IJE Advance Access originally published online on February 21, 2006
International Journal of Epidemiology 2006 35(3):683-688; doi:10.1093/ije/dyl023
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Child Health |
Prevalence and risk factors of neurological disability and impairment in children living in rural Kenya
1 Center for Geographic Medicine-Coast, Kenya Medical Research Institute, PO 428, Kilifi, Kenya
2 Department of Psychology, University of Lancaster, Lancaster, UK
3 Neurosciences Unit, Institute of Child Health, University College, London, UK
* Corresponding author. E-mail: vodera{at}kilifi.mimcom.net
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Background There is little data on the burden of neurological impairment (NI) in developing countries, particularly in children of Africa.
Methods We conducted a survey of NI in children aged 6–9 years in a rural district of Kenya. First, we screened for neurological disability by administering the Ten Questions Questionnaire (TQQ) to parents/guardians of children in a defined population. In phase two, we performed a comprehensive clinical and psychological assessment on children who tested positive on TQQ and on a similar number of children who tested negative.
Results A total of 10 218 children were screened, of whom 955 (9.3%) were positive on TQQ. Of these, 810 (84.8%) were assessed, and of those who tested negative 766 (8.3%) were assessed. The prevalence for moderate/severe NI was 61/1000 [95% confidence interval (95% CI) 48–74]. The most common domains affected were epilepsy (41/1000), cognition (31/1000), and hearing (14/1000). Motor (5/1000) and vision (2/1000) impairments were less common. Of the neurologically impaired children (n = 251), 56 (22%) had more than one impairment. Neonatal insults were found to have a significant association with moderate/severe NI in both the univariate [odds ratio (OR) = 1.70; 95% CI 1.12–2.47] and multivariate analyses (OR = 1.30; 95% CI 1.09–1.65).
Conclusions There is a considerable burden of moderate/severe NI in this area of rural Kenya, with epilepsy, cognition, and hearing being the most common domains affected. Neonatal insults were identified as an important risk factor.
Keywords Prevalence, risk factors, neurological impairment, children
Accepted 30 January 2006
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Neurological impairment (NI) is an important cause of disability and death. It is estimated to account for >28% of years lived with disability and to be responsible for at least one in every nine deaths.1 It has been suggested that 85% of children with disability live in resource poor countries (RPCs),2 but there is little data to support this.
Studies estimating the burden of NI in developing countries have been hampered by lack of relevant medical records and registries. In places where these sources of data are available, they are often of insufficient quality to estimate the burden accurately. The Ten Questions Questionnaire (TQQ)3 has been developed to screen for neurological disability in children in resource poor settings. The tool consists of 10 questions in a yes/no format and has been validated in studies in Bangladesh,4 Jamaica,5 Pakistan,6 and, more recently, in Kenya.7
We measured the prevalence and investigated the risk factors for NI in older children living in a rural area in Africa, and compared our data with data from other studies in Africa and other continents.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Study setting
This study was conducted in a demarcated area in Kilifi District, one of the poorest areas in Kenya,8 that is mainly inhabited by subsistence farmers. The formal health care system comprises a government hospital, five government dispensaries and 15 private clinics. The neonatal and under-five mortality in this area are 41 and 141 per 1000,9 respectively. The study area was sub-divided into 87 enumeration zones, with sketch maps of each zone drawn in 1992 and recently updated, showing major landmarks, footpaths, and the relative positions and survey numbers of each homestead.7 The sketch maps were used to locate each household during a census in October 2000, and the survey was conducted by us from June 2001 to March 2002.
Population
The study population consisted of children aged 6–9 years identified from a census of 108 896 people. The population consists mainly of the Mijikenda ethnic group, in which the Giriama sub-group predominates. Children aged 6–9 years were selected because of difficulties in identifying NI in children younger than 6 years (particularly assessing hearing and vision) and the lack of culturally appropriate assessment tools for cognition in young children from this area. Furthermore, since one of the aims of this study was to identify acquired NI, whose most likely causes are infections (bacterial meningitis and cerebral malaria) and which commonly occur post-natally (neonatal insults and those after the neonatal period) until age 5, it was appropriate to study children in this age group. Only children who had been residing in the area for at least 6 months prior to the survey were included.
In this study, we defined NI as any disorder resulting from damage to the nervous system or the brain, and neurological disability as the inability (due to Neurological Impairment) to perform an activity in a manner or range considered normal for an individual. The TQQ is administered to the parents or guardians of children and detects disability as perceived by these adults. The assessments measure impairment in the following domains: cognition, motor, hearing, vision, and epilepsy.
Study design
This study consisted of two stages. In Stage 1, the TQQ was administered to the parents or guardians of children aged 6–9 years. In Stage 2, clinicians and psychological assessors assessed children with at least one positive response on TQQ. A similar number of children from those who tested negative were also invited for assessment (we chose one in every 12 children as this was the prevalence of disability in our pilot survey and was similar to that observed in other studies of NI4–6 and would be expected to give us broadly similar numbers of affected and unaffected children. All these children underwent comprehensive clinical and psychological assessments at the research centre, to detect epilepsy and impairments of vision, hearing, motor function, and cognition. The assessments were performed within 1 week of the screening. None of the clinicians or assessors involved in the study was aware of the result of the TQQ screen at the time of the examination.
TQQ screening instrument
The TQQ was developed for identifying children with disability in RPCs and utilizes non-professionals in its execution. It is a short questionnaire in a yes/no format and consists of 10 questions, one question each addressing the child's vision, hearing, motor skills, and seizures, six concerning cognitive development, and one extra question regarding other serious health problems.3 It is designed to detect moderate to severe disability and impairment but not the mild ones. Previous studies have shown it to be useful in the detection of severe and moderate cognitive, motor and seizure impairment, and poor in the detection of vision and hearing, and mild impairment.4–7 The questionnaire was translated into Kigiriama and back to English to ensure the intended meaning remained the same, before being piloted.
Household screening
Five trained field interviewers fluent in Kigiriama performed the household screening after undergoing a week's training in field methods and administration of the questionnaire used in the survey.7 The field interviewers administered the TQQ to the parents or guardians of 80–100 children daily.
Psychological and neurological assessment
A team of three clinicians and five psychological assessors performed the assessments. The assessments included vision (Sonksen-Silver Acuity system),10 hearing (Kamplex screening audiometer),11 and motor impairments (clinical examination). The diagnosis of epilepsy was based upon the history. Local adaptations of assessments of cognitive abilities, speech, and language were used. The cognitive assessment involved a seven-item battery testing verbal and non-verbal skills, which included: ‘panga mutu’ (‘build a man’, testing intellectual and developmental maturity of a child through observing how a child deploys a set of basic skills to represent their knowledge of the human form)12; a local adaptation of the matching familiar figures (assessing information processing speed and impulsivity)13; digit span (auditory short-term memory)14; a construction task using wooden sticks (simultaneous processing, visual-motor co-ordination, visual-spatial perception and reasoning)15; category fluency (retrieval from long-term memory)14; information questions15 and a picture vocabulary test (receptive vocabulary, verbal comprehension, achievement and association of pictures and words).15 The speech and language assessment battery included measures of comprehension, expression, and the child's phonological system.16 The neurological assessment included a detailed medical history, treatment, and family history. A full clinical and neurological examination was performed. For the definition of impaired cognition, Z-scores for each of the tasks in the battery were calculated initially for each age band (6 through 9 years of age) on a sample representative of the overall population in their screening status (positive or negative). The means and standard deviations were then calculated from this sample. Z-scores were then calculated for each age band on all children who had been tested and these scores were used to determine impairment and level of severity (Table 1). The Z-scores used are based on the age-banded scores of at least 360 children in each age band. For other domains, the definition of severity was based on the criteria developed by the World Health Organization (Table 1).17
|
Details of previous admissions
The children's detailed medical history, which included information on whether the child had been admitted to Kilifi District Hospital (KDH), was used to track hospital records to determine the causes of admission for these children.
Potential risk factors for NI
Information was collected on numerous potential risk factors for NI by parent report using structured questionnaires during the medical assessments. For this analysis, 25 variables were selected as potential risk factors for NI. We have categorized these into socio-demographic factors and factors operating primarily during one of the three stages of fetal and child development: prenatal, perinatal, and post-natal. After the initial univariate analysis, 13 variables were identified as potential risk factors for the development of NI. For the purposes of this study, neonatal insult was defined by a positive history of birth asphyxia, tetanus, jaundice, or sepsis or any other potential NI causing infection during the neonatal period. The information was based on hospital and maternal/guardian report.
Data storage and analysis
All phase 1 and 2 data were double-entered and stored using Fox-pro software after verification of the data had been performed. Statistical analysis was performed with Stata version 8 (Stata Corporation, TX, USA). For the purpose of this study, NI was categorized depending on the domain affected: cognition, motor, epilepsy, hearing, and vision. Adjusted estimates of prevalence and its variance were computed using formulae that took account of the two-phase design of the study.18
Logistic regression analysis was used to investigate associations between potential risk factors and moderate/severe NI. Variables were entered into logistic regression models if the P-values associated with their regression coefficients were <0.25; they were retained if their respective P-values were <0.10 and/or their removal substantially affected the magnitude of the regression coefficients for other variables in the model. In the multivariate analysis, we used factors that fulfilled the statistical criteria P < 0.10 and controlled them for child's age and schooling, since these have major effects on cognitive performance.
The National Ethics Committee of Kenya Medical Research Institute approved this study.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
A total of 10 218 out of 11 416 children aged 6–9 years resident in the study area were screened; of whom 50.5% were males. For the screened children, 29% were aged 6 years, 25% were aged 7 years, 23% were aged 8 years, and 23% were aged 9 years. Out of those screened, 955 (9.3%) were positive on TQQ (delayed milestones and hearing difficulties being the most common types of problems reported).7 Of those identified as positives, 810 (84.8%) were assessed in the second phase of the study. Assessment was not performed on 32 who did not consent; 36 who were under or over age on review of their date of birth; 38 who had moved out of the study area; and 39 who could not converse in the language of the cognitive tests. Eight hundred and sixty-one children who were negative on the TQQ were also invited for assessment, 89% (766) of whom were assessed. Assessment was not performed on 95 children who failed to attend after being invited.
In the sample of children assessed, 1324 (84%) of them were born at home. Seventy-eight per cent of these were births assisted by untrained birth attendants.
Prevalence
Two hundred and fifty-one children out of the total assessed were found to have NI with an adjusted prevalence of moderate/severe NI of 61/1000 [95% confidence interval (95% CI) 48–74] (Table 2). There were no significant differences in prevalence between boys and girls (
2 = 0.5; P = 0.5) or within the age groups (2 = 1.5; P = 0.2). In addition, no significant geographical variations in prevalence were found within the study area.
|
The prevalence for specific impairments varied widely (Table 2), with epilepsy being the most common followed by cognitive and hearing impairments. Fifty-six children (22% of all children with impairment) were found to have one or more NIs. Seventeen of them (30.4%) had epilepsy with cognitive impairment, seven (13%) had cognitive with motor impairment, and six (10.8%) had hearing impairment with epilepsy. Forty-four (17.5%) of the children with NI had speech and language impairments.
Risk factors
In both the univariate [odds ratio (OR) 1.70; 95% CI 1.12–2.47] (Table 3) and multivariate NI (OR 1.30; 95% CI 1.09–1.65), neonatal insults were found to have a significant association with moderate/severe NI.
|
Causes of admission
Of the children with NI (n = 251), 130 (52%) were reported to have been admitted to hospital. We were able to confirm 73 admissions to KDH from the hospital database. Central nervous system infections (n = 6), neonatal sepsis (n = 1), neonatal jaundice (n = 1), and birth trauma (n = 1) were identified as potential causes of NI in these children. The other causes of admission included were non-severe malaria (n = 30), respiratory tract infections (n = 19), febrile convulsions (n = 4), and miscellaneous causes (n = 11).
In contrast, 280 children who tested negative on TQQ were reported to have been admitted to KDH. We were able to confirm 113 of them from the hospital database. Central nervous system infections (n = 11), neonatal sepsis (n = 3), neonatal jaundice (n = 1), and birth trauma (n = 1) were identified as potential causes of NI in these children. Other causes identified were non-severe malaria (n = 30), respiratory tract infections (n = 29), anaemia (n = 10), gastroenteritis (n = 10), febrile convulsions (n = 5), and miscellaneous causes (n = 13).
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This study identifies a significant burden of NI in a rural area of Africa, with 6% of older children having moderate or severe NI. Epilepsy, cognitive, and hearing are the domains most affected, and 22% of the children had multiple impairments. There were no significant differences in prevalence between boys and girls.
The prevalence of NI in this study is comparable with the estimate from South Africa19 but was considerably higher than those reported in Bangladesh, Pakistan, Jamaica, Saudi Arabia, and Ghana4–6,20,21 (Table 4). The differences may be explained by the use of different assessment tools or real differences in the prevalence. Since, the studies in Ghana,21 South Africa,19 and Saudi Arabia20 did not assess a sample of children testing negative on TQQ, the sensitivities and specificities of the TQQ used in these studies are unknown. Furthermore as this study assessed children aged 6–9 years, it may not provide an estimate of the full burden of disability, since many children with severe impairments will have died before they reach this age. This may explain the low prevalence of cerebral palsy in this study.
|
The prevalence of cognitive impairment was similar to that in the South African study, but considerably higher than those reported in the other studies outlined in Table 4. These differences could be related to differences in definition and/or tools used in assessing cognition. It is important to note that for any broad psychological assessment, psychologists should be used to administer standardized tests when assessing cognition; however, absence of standardized tools and inadequate trained psychologists in our situation, forced us instead to use trained assessors and adapted tools that were culturally modified.
The high prevalence of epilepsy in this area may be attributed to the high prevalence of CNS infections, in particular severe falciparum malaria and acute bacterial meningitis in this area. It is well established that CNS infections, including severe falciparum malaria, can lead to epilepsy.22,23 The low prevalence of moderate/severe motor impairment can be attributed partly to the increased mortality of children with these impairments in this community. We have found that children with severe neurological deficits following severe malaria (mainly motor deficits) are at an increased risk of dying after discharge from hospital.24
It is worth noting that the prevalence rates reported in this study are minimum estimates since the TQQ only reliably detects moderate to severe disability and not the mild conditions. Furthermore since it depends upon the parental reporting, it has a low sensitivity in detecting vision and hearing impairment. However, the TQQ still remains a useful tool for identifying children with disability in RPCs in which records are either non-existent or poor quality.
The risk factors associated with moderate/severe NI were potential neonatal insults. Similar findings have been reported in studies in Nepal and Kenya, where it was found that neonatal encephalopathy25 and tetanus26 increased the risk of NIs among survivors. Although 130 (52%) of children with NI were reported by the parents/guardians to have been admitted to hospital, we could only confirm 73 of these admissions to KDH, the main government hospital that serves this area. Some of the children may have been admitted to private clinics or other clinics. In those children admitted to KDH, CNS infections and birth trauma were identified as potential causes of NI. However, these had no association with NI, possibly because of the small sample size. Other risk factors such as lack of prenatal care, birth difficulties, home births, lack of maternal education, and maternal age >30 years at birth of child, which have been identified in other studies as risk factors for NI, especially cognitive impairment,27–29 were not identified in this study. This may have been caused by poor recall of birth events by mothers. We have found that mothers who deliver at hospital often forget many of the events surrounding the delivery of the child 4 years later.30
In addition, a selection bias caused by poor survivorship of infants born under these conditions may have resulted to an underestimation of neurologically impaired children in this community.
The significant burden of neuro-cognitive impairment in rural Africa is likely to increase, with the reduction in childhood mortality. However, there is need to establish the causes of NI in these areas, although prevention and aggressive treatment of neonatal insults would reduce the prevalence in this area. Furthermore, sustainable community-based rehabilitation services need to be instituted to support people with disabilities arising from this impairment.
|
Acknowledgments |
|---|
The Wellcome Trust-UK and the Kenya Medical Research Institute supported this study. We thank the mapping and census team, field staff, and assessors who made this study possible. In particular we thank Joseph Gona, Godffrey Otieno, Elizabeth Obiero, Khamis Katana, Kenneth Rimba, Gladys Murira, Judy T Dzombo, Francis Yaa, Douglas Konde, Mary Karisa, Francis Kanyetta, Silas Haro, Karen Konde, and Janet Chea. We also thank Drs Penny Holding and Julie Carter for their assistance in the design and development of assessment tools; Tansy Edwards, Dr Neal Alexander and Dr Anthony Scott for their comments on the manuscript; and Prof Kevin Marsh and Dr Norbert Peshu for their advice on the study design. This paper is published with the permission of the director of KEMRI. C.R.J.C.N. holds a Wellcome Trust Career Post in Clinical Tropical Medicine (No. 050533).
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1 Murray C, Lopez AD. Quantifying disability: data, methods and results. Bull World Health Organ 1994;71:481–94.
2 Helander E. Prejudice and Dignity: An Introduction to Community Based Rehabilitation. New York: United Nations Development Programme, 1993.
3 Belmont L. Final Report of the International Pilot Study of Severe Childhood Disability. New York: Gertrude Sergievsky Center, Columbia University, 1984.
4 Zaman SS, Khan N, Islam S, Durkin M. Childhood disabilities in Bangladesh: Report on rapid epidemiological assessment of childhood disabilities in Bangladesh 1992. Bangladesh: Bangladesh Protibondhi Foundation, 1992.
5 Thorburn M, Desai P, Paul TJ, Malcolm L, Durkin M, Davidson L. Identification of childhood disability in Jamaica: the ‘ten questions’ screen. Int J Rehabil Res 1992;15:115–27.[Medline]
6 Durkin, MS, Hasan, MZ, Hasan Z. The ten questions screen for childhood disabilities: Its uses and limitations in Pakistan. J Epidemiol Community Health 1995;49:431–36.
7 Mung'ala-Odera V, Meehan R, Njuguna P, Mturi N, Alcock K, Carter JA et al. Validity and reliability of the ‘Ten Questions’ questionnaire for detecting moderate to severe neurological impairment in children aged 6–9 years in rural Kenya. Neuroepidemiology 2004;23:67–72.[CrossRef][Web of Science][Medline]
8 Central Bureau of Statistics. The Kenya Economic survey 1999. Nairobi: Government Printers, 2000.
9 Central Bureau of Statistics. The 1999 Kenya Population Census. Nairobi: Government Printers, 2000.
10 Salt AT, Sonksen PM, Wade A, Jayatunga R. The maturation of linear acuity and compliance with the Sonksen-Silver Acuity System in young children, Dev Med Child Neurol 1995;37:505–14.[Medline]
11 Kamplex screening audiometer, P.C. Werth, London, UK.
12 Kathuria R, Serpell R. Standardization of the Panga Muntu test: a nonverbal cognitive test developed in Zambia. J Negro Educ 1999;67:228–41.
13 Kagan J. Matching Familiar Figures Test. Cambridge: Harvard University press, 1964.
14 Baddeley A, Gardner JM, Graham-Mcgrgor S. Cross-cultural cognition: developing tests for developing countries: special issues: Donald Broadbent and Applied Cognitive psychology. Appl Cogn Psychol 1995;9:S173–95.[CrossRef][Web of Science]
15 Holding PA, Taylor HG, Kazungu SD, Mkala T, Gona J, Mwamuye B et al. Assessing cognitive outcomes in a rural African population: Development of a neuropsychological battery in Kilifi District, Kenya. J Int Neuropsychol Soc 2004;10:246–60.[CrossRef][Web of Science][Medline]
16 Carter JA, Murira GM, Ross AJ, Mung'ala-Odera V, Newton CRJC. Speech and language sequalae of severe malaria in Kenyan children. Brain Inj 2003;17:217–24.[CrossRef][Web of Science][Medline]
17 WHO. International classification of Impairments, Disabilities, and Handicaps. Geneva: WHO, 1980.
18 Shrout P, Newman S. Design of two-phase prevalence surveys of rare disorders. Biometrics 1989;45:549–55.[CrossRef][Web of Science][Medline]
19 Couper J. Prevalence of childhood disability in rural KwaZulu-Natal. S Afr Med J 2002;92:549–52.[Web of Science][Medline]
20 Biritwum RB, Devres JP, Ofosu-Amaah S, Marfo C, Essah ER. Prevalence of children with disabilities in central region, Ghana. West Arf J Med 2001;20:249–55.
21 Milaat AM, Ghabrah MT, Al-Bar SMH, Abalkhail AB, Kordy NM. Population-based survey of childhood disability in Eastern Jeddah using the ten questions tool. Disabil Rehabil 2001;22:199–203.
23 Carter JA, Neville BG, White S, Ross AJ, Otieno G, Mturi N et al. Increased prevalence of epilepsy associated with severe falciparum malaria in children. Epilepsia 2004;45:978–81.[CrossRef][Web of Science][Medline]
23 Annegers JF, Rocca WA, Hauser WA. Causes of epilepsy: contributions of Rochester epidemiology project. Mayo Clin Proc 1996;71:570–75.[Abstract]
24 Carter JA, Mung'ala-Odera V, Neville BG, Murira G, Mturi N, Musumba C et al. Persistent neurocognitive impairments associated with falciparum malaria in Kenyan children. J Neurol Neurosurg Psychiatry 2005;76:476–81.
25 Ellis M, Manandhar N, Shrestha PS, Shrestha L, Manandhar DS, Costello AM. Outcome at one year of neonatal encephalopathy in Kathmandu. Nepal. Dev Med Child Neurol 1999;41:689–95.[CrossRef][Web of Science][Medline]
26 Barlow JL, Mung'ala-Odera V, Gona J, Newton CRJC. Brain damage after neonatal tetanus in a rural Kenyan hospital. Trop Med Int Health 2001;6:305–08.[CrossRef][Medline]
27 Durkin MS, Hasan MZ, Hasan ZK. Prevalence and correlates of mental retardation among children in Karachi, Pakistan. Am J Epidemiol 1998;147:281–87.
28 Durkin MS, Khan NZ, Davidson LL, Huq S, Munir S, Rasul E et al. Prenatal and postnatal risk factors for mental retardation among children in Bangladesh. Am J Epidemiol 2000;152:1024–32.
29 Kandamuthan M. Socio-economic factors and childhood disability in Trivandrum of South India. Int J Rehabil Res 1997;20:335–39.[Medline]
30 Mung'ala-Odera V, Newton CR. Recall of perinatal events by mothers living in rural Kenya. Epidemiology 2001;12:366.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  B. O OLUSANYA and A. A OKOLO Revisiting the ten questions questionnaire for developing countries Int. J. Epidemiol., August 1, 2006; 35(4): 1103 - 1103. [Full Text] [PDF]
|
|
|
|
|
|
S. Kauchali and L. L Davidson Commentary: The epidemiology of neurodevelopmental disorders in Sub-Saharan Africa--moving forward to understand the health and psychosocial needs of children, families, and communities Int. J. Epidemiol., June 1, 2006; 35(3): 689 - 690. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||