY cmparing cuntries. Therefre relying n fficial statistics will ften be misleading. T enable mre accurate analyses with a few t addressing the prblem, imprvements are needed. What we recmmend Review hw data n alchl-related rad crashes is cllected In rder t cme t mre reliable and cmparable data n alchl-related serius rad injuries and fatalities, cuntries shuld begin by assessing their current status n the recrding f data n alchlrelated rad fatalities and serius injuries. In this assessment we suggest t check fr cmpliance with the fllwing gd-practices: Aim fr a systematic alchl testing f every rad user actively invlved in a serius crash Ideally, 100% f active rad users that are invlved in a rad crash that resulted in death r serius injury shuld be tested fr alchl. If a systematic alchl testing at this level is nt pssible, cuntries shuld apply additinal methds fr adjusting the fficial numbers f alchl-related rad fatalities and serius rad injuries. Use statistical analysis methds t better estimate the number f alchl-related rad fatalities Additinal statistical analysis methds as described in this study can help t btain better estimates f alchl-related serius rad crashed. Methds shuld be develped and applied that align with the legal system and data cllectin framewrk f individual cuntries, rather than harmnise methds internatinally. Harmnise definitins f alchl-related rad casualties T make fficial cuntry statistics cmparable, definitins f alchl-related rad casualties shuld be harmnised. A number f Eurpean cuntries have already adpted the definitin prpsed by the 2009 SafetyNet prject, as any death ccurring within 30 days as a result f a fatal rad crash in which any active participant was fund with a bld alchl level abve the legal limit. A similar apprach shuld be used t define a persn seriusly injured in an alchl related crash, based fr example n the severity level f 3+ n the Maximum Abbreviated Injury Scale (MAIS3+), s that it wuld be defined as any serius injury at MAIS3+ that ccurred as a result f a rad crash in which any active participant was fund with a bld alchl level abve the legal limit. If cuntries are unable t apply these recmmended definitins, develping algrithms t allw fr cnversin f these definitins is recmmended. Cnduct future research n hw t measure alchl-related rad crashes invlving pedestrians and cyclists In rder t make sure pedestrians and cyclists are als cunted as any active participant in the definitins n alchl-related fatalities and serius injuries further research is needed regarding the pssibilities f measuring bld alchl cncentratin fr pedestrians and cyclists invlved in rad crashes.

10 1. INTRODUCTION 9 Chapter 1. Intrductin Accrding t the Glbal Status Reprt n Rad Safety 2015 f the Wrld Health Organizatin (2015), every year 1.25 millin rad deaths ccur wrldwide. This is an estimate based n a survey amngst WHO member cuntries cmpleted by using a mdelling technique when reliable data was nt available. In cmparisn t 2007, when an earlier reprt was published, this number did nt change much despite glbal and natinal effrts t reduce the amunt f rad traffic deaths. One imprtant rad safety issue is drink driving. It is well dcumented that drink driving increases risks (Keall et al., 2004; Blmberg et al., 2005; Hels et al., 2011). Hwever, it is nt pssible t indicate precisely hw much, because the exact number f alchl-related casualties is unknwn. Official natinal statistics n rad fatalities related t impaired driving differ cnsiderably. Sme cuntries attribute a relatively small prprtin f rad fatalities t alchl use whereas fficial data in ther cuntries attributes almst half f all rad traffic deaths t alchl-related accidents. The fficial percentages f alchl-related rad fatalities in different cuntries fr the year 2010 range between 2% and 38% f all rad traffic deaths (WHO, 2013; WHO, 2015). Hw can this range be s wide and des it crrectly reflect reality?

11 10 1. INTRODUCTION The underreprting issue Several cuntries, particularly thse in which the fficial rad crash registratin is mainly based n plice crash reprts, face serius data quality prblems. The plice are nt infrmed abut every crash, and if they are infrmed, they d nt always fill ut a crash frm r make a reprt. This phenmenn is called underreprting and frm a data perspective this is a widespread and unfrtunate prblem (Derriks and Mak, 2007). Because certain crash types tend t be underreprted mre than thers, the resulting statistics are biased. Additinally, sme dubts are expressed abut the quality f fficial statistics cncerning the number f peple killed r seriusly injured attributed t drinking and driving. Many studies suggest that the fficial numbers f alchl-related rad casualties d nt seem reliable due t the prblem f general underreprting f rad crashes (Derriks and Mak, 2007; ETSC, 2010; ITF, 2011), r due mre specifically t underreprting in alchl-related rad casualties (Assum and Sørensen, 2010; COWI et al., 2014). This issue is ften nt mentined in fficial dcuments, which makes it difficult t knw whether these fficial statistics n alchl-related rad casualties are reliable (COWI et al., 2014). While develping a safety perfrmance indicatr fr drinking and driving, Assum and Sørensen (2010) cnclude that the validity f such an indicatr is pr and imprvements in reprting f drink-drivers are necessary t establish a reliable and valid safety perfrmance indicatr (SPI). Nt all cuntries systematically test bld alchl cncentratin (BAC) n all rad users invlved in rad crashes. In sme cuntries, such as Belgium and the Netherlands, drivers wh are killed n the spt are nt tested fr alchl, which cntributes t underreprting (COWI et al., 2014). Assum and Sørensen (2010) state that all drivers invlved in fatal crashes, including thse wh are killed and thse nt cnsidered t have caused the crash, shuld be tested t get cmplete data. This is the nly way t btain a cmprehensive picture f the share f alchl-related rad casualties. The legal requirements fr reprting crashes t and by the plice vary between cuntries. Furthermre, plice fficers d nt always understand the imprtance f data cllectin. Data cllectin is smetimes assciated with a lt f paperwrk and seen as just an administrative burden (ITF, 2011). In additin, recrdings f alchl tests results smetimes get lst in the prcess f registratin and changes in the registratin prcess can als affect the reliability f the data (Derriks and Mak, 2007). Finally, Gundy and Verschuur (1986) fund that plice fficers at the scene f the rad crash have a tendency t underestimate high BAC. They fund that at least a quarter t ne-third f the drivers with a BAC abve the legal limit are nt breath tested and therefre nt recrded. T address this issue f underreprting f alchl-related rad crashes in the plice registratin, many cuntries use hspital data n rad casualties t supplement the plice data (ITF, 2011; WHO, 2013). In a number f cuntries a substantial number f rad casualties are admitted t hspitals withut being knwn t the plice. The issue f drinking and driving shuld nt be limited t rad fatalities and plice data nly. Hspital data can cntribute t a mre cmprehensive picture f alchl-related serius injuries. Thus, fr a better understanding f bth the injury severity and the ttal number f alchlrelated rad casualties, the use f hspital data in additin t plice data is highly imprtant and cntributes t better internatinal cmparisns (ITF, 2011). Hwever, hspitals in many cuntries d nt perfrm a standard alchl test n casualty admissins and ften they are nly tested when the use f alchl leads t cmplicatins during surgery. The issue f underreprting nt nly characterises plice registratin, but als hspital data. When linking recrds in bth databases the prblems related t underreprting can be by-passed by using the capture-recapture apprach t estimate the ttal number f subjects (ITF, 2011).

12 1. INTRODUCTION 11 Attempts have been made t make a mre reliable and realistic estimate n the issue f alchl-related rad casualties. A study cmmissined by the Eurpean Cmmissin (COWI et al., 2014) uses additinal surrgate surces t cme t better estimates f real numbers f drinking and driving related fatalities and injuries. Surces such as epidemilgical studies, expert estimates, data n alchl use in the general ppulatin, the results frm plice enfrcement activities and self-reprted use f alchl in traffic were used in this attempt t cme t a Eurpean estimate. This study cncludes that an estimated 20-28% (25% average) f all rad fatalities in Eurpe was related t alchl use. Hwever, the fficial statistics fr the same grup f cuntries, suggest that n average 12.9% f all rad fatalities were due t alchl cnsumptin. Keeping in mind that fficial figures tend t underestimate the share f rad fatalities related t alchl (Assum and Sørensen, 2010) the authrs reprt that the actual share is clser t the higher end f the estimate (28%) than t the lwer end (20%). Defining an alchl-related crash S far we learned that the prcedures and methds that seem t be nrmal practice in many cuntries tday, lead t underreprting. Anther prblem cncerns the definitins and registratin methds and their differences between cuntries. These differences cmplicate meaningful internatinal cmparability f fficial data n alchl-related rad casualties. The Wrld Health Organizatin states in their Glbal Status Reprt n Rad Safety (2015) that much needs t be imprved t further harmnise the data cllectin n rad fatalities and serius injuries. The Eurpean SafetyNet prject recmmended a definitin fr an alchl fatality: Any death ccurring within 30 days as a result f a fatal rad crash in which any active participant was fund with a bld alchl cncentratin level abve the legal limit. IRTAD has adpted this definitin. Hwever, this definitin des nt guarantee that pedestrians and cyclists are cnsidered as any active participant, because prbably in mst cuntries a legal limit n bld alchl cncentratin fr these grups des nt exist. This wuld imply that pedestrians as well as cyclists are nt recrded as an alchl-related casualty as lng as a legal limit fr these grups is missing. This cntributes t the issue f underreprting mentined earlier. If such a definitin is accepted internatinally, crrectin factrs need t be develped t make pssible meaningful cmparisns that cnsider differences in legal limits. As yet, nt many cuntries use this definitin, and fr the cuntries wh d, data n alchl-related rad casualties can vary substantially due t ther factrs. Fr example, Gjerde et al. (2014) cmpared the use f alchl amng drivers in Brazil and in Nrway and fund that drink driving is mre cmmn in Brazil, althugh bth cuntries use the same legal alchl limit. This study cncludes that differences in the histry f legislatin, enfrcement and penalties may als be f influence n the dissimilarities in the prevalence f alchl-related rad casualties. In the present study we aim t set ut the current size f the drink driving prblem based n the fficial statistics n bth rad fatalities and serius rad injuries. Furthermre, we want t get insight int the definitins and the recrding methds used by cuntries t arrive at their fficial statistics. By reviewing registratin methds we aim t get a gd understanding f the quality f fficial statistics n alchl-related fatalities and serius rad injuries. This study als aims t get insight int the methds and prcedures that cuntries use t imprve the quality f their alchl-related casualty statistics. Our findings will be used as a basis fr recmmendatins n hw t make high quality estimates f drink driving in fficial statistics. Cntent f the reprt Chapter 2 describes the research methd used in this study. Chapter 3 presents the results f the survey regarding the respndents, the fficial statistics, the registratin methds f alchl-related casualties and the quality f the data. Chapter 4 prvides an verview f methds that can be used t

13 12 1. INTRODUCTION adjust the fficial statistics f alchl-related rad casualties. Chapter 5 presents ur cnclusins and recmmendatins.

14 1. INTRODUCTION 13 References Assum, T. and Sørensen, M. (2010), Safety Perfrmance Indicatr fr alchl in rad accidents - Internatinal cmparisn, validity and data quality. Accident Analysis & Preventin, Vl. 42/2, pp Blmberg, R.D., R.C. Peck, H. Mskwitz, M. Burns and D. Firentin (2005), Crash risk f alchl invlved driving: A case-cntrl study. Stamfrd, Dunlap and Assciates, Inc. COWI, SWOV and ADV cnsultancy and research (2014), Study n the preventin f drink driving by the use f alchl interlck devices. Rtterdam. Ecrys. Derriks, H. and P. Mak (2007), Underreprting f Rad Traffic Casualties. IRTAD special reprt, Organisatin fr Ecnmic Cperatin and Develpment, Internatinal Transprt Frum, Paris ETSC (2010), Rad safety target in sight: making up fr lst time; 4th rad safety PIN reprt. ETSC, Brussels. Hels, T., I.M Bernhft, A. Lyckegaard S. Huwing M. Hagenzieker, S.-A Legrand, C. Isalberti, T. Van der Linden, and A. Verstraete (2011), Risk f injury by driving with alchl and ther drugs DRUID Driving under the Influence f Drugs, Alchl and Medicines, D ITF (2012), Rad Safety Annual Reprt 2011, OECD Publishing, Paris. DOI: ITF (2011). Reprting n Serius Rad Traffic Casualties. Cmbining and using different data surces t imprve understanding f nn-fatal rad traffic crashes. Organisatin fr Ecnmic Cperatin and Develpment. Paris. Internatinal Transprt Frum. Accessible at: Keall, M., W. Frith, and T. Pattersn (2004), The influence f alchl, age and number f passengers n the night-time rate f driver fatal injury in New Zealand, Accident Analysis & Preventin, Vl. 36, pp Gjerde H., T.R. Susa, R. De Bni, A.S. Christphersen, R.P. Limberger, I. Zancanar, E.L. Oiestad, P.T. Nrmann, J. Mørland, F. Pechansky (2014), A cmparisn f randm drug use by randm mtr vehicle drivers in Brazil and Nrway. The Internatinal Jurnal n Drug Plicy, Vl. 25/3, pp Gundy, C. M., and W.L.G. Verschuur (1986), Plice enfrcement f drinking and driving laws: A field study f plice decisins fr requiring a radside breath test. In: Prceedings 10 th Internatinal Cnference n Alchl, Drugs and Traffic Safety. Amsterdam, 9-12 September Exerpta Medica, Internatinal Cngress series 721, Elsevier Science Publishers B.V. (Bimedical Divisin) pp SafetyNet (2008), Building the Eurpean Rad Safety Observatry. Deliverable Final reprt n task 1.5, SafetyNet, umber_f_rad_accident_casualties_final%20reprt_3.pdf.

15 14 1. INTRODUCTION Wrld Health Organizatin (2013), Glbal Status Reprt n Rad Safety Wrld Health Organizatin, Geneva. Wrld Health Organizatin (2015), Glbal Status Reprt n Rad Safety Wrld Health Organizatin, Geneva.

16 2. RESEARCH METHOD 15 Chapter 2. Research methd T cllect the data required fr this study, a questinnaire was circulated t several cuntries, thrugh their members in the IRTAD Grup, the Iber American Rad Safety Observatry (OISEVI) and the Internatinal Center fr Alchl Plicies (ICAP). This chapter describes the research methd.

17 16 2. RESEARCH METHOD Descriptin f the wrk T cllect all relevant data fr the study we develped and distributed an nline questinnaire using the nline questinnaire applicatin LimeSurvey (v2.05). Respndents culd cmplete the questinnaire by fllwing a web link. This pen surce prgram makes it pssible t dwnlad the cmplete data file in Excel frmat. We therefre used Excel t analyse the data; this is discussed in Chapter 3. Data cllectin and participants The IRTAD database includes crash and casualty data, but n data abut crash causatin factrs, including drink driving. With the help f the secretariats f IRTAD and OISEVI, 50 respndents were selected, based n membership in ne f the tw rganisatins. Bth rganisatins include institutins which have a direct interest in rad safety. Fr each cuntry, we cntacted ne rad safety expert with an invitatin letter which cntained a web link t the nline questinnaire. The initial questinnaire was pretested with five rad safety experts wh did nt respnd t the final questinnaire. Based n their remarks sme questins were added and sme were refrmulated. The first grup f respndents are IRTAD members whse respnses were cllected between March and June After apprximately six weeks a reminder was sent t nn-respnders. The secnd and final reminder was sent ne mnth after the first ne. The secnd grup f respndents are members f OISEVI and are frm Spanish speaking cuntries. Therefre the questinnaire was translated in Spanish with the help f the OISEVI secretariat. This secnd grup f respndents was apprached in mid-april Their respnses were cllected between mid-april and mid-june and a reminder was sent by the OISEVI secretariat after ne mnth. A third grup f respndents cnsisted f representatives frm six cuntries participating in a prject f the Internatinal Center fr Alchl Plicies (ICAP 1 ). Tw f these cuntries, Clmbia and Mexic, are als member f the OISEVI grup. A ttal f 54 cuntries therefre received an invitatin t participate. Questinnaire Our bjective was t cllect nt nly accurate natinal data n alchl-related rad casualties, but als t cllect backgrund infrmatin n natinal methds and prcedures used t btain natinal statistics. The questinnaire (Annex B) distinguishes fur main tpics: General backgrund infrmatin n the respndent (see Sectin Descriptin f respndents). Legislatin n maximum authrised bld alchl cncentratin (BAC), definitins f alchl-related rad casualties, and fficial casualty data (see Sectin Legislatins, definitins and fficial statistics). Definitins used as a basis fr the fficial figures n alchl-related rad fatalities and serius rad injuries. Definitins f rad fatalities and serius rad injuries attributable t drink driving. Unit f measurement fr the legal limit f bld alchl cncentratin. Existence f differentiated legal limits fr the general driving ppulatin and ther driver grups such as nvice r prfessinal drivers.

18 2. RESEARCH METHOD 17 Changes in natinal legislatin regarding drinking and driving in the perid 2000 t Official statistics n the number f rad fatalities and serius rad injuries related t alchl in 2000, 2005 and 2010 thus cvering a time perid f ten years. It was a predetermined chice nt t ask fr fficial numbers n mre recent years since ften these numbers are nt yet (cmpletely) available. Methd f recrding alchl-related crashes by plice and medical institutins (see Sectin Data cllectin methds and surces). Prcedures used t prduce plice reprts n rad crashes. Cnditins and prtcl fr carrying ut alchl tests n the scene f crashes. Cnditins and prtcl fr carrying ut alchl test at hspitals. Availability f standard tests. Prcess fr registering results f alchl tests. Linkage prcedure t link hspital and plice data. Quality f the data and the respndent s expert estimate n alchl-related rad casualties (see Chapter 3). Existence f prcedures t link and cmbine plice data and hspital data n serius rad injuries t crrect fr underreprting. Cnditins fr pst-mrtem testing. Experts best estimate f the prprtin f rad traffic deaths and serius rad injuries attributable t drink driving. Experts cmments n differences between the reprted fficial data and their persnal best estimate. Drug related fatalities and injuries with a distinctin between illicit and prescribed drugs. The questinnaire mainly cnsists f clsed questins, mstly with mre than ne pssible answer. In August 2014, six cuntries (Hungary, Iceland, Japan, Lithuania, Spain and Sweden) were again apprached via fr sme additinal fllw-up questins regarding their respnses. In December 2014, sme mre cuntries (Canada, France, Nrway, the United Kingdm, the United States f America and Sweden) were apprached via t inquire abut additinal methds used in these cuntries t imprve data n rad casualties related t alchl. In Autumn 2015, in respnse t the peer reviews, and in rder t cme t a mre cmplete picture f the fficial numbers f alchl-related fatalities and serius rad injuries, additinal infrmatin was asked fr and received frm Australia, Greece, Ireland, Italy, Jamaica, Sweden and the United States. Nte 1 Since the name f ICAP is changed int Internatinal Alliance fr Respnsible Drinking (IARD).

19

20 3. RESULTS OF THE SURVEY 19 Chapter 3. Results f the survey This chapter presents the main results f the survey regarding the respndents, the definitins, fficial statistics, the registratin methds f alchl-related casualties and the quality f the data.

21 20 3. RESULTS OF THE SURVEY Descriptin f respndents First, this sectin presents the details f the respndents in the cuntries included in this study. Table 3.1. Respnse f cuntries by rganisatin membership in percentage Respnse N respnse Ttal IRTAD (34) % 0 0.0% 63.0% OISEVI (16) % % 29.6% ICAP (4) 3 5.6% 1 1.9% 7.4% Ttal % % 100% Nte: Clmbia and Mexic are members f bth OISEVI and ICAP. In this table they are included under OISEVI. In ttal, 54 cuntries were apprached t participate in this survey: 34 cuntries are members f IRTAD, 16 cuntries are members f OISEVI and 4 cuntries participate in a drink driving prgramme rganised by ICAP. A ttal f 45 cuntries filled ut the questinnaire. This results in a respnse rate f 83 %, with a 100% respnse rate frm IRTAD cuntries. Table 3.2. Organisatin type f the respndents Organisatin Frequency Rad Safety / Transprt Research Institute (16) 35.6% Ministry (Transprt/Infrastructure/Interir) (8) 17.8% Natinal Rad Safety Authrity (10) 22.2% Natinal Statistics Bureau (2) 4.4% Natinal Plice (2) 4.4% Other (University / Cuncil / ICAP) (7) 15.6% Ttal (45) 100.0% The cuntry representatives are all rad traffic (safety) experts. Almst 38% f the respndents wrk fr natinal research institutes, a quarter fr the natinal rad safety authrity and apprximately 20% fr the Ministry f Transprt. The majrity f respndents are researchers r statisticians (44%) and managers r directrs (36%). Legislatins, definitins and fficial statistics Unit f measurement f bld alchl cncentratin Cuntries use varius units f measurement fr the bld alchl cncentratin in their legislatin n drink driving.

22 3. RESULTS OF THE SURVEY 21 Table 3.3. Unit f measurement Unit g/dl (6) Australia, Jamaica, Krea, Malaysia, Nigeria and the United States g/l (23) Argentina, Austria, Cambdia, Chile, Clmbia, Csta Rica, Ecuadr, France, Greece, Hungary, Ireland, Israel, Italy, Japan, Lithuania, Luxemburg, Nicaragua, Peru, Pland, Prtugal, Serbia, Spain and Sweden g/kg (3) Germany, Slvenia and Switzerland Other (13) Belgium (mg/ual), Blivia (mg/l), Canada (mg), Czech Republic ( ), Denmark ( ), Finland (mg/l), Great Britain (mg/100 ml), Guatemala (mg/ml), Iceland ( ), Netherlands (mg/ml), New Zealand (mg/100 ml), Nrway (mg/ual) and Russia (mg/ual) Frequency 13.3% 51.1% 6.7% 28.9% Ttal (45) 100.0% Mre than half f the cuntries use gramme per litre as unit f measurement (51%) cmpared t g/dl (13%) and g/kg (6.7%). Thirteen cuntries use ther units f measurement such as mg/ml, mg/l r mg/ual (exhaled alvelar air). This variety shuld nt have a negative effect n the cmparability f fficial data n alchl-related rad casualties when definitins n rad casualties are equal, as cnversin factrs can be applied t make cmparisn pssible. Legal limits f bld alchl cncentratin With regard t the legal limits f bld alchl cncentratin (BAC), 43 cuntries (96%) have a legal alchl limit. Blivia and Guatemala are the nly cuntries in this study withut such a limit. When cnverted in g/l, the highest BAC limit is 0.8 g/l and is fund in Canada, Great Britain, Jamaica, Malaysia, New Zealand and the United States. It shuld be nted that prvinces in Canada and states in the United States can set their wn limits and that these legal limits may be lwer than 0.8 g/l. 22 cuntries (49%) have a differentiated legal limit fr yung r nvice drivers, 23 cuntries (53%) have a differentiated legal limit fr prfessinal drivers and 19 cuntries (42%) have differentiated legal limits fr bth yung r nvice drivers and prfessinal drivers. The Czech Republic and Hungary have a legal limit f 0.0 g/l fr all rad users. See Table 3.4 fr the legal BAC limits per driver grup per cuntry.

23 22 3. RESULTS OF THE SURVEY Cuntry General ppulatin Table 3.4. Legal limits n bld alchl cncentratin levels definitin f an alchl-related rad fatality Yung nvice drivers Prfessinal drivers Argentina 0.5 g/l 0 g/l SafetyNet definitin Australia 0.5 g/l 0.2 g/l 0 g/l SafetyNet definitin Austria 0.5 g/l 0.1 g/l 0.1 g/l SafetyNet definitin Definitin f alchl-related rad fatality SafetyNet definitin: Any death ccurring within 30 days as a result f a fatal crash in which any active participant was fund with a BAC abve the legal limit Belgium 0.5 g/l 0.2 g/l Driver under the influence f alchl and drivers wh refuse t be tested. Drivers killed n the spt might nt be tested Blivia Deaths ccurring at the scene f rad accident where with a BAC abve the legal limit Cambdia 0.5 g/l SafetyNet definitin Canada 0.8 g/l 0 g/l 0.8 g/l SafetyNet definitin Chile 0.3 g/l SafetyNet definitin Clmbia 0.2 g/l N definitin Csta Rica 0.5 g/l 0.2 g/l 0.2 g/l Any death (driver, pedestrian r cyclist) ccurred in a rad accident with a BAC abve the legal limit Czech Republic 0 g/l SafetyNet definitin Denmark 0.5 g/l 0.5 g/l 0.5 g/l Any death ccurring within 30 days as a result f a fatal rad crash in which any mtr vehicle driver was fund with a bld alchl level abve the legal limit, r where the reprting fficer suspected alchl was a cntributing factr Ecuadr 0.3 g/l 0.1 g/l SafetyNet definitin Finland 0.5 g/l SafetyNet definitin France 0.5 g/l SafetyNet definitin Germany 0.5 g/l 0 g/l SafetyNet definitin Great Britain 0.8 g/l SafetyNet definitin Greece 0.5 g/l 0.2 g/l 0.2 g/l SafetyNet definitin Guatemala N definitin

24 23 3. RESULTS OF THE SURVEY Cuntry General ppulatin Yung nvice drivers Prfessinal drivers Definitin f alchl-related rad fatality SafetyNet definitin: Any death ccurring within 30 days as a result f a fatal crash in which any active participant was fund with a BAC abve the legal limit Hungary 0.0 g/l Any death ccurring within 30 days as a result f a fatal rad crash in which at least ne f the participants at fault was tested with a BAC abve the legal limit (0.0 g/l) Iceland 0.5 g/l SafetyNet definitin Ireland 0.5 g/l 0.2 g/l 0.2 g/l SafetyNet definitin Israel 0.5 g/l 0.1 g/l 0.1 g/l SafetyNet definitin Italy 0.5 g/l 0 g/l 0 g/l N definitin. Has nt published fficial statistics n alchl-related rad fatalities since Jamaica 0.8 g/l 0.8 g/l 0.8 g/l SafetyNet definitin Japan 0.3g/l Only the number f fatal crashes due t drink driving regardless f legal limit is reprted Lithuania 0.4 g/l 0.2 g/l 0.2 g/l SafetyNet definitin Luxemburg 0.5 g/l 0.2 g/l 0.2 g/l SafetyNet definitin Malaysia 0.8 g/l SafetyNet definitin Netherlands 0.5 g/l 0.2 g/l SafetyNet definitin New Zealand 0.8 g/l 0 g/l Any death ccurring within 30 days as a result f a fatal crash in which any active participant was fund with a BAC abve the legal limit r suspected t be under the influence f alchl. Nicaragua 0.5 g/l N definitin Nigeria 0.5 g/l 0.5 g/l Any death ccurring as a result f a fatal crash which was fund with a BAC abve the legal limit Nrway 0.2 g/l SafetyNet definitin Peru 0.5 g/l Any death ccurring as a result f a fatal crash in which the driver f a mtr vehicle was fund with a BAC abve the legal limit Pland 0.2 g/l Any death ccurring within 30 days as a result f a fatal crash in which any driver/cyclist was fund with a BAC abve the legal limit Prtugal 0.5 g/l 0.2 g/l 0,2 g/l SafetyNet definitin Russia 0.35 g/l 0.35 g/l 0.35 g/l Any death ccurring within 30 days as a result f a fatal rad crash in which any active participating driver r pedestrian was fund with a bld alchl level abve the legal limit Serbia 0.3 g/l 0.0 g/l 0.0g/l SafetyNet definitin Krea 0.5 g/l SafetyNet definitin Slvenia 0.5 g/l 0 g/l 0 g/l Any death ccurring within 30 days as a result f a fatal rad crash in which at least ne f the participants at fault was tested with a BAC abve the legal limit.

25 24 3. RESULTS OF THE SURVEY Cuntry General ppulatin Yung nvice drivers Prfessinal drivers Spain 0.5 g/l 0.3 g/l 0.3 g/l Sweden 0.2 g/l SafetyNet definitin Switzerland 0.5 g/l 0.1 g/l 0 g/l SafetyNet definitin United States 0.8 g/l g/l 0.4 g/l SafetyNet definitin * Expanded the SafetyNet definitin with inclusin f alchl-suspected rad users in crashes Definitin f alchl-related rad fatality SafetyNet definitin: Any death ccurring within 30 days as a result f a fatal crash in which any active participant was fund with a BAC abve the legal limit There is n fficial definitin used regarding alchl-related rad fatalities. In practice, any death ccurring within 30 days as a result f a fatal rad crash in which any active participating driver r pedestrian was fund with a bld alchl level abve 0.2 g/l (legal limit = 0.5 g/l)

26 3. RESULTS OF THE SURVEY 25 It wuld be desirable t have harmnised definitins f alchl-related rad fatalities and serius rad injuries t enable meaningful, reliable, internatinal cmparisns. Hwever, cuntries have different BAC limits and define differently an alchl-related crash, a yung r nvice driver, and prfessinal drivers. Mst cuntries define nvice drivers as thse wh have had their licenses fr less than tw, three r five years. Yung drivers are defined as yunger than a certain age (17, 18, 20, 21 r 25 years ld). Prfessinal drivers are mainly truck, van, bus and taxi drivers; fur cuntries (Israel, Luxemburg, Prtugal and Spain) als include drivers f emergency vehicles in this grup. Natinal legislatins fr yung nvice drivers are bviusly nt harmnised, which cmplicates meaningful internatinal cmparisns. Therefre an algrithm shuld be develped t make these definitins between cuntries cmparable. All respndents were als asked fr recent changes in legislatin in their cuntry. Since 2000, twenty-tw f the cuntries that publish fficial data n alchl-related rad fatalities have carried ut change in legislatin regarding BAC limits. We fund a tendency twards lwer legal limits n alchl fr the general driving ppulatin (31%) as well as fr the yung nvice drivers grup and the prfessinal drivers (17%). Canada, Germany, Italy, New Zealand, Serbia and Slvenia even have a zer-tlerance legislatin (0.0 g/l) fr yung nvice drivers (See Annex A Table A1 fr detailed infrmatin n changes in natinal legislatin). Hwever, despite the tendency twards lwer legal limits n drink driving, the share f alchl-related rad fatalities has nt shwn a decrease (See the Official statistics n alchl-related fatalities paragraph). Definitins f rad casualties As shwn in Table 3.4, several cuntries define an alchl-related fatality as any death ccurring within 30 days as a result f a fatal rad crash in which any active participant was fund with a bld alchl level abve the legal limit. This definitin was als adpted and prpsed by the SafetyNet prject. The chice f 30 days is based n the internatinal definitin f a rad crash fatality (UNECE 2009; ITF 2012; WHO, 2013). Table 3.5 shws the number f cuntries using the SafetyNet definitin. Table 3.5. Definitin f an alchl-related rad fatality Frequency SafetyNet definitin (28) 62.2% Argentina, Australia, Austria, Cambdia, Canada, Chile, Czech Republic, Ecuadr, Finland, France, Germany, Great Britain, Greece, Iceland, Ireland, Israel, Jamaica, Lithuania, Luxemburg, Malaysia, Netherlands, Nrway, Prtugal, Serbia, Krea, Sweden, Switzerland and the United States Other definitin (12) 26.7% Belgium, Blivia, Csta Rica, Denmark, Hungary, Japan, New-Zealand, Nigeria, Peru, Pland, Russia and Slvenia N definitin at all (5) 11.1% Clmbia, Guatemala, Italy, Nicaragua and Spain Ttal (45) 100% 28 f the participating cuntries (62.2%) apply this definitin and 5 cuntries d nt have any definitin n alchl-related rad fatalities. Denmark and New Zealand expand the cmmnly accepted definitin with inclusin f rad fatalities that are suspected t be alchl-related. The United States is

27 26 3. RESULTS OF THE SURVEY the nly cuntry that des nt use any active participant abve the legal limit in their definitin, but nly includes drivers and mtrcyclist. This cuntry des nt have legal limits n alchl fr pedestrians and cyclists and the SafetyNet definitin wuld therefre nt apply fr these grups (see als sectin 1 Defining an alchl-related crash ). Ten cuntries use ther definitins which in mst cases d nt include a time perid. Sme definitins cncern crashes but nt fatalities, r are unclear regarding the categry f rad users included. There is n generally accepted definitin f alchl-related serius rad injuries. In ur study we distinguish between a cmplete and an incmplete definitin presented in Table 3.6. Table 3.6. Definitin used fr alchl-related serius rad injuries Frequency Cmplete definitin 24.4% (11) Incmplete definitin 11.1% (5) N definitin 64.5% (29) Ttal 100% (45) Nte: A cmplete definitin includes bth a definitin n the severity f the injury and a definitin n the alchl-related crash. An incmplete definitin lacks ne f these definitins. Only eleven cuntries have a cmplete definitin which includes bth a definitin n serius rad injuries as well as whether r nt the crash is alchl-related. Hwever, these definitins still vary cnsiderably. Five cuntries d nt specify the severity f an injury (severe r slight) and the majrity (29 cuntries) d nt have any definitin at all. Official statistics n alchl-related casualties Alchl-related fatalities Tables 3.7 t 3.11 present the develpment in time f the fficial statistics n alchl-related rad fatalities fr cuntries with the same current legal limit n bld alchl cncentratin: Table 3.7 fr cuntries with a legal limit f 0.0 g/l. Table 3.8 fr cuntries with a legal limit f 0.2 g/l. Table 3.9 fr cuntries with a legal limit between 0.3 and 0.4g/l. Table 3.10 fr cuntries with a legal limit f 0.5g /l. Table 3.11 fr cuntries with a legal limit f 0.8 g/l. Data were cllected fr 2000, 2005 and 2010 thus cvering a time perid f ten years. It was a predetermined chice nt t ask fr data n mre recent years since ften these numbers are nt yet (cmpletely) available. Table 3.7. Share f alchl-related rad fatalities in cuntries with a present legal limit f 0.0 g/l frm fficial statistics Cuntry Remark Czech Republic 8.5% 5.5% 13.5% - Hungary 10.7% 12.8% 8.2% -

28 3. RESULTS OF THE SURVEY 27 Table 3.8. Share f alchl-related rad fatalities in cuntries with a present legal limit f 0.2 g/l frm fficial statistics Cuntry Remark Nrway 13.0% (0.5 g/l) 24.0% 21.0% In 2001 the legal limit reduced frm 0.5 g/l t 0.2 g/l Pland 11.9% 9.7% 7.9% - Sweden 18.5%* - * The numbers in Sweden are based n a study n alchl-related rad fatalities abve the legal limit between 2005 and 2013 (VTI, 2015). Table 3.9. Share f alchl-related rad fatalities in cuntries with a present legal limit between 0.3 g/l and 0.4 g/l frm fficial statistics Cuntry Remark Chile (0.3 g/l) 17.0% (0.5 g/l) 23.0% (0.5 g/l) 18.0% (0.5 g/l) In 2012 the legal limit reduced frm 0.5 g/l t 0.3 g/l. Japan (0.3 g/l) 14.9% (0.5 g/l) 10.8% 6.2% In 2002 the legal limit reduced frm 0.5 g/l t 0.3 g/l. Serbia (0.3 g/l) 4.3% (0.5 g/l) 4.9% (0.5 g/l) 5.5% In 2009 the legal limit reduced frm 0.5 g/l t 0.3 g/l. Russia (0.35 g/l) 10.8% 7.4% (0.3 g/l) Up t 2010 the legal limit was 0.3 g/l. It was 0.0 g/l frm 2010 t Since 2013, it is 0.35 g/l. Lithuania (0.4 g/l) 13.0% 14.0% 18.0% - Table Share f alchl-related rad fatalities in cuntries with a present legal limit f 0.5 g/l frm fficial statistics Cuntry Remark Australia 30%¹ - Austria 6.8% 7.3% 5.8% - Belgium 5.1% 3.5% 5.8% - Cambdia 15.4% - Csta Rica 17.3% (0.75 g/l) In 2012 the legal limit reduced frm 0.75 g/l t 0.5 g/l Denmark 18.0% 20.0% 22.0% - Finland 17.5% 22.2% - France 30.6% 28.8% 30.8% - Germany 13.6% 11.2% 9.4% - Greece² 39.4% 34.2% 33.7% - Iceland⁴ 15.5% 31.6% 37.5% - Ireland 36.5% (0.8 g/l)³ 28.3% (0.8 g/l) 15.5% (0.8 g/l) ⁵ In 2009 the legal limit reduced frm 0.8 g/l t 0.5 g/l Israel 1.9% 4.3% 3.7% - Italy 22.6%⁶ - Luxemburg 16% (0.8 g/l) 6% (0.8 g/l) 34.0% In 2007 the legal limit reduced frm 0.8 g/l t 0.5 g/l Netherlands 8.3% 8.3% 4.8% - Nicaragua 8.0% 7.0% 7.0% - Prtugal 2.7% 4.7% 7.0% - Krea 11.9% 14.3% 14.2% - Slvenia 36.3% 19.8% 35.5% - Spain 36% (0.8 g/l) 34.1% 30.9% Switzerland 19.3% 19.3% 19.3% In 2005 the legal limit reduced frm 0.8 g/l t 0.5 g/l - Ntes: ¹ The number f Australia is based n the Natinal Rad Safety Actin Plan (2010).

29 28 3. RESULTS OF THE SURVEY ² The numbers f Greece are based n ELSTAT and NTUA. ³ This number is based n the Ppulatin Health Directrate f the Health Service Executive (2006) and is the fficial prprtin fr the year ⁴ The shares f Iceland are based n very lw numbers f fatalities: 2000 N=5, 2005 N= N=3. ⁵This number is based n a reprt f the Research Department f the Rad Safety Department (2011) and is the fficial prprtin fr the year ⁶ This number is based n the DRUID prject n prevalence f alchl and ther psychactive substances in drivers killed and injured (Isalberti et al., 2011). Table Share f alchl-related rad fatalities in cuntries with a present legal limit f 0.8 g/l Frm fficial statistics Cuntry Remark Canada 30.2% 29.9% 33.6% - Great Britain 16.0% 17.0% 13.0% - Malaysia 1.3% 0.9% 5.0% - New Zealand 22.0% 26.0% 32.0% - United States f America 32.0% 31.0% 31.0% - Nte: Argentina, Blivia, Ecuadr, Guatemala, Nigeria and Peru did nt prvide data n alchl-related rad fatalities Lking at the develpment ver the years, 16 cuntries have an increased prprtin f alchl-related fatalities. In 2010, the share f alchl-related fatalities ranged frm apprximately 5% t 35% and in ten cuntries mre than 30% f rad fatalities were alchl-related. Japan and Ireland are the nly tw cuntries with a substantial decrease in the prprtin f alchl-related fatalities between 2000 and In general, the share f alchl-related fatalities has remained stable ver the years. The weighted average in 2000 was 21.95% and in 2010 this remained n the same level with a weighted average f 21.80% (see Figure 3.1). On the basis f an average share f 21.95% f alchl-related fatalities, and taking int accunt the ttal number f rad deaths f 1.25 millin rad fatalities wrldwide (WHO, 2015), the alchl-related rad tll amng fatally injured rad users was arund persns in 2010.

30 3. RESULTS OF THE SURVEY 29 Figure 3.1. Share f alchl-related rad fatalities in 2000, 2005 and 2010 frm fficial statistics 45.0% 40.0% 35.0% 30.0% 25.0% 20.0% 15.0% 10.0% 5.0% 0.0%

31 30 3. RESULTS OF THE SURVEY Alchl-related serius injuries 26 f the respndent cuntries have fficial data n alchl-related serius rad injuries (see Table 3.12). Table Cuntries publishing fficial statistics n alchl-related serius rad injuries Frequency Official statistics 57.8% (26) N fficial statistics 42.2% (19) Ttal 100% (45) Table 3.13 illustrates the evlutin in time f the share f alchl-related serius rad injuries.

32 3. RESULTS OF THE SURVEY 31 Table Share f alchl-related serius rad injuries 2000, 2005 and 2010 Frm fficial statistics Cuntry Australia 9.0% Austria 8.3% 9.9% 8.2% Belgium 8.6% 7.0% 11.0% Canada 19.7% 17.7% 18.0% Chile 13.0% 13.0% 15.0% Czech Republic 15.0% 8.9% 11.1% Denmark 15.0% 13.0% 13.0% France 12.6% 15.1% Germany 11.1% 10.4% 8.0% Great Britain 7.0% 7.0% 5.0% Greece 36.7% 20.6% 23.0% Hungary 13.3% 14.0% 12.2% Iceland 9.5% 3.1% 4.4% Israel 1.8% 1.9% 2.8% Japan 4.7% 2.7% 1.6% Luxemburg 25.0% 23.0% 15.0% Netherlands 9.3% 10.5% 8.7% New Zealand 21.0% 20.0% 23.0% Nicaragua 5.0% 7.0% 1.0% Pland 11.5% 10.9% 9.3% Prtugal 4.6% 12.8% 17.3% Serbia 5.3% 5.0% 9.3% Russia 10.3% 6.8% Krea 10.8% 13.5% 12.9% Slvenia 21.0% Switzerland 14.0% 12.9% 12.7% Nte: Argentina, Blivia, Cambdia, Clmbia, Csta Rica, Ecuadr, Finland, Guatemala, Ireland, Italy, Jamaica, Lithuania, Malaysia, Nigeria, Nrway, Peru, Spain, Sweden and the USA did nt prvide numbers n alchl-related serius injuries. The average share (mean) remained quite stable between 2000 (12.3%) and 2010 (11.3%). The highest shares f alchl-related serius rad-injuries are fund in New Zealand (23%) and Greece (23%) and the lwest share is reprted in Japan (1.6%). Figure 3.2 summarises these figures fr the years 2000, 2005 and 2010.

33 32 3. RESULTS OF THE SURVEY Figure 3.2. Share f alchl-related serius rad injuries in 2000, 2005 and 2010 frm fficial statistics 40.00% 35.00% 30.00% 25.00% 20.00% 15.00% 10.00% 5.00% 0.00%