La prise de conscience de l’implication des médicaments dans la genèse des accidents de la route date d’une vingtaine d’années. Les médicaments psycho-actifs peuvent altérer les capacités de conduite par leur action sur le système nerveux (par exemple, un effet sédatif le lendemain d’une prise d’hypnotique). D’autres médicaments sont susceptibles d’affecter les fonctions psychomotrices par leur action sur les fonctions physiologiques (tel que les hypoglycémies liées à un traitement antidiabétique). L’étude CESIR-A a été mise en place pour contribuer à la connaissance du lien épidémiologique entre médicaments et accidents de la route. L’étude utilise trois bases de données françaises : le Système National d’Information Inter-Régimes de l’Assurance Maladie (SNIIR-AM), les Procès Verbaux d’accidents (PV) et les Bulletins d’Analyse des Accidents Corporels de la circulation (BAAC). L’appariement de ces données a conduit à l’inclusion de 72,685 conducteurs impliqués dans un accident corporel sur la période juillet 2005-mai 2008. L’analyse a été réalisée grâce à deux méthodes: une analyse cas-témoin comparant les responsables aux non-responsables des accidents et une analyse dite en case-crossover. Les périodes d’exposition aux médicaments ont été estimées à partir des dates de délivrances de médicaments prescrits, puis remboursés par l’assurance maladie. L’étude des médicaments regroupés selon les quatre niveaux de risque sur la conduite définis par l’Agence Française de Sécurité Sanitaire des Produits de Santé (AFSSAPS) [du niveau 0 (pas de risque) au niveau 3 (risque élevé)], a montré que les utilisateurs de médicaments prescrits de niveau 2 et de niveau 3 ont un risque significativement plus élevé d’être responsables de leur accident (OR=1,31 [1,24-1,40] et OR=1,25 [1,12-1,40], respectivement). La fraction de risque attribuable à l’utilisation de ces médicaments était de 3,3% [2,7%-3,9%]. Le risque d’être responsable d’un accident était augmenté chez les utilisateurs de zolpidem (OR=1,28 [1,07-1,53]) mais pas chez les utilisateurs de zopiclone ou de benzodiazépines hypnotiques. Plus particulièrement, ce risque était augmenté chez les 139 conducteurs ayant eu plus d’un comprimé de zolpidem délivré par jour au cours des cinq mois précédant l’accident (OR=2,38 [1,61-3,52]). L’analyse case-crossover a mis en évidence un sur-risque d’accident de la route chez les utilisateurs de benzodiazépines hypnotiques seulement (OR=1,42 [1,09-1,85]). Les conducteurs exposés aux hypnotiques partagent les mêmes caractéristiques au regard du type d’accident, qui survenaient plus fréquemment sur autoroute. Dans notre base de données, 196 conducteurs ont été exposés à la buprénorphine et/ou à la méthadone, le jour de leur accident. Cette population spécifique était jeune, essentiellement masculine, avec d’importantes co-consommations, notamment d’alcool de médicaments de niveau 3. Les conducteurs exposés à la buprénorphine et/ou à la méthadone présentaient un risque accru d’être responsables de leur accident (OR= 2,19 [1,51-3,16]). Notre étude fournit des informations importantes sur la contribution des médicaments au risque d’accident de la route. D’après nos résultats, la classification de l’AFSSAPS semble appropriée concernant les médicaments de niveaux 2 et 3. Les sur-risques d’être responsable d’un accident chez les exposés au zolpidem ou aux traitements de substitution pourraient être liés, au moins en partie, au comportement à risque de ces conducteurs. L’amélioration du comportement des conducteurs représente un des défis pour la sécurité routière. L’objectif de la classification française et de la signalétique apposée sur les boîtes de médicaments est donc de fournir aux patients une information appropriée sur les effets des médicaments sur leur capacité de conduite. / In recent decades, attention has been increasingly focused on the impact of disabilities and medicinal drug use on road safety. Psychoactive medicines may impair driving abilities due to their action on the central nervous system (e.g. sedation in the morning following administration of a hypnotic), while other medicines may affect psychomotor functions by their action on physiological functions (e.g hypoglycaemic seizures related to diabetic treatment). The CESIR-A project was set up to improve the epidemiological knowledge on medicines and the risk of road traffic crashes. The study matched three French nationwide databases: the national healthcare insurance database, police reports, and the police national database of injurious crashes, leading to the inclusion of 72,685 drivers involved in an injurious road traffic crash from July 2005 to May 2008. Two methods were performed for data analysis: a case-control analysis in which cases where responsible drivers and controls non-responsible ones and a case-crossover analysis. Medicine exposures were estimated from prescription drug dispensations in the healthcare reimbursement database. The study of medicines grouped according to the four levels of driving impairment risk of the French classification system [from 0 (no risk) to 3 (high risk)], showed that users of level 2 and level 3 prescribed medicines were at higher risk of being responsible for the crash (OR=1.31 [1.24-1.40] and OR=1.25 [1.12-1.40], respectively). The fraction of road traffic crashes attributable to levels 2 and 3 medicines was 3.3% [2.7%-3.9%]. Zolpidem use was associated with an increased risk of being responsible for a road traffic crash (OR=1.28 [1.07-1.53]) whereas use of zopiclone and benzodiazepine hypnotics use was not. Responsibility risk was only increased in the 139 drivers with dispensing of more than one pill of zolpidem a day during the five months before the crash (OR=2.38 [1.61-3.52]). Case-crossover analysis showed an increased risk of crash for benzodiazepine hypnotic users only (OR=1.42 [1.09-1.85]). Hypnotic users shared similar crash characteristics, with crashes more likely to occur on highways. In our database, 196 drivers were exposed to buprenorphine and/or methadone on the day of crash. This specific population was young, essentially males, with important co-consumption of other substances, in particular alcohol and level 3 medicines. Injured drivers exposed to buprenorphine and/or methadone on the day of crash, had an increased risk of being responsible (OR=2.19 [1.51-3.16]). The case cross-over analysis did not demonstrate any association (OR=1.26 [0.93 - 1.70]). Our study provides evidence of the contribution of medicines to the risk of road traffic crashes. According to our results, the French risk classification seems relevant regarding medicines classified as levels 2 and 3 of risk for road traffic crashes. The observed increased risks of being responsible for a crash for zolpidem and substitution maintenance treatment users may be linked to risky behaviors. Improving driver behaviour is one of the challenges for road safety. Providing patients with proper information on the potential effect of medicines on their driving abilities is the main objective of drug and risk classifications such as the French one.

Abstract in English, Tsonga and Zulu / Mitlumbo ya mifambafambo ya le magondzweni i xin’wana xa miringeto (risks) yo biha ku tlula hinkwayo ya swifambo swa le gondzweni emisaveni hinkwayo, leswi yimelaka xiphiqo lexikulu xa swohanyaswin’we-ikhonomi ngopfu-ngopfu eka matiko lama ya ha hluvukaka tanihi Afrika-Dzonga. Ku va ku nyikiwa masungulo yo tiya ya xiikhonomi eka swiboho swa mbekiso ku tirhana na ntlhontlho lowu, i swa nkoka swinene ku hlela ndhurho wa mitlumbo leyi. Mipimanyeto leyi yi tirha tanihi nxopaxopo wa swinghenisiwa swa mbuyelo wa ndhurheriwo ku endlela ku kuma mphakelo wa switirhisiwa wo tirha kahle eka ku nghenelela eka ku tirhana na mitlhontlho leyi vangiwaka hi mitlumbano ya le magondzweni. Tiko ra Afrika-Dzonga a ri nga ri ku pfuxeteni ka mahungu ya mipimanyeto ya midurho ya mitlumbano ya le magondzweni nkarhi na nkarhi, naswona leyi a yi endliwa a yi tirhisa maendlelo lamo soriwa ngopfu yo languta nkoka wa vanhu (human capital). Hikwalaho, mipimanyeto leyi nga kona a yi nga ta va leyi tshembekaka eka ku kunguhata na ku pimaniseka na mipimanyeto ya matiko man’wana. Hi le ka ku landzelela vundzhaku lebyi laha dyondzo leyi yi nga tumbuluka na ku kombisa matirhiselo ya rimba ra ntirho wo katsa (hybrid) ku kambela ndhurho wa mitlumbo ya swifambo swa le magondzweni eAfrika-Dzonga. Rimba leri ri tirhisa endlelo ro kongomisa eka nkoka wa vanhu na ku pfumela ku hakela (willingness-to-pay), eka dyondzo yi ri yin’we. Mipimanyeto ya midurho ya nkoka wa vanhu ya laveka ku va yi pfuna eka ku kunguhata leswaku yi tlakusa swinenenene swihumesiwa swa rixaka, loko hala tlhelo mipimanyeto yo pfumela ku hakela yona yi ri yona yi fanelaka swinene eka ku pfuneta minghenelelo yo tlakusa nhlayiseko wa vanhu hi ku hunguta ku vaviseka na ku fa. Endlelo ro pfumela ku hakela ri tirhisa maendlelo ya swo ka swi nga ri swa makete (contingent valuation) na ya maendlelo yo langa (preference). Khwexinere yo valanga leyi a yi ri na maendlelo yo ka ya nga ri ya swa makete na swilangiwa leswi a swi boxiwile, yi tirhisiwile hi magoza mambirhi ku sampula vaanguri va 273 eka ntirho wa swo tleketla. Eka mhaka ya endlelo ro kongomisa eka nkoka wa vanhu, mipimanyeto ya ndhurho eka xiviko xa 2016 xa Cost of Crashes in South Africa yi hundzuluxiwile hi ku katsa inifulexini, ku tirhisiwa mpimo wa 2017 wa 5.3% ku kuma mipimanyeto ya ndhurho ya 2017. Dyondzo leyi yi paluxile leswaku endlelo ro kongomisa eka nkoka wa vanhu ri kayiveta ku vona ndhurho wa mitlumbo ya le magondzweni. Dyondzo leyi yi hoxa xandla eka ntsengo wa vutivi hi ku tirhisa endlelo ro languta nkoka wa vanhu na endlelo ro pfumela ku hakela eka dyondzo yi ri yin’we ku kombisa ntirhiseko wa endlelo leri ra nkatso (hybrid)/nhlanganiso eka xiyimo xa Afrika-Dzonga. Ndzavisiso wa nkarhi lowu taka wu fanele ku engeta dyondzo leyi hi sampulu leyi humaka eka swifundzakulu hinkwaswo swa nkaye swa Afrika-Dzonga, leswaku mipimanyeto ya ndhurho yi yimela vanhu va tiko hinkwaro. / Road traffic crashes are one of the worst risks of road mobility worldwide, representing a huge socio-economic problem particularly in developing countries such as South Africa. In order to provide a sound economic basis for investment decisions to address this challenge, it is critical to assess the cost of these crashes. These estimates serve cost-benefit analysis inputs to facilitate a more efficient resources allocation for interventions to address the challenge posed by road crashes. South Africa has not been updating crash cost estimates on a regular basis, and those that were conducted used the much criticised human capital approach. Therefore, the available estimates could not be relied upon for planning purposes and comparison with the estimates of other countries. It is against this background that this study developed and illustrated the application of a hybrid framework for assessing the cost of road traffic crashes in South Africa. The framework uses the human capital approach and the willingness-to-pay approach in one study. Human capital approach cost estimates are needed to inform planning to maximize the national output, while the willingness-to-pay estimates are more suitable when the main concern is to inform interventions to increase social welfare by reducing injuries and fatalities. The willingness-to-pay approach uses the contingent valuation and the stated preference methods. A survey questionnaire with contingent valuation and stated preference questions was administered in two phases to a sample of 273 respondents within the transport industry. For the human capital approach, the cost estimates in the 2016 Cost of Crashes in South Africa report were adjusted for inflation using the 2017 rate of 5.3% to obtain 2017 cost estimates. This study revealed that the human capital approach underestimates the cost of road crashes. The study contributes to the body of knowledge by using the human capital approach and the willingness-to-pay approach in one study to illustrate the applicability of this hybrid/ combination within the South African context. Future research needs to replicate this study on a sample drawn from all nine provinces of South Africa, so that the cost estimates are representative of the country’s population. / Ukuphazamiseka komgwaqo kungenye yezingozi ezimbi kakhulu zokuhamba komgwaqo emhlabeni jikelele, ezimele inkinga enkulu yenhlalo nezomnotho ikakhulukazi emazweni asathuthuka njengeNingizimu Afrika. Ukuze unikeze isisekelo sezomnotho esizwakalayo ezinqumeni zokutshala izimali ukubhekana nale nselele, kubalulekile ukuhlola izindleko zalezi zingozi. Lezi zilinganiso zisebenza njengeziphakamiso zokuhlaziywa kwezindleko zokuhlomula ukuze kube lula ukunikezwa kwezinsiza ezenzelwe ukuxazulula inselele ebangelwa ukuphazamiseka komgwaqo. INingizimu Afrika ayizange ibuyekeze ukulinganisa izindleko zezingozi njalo, futhi lezo ezenziwa zisebenzise indlela enkulu yokugxeka ukusebenzisa abantu. Ngakho-ke, izilinganiso ezitholakalayo azikwazanga ukuthenjelwa kuzona ngezinjongo zokuhlela nokuqhathaniswa nezilinganiso zamanye amazwe. Lokhu kuphikisana nalesi sigaba ukuthi lolu cwaningo lusungulwe futhi luboniswe ukusetshenziswa kohlaka oluxubile lokuhlola izindleko zokuphazamiseka komgwaqo eNingizimu Afrika. Uhlaka lusebenzisa indlela yokusebenzisa abantu kanye nendlela yokuzimisela-ukukhokha ocwaningweni olulodwa. Ukulinganiselwa kwezindleko zokusebenzisa abantu kuyadingeka ukuze kwaziswe ukuhlela ukwandisa umkhiqizo kazwelonke, kanti ukulinganiselwa kokuzimisela-ukukhokhela kukulungele kakhulu ukwazisa ukungenelela ukwandisa inhlalakahle yomphakathi ngokunciphisa ukulimala nokubulawa kwabantu. Indlela yokuzimisela-ukukhokha isebenzisa ukuhlaziywa kwesilinganiso kanye nezindlela okukhethwa ngazo. Imibuzo yokuhlola ngokuhlaziywa kwesilinganiso kanye nemibuzo ekhethwe ngayo yenziwa ngezigaba ezimbili embonakalisweni yabaphendulile abangama-273 embonini yezokuthutha. Ngokwendlela yokusebenzisa abantu, izindleko ezilinganiselwa ku-2016 Izindleko Zokushayisana eNingizimu Afrika kubikwa ukuthi zalungiselwa ukwenyuka kwamandla emali, kusetshenziswa isilinganiso sango-2017 esingu-5.3% ukuthola izindleko zango-2017. Lolu cwaningo luveze ukuthi indlela yokusebenzisa abantu ithatha kancane izindleko zokuphazamiseka komgwaqo. Ucwaningo lunomthelela emzimbeni wolwazi ngokusebenzisa indlela yokusebenzisa abantu kanye nendlela yokuzimisela-ukukhokha ocwaningweni olulodwa ukukhombisa ukufaneleka kwalesi sivumelwano / inhlanganisela ngaphakathi komongo waseNingizimu Afrika. Ucwaningo lwesikhathi esizayo ludinga ukuphindaphinda lolu cwaningo embonakalisweni othathwe kuzo zonke izifundazwe eziyisishiyagalolunye zaseNingizimu Afrika, ukuze ukulinganiswa kwezindleko kummele abantu bezwe / Business Management / D. Phil. (Management Studies)