Epilepsy & Behavior 41 (2014) 179–182

Contents lists available at ScienceDirect

Epilepsy & Behavior

j ourna l homepage: www.e lsev ie r .com/ locate /yebeh

Disobedience and driving in patients with epilepsy in Greece

Panagiotis Zis a,⁎, Anna Siatouni b, Vassilios K. Kimiskidis c, Anastasia Verentzioti b, Georgios Kefalonitis a,Nikolaos Triantafyllou d, Stylianos Gatzonis b

a Department of Neurology, Evangelismos Hospital, Athens, Greeceb 1st Department of Neurosurgery, Epilepsy Surgery Unit, University of Athens, Evangelismos Hospital, Athens, Greecec Laboratory of Clinical Neurophysiology, AHEPA Hospital, Aristotle University of Thessaloniki, Thessaloniki, Greeced 1st Department of Neurology, University of Athens, Eginition Hospital, Athens, Greece

⁎ Corresponding author at: Evangelismos General Hosp45-47 Ipsilantou Street, 10676 Athens, Greece. Tel.: +32041403.

E-mail address: takiszis@gmail.com (P. Zis).

http://dx.doi.org/10.1016/j.yebeh.2014.09.0791525-5050/© 2014 Published by Elsevier Inc.

a b s t r a c t

a r t i c l e i n f o

Article history:

Received 19 August 2014Revised 20 September 2014Accepted 28 September 2014Available online xxxx

Keywords:DrivingEpilepsySeizuresSafetyRiskEmploymentMale gender

Objective: Regulations and guidelines regarding driving privileges of patients with epilepsy vary greatlyworldwide. The aim of our study was twofold: firstly, to evaluate disobedient drivers in Greece and to elucidatetheir awareness of the law, emotional responses, and seizure profile and, secondly, to identify determinants ofdisobedience regarding driving among patients with epilepsy.Methods:All consecutive patients with epilepsywho visited the epilepsy outpatient clinic of two tertiary epilepsycenters were invited to participate in the study. One hundred ninety patients met our inclusion criteria.Results: Fifty-two percent of our study population was aware of the driving restrictions. More than one out ofthree patients were disobedient (35.8%). Being a male was associated with a 6.07-fold increase in the odds ofbeing disobedient (95% CI: 2.73–13.47, p b 0.001); being employed was associated with a 4.62-fold increase inthe odds of being disobedient (95% CI: 2.20–9.68, p b 0.001); and each extra antiepileptic drug (AED) wasassociated with a decrease in the odds of disobedience by a factor of 0.41 (95% CI: 0.26–0.63, p b 0.001).Conclusion: Male gender, employment, and number of AEDs are important determinants of disobedienceregarding driving among patients with epilepsy.

© 2014 Published by Elsevier Inc.

1. Introduction

Regulations and guidelines regarding driving privileges of patientswith epilepsy vary greatly worldwide [1,2]. Since 2012, Greece includedthe driving licensure standards as defined by the European Commissionin the Greek legislation. The basic points include that (1) only patientswho had received antiepileptic treatment in the past and after itsdiscontinuation and for a period of 3 years are seizure-free can receiveor renew a driving license and (2) only patientswho still receive antiep-ileptic treatment and have been seizure-free for a period of 2 years andhave a normal electroencephalogram (EEG) can receive or renew theirdriving license [3].

Before obtaining a driving license, each candidate for a drivinglicense has to go through a medical examination by an independentcommittee, which includes an examination by a physician and anophthalmologist. At this point, it is the candidate's responsibility to re-port any medical problems and all medications he/she may receive.Therefore, a history of epilepsy is self-reported, and the neurologist ofindividuals with epilepsy, apart from informing the patient of the law,

ital, Department of Neurology,0 697 4105446; fax: +30 213

cannot report patients to the relevant authorities. Unfortunately, untilnow, there are no reporting rules, even in the case that the neurologistfinds out that one of his or her patients with epilepsy illegally drives.The only “safety net” is that the insurance companies only insure pa-tients who are legally driving, and in case of an accident, they mayhave access to themedication regime of the driver (through a newly de-veloped electronic database) and, subsequently, not cover the expensesif they find out that the patient was illegally driving.

Several studies worldwide have tried to capture the behavior of pa-tients with epilepsy regarding driving, and, interestingly, results differfrom country to country [4–9]. This may not only reflect the culturaldifferences but also highlights the necessity of investigating thesebehaviors in different countries in order to raise awareness.

In a recent study conducted in the USA, Tatum et al. used a 12-itemquestionnaire in order to identify illegal and disobedient driving prac-tices among patients with epilepsy [4]. They found that, overall, asmall number of patients with seizures were disobedient and illegallydriving, and they suggested that a targeted approach to high-riskdrivers with repeated verbal and supplemental driving informationmay help avoid unnecessary universal physician reporting for patientswith seizures [4].

The aim of our study was twofold: firstly, using the same methodol-ogy as Tatum et al. [4], to evaluate disobedient drivers in Greece and toelucidate their awareness of the law, emotional responses, and seizure

Table 1Demographic and clinical characteristics of the total study population and the subgroups.

Total(n = 190)

Disobedientpatients(n = 68)

Obedientpatients(n = 122)

p-Value

Demographic characteristicsMale sex (%) 107 (56.3) 52 (76.5) 55 (45.1) b0.001Age, in years (SD) 37.7 (12.4) 38.2 (10.5) 37.4 (13.5) 0.685Married (%) 12 (6.3) 5 (7.4) 7 (5.7) 0.661Education level (%)

Primary 18 (9.5) 4 (5.9) 14 (11.5)Secondary 98 (51.6) 30 (44.1) 68 (55.7) 0.052Higher 74 (38.9) 34 (50) 40 (32.8)

Employment statusPaid employment 86 (45.3) 45 (66.2) 41 (33.6)Employed part-time (%) 19 (10.0) 7 (10.3) 12 (9.8)Employed full-time (%) 67 (35.3) 38 (55.9) 29 (23.8) b0.001

Homemaker (%) 17 (8.9) 2 (2.9) 15 (12.3)Full-time student (%) 17 (8.9) 6 (8.8) 11 (9.0)Retired (%) 31 (16.3) 8 (11.8) 23 (18.9)Unemployed (%) 39 (20.5) 7 (10.3) 32 (26.2)

Epilepsy-related characteristicsNumber of lifetime seizures (%)

1 15 (7.9) 7 (10.3) 8 (6.6)b10 58 (30.5) 24 (35.3) 34 (27.9)10–20 29 (15.3) 12 (17.6) 17 (13.9)N20 32 (16.8) 13 (19.1) 19 (15.6) 0.181N100 43 (22.6) 10 (14.7) 33 (27.0)Daily 13 (6.8) 2 (2.9) 11 (9.0)

Time of last seizureToday 5 (2.6) 1 (1.5) 4 (3.3)This week 31 (16.3) 7 (10.3) 24 (19.7) 0.132This month 31 (16.3) 7 (10.3) 24 (19.7)2–3 months ago 18 (9.5) 8 (11.8) 10 (8.2)4–6 months ago 19 (10.0) 7 (10.3) 12 (9.8)N1 year ago 86 (45.3) 38 (55.9) 48 (39.3)

Number of AEDs (SD) 2.0 (0.9) 1.6 (0.7) 2.2 (1.0) b0.001Compliant with medication (%) 181 (95.3) 66 (97.1) 115 (94.3) 0.384

180 P. Zis et al. / Epilepsy & Behavior 41 (2014) 179–182

profile and, secondly, to identify determinants of disobedience regardingdriving among patients with epilepsy.

2. Methods

2.1. Participants

All consecutive patients with epilepsy who visited the epilepsy out-patient clinics of two tertiary hospitals, the Evangelismos GeneralHospital in Athens and theAHEPAHospital in Thessaloniki, were invitedto participate in the study.

To be enrolled, the patients had to meet the following inclusioncriteria: (1) having a confirmed diagnosis of any type of epilepsyaccording to the International League Against Epilepsy (ILAE) criteria[10,11], documented clinically and confirmed with EEG studies; (2)having age equal to or greater than 18 years; (3) having no gross cogni-tive deficits or intellectual disability; (4) being a native Greek speaker;and (5) being willing to provide written informed consent. Approvalwas gained from the local Research Ethics Committees to conduct thestudies.

2.2. Procedures

An anonymous questionnaire was administered to all eligiblecandidates who were asked to return the completed questionnaires ina sealed envelope that was placed in a nontransparent empty box bythe participant in order to ensure the anonymity of the questionnaire.As the participation in the study was voluntary, there was no interfer-ence in the study by anyone outside the team of researchers, whichmeans that no pressure was applied to take part in the study.

We followed the same methodology that Tatum et al. used in theirstudy in the USA [4]. Our questionnaire was divided into two parts. Thefirst part included a survey that elicited the following demographic char-acteristics: age, gender, level of education, place of residence,marital sta-tus, professional status, and driving skills. The second part was anadapted version of the 12-item questionnaire that was used by Tatumet al. [4]. After obtaining the relevant permission from Professor Tatum,the original English version of the questionnairewas translated indepen-dently by two authors (AS and NV) into Greek. The two independenttranslations matched absolutely. The fifth author (GK), who is fluent inEnglish and initially had no access to the original instrument, thenback-translated the translated version into English. Pilot testing of theGreek version was implemented in 10 volunteer patients to ensurethat participants would interpret each item as originally intended. Nofurther adaptations were required. In summary, the questionnaire is de-signed to provide responses that would reflect information regardingmedical, psychological, legal, and compliance issues. Six of these ques-tions addressed the total number of lifetime seizures, time of last occur-rence, emotional reaction to driving restriction, initial impulse torestricted driving, current driving lifestyle, and knowledge of the Greekregulationwith respect to driving in a six-optionmultiple choice format.Six additional questions addressed attitudes toward compliancewith thelaw, physician delivery of driving information, country of licensure andvalidity, number of antiepileptic drugs (AEDs) taken, and compliancewith treatment in a yes–no or a write-in format.

Patients whowere not driving, either having or not having a drivinglicense, were considered to be obedient, while patients who weredriving, either having or not having a driving license, were consideredto be disobedient.

2.3. Statistical analyses

A databasewas developed using the Statistical Package for the SocialSciences (version 16.0 for Mac; SPSS). Frequencies and descriptive sta-tistics were examined for each variable. Comparisons between obedientand disobedient patientsweremadeusing Student's t-tests for normally

distributed continuous data, Mann–Whitney's U test for nonnormallydistributed data, and chi-square test for categorical data.

Variables that showed either significant differences or trend fordifference (p b 0.10) were entered into a logistic regression model toidentify determinants of disobedience among patients with epilepsy.The variables were entered as independent variables, and disobediencewas entered as the dependent variable.

The Wald statistical test was used to investigate the differencebetween the maximum likelihood estimates of the logistic regressionof the parameters of interest compared with that of the null hypothesisassumptions.

A p-value of b0.05 was considered to be statistically significant.

3. Results

3.1. Study population

Between November 2013 and May 2014, 193 consecutive patientsfulfilling inclusion criteria (1) to (4) were invited to participate inthe study. Three patients did not provide written informed consent;thus, they were excluded. In total, 190 patients fulfilled all theabovementioned inclusion criteria. In total, 111 (58.4%) patients had adriving license. None of them had requested the special permissionfrom the relevant authorities in order to make their license valid afterbeingdiagnosedwith epilepsy. In total, 68 (35.8%) patientswere drivingand were considered disobedient, while 122 were not driving and wereconsidered obedient regarding driving. Interestingly, 3 disobedient pa-tients were driving despite the fact that they had never taken a drivinglicense examination.

Table 2Initial reactions of the studypopulation and the subgroupswhen theywere first told of thedriving restrictions.

Number (%) ofpatients of thetotal population(n = 190)

Number (%)of disobedientpatients(n = 68)

Number (%) ofobedient patients(n = 122)

Anger 26 (13.7) 11 (16.2) 15 (12.3)Denial or disagreement 32 (16.8) 23 (33.8) 9 (7.4)Fear 17 (8.9) 3 (4.4) 14 (11.5)Isolation 8 (4.2) 3 (4.4) 5 (4.1)Relief 20 (10.5) 1 (1.5) 19 (15.6)Sadness 55 (28.9) 19 (27.9) 36 (29.5)No answer 32 (16.8) 8 (11.8) 24 (19.7)

Table 4Patient characteristics investigated for their associationwith disobedience regarding driv-ing restrictions.

Variable ORa (95% CI) Wald p-Value

Male sex 6.069 (2.734–13.471) 19.648 b0.001Higher education 0.626 (0.297–1.321) 1.512 0.219Being employed 4.618 (2.204–9.675) 16.432 b0.001Number of AEDs (per extra agent) 0.406 (0.262–0.630) 16.223 b0.001

a Adjusted odds ratios (ORs) presented.

181P. Zis et al. / Epilepsy & Behavior 41 (2014) 179–182

Table 1 summarizes the demographic and clinical characteristicsof the total study population as well as the two subgroups: obedientpatients and disobedient patients.

3.2. Awareness of the law and emotional responses

Thirty-two (16.8%) patients reported unawareness of the law re-garding driving and epilepsy. Interestingly, of the patientswho reportedawareness of the law, only 63% knew the law correctly. This means thatonly 52% of the study population was, in fact, aware of the law.

One hundred (52.6%) patients of the study population indicated thattheir doctor had never discussed the driving laws with them. Forty-one(21.6% of the total study population) of the patients who reported notbeing informed of the driving laws by their provider were illegallydriving.

As seen in Table 2, regarding the initial emotional reaction to drivingrestriction, sadness was the most common reaction reported in thegroup as a whole (28.9%) and in the subgroup of the obedient drivers(29.5%). However, among the disobedient drivers, the most commoninitial emotional reaction was denial or disagreement (33.8%).

Table 3 summarizes the responses to the question “which statementbest represents your initial impulse?”. Interestingly, the initial impulseof 69.6% of the obedient drivers was that they would not drive as theymay hurt self or others, while, respectively, only 29.4% of the disobedi-ent drivers reported the same impulses.

3.3. Determinants of disobedience

The following independent variables were entered into the multi-variate logistic regressionmodel: sex, education level (higher educationor not), employment status (paid employment or not), and number ofAEDs. The results of the multivariate logistic regression are shown inTable 4. Adjusted odds ratios are presented. The full model significantlypredicted disobedience (χ2 = 62.18, df = 4, p b 0.001), with 27.9% to38.3% of the variance being explained by three variables: sex, employ-ment status, and number of AEDs. Being a male was associated with a

Table 3Responses of the study population and the subgroups to the question “which statement best r

Impulse

This is a stupid rule, my doctor doesn't know what he/she is talking about.I am still going to drive because I usually get a warning before a seizure and can pull overI am still going to drive because I have to get to work, drive my children, or run errands andother people all the time.I will wait a few months to see what happens and then start driving even if my doctor or tauthorities say I should not.I will not drive until my doctor or the relevant authorities say it is okay because I may hurI will not drive until my doctor or the relevant authorities say it is okay because I may hurNo answer

6.07-fold increase in the odds of being disobedient (95% CI: 2.73–13.47, p b 0.001); being employed was associated with a 4.62-fold in-crease in the odds of being disobedient (95% CI: 2.20–9.68, p b 0.001);and each extra AED was associated with a decrease in the odds ofdisobedience by a factor of 0.41 (95% CI: 0.26–0.63, p b 0.001).

4. Discussion

Our cross-sectional study aimed, firstly, to evaluate disobedientdrivers in Greece and to elucidate their awareness of the law, emotionalresponses, and seizure profile in order to help identify this at-riskpopulation and, secondly, to identify determinants of disobedience re-garding driving among patients with epilepsy. Out of the 193 patientswho were invited to participate in the study, 190 were willing toprovide informed consent (response rate: 98.4%).

Using the same methodology that Tatum et al. used, we found that35.3% of our study population was disobedient, which is higher com-pared with the 23.8% of ineligible drivers who were violating the lawin the state of Florida [4]. This disagreement, apart from a very differentdriving legislation and possible cultural differences, could be attributedto the fact that, within our study population, 52.6% of the patientsreported that their doctor had never discussed with them drivingrestriction, a percentage which is significantly higher compared withthe respective 20.8% of patients in Florida [4].

Regarding the initial reactions of the patients when they were firsttold of the driving restrictions, in both study populations, as a whole,sadness was the most frequent initial reaction. However, the first re-sponse among Greek disobedient patients was denial or disagreement,while among American patients in Florida, the first response wasanger [4]. This is a fact that also highlights possible cultural differencesbetween the two populations.

With regard to the initial impulse, both Greek andAmerican patientswith epilepsy gave similar answers. The most common impulse of thetotal cohort was that they would not drive as they may hurt someoneelse, while the most common impulse of the disobedient drivers wasthat they would still go driving because they had to get to work, drivetheir children, or run errands and that they could not rely on otherpeople all the time.

Apart fromdescribing the awareness of the law and the emotional re-sponses of patients with epilepsy, we aimed to identify determinants ofdisobedience. To our knowledge, this is the first Greek study to address

epresents your initial impulse?”

Number (%) ofpatients of thetotal population(n = 190)

Number (%) ofdisobedient patients(n = 68)

Number (%) ofobedient patients(n = 122)

3 (1.6) 1 (1.5) 2 (1.6)if I need to. 23 (12.1) 17 (25.0) 6 (4.9)I can't rely on 21 (11.1) 18 (26.5) 3 (2.5)

he relevant 11 (5.8) 6 (8.8) 5 (4.1)

t myself. 33 (17.4) 6 (8.8) 27 (22.1)t someone else. 72 (37.9) 14 (20.6) 58 (47.5)

27 (14.2) 6 (8.8) 21 (17.2)

182 P. Zis et al. / Epilepsy & Behavior 41 (2014) 179–182

this double aim. Another Greek study conducted by Polychronopouloset al. aimed to identify factors that are independently associatedwith noncompliance with driving restrictions and showed that patientswith epilepsy do not comply with driving restrictions mainly foremployment-related reasons [2]. However, Polychronopoulos et al. con-ducted their study in a case–control format, after age- and sex-matching60 disobedient patients with 60 obedient patients. We followed a differ-ent methodology as this paper describes a cross-sectional study, whichpermitted us not only to calculate the percentage of disobedient patientsbut also to unmask other possible factors associated with disobediencesuch as gender. This is another novelty of our study as, apart from theepidemiological importance for Greece, our results may provide a betterunderstanding of at-risk population to be disobedient. Specifically, weshowed that male sex, employment, and number of AEDs are factorsassociated with disobedience.

Gender differences in behaviors among people with epilepsy havebeen described in many studies. Among patients with epilepsy, themale gender has also independently been associated with driving inBrazil [12], West China [9], the USA [13], and Korea [14].

Employment has also been an important factor for patients withepilepsy. Patients with epilepsy may have difficulty in finding andmaintaining regular employment [15,16]. This may be caused by thefact that they face appropriate restrictions, such as those relating todriving or working in situations in which they might be liable to injury.Theymay also be the victims of ignorance and stigma [17]. Interestingly,for self-reported driving or licensure status, employment and epilepsysurgery are predictive of driving [18]. Our results also showed thatemployment is a risk factor for patients with epilepsy to illegally drive.This observation is concordant with the study of Polychronopouloset al. who reported that the occupational purpose for driving and themaintenance of a paid job are clinical variables linked to noncompliancewith driving restrictions in Greece [2] and the one of Elliott et al. in theUSA [6].

The lack of antiepileptic treatment comprises an independent riskfactor for driving [9], which may reflect the fact that nonobedient pa-tients may also be noncompliant with their medication regime. Ouranalysis showed that there is no difference in the self-reported compli-ance between the obedient patients and the disobedient patients. How-ever, we showed that the less AEDs the patient receives, the higher thelikelihood of being disobedient regarding driving. This observation isconcordant with observations in Brazil [12], West China [9], and Korea[14]. We hypothesize that patients receiving less AEDs may considertheir illness as not severe enough and, thus, may be violating the lawbelieving that they will not have a seizure when driving.

Although Greece included the driving licensure standards as definedby the European Commission in the Greek legislation, in some cases, theGreek legislation is stricter. For example, patients with epilepsy whostill receive treatment, apart from being seizure-free for a period of2 years, they must have a normal EEG in order to receive or renewtheir driving license. In our opinion, this is overly harsh and inappropri-ate and needs to be reviewed. Moreover, the fact that the neurologisthas no authority to report his/her illegally driving patients, incase they are disobedient despite recommendations, should also bereconsidered.

In conclusion, male gender, employment, and fewer AEDs aredeterminants of disobedience regarding driving in Greek patients with

epilepsy. These characteristics may indicate the at-risk population.Clinicians should also acknowledge the possibility that only half oftheir patients correctly know the driving restrictions, and, therefore,providing the correct information may increase awareness and, thus,decrease the percentage of disobedient patients. Importantly, morestudies in different geographical areas are needed to map cultural andbehavioral differences with regard to driving among people withepilepsy.

Acknowledgments

We are sincerely thankful to Professor Tatumwho allowed us to usehis questionnaire. We are also sincerely thankful to the patients whoparticipated in the study.

Conflict of interest

The authors declare that there are no conflicts of interest.

References

[1] Fisher RS, Parsonage M, Beaussart M, Bladin P, Masland R, Sonnen AE, et al. Epilepsyand driving: an international perspective. Joint Commission on Drivers' Licensing ofthe International Bureau for Epilepsy and the International League Against Epilepsy.Epilepsia 1994;35(3):675–84.

[2] Polychronopoulos P, Argyriou AA, Huliara V, Sirrou V, Gourzis P, Chroni E. Factorsassociated with poor compliance of patients with epilepsy driving restrictions.Neurology 2006;67(5):869–71.

[3] Gazette of Government of Greek Democracy, 1:101/27; 2012.[4] Tatum WO, Worley AV, Selenica ML. Disobedience and driving in patients with

epilepsy. Epilepsy Behav 2012;23(1):30–5.[5] Bautista RE, Wludyka P. Driving prevalence and factors associated with driving

among patients with epilepsy. Epilepsy Behav 2006;9(4):625–31.[6] Elliott JO, Long L. Perceived risk, resources, and perceptions concerning driving and

epilepsy: a patient perspective. Epilepsy Behav 2008;13(2):381–6.[7] Drazkowski JF, Neiman ES, Sirven JI, McAbee GN, Noe KH. Frequency of physician

counseling and attitudes toward driving motor vehicles in people with epilepsy:comparing a mandatory-reporting with a voluntary-reporting state. EpilepsyBehav 2010;19(1):52–4.

[8] Bielen I, Hajnšek S, Krmpotić P, Petelin Z, Sušak R, Sepić-Grahovac D, et al. Impact ofpartial liberalization of driver's license regulations on the driving behavior of peoplewith epilepsy: experience from Croatia. Epilepsy Behav 2011;21(4):459–61.

[9] Chen J, Yan B, Lu H, Ren J, Zou X, Xiao F, et al. Driving among patients with epilepsyin West China. Epilepsy Behav 2014;33:1–6.

[10] Commission on Classification and Terminology of the International League againstEpilepsy. Proposal for revised clinical and electroencephalographic classification ofepileptic seizures. Epilepsia 1981;22:489–501.

[11] Commission on Classification and Terminology of the International League againstEpilepsy. Proposal for classification of epilepsies and epileptic syndromes. Epilepsia1985;26:268–78.

[12] Bicalho MA, Sukys-Claudino L, Guarnieri R, Lin K, Walz R. Socio-demographic andclinical characteristics of Brazilian patients with epilepsy who drive and their asso-ciation with traffic accidents. Epilepsy Behav 2012;24(2):216–20.

[13] Hansotia P, Broste SK. Epilepsy and traffic safety. Epilepsia 1993;34(5):852–8.[14] No YJ, Lee SJ, Park HK, Lee SA. Factors contributing to driving by people with uncon-

trolled seizures. Seizure 2011;20(6):491–3.[15] Zis P, Yfanti P, Siatouni A, Tavernarakis A, Gatzonis S. Determinants of depression

among patients with epilepsy in Athens, Greece. Epilepsy Behav 2014;33:106–9.[16] Zis P, Yfanti P, Siatouni A, Tavernarakis A, Gatzonis S. Validation of the Greek version

of the Neurological Disorders Depression Inventory for Epilepsy (NDDI-E). EpilepsyBehav 2013;29(3):513–5.

[17] Gloag D. Epilepsy and employment. Br Med J (Clin Res Ed) 1985;291(6487):2–3.[18] Classen S, Crizzle AM, Winter SM, Silver W, Eisenschenk S. Evidence-based review

on epilepsy and driving. Epilepsy Behav 2012;23(2):103–12.