www.elsevier.com/locate/yebeh

Epilepsy & Behavior 10 (2007) 456–462

Epilepsy-related knowledge, attitudes, and practicesamong Zambian police officers

Edward Mbewe a,*, Alan Haworth b, Masharip Atadzhanov c,Elwyn Chomba d, Gretchen L. Birbeck e,f

a Chainama Hills College Hospital, P.O. Box 30043, Lusaka, Zambiab Department of Psychiatry, University of Zambia School of Medicine, Lusaka, Zambiac Department of Medicine, University of Zambia School of Medicine, Lusaka, Zambia

d Department of Pediatrics and Child Health, University of Zambia School of Medicine, Lusaka, Zambiae Michigan State University’s International Neurologic and Psychiatric Epidemiology Program, East Lansing, MI, USA

f Chikankata Health Services Epilepsy Care Team, Mazabuka, Zambia

Received 23 September 2006; revised 26 December 2006; accepted 28 December 2006Available online 23 March 2007

Abstract

Objective. In Zambia, where emergency medical services are very limited, the police are frequently called to the scene for unaccom-panied people experiencing seizures or exhibiting disturbed behaviors during a seizure. Police officers receive no formal medical trainingto manage such encounters. We developed and administered a police-specific survey to assess knowledge, attitudes, and practices (KAP)regarding epilepsy among police officers in Zambia.

Methods. In 2004, a 28-item KAP questionnaire that included queries specific to police encounters with seizures and epilepsy wasdeveloped and delivered to a random sample of 200 police officers stationed in Lusaka. Descriptive data were reviewed and open textquestions postcoded and categorized.

Results. The response rate was 87.5% (n = 175). Police were familiar with epilepsy, with 85% having witnessed a seizure. Although77.1% recognized epilepsy as a brain disorder, almost 20% blamed spirit possession, 13.9% associated epilepsy with witchcraft, and morethan half the respondents believed epilepsy is contagious. When asked how they would treat someone brought in for disturbing the peaceduring a seizure, most police provided supportive or neutral responses, but 8% reported taking harmful actions (arrest, detain, handcuff,restrain), and 14.3% indicated that people with epilepsy in police custody require quarantine.

Conclusions. A significant number of police officers in Zambia lack critical knowledge regarding epilepsy and self-report detrimentalactions toward people with seizures. In regions of the developing world where the police provide emergency medical services, policeofficers need to be a target for educational and social intervention programs.� 2007 Elsevier Inc. All rights reserved.

Keywords: Stigma; Police; Tolerance; Knowledge; Epilepsy; Law enforcement; Prison

1. Introduction

Zambia, a land-locked country located in southern Afri-ca (Fig. 1), has a population of �10 million people, andepilepsy prevalence rates in Zambia appear to substantiallyexceed rates in more developed countries [1]. As with most

1525-5050/$ - see front matter � 2007 Elsevier Inc. All rights reserved.

doi:10.1016/j.yebeh.2006.12.010

* Corresponding author. Fax: +260 1 283824.E-mail address: embewe2001@yahoo.com (E. Mbewe).

other countries of sub-Saharan Africa (SSA), Zambia’smajor public health challenges include HIV/AIDS. Eco-nomic development in the region is slow, and the WorldBank ranks Zambia among the poorest of nations [2].

A substantial body of data supports the existence of adisproportionate burden of epilepsy in SSA relative tomore developed regions and a substantial social and eco-nomic burden from the disease in the region [1,3–5].Research in Kenya and Zambia has found that neurological

Fig. 1. Location of Zambia in Africa.

E. Mbewe et al. / Epilepsy & Behavior 10 (2007) 456–462 457

impairment is quite common among both children andadults with epilepsy [4,6]. Misperceptions regarding thenature of epilepsy may cause persons with epilepsy to seekcare from traditional healers or not to seek care at all [7,8].A 2001 study of public health priorities in mental healthfunding identified epilepsy as one of the key conditionsin need of greater support [9]. A community-basedapproach to epilepsy care is recognized as the most likelymethod of decreasing the epilepsy treatment gap in SSA[10–14].

In urban Zambia, where public emergency medical ser-vices are very limited, the police are frequently called whenan unaccompanied person experiences a seizure. This isespecially likely if the episode includes associated confusionwith aggressive or psychotic features. In New York City,the frequency of police aid to persons experiencing seizuresin public in 1977 was examined as an index of uncontrolledseizures and to assess variations in seizure frequency by age,sex, and ethnicity [15]. That study involving 5500 peoplescreened for a history of violence during their seizures iden-tified only 7 who exhibited significant aggression duringtheir seizures. Training for police in Zambia includes basicfirst-aid medical training, but no specific information on theacute management or nature of seizures (Personal commu-nication, Superintendent Staff-Training Department, PoliceHeadquarters, Lusaka). Zambia has a centralized policeservice and personnel, and officers may be posted to anypart of the country. The police are members of an organiza-tion intended to serve the public at large and take a leadingrole in law enforcement. In addition, the police are expectedto provide a wider range of emergency and other services.

Among the norms that are intended to guide Zambianpolice actions is an expectation that the police be equitablein their treatment of citizens (Zambian Bill of Rights PartIII, Articles 21 and 23).

A series of 10 focus group discussions were conductedby our research team and involved both rural and urbanpopulations segregated into men, women, youths, and par-ents of children with epilepsy. From these discussions aswell as our own clinic experiences, it was noted that Zam-bian police are also contacted for people who are uncon-scious or confused and those who are found wanderingduring partial seizures, indicating that police encounterswith people with epilepsy in Zambia are not limited toaggressive or violent actions. Several focus group partici-pants reported encounters with police, many of them of anegative nature. For example, the father of a child with epi-lepsy whose seizures often manifested with ‘‘wandering’’behavior was threatened by the police with imprisonmentif he could not manage to keep his child at home. Anotherman with epilepsy reported physical assault at the hands ofpolice after being detained at police headquarters after apublic seizure. Therefore, we developed and administereda survey for Zambian police officers to assess their knowl-edge, attitudes, and practices (KAP) regarding epilepsy.The survey includes role-specific questions regarding pro-fessional behavior. Although 72 languages are spoken inZambia, English is the official language and the languageused for higher education.

2. Methods

2.1. Instrument

To examine how the police in Zambia understand and deal with peoplewith epilepsy, we developed a 28-item, self-administered KAP survey thatincludes generic questions used in other studies of epilepsy-associated stig-ma [16–24]. The survey assesses demographic characteristics (6 items), per-sonal proximity/experience with epilepsy (4 items), knowledge of epilepsy(11 items), recommendations for care (1 item), and social tolerance(4 items), and poses two open-ended text questions asking how the policeofficers would respond to someone with a seizure and someone they knewhad epilepsy (see the Appendix A for the complete questionnaire). The sur-vey was piloted in series with three urban Lusaka police officers who werethen removed from the sampling frame. Questions were read to them anddiscussions undertaken to ensure that the intent of the questions was appre-ciated. Minor changes in language were made to facilitate clarification. Bythe third piloting, no further queries or clarifications were necessary. Nomajor structural changes in the original instrument were necessary.

2.2. Respondent sample

Using the register of police officers from the Police Central Headquar-ters in Lusaka, a random sample of 200 were selected using a randomnumber-generating program and the police officers’ registration numbers,which are assigned sequentially at the time of hire. Unfortunately, accu-rate demographic data for the overall sampling frame were not available.The senior police officer in charge of research distributed the question-naires to the respective stations where sampled officers were on duty.Questionnaires were completed anonymously after duty hours andreturned to the Central Police Headquarters to be collected by the localinvestigators. ZMK 10,000, equivalent value to �$2.00 US, was offeredto potential participants for their time.

458 E. Mbewe et al. / Epilepsy & Behavior 10 (2007) 456–462

2.3. Analysis

Data were double-entered into Microsoft Access for accuracy beforeimportation into EPI INFO Version 3.2.2 for descriptive analysis. Inves-tigators reviewed the two open-ended police-specific scenarios and post-coded responses. To facilitate postcoding, each of the co-investigators

Table 1Descriptive data (n = 175)

DemographicAge 31.8 (7.0) [19–25]a

Males 71.3 %Years of formal education 12 (1.0) [9–13]Marital status (married) 72.3%Urban residence 95.4 %Fluent in Englishb 90.0%

Personal proximityHas heard of epilepsy 97.7%Knows someone with epilepsy 91.7%Has witnessed a seizure 85.0%Has a family member with epilepsyc 35.5%Is not familiar with epilepsyd 2.3%

KnowledgeEpilepsy is a form of:

Madness 1.2%Mental retardation 30.1%Spirit or demon possession 19.3%Brain disease 77.1%

Epilepsy is contagious:Always 1.8%Sometimes 52.1%Never 46.1%

The cause of epilepsy is:Brain injury 67.2%Hereditary (runs in families) 35.5%Birth injury 21.1%A blood disorder 31.3%Witchcraft 13.9%A curse from God 4.2%Spirit possession 18.7%Excessive worry 7.2%

Would recommend treatment viae:Doctor 85.6%Traditional healer 24.6%Church healing session 27.5%

ToleranceWould allow their child to play with a child withepilepsy

83.6%

Would allow son to marry a PWEf 56.2%Would allow daughter to marry a PWE 57.4%Believes a child with epilepsy shouldnever attend school 4.8%

a Mean (SD) [range].b Self-reported assessment.c Within the Zambian context, family is more inclusive, and this answer

reflects both first-degree (parents, brothers, sisters) and second-degree(aunts, uncles, cousins) relatives.

d Data for these individuals were dropped from further KAP-relatedanalysis.

e Health care-seeking patterns for many conditions in the region involvedual care from both a traditional healer and a physician.

f PWE, person with epilepsy.

was asked to review the transcript and independently assigned theresponse to one of three categories: ‘‘supportive,’’ ‘‘harmful,’’ or ‘‘neu-tral.’’ For the question ‘‘Someone is brought into the police station who

has had a convulsion. You do not know if the person has epilepsy. What

would you do?’’ answers were postcoded into one of these three ranked cat-egories (supportive, harmful, or neutral). For the question, ‘‘Someone with

known epilepsy is arrested and brought into the police station. What would

you do?’’ frequent responses included ambiguous actions, so this questionwas postcoded into six descriptive categories (1, proceed with arrest; 2,bond them/post bail; 3, help; 4, neutral; 5, quarantine; 6, harmful), andthe nature of the explicit answer is provided.

3. Results

A total of 175 surveys were returned for a response rate of87.5%. Descriptive data regarding demographics, personalproximity, knowledge, and tolerance are summarized inTable 1. Among completed questionnaires, there were <5%missing data for any single question. Local terms for epilepsyincluded the following: akakosh, cabana, chikonikoni, chiko-

nya, chipumputu, chituwani, kaganyuka, kakoshi, kanono,kashita, kasuntu, kifwafwa, kikonyi, kishimo, kuganyuka,

kunyu, kuwa, makwa, musamfu, nafwa, njilinjili and vizilisi.Police officers were asked ‘‘Someone is brought into the

police station who has had a convulsion. You do not know

if the person has epilepsy. What would you do?’’ Investiga-tors postcoded responses into three categories: supportive,neutral, or harmful practices (Table 2). Although 77%reported that they would offer supportive care (seek orrefer to medical services or provide first-aid), 8% indicatedthat they would (1) detain and/or arrest the individual, or(2) isolate him or her (i.e., place the person in a closed cellwithout windows, access to other persons, or free atmo-sphere), or (3) restrain the person with handcuffs or straps.Additional text provided by the respondents indicated thatfear of catching epilepsy was apparently uppermost in theminds of the officers responding in this manner.

The second open-ended, police-specific question was,‘‘Someone with known epilepsy is arrested and brought into

the police station. What would you do?’’ Results are provid-ed in Table 3. Although overtly harmful responses wererare (<1%), 14% indicated that they would place such aperson in quarantine.

Table 2Responses to the question, ‘‘Someone is brought into the police station who

has had a convulsion. You do not know if the person has epilepsy. What

would you do?’’ (n = 159)

Postcoded category N (%) Example

Supportive 115 (72.3) Seek or refer to medical careProvide first aid

Neutral 32 (20.1) ObserveWait until recoveryRefer to superiorPray for him

Harmful 12 (7.5) IsolateArrestDetainHandcuff and strap

Table 3Responses to the question, ‘‘Someone with known epilepsy is arrested and brought into the police station. What would you do?’’ (n = 153)a

Response N (%) Example

Proceed with arrest 23 (15.0) Charge him accordinglyTreat as any other criminalArrest as usual

Bond them/post bail 54 (35.3) Arrange bondArrange bond, if crime not dangerousArrange bond as quickly as possible

Help 32 (20.9) Refer to clinicTake to hospitalCall doctor

Neutral 22 (14.4) Treat fairlyPray for himSeek advice from superior

Quarantine 22 (14.4) All respondents literally said ‘‘quarantine’’

Harmful action 1 (0.7) Handcuff and strap down

a Answers were not mutually exclusive and respondent may have been coded for more than one type of response so total is greater than 100%.

E. Mbewe et al. / Epilepsy & Behavior 10 (2007) 456–462 459

4. Discussion

Although we took a random sample of registered,employed Zambian police in Lusaka, we were not able tocompare the demographic characteristics of our sample of200 with those for the overall sampling frame. Formalrecords of how frequently Zambian police encounter some-one with a seizure or epilepsy are not maintained. Our ownpotential biases in postcoding open-ended questions couldpotentially lead to a more negative perception of policeself-reported actions. Therefore, when we felt responses wereambiguous, we provided details of the response rather thansimple categories. Data presented here provide insight intopolice officers’ understanding of epilepsy in urban Lusaka.Within Zambia, there are few police in the rural regions,and extrapolation of these data to other African countriesmay not be appropriate, but there are few prior studies thathave assessed the encounters that occur between people withepilepsy and police officers, and none of these have been con-ducted in developing countries where the police are the de

facto emergency medical service providers. Although easierto analyze, the content of simple multiple-choice questionsis by its nature very limited, so we also asked open-endedquestions. Interpretation of the text in such responses isfraught with the potential for misunderstanding, but canoffer some valuable insights not available when using closedor limited responses. Clearly, further qualitative work wouldoffer more valuable insights into our understanding ofhow Zambian police conceptualize epilepsy and respond topeople with the disorder.

In the early 1990s, the police in Zambia underwent vigor-ous reforms, including a change in title from ‘‘Police Force’’to ‘‘Police Service’’ and an increase in level of educationrequired to join the Police Service from grade 7 to grade 12(i.e., secondary school completion). In 2002, 26% of womenand 14% of men in Zambia were reported to be illiterate [25].In our sample of police, more than 90% of respondents had

completed secondary school, illustrating that police officersin Zambia are substantially more educated than the generalpopulation they serve. Despite their educational status, theZambian police officers surveyed demonstrate intolerancefor people with epilepsy and knowledge gaps regarding thebiomedical nature of the disease, with supernatural causesand contagion fears being common. Although the majorityof police indicated that they would provide assistance to peo-ple they encounter with seizures while on duty, a substantialminority (�20%) reported that they would undertakeactions that could be harmful and/or stigmatizing to some-one with epilepsy, with some believing that people with epi-lepsy require isolation or quarantine. In this survey, thepolice officers commonly indicated a desire to post bond/bailfor people with epilepsy who have been arrested. This is dif-ficult to interpret, as it could reflect either a desire to help theindividual or contagion fears, and deserves further qualita-tive study.

Despite its limitations, to the best of our knowledge, thisstudy represents the first systematic evaluation of policeofficers attitudes toward seizures and people with epilepsyin a resource-poor environment. Our findings indicate thatZambian police officers require further education regardingthe noncontagious nature of epilepsy and acute seizuremanagement, as well as some social sensitization regardingthe condition. In less developed regions where emergencymedical services are extremely limited, law enforcementpersonnel should be an important target for educationalprograms aimed at decreasing epilepsy-associated morbid-ity, mortality, and stigma.

Acknowledgments

Funding for this work was provided by the US NationalInstitutes of Health (NINDS R21 NS48060). We appreci-ate the cooperation of the Zambian Police Service insupporting the conduct of this study.

460 E. Mbewe et al. / Epilepsy & Behavior 10 (2007) 456–462

Appendix A. KAP survey police version

E. Mbewe et al. / Epilepsy & Behavior 10 (2007) 456–462 461

Thank you for participating in this survey. We welcome any comments you have regarding this survey. Please placethem below.

462 E. Mbewe et al. / Epilepsy & Behavior 10 (2007) 456–462

References

[1] Birbeck GL, Kalichi EM. Epilepsy prevalence in rural Zambia: adoor-to-door survey. Trop Med Int Health 2004;9:92–5.

[2] WorldBank. Zambia—Country Brief. Available from: http://web.worldbank.org/WBSITE/EXTERNAL/COUNTRIES/AFRI-CAEXT/ZAMBIA. Last accessed: 9 December 2006.

[3] Snow RW, Williams RE, Rogers JE, Mung’ala VO, Peshu N. Theprevalence of epilepsy among a rural Kenyan population: itsassociation with premature mortality. Trop Geogr Med1994;46:175–9.

[4] Birbeck G, Kalichi E. The functional status of people with epilepsy inrural sub-Saharan Africa. J Neurol Sci 2003;209(1–2):65–8.

[5] Birbeck GL, Kalichi EM. Famine-associated AED toxicity in ruralZambia. Epilepsia 2003;44:1127.

[6] Mung’ala-Odera V, Meehan R, Njuguna P, Mturi N, Alcock KJ,Newton CR. Prevalence and risk factors of neurological disabilityand impairment in children living in rural Kenya. Int J Epidemiol2006;35:683–8.

[7] Baskind R, Birbeck GL. Epilepsy-associated stigma in sub-SaharanAfrica: the social landscape of a disease. Epilepsy Behav2005;7:68–73.

[8] El Sharkawy G, Newton C, Hartley S. Attitudes and practices offamilies and health care personnel toward children with epilepsy inKilifi, Kenya. Epilepsy Behav 2006;8:201–12.

[9] Modiba P, Schneider H, Porteus K, Gunnarson V. Profile ofcommunity mental health service needs in the Moretele District(North-West Province) in South Africa. J Ment Health Policy Econ2001;4:189–96.

[10] Danesi MA. Patient perspectives on epilepsy in a developing country.Epilepsia 1984;25:184–90.

[11] De Boer HM, Engel Jr J, Prilipko LL. Out of the shadows: apartnership that brings progress! Epilepsia 2005;46(Suppl. 1):61–2.

[12] Diop AG, de Boer HM, Mandlhate C, Prilipko L, Meinardi H. Theglobal campaign against epilepsy in Africa. Acta Trop2003;87:149–59.

[13] IOM. Neurological, psychiatric, and developmental disorders: meet-ing the challenge in the developing world. Washington, DC: Natl.Academy Press; 2001.

[14] Leonardi M, Ustun TB. The global burden of epilepsy. Epilepsia2002;43(Suppl. 6):21–5.

[15] Neugebauer R, Oppenehimer G, Susser M. Seizures in public placesin New York City. Am J Public Health 1986;76:1115–9.

[16] Jacoby A, Gorry J, Gamble C, Baker GA. Public knowledge, privategrief: a study of public attitudes to epilepsy in the United Kingdomand implications for stigma. Epilepsia 2004;45:1405–15.

[17] Prpic I, Korotaj Z, Vlasic-Cicvaric I, Paucic-Kirincic E, Valerjev A,Tomac V. Teachers’ opinions about capabilities and behavior ofchildren with epilepsy. Epilepsy Behav 2003;4:142–5.

[18] Hills MD, MacKenzie HC. New Zealand community attitudestoward people with epilepsy. Epilepsia 2002;43:1583–9.

[19] Desai P, Padma MV, Jain S, Maheshwari MC. Knowledge, attitudesand practice of epilepsy: experience at a comprehensive rural healthservices project. Seizure 1998;7:133–8.

[20] Fong CY, Hung A. Public awareness, attitude, and understanding ofepilepsy in Hong Kong Special Administrative Region, China.Epilepsia 2002;43:311–6.

[21] Dent W, Helbok R, Matuja WB, Scheunemann S, Schmutzhard E.Prevalence of active epilepsy in a rural area in South Tanzania: adoor-to-door survey. Epilepsia 2005;46:1963–9.

[22] Novotna II, Rektor I. The trend in public attitudes in the CzechRepublic towards persons with epilepsy. Eur J Neurol2002;9:535–40.

[23] Atadzhanov M, Chomba E, Haworth A, Mbewe E, Birbeck G.Knowledge, attitudes, behaviors, and practices (KABP) regardingepilepsy among Zambian clerics. Epilepsy Behav 2006;9:83–8.

[24] Birbeck G, Chomba E, Atadzhanov M, Mbewe E, Haworth A.Zambian teachers: what do they know about epilepsy and how canwe work with them to decrease stigma? Epilepsy Behav2006;9:275–80.

[25] Earthtrends. earthtrends.wri.org/pdf_library/country_profile/pop_cou)894.pdf. Last accessed: 9 December 2006.