search text   Advanced Search

This Article Abstract Full Text (PDF) P3Rs: Submit a response Alert me when this article is cited Alert me when P3Rs are posted Alert me if a correction is posted Citation Map Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My File Cabinet Download to citation manager Citing Articles Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Maulik, P. K. Articles by Darmstadt, G. L. Search for Related Content PubMed PubMed Citation Articles by Maulik, P. K. Articles by Darmstadt, G. L. Related Collections Developmental/Behavior

Childhood Disability in Low- and Middle-Income Countries: Overview of Screening, Prevention, Services, Legislation, and Epidemiology

^a Departments of Mental Health
^b International Health, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, Maryland

	ABSTRACT

TOP ABSTRACT METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
BACKGROUND. Childhood disability affects millions of children around the world, most of whom are in low- and middle-income countries. Despite the large burden on child development, family life, and economics, research in the area of childhood disability is woefully inadequate, especially from low- and middle-income countries.

OBJECTIVE. The objective of this review was to generate information about current knowledge on childhood disability in low- and middle-income countries and identify gaps to guide future research.

METHODS. Electronic databases (PubMed, Embase, PsycInfo) were searched by using specific search terms related to childhood disability in developing countries. The Cochrane Library was also searched to identify any similar reviews. Whole texts of articles that met study criteria were scrutinized for information regarding research method, screening tools, epidemiology, disability-related services, legislation, and prevention and promotion activities. Quantitative and qualitative information was collated, and frequency distributions of research parameters were generated.

RESULTS. Eighty articles were included in the review (41 from low-income countries). Almost 60% of the studies were cross-sectional; case-control, cohort, and randomized, controlled trials accounted for only 15% of the studies. Of the 80 studies, 66 focused on epidemiologic research. Hearing (26%) and intellectual (26%) disabilities were the commonly studied conditions. The Ten Questionnaire was the most commonly used screening tool. Information on specific interventions, service utilization, and legislation was lacking, and study quality generally was inadequate. Data on outcomes of morbidities, including delivery complications and neonatal and early childhood illness, is particularly lacking.

CONCLUSIONS. With this review we identified potential gaps in knowledge, especially in the areas of intervention, service utilization, and legislation. Even epidemiologic research was of inadequate quality, and research was lacking on conditions other than hearing and intellectual disabilities. Future researchers should not only address these gaps in current knowledge but also take steps to translate their research into public health policy changes that would affect the lives of children with disabilities in low- and middle-income countries.

Key Words: childhood disability • developing countries • impairment • sense-organ disorders • mental retardation • low- and middle-income countries

Abbreviations: LAMI—low- and middle-income • RCT—randomized, controlled trial • TQ—Ten Questionnaire • WHO—World Health Organization

International effort and research have led to substantial reductions in the mortality rates of children <5 years old¹; however, research and progress in the area of childhood disability has been seriously lagging, particularly in low- and middle-income (LAMI) countries. An estimated 150 million children suffer from some kind of disability, and most live in the poorest parts of the world.² Moreover, a majority of these children suffer the double burden of disability and its associated stigmatization, leading to a marginalized life. The Bellagio Group on Child Survival called on all international organizations and funding agencies to support child-survival programs and outlined 4 steps: (1) develop worldwide leadership; (2) generate evidence-based practices; (3) increase country capacity; and (4) implement programs that are based on principles of equality and equity to reach Millennium Development goal 4.³ The Lancet's neonatal survival series identified knowledge of long-term developmental outcomes as a key public health gap.⁴ The Lancet subsequently published a 3-part series of articles that explored global indicators and burden of poor child development,⁵ risk factors,⁶ and potential strategies for addressing these problems.⁷ Thus, the importance of child development has been increasingly recognized in recent years. To promote efficient and effective progress in the introduction of programs to reduce the burden of childhood neurodevelopmental disabilities, we undertook this review to identify gaps in knowledge regarding the epidemiology, screening methods, prevention, service provision, policies, and legislation related to childhood disabilities in LAMI countries.

	METHODS

TOP ABSTRACT METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
We performed a comprehensive review of available literature to identify data on the prevalence of disabilities and impairments in children <5 years old in LAMI countries. LAMI countries were as defined by the World Bank (2006)⁸ on the basis of per-capita gross national income (in US dollars): low income, $875 and lower; lower-middle income, $876 to $3465; higher-middle income, $3466 to $10725; and high income, $10726 and higher.

Eligibility Criteria
The abstract from any study from a LAMI country⁸ that discussed childhood disability was reviewed for possible inclusion in the study database. There was no limitation to the year or type of study. This was done to cover as much literature as possible from developing countries, keeping in mind that research from developing countries is limited. Studies unrelated to childhood disabilities or based in high-income countries⁸ were excluded. Emphasis was placed on articles that provided information on research that addressed intellectual, hearing, speech, vision, motor, and neurologic impairment in a broader sense; less emphasis was placed on studies that assessed an intervention for the management of a specific syndrome (eg, Prader-Willi syndrome, cri-du-chat syndrome, Down syndrome, etc).

Search Strategy
Online medical databases were searched by using specific search strategies. PubMed was searched by combining the Medical Subject Heading (MeSH) terms "disabled children," "developing countries," "mental retardation," and "sensation disorders." The search was limited to infants, preschool-aged children, and children. Embase was searched by combining MeSH words "childhood disability," "sensation disorders," "sensory disorders," "mental deficiency," and "developing countries." PsycInfo was searched by combining thesaurus words "developmental disabilities," "mental retardation," and "sense organ disorders" with "developing countries" and limiting it to childhood (0–12 years). The Cochrane Library was also searched for any review on childhood disability with a focus on developing countries.

Initially, the abstracts of all relevant articles that matched the search terms were screened to identify articles that provided information on childhood disability related to screening tools, services, prevention and promotion, legislation, and epidemiology. Electronic and/or hard copies of studies that were found to provide information on any of these areas were obtained, and a snowballing hand search was performed of reference lists in relevant articles to identify any other study that potentially met our inclusion criteria.

More focus was given to relatively newer studies (conducted after 1990) and those that reported on neurocognitive disorders, including motor disabilities. Emphasis was given to articles that reported overall disability estimates. No attempt was made to search articles that focused on specific syndromes associated with any type of impairment. Mental disability, per se, was not a focus of this review, although intellectual disability was reviewed. Although hearing-, speech-, and vision-related disabilities were also included in the search and are reported here, relatively less emphasis was placed on those disabilities, and some articles related to those disabilities were not searched for once electronic and hard copies were found to be unavailable.

Data Management
Each full article was further screened to judge its relevance to the study objectives. Quantitative information pertaining to disability screening tools, services, prevention and promotion, legislation, and epidemiology were entered into an electronic database. The income group of the country in which the study was based was determined according to the current World Bank income groups.⁸ A fifth group of multicountry studies was identified that included >1 LAMI country. The research method used by each study was coded: cross-sectional studies were coded as 1, case-control studies as 2, cohort studies as 3, randomized, controlled trials (RCTs) as 4, and review articles that did not contain original data as 5. The review articles spanned comprehensive reviews and brief commentaries about certain aspects of childhood disability in a country. The study settings were also coded: community-based studies were coded as 1, clinic-based studies as 2, and special-population–or special-school–based studies as 3. When studies used a combination of study settings/populations, for the sake of quantitative analysis the higher setting (lower number) was chosen, provided subjects with disability (not controls) were drawn from that setting (eg, a study using both community-based [code 1] and clinic-based [code 2] samples was coded as community based). The rationale for doing this was that the study population was more inclusive in community-based studies compared with that of clinic-based studies, which in turn was more inclusive than that of special-population and special schools. Parameters for screening tools and/or tests used for screening, services, prevention and promotion, legislation, and epidemiology were each coded dichotomously as yes or no depending on whether the focus of the study was related to any of these topics. A mere mention of the topic in the discussion section was not considered as being a focus of the study.

Epidemiologic estimates of total disability and disability of neurologic, intellectual, hearing, visual, speech, and motor function were noted. Qualitative information on sampling method, use of standardized tools, discussion of bias and confounding, use of appropriate statistical analysis including provision of confidence intervals, and discussion of power/sample-size calculations was also entered into the database. Because the aim of this study was to identify potential gaps in information from LAMI countries, a rigorous qualitative assessment of each study on the basis of established guidelines was not performed while selecting them. The objective was to be less stringent on study quality as part of inclusion criteria and gather more information on the variety of content and range of quality of knowledge available about childhood disability in LAMI countries. A brief synopsis of the studies was included under the areas of focus covered by the article.

Data Analysis
The frequency distribution of quantitative data, except the epidemiologic estimates, was tabulated by using Stata 9.⁹

	RESULTS

TOP ABSTRACT METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
The results section is presented under the headings "Search Profile," "Study Setting," "Research Method Used in Studies Reviewed," "Screening Method," "Disability-Related Services," "Prevention and Promotion," "Legislation," and "Epidemiology." Some larger sections (eg, "Research Method Used in Studies Reviewed" and "Epidemiology") are further subdivided and discussed separately under each subsection. Community-based epidemiologic studies are further subdivided according to the types of disabilities addressed by the studies, given the importance of community-based epidemiologic research for large-scale programs. Both quality of the research and important observations made by different researchers are highlighted under each section, with an aim at placing the results in the context of the quality of the evidence available.

Search Profile
The search strategy is outlined in Fig 1. No reviews were identified in the Cochrane database that fulfilled our specific inclusion criteria of studies conducted in developing countries. The PubMed database search resulted in the identification of 148 articles, of which 75 were initially thought to be relevant after examination of the abstracts, and 47 were found to fulfill eligibility criteria after review of the full article. The Embase search resulted in 172 articles, of which 25 were initially identified on the basis of their abstract, and 16 were eventually selected after reading the whole article. Of the 4 articles identified in the PsycInfo database, 3 were found to be relevant to our review after close scrutiny. There was substantial overlap among the 3 databases. Hard copy of 1 probable article could not be located.¹⁰ Another study from the Dominican Republic was not included because it was in Spanish.¹¹ Overall, these databases generated 66 articles, and another 14 were included on the basis of a hand search of the reference lists of each of the articles, which resulted in a total of 80 articles identified for in-depth analysis.

View larger version (30K): [in this window] [in a new window]   FIGURE 1 Flowchart of search strategy.

Research Methods Used in Studies Reviewed
Table 1 shows that a cross-sectional design was used in 59% of the studies, case-control in 6% of the studies, and cohort and RCT designs in 5% of the studies. Almost 58% of all the cross-sectional studies were conducted in low-income countries, and approximately one third was conducted in middle-income countries. More than 77% of the studies were community based and used subjects who were chosen from either the population or general schools. More than 45% of the 62 community-based studies were from low-income countries. Clinic-based studies accounted for 10% of the studies, and 13% of the studies used subjects who had a specific disability and selected them from specialty clinics or schools.

Cross-sectional Studies
Among the 47 cross-sectional studies, 36 were community based. Study populations of children were identified either through population-based sampling^12–25 or schools.^26–33 Some studies focused on the parents of children with disabilities and tried to assess their needs and attitudes with respect to disability.^34–36 Among the others, 3 studies^37–39 were clinic based, with a primary focus on the epidemiology and available services for cerebral palsy. One group of investigators⁴⁰ used participants from clinics and special institutions, and a control group of normal children, to assess maternal risk factors for disability. All 6 of the special-population–or special-school–based studies had an epidemiologic focus on the prevalence, type, and severity of disabilities in the study population,⁴¹ causes of cerebral palsy,⁴² validation of tools,⁴³ or effect on caregivers.^16,44 Most of the studiesused some kind of sampling framework such as consecutive sampling, randomized sampling, or inclusion of all those within a delineated population. A process of randomization was used in the sampling stage by many researchers.^* Most used simple random sampling, whereas some studies used a stratified sampling technique based on various criteria such as type and level of schools^27,40 or socioeconomic criteria.²⁰ Only a few studies discussed both bias and confounding.^{16,18,20,21,27,37,48}

Case-Control Studies
Among the 5 case-control studies, 3 examined etiologic factors.^49–51 Matching was used in some of the case-control studies.^46,49–51 One study⁴⁹ used medical data and a questionnaire prepared for the study to assess the etiology of mild intellectual disability. No physical examination of the children was conducted, and there was no discussion of bias or confounders. Another study⁵⁰ was a population-based study that used screening methods, questionnaires, and standardized definitions in addition to matched controls on the basis of certain sociodemographic characteristics. The authors failed to discuss any bias or confounders that may have affected the results. The study focused on perinatal and maternal factors related to intellectual disability. Another community-based study⁴⁶ examined the effect of multiple early childhood intervention strategies, including nutritional supplementation regimens and neurophysical stimulation on children with stunted growth, used standardized tools and sound statistical analysis, and discussed potential bias, confounders, and other limitations of the study.

Social integration of children with epilepsy was the area of focus in a study in rural India.⁴⁷ The study used questionnaires prepared for the research to collect semiqualitative data on causes for poor social integration of affected children and interviewed parents on causes for nonparticipation of their children in social activities. The information was collected across different age groups and compared against age- and gender-matched controls. Nonparticipatory observation of the children in their societies was also done. The researchers discussed bias, confounding, and other methodologic limitations in their study and used appropriate statistical analysis.

Cohort Studies
Among the 4 cohort studies, 3 studied the prevalence of etiologic factors for disabilities,^52–54 and the other assessed outcome of cerebral palsy.⁵⁵ Gustavson⁵³ studied the health outcome of children born to a certain cohort of mothers residing in a circumscribed area. The families were followed up periodically for 12 years, and health outcomes, including neonatal health and mortality, were recorded. The children were examined clinically, and congenital disorders were diagnosed. Izuora⁵⁴ used prospective and retrospective clinical data to study etiologic factors for mental retardation. Both clinical and laboratory tests were used along with standard tools for assessing cognitive abilities. The study was conducted over a 4-year period. Bashir et al⁵² studied the prevalence of intellectual disability in children. Khan et al⁵⁵ studied a group of children from a cerebral palsy clinic over a 3-year period and ascertained their health outcomes. Ninety-two consecutive children were enrolled from the clinic and assessed clinically for physical problems; the children were assessed psychologically with standardized tools for different aspects of cognitive and social development. A common drawback of each of these studies was that none reported on bias, and only 1 study discussed confounding.⁵³

Randomized, Controlled Trials
McConachie et al⁴⁵ used an RCT design to evaluate the effect of 3 types of service-delivery strategies for assisting mothers of children with cerebral palsy. Children from both rural and urban settings were selected from special clinics and schools. Although the process of randomization was not described in detail, the study included allocation to 3 types of interventions. Participants from a rural community were divided into 2 groups: (1) a distance training group in which the parents were taught about child development and use of simple tools and aids to support their children's development, and (2) a health advice group wherein the parents were given simple toys for their children to play with but no special information about child positioning or other techniques. The urban community was also divided into 2 groups: (1) a distance training group, as described above, and (2) a mother-child group in which daily living skills were taught to the mothers by specially trained therapists. Verbal consent of the mother was obtained, and detailed clinical assessment was performed by a pediatrician. Statistical analysis provided confidence intervals and test statistics. The authors discussed the implications of the results but did not report on any weaknesses of the study design. Two other RCTs^56,57 studied the effect of zinc supplements given to both pregnant women and their infants on level of intellectual development at 13 months of age. Both studies failed to find any significant benefit. On the contrary, zinc supplements seemed potentially harmful. Russell et al⁵⁸ studied the effect of specific integrated group psychoeducation on families with children with disabilities. The parents were taught problem-solving and parenting skills, and they did better compared with a group whose families were not taught problem-solving skills, although they were also provided with parenting skills.

Reviews
The review articles were a mixture of studies that discussed various issues related to childhood disabilities, such as problems in conducting research in developing countries,^59–61 screening methods,^62,63 risk factors and prevention strategies, including early childhood strategies,^64–69 and available services.^{61,67,70–72} None were systematic reviews or included critical comments about individual articles. The review by McPherson and Swart⁷³ was on hearing impairment and provided an overview of prevalence of hearing disability in Sub-Saharan African countries; they outlined the etiology for disabilities and research needs in that area. While describing the epidemiology, the researchers subdivided the Sub-Saharan region into smaller geographical areas to describe the etiology regionally. The problem of hearing impairment and the role of nonspecialists, especially in developing countries, was discussed by other researchers.⁷¹ Yousef⁷² outlined available services in different Arab countries, with a special focus on education. Some reviews^61,67–69 discussed various causes for intellectual disability among children in developing countries and provided a framework for implementing preventive strategies to reduce the impact.

Screening Method
Approximately two thirds of the studies discussed screening tools or assessment methods, including clinical investigations, and more than half of those studies were performed in low-income countries. A multitude of screening tools were used in the studies, but most were related to assessing cognitive dysfunction and intellectual disability. These screening tools were generally standardized tools or adapted versions of the American Association on Mental Retardation Adaptive Behavior Scale⁷⁴; the Vineland Adaptive Behavior Scale⁷⁵; Griffith's Scale of Mental Development⁷⁶; or the Denver Development Screening Test.⁷⁷ Assessment of hearing impairment often included use of the Liverpool Field Audiometer, whereas visual impairment was assessed by using Snellen's chart and E-charts. Some researchers developed questionnaires for their study but provided inadequate information on psychometric properties of the instruments.^{12,25,30,34,36,47} The instruments gathered information on signs and symptoms of various disorders²⁵ and qualitative information on a child's social integration into society,⁴⁷ attitude of families toward their disabled children,^43,58 hearing ability,^30,34 and availability of human resources to provide service for disabled children.³⁶ Some of the researchers provided limited information on characteristics of the questionnaires and also compared them to more standard tools.^30,34,36 Little is known about the development of the questionnaires except for the Indonesian adaptation of the Vineland Adaptive Behavior Scale⁷⁸ and the scale to measure social integration of children with epilepsy in the Indian context.⁴⁷

Table 2 outlines the screening tools that were either validated or adapted by different researchers. Couper⁷⁹ modified the Ten Questionnaire (TQ) and added 6 additional questions to identify developmental impairments in children below 2 years of age. They piloted their questionnaire before using it, but no proper reliability or validity study was performed. The TQ, developed as a part of the International Pilot Study of Severe Childhood Disability,⁸⁰ was the most commonly used tool to assess disability in large populations.²⁴ It was found to have good specificity for identifying severe forms of mental retardation in the study by Belmont⁸⁰; however, in another study,¹⁵ it was not found to be a suitable screening tool for mild-to-moderate degrees of mental retardation. A detailed assessment of the tool was performed by Thorburn et al,⁴⁸ who found that the TQ was a good tool for assessing severe disabilities of all types except cognitive disabilities and tended to miss moderate degrees of intellectual impairment. However, they inferred that the TQ identified more severe cases but was limited by it being just a screening tool that provided little information on the degree of impairment and the type of services required. Thus, in most settings, the TQ needs to be supplemented by another, more detailed assessment, including 1 or more disability-specific tools to capture a broader range of disorders and to help identify the degree of impairment.

Disability-Related Services
Forty percent of studies provided some information on disability-related services, and of these, 44% were from low-income countries and 38% were from middle-income countries. Some country-specific details about services and needs are provided in Table 3.

Three comprehensive reviews,^68,85,86 completed over a span of 10 years, provided similar recommendations such as increasing intersectoral collaboration, involvement of national and international agencies, developing community-based services, and increasing training of staff as some of the steps for improving disability-related services.

Prevention and Promotion
Information about prevention and promotion activities in the area of childhood disability was reported by 21 (26%) of the 80 studies, 48% of which were from low-income countries. Strategies identified by various researchers are outlined in Table 4 and most incorporated techniques applicable to early childhood. Some of the prevention and promotion activities outlined in the studies included improving primary health care^63,67; increasing immunization coverage to protect against infections such as poliomyelitis and meningitis^41,54; implementing programs that provide nutritional supplements such as vitamin A, iron, and zinc⁶⁶; promoting effective health education programs that highlight the effect of certain genetic factors in causing different types of impairments; the importance of hypothyroidism and iodine deficiencies in causing intellectual impairment^53,69,87; increasing parental knowledge about available services related to different types of disabilities; and improving mother-child interaction.^68,72 The importance of avoiding iodine deficiency and measures to prevent hypothyroidism, especially within the Indian context, has been highlighted by others.^65,87

Legislation
Only 6 studies provided any information on legislation related to childhood disability. Two studies from China^14,69 discussed issues such as strengthening laws that pertain to protection of rights of children with disabilities, banning consanguineous marriage, and making immunization compulsory within a Chinese perspective. Yousef⁷² highlighted the importance of laws for protecting children with intellectual disabilities and developing national policies for integrated schooling facilities for such children. Nair and Radhakrishnan⁶⁷ discussed governmental policies regarding implementation of preventive measures, especially those that target girls. The importance of policies to reduce iodine deficiency in India was outlined by others.⁶⁵

Epidemiology
Epidemiology was the focus of 82% of the studies. Of these, 56% were from low-income countries and 27% were from middle-income countries. Of the 66 studies that reported on epidemiology, 45 were cross-sectional, 5 were case-control, 4 were cohort, 4 were RCTs, and 8 were reviews.

Research methods and selection of study population have been elaborated in earlier sections. Only 8 studies provided some information on all the criteria used to assess quality (sampling method, use of standardized tools, discussion of power/sample size, bias and confounding, and use of appropriate statistical analysis, including confidence intervals for estimates).^{20,21,27,37,48,56–58} Even among these 8 studies, detailed information on sample-size estimation was provided by only 2 research groups.^56,58 It is possible that the authors of these studies had performed these calculations but did not report them in the articles that were reviewed. Some studies discussed specific issues related to different biases such as selection^41,88 and information^{20,27,37,46,48} bias. These studies not only reported possible sources of bias but also discussed the strengths of the studies in their ability to reduce selection bias by randomization of the selection process or by using trained interviewers who applied standard instruments and performed reliability checks to control for information bias. Confounding and statistical adjustment for it using regression or stratification techniques were detailed by some researchers.^{20,21,27,40,46} Hartley⁸⁹ reported extremely high prevalence rates for all types of impairments. However, the study failed to provide information on sampling method, bias, confounding, and power of the study. Hartley also used a modified version of the TQ that was not properly validated. All these drawbacks made it difficult to correlate the high estimates the author obtained with other studies. There was wide variation in the sample size of the studies, ranging between 30 and 550000.

Community-Based Studies
Key information about some of the community-based epidemiologic studies is provided in Table 5.

Al-Ansari¹² used a door-to-door household-survey technique in Bahrain and administered a questionnaire that was developed for the study. The questions were answered by the head of the household and not corroborated by interviewing the affected person. The study included both children and adults; the prevalence of disability in male children was 0.4%, and that in female children was 0.3%.

Sauvey et al²² also used a door-to-door survey of a rural population in Nepal and asked the respondents about the presence of any member in the household aged <20 years who had a disability. They were also asked to name the type of disability. This simple, 2-question survey gave an overall prevalence of 1%, with the majority (89%) affected by a motor disability.

Using a random, stratified sample of rural households in Ethiopia, it was found that there was a 3.1% prevalence of disability there.⁹¹ Chen and Simeonsson,¹⁴ in their study in China, also used a house-to-house survey technique, but there was no information provided on the type of questionnaire used. The study was part of a national study and also provided estimates for specific disabilities. Intellectual disability had the highest prevalence at 1.8%.

The studies in Jamaica⁹² and South Africa,⁷⁹ used a 2-stage screening method, applying a standardized tool such as the TQ in the first stage and an evaluation protocol for those who screened positive in the second stage. Although Paul et al⁹² found that intellectual disability had the highest prevalence at 8.1%, Couper⁷⁹ found that among children with disability, the most common were neurocognitive and hearing disabilities.

Natale et al²⁰ studied a specific population group in India that comprised the 2 lowest income strata of the population. The study included children aged 2 to 9 years and used an adapted Tamil version of the TQ; the highest prevalence of disability was among the 2-year-olds (26%) followed by the 7- to 9-year-olds (15%) and 3- to 6-year-olds (9%). The authors also found that disability prevalence was greater among the lower of the 2 strata. They commented that the higher estimates in their study were most likely a result of inadequate validation of their instrument.

In Ghana, researchers⁹⁰ found that in their sample of children (<15 years), the overall disability rate was 1.8% and disability was highest among the 6- to 9-year-old group. Inadequate immunization, especially against measles, meningitis, rubella, and poliomyelitis, were important causes for disability. Reviews^59,85 highlighted the importance of epidemiologic issues such as standard definitions, information on different childhood morbidity and mortality indicators, and different risk factors for disability in LAMI countries.

Hearing Disabilities
Among studies that used a cross-sectional design, hearing impairment was the most frequently studied disability; prevalence estimates ranged from 0.4%^14,48 to 19.7%.⁸⁹ Higher prevalence rates were generally obtained in studies on schoolchildren compared with non–school-based community studies except for the study by Hartley,⁸⁹ which suffered from poor method and failed to control for bias and confounders and used a nonvalidated adaptation of the TQ. In a review on hearing impairment in Sub-Saharan African countries,⁷³ it was found that prevalence rates varied across countries from 0.3% in rural Gambian children to 13.5% among schoolchildren in Nigeria. The sample sizes varied across countries. The authors found that the most common etiologies were meningitis, measles, maternal rubella, febrile illnesses, and genetic causes; there was a large proportion of unknown etiology. The need for more epidemiologic studies, especially community-based surveys, studies on cultural healing practices, use of systematic research methods, standard definitions to define hearing impairment, and use of good instruments to assess hearing loss, was emphasized.

Intellectual Disabilities
The prevalence of intellectual disability varied from 0.09%¹⁷ to 18.3%.⁸⁹ The large variation could be a result of sampling framework, degree to which confounders and biases were accounted for in the study, and use of reliable and valid tools. A multicountry, 2-stage study design that included the TQ²⁴ found that prevalence of intellectual disability varied according to severity across countries. For severe mental retardation, the rates varied from 5 in 1000 in the Philippines to 40.3 in 1000 in India; and for mild mental retardation, prevalence varied from 4 in 1000 in the Philippines to 138 in 1000 in Bangladesh. Among community-based studies, 1 case-control study⁵⁰ reported an intellectual disability rate of 0.8%, with 62.5% of the cases falling in the mild area of the spectrum. Cohort studies^52,53 reported intellectual disability to have a prevalence of 2.8%, with mild mental retardation having a prevalence of 6.2% and severe mental retardation a prevalence of 1.1%. Although half of the mild cases had unknown etiology, 28% were a result of postnatal causes and the remaining resulted from prenatal causes. The most common causes for severe mental retardation were congenital problems such as Down syndrome, consanguineous marriage, and birth trauma. Shah⁶⁸ reviewed intellectual disabilities in 5 south-Asian countries and reported a prevalence between 0.5% and 1.5%. The most common causes were birth asphyxia and trauma, intrauterine growth retardation, infection, malnutrition, iodine deficiency, iron deficiency, neonatal jaundice, genetic disorders, and metabolic disorders. The neonatal period was extremely vulnerable. Another review⁶⁹ reported intellectual disability in 2% of the Chinese population, with a slightly higher male predominance. Iodine deficiency, especially in the hilly regions, was found to be prevalent. Tao⁶¹ also reviewed intellectual disability in China and reported a prevalence of mental retardation between 0.1% and 0.8%, with higher prevalence in rural areas. Perinatal factors associated with birth trauma, congenital factors, and maternal infections were some of the factors they identified as being associated with intellectual disability.

Visual Disabilities
Visual impairment varied between 0.1%¹⁴ and 12.5%.⁸⁹ An Indian study²¹ found a prevalence of 9.2% and used a vision-specific LV Prasad Functional Vision Questionnaire in contrast to a more generic questionnaire such as the TQ used by others.^48,92 A 2-stage design with an initial assessment by a trained community worker and a confirmatory assessment by a specialist was a method often used in studies that reported visual impairment.

Motor Disabilities
Motor disability was reported by 5 studies.^{14,25,48,89,92} The rates generally varied between 0.1%⁴⁸ and 0.4%⁹² except for 1 study,⁸⁹ which reported a rate of 62.2%. All the studies except the study by Chen and Simeonsson¹⁴ used a 2-stage design.

Speech Disabilities
Speech or neurologic disabilities were reported in only a few studies. Two were based on the same study population in Jamaica.^48,92 The others were performed in Uganda,⁸⁹ South Africa,⁷⁹ and Ghana.⁹⁰ The rates reported in the studies from Jamaica were 0.2% for neurologic disorders and 1.4% for speech disability. Hartley⁸⁹ reported a verbal communication problem in 49.4% of the population. However, the criteria for defining verbal communication problems were not clear. The study from South Africa⁷⁹ reported neurologic disability in 4.7% of the population and included both epilepsy and perceptual problems other than vision and hearing. They also reported speech problems in 2.4% of the population. They confirmed impairment levels on the basis of assessments made by specialists on those who screened positive.

Clinic-Based Studies
One old study from east Africa examined intellectual and speech disability, but it was of poor quality.⁹³ More recently, 4 studies focused on children with cerebral palsy.^37–39,45 Two studies^38,39 found that spastic diplegia was the most common type of cerebral palsy. Associated visual defect was present in 54% of the children assessed by Bhatia and Joseph,³⁹ but the parents were unaware of the problem. In an RCT and a follow-up study, McConachie et al^37,45 evaluated 3 different types of service-related interventions for improving the condition of such children in rural and urban settings. Both these studies are discussed earlier in this article. Follow-up was found to be affected primarily by 2 factors: male gender of the child and the parents being less adapted to their children's condition. Higher level of education and being from an urban community also predicted better outcome. The RCT showed that the mother-child group, which provided the most intensive package, benefited the most, but outreach interventions were also effective. One study⁴⁷ divided 88 children with epilepsy into 5 groups and compared them with controls to look for social integration. School attendance and social interaction were more severely impaired among girls, and the most common determinants of integration of these children were societal and parental attitudes. A cohort study⁵⁴ was conducted over a 4-year period and found that the most common causes were acquired (44%), congenital (33%), and idiopathic (23%). Although the most common congenital cause was Down syndrome, the most common acquired causes were birth trauma and neonatal jaundice (19.5%). Some of the early childhood preventive strategies highlighted in the authors' discussion were immunization, chromosomal screening during the antenatal period, prevention of malnutrition, and better antenatal care.

Special-Population or Special-School–Based Studies
Two of the special-population–based studies were of children with cerebral palsy.^42,55 Both studies had a higher number of boys. Spastic diplegia and quadriplegia were the most common types of cerebral palsy. Adverse outcomes (eg, malnutrition in children suffering from cerebral palsy) were common. In the 3-year cohort study⁵⁵ in Bangladesh, 93% of the children were suffering from malnutrition per Western standards, with more than double the rate among rural compared with urban children. The case fatality rate was 4% among urban children and 14% among rural children with cerebral palsy. The other study⁴² found that more than half of the children in the study cohort in India were suffering from malnutrition. The authors also found that although intellectual disability was the most common disability and affected 73% of the 1000 children assessed in the study, visual impairment affected 41% and epilepsy 32% of the children. Another study⁴⁰ included children with 3 types of impairment (visual, auditory, and intellectual) in addition to a normal cohort. They studied the maternal risk factors associated with disability and found that maternal age of <16 or >30 years and multiparity were some of the maternal risks associated with the different impairments. Illiteracy, unemployment, and consanguinity were other associated factors. The authors of a case-control study⁴⁹ found that prenatal causes were responsible for 39% of the cases of mild mental retardation, and consanguinity, illiteracy, and family history of mental retardation were associated risk factors. A case-control design was used to assess the etiology of cerebral palsy, mental retardation, and visual and hearing impairment in an Afghan clinic.⁵¹ High rates of consanguinity and lack of universal and comprehensive antenatal care resulting from lack of accessibility, inadequate services, and illiteracy were some of the underlying causes of the impairments.

Four studies looked at the impact of children with disabilities on the family.^16,43,44,58 All but 1 of the studies⁵⁸ used a cross-sectional design and interviewed parents of children with disabilities regarding the effect on them both psychologically and financially. Negative attitudes toward their children, high expressed emotion, and concern regarding the effect of the child's illness on the overall functioning of the family were found in 2 studies.^16,43 All the studies showed that parents wanted more information regarding available professional services, job opportunities, education, and financial support. Another study⁴¹ found that vitamin A deficiency and measles were the most common causes of blindness in 50% of the children in a school for the blind. Avoidable causes were identified in 68% of the cases.

	DISCUSSION

TOP ABSTRACT METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
This study is, to our knowledge, the first attempt to review research on childhood disability from LAMI countries. The aim of the study was to identify the gaps in knowledge that could be addressed by future research.

Overall, the results show that researchers in LAMI countries have primarily focused on the epidemiology of childhood disability using cross-sectional community-based studies. Few studies used robust RCT designs, there is much variation in the tools used to study the problems, and few researchers have addressed issues such as confounding, bias, sample size, and use of appropriate statistics. Thus, if one were to apply critical measures of quality, many of the studies would not live up to acceptable standards of evidence-based scientific research. We were able to identify potential areas of future research, especially in the areas of prevention and promotion, services, and legislation in addition to the need for more robust studies on epidemiology.

Although every effort was made to conduct a search using a broad search strategy to identify all relevant research, ours was not a systematic review; hence, the scope of the search had limitations. Thus, it is possible that some pertinent studies may not have been included in this review, although the initial search results were supplemented by additional reports gleaned from the references of each article. Including only studies in the English language implies that any non–English-language publications have been missed, although they were included if an abstract was available in English. This also means that this review identified literature from countries that disseminate scientific research primarily in English; thus, the countries listed in this review may reflect that bias. Also, a number of articles related to specific syndromes might have been missed by not searching for them because it was not within our scope of research. The disability-causing conditions examined in this study included neurologic, intellectual, hearing, visual, speech, and motor disabilities. However, there are other conditions that could be included such as severe mental disorders and severe neurologic disorders. Future research could use a broader definition and include specific disability-producing conditions and syndromes (eg, Down syndrome) that are not addressed in this review; however, it is likely that our search strategy captured much of this literature from LAMI countries. Because a number of severe neurologic disorders also lead to increased mortality, a future review that addresses such an outcome is also possible. The definitions of disability used by different researchers also vary and made it difficult to compare the results across studies. Although the studies were reviewed critically, no predefined quality criteria were used to include or exclude studies, because the aim of the study was to identify gaps in knowledge; thus, the broadest possible lens for study inclusion was maintained. Future studies can build on our review by expanding the scope and making it more stringent with regards to quality of studies; however, we felt it was important at this stage in the genesis of evidence-based approaches to child development in LAMI countries to take a more inclusive approach. Finally, the statistical analyses were purely descriptive, and no attempt was made to report pooled estimates because of the huge variation in the methods used in the studies. Only ranges of different epidemiologic estimates are provided.

Need for Research in Childhood Disability in LAMI Countries
In 1990 the United Nations published a document that outlined the different methods to collect data on disabilities across countries, prompted by a review of existing literature at that time.⁹⁴ The need for such a document was determined after reviewing existing research at that time. The United Nations study also found that disability estimates varied between 0.2% and 20.9%. Since then, the World Health Organization (WHO) has taken major steps in defining disabilities and developing tools to assess the level of disability as part of its disability-related work. A recent concept article⁹⁵ by the WHO on disability and rehabilitation reported that there are an estimated 600 million disabled individuals in the world, of whom 200 million are children. Poverty was the major determinant of disability, which was most common in LAMI countries. Acknowledging the gravity of the situation, the 58th World Health Assembly⁹⁶ came up with a resolution that urged member states to develop their knowledge base about disability, implement evidence-based programs for rehabilitation, and formulate policies and legislation to strengthen the rights of individuals who are living with disabilities. Although some knowledge is available about disability-related issues among adults, the same is not true for childhood disability, especially within LAMI countries, in which both knowledge and activities directed toward childhood disabilities are severely limited.

Table 6 highlights some of the gaps in knowledge identified through this review and suggests some recommendations for future endeavors. Often, the recommendations to overcome potential gaps in knowledge are similar across different problem areas. The discussion below outlines some of those gaps and suggests measures to overcome those gaps. The intent is not to be prescriptive but to provide recommendations across different problem areas in an overarching manner without being too specific and repetitive.

Special Need for Research Related to the Neonatal Period
Information available from both developed and developing countries point to the special importance of the neonatal period for future growth and development of the child. Researchers have identified various etiologic conditions, such as birth asphyxia, meningitis, jaundice, hypothyroidism, prematurity, etc, which are particularly relevant to future disabilities and are of utmost importance during the neonatal period. Appendices 1 and 2 provide details about such research from LAMI countries. A number of preventive strategies highlighted in Table 4 are especially relevant to the neonatal period, including screening for genetic disorders and neonatal hypothyroidism, immunization, nutritional supplements, and appropriate antenatal and postnatal care. However, information about long-term cognitive and other impairments of children with neonatal high-risk conditions is particularly lacking from LAMI countries; hence, more research is needed to generate such knowledge. Because most of the neonatal high-risk conditions can be avoided or minimized by providing appropriate maternal and child health services, it makes sense for policy makers to focus on strategies that are needed in their countries to reduce neonatal risk factors. Besides this, policy makers should also encourage programs that provide knowledge and supportive services to families with at-risk children.

Knowledge About Childhood Disability and Regional Variation
This review highlights the paucity of information on childhood disability available from LAMI countries. A cursory search of medical databases for childhood disability shows the vast discrepancies in knowledge between LAMI and high-income countries. Researchers in different high-income countries have a better sense about the burden of the problem, its economic impact, and trends over time.^97–100

It is evident from the review that there is a paucity of research from some parts of the world, especially South and Central America, southeast Asia, central Asia, and large parts of Africa. Some research is available from Asian countries such as India, Bangladesh, Pakistan and China and African countries such as South Africa and Nigeria.

Although one reason for this asymmetry could be that most indexed medical journals accept only English-language articles, it is possible that the true reason is an actual deficit in quality research; for example, China is able to publish a number of studies in international journals despite its medical fraternity not being taught in English. However, a better understanding of this is only possible through a multilanguage search of regional databases.

Definitions and Screening Instruments
One issue that has often been a major concern among researchers in the area of childhood disability is the definitions used to define the problems and standardized methods of measuring them. The studies reviewed failed to address this issue and used varying definitions of disability that were based on either cutoff scores on different measures of intellectual impairment or levels of sound above which hearing loss was assumed. This results in a medical diagnosis, which is not the best way to ascertain level of disability in children because it does not take into account the level of functional limitations, degree of service utilization, or impairment of role performances.¹⁰¹ The WHO¹⁰² has developed a specific tool and assigned specific definitions for standardized assessment of disability and impairment that incorporate all the above-mentioned concerns in the measure. However, none of the recent studies from LAMI countries used a comprehensive definition or used any measure that provides a wholesome estimate of the problem of childhood disability.

Although the TQ was the most common screening tool used by researchers, its validity is highest for identifying forms of disabilities on the more severe end of the spectrum but has limitations in detecting milder disabilities. Research has also found that the questions on hearing, vision, and seizures in the TQ have lower reliability compared with other questions, and the question on "slowness" is highly dependent on cultural understanding and parental perceptions and is most likely to result in heterogeneity across different cultures.¹⁰³ Moreover, TQ-positive individuals must undergo secondary testing to identify the precise nature of the disability. Thus, more research needs to be done to develop standardized, culturally sensitive, valid instruments for use by professionals that can conform to the standard definitions of the concept of disability as outlined by researchers in the area of childhood disability, or researchers in LAMI countries need to use the standard WHO instrument in conducting research. Moreover, strategies that link community-based screening (eg, with the TQ) with gold-standard professional assessment and development of a tailored intervention approach need to be worked out.

An oft-used technique by researchers in LAMI countries was to use a standard instrument and translate and back-translate it into their native language and use that translated version for their research. This fails to address cultural differences in expression of symptoms, and a more complex process that requires generating new questions and performing extensive psychometric tests to assess reliability and validity is needed. Unfortunately, this would take time and money, both of which are limited in LAMI countries, given their limited pool of expertise and funding. Collaboration with international institutes with experience in such research could be a way of reducing the implication of inadequate human resources. Also, given the expense of modern research, collaborations with other institutes and even across sectors, both nationally and internationally, might be a way of sharing the cost and burden of research.

Epidemiology
Another issue that prevents researchers in LAMI countries from developing a better understanding of the burden of childhood disability is the type of research undertaken. A majority of the studies from LAMI countries are cross-sectional in nature and provide some epidemiologic estimate; however, few researchers have conducted cohort studies, and none have addressed the issue of developmental changes over time and their effect on disability. All the cohort studies focused on intellectual disabilities and primarily studied clinical outcomes. Because cohort studies provide good estimates of causality and allow one to track changes of different predictors over time, it is imperative that more research in LAMI countries focus on conducting longitudinal studies to improve the understanding of childhood disability. Other advantages of longitudinal studies are that they allow one to assess multiple outcomes such as different types of disabilities or different types of service utilization. This, in turn, would allow better estimation of the burden of disability and the gap in health-related service utilization. The limited number of RCTs make it difficult to evaluate the efficacy of any intervention. All the RCTs mainly focused on intellectual disability, primarily from a south-Asian perspective. Although these studies are invaluable, especially given the paucity of quality research, others will need to use this information to develop studies that are suitable for their own setting and area of focus.

This review shows that a number of researchers had focused on certain epidemiologic aspects of childhood disability. However, most of the epidemiologic research is related to descriptive epidemiology that outlines the prevalence of different types of disabilities and enumerates their association with certain etiologies. Less is known about the impact of childhood development on disability or the variation across age groups according to severity, although the importance of the neonatal period is evident. Even the etiology is mere association, because most studies were cross-sectional in nature. One way in which the epidemiologic research could be improved is by generating more analytically oriented research. Some of the areas of research that epidemiologic studies need to address are the association between different types of disabilities and child development and the association between service utilization and type or severity of impairment. Parental needs and expectations and their link with available service-utilization rates are other areas that need to be investigated. More qualitative research on culturally appropriate assessment techniques, perceptions of stigma, and needs assessment is also required.

Prevention and Promotion and Services Related to Disability
Although researchers have identified a number of prevention and promotion activities that are deemed beneficial in alleviating childhood disability in the context of LAMI countries (Table 4), few have actually reported on specific early childhood intervention–related research that they have conducted, the exception being RCTs in the area of intellectual disability^45,56,58 or involvement in national programs on salt iodization in India,⁸⁷ which have been found to be beneficial in reducing neonatal hypothyroidism.⁶⁵ However, researchers realize the importance of community-based interventions and services that involve families as the most sustainable of activities. This is encouraging, given that recent evaluation of the Integrated Management of Childhood Illness program in Bangladesh, Brazil, Peru, Tanzania, and Uganda has highlighted the importance of community-based, family-oriented programs that are based on a sound epidemiologic and cultural framework.¹⁰⁴ One strategy could be to include key indicators of neurodevelopment for preschool children across a range of domains including motor, vision, hearing, and intellectual within larger maternal and child health care programs; this would require strong linkage between community-based programs and clinic-based family care services, including diagnostic and intervention services. The need for proper supervision and effective referral services was also stressed by others when they assessed the importance of trained birth attendants in reducing perinatal complications of mothers and newborns.¹⁰⁵ The importance of all these factors has also been realized by researchers working in a LAMI setup in which community-based clinics that provide family-based interventions are being promoted along with more emphasis on intersectoral collaboration to develop comprehensive programs.^67,68,85,86

Another important issue related to disabilities, especially in LAMI countries, is the stigma attached to certain disabilities such as intellectual disability. Intellectual disability along with other mental disorders and epilepsy are highly stigmatizing.^106,107 Stigma attached to hearing impairment has also been found.¹⁰⁸ Some researchers^47,69,83 have identified this issue and have tried to address it through their research, but more needs to be done. Stigma often unknowingly begins at home, where families undermine the efforts of their children and restrict activities in which they would like to be involved. This is further amplified within the community when there is failure to assimilate individuals, particularly those with mild disabilities within the mainstream, and they are marginalized to special schools or offered restricted job opportunities suitable only for people with disabilities. Evidences of such an attitude are development of special schools and admission of any child with a minimal level of disability in such schools. Educators often realize the importance of including these children in normal schools and providing special care, but advocacy is lacking. In LAMI countries, inadequate assessment of the child's degree of impairment often results in faulty choices. In some cases, simple readjustments such as using microphones in class or placing a child near the teacher can alleviate the problem associated with hearing or vision impairment.³¹ Thus, childhood disability needs to be recognized and addressed by the society. Although milder forms of disability can be managed better by providing wholesome community-based services that provide education to the children besides other services, the more severe forms can be tackled by addressing the different etiologic factors that lead to severe disabilities (eg, encephalitis, meningitis, cerebral malaria, birth asphyxia/intrapartum hypoxia, iodine deficiency, iron-deficiency anemia, consanguinity, etc).⁶

However, some children who suffer from severe disability do require special training. The need for additional adequately trained staff to deliver such special care is highlighted by many researchers. The activities could entail services related to screening^17,53 or specialist activities related to rehabilitation⁷⁰ or establishment of special schools.⁷² The importance of developing managerial skills to coordinate effectively is an important component of any intersectoral activity and is required in services related to disability. The need for staff trained in the coordination of services between different professionals and services has been highlighted by others.^82,85

From a research perspective, no studies could be identified that addressed the issue of cost evaluations of specific intervention in LAMI countries. The reasons are inadequate knowledge about the efficacy of interventions in LAMI countries and, where information is available, lack of knowledge about their effectiveness in community settings. In the presence of limited budgets dedicated to specific interventions, the importance of more research in the area of cost-effective interventions using rigorous study designs is further underlined.

Legislation
Almost no information is available about existing policies and legislation with regards to childhood disability in LAMI countries. The little information available pertains to the implementation of immunization programs or iodine-supplementation programs. Most of the available discussion focuses on the need to develop legislation related to promotion of immunization and prohibition of consanguineous marriage in China¹⁴ or the rights of children with disability.^69,72 Only 2 recent studies^65,67 have discussed legislative issues or policy implications; they discussed screening for neonatal hypothyroidism and existing education and child development schemes from an Indian perspective. The importance of legislation cannot be overestimated given the stigma attached to certain conditions such as intellectual disabilities. National health and education policies should also address the issue of disability in a more comprehensive and collective manner to enable effective utilization of available resources.

	CONCLUSIONS

TOP ABSTRACT METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Most researchers in LAMI countries have focused on cross-sectional community-based epidemiologic studies that have ascertained either the prevalence of certain types of disabilities or the etiology of those disabilities. The methods used have varied and often failed to satisfy basic parameters by which the qualities of the studies were judged. Different instruments have been used to measure different types of impairments, but the TQ remained the most widely used screening tool even after 20 years since its introduction. RCTs were few in number; thus, information about effective evidence-based interventions was limited. Research in the area of services and legislation is negligible. Thus, there are huge gaps in knowledge regarding childhood disability in LAMI countries. Although it is desirable to collect information from all countries, this might not be a feasible option for many countries with limited resources and research capabilities. Hence, countries that have better research capacities should strive to generate quality research that could be extrapolated to some extent to other similar situations. More focused, organized, and higher-quality research that embodies greater international collaboration is needed before implementation of sound public health practices for prevention and management of childhood disabilities and impairments in such countries.

	ACKNOWLEDGMENTS

 
This review was supported by the Thrasher Research Fund, March of Dimes, Save the Children-US (through a grant from the Bill & Melinda Gates Foundation to the Saving Newborn Lives program), the Wellcome Trust Burroughs Wellcome Fund, and the Programme for Global Paediatric Research.

	FOOTNOTES

 
Accepted Mar 15, 2007.

Address correspondence to Gary L. Darmstadt, MD, MS, Department of International Health, E8153, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, MD 21205. E-mail: gdarmsta{at}jhsph.edu

The authors have indicated they have no financial relationships relevant to this article to disclose.

^* Refs 20, 25–27, 30, 32, 36, 40, and 45–47.

	REFERENCES

TOP ABSTRACT METHODS RESULTS DISCUSSION