- Email: [email protected]
Prevalence of active epilepsy in rural Tanzania: A large community-based survey in an adult population
Seizure 21 (2012) 691–698
Contents lists available at SciVerse ScienceDirect
Seizure journal homepage: www.elsevier.com/locate/yseiz
Prevalence of active epilepsy in rural Tanzania: A large community-based survey in an adult population Ewan Hunter a,d,*, Jane Rogathi b, Simukai Chigudu c, Ahmed Jusabani b, Margaret Jackson d, Richard McNally e, William Gray a, Roger G. Whittaker d, Ahmed Iqbal d, Daniel Birchall d, Eric Aris f, Richard Walker a a
Northumbria Healthcare NHS Foundation Trust, North Shields, NE29 8NH, UK Kilimanjaro Christian Medical Centre, PO Box 3010, Moshi, Tanzania c Newcastle University Medical School, Newcastle NE2 4HH, UK d Royal Victoria Infirmary, Newcastle NE1 4LP, UK e Institute of Health and Society, Newcastle University, NE2 4AX, UK f Muhimbili University College of Health Sciences, P.O. Box 10028, Dar es Salaam, Tanzania b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 29 April 2012 Received in revised form 18 July 2012 Accepted 20 July 2012
Purpose: To estimate the prevalence of active epilepsy in adults in an established demographic surveillance site in rural Tanzania. To describe the clinical characteristics of epilepsy and to estimate the treatment gap in this population. Methods: A pilot study established that a previously validated screening questionnaire was sensitive for detecting cases of epilepsy in a Kiswahili-speaking Tanzanian population. A door-to-door census of the adult population (total 103,026) used the screening questionnaire to identify possible cases of epilepsy, who were then assessed by a research doctor to establish a diagnosis of epilepsy or otherwise. The prevalence of active epilepsy in this population was estimated with age-standardisation to the WHO standard population. Seizure types and epilepsies were classified according to current recommendations of the International League Against Epilepsy. The treatment gap for epilepsy was estimated based on antiepileptic drug use as reported by cases. Results: Two hundred and ninety-one cases of active epilepsy, all with convulsive seizures, were identified. The age-standardised prevalence was 2.91/1000 adults (95% CI 2.58–3.24); the crude prevalence adjusted for non-response was 3.84/1000 adults (95% CI 3.45–4.20). Focal-onset seizures accounted for 71.5% of all cases identified. The treatment gap was 68.4% (95% CI 63.0–73.7). Conclusions: This is one of the largest community-based studies of the prevalence of epilepsy in adults conducted in sub-Saharan Africa to date. We identified a lower prevalence than has previously been described in this region. The high proportion of focal onset seizures points to a large burden of acquired, and possibly preventable, epilepsy in this population. A treatment gap of 68.4% confirms that interventions to raise awareness of the treatable nature of epilepsy are warranted in this and similar populations. ß 2012 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved.
Keywords: Classification Epidemiology Epilepsy Prevalence Sub-Saharan Africa Treatment gap
1. Introduction 1.1. Background Epilepsy accounts for up to 1% of the global burden of disease.2 An estimated 32.7 million people suffer from active epilepsy worldwide,3 with 80% or more of people with epilepsy (PWE) living
* Corresponding author at: Department of Infection and Tropical Medicine, Elliott Building, Royal Victoria Infirmary, Newcastle-upon-Tyne, NE1 4LP, United Kingdom. Tel.: +44 0 191 282 9751; fax: +44 0 191 282 0810. E-mail address: [email protected] (E. Hunter).
in low and middle income countries (LMICs).4 A further 500 million people (including families and carers of PWE) are thought to be indirectly affected by epilepsy.5 It is generally accepted that the incidence, prevalence and mortality associated with epilepsy are all higher in LMICs than in high income countries (HICs), although estimates have varied widely and findings from different studies are difficult to compare due to differing inclusion criteria, clinical definitions and the difficulties associated with establishing reliable baseline demographic data.3,6–9 While up to four-fifths of the potential market for anti-epileptic drugs (AEDs) is in LMICs, 90% or more of PWE in these countries may receive no treatment at all.5,10,11 The proportion of PWE who require treatment but who are not receiving it is known as the treatment gap (TG).12
1059-1311/$ – see front matter ß 2012 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.seizure.2012.07.009
692
E. Hunter et al. / Seizure 21 (2012) 691–698
1.2. Objectives We conducted a door-to-door survey in an established demographic surveillance site (DSS) in rural Tanzania. We sought to estimate the prevalence of active epilepsy, to describe the pattern of epilepsy and to quantify the TG in this population. With a total denominator population of 161,119, this is one of the largest community-based epidemiological studies of epilepsy to be conducted in sub-Saharan Africa (SSA) to date. Paediatric and adult studies were conducted in parallel. Findings from the paediatric population have been published elsewhere13; we report here on findings in the adult population, defined as individuals aged 15 years and above.
enumerators were also asked to identify any individuals within their communities who they knew, or suspected, suffered from a seizure disorder. Finally, results of a contemporaneous study utilising data from the same census to screen for neurological disability, including epilepsy, in 2232 individuals aged over 70 provided further cases.21 A collateral history was sought in all instances to aid in diagnosis and classification of seizures. To be included in the study people diagnosed with epilepsy had to be alive and resident in the district on the prevalence date. The age cut-off for adults of 15 years and above was chosen to reflect local practice in terms of whether individuals are managed by adult or paediatric medical services. 2.4. Variables
2. Methods 2.1. Study design A door-to-door census of the entire DSS population was completed between January and June 2009 and a previously validated screening questionnaire was used to detect possible cases of epilepsy.1 The questionnaire was forward and back translated between English and Kiswahili (KS) and a pilot study was conducted to assess the questionnaire’s sensitivity for cases of epilepsy in a KS-speaking Tanzanian population. Volunteers for the pilot study included patients with and without epilepsy at Muhimbili National Hospital, Dar es Salaam, and locally at Kilimanjaro Christian Medical Centre (KCMC), Moshi. During preparations for the census, all enumerators received training in the identification of likely cases of epilepsy and were also involved in focus group discussions of each item in the screening questionnaire, with further refinements to the KS translation being made to optimise its utility in the field. Details of the screening questionnaire and KS translation are given in Appendix. 2.2. Study setting The Hai district lies on the slopes of Mount Kilimanjaro in northeast Tanzania, covering an area of approximately 13,000 km2. Villages are served by a network of unpaved roads of varying quality, many of which are impassable to motorised vehicles during the twice-yearly rainy seasons. Agriculture, livestock keeping, dairy farming, commercial mining and cottage industries are the main economic activities.14 Facilities for Electroencephalography (EEG) and computed tomography (CT) are available at KCMC, a large referral hospital in the nearby town of Moshi. Hai was established as one of three DSSs in Tanzania by the Adult Morbidity and Mortality Project (AMMP), an epidemiological programme funded by the UK Department for International Development and conducted in partnership between the Tanzanian Ministry for Health and Newcastle University, UK.15 Between 1992 and 2004 the population of each DSS was enumerated by an annual census, with all households and household members being assigned unique identifying numbers. The DSS infrastructure in Hai has subsequently been maintained through a series of epidemiological studies, and there is currently at least one trained census enumerator per village.16–21 These enumerators facilitate access to their communities by visiting researchers and are coordinated by a team of supervisors who are employed in clinical posts by the district health authority.
All individuals diagnosed with epilepsy were interviewed in depth to establish seizure history, possible risk factors, treatment history and socio-demographic features. The interview format was drawn from the clinical experience of the research team and from previous field work experience in this study population.16,18,22 A semistructured element was included to facilitate subsequent review and validation of diagnoses.23 All cases were offered further investigation with CT head scan and EEG. CT scans were reported locally to exclude any acute pathology, and then reviewed by neuroradiologists in the UK (DB, AI). There are no facilities for interpreting EEGs locally and all recordings were read in the UK by a consultant neurophysiologist (RGW). Any individuals identified with epilepsy who were not being treated adequately were started on treatment and referred to services locally for follow-up. Clinical details and investigation findings for all patients were reviewed by a consultant neurologist in the UK with an interest in epilepsy (MJ). CT and EEG findings were correlated with clinical details to classify epilepsies according to currently accepted criteria.24,25 Active epilepsy was defined as at least two unprovoked seizures occurring at least 24 h apart, with the most recent seizure occurring within the 5 years prior to the prevalence date of 1st of June 2009. Individuals who did not satisfy these criteria, but who were maintained on AEDs for an appropriate indication, were also considered to have active epilepsy. Estimates for the TG were based on individuals who reported either no or inadequate AED therapy at the time of the study. All data were entered into Microsoft Excel 2007 as field work progressed and then re-entered into Microsoft Access 2007, with any discrepancies being corrected with reference to the original case notes. 2.5. Statistical methods Statistical analyses were performed using SPSS-17 for Windows (SPSS, Chicago, IL, USA). Age- and sex-specific prevalence rates for the Hai adult population were calculated in five-year age-bands. Prevalence was age-standardised to the WHO world standard population using the direct method26; 95% confidence intervals (CIs) were calculated based on the assumptions of the binomial distribution. As this study is concerned only with individuals aged 15 years and above, the Hai and WHO populations were weighted accordingly (Table 5). To look for any evidence of clustering, oneway analysis of variance (ANOVA) was used to compare normally distributed mean prevalences from administrative divisions within the district. 3. Results
2.3. Participants and data sources 3.1. Pilot study All individuals who responded positively to the screening questionnaire at census were assessed in detailed by a research doctor trained in the diagnosis of epilepsy (EH). In addition,
Forty-nine subjects (27 cases of epilepsy and 22 controls) were recruited into the pilot study. The KS translation of the screening
E. Hunter et al. / Seizure 21 (2012) 691–698
693
Table 1 Response rates to individual screening questions.
1 2
3 4 5 Fig. 1. Hai DSS adult population structure (2009 census). 6
questionnaire performed with 100% sensitivity and 54.5% specificity, with positive and negative predictive values (PPV, NPV) of 73% and 100% respectively.
7 8 9
3.2. Participants The 2009 census identified a total population of 161,119, of whom 103,026 (63.9%) were aged 15 years and over (54,386 female, 48,640 male) (Fig. 1). Five hundred and sixty-eight adult individuals (286 female, 282 male) responded positively to the screening questionnaire. Response rates to individual questions are given in Table 1. Three hundred and ninety-five (69.5%) of these were clinically assessed, with a diagnosis of active epilepsy being made in 209 (PPV 52.9%). Of the 173 individuals that were not available for assessment, 29 (16.8%) had moved or travelled from the district at the time of the study, 21 (12.1%) could not be traced, 18 (10.4%) refused, 12 (6.9%) had died since the census (four deaths prior to the prevalence date) and four (2.3%) were accounted for by data entry errors (duplicate names or wrong dates of birth). A further 89 (51.4%) did not attend for assessment on two or more occasions. In addition to census responders, a further 130 individuals (60 female, 70 male) identified by village enumerators as possible cases were assessed, with active epilepsy being diagnosed in 80 (61.5%). The concurrent study of neurological disability in the elderly identified five cases of active epilepsy, of
Question
Positive responses n
(%)
Have you ever had attacks of shaking of the arms or legs which you could not control? Have you ever had attacks in which you fall suddenly, without any reason, changing colour in the palms, lips or face? Have you ever lost consciousness? Have you ever had attacks in which you fall with loss of consciousness? Have you ever had attacks in which you fall and bite your tongue? Have you ever had attacks in which you fall and lose control of your bladder? Have you ever had brief attacks of shaking or trembling in one arm or leg or in the face? Have you ever had attacks in which you lose contact with the surroundings or experience abnormal smells? Have you ever been told that you have or have had convulsions, epilepsy or epileptic fits?
259
(45.6)
363
(63.9)
362 316
(63.7) (55.6)
213
(37.5)
177
(31.2)
200
(35.2)
152
(26.8)
173
(30.5)
which one was identified independently by the epilepsy study, with a further two being available for assessment and inclusion in the final case series. Epilepsy was suspected in a further 14 individuals, although insufficient history was obtained, either directly or from an eye witness, to confirm the diagnosis. These individuals remain under follow-up. In summary, 527 individuals were assessed clinically with 291 cases of active epilepsy being identified (Fig. 2). An eye witness was available in 202 (69.4%) of 291 confirmed cases. CT head scans were available in 200 (68.7%) of 291 confirmed cases, with cerebral abnormalities on 51 (25.5%) scans. EEGs were available from 216 (74.2%) of 291 cases, with abnormalities consistent with epilepsy seen in 84 (38.9%) recordings. CT and EEG findings are summarised in Tables 2 and 3. 3.3. Clinical features and treatment history All 291 cases identified were suffering from active epilepsy with convulsive seizures (AEWCS). Focal onset seizures were identified
Fig. 2. Summary of case ascertainment.
E. Hunter et al. / Seizure 21 (2012) 691–698
694 Table 2 Abnormalities on CT scans. CT scan finding
n
(%)
Vascular Cysticercosis Trauma Encephalomalacia Ventricular dilatation Hemispheric atrophy Hydrocephalus Lobar atrophy Tumour Generalised atrophy Lobar calcification Porencephalic cyst Total
13 8 5 5 5 4 3 3 2 1 1 1 51
(25.5) (15.7) (9.8) (9.8) (9.8) (7.8) (5.9) (5.9) (3.9) (2.0) (2.0) (2.0) (100)
in 208 (71.5%) of 291 cases, generalised onset seizures in 7 (2.4%) cases, with seizures remaining unclassified in 76 (26.1%) cases. Six (85.7%) of seven cases with generalised onset seizures had EEGs suggestive of genetic generalised epilepsy, two of which could be sub-classified as juvenile myoclonic epilepsy. Sixty-eight (32.7%) of 208 cases with focal onset seizures were classified as structural/ metabolic. Epilepsy remained unclassified in 274 (74.6%) of 291 cases: 140 (67.3%) of 208 focal cases, one (14.3%) of seven generalised cases, and in all 76 cases (100.0%) in which seizures could not be classified (Table 6). The relative contributions of diagnostic modalities to identifying focal onset seizures are illustrated in Fig. 3. Two-hundred and fifty-three (86.9%) of 291 cases had previously sought medical attention for their seizures, of whom 122 (48.2%) reported using AEDs at the time of the study. Of these, 30 (24.6%) were using doses likely to be sub-therapeutic, based on published prescribing information27,28 and guidance on managing epilepsy in resource-poor settings.29 Ninety-two (31.6%) of 291 cases were therefore considered as adequately treated with AEDs at the time of the study.
Fig. 3. Relative contributions of diagnostic modalities to identifying focal-onset seizures in Hai.
deferred in 14 individuals and two cases identified by the other study were not available for assessment by the epilepsy study team. A confirmed diagnosis in these 16 would bring the total case series to 394 (crude prevalence 3.82 per 1000; 95% CI 3.45–4.20). While the prevalence of epilepsy varied from village to village (0–6.24/1000), there were no significant differences in mean prevalence between the five administrative divisions within the district (ANOVA, p = 0.991). A conservative estimate of the TG for epilepsy in the adult Hai population, based on AED use, was 68.4% (95% CI 63.0–73.7). 4. Discussion 4.1. Key results and interpretation
3.4. Prevalence and treatment gap The crude prevalence of AEWCS in the Hai adult population on 1st June 2009 was 2.82 per 1000 adults (95% CI 2.50–3.15). Ageand sex-specific prevalence rates are given in Table 4. The crude prevalence was higher in males than in females (difference in proportions 0.69/1000; 95% CI 0.03–1.34). To accommodate small numbers in the older population, cases aged 70 years or more were combined into a single age band of 70+ prior to age-standardisation (Table 5). The age-standardised prevalence of AEWCS in the adult Hai population was 2.91 per 1000 adults (95% CI 2.58–3.24). After accounting for deaths prior to the prevalence date and data entry errors, a further 165 (95.4%) of the 173 individuals who screened positive during the census but who were not assessed clinically were potentially eligible for the study. Applying the PPV of 52.9% for detection of active epilepsy by the screening questionnaire, an additional 87 cases may have been missed, which would bring the total number of active cases to 378 (crude prevalence 3.67 per 1000; 95% CI 3.3–4.0). Diagnoses were Table 3 Abnormalities on EEG. EEG finding
n
(%)
Temporal lobe abnormalities Non-epileptiform abnormalities Extra-temporal focal abnormalities Multifocal epileptiform abnormalities Generalised epileptiform abnormalities Total
27 25 16 10 6 84
(32.1) (29.8) (19.0) (11.9) (7.1) (100)
This study is one of the largest community-based surveys of the prevalence of epilepsy in an adult population to be conducted in SSA to date. The age-standardised prevalence for AEWCS in adults of 2.91/1000 (95% CI 2.58–3.24), with crude prevalence adjusted for non-response of 3.84/1000 (95% CI 3.47–4.22), is considerably lower than both the median prevalence of 15/1000 previously reported from SSA9 and the range for active prevalence of 8–10/ 1000 previously reported from Tanzania.30–32 It is, however, in line with a recent study of comparable size and design in a rural Kenyan population which found a crude prevalence of 2.9/1000 (95% CI 2.6–3.2), rising to 4.5/1000 (95% CI 4.1–4.9) after adjustment for non-response and sensitivity.33 This latter study sought to identify active convulsive epilepsy only, and the lower prevalence figures of both studies may reflect this. In community-based studies from SSA heterogeneity of prevalence within study populations has been frequently observed.9,31,33 Whether this reflects clustered exposure to environmental or genetic risk factors or a tendency for PWE to live closer to locally available treatment services remains unclear. In our study case ascertainment was grossly uniform across administrative divisions, with no observation of clustering. We found a higher prevalence in males than in females, although the difference was marginal. A review of 19 studies of the prevalence in SSA in which data on gender were available found a mean ratio of males to females of 1.4, although the ratio was significantly different in only a quarter of studies.9 It has been suggested that a higher prevalence of epilepsy in males in Tanzania may reflect a higher risk of cranial trauma through a tendency to
E. Hunter et al. / Seizure 21 (2012) 691–698
695
Table 4 Age- and sex-specific prevalence of epilepsy in adults in Hai. Male
Female Age-band (years) 15–19 20–24 25–29 30–34 35–39 40–44 45–49 50–54 55–59 60+ Total
Cases (n)
Denominator population (n)
23 25 23 12 9 10 13 7 6 8
8114 5225 5487 5381 5216 4800 4156 3727 2911 9369
136
54,386
Prevalence per 1000 (95% CI) 2.83 4.78 4.19 2.23 1.73 2.08 3.13 1.88 2.06 0.85
Cases (n)
Total Denominator population (n)
(1.68–3.99) (2.91–6.66) (2.48–5.90) (0.97–3.49) (0.60–2.85) (0.79–3.37) (1.43–4.83) (0.49–3.27) (0.41–3.71) (0.26–1.45)
19 36 14 18 13 17 11 7 8 12
8648 4719 4467 4961 4852 4234 3630 2986 2437 7706
2.50 (2.08–2.92)
155
48,640
engage in more dangerous occupations.32 Conversely, other studies from Tanzania have found higher proportions of females with epilepsy.30,31 It has been postulated that this may be accounted for by a higher mortality rate among males, again because of dangerous occupations,31 although Tanzania is a country with a high maternal mortality rate, and the strength of this argument could be questioned.34 It has been observed that women in one study population were more health-conscious then men, being accustomed to attending maternity and child-care clinics and speaking openly about health issues, possibly contributing to higher ascertainment rates in women.30 Overall there remains no clear consensus on whether observed gender differences in the prevalence of epilepsy in SSA are due to genderspecific aetiological factors, bias arising through artefact in caseascertainment or competing mortality risks.35 Studies from Europe and the USA have identified peaks in the incidence of epilepsy in young children and in the elderly.36–38 We did not find a peak in prevalence in those aged 60 and older, although this age group represents only 10.6% of the total Hai population, and it is difficult to draw comparisons with populations from HICs which have very different population structures. We did find a peak in prevalence in those aged 20–24 years (6.13/1000), with a concentration of male cases in this age group (crude prevalence in males 7.63/1000). This is in keeping with a higher prevalence in adolescents and young adults identified by other studies in rural SSA.9,30–33 Whether the lower prevalence in later adult life points to early mortality or spontaneous remission remains unclear.33 A similar distribution in both the active and lifetime prevalence of epilepsy has been observed in Tanzania,
Prevalence per 1000 (95% CI) 2.20 7.63 3.13 3.63 2.68 4.02 3.03 2.34 3.28 1.56
(1.21–3.18) (5.15–10.11) (1.49–4.77) (1.96–5.30) (1.22–4.13) (2.11–5.92) (1.24–4.82) (0.61–4.08) (1.01–5.55) (0.68–2.44)
3.19 (2.69–3.69)
Cases (n)
Denominator population (n)
42 61 37 30 22 27 24 14 14 20
16,762 9944 9954 10,342 10,068 9034 7786 6713 5348 17,075
291
103,026
Prevalence per 1000 (95% CI) 2.51 6.13 3.72 2.90 2.19 2.99 3.02 2.09 2.62 1.17
(1.75–3.26) (4.60–7.67) (2.52–4.91) (1.86–3.94) (1.27–3.10) (1.86–4.11) (1.85–4.31) (0.99–3.18) (1.25–3.99) (0.66–1.68)
2.82 (2.50–3.99)
suggesting that those in remission may not report their previous seizures.31,39 Conversely, there is mounting evidence of early mortality in epilepsy in LMICs, in excess of that seen in HICs.40 Observational data suggest that this is likely to be the case in SSA,41–43 although there are very few formalised data with comparable standardised mortality ratios.43 All PWE identified in our study had AEWCS, with focal-onset seizures accounting for 71.5% of all cases. High rates of focal onset seizures were found in 71% of all cases identified in a recent community-based study of people with AEWCS in Kenya,44 and in 55% of all cases in a community-based study from Nigeria, although it is not clear how focal seizures with generalised clonic movements were classified in this latter study.45 Conversely, higher rates of generalised-onset seizures have previously been reported from community-based studies in SSA, accounting for between 58% and 82% of diagnoses.31,32,46 Differing approaches to case-ascertainment, differing diagnostic and classification criteria, and lack of EEG and CT to aid with classification make a strict comparison between studies difficult, although all authors comment that focal seizures becoming rapidly generalised may have erroneously been classified as generalised onset seizures. The high rate of focal onset seizures identified in our study suggests that a high proportion of epilepsy in this population may be due to identifiable, and possibly preventable, causes. We know that epilepsy in children from Hai is associated with adverse perinatal events13, while exposure to Taenia solium, the parasitic infection responsible for neurocysticercosis (NCC), is associated with epilepsy in a number of African populations,47 including Tanzania.48,49 The role that this latter aetiology might play in an area where the prevalence of epilepsy is low, such as Hai, is not clear,
Table 5 Age distributions of Hai and WHO standard populations, weighted for adults (15 years). Age band (years)
Hai population (% of total)
Hai adult weighted (%)
WHO (%)
WHO adult weighted (%)
0–14 15–19 20–24 25–29 30–34 35–39 40–44 45–49 50–54 55–59 60–64 65–69 70+ All-ages Adults
58,093 16,762 9944 9954 10,342 10,068 9034 7786 6713 5348 4339 3867 8869 161,119 103,026
– 16.3 9.7 9.7 10.0 9.8 8.8 7.6 6.5 5.2 4.2 3.8 8.6 – 100.2a
26.2 8.5 8.2 7.9 7.6 7.2 6.6 6.0 5.4 4.6 3.7 3.0 5.3 100.2a –
– 11.5 11.1 10.7 10.3 9.7 8.9 8.2 7.3 6.2 5.0 4.0 7.2 – 100.1a
a
(36.1) (10.4) (6.2) (6.2) (6.4) (6.3) (5.6) (4.8) (4.2) (3.3) (2.7) (2.4) (5.5) (100.1)a (63.9)
Percentages may not add up to 100% because of rounding.
E. Hunter et al. / Seizure 21 (2012) 691–698
696
Table 6 Classification of epilepsies in adult PWE in Hai district, Tanzania. Final classification
n
(%)
Focal – structural/metabolic Focal – unclassified Generalised – GGEa Generalised – JMEb Generalised – unclassified Unclassified convulsive epilepsy
678 140 4 2 1 76
(23.4) (48.1) (1.4) (0.7) (0.3) (26.1)
Total
291
(100.0)
a b
GGE – Generalised Genetic Epilepsy. JME – Juvenile Myoclonic Epilepsy.
and further investigation of risk factors in this population is warranted. The TG for epilepsy in the Hai population, based on reported use of adequate and appropriate AED therapy, was 68.4% (95% CI 63.0– 73.7). This is lower than previous estimates of TG in Tanzania of between 76% and 95.8%,30,32 although is in line with more recent systematic reviews of TG in developing countries worldwide.12,50 The lower TG in Hai could be explained by the relative proximity to a large referral hospital (KCMC) and to the nearby large towns of Arusha and Moshi, where numerous private clinics and non-governmental healthcare providers are located. Conversely, we know that there is a high degree of stigma associated with epilepsy in this population,20 and more stigmatised and marginalised individuals may not have presented to the study. It can be assumed that such individuals would be less likely to be treated, and would therefore contribute to the estimate of TG if they were known. 4.2. Limitations While a small pilot study established the sensitivity of the screening tool in a KS-speaking Tanzanian population, a fieldbased validation study with complete ascertainment in a larger sample would have facilitated a sensitivity analysis of the screening questionnaire and would also have allowed us to comment on the likely prevalence of non-convulsive epilepsies, given that these were not detected by our screening protocol. Epilepsy with non-convulsive seizures can account for up to 50% of all epileptic disorders in community-based studies, and the true prevalence of all epilepsies in the Hai population may therefore approach 6/1000.36,37 Only cases of generalised convulsive epilepsy were detected in our study, despite the aim of detecting all seizure types. Community-based studies from Zambia and Tanzania have also made use of the same screening instrument, with modifications. In Zambia, three additional questions were used to reduce the falsepositive rate due to detection of febrile and malaria-associated seizures in children under the age of seven years.35 The paediatric study from Hai excluded children under the age of six years for this reason,13 and these additional questions were not used in our study. In Tanzania, a 15-item questionnaire has been used,30 although results of this study were not available at the time of planning and initiating our population census. We anticipated difficulties with specific questions, particularly questions two and eight (see Appendix), and sought to address these through focus group work with census enumerators administering the questionnaire. The difficulties of identifying focal onset epilepsies without secondary generalisation in community-based studies in SSA have previously been acknowledged and reflected in study design.33 Based on our experience, it may be that this is a more realistic approach for large-scale surveys conducted in SSA. Nearly one third of individuals (30.5%) who responded positively to the screening questionnaire were not available for clinical assessment. While this decreases the internal validity
of our study, and must be acknowledged as a limitation, Hai is an established DSS, and this high rate of non-attendance may also reflect the high degree of stigma associated with epilepsy in this population.20 The use of capture-recapture techniques has previously been used to maximise case ascertainment of noncommunicable diseases in a clinic-based setting in South Africa,51 although has been less successful in community-based settings in Tanzania, including Hai,52 and was not pursued in our study. While we report here on the prevalence in adults only, the estimate has been age-standardised to facilitate comparison with other studies. The crude prevalence in children over the age of six years in the same population was 2.91 per 1000.13 5. Conclusions Active epilepsy with convulsive seizures has a lower prevalence in this part of Tanzania than elsewhere in SSA, and the previously reported high prevalence rates for epilepsy in SSA cannot be generalised to the whole region. The high proportion of focal onset seizures points to a large burden of acquired, and possibly preventable, epilepsy. The treatment gap of 68.4% confirms that interventions to improve the management of epilepsy in this population are warranted. With a low prevalence, however, improved outcomes are potentially achievable for the majority of PWE in this and similar populations. Ethics The ethics review committee of the Tanzanian National Institute of Medical Research formally approved the study (ref. NIMR/HQ/R.8a/Vol.IX/786; 09/02/2009). Informed consent was obtained from all cases participating in the study. Conflict of interest None of the authors have any conflicts of interest. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. Acknowledgements Neuroimaging performed during this study was funded by the Helen H Lawson Grant, administered by BMA Charities. We thank the census team and field staff in Hai, along with members of the Hai District Community Health Management Team and members of staff in the radiology and EEG departments at KCMC. The study was sponsored by Northumbria Healthcare NHS Foundation Trust, UK. Appendix A Screening questionnaire used to detect possible cases of epilepsy, adapted from Placencia et al.1 (Kiswahili translations in italics) 1. Have you ever had attacks of shaking of the arms or legs which you could not control? Je, ulishawahi kupatwa na hali ya kutetemeka mikono au miguu ambayo hukuweza kuizuia? 2. Have you ever had attacks in which you fall suddenly, without any reason, changing colour in the palms, lips or face? Je, ulishawahi kupatwa na hali ambayo ilikufanya uanguke kubadilika rangi na kuwa mweupe kwenye viganja, midomo na usoni?? 3. Have you ever lost consciousness?
E. Hunter et al. / Seizure 21 (2012) 691–698
Je, ulishawahi kupoteza fahamu? 4. Have you ever had attacks in which you fall with loss of consciousness? Je, ulishawahi kupatwa na hali ambayo ilisababisha uanguke na kupoteza fahamu? 5. Have you ever had attacks in which you fall and bite your tongue? Je, ulishawahi kupatwa na hali iliyokusababisha uanguke na kuuma/kung’ata ulimi wako? 6. Have you ever had attacks in which you fall and lose control of your bladder? Je, ulishawahi kupatwa na hali ambayo ilikufanya uanguke na kutoa mkojo? 7. Have you ever had brief attacks of shaking or trembling in one arm or leg or in the face? Je, ulishawahi kupatwa na hali ya muda mfupi ya kutetemeka au kutikisika mkono au mguu mmoja au kwenye uso? 8. Have you ever had attacks in which you lose contact with the surroundings and experience abnormal smells? Je, ulishawahi kupatwa na hali ambayo ilikufanya upoteze ufahamu wa mahali ulipo na kusikia harufu zisizo za kawaida? 9. Have you ever been told that you have or had epilepsy or epileptic fits? Je, ulishawahi kuelezwa kuwa una au ulishawahi kuwa na kifafa au kuanguka kifafa?
References 1. Placencia M, Sander JWAS, Shorvon SD, Ellison RH, Cascante SM. Validation of a screening questionnaire for the detection of epileptic seizures in epidemiological studies. Brain 1992;115:783–94. 2. WHO. Atlas: epilepsy care in the world. Geneva: WHO; 2005. 3. Ngugi AK, Bottomley C, Kleinschmidt I, Sander JW, Newton CR. Estimation of the burden of active and life-time epilepsy: a meta-analytic approach. Epilepsia 2010;51(5):883–90. 4. Diop AG, de Boer HM, Mandlhate C, Leonid P, Meinardi H. The global campaign against epilepsy in Africa. Acta Tropica 2003;87:149–59. 5. Kale R. Global campaign against epilepsy: the treatment gap. Epilepsia 2002;43(Suppl. 6):31–3. 6. Sander JWAS, Shorvon SD. Incidence and prevalence studies in epilepsy and their methodological problems: a review. Journal of Neurology Neurosurgery and Psychiatry 1987;50:829–39. 7. Sander JW. The epidemiology of epilepsy revisited. Current Opinion in Neurology 2003;16:165–70. 8. Senanayake N, Roman GC. Epidemiology of epilepsy in developing countries. Bulletin of the World Health Organization 1993;71:247–58. 9. Preux PM, Druet-Cabanac M. Epidemiology and aetiology of epilepsy in subSaharan Africa. Lancet Neurol 2005;4:21–31. 10. Scott RA, Lhatoo SD, Sander JWAS. The treatment of epilepsy in developing countries: where do we go from here? Bulletin of the World Health Organization 2001;79:344–51. 11. Meinardi H, Scott RA, Reis R, Sander JWAS. The treatment gap in epilepsy: the current situation and ways forward. Epilepsia 2001;42:136–49. 12. Mbuba CK, Ngugi AK, Newton CR, Carter JA. The epilepsy treatment gap in developing countries: a systematic review of the magnitude, causes, and intervention strategies. Epilepsia 2008;49:1491–503. 13. Burton K, Rogathe J, Whittaker R, Mankad K, Hunter E, Burton M, et al. Epilepsy in Tanzanian children: Association with perinatal events and other risk factors. Epilepsia 2012;53(4. 14. Mswia R, Whiting D, Kabadi G, Honorati M, Setel P, Hai DSS. Hai DSS, Tanzania. The policy implications of Tanzania’s mortality burden volume 1: a ten-year community-based perspective: Ministry of Health. United Republic of Tanzania; 2004. pp. 228–235. 15. Mswia RD, Whiting D, Kabadi G. Hai district demographic surveillance system. In: INDEPTH Network, editor. Population and health in developing countries, vol. 1: population, health, and survival in INDEPTH sites. Ottowa: International Development Research Centre; 2002. p. 151–8. 16. Dotchin C, Msuya O, Kissima J, Massawe J, Mhina A, Moshi A, et al. The prevalence of Parkinson’s disease in rural Tanzania. Movement Disorders 2008;23(11):1567–672. 17. Dotchin CL, Walker RW. The prevalence of essential tremor in rural northern Tanzania. Journal of Neurology Neurosurgery and Psychiatry 2008;79:1107–9. 18. Walker R, Whiting D, Unwin N, Mugusi F, Swai M, Aris E, et al. Stroke incidence in urban and rural Tanzania. Lancet Neurology 2010;9(8):786–92.
697
19. Howitt SC, Jones MP, Jusabani A, Gray WK, Aris E, Mugusi F, et al. A crosssectional study of quality of life in incident stroke survivors in rural northern Tanzania. Journal of Neurology 2011;258:1422–30. 20. Mushi D, Hunter E, Mtuya C, Mshana G, Aris E, Walker R. Social–cultural aspects of epilepsy in Kilimanjaro Region, Tanzania: knowledge and experience among patients and carers. Epilepsy & Behavior 2010;20(2):338–43. 21. Dewhurst F, Dewhurst MJ, Gray WK, Orega G, Howlett W, Chaote P, et al. The prevalence of disability in older people in Hai, Tanzania. Age Ageing 2012;41(4):517–23 [Epub 2012 Apr 19]. 22. Walker RW, McLarty DG, Masuki G, Kitange HM, Whiting D, Moshi AF, et al. Age specific prevalence of impairment and disability relating to hemiplegic stroke in the Hai District of northern Tanzania. Journal of Neurology Neurosurgery and Psychiatry 2000;68(6):744–9. 23. Reutens D, Howell R, Gebert K, Berkovic S. Validation of a Questionnaire for Clinical Seizure Diagnosis. Epilepsia 1992;33(6):1065–71. 24. Berg AT, Berkovic SF, Brodie MJ, Buchhalter J, Cross HJ, Boas W, et al. Revised terminology and concepts for organization of seizures and epilepsies: report of the ILAE Commission on Classification and Terminology, 2005–2009. Epilepsia 2010;51(4):676–85. 25. Thurman DJ, Beghi E, Begley CE, Berg AT, Buchhalter J, Ding D, et al. ILAE Commission Report: standards for epidemiolgoci studies and surveillance of epilepsy. Epilepsia 2011;52(Suppl. 7):2–26. 26. Ahmad OB, Boschi-Pinto C, Lopez AD, Murray CJL, Lozano R, Inoue M. Age standardization of rates: a new WHO standard (GPE discussion paper Series: No. 31). Geneva: World Health Organisation; 2001. 27. British National Formulary. 2012 [07/06/2012]. Available from: http:// www.medicinescomplete.com/mc/bnf/current/3577.htm. 28. Tanzanian National Formulary. Tanzanian National Formulary. 2nd ed. Dar es Salaam: Tanzanian Food and Drugs Authority; 2005. 29. Shorvon SD, Hart YM, Sander JWAS, van Andel FLRSoM. The management of epilepsy in developing countries – an ICEBERG manua. London: Royal Society of Medicine; 1991. 30. Winkler AS, Kerschbaumsteiner K, Stelzhammer B, Meindl M, Kaaya J, Schmutzhard E. Prevalence, incidence, and clinical characteristics of epilepsy – a community-based door-to-door study in northern Tanzania. Epilepsia 2009;50:2310–3. 31. Rwiza HT, Kilonzo GP, Haule J, Matuja WBP. Prevalence and incidence of epilepsy in ulanga, a rural tanzanian district: a community-based study. Epilepsia 1992;33:1051–6. 32. Dent W, Helbok R, Matuja WBP, Scheunemann S, Schmutzhard E. Prevalence of active epilepsy in a rural area in South Tanzania: a door-to-door survey. Epilepsia 2005;46:1963–9. 33. Edwards T, Scott AG, Munyoki G, Mung’ala O, Chengo E, Bauni E, et al. Active convulsive epilepsy in a rural district of Kenya: a study of prevalence and possible risk factors. Lancet Neurol 2008;7:50–6. 34. Janson A. Shed some light on darkenss: will Tanzania reach the millenium development goals? Acta Paediatrica 2007;96(6):781–6. 35. Birbeck GL, Kalichi EMN. Epilepsy prevalence in rural Zambia: a door-to-door survey. Tropical Medicine and International Health 2004;9:92–5. 36. Cockerell OC, Eckle I, Goodridge DMG, Sander JWA, Shorvon SD. Epilepsy in a population of 6000 re-examined: secular trends in first attendance rates, prevalence and prognosis. Journal of Neurology Neurosurgery and Psychiatry 1995;58:570–6. 37. Banerjeea PN, Filippi D, Hauser WA. The descriptive epidemiology of epilepsy – a review. Epilepsy Research 2009;85:31–45. 38. Hauser WA, Annegers JF, Kurland LT. Incidence of epilepsy and unprovoked seizures in Rochester, Minnesota: 1935–1984. Epilepsia 1993;34:453–68. 39. Rwiza HT. The Muhimbili epilepsy project, a three pronged approach. Tropical and Geographical Medicine 1994;46(Suppl. 3):22–4. 40. Ding D, Wang W, Wu J, Ma G, Dai X, Yang B, et al. Premature mortality in people with epilepsy in rural China: a prospective study. The Lancet Neurology 2006;5:823–7. 41. Snow RW, Williams REM, Rogers JE, Mung’ala VO, Peshu N. The prevalence of epilepsy among a rural Kenyan population. Its association with premature mortality. Tropical and Geographical Medicine 1994;46:175–9. 42. Jilek-Aall L, Rwiza HT. Prognosis of epilepsy in a rural African Community: a 30year follow-up of 164 patients in an outpatient clinic in rural Tanzania. Epilepsia 1992;33(4):645–50. 43. Diop AG, Hesdorffer DC, Logroscino G, Hauser WA. Epilepsy and mortality in Africa: a review of the literature. Epilepsia 2005;46(Suppl. 11):33–5. 44. Munyoki G, Edwards T, White S, Kwasa T, Chengo E, Kokwaro G, et al. Clinical and neurophysiologic features of active convulsive epilepsy in rural Kenya: a population-based study. Epilepsia 2010;51(12):2370–6. 45. Osuntokun BO, Adeuga AOG, Nottidge VA, Bademosi O, Olumide A, Ige O, et al. Prevalence of the epilepsies in Nigerian Africans: a community-based study. Epilepsia 1987;28:272–9. 46. Almu S, Tadesse Z, Cooper P, Hackett R. The prevalence of epilepsy in the Zay society, Ethiopia – an area of high prevalence. Seizure 2006;15: 211–3. 47. Quet F, Guerchet M, Pion S, Ngoungou E, Nicoletti A, Preux P. Meta-analysis of the association between cysticercosis and epilepsy in Africa. Epilepsia 2010;51(5):830–7. 48. Winkler A, Blocher J, Auer H, Gotwald T, Matuja W, Schmutzhard E. Anticysticercal and antitoxocaral antibodies in people with epilepsy in rural Tanzania. Transactions of the Royal Society of Tropical Medicine and Hygiene 2008;102:1032–8.
698
E. Hunter et al. / Seizure 21 (2012) 691–698
49. Winkler A, Willingham III A, Sikasunge C, Schmutzhard E. Epilepsy and nuerocysticercosis in rural Tanzania – an imaging study. Epilepsia 2009;50(5): 987–93. 50. Meyer AC, Dua T, Ma J, Shekhar S, Birbeck G. Global disparities in the epilepsy treatment gap: a systematic review. Bulletin of the World Health Organization 2010;88:260–6.
51. Gill G, Scott B, Beeching N, Wilkinson D, Ismail A. Enumeration of non-communicable disease in rural South Africa by electronic data linkage and capture– recapture techniques. Tropical Medicine and International Health 2001; 6 (6):435–41. 52. Black J, McLarty D, Mtasiwa D. Capture–recapture techniques – difficult to use in developing countries (letter). BMJ 1994;308:531.
Contents lists available at SciVerse ScienceDirect
Seizure journal homepage: www.elsevier.com/locate/yseiz
Prevalence of active epilepsy in rural Tanzania: A large community-based survey in an adult population Ewan Hunter a,d,*, Jane Rogathi b, Simukai Chigudu c, Ahmed Jusabani b, Margaret Jackson d, Richard McNally e, William Gray a, Roger G. Whittaker d, Ahmed Iqbal d, Daniel Birchall d, Eric Aris f, Richard Walker a a
Northumbria Healthcare NHS Foundation Trust, North Shields, NE29 8NH, UK Kilimanjaro Christian Medical Centre, PO Box 3010, Moshi, Tanzania c Newcastle University Medical School, Newcastle NE2 4HH, UK d Royal Victoria Infirmary, Newcastle NE1 4LP, UK e Institute of Health and Society, Newcastle University, NE2 4AX, UK f Muhimbili University College of Health Sciences, P.O. Box 10028, Dar es Salaam, Tanzania b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 29 April 2012 Received in revised form 18 July 2012 Accepted 20 July 2012
Purpose: To estimate the prevalence of active epilepsy in adults in an established demographic surveillance site in rural Tanzania. To describe the clinical characteristics of epilepsy and to estimate the treatment gap in this population. Methods: A pilot study established that a previously validated screening questionnaire was sensitive for detecting cases of epilepsy in a Kiswahili-speaking Tanzanian population. A door-to-door census of the adult population (total 103,026) used the screening questionnaire to identify possible cases of epilepsy, who were then assessed by a research doctor to establish a diagnosis of epilepsy or otherwise. The prevalence of active epilepsy in this population was estimated with age-standardisation to the WHO standard population. Seizure types and epilepsies were classified according to current recommendations of the International League Against Epilepsy. The treatment gap for epilepsy was estimated based on antiepileptic drug use as reported by cases. Results: Two hundred and ninety-one cases of active epilepsy, all with convulsive seizures, were identified. The age-standardised prevalence was 2.91/1000 adults (95% CI 2.58–3.24); the crude prevalence adjusted for non-response was 3.84/1000 adults (95% CI 3.45–4.20). Focal-onset seizures accounted for 71.5% of all cases identified. The treatment gap was 68.4% (95% CI 63.0–73.7). Conclusions: This is one of the largest community-based studies of the prevalence of epilepsy in adults conducted in sub-Saharan Africa to date. We identified a lower prevalence than has previously been described in this region. The high proportion of focal onset seizures points to a large burden of acquired, and possibly preventable, epilepsy in this population. A treatment gap of 68.4% confirms that interventions to raise awareness of the treatable nature of epilepsy are warranted in this and similar populations. ß 2012 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved.
Keywords: Classification Epidemiology Epilepsy Prevalence Sub-Saharan Africa Treatment gap
1. Introduction 1.1. Background Epilepsy accounts for up to 1% of the global burden of disease.2 An estimated 32.7 million people suffer from active epilepsy worldwide,3 with 80% or more of people with epilepsy (PWE) living
* Corresponding author at: Department of Infection and Tropical Medicine, Elliott Building, Royal Victoria Infirmary, Newcastle-upon-Tyne, NE1 4LP, United Kingdom. Tel.: +44 0 191 282 9751; fax: +44 0 191 282 0810. E-mail address: [email protected] (E. Hunter).
in low and middle income countries (LMICs).4 A further 500 million people (including families and carers of PWE) are thought to be indirectly affected by epilepsy.5 It is generally accepted that the incidence, prevalence and mortality associated with epilepsy are all higher in LMICs than in high income countries (HICs), although estimates have varied widely and findings from different studies are difficult to compare due to differing inclusion criteria, clinical definitions and the difficulties associated with establishing reliable baseline demographic data.3,6–9 While up to four-fifths of the potential market for anti-epileptic drugs (AEDs) is in LMICs, 90% or more of PWE in these countries may receive no treatment at all.5,10,11 The proportion of PWE who require treatment but who are not receiving it is known as the treatment gap (TG).12
1059-1311/$ – see front matter ß 2012 British Epilepsy Association. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.seizure.2012.07.009
692
E. Hunter et al. / Seizure 21 (2012) 691–698
1.2. Objectives We conducted a door-to-door survey in an established demographic surveillance site (DSS) in rural Tanzania. We sought to estimate the prevalence of active epilepsy, to describe the pattern of epilepsy and to quantify the TG in this population. With a total denominator population of 161,119, this is one of the largest community-based epidemiological studies of epilepsy to be conducted in sub-Saharan Africa (SSA) to date. Paediatric and adult studies were conducted in parallel. Findings from the paediatric population have been published elsewhere13; we report here on findings in the adult population, defined as individuals aged 15 years and above.
enumerators were also asked to identify any individuals within their communities who they knew, or suspected, suffered from a seizure disorder. Finally, results of a contemporaneous study utilising data from the same census to screen for neurological disability, including epilepsy, in 2232 individuals aged over 70 provided further cases.21 A collateral history was sought in all instances to aid in diagnosis and classification of seizures. To be included in the study people diagnosed with epilepsy had to be alive and resident in the district on the prevalence date. The age cut-off for adults of 15 years and above was chosen to reflect local practice in terms of whether individuals are managed by adult or paediatric medical services. 2.4. Variables
2. Methods 2.1. Study design A door-to-door census of the entire DSS population was completed between January and June 2009 and a previously validated screening questionnaire was used to detect possible cases of epilepsy.1 The questionnaire was forward and back translated between English and Kiswahili (KS) and a pilot study was conducted to assess the questionnaire’s sensitivity for cases of epilepsy in a KS-speaking Tanzanian population. Volunteers for the pilot study included patients with and without epilepsy at Muhimbili National Hospital, Dar es Salaam, and locally at Kilimanjaro Christian Medical Centre (KCMC), Moshi. During preparations for the census, all enumerators received training in the identification of likely cases of epilepsy and were also involved in focus group discussions of each item in the screening questionnaire, with further refinements to the KS translation being made to optimise its utility in the field. Details of the screening questionnaire and KS translation are given in Appendix. 2.2. Study setting The Hai district lies on the slopes of Mount Kilimanjaro in northeast Tanzania, covering an area of approximately 13,000 km2. Villages are served by a network of unpaved roads of varying quality, many of which are impassable to motorised vehicles during the twice-yearly rainy seasons. Agriculture, livestock keeping, dairy farming, commercial mining and cottage industries are the main economic activities.14 Facilities for Electroencephalography (EEG) and computed tomography (CT) are available at KCMC, a large referral hospital in the nearby town of Moshi. Hai was established as one of three DSSs in Tanzania by the Adult Morbidity and Mortality Project (AMMP), an epidemiological programme funded by the UK Department for International Development and conducted in partnership between the Tanzanian Ministry for Health and Newcastle University, UK.15 Between 1992 and 2004 the population of each DSS was enumerated by an annual census, with all households and household members being assigned unique identifying numbers. The DSS infrastructure in Hai has subsequently been maintained through a series of epidemiological studies, and there is currently at least one trained census enumerator per village.16–21 These enumerators facilitate access to their communities by visiting researchers and are coordinated by a team of supervisors who are employed in clinical posts by the district health authority.
All individuals diagnosed with epilepsy were interviewed in depth to establish seizure history, possible risk factors, treatment history and socio-demographic features. The interview format was drawn from the clinical experience of the research team and from previous field work experience in this study population.16,18,22 A semistructured element was included to facilitate subsequent review and validation of diagnoses.23 All cases were offered further investigation with CT head scan and EEG. CT scans were reported locally to exclude any acute pathology, and then reviewed by neuroradiologists in the UK (DB, AI). There are no facilities for interpreting EEGs locally and all recordings were read in the UK by a consultant neurophysiologist (RGW). Any individuals identified with epilepsy who were not being treated adequately were started on treatment and referred to services locally for follow-up. Clinical details and investigation findings for all patients were reviewed by a consultant neurologist in the UK with an interest in epilepsy (MJ). CT and EEG findings were correlated with clinical details to classify epilepsies according to currently accepted criteria.24,25 Active epilepsy was defined as at least two unprovoked seizures occurring at least 24 h apart, with the most recent seizure occurring within the 5 years prior to the prevalence date of 1st of June 2009. Individuals who did not satisfy these criteria, but who were maintained on AEDs for an appropriate indication, were also considered to have active epilepsy. Estimates for the TG were based on individuals who reported either no or inadequate AED therapy at the time of the study. All data were entered into Microsoft Excel 2007 as field work progressed and then re-entered into Microsoft Access 2007, with any discrepancies being corrected with reference to the original case notes. 2.5. Statistical methods Statistical analyses were performed using SPSS-17 for Windows (SPSS, Chicago, IL, USA). Age- and sex-specific prevalence rates for the Hai adult population were calculated in five-year age-bands. Prevalence was age-standardised to the WHO world standard population using the direct method26; 95% confidence intervals (CIs) were calculated based on the assumptions of the binomial distribution. As this study is concerned only with individuals aged 15 years and above, the Hai and WHO populations were weighted accordingly (Table 5). To look for any evidence of clustering, oneway analysis of variance (ANOVA) was used to compare normally distributed mean prevalences from administrative divisions within the district. 3. Results
2.3. Participants and data sources 3.1. Pilot study All individuals who responded positively to the screening questionnaire at census were assessed in detailed by a research doctor trained in the diagnosis of epilepsy (EH). In addition,
Forty-nine subjects (27 cases of epilepsy and 22 controls) were recruited into the pilot study. The KS translation of the screening
E. Hunter et al. / Seizure 21 (2012) 691–698
693
Table 1 Response rates to individual screening questions.
1 2
3 4 5 Fig. 1. Hai DSS adult population structure (2009 census). 6
questionnaire performed with 100% sensitivity and 54.5% specificity, with positive and negative predictive values (PPV, NPV) of 73% and 100% respectively.
7 8 9
3.2. Participants The 2009 census identified a total population of 161,119, of whom 103,026 (63.9%) were aged 15 years and over (54,386 female, 48,640 male) (Fig. 1). Five hundred and sixty-eight adult individuals (286 female, 282 male) responded positively to the screening questionnaire. Response rates to individual questions are given in Table 1. Three hundred and ninety-five (69.5%) of these were clinically assessed, with a diagnosis of active epilepsy being made in 209 (PPV 52.9%). Of the 173 individuals that were not available for assessment, 29 (16.8%) had moved or travelled from the district at the time of the study, 21 (12.1%) could not be traced, 18 (10.4%) refused, 12 (6.9%) had died since the census (four deaths prior to the prevalence date) and four (2.3%) were accounted for by data entry errors (duplicate names or wrong dates of birth). A further 89 (51.4%) did not attend for assessment on two or more occasions. In addition to census responders, a further 130 individuals (60 female, 70 male) identified by village enumerators as possible cases were assessed, with active epilepsy being diagnosed in 80 (61.5%). The concurrent study of neurological disability in the elderly identified five cases of active epilepsy, of
Question
Positive responses n
(%)
Have you ever had attacks of shaking of the arms or legs which you could not control? Have you ever had attacks in which you fall suddenly, without any reason, changing colour in the palms, lips or face? Have you ever lost consciousness? Have you ever had attacks in which you fall with loss of consciousness? Have you ever had attacks in which you fall and bite your tongue? Have you ever had attacks in which you fall and lose control of your bladder? Have you ever had brief attacks of shaking or trembling in one arm or leg or in the face? Have you ever had attacks in which you lose contact with the surroundings or experience abnormal smells? Have you ever been told that you have or have had convulsions, epilepsy or epileptic fits?
259
(45.6)
363
(63.9)
362 316
(63.7) (55.6)
213
(37.5)
177
(31.2)
200
(35.2)
152
(26.8)
173
(30.5)
which one was identified independently by the epilepsy study, with a further two being available for assessment and inclusion in the final case series. Epilepsy was suspected in a further 14 individuals, although insufficient history was obtained, either directly or from an eye witness, to confirm the diagnosis. These individuals remain under follow-up. In summary, 527 individuals were assessed clinically with 291 cases of active epilepsy being identified (Fig. 2). An eye witness was available in 202 (69.4%) of 291 confirmed cases. CT head scans were available in 200 (68.7%) of 291 confirmed cases, with cerebral abnormalities on 51 (25.5%) scans. EEGs were available from 216 (74.2%) of 291 cases, with abnormalities consistent with epilepsy seen in 84 (38.9%) recordings. CT and EEG findings are summarised in Tables 2 and 3. 3.3. Clinical features and treatment history All 291 cases identified were suffering from active epilepsy with convulsive seizures (AEWCS). Focal onset seizures were identified
Fig. 2. Summary of case ascertainment.
E. Hunter et al. / Seizure 21 (2012) 691–698
694 Table 2 Abnormalities on CT scans. CT scan finding
n
(%)
Vascular Cysticercosis Trauma Encephalomalacia Ventricular dilatation Hemispheric atrophy Hydrocephalus Lobar atrophy Tumour Generalised atrophy Lobar calcification Porencephalic cyst Total
13 8 5 5 5 4 3 3 2 1 1 1 51
(25.5) (15.7) (9.8) (9.8) (9.8) (7.8) (5.9) (5.9) (3.9) (2.0) (2.0) (2.0) (100)
in 208 (71.5%) of 291 cases, generalised onset seizures in 7 (2.4%) cases, with seizures remaining unclassified in 76 (26.1%) cases. Six (85.7%) of seven cases with generalised onset seizures had EEGs suggestive of genetic generalised epilepsy, two of which could be sub-classified as juvenile myoclonic epilepsy. Sixty-eight (32.7%) of 208 cases with focal onset seizures were classified as structural/ metabolic. Epilepsy remained unclassified in 274 (74.6%) of 291 cases: 140 (67.3%) of 208 focal cases, one (14.3%) of seven generalised cases, and in all 76 cases (100.0%) in which seizures could not be classified (Table 6). The relative contributions of diagnostic modalities to identifying focal onset seizures are illustrated in Fig. 3. Two-hundred and fifty-three (86.9%) of 291 cases had previously sought medical attention for their seizures, of whom 122 (48.2%) reported using AEDs at the time of the study. Of these, 30 (24.6%) were using doses likely to be sub-therapeutic, based on published prescribing information27,28 and guidance on managing epilepsy in resource-poor settings.29 Ninety-two (31.6%) of 291 cases were therefore considered as adequately treated with AEDs at the time of the study.
Fig. 3. Relative contributions of diagnostic modalities to identifying focal-onset seizures in Hai.
deferred in 14 individuals and two cases identified by the other study were not available for assessment by the epilepsy study team. A confirmed diagnosis in these 16 would bring the total case series to 394 (crude prevalence 3.82 per 1000; 95% CI 3.45–4.20). While the prevalence of epilepsy varied from village to village (0–6.24/1000), there were no significant differences in mean prevalence between the five administrative divisions within the district (ANOVA, p = 0.991). A conservative estimate of the TG for epilepsy in the adult Hai population, based on AED use, was 68.4% (95% CI 63.0–73.7). 4. Discussion 4.1. Key results and interpretation
3.4. Prevalence and treatment gap The crude prevalence of AEWCS in the Hai adult population on 1st June 2009 was 2.82 per 1000 adults (95% CI 2.50–3.15). Ageand sex-specific prevalence rates are given in Table 4. The crude prevalence was higher in males than in females (difference in proportions 0.69/1000; 95% CI 0.03–1.34). To accommodate small numbers in the older population, cases aged 70 years or more were combined into a single age band of 70+ prior to age-standardisation (Table 5). The age-standardised prevalence of AEWCS in the adult Hai population was 2.91 per 1000 adults (95% CI 2.58–3.24). After accounting for deaths prior to the prevalence date and data entry errors, a further 165 (95.4%) of the 173 individuals who screened positive during the census but who were not assessed clinically were potentially eligible for the study. Applying the PPV of 52.9% for detection of active epilepsy by the screening questionnaire, an additional 87 cases may have been missed, which would bring the total number of active cases to 378 (crude prevalence 3.67 per 1000; 95% CI 3.3–4.0). Diagnoses were Table 3 Abnormalities on EEG. EEG finding
n
(%)
Temporal lobe abnormalities Non-epileptiform abnormalities Extra-temporal focal abnormalities Multifocal epileptiform abnormalities Generalised epileptiform abnormalities Total
27 25 16 10 6 84
(32.1) (29.8) (19.0) (11.9) (7.1) (100)
This study is one of the largest community-based surveys of the prevalence of epilepsy in an adult population to be conducted in SSA to date. The age-standardised prevalence for AEWCS in adults of 2.91/1000 (95% CI 2.58–3.24), with crude prevalence adjusted for non-response of 3.84/1000 (95% CI 3.47–4.22), is considerably lower than both the median prevalence of 15/1000 previously reported from SSA9 and the range for active prevalence of 8–10/ 1000 previously reported from Tanzania.30–32 It is, however, in line with a recent study of comparable size and design in a rural Kenyan population which found a crude prevalence of 2.9/1000 (95% CI 2.6–3.2), rising to 4.5/1000 (95% CI 4.1–4.9) after adjustment for non-response and sensitivity.33 This latter study sought to identify active convulsive epilepsy only, and the lower prevalence figures of both studies may reflect this. In community-based studies from SSA heterogeneity of prevalence within study populations has been frequently observed.9,31,33 Whether this reflects clustered exposure to environmental or genetic risk factors or a tendency for PWE to live closer to locally available treatment services remains unclear. In our study case ascertainment was grossly uniform across administrative divisions, with no observation of clustering. We found a higher prevalence in males than in females, although the difference was marginal. A review of 19 studies of the prevalence in SSA in which data on gender were available found a mean ratio of males to females of 1.4, although the ratio was significantly different in only a quarter of studies.9 It has been suggested that a higher prevalence of epilepsy in males in Tanzania may reflect a higher risk of cranial trauma through a tendency to
E. Hunter et al. / Seizure 21 (2012) 691–698
695
Table 4 Age- and sex-specific prevalence of epilepsy in adults in Hai. Male
Female Age-band (years) 15–19 20–24 25–29 30–34 35–39 40–44 45–49 50–54 55–59 60+ Total
Cases (n)
Denominator population (n)
23 25 23 12 9 10 13 7 6 8
8114 5225 5487 5381 5216 4800 4156 3727 2911 9369
136
54,386
Prevalence per 1000 (95% CI) 2.83 4.78 4.19 2.23 1.73 2.08 3.13 1.88 2.06 0.85
Cases (n)
Total Denominator population (n)
(1.68–3.99) (2.91–6.66) (2.48–5.90) (0.97–3.49) (0.60–2.85) (0.79–3.37) (1.43–4.83) (0.49–3.27) (0.41–3.71) (0.26–1.45)
19 36 14 18 13 17 11 7 8 12
8648 4719 4467 4961 4852 4234 3630 2986 2437 7706
2.50 (2.08–2.92)
155
48,640
engage in more dangerous occupations.32 Conversely, other studies from Tanzania have found higher proportions of females with epilepsy.30,31 It has been postulated that this may be accounted for by a higher mortality rate among males, again because of dangerous occupations,31 although Tanzania is a country with a high maternal mortality rate, and the strength of this argument could be questioned.34 It has been observed that women in one study population were more health-conscious then men, being accustomed to attending maternity and child-care clinics and speaking openly about health issues, possibly contributing to higher ascertainment rates in women.30 Overall there remains no clear consensus on whether observed gender differences in the prevalence of epilepsy in SSA are due to genderspecific aetiological factors, bias arising through artefact in caseascertainment or competing mortality risks.35 Studies from Europe and the USA have identified peaks in the incidence of epilepsy in young children and in the elderly.36–38 We did not find a peak in prevalence in those aged 60 and older, although this age group represents only 10.6% of the total Hai population, and it is difficult to draw comparisons with populations from HICs which have very different population structures. We did find a peak in prevalence in those aged 20–24 years (6.13/1000), with a concentration of male cases in this age group (crude prevalence in males 7.63/1000). This is in keeping with a higher prevalence in adolescents and young adults identified by other studies in rural SSA.9,30–33 Whether the lower prevalence in later adult life points to early mortality or spontaneous remission remains unclear.33 A similar distribution in both the active and lifetime prevalence of epilepsy has been observed in Tanzania,
Prevalence per 1000 (95% CI) 2.20 7.63 3.13 3.63 2.68 4.02 3.03 2.34 3.28 1.56
(1.21–3.18) (5.15–10.11) (1.49–4.77) (1.96–5.30) (1.22–4.13) (2.11–5.92) (1.24–4.82) (0.61–4.08) (1.01–5.55) (0.68–2.44)
3.19 (2.69–3.69)
Cases (n)
Denominator population (n)
42 61 37 30 22 27 24 14 14 20
16,762 9944 9954 10,342 10,068 9034 7786 6713 5348 17,075
291
103,026
Prevalence per 1000 (95% CI) 2.51 6.13 3.72 2.90 2.19 2.99 3.02 2.09 2.62 1.17
(1.75–3.26) (4.60–7.67) (2.52–4.91) (1.86–3.94) (1.27–3.10) (1.86–4.11) (1.85–4.31) (0.99–3.18) (1.25–3.99) (0.66–1.68)
2.82 (2.50–3.99)
suggesting that those in remission may not report their previous seizures.31,39 Conversely, there is mounting evidence of early mortality in epilepsy in LMICs, in excess of that seen in HICs.40 Observational data suggest that this is likely to be the case in SSA,41–43 although there are very few formalised data with comparable standardised mortality ratios.43 All PWE identified in our study had AEWCS, with focal-onset seizures accounting for 71.5% of all cases. High rates of focal onset seizures were found in 71% of all cases identified in a recent community-based study of people with AEWCS in Kenya,44 and in 55% of all cases in a community-based study from Nigeria, although it is not clear how focal seizures with generalised clonic movements were classified in this latter study.45 Conversely, higher rates of generalised-onset seizures have previously been reported from community-based studies in SSA, accounting for between 58% and 82% of diagnoses.31,32,46 Differing approaches to case-ascertainment, differing diagnostic and classification criteria, and lack of EEG and CT to aid with classification make a strict comparison between studies difficult, although all authors comment that focal seizures becoming rapidly generalised may have erroneously been classified as generalised onset seizures. The high rate of focal onset seizures identified in our study suggests that a high proportion of epilepsy in this population may be due to identifiable, and possibly preventable, causes. We know that epilepsy in children from Hai is associated with adverse perinatal events13, while exposure to Taenia solium, the parasitic infection responsible for neurocysticercosis (NCC), is associated with epilepsy in a number of African populations,47 including Tanzania.48,49 The role that this latter aetiology might play in an area where the prevalence of epilepsy is low, such as Hai, is not clear,
Table 5 Age distributions of Hai and WHO standard populations, weighted for adults (15 years). Age band (years)
Hai population (% of total)
Hai adult weighted (%)
WHO (%)
WHO adult weighted (%)
0–14 15–19 20–24 25–29 30–34 35–39 40–44 45–49 50–54 55–59 60–64 65–69 70+ All-ages Adults
58,093 16,762 9944 9954 10,342 10,068 9034 7786 6713 5348 4339 3867 8869 161,119 103,026
– 16.3 9.7 9.7 10.0 9.8 8.8 7.6 6.5 5.2 4.2 3.8 8.6 – 100.2a
26.2 8.5 8.2 7.9 7.6 7.2 6.6 6.0 5.4 4.6 3.7 3.0 5.3 100.2a –
– 11.5 11.1 10.7 10.3 9.7 8.9 8.2 7.3 6.2 5.0 4.0 7.2 – 100.1a
a
(36.1) (10.4) (6.2) (6.2) (6.4) (6.3) (5.6) (4.8) (4.2) (3.3) (2.7) (2.4) (5.5) (100.1)a (63.9)
Percentages may not add up to 100% because of rounding.
E. Hunter et al. / Seizure 21 (2012) 691–698
696
Table 6 Classification of epilepsies in adult PWE in Hai district, Tanzania. Final classification
n
(%)
Focal – structural/metabolic Focal – unclassified Generalised – GGEa Generalised – JMEb Generalised – unclassified Unclassified convulsive epilepsy
678 140 4 2 1 76
(23.4) (48.1) (1.4) (0.7) (0.3) (26.1)
Total
291
(100.0)
a b
GGE – Generalised Genetic Epilepsy. JME – Juvenile Myoclonic Epilepsy.
and further investigation of risk factors in this population is warranted. The TG for epilepsy in the Hai population, based on reported use of adequate and appropriate AED therapy, was 68.4% (95% CI 63.0– 73.7). This is lower than previous estimates of TG in Tanzania of between 76% and 95.8%,30,32 although is in line with more recent systematic reviews of TG in developing countries worldwide.12,50 The lower TG in Hai could be explained by the relative proximity to a large referral hospital (KCMC) and to the nearby large towns of Arusha and Moshi, where numerous private clinics and non-governmental healthcare providers are located. Conversely, we know that there is a high degree of stigma associated with epilepsy in this population,20 and more stigmatised and marginalised individuals may not have presented to the study. It can be assumed that such individuals would be less likely to be treated, and would therefore contribute to the estimate of TG if they were known. 4.2. Limitations While a small pilot study established the sensitivity of the screening tool in a KS-speaking Tanzanian population, a fieldbased validation study with complete ascertainment in a larger sample would have facilitated a sensitivity analysis of the screening questionnaire and would also have allowed us to comment on the likely prevalence of non-convulsive epilepsies, given that these were not detected by our screening protocol. Epilepsy with non-convulsive seizures can account for up to 50% of all epileptic disorders in community-based studies, and the true prevalence of all epilepsies in the Hai population may therefore approach 6/1000.36,37 Only cases of generalised convulsive epilepsy were detected in our study, despite the aim of detecting all seizure types. Community-based studies from Zambia and Tanzania have also made use of the same screening instrument, with modifications. In Zambia, three additional questions were used to reduce the falsepositive rate due to detection of febrile and malaria-associated seizures in children under the age of seven years.35 The paediatric study from Hai excluded children under the age of six years for this reason,13 and these additional questions were not used in our study. In Tanzania, a 15-item questionnaire has been used,30 although results of this study were not available at the time of planning and initiating our population census. We anticipated difficulties with specific questions, particularly questions two and eight (see Appendix), and sought to address these through focus group work with census enumerators administering the questionnaire. The difficulties of identifying focal onset epilepsies without secondary generalisation in community-based studies in SSA have previously been acknowledged and reflected in study design.33 Based on our experience, it may be that this is a more realistic approach for large-scale surveys conducted in SSA. Nearly one third of individuals (30.5%) who responded positively to the screening questionnaire were not available for clinical assessment. While this decreases the internal validity
of our study, and must be acknowledged as a limitation, Hai is an established DSS, and this high rate of non-attendance may also reflect the high degree of stigma associated with epilepsy in this population.20 The use of capture-recapture techniques has previously been used to maximise case ascertainment of noncommunicable diseases in a clinic-based setting in South Africa,51 although has been less successful in community-based settings in Tanzania, including Hai,52 and was not pursued in our study. While we report here on the prevalence in adults only, the estimate has been age-standardised to facilitate comparison with other studies. The crude prevalence in children over the age of six years in the same population was 2.91 per 1000.13 5. Conclusions Active epilepsy with convulsive seizures has a lower prevalence in this part of Tanzania than elsewhere in SSA, and the previously reported high prevalence rates for epilepsy in SSA cannot be generalised to the whole region. The high proportion of focal onset seizures points to a large burden of acquired, and possibly preventable, epilepsy. The treatment gap of 68.4% confirms that interventions to improve the management of epilepsy in this population are warranted. With a low prevalence, however, improved outcomes are potentially achievable for the majority of PWE in this and similar populations. Ethics The ethics review committee of the Tanzanian National Institute of Medical Research formally approved the study (ref. NIMR/HQ/R.8a/Vol.IX/786; 09/02/2009). Informed consent was obtained from all cases participating in the study. Conflict of interest None of the authors have any conflicts of interest. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. Acknowledgements Neuroimaging performed during this study was funded by the Helen H Lawson Grant, administered by BMA Charities. We thank the census team and field staff in Hai, along with members of the Hai District Community Health Management Team and members of staff in the radiology and EEG departments at KCMC. The study was sponsored by Northumbria Healthcare NHS Foundation Trust, UK. Appendix A Screening questionnaire used to detect possible cases of epilepsy, adapted from Placencia et al.1 (Kiswahili translations in italics) 1. Have you ever had attacks of shaking of the arms or legs which you could not control? Je, ulishawahi kupatwa na hali ya kutetemeka mikono au miguu ambayo hukuweza kuizuia? 2. Have you ever had attacks in which you fall suddenly, without any reason, changing colour in the palms, lips or face? Je, ulishawahi kupatwa na hali ambayo ilikufanya uanguke kubadilika rangi na kuwa mweupe kwenye viganja, midomo na usoni?? 3. Have you ever lost consciousness?
E. Hunter et al. / Seizure 21 (2012) 691–698
Je, ulishawahi kupoteza fahamu? 4. Have you ever had attacks in which you fall with loss of consciousness? Je, ulishawahi kupatwa na hali ambayo ilisababisha uanguke na kupoteza fahamu? 5. Have you ever had attacks in which you fall and bite your tongue? Je, ulishawahi kupatwa na hali iliyokusababisha uanguke na kuuma/kung’ata ulimi wako? 6. Have you ever had attacks in which you fall and lose control of your bladder? Je, ulishawahi kupatwa na hali ambayo ilikufanya uanguke na kutoa mkojo? 7. Have you ever had brief attacks of shaking or trembling in one arm or leg or in the face? Je, ulishawahi kupatwa na hali ya muda mfupi ya kutetemeka au kutikisika mkono au mguu mmoja au kwenye uso? 8. Have you ever had attacks in which you lose contact with the surroundings and experience abnormal smells? Je, ulishawahi kupatwa na hali ambayo ilikufanya upoteze ufahamu wa mahali ulipo na kusikia harufu zisizo za kawaida? 9. Have you ever been told that you have or had epilepsy or epileptic fits? Je, ulishawahi kuelezwa kuwa una au ulishawahi kuwa na kifafa au kuanguka kifafa?
References 1. Placencia M, Sander JWAS, Shorvon SD, Ellison RH, Cascante SM. Validation of a screening questionnaire for the detection of epileptic seizures in epidemiological studies. Brain 1992;115:783–94. 2. WHO. Atlas: epilepsy care in the world. Geneva: WHO; 2005. 3. Ngugi AK, Bottomley C, Kleinschmidt I, Sander JW, Newton CR. Estimation of the burden of active and life-time epilepsy: a meta-analytic approach. Epilepsia 2010;51(5):883–90. 4. Diop AG, de Boer HM, Mandlhate C, Leonid P, Meinardi H. The global campaign against epilepsy in Africa. Acta Tropica 2003;87:149–59. 5. Kale R. Global campaign against epilepsy: the treatment gap. Epilepsia 2002;43(Suppl. 6):31–3. 6. Sander JWAS, Shorvon SD. Incidence and prevalence studies in epilepsy and their methodological problems: a review. Journal of Neurology Neurosurgery and Psychiatry 1987;50:829–39. 7. Sander JW. The epidemiology of epilepsy revisited. Current Opinion in Neurology 2003;16:165–70. 8. Senanayake N, Roman GC. Epidemiology of epilepsy in developing countries. Bulletin of the World Health Organization 1993;71:247–58. 9. Preux PM, Druet-Cabanac M. Epidemiology and aetiology of epilepsy in subSaharan Africa. Lancet Neurol 2005;4:21–31. 10. Scott RA, Lhatoo SD, Sander JWAS. The treatment of epilepsy in developing countries: where do we go from here? Bulletin of the World Health Organization 2001;79:344–51. 11. Meinardi H, Scott RA, Reis R, Sander JWAS. The treatment gap in epilepsy: the current situation and ways forward. Epilepsia 2001;42:136–49. 12. Mbuba CK, Ngugi AK, Newton CR, Carter JA. The epilepsy treatment gap in developing countries: a systematic review of the magnitude, causes, and intervention strategies. Epilepsia 2008;49:1491–503. 13. Burton K, Rogathe J, Whittaker R, Mankad K, Hunter E, Burton M, et al. Epilepsy in Tanzanian children: Association with perinatal events and other risk factors. Epilepsia 2012;53(4. 14. Mswia R, Whiting D, Kabadi G, Honorati M, Setel P, Hai DSS. Hai DSS, Tanzania. The policy implications of Tanzania’s mortality burden volume 1: a ten-year community-based perspective: Ministry of Health. United Republic of Tanzania; 2004. pp. 228–235. 15. Mswia RD, Whiting D, Kabadi G. Hai district demographic surveillance system. In: INDEPTH Network, editor. Population and health in developing countries, vol. 1: population, health, and survival in INDEPTH sites. Ottowa: International Development Research Centre; 2002. p. 151–8. 16. Dotchin C, Msuya O, Kissima J, Massawe J, Mhina A, Moshi A, et al. The prevalence of Parkinson’s disease in rural Tanzania. Movement Disorders 2008;23(11):1567–672. 17. Dotchin CL, Walker RW. The prevalence of essential tremor in rural northern Tanzania. Journal of Neurology Neurosurgery and Psychiatry 2008;79:1107–9. 18. Walker R, Whiting D, Unwin N, Mugusi F, Swai M, Aris E, et al. Stroke incidence in urban and rural Tanzania. Lancet Neurology 2010;9(8):786–92.
697
19. Howitt SC, Jones MP, Jusabani A, Gray WK, Aris E, Mugusi F, et al. A crosssectional study of quality of life in incident stroke survivors in rural northern Tanzania. Journal of Neurology 2011;258:1422–30. 20. Mushi D, Hunter E, Mtuya C, Mshana G, Aris E, Walker R. Social–cultural aspects of epilepsy in Kilimanjaro Region, Tanzania: knowledge and experience among patients and carers. Epilepsy & Behavior 2010;20(2):338–43. 21. Dewhurst F, Dewhurst MJ, Gray WK, Orega G, Howlett W, Chaote P, et al. The prevalence of disability in older people in Hai, Tanzania. Age Ageing 2012;41(4):517–23 [Epub 2012 Apr 19]. 22. Walker RW, McLarty DG, Masuki G, Kitange HM, Whiting D, Moshi AF, et al. Age specific prevalence of impairment and disability relating to hemiplegic stroke in the Hai District of northern Tanzania. Journal of Neurology Neurosurgery and Psychiatry 2000;68(6):744–9. 23. Reutens D, Howell R, Gebert K, Berkovic S. Validation of a Questionnaire for Clinical Seizure Diagnosis. Epilepsia 1992;33(6):1065–71. 24. Berg AT, Berkovic SF, Brodie MJ, Buchhalter J, Cross HJ, Boas W, et al. Revised terminology and concepts for organization of seizures and epilepsies: report of the ILAE Commission on Classification and Terminology, 2005–2009. Epilepsia 2010;51(4):676–85. 25. Thurman DJ, Beghi E, Begley CE, Berg AT, Buchhalter J, Ding D, et al. ILAE Commission Report: standards for epidemiolgoci studies and surveillance of epilepsy. Epilepsia 2011;52(Suppl. 7):2–26. 26. Ahmad OB, Boschi-Pinto C, Lopez AD, Murray CJL, Lozano R, Inoue M. Age standardization of rates: a new WHO standard (GPE discussion paper Series: No. 31). Geneva: World Health Organisation; 2001. 27. British National Formulary. 2012 [07/06/2012]. Available from: http:// www.medicinescomplete.com/mc/bnf/current/3577.htm. 28. Tanzanian National Formulary. Tanzanian National Formulary. 2nd ed. Dar es Salaam: Tanzanian Food and Drugs Authority; 2005. 29. Shorvon SD, Hart YM, Sander JWAS, van Andel FLRSoM. The management of epilepsy in developing countries – an ICEBERG manua. London: Royal Society of Medicine; 1991. 30. Winkler AS, Kerschbaumsteiner K, Stelzhammer B, Meindl M, Kaaya J, Schmutzhard E. Prevalence, incidence, and clinical characteristics of epilepsy – a community-based door-to-door study in northern Tanzania. Epilepsia 2009;50:2310–3. 31. Rwiza HT, Kilonzo GP, Haule J, Matuja WBP. Prevalence and incidence of epilepsy in ulanga, a rural tanzanian district: a community-based study. Epilepsia 1992;33:1051–6. 32. Dent W, Helbok R, Matuja WBP, Scheunemann S, Schmutzhard E. Prevalence of active epilepsy in a rural area in South Tanzania: a door-to-door survey. Epilepsia 2005;46:1963–9. 33. Edwards T, Scott AG, Munyoki G, Mung’ala O, Chengo E, Bauni E, et al. Active convulsive epilepsy in a rural district of Kenya: a study of prevalence and possible risk factors. Lancet Neurol 2008;7:50–6. 34. Janson A. Shed some light on darkenss: will Tanzania reach the millenium development goals? Acta Paediatrica 2007;96(6):781–6. 35. Birbeck GL, Kalichi EMN. Epilepsy prevalence in rural Zambia: a door-to-door survey. Tropical Medicine and International Health 2004;9:92–5. 36. Cockerell OC, Eckle I, Goodridge DMG, Sander JWA, Shorvon SD. Epilepsy in a population of 6000 re-examined: secular trends in first attendance rates, prevalence and prognosis. Journal of Neurology Neurosurgery and Psychiatry 1995;58:570–6. 37. Banerjeea PN, Filippi D, Hauser WA. The descriptive epidemiology of epilepsy – a review. Epilepsy Research 2009;85:31–45. 38. Hauser WA, Annegers JF, Kurland LT. Incidence of epilepsy and unprovoked seizures in Rochester, Minnesota: 1935–1984. Epilepsia 1993;34:453–68. 39. Rwiza HT. The Muhimbili epilepsy project, a three pronged approach. Tropical and Geographical Medicine 1994;46(Suppl. 3):22–4. 40. Ding D, Wang W, Wu J, Ma G, Dai X, Yang B, et al. Premature mortality in people with epilepsy in rural China: a prospective study. The Lancet Neurology 2006;5:823–7. 41. Snow RW, Williams REM, Rogers JE, Mung’ala VO, Peshu N. The prevalence of epilepsy among a rural Kenyan population. Its association with premature mortality. Tropical and Geographical Medicine 1994;46:175–9. 42. Jilek-Aall L, Rwiza HT. Prognosis of epilepsy in a rural African Community: a 30year follow-up of 164 patients in an outpatient clinic in rural Tanzania. Epilepsia 1992;33(4):645–50. 43. Diop AG, Hesdorffer DC, Logroscino G, Hauser WA. Epilepsy and mortality in Africa: a review of the literature. Epilepsia 2005;46(Suppl. 11):33–5. 44. Munyoki G, Edwards T, White S, Kwasa T, Chengo E, Kokwaro G, et al. Clinical and neurophysiologic features of active convulsive epilepsy in rural Kenya: a population-based study. Epilepsia 2010;51(12):2370–6. 45. Osuntokun BO, Adeuga AOG, Nottidge VA, Bademosi O, Olumide A, Ige O, et al. Prevalence of the epilepsies in Nigerian Africans: a community-based study. Epilepsia 1987;28:272–9. 46. Almu S, Tadesse Z, Cooper P, Hackett R. The prevalence of epilepsy in the Zay society, Ethiopia – an area of high prevalence. Seizure 2006;15: 211–3. 47. Quet F, Guerchet M, Pion S, Ngoungou E, Nicoletti A, Preux P. Meta-analysis of the association between cysticercosis and epilepsy in Africa. Epilepsia 2010;51(5):830–7. 48. Winkler A, Blocher J, Auer H, Gotwald T, Matuja W, Schmutzhard E. Anticysticercal and antitoxocaral antibodies in people with epilepsy in rural Tanzania. Transactions of the Royal Society of Tropical Medicine and Hygiene 2008;102:1032–8.
698
E. Hunter et al. / Seizure 21 (2012) 691–698
49. Winkler A, Willingham III A, Sikasunge C, Schmutzhard E. Epilepsy and nuerocysticercosis in rural Tanzania – an imaging study. Epilepsia 2009;50(5): 987–93. 50. Meyer AC, Dua T, Ma J, Shekhar S, Birbeck G. Global disparities in the epilepsy treatment gap: a systematic review. Bulletin of the World Health Organization 2010;88:260–6.
51. Gill G, Scott B, Beeching N, Wilkinson D, Ismail A. Enumeration of non-communicable disease in rural South Africa by electronic data linkage and capture– recapture techniques. Tropical Medicine and International Health 2001; 6 (6):435–41. 52. Black J, McLarty D, Mtasiwa D. Capture–recapture techniques – difficult to use in developing countries (letter). BMJ 1994;308:531.